Frisians of the Early Middle Ages 1783275618, 9781783275618

Citation preview

The thematic studies in this volume foreground how diverse ‘Frisians’ in different places and contexts could be. They draw on a range of multi-disciplinary sources and methodologies to explore a comprehensive range of social, economic and ideological aspects of early Frisian culture, from the Dutch province of Zeeland in the south-west to the North Frisian region in the north-east. Chronologically, there is an emphasis on the crucial developments of the seventh and eighth centuries AD, alongside demonstrations of how later evidence can retrospectively clarify long-term processes of group formation. The essays here thus add substantial new evidence to our understanding of a crucial stage in the evolution of an identity which had to develop and adapt to changing influences and pressures. JOHN HINES is Professor of Archaeology at Cardiff University. NELLEKE IJSSENNAGGER-VAN DER PLUIJM is Director of the Fryske Akademy, Leeuwarden. Contributors: Robert Flierman, John Hines, Nelleke IJssennagger-van der Pluijm, Egge Knol, G.J. de Langen, Tineke Looijenga, Bente Sven Majchczack, J.A. Mol, Johan Nicolay, Annet Nieuwhof, Han Nijdam, Arjen Versloot, Ian Wood

Cover image: Frisian creek (reconstruction). Illustration by Ulco Glimmerveen.

Studies in Historical Archaeoethnology

FRISIANS OF THE EARLY MIDDLE AGES

Frisian is a name that came to be identified with one of the territorially expansive, Germanic-speaking peoples of the Early Middle Ages, occupying coastal lands south and south-east of the North Sea. Highly varied manifestations of Frisian-ness can be traced in and around the north-western corner of the European continent in cultural, linguistic, ethnic and political forms across two thousand years to the present day.

Frisians

of the Early Middle Ages Edited byJohn Hines and Nelleke IJssennagger-van der Pluijm

Studies in Historical Archaeoethnology

Volume 10

FRISIANS OF THE EARLY MIDDLE AGES

Z01 Hines Book B.indb 1

07/05/2021 13:29

Studies in Historical Archaeoethnology FOUNDING EDITOR: GIORGIO AUSENDA Already published: After Empire: Towards an Ethnology of Europe’s Barbarians  edited by Giorgio Ausenda The Anglo-Saxons from the Migration Period to the Eighth Century: An Ethnographic Perspective  edited by John Hines Franks and Alamanni in the Merovingian Period: An Ethnographic Perspective  edited by Ian Wood The Visigoths from the Migration Period to the Seventh Century: An Ethnographic Perspective  edited by Peter Heather The Scandinavians from the Vendel Period to the Tenth Century: An Ethnographic Perspective  edited by Judith Jesch The Continental Saxons from the Migration Period to the Tenth Century: An Ethnographic Perspective  edited by Dennis H. Green and Frank Siegmund The Ostrogoths from the Migration Period to the Sixth Century: An Ethnographic Perspective  edited by Samuel Barnish and Federico Marazzi The Langobards before the Frankish Conquest: An Ethnographic Perspective  edited by Giorgio Ausenda, Paolo Delogu and Chris Wickham The Baiuvarii and Thuringi: An Ethnographic Perspective  edited by Janine Fries-Knoblach and Heiko Steuer

Z01 Hines Book B.indb 2

07/05/2021 13:29

FRISIANS OF THE EARLY MIDDLE AGES Edited by

John Hines and Nelleke IJssennagger-van der Pluijm

THE BOYDELL PRESS

Z01 Hines Book B.indb 3

07/05/2021 13:29

© Contributors 2021 All Rights Reserved. Except as permitted under current legislation no part of this work may be photocopied, stored in a retrieval system, published, performed in public, adapted, broadcast, transmitted, recorded or reproduced in any form or by any means, without the prior permission of the copyright owner

First published 2021 The Boydell Press, Woodbridge ISBN  978 1 78327 561 8 hardback ISBN  978 1 80010 132 6 ePDF A CIP catalogue record of this publication is available from the British Library The Boydell Press is an imprint of Boydell & Brewer Ltd PO Box 9, Woodbridge, Suffolk IP12 3DF, UK and of Boydell & Brewer Inc. 668 Mt Hope Avenue, Rochester, NY 14620-2731, USA website: www.boydellandbrewer.com

This volume contains the papers presented at a conference on ‘Studies in Historical Archaeoethnology’ organized by the Center for Interdisciplinary Research on Social Stress Cover image: Frisian creek (reconstruction). Illustration by Ulco Glimmerveen.

CONTENTS

List of Figures

vii



List of Tables

xi

Acknowledgements

xii

Abbreviations

xiii



xiv

Conventions for the Representation of Names

1. Frisians of the Early Middle Ages: An Archaeoethnological Perspective Nelleke IJssennagger-van der Pluijm, John Hines and Ian Wood 2. For Daily Use and Special Moments: Material Culture in Frisia, AD 400–1000 Egge Knol 3. The Frisians and their Pottery: Social Relations before and after the Fourth Century AD Annet Nieuwhof

1 13

45

4. Landscape, Trade and Power in Early-Medieval Frisia Gilles de Langen and J. A. Mol

79

5. Law and Political Organization of the Early Medieval Frisians (c. AD 600–800) Han Nijdam

137

6. Recent Developments in Early-Medieval Settlement Archaeology: The North Frisian Point of View Bente Sven Majchczack

171

7. Franks and Frisians Ian Wood

203

8. Mirror Histories: Frisians and Saxons from the First to the Ninth Century AD Robert Flierman

223

9. Structured by the Sea: Rethinking Maritime Connectivity of the Early-Medieval Frisians Nelleke IJssennagger-van der Pluijm

249

10. Art, Symbolism and the Expression of Group Identities in Early-Medieval Frisia J. A. W. Nicolay

273

v

Z01 Hines Book B.indb 5

07/05/2021 13:29

Contents 11. Religion and Conversion amongst the Frisians John Hines

311

12. Traces of a North Sea Germanic Idiom in the Fifth–Seventh Centuries AD Arjen P. Versloot

339

13. Runic Literacy in North-West Europe, with a Focus on Frisia Tineke Looijenga

375



Final Discussion

401



List of Contributors

407

Index

409

vi

Z01 Hines Book B.indb 6

07/05/2021 13:29

FIGURES Frisians of the Early Middle Ages: An Archaeoethnological Perspective, Nelleke IJssennagger-van der Pluijm, John Hines and Ian Wood 1.1

The Frisian zone in the southern North Sea area.

3

For Daily Use and Special Moments: Material Culture in Frisia, AD 400–1000, Egge Knol 2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.9 2.10 2.11 2.12

Part of the mound of Garnwerd, Groningen, being quarried. Textile fragment from Rasquert in Groningen. Leather shoe from Hallum-Jousemaburen. A semi-manufactured comb fromWirdum, Groningen. A wooden stool found in grave 44 at Hogebeintum. Two bowls of alder wood from graves at Hogebeintum. A wooden trough of the fifth century AD re-used as a coffin. Two beams of a harrow, re-used for reinforcement at the base of a well at Hattersum, northern Germany. (a) The half dug-out canoe from Jemgum; (b) A ceramic boat found in the terp of Pogum. Two yew-wood bows, from Aalsum and Hoogterp. A pectoral cross of the ninth century. The inscribed whalebone staff from Bernsterburen.

14 17 18 20 21 21 23 24 27 28 31 32

The Frisians and their Pottery: Social Relations before and after the Fourth Century AD, Annet Nieuwhof 3.1 Areas of different pottery styles. 3.2 Small pots of the first century AD from terps in the province of Groningen in the new, ‘Chaucian’ style. 3.3 One of the excavated levels in the terp of Ezinge. 3.4 Pottery from the fourth-century transition period in Ezinge. 3.5 Annular brooches with backward-looking animal heads from the northern Netherlands and north-western Germany dated to the fourth or early fifth century. 3.6 The distribution of Anglo-Saxon style pottery in the northern Netherlands. 3.7 Silver ‘hand pin’ of British origin found in a terp near Hallum. 3.8 Three sherds, allegedly of Oxfordshire Ware, dated to the late third and fourth centuries, from Groningen terps. 3.9 Finds from a fifth-century sunken hut in the terp of Ezinge, probably a ritual deposit.

48 50 52 53 55 57 60 61 62

vii

Z01 Hines Book B.indb 7

07/05/2021 13:29

Figures Landscape, Trade and Power in Early-Medieval Frisia, Gilles de Langen and J. A. Mol 4.1 Early-medieval Frisia and its districts c. AD 700. 84 4.2 High-medieval manors in the county of Holland projected on the landscape of c. AD 700. 85 4.3 Territorial parishes and their churches c. AD 1000 in Westergo and Oostergo. 89 4.4 Territorial parishes and their churches c. AD 1000 in Eemsgo. 90 4.5 Territorial parishes and their churches c. AD 1000 in Kennemerland and Rijnland. 92 4.6 Distribution of different forms of field parcellation in Central Frisia and Groningen East Frisia. 104 4.7 Modern elevation map of a part of Westergo in 1832 showing regular block-field parcellation. 105 4.8 Modern elevation map of a part of Westergo in 1832 showing irregular block-field parcellation. 105 4.9 Hogebeintum in Oostergo in 1832. 106 4.10 Firdgum in Westergo in 1832. 106 4.11 Biessum in Fivelingo in 1832. 107 4.12 Foudgum in Oostergo in 1832. 108 4.13 Wijnaldum in Westergo in 1832. 108 4.14 Eleventh- and twelfth-century (pre-)urban centres in or near Frisia. 109 4.15 Sexbierum in Westergo in 1832. 110 4.16 Oldeboorn in Oostergo in 1832. 111 4.17 Middelstum in Hunsingo in 1832. 114 4.18 The Wijnaldum and Dongjum terp complexes and the terps of Westerbierum (Pietersbierum and Sexbierum) and Oosterbierum and their surroundings in 1832. 114 Law and Political Organization of the Early Medieval Frisians (c. AD 600–800), Han Nijdam 5.1

A diagram representing the nested circles of the embodied honour of an individual under Old Frisian law. 5.2 The Totius Frisiae seal of 1323.

146 153

Recent Developments in Early-Medieval Settlement Archaeology: the North Frisian Point of View, Bente Majchczak 6.1 Map of North Frisia with sites mentioned in the text. 6.2 Aerial view of the Migration-period settlement at Wrixum/Föhr. 6.3 Schematic model of the system of trade and communication in the southern North Sea area. 6.4 Magnetic map (±9 nT) of the settlement at Witsum with excavation trenches.

172 175 176 179

viii

Z01 Hines Book B.indb 8

07/05/2021 13:29

Figures 6. 5 Example of a sunken-feature building from Witsum/Föhr with traces of a standing loom. 6.6 Examples of vessel-glass sherds and glass beads from Witsum/Föhr. 6.7 Magnetic map (±9 nT) of the settlement at Goting. 6.8 Aerial photograph with a view of the Borgsumburg in the foreground by the southern coast of Föhr.

180 181 182 186

Franks and Frisians, Ian Wood 7.1

Map of the Frisian region showing the principal sites referred to.

204

Art, Symbolism and the Expression of Group Identities in Early-Medieval Frisia, J. A. W. Nicolay 10.1 The leather scabbard of a seax from Usquert (Groningen). 274 10.2 Copper-alloy brooches deposited as grave goods at the cemetery of Oosterbeintum. 278–9 10.3 Copper-alloy, silver and pewter brooches from settlement sites in Friesland. 282–3 10.4 Copper-alloy brooches from a settlement site and two cemeteries in South-Holland. 284 10.5 The stylistic development of Domburg brooches, showing a gradual ‘debasement’ of animal motifs. 285 10.6 Chronology of copper-alloy brooches that have been found in the area of Early-medieval Frisia. 287 10.7 Gold bracteates of Scandinavian types A, B and D found in Friesland. 289 10.8 Saxon-style silver brooches and regional-style silver-gilt brooches found at settlement sites in Friesland. 290 10.9 A selection of high-status ornaments. 291 10.10 Religious motifs on ‘Frisian’ ornaments. 294 Religion and Conversion amongst the Frisians, John Hines 11.1 Map of the Frisian region showing the principal sites referred to.

315

Traces of a North Sea Germanic Idiom in the Fifth–Seventh Centuries AD, Arjen P. Versloot 12.1 A range of name-types commonly held to represent settlements from the Early Middle Ages, together with the distribution of runic finds. 12.2 The relative proportions of the name-types in Figure 12.1 (pale green: ‘overall’) and the distribution of runic finds of the third–eighth centuries correlated with the name-types. Areas and runic finds that are not associated with any of the six name-types are not included. 12.3 Correlation between the Anglo-Frisian and Continental runic finds of the third–eighth centuries and the various name-types.

351

352 352

ix

Z01 Hines Book B.indb 9

07/05/2021 13:29

Figures 12.4 Evidence of Germanic settlement and the distribution of early runic inscriptions of the fifth and sixth centuries. 12.5 (a) The overall areal proportion of names ending in *-haim or *-ing in the West Germanic speaking countries; (b) the proportional correlation of these points with a runic find. 12.6 The distribution of cruciform and Saxon-style brooches against the background of Migration-period place-names. 12.7 Possible evidence of Anglo-Saxon and British populations in England during the fifth–seventh centuries. 12.8 Anglo-Frisian runic inscriptions of the fifth–seventh centuries against a traditional ethno-political map of England in the sixth century.

353 354 356 358 360

Runic Literacy in North-West Europe, with a Focus on Frisia, Tineke Looijenga 13.1 The ‘Frisian’ and ‘Frankish’ runic items. 377 13.2 Westeremden A. An inscribed ‘weaving-slay’ (or ‘reed’). 379 13.3 Runes inscribed on the plates of the Oostum comb. 380 13.4 The runes on a miniature yew-wood sword from Arum. 381 13.5 The word kobu (‘comb’) inscribed in runes on the Toornwerd comb. 381 13.6 Westeremden B. A stick of yew-wood, inscribed with runes. 382 13.7 Ring-sword from Saint-Dizier, inscribed with the runes alu on the pommel. 384 13.8 The Amay comb and its runic inscription. 385 13.9 The ‘Schweindorf ’ runic solidus. 387 13.10 The æniwulufu tremissis. 387 13.11 The Bergakker scabbard mount and its runic inscription. 388 13.12 The Borgharen belt buckle and its runic inscription. 388 13.13 A core area in which runes occur to the south and south-west of Scandinavia in the fifth and sixth centuries. 389 13.14 A stick of yew-wood from Britsum with double carved runes. 390 13.15 An inscribed whalebone tau-staff from Bernsterburen inscribed with runes reading tuda. 391 The editors, contributors and publisher are grateful to all the institutions and persons listed for permission to reproduce the materials in which they hold copyright. Every effort has been made to trace the copyright holders; apologies are offered for any omission, and the publisher will be pleased to add any necessary acknowledgement in subsequent editions.

x

Z01 Hines Book B.indb 10

07/05/2021 13:29

TABLES 5.1 10.1 12.1 12.2 12.3

Wergild ratios in Lex Frisionum and numbers of oaths of innocence per estate. 149 Copper-alloy brooches and a gold-filigree and silver-cloisonné brooch from grave contexts in Early-medieval Frisia, in the most probable chronological order. 280 Examples of similarities between North Sea Germanic language varieties. 339 Selected ‘Anglo-Frisian’ sound changes and their relative chronological ordering in English and Frisian according to the current conventional understanding. 341 An overview of the dating and geographical spread of prominent North Sea Germanic linguistic developments. 345

xi

Z01 Hines Book B.indb 11

07/05/2021 13:29

ACKNOWLEDGEMENTS This volume presents papers delivered at a symposium in Leeuwarden on the various ‘Frisian’ communities and regions, in 2018. The event could not have taken place without the Fries Museum having provided the venue, together with technical and administrative support, for which we are immensely grateful. We are equally thankful for the financial and practical support provided for the symposium by the College of Arts, Humanities and Social Sciences, Cardiff University, the Koninklijk Fries Genootschap and the Society for the Study of Medieval Languages and Literature. All those who have provided and permitted the reproduction of images for the current volume or contributed in another way by providing access to material receive our gratitude too. We specifically want to thank those who have made the publication of this volume possible through financial subventions, notably the Marc Fitch Fund and the Vereniging voor Terpenonderzoek, along with a private donation. Last but not least, the editors and contributors wish to record their thanks to Boydell and Brewer for their continued support for this series and the publication of this volume, and also to the anonymous referee for his or her meticulous, thoughtful and consistently constructive reading and comments on the materials submitted.

xii

Z01 Hines Book B.indb 12

07/05/2021 13:29

ABBREVIATIONS ARF Annales Regni Francorum AS Additiones Sapientum (Lex Frisionum) DB Diplomatica Belgica ep epistolae HE [Bede] Historia Ecclesiastica Gentis Anglorum LH [Gregory] Decem librum historiarum LHF Liber Historiae Francorum MGH Monumenta Germaniae Historica MGH SS Monumenta Germaniae Historica, Scriptores SFB sunken-feature building

Language varieties ModDu Modern Dutch ModE Modern English ModHG Modern High German ModINFris Modern Insular North Frisian ModWFris Modern West Frisian OE Old English OFris Old Frisian OHG Old High German ON Old Norse OSax Old Saxon PGmc Proto-Germanic WGmc West Germanic Abbreviated references to further language varieties are provided with Tables 12.2 and 12.3.

xiii

Z01 Hines Book B.indb 13

07/05/2021 13:29

CONVENTIONS FOR THE REPRESENTATION OF NAMES Regional place-names have largely been anglicized, but with certain limits so as to maintain precision in what is referred to. The Dutch provinces of Noord- and Zuid-Holland appear as North-Holland and South-Holland respectively. Friesland may appear in this form (the Dutch) or the Frisian form Fryslân as the authors have chosen. Where ‘East Friesland’ could be misleading or confusing, it was decided that it was better to keep the German Ostfriesland for that specific region; Nordfriesische, however, is usually translated as North Frisian, not least as part of the North Frisian zone is actually in Denmark beyond the corresponding region of Nordfriesland in Germany. The more common modern form Helgoland is used rather than Heligoland. The Frisian King Radbod is also widely known as Redbad. We have standardized on Radbod, except where a specific vernacular source refers to the king as Redbad.

xiv

Z01 Hines Book B.indb 14

07/05/2021 13:29

• 1 • FRISIANS OF THE EARLY MIDDLE AGES: AN ARCHAEOETHNOLOGICAL PERSPECTIVE Nelleke IJssennagger-van der Pluijm, John Hines and Ian Wood

I

t is a great pleasure to be able to add a volume dedicated to Frisians of the Early Middle Ages to the Studies in Historical Archaeoethnology series. All of the previous titles in this series derived from symposia organized by and presided over by the late Professor Giorgio Ausenda. From the early 1990s, the Center for Interdisciplinary Research on Social Stress (CIROSS) affiliated to the Università di San Marino ran a series of symposia examining, individually, major ethno-cultural groups which both formed in and were responsible for the re-shaping of Europe following the collapse of the Roman Empire. This is the tenth volume in that series, and the first representing a revival of the format and objectives defined by Giorgio Ausenda following the founder’s death in 2007. Hosted by the Fries Museum, a symposium on ‘The Frisians’ was organized and held in Leeuwarden in the province of Fryslân/Friesland, the Netherlands, from 11 to 14 September 2018, designed to coincide with the status of Leeuwarden as European City of Culture in that year. The concept of historical archaeoethnology is, as described by Ausenda, to ‘recover knowledge of several aspects of the life style and sociocultural conditions of past populations’ (Ausenda ed. 1995, 3, 295–304). With its focus on the major ethno-cultural groups whose rise defines the emergence of an Early Middle Ages/Medieval Period after the end of the Roman Period, it may surprise us that Frisians have not been the subject of this series up to now. Their Continental and North Sea neighbours have been addressed: the Continental Saxons, the Franks and Alemanni, the Anglo-Saxons and the Scandinavians. But perhaps where those Early-medieval groups have been or can be more clearly defined and therefore demarcated, the Frisian group, although well known, has been rather elusive and more challenging to capture. This is not least the case because historically ‘Frisian’ has been an identity that has stretched through the territory of various modern states, notably Germany, the Netherlands and Belgium, all of which have different research traditions, with the outputs of research published in different languages – including Frisian. Recent years have seen an increase in research on various aspects of the Early-medieval Frisian identity and culture in an international and interdisciplinary perspective, based primarily in historical, archaeological and linguistic sources. This is exemplified by the publication of Frisians and their North Sea Neighbours from the Fifth Century to the Viking Age (Hines and IJssennagger eds 2017), derived from an earlier conference held in Leeuwarden in June 2014, which examined aspects of the Frisian population and culture in a 1

Z01 Hines Book B.indb 1

07/05/2021 13:29

Frisians of the Early Middle Ages comparative perspective and focussed on overseas connections. Complementarily, the present volume is able to use the archaeoethnological perspective to bring greater and informative focus to bear on the internal diversity of the geographically Frisian areas, and in a way that also almost automatically generates a different chronological focus. The Frisians of the Early Middle Ages is the result of a fruitful discussion of studies from leading scholars, including early career researchers, covering a range of economic, social and ideological fields and using historical, archaeological and linguistic evidence on an equal footing, on the question of the early medieval Frisians. The ‘San Marino model’, as it became known, was in fact quite strongly ‘directive’ in the framework it imposed on the participants: not only the given format, of precirculated papers followed by intense discussion, that was recorded for edited transcripts to be published in due course, but also in demanding attention to a specified and comprehensive range of topics. The character of this approach is explained and evaluated in more detail below. All who experienced the San Marino symposia would agree that the framework was challenging, and also that it proved successful and productive. In certain ways, as is also explained in this introduction, we sought to adapt the model constructively to new opportunities and in the light of experience.

The present volume By the ‘Early Middle Ages’ or ‘Early Medieval’ is meant, as is conventional, the period between the fall of the Roman Empire in the West and the consolidation of a more feudal, manorialized and Christian culture across the greater part of Europe, including Britain and Ireland and virtually all of the Continent and Scandinavia, by the end of the eleventh century. The Frisian zone had never (fully?) been part of the Roman Empire, but Frisii and Frisiavones were close neighbours of the Gallic and Germanic provinces and interacted with the Empire socially and economically. Chronologically, there is in fact a core focus in the present volume on the seventh to ninth centuries AD: the historical horizon of the two known Frisian leaders Aldgisl and Radbod in the late Merovingian and early Carolingian Periods, and also the context in which important legal collection Lex Frisionum was generated around AD 800. This is a little later as well as being very different in perspective from what we may call the centre of gravity of Frisians and their North Sea Neighbours. What must particularly be stressed, however, is the strength and persistency with which the following papers and discussions demonstrate how a much longer perspective is essential: from the fourth century AD through to the ‘retrospective’ use of High- and Late-medieval evidence to uncover substantial clues about the Early Middle Ages. Just as considerably later evidence can have surprisingly substantial implications for the seventh- to ninth-century horizon, so the intermediary periods of survival, transmission or development are crucially important too. Geographically, the studies included in this volume cover not only the area of the Netherlands and north-western Germany that is identified as ‘Frisia’ in the Lex Frisionum, but also the latest substantial research results on the North Frisian Islands and the adjacent North Frisian coastal zone. This collection thus examines, discusses and compares the various Frisian lands around the southern North Sea area (Fig. 1.1). This is an important innovation, as collectively – not least in the discussions – it is shown that there is real validity in considering these apparently disconnected, and all too readily 2

Z01 Hines Book B.indb 2

07/05/2021 13:29

An Archaeoethnological Perspective

Fig. 1.1  The Frisian zone in the southern North Sea area, with a selection of principal sites and locations referred to in the text.

separated, Frisian zones together in relation to this period. The ‘Law of the Frisians’, Lex Frisionum, as one of the Leges Barbarorum recorded in the late eighth century, states that Frisia is the area ranging from the River Sincfal on the present Dutch-Belgian border to the River Weser in Germany. This landscape is analysed and characterized in new, historically interpretative detail, by Gilles de Langen and J. A. (Hans) Mol (Ch. 4) as the setting of Frisian society, trade, power bases and internal districts and variation. By looking back through layers of landscape organization and land-ownership, they confirm a perception of this landscape as a rather free and non-hierarchical context of rural production and exchange. This resonates strongly with the paper by Nelleke IJssennagger-van der Pluijm (Ch. 9) which suggests that the maritime mobility of the Frisians, facilitated by the landscape and through which they were well connected, had the effect of their physical footprint being more scattered and light-touch; consequently particular care and determined focus are required to elucidate its presence and character. Outside of what Lex Frisionum considers the Frisian lands, the North Frisian islands of Föhr, Amrum and Sylt as well as their near mainland were also considered Frisian, as they still are nowadays. The new information in the contribution of Bente Majchczack (Ch. 6) shows how, in the earliest settlement phases of the seventh and eighth centuries, the North Frisian Islands were apparently re-settled and became part of the Frisian maritime trading networks, being strategically located between the North and the Continent, and indeed being able to define how they became part of the Frisian sphere. The question that remains is how far inland the Frisian sphere stretched in this extended region, and this can partly be answered by realizing how inaccessible the further hinterland was along most of the coast apart from along the waterways. 3

Z01 Hines Book B.indb 3

07/05/2021 13:29

Frisians of the Early Middle Ages Lex Frisionum does not, in fact, give us an inland border for Frisia, but we can be fairly certain that the inaccessible mires formed a natural boundary of Frisian society land inwards until they started to be reclaimed in the tenth century. On the map in Figure 4.1 De Langen and Mol have indicated the expected internal boundary based on their spatial analysis, partly based on the earliest parish boundaries. The issues of how far borders were linear, or were transitional zones in ways that could take various practical forms, is especially pertinent to the irrepressible question of whether the justly famous trading wīc of Dorestad, at the location of modernWijk bij Duurstede, was Frisian or Frankish, if such a dichotomy ever existed, or indeed was neutral and related to both polities and their trade. The papers in this volume constantly revert to the challenging questions of what is Frisia and what is Frisian: issues of relevance to a broad range of disciplines, and questions on which light can be shed from a diversity of perspectives. It is impossible to disentangle the Early-medieval Frisians from interactions with surrounding areas – by land on the Continent, and overseas to Scandinavia and Britain – and undesirable, of course, even to try to do such a thing. As a historic label, ‘Frisian’ arose in the writings of Roman authors who described an area in which a people called ‘Frisians’ lived between land and water. Annet Nieuwhof (Ch. 3) shows that in discussing Frisians of the seventh and eighth centuries who essentially are ‘New Frisians’ descending from the fifth-century migrations of other Germanic groups sometimes (especially in the Frisian context) called Anglo-Saxon, it is also essential to discuss the ‘Old Frisians’ that are first mentioned by the classical and later Empire authors from before the fourth-century abandonment of the coastal area. Robert Flierman (Ch. 8) highlights how, after being first documented as one of the barbarian groups on the North Sea edges, ‘Frisians’ also disappear from the textual record only to re-appear in the seventh century, whilst the related Saxons are present in the coastal area in the textual record throughout the period. The historical label that re-appeared in the seventh century was applied to people – the northern continental coastal dwellers, and to a geographical area. Apart from runic inscriptions, we have no Frisian written documents at the time, but from the appearance of the term ‘Frisian’ in writings all around the North Sea we get a sense of a group’s location and their relations. On the Continent at this time, it was first and foremost the Frankish powers who had a particular use for the term Frisian to differentiate between themselves and the barbarian coastal dwellers; such as when at the start of Carolingian times the laws of these people were recorded in the Lex Frisionum code. It is highly apparent that there is a large range of historical sources of relevance to this subject, although the reader will also see that analysis and discussion frequently recur to a core set of texts and documents. Lex Frisionum is perhaps the key historical source that is used in a multitude of contributions and that forms the basis of a lot of our understanding of what is considered ‘Frisian’ and how they lived. This is the case, not only because it provides us with the geographical area of Frisia that we can subsequently investigate with archaeological methods, such as through the material culture collated and highlighted by Egge Knol (Ch. 2), but also because it gives us insights into core elements of lifestyle and sociocultural conditions through a certain lens and at a moment in time. Han Nijdam (Ch. 5) uses this text in the context of later medieval Frisian law texts that can be applied retrospectively to analyse law and political organization in Frisia around AD 800. In the Carolingian period, Frisia underwent conversion to Christianity in a long and complicated process which is scrutinized in 4

Z01 Hines Book B.indb 4

07/05/2021 13:29

An Archaeoethnological Perspective John Hines’s paper (Ch. 11). Here, religio-juridical procedures in Lex Frisionum are studied as regional variation of Germanic pre-Christian religion, and with a distinctive approach of the Church to it, placed in the context of the information various hagiographical Vitae, such as those of Willibrod, Wulfram, Boniface and Liudger, provide us with. Similarly, Ian Wood (Ch. 7) places Lex Frisionum alongside the contemporary Vitae and the various Annales and political historical documents of the Carolingian power sphere, whilst Robert Flierman looks at the same text in relation to the Lex Saxonum. The work of various classical authors, from Ptolemy to Tacitus, forms part of the sources discussed to understand the first historical documentation of the ‘Frisians’ and the development in it. Communication and interaction are the activities which predictably may leave invaluable traces of the activities and values of this population. On the basis of metal finds, Johan Nicolay (Ch. 10) argues that expressed identity shifts over time from a regional Frisian entity related primarily to a North Sea context to a supra-regional Frankish one. In the seventh and eighth centuries particularly, the question of the relation between Frisian and Frankish is a key one, and Ian Wood scrutinizes the relation between these groups on the political stage through the historic sources. Equally important is the question of overseas relations that can also be seen as formative for ‘Frisian’, who are often identified as seafaring. The contribution of Nelleke IJssennagger-van der Pluijm considers the impact of the maritime on society, mindset and imagery more widely following from this. Chapter 12, by Arjen Versloot, shows how those overseas connections created a single speech community on either side of the North Sea in the fifth and sixth centuries, suggesting the range and power of connectivity at this time, which is similarly reflected in the question of Frisian literacy, a runic literacy, as discussed by Tineke Looijenga (Ch. 13). In both studies we encounter ‘Frisian’ as a linguistic term, as one of the North Sea Germanic languages and at some point so closely connected to English that varieties of both language and runic text can be described as Anglo-Frisian. Looijenga, however, also presents a powerful reminder of the fact that it can be seductively easy to look only at a North Sea context that might appear to be capable of explaining all the observable phenomena, while southerly connections with an early Frankish region continue to require attention. The overseas connections are clearly visible also in the material culture that is exchanged or influenced, as highlighted in the finds that Nicolay shows. But they are also visible in further categories of finds through the Early Medieval Period, such as various metal-detected portable objects, antler combs and coins. Clearly, this elongated Frisian sphere was relatively diffuse and heterogeneous, and can be sub-divided geographically with reference to landscape and topography as well as on the basis of legal nuances, dialectal differences, patterns of material culture, documentary evidence and historical references. One distinction made by contemporary writers such as Bede (e.g. HE V.10) is that between Frisia Ulterior, the coastal zone, and Frisia Citerior, the old Roman frontier zone along the rivers that now defined the Frankish frontier and the area of the trading towns and power centres such as Dorestad and Utrecht. Research traditions have created further differentiation between areas, particularly exemplified by the approach to the investigation of the terpen/wierden or dwelling mounds, and metal-detecting. As Egge Knol explains (Ch. 2), the large-scale commercial exploitation of the terps in central Frisia, the current provinces of Friesland 5

Z01 Hines Book B.indb 5

07/05/2021 13:29

Frisians of the Early Middle Ages and Groningen, have provided us with a substantial number of well-preserved finds. Because of the physical conditions within the terps, these include organic material. The terps in the German Frisian areas, where they are called Würten, were not commercially extracted in the same way and therefore often remain more intact, providing us with a different picture. One field that Giorgio Ausenda would have insisted on a paper on which we do not have is ‘family relationships’ and ‘jural relations’: households, however, are well covered, especially by Annet Nieuwhof (Ch. 3). Numismatics is not covered as a topic on its own in this volume, nor are place-names as such, but the subjects of coinage, economy and the patterns of distribution of wealth run through a number of papers, and are particularly carefully addressed in discussions. De Langen and Mol thus combine agrarian and urban settlement in an especially comprehensive chapter that renders a traditional ‘Trade and Towns’ chapter (as was, in fact, initially envisaged when planning the symposium) unneeded. In respect of the monetization of economic life and overseas connections, a traditional topic in the study of Early-medieval Frisia is the sceattas, the silver coinage found as a common type in Frisia, Denmark and England. Sceattas have been studied by numismatists, archaeologists and art historians alike, leading to a range of ideas of depicted symbolism – from pre-Christian to Christian – and mechanisms of distribution (Blackburn 2003; Gannon 2003; Ulmschneider 2000; Theuws 2018). The North Sea distribution of the sceattas certainly fits with the idea of Frisian middlemen in trade. ‘Frisian’ has sometimes even been suggested to be synonymous with trader in the Early-medieval period, and clearly overseas and riverine trade, as well as agricultural productions, was important in the Frisian area. But even if ‘Frisian’ indeed came to mean ‘trader’ in some contexts, the term continued to serve first and foremost as an ethnic and geographical label. Over time, too, what is considered ‘Frisian’ has demonstrably evolved and changed, probably both for Frisians themselves and for those around them. Contemporary ideas of identity and affiliation must also be taken into account, both out of respect for the public, but also because presuppositions can muddy the waters too – and be they rivers, tidal zones or sea, the Frisians lived lives extremely close to and very much regulated by the waterscapes. The choice for the title of this volume, Frisians of the Early Middle Ages rather than ‘The Early-medieval Frisians’, deliberately reflects multiple perspectives on ‘Frisian’ in this period. The ­Early-medieval Frisians is not a group we would wish to assert existed; but there certainly were Frisians of the Early Middle Ages.

The Symposia in Historical Archaeoethnology1 Historians and archaeologists of the Early Middle Ages have in fact long made use of the classics of anthropology to help elucidate their material. What we might call a manifesto for ‘Historical Ethno-archaeology’ was set out in the Introduction to a volume entitled After Empire: Towards an Ethnology of Europe’s Barbarians (Ausenda ed. 1995, 1–13). This began with a short history of the discipline: for Giorgio Ausenda – and his 1

This section is based primarily on a public lecture, ‘History, historiography and anthropology: the case of the Frisians’, given by Ian Wood in Tresoar, Leeuwarden, in collaboration with the Fryske Akademy, on Thursday 13 September 2018.

6

Z01 Hines Book B.indb 6

07/05/2021 13:29

An Archaeoethnological Perspective was a very personal reading – it originated with the publication of the first volume of The Journal of the American Society of Ethnohistory in 1954, which was followed two years later by an article of Maxine Kleindienst and Patty Jo Watson, ‘Action archaeology: the archaeological inventory of a living community’ in Anthropology Tomorrow (5,1: 1956): a short piece which opens with the declaration that the ‘material remains of … dead cultures are worse than useless if they are not subjected to meaningful interpretation and eventual inclusion in the growing body of knowledge of culture, society and human behaviour’ (Kleindienst and Watson 1956, 75). According to Ausenda, a French response to this came in 1976, with Jacques Le Goff ’s seminar on Anthropologie historique de l’Occident médiéval. This French development of the discipline was then taken up in the United States by Marshall Sahlins in Historical Metaphors and Mythical Realities (Association for Social Anthropology in Oceania, Special Publications 1, 1981). In Ausenda’s reading, the concept of Historical Anthropology was further developed by Aletta Biersack in her introduction to Clio in Oceania: Towards a Historical Anthropology, published in 1991 by the Smithsonian Institution Press. Both to understand and to appreciate Ausenda’s approach and his model for the integration of ethnographic anthropology and the historical disciplines (as most broadly conceived), it is useful to outline key points of his biography. He was born in 1925, his father being an Italian engineer, his mother an American of Austrian extraction. He studied electronic engineering at the Politecnico in Milan. After the end of the Second World War he took advantage of his mother’s origins to go to the USA, where he embarked on and gained a degree in electronic engineering at Cornell, returning to Italy and to join his father’s business in 1951. His work took him back to the USA, and also to Europe, Africa, the Middle and Far East and South America. In 1960 he returned to Italy, and while continuing his work as an engineer developed his interests in the humanities and anthropology. In 1970 he wrote a thesis at the Catholic University of Milan on the Roma people – an early indication of his interest in non-sedentary peoples – leading to publications on social structure and culture, and a study of a Roma individual (Ausenda 1974; Levakovich and Ausenda 1976). His first book, on Strutture sociale e contenuti culturali was published simultaneously in Italian and English in 1974. In 1977 his engineering work took him to Cairo, where he became increasingly fascinated by African anthropology. He applied to do a doctorate in Cultural Anthropology at Columbia University in New York in 1980, and the following year gave up his engineering career to study the Hadendowa, a nomadic people of the Sudan and Eritrea, pursuing extensive fieldwork in the region in 1987. The result was a doctoral thesis completed in that year entitled ‘Leisurely Nomads: The Hadendowa (Beja) of the Gash Delta and their Transition to Sedentary Village Life’. In 1990, after he had returned to Italy, he founded CIROSS, which provided a forum for seminars and conferences on his major areas of interest. The first meeting was devoted to War Studies, and resulted in a volume on The Effects of War on Society, which was published in 1992. Other workshops and volumes concerned with modern warfare followed. The first of the meetings dedicated to the ethnology of Europe’s barbarians took place in 1993. This biographical sketch of the years up to 1993 perhaps helps clarify some aspects of the manifesto laid down by Ausenda at the start of After Empire. Ausenda was consciously setting out what he regarded as a new discipline, and it is worth quoting his goal in full (Ausenda ed. 1995, 3): 7

Z01 Hines Book B.indb 7

07/05/2021 13:29

Frisians of the Early Middle Ages to recover knowledge of several aspects of the life style and sociocultural conditions of past populations … This endeavour should be capable of acquiring momentum from an interdisciplinary approach because of the complementarity of the disciplines that belong to the tripartite division of the field between: (1) historical disciplines that rely on documents, i.e. history, philology and literature; (2) those that can shed light on long-forgotten customs and procedures by comparing them with examples from contemporary societies at similar levels of complexity, i.e. ethnology, folklore, and socio-cultural anthropology; and (3) archaeology and physical anthropology, the disciplines whose practitioners recover and interpret the actual remains of the past reality of the populations under study. In the process the practitioners of all the disciplines involved would certainly benefit from the cross-fertilisation and feedback inherent in such interdisciplinary encounters.

To pursue this end, Ausenda gathered a group of ten interested scholars, including archaeologists, historians and a linguist and an anthropologist with a distinguished record of field research, to discuss the ethnology of Europe’s Late Roman and Early-medieval barbarians, in San Marino in 1993. The range of disciplines represented was important – specialists could check that their own discipline was not misused or misunderstood. Ausenda’s training as an engineer probably contributed to his approach to the organization of a seminar or workshop as a highly regimented exercise. The order in which themes were dealt with remained a constant in all the workshops on Early-medieval peoples that he organized, passing from social structure and the economy to politics and religion. This was derived from the school of American anthropology in which he had been trained. He saw it as the proper scientific way to organize the study of an early people. This approach, established by Franz Boas (1858–1942), Professor of Anthropology at Columbia University, where Ausenda took his doctorate, from 1899, also encouraged the application of anthropology to ancient peoples. Given the circumstances in which an indigenous American archaeology could develop in the later nineteenth to earlier twentieth centuries – and the presumption that Archaeology had a useful role to play primarily in undocumented, pre-historic contexts – it is no surprise that Archaeology is typically housed as a sub-division of Anthropology in US universities. Archaeology and anthropology became standard enough a pairing to become the subject of a joint degree programme at Cambridge (England) by the 1960s. Ausenda was in fact primarily a social anthropologist in his own field of research specialization, and in his perception of the informative order of priorities in understanding both present and past cultures. He had a particular interest in the practicalities of relationships between and amongst the individual members of societies on a community basis. His own researches on the Roma in Europe and the Hadendowa in Sudan reinforced his conviction that living societies and observable contemporary human behaviour in given situations could cast a powerful explanatory light on to past contexts and evidence. In contributions to volumes in the series of Studies in Historical Archaeoethnology, he applied insights into segmentary lineage (the segments being families and kindreds) to unravel the meaning of the term fara that appears in sources on the Langobards and so to describe a key unit in the structure of that people (Ausenda 1995), while in relation to the Visigoths, Baiuvarii and Thuringians he analysed the evidence for their kinship and marriage systems from the point of view 8

Z01 Hines Book B.indb 8

07/05/2021 13:29

An Archaeoethnological Perspective of modern anthropology (Ausenda 1999; 2014 [the latter published posthumously]). The programmes for the symposia organized by Ausenda consequently always began with social structure, and passed through the economy and relations between the cities and the countryside to end with the ‘superstructure’ of politics and religion. Our own approach is markedly different in starting with landscape and resources, then, although the omission of commissioned contributions specifically on kinship and relationships may, as noted above, give a misleading impression of the extent to which we differ from the ‘San Marino model’ in range. Ausenda found, to his frustration, that it was a challenge to interest specialists in anthropology in applying their expertise to past societies; from the experience of discussing issues with him in the original symposia, it is probably realistic to postulate that most of such anthropologists were more reticent to tackle the problems of evaluating and interpreting particularly old historical and archaeological records than he was himself. It is in fact by no means difficult to outline a creditable history of more a than a century’s willingness within the fields of Early Medieval cultural history generally to draw upon and apply insights gained in contemporary anthropological and ethnological studies. Bertha Philpotts’s groundbreaking study, Kindred and Clan in the Middle Ages and After, was published by the Cambridge University Press as early as 1913; her own subsequent research, alongside a glittering academic career at Westfield College, London, and Girton College, Cambridge, was primarily in Old Norse literature, and her first book has been used primarily in relation to Scandinavian and Icelandic studies. Kinship, however, was firmly established as an issue for historians of the period in various areas and linguistic families. Jack Goody (1983) dealt directly with Early-medieval kinship, drawing upon an extensive argument in Past and Present in the years 1968–70 over agnatic and cognatic kinship between Karl Leyser and Donald Bullough, which itself was a response to Georges Duby’s study of the county of Mâcon of 1953. Comparably, a study of feuding and pacification in Africa (Gluckmann 1955), has been treated as providing a key to the preservation of order in Anglo-Saxon society since the 1950s. In the field of economic life and relationships , meanwhile, Malinowski’s account of the Trobriand Islanders, Argonauts of the Western Pacific, appeared in the same year as Radcliffe-Brown’s study of The Andaman Islanders (Malinowski 1922; Radcliffe-Brown 1922). Soon supported by Marcel Mauss’s study of gift-exchange and reciprocity (1925), these foundational works have provided an important interpretative model for the understanding of the Early-medieval economy. The concepts thus presented of the circulation and distribution of goods according to order and authority within a community became central, to, amongst others, Philip Grierson in reading of the Early-medieval economy, which set about trying to define exchange in non-monetary terms (Grierson 1959). This line of approach was pursued in further influential studies of the 1970s and 1980s (Sawyer 1977; Hodges 1982). At that same time, Mary Douglas’s book Purity and Danger of 1966 proved as influential in the field of Early-medieval religious history, particularly in studies of saints. It is striking that it was Evans-Pritchard’s (1940) work on the Nuer that particularly attracted Ausenda’s interest and explicit praise rather than his slightly earlier study of Witchcraft, Oracles and Magic among the Azande (1937), which had become a major point of reference for students of Early-medieval religion. The consequence of an approach to the range of subjects that placed social issues first and such ideological and cognitive issues at the other end of things was in fact to turn the spotlight away from a 9

Z01 Hines Book B.indb 9

07/05/2021 13:29

Frisians of the Early Middle Ages field of scholarship in which anthropology may in fact have had its greatest influence to date in Early Medieval studies. We also know, of course, that anthropologists can be misled and misinterpret what they believe they can see in field data as much as archaeologists, historians and philologists can with their source data. If anything, the present volume may be held to vindicate the position that it does not really matter in which order one approaches the topics, provided that the context of analysis is as broad in terms of disciplinary expertise and attention to the whole range of cultural history as it can be. Ausenda’s recurrent insistence to the members of his workshops that they should approach these peoples as would an anthropologist of a particular school was undoubtedly influential and productive, and he was never faced with a body of scholars who simply accepted any particular approach without question. Anthropological descriptions are not blueprints that can be transferred directly as ‘models’ on to early societies. But the questions asked of comparable phenomena can assist in the study of the evidence of the Early Middle Ages to identify issues and formulate relevant question. And anthropologists can also learn historians, and not least research on the Early Middle Ages such as that collated in this volume, drawing on and integrating the full range of available source material.

Conclusions The symposium in Leeuwarden on the various ‘Frisian’ communities and regions took place almost exactly quarter of a century after the pioneering ‘After Empire’ meeting in San Marino of 1993. Although, as the review and critique above make abundantly clear, the idea of such focused, multi- and inter-discliplinary study of a particular population, area and period was itself nothing new, the practical opportunity to attempt that provided by the CIROSS symposia was something special, and it is as much a tribute to Ausenda’s initiative that intellectually and methodologically far less reflection on the why and how seemed necessary at all in 2018. No less a tribute to the special achievement of the ‘San Marino’ symposia, however, was the appreciation and excitement of the participants in Leeuwarden at such a rare opportunity to discuss, in real dialogue, and with time to do so, various ideas and points of debate with colleagues. That is no less rare and valuable an experience when it can be brought about. Crucial aspects of the current volume include, as noted above, its chronological and geographical range, as well as its focus on a historically crucial period, which seems in some respects ‘nearly formative’ in the course of development of a Frisian identity and self-understanding – and where it just might be the inchoate nature of the development of a Frisian polity in the eighth century that has preserved invaluable clues for us, even if still detected through considerably later sources. Without doubt, whether we conceive of them as a ‘topic’ or as a population, the Frisians provide a valuable case study in respect of the general development of Europe in the Early Middle Ages. State formation was a key process, but Frisia’s decentralized internal structure and position in relation to the major powers of continental Europe may be what left us with the anomalous ‘Frisian Freedom’ of the High and Late Middle Ages. It is fundamental to an archaeoethnology that the volume offers a keen focus on practical questions, and so in this case starts plainly with material resources and how they were used; this subject, however, comprises not only subsistence and other economic matters but also social 10

Z01 Hines Book B.indb 10

07/05/2021 13:29

An Archaeoethnological Perspective arrangements and communication. Both secular and religious ideologies can then be correlated with these concerns. It is fitting, finally, that the concluding Discussion (Ch. 14) specifically addressed the challenges of public engagement with this sort of scholarly research and debate. As noted, the event was deliberately designed to correlate with Leeuwarden’s programme as European City of Culture in 2018. It had been hoped that modern technology would allow us to modify the inevitably ‘closed’ nature of a small-group symposium – so productive for those privileged to take part, but at the unfortunate cost of appearing exclusive – by broadcasting the proceedings live on the web, and so inviting contributions in the form of comments and questions that could be taken up in the final discussion. That proved technically too ambitious, alas. But the public lecture proved a valuable exercise in outreach, and it is hoped too that this publication will offer inspiration and ideas to the credit of the late Giorgio Ausenda, in the respectful and affectionate memory of whom it has been produced.

References Ausenda, Giorgio 1974, Strutture sociale e contenuti culturali (Milan). — ed. 1995, After Empire: Towards an Ethnology of Europe’s Barbarians. Studies in Historical Archaeoethnology (Woodbridge). — 1995, ‘The segmentary lineage in contemporary anthropology and among the Langobards’. In After Empire: Towards an Ethnology of Europe’s Barbarians, ed. G. Ausenda, Studies in Historical Archaeoethnology (Woodbridge), 15–45. — 1999, ‘Kinship and marriage among the Visigoths’. In The Visigoths from the Migration Period to the Seventh Century: An Ethnographic Perspective, ed. P. Heather, Studies in Historical Archaeoethnology (Woodbridge), 129–90. — 2014, ‘Kinship and marriage among the Baiuvarii and Thuringi’. In The Baiuvarii and Thuringi: An Ethnographic Perspective, eds. J. Fries-Knoblach, H. Steuer and J. Hines, Studies in Historical Archaeoethnology (Woodbridge), 103–10. Blackburn, Mark 2003, ‘“Productive sites” and the pattern of coin loss in England, 600–1180’. In Markets in Early Medieval Europe: Trading and ‘Productive’ Sites, eds. T. Pestell and K. Ulmschneider (Macclesfield), 20–36. Douglas, Mary 1966, Purity and Danger: An Analysis of Concepts of Pollution and Taboo (London). Duby, George 1953, La société aux XIe et XIIe siècles dans la région mâconnaise (Paris). Evans-Pritchard, Edward E. 1937, Witchcraft, Oracles and Magic among the Azande (Oxford). — 1940, The Nuer: A Description of the Modes and of Livelihood and Political Institutions of a Nilotic People (Oxford). Gannon, Anna 2003, The Iconography of Anglo-Saxon Coinage: Sixth to Eighth Centuries (Oxford). Gluckman, Max 1955, Custom and Conflict in Africa (Glencoe, IL). Goody, Jack 1983, The Development of the Family and Marriage in Europe (Cambridge). Grierson, Philip 1959, ‘Commerce in the Dark Ages: a critique of the evidence’, Transactions of the Royal Historical Society 9, 124–40. Hines, John and Nelleke IJssennagger eds. 2017, Frisians and their North Sea Neighbours: From the Fifth Century to the Viking Age (Woodbridge).

11

Z01 Hines Book B.indb 11

07/05/2021 13:29

Frisians of the Early Middle Ages Hodges, Richard 1982, Dark Age Economics: The Origins of Towns and Trade A.D. 600–1000 (London). Kleindienst, Maxine R. and Patty Jo Watson 1956, ‘“Action archaeology”: the archaeological inventory of a living community’, Anthropology Tomorrow 5, part 1, 75–8. Levakovich, Giuseppe and Giorgio Ausenda 1976, Tzigari: vita di una nomade (Milan). Malinowski, Bronisław 1922, Argonauts of the Western Pacific: An Account of Native Enterprise and Adventure in the Archipelagoes of Melanesian New Guinea (London). Mauss, Marcel 1925, ‘Essai sur le don: Forme et raison de l’échange dans les sociétés archaïques’, L’Année sociologique 1925. Philpotts, Bertha 1913, Kindred and Clan in the Middle Ages and After (Cambridge). Radcliffe-Brown, Alfred R. 1922, The Andaman Islanders (Cambridge). Sawyer, Peter 1977, ‘Kings and merchants’. In Early Medieval Kingship, eds. P. Sawyer and I. Wood (Leeds), 139–58. Theuws, Frans 2018, ‘Reversed directions: re-thinking sceattas in the Netherlands and England’, Zeitschrift für Archäologie des Mittelalters 46, 27–84. Ulmschneider, Katharina 2000, Markets, Minsters, and Metal-Detectors: The Archaeology of Middle Saxon Lincolnshire and Hampshire Compared, BAR British Series 307 (Oxford).

12

Z01 Hines Book B.indb 12

07/05/2021 13:29

• 2 • FOR DAILY USE AND SPECIAL MOMENTS: MATERIAL CULTURE IN FRISIA, AD 400–1000 Egge Knol

T

he salt marshes of the Wadden Sea, silted up to well above sea level, are very fertile and rich in protein-rich grasses. There was ample scope for the inhabitants to provide themselves with food. It was a good place to live, on artificially raised mounds known as terpen or wierden. In the Early Middle Ages, this resulted in a high population density. By the eighth and ninth centuries, the salt marshes, covering an area of about 2,000 square kilometres in the provinces of Groningen and Friesland, supported a population of at least 40,000 inhabitants, and probably a great deal more (Knol 2019a). It was the good farmland that allowed this high density. The settlements – in Carolingian times at least a thousand of them – were continuously occupied or frequently re-occupied, which meant that the sites became accumulations of settlement debris and artefacts. The lime-rich clay, and deposits of farm manure, provided good conditions for the preservation of organic remains. In short, the coastal zone is a rich source for archaeology. Unfortunately, the circumstances in which many finds were recovered – during commercial quarrying of the fertile soil – mean that contextual evidence tends to be limited. Lex Frisionum, written at the end of the eighth century at the order of Charlemagne, governing the territory of the Frisians, indicates that this territory extended well beyond the current provinces of Groningen and Friesland. It ran from the River Zwin (formerly the Sincfal – on the Belgian border) up to the Weser estuary in northern Germany. In the western Netherlands, the Frisians mostly occupied the coastal dunes; the marshy interior long remained of little significance. The dune zone was cut through by the great rivers Scheldt, Meuse and Rhine. Trading settlements arose around their estuaries (Dijkstra 2011). In these parts there were no artificially raised settlement mounds. Here the archaeological record is known mostly from archaeological excavations. The German coast between the rivers Ems and Weser does have some salt marsh between the Wadden Sea and the higher Pleistocene interior, but there is less of it, as in many parts the Pleistocene geology comes quite close to the sea. There are barely any native written sources from the Frisians. Indeed they are limited to some sixteen runic inscriptions from Frisia itself, and a few more found elsewhere (Looijenga 2003; this vol.). For evidence from Frisia, the researcher must resort to the material remains. These do allow far-reaching inferences, as has been proved by Nieuwhof studying the pottery, and as Nicolay has shown with reference to mostly copper-alloy brooches (see both, this vol.). My story will first focus on the conditions of 13

Z01 Hines Book B.indb 13

07/05/2021 13:29

Frisians of the Early Middle Ages conservation and the context of many finds. It should be remembered that, even though much has been preserved, an unmeasurable amount has been lost. I shall then discuss some groups of finds, to demonstrate the wide variety of material remains. Of course it will not be possible in this summary to discuss all aspects of the material culture.

Preservation and context The conditions for conservation and preservation in the dwelling mounds are comparatively good. Clay and dung shut out oxygen. This and the lime-richness of the recent marine deposits have ensured first-rate preservation of bone and antler. Skeletal parts of animals, but also humans, and of course artefacts of bone and antler, are perfectly preserved. This allows us to study the fauna in and around the dwelling mounds. There are fine examples of bone combs and many other implements. The dung has preserved fragments of woven fabric and, in some cases, leather. Articles of wood have survived on a more modest scale. Foundations of wooden structures have also been preserved. Of course any stone and fired clay is still present. Bronze artefacts are well preserved, as, naturally, are gold and silver. Iron, however, is often in a poorer state of preservation. However, a drawback of these continuously inhabited sites is that the stratigraphy of successive occupation layers is difficult to unravel. Digging would cause older items to end up in younger layers. People living in the saline salt marsh were confronted with the problem of fresh-water supply; so through the ages they have dug many wells,

Fig. 2.1  Part of the mound of Garnwerd, Groningen, being quarried. Tipping wagons carried the soil on a narrow-gauge railway to the nearby River Reitdiep. From there it was transported on small sailing barges to farms in regions with poor soils. Antiquities would occasionally be retrieved in the course of such quarrying. These usually were the intact and more valuable items. Groninger Archieven: NL-GnGRA_1173_61_87.

14

Z01 Hines Book B.indb 14

07/05/2021 13:29

Material Culture in Frisia which cut right through the stratigraphy. In short, excavating a dwelling mound means being faced with a confusion of features. This often makes it hard to date particular features and the associated finds. So far, I have been referring to excavated settlements. However, the majority of these dwelling mounds have not been excavated, but were quarried. The soil from these man-made mounds, where people and their livestock had lived for at least two thousand years, was very fertile. In the 1840s, it was discovered that this soil could be used successfully as a fertiliser for poor soils. This sparked a large-scale quarrying industry. Hundreds of dwelling mounds were dug away and their soil transported to other regions (Fig. 2.1). This practice was banned in 1942. Quite frequently, the manual quarrying would reveal antiquities that drew the attention of interested scholars, such as doctors and clergymen. This was the start of the collecting of antiquities in the northern Netherlands. In Friesland (in 1827) and Groningen (in 1874) formal collections were set up. The mounds even received international attention. Both the Italian antiquarian Pigorini and his Scottish colleague Dr Robert Munro visited the sites and took home a number of items for their own museums (Jensma and Knol 2005). In those days of quarrying, the emphasis lay on intact items, such as complete pots, fine combs and of course treasure of gold and silver. Large numbers of remarkable finds were recovered, but their context was poorly documented, if at all. Yet these objects do offer a distinct impression of life on the mounds. Detectorists’ finds of the past thirty years contribute little more in terms of context, but still are valuable additions to the archaeological record: for example, the finds of precious metals that have allowed Nicolay’s survey of gold and silver ornaments (Nicolay 2014, tabs. 4.1, 4.3, 4.4 and 4.5). Eleven of the twelve hoards containing gold are old quarrying finds. A thirteenth hoard can added to this list, comprising at least two gold solidi from the quarrying at Ezinge (Knol 2014, 190). Among the forty-one gold pendants, twenty-four are quarrying finds, two come from excavations and fifteen are detector finds. Of twenty-three gold and silver brooches, one is a quarrying find and the others are detector finds. Eleven of the fifteen gold and silver rings are detector finds. Though these are mostly small items and fragments that build up the record, they significantly clarify and differentiate the picture. On the whole it is known from which mound the early finds derive, but any further contextual information is lacking. Their dating then depends on similar, but better-documented items found elsewhere. A survey in 1993 listed twenty-nine dwelling mounds in Groningen and Friesland with well-documented occupation traces of the Early Middle Ages. However, Early-medieval material was recorded from 499 mounds (Knol 1993, 147–50, tab. 6). Around these demonstrably inhabited mounds lay many more in similar geographical positions. Without doubt, these must also have been inhabited. Modern surveys, detector finds and excavations repeatedly confirm this. A number of modern excavations have uncovered distinct Early-medieval settlement traces, for instance at Leeuwarden-Oldehove, Anjum, Hallum, Midlum and Ulrum (Dijkstra and Nicolay 2008; Nicolay 2010; Tuinstra et al. 2011; Hielkema 2017; Groenendijk 2006). Other excavations have produced Early-medieval finds, but lack clear settlement features other than pits or ditches. It is important to realise that all early or modern excavations have uncovered only small areas of any mound, sometimes no more than a small trench along a section exposed by quarrying. Always it has been just a small sector that is investigated; this even applies to the famous excavations of Ezinge or Wijnaldum. In brief, nowhere do we have a full overview of a settlement. The dating 15

Z01 Hines Book B.indb 15

07/05/2021 13:29

Frisians of the Early Middle Ages of many objects depends on parallels found elsewhere in dated contexts. Categories studied typochronologically are especially pottery, metal ornaments, coins and combs; these allow stray finds to be dated. Other categories have been less intensively studied, or lack dated contexts, or have been used for long periods without typological change. These remain difficult to date. Along the Wadden Sea coast of northern Germany, the dwelling mounds were quarried to a much lesser extent. Most knowledge about this area derives from archaeological excavations carried out from the 1930s onwards (Behre and Schmid 1998; Bärenfänger 1999). These are investigations resulting from scholarly research goals, by the Niedersächsisches Institut für historische Küstenforschung at Wilhelmshaven, as well as by numerous rescue excavations by the Ostfriesische Landschaft and others. These excavations have produced a great deal of knowledge, but in comparison to Friesland and Groningen there are many mounds about which nothing at all is known. The evidence relating to the sixth century AD is so limited that depopulation has been suggested (Schmid 1995). However, findings from various recent excavations point to some continuity of habitation (Bärenfänger 2002, 288–90; Nüsse 2004a; 2004b). As mentioned already, the coast of the western Netherlands is of a different nature. Settlement was concentrated in the zone of sand dunes, the northern part of which was eroded, and also on the Pleistocene islands of Texel and Wieringen, and of course around the estuaries of the rivers Rhine, Meuse and Scheldt. Around the Rhine estuary, excavations at Katwijk, Rijnsburg and Oegstgeest have yielded a great deal of evidence (Van der Velde 2008; Dijkstra 2011). For the study of material culture it is very important to publish material studies and excavation reports. Miedema made a systematic inventory of all finds known from two areas in the dwelling-mound region of Groningen (Miedema 1983; 1989; 1990; 2000). The finds from the excavations of Ezinge, Leens and Godlinze have also been published (Van Giffen 1920; 1940; Knol and Bardet 1999; Nieuwhof 2014). In Friesland, apart from the earlier-mentioned excavations, the evidence from the cemeteries of Oosterbeintum, Hogebeintum, Beetgum and Midlum is available (Knol et al. 1996; Knol 2019b; Knol et al. 2019; Knol 2008a; Hielkema 2017). For Ostfriesland we have publications on excavations, at settlements such as Oldorf, but also the cemetery of Dunum (Schmid 1969; 1970a; 1970b; 1972; Tidow and Schmid 1979; Schmid 1994). A compilation of older material was published by Schwarz (1990; 2016). For the coast of the western Netherlands, there is the survey by Dijkstra (2011).

Garments In the dung deposits of the dwelling mounds, woollen fabrics are well preserved (Fig. 2.2; Schlabow 1953; 1974; Brandenburgh 2012; Groenendijk 2006). Linen, however, will rapidly decay. From analysis of seeds, it has become clear that flax was grown, presumably both for the oily seeds and for the textile fibres. The processing of flax into linen leaves few specific features or utensils: there may be bundles of flax stems in shallow retting pits, or smoothed ribs used as swinglebars for cleaning the scutched fibres, as well as some combs. The spinning of flax required spindle whorls, but these were also used for wool (Nieuwhof 2017). The salt marsh was highly suitable for sheep farming. Even well after dikes were constructed, flocks of sheep would graze the unendiked marsh beyond. Wool was a major 16

Z01 Hines Book B.indb 16

07/05/2021 13:29

Material Culture in Frisia

Fig. 2.2  Textile fragment from Rasquert in Groningen. Groninger Museum (photograph: Marten de Leeuw).

product; every household must have been involved in wool processing and weaving. This was very much a woman’s work. Spinning whorls point to spinning; many of them have been found, made of pottery, antler or bone. Mostly the threads were 0.5–1.5 millimetres thick, but thinner or thicker yarn was made as well (Brandenburgh 2010, tab. 6a). 17

Z01 Hines Book B.indb 17

07/05/2021 13:29

Frisians of the Early Middle Ages No loom has been recovered, but there are plenty of loomweights. Weaving battens for beating down the weft were made of wood or bone. Presumably almost every household in the coastal zone had its loom. The importance of woven fabric from these parts is evident also from the tenancy records of the monastery of Werden and Fulda (Kötschke 1906, 96, 110–13; Peek and Siegmüller 2007). After Christianization, much farmland, or rather its produce, was given to monastic institutions. Some of the rent had to be paid in woollen mantles (pallia). These evidently were very desirable, but also constitute a form of payment easy to collect and to transport to the monasteries situated far into Germany (Tidow and Schmid 1979; Peek and Siegmüller 2007). The Frisian pallia were quite famous at the time. The wool was probably from a special breed of long-haired sheep which were washed before shearing. A sheep-washing set-up in an artificial pond (fething) was discovered in the dwelling mound of Hessens near Wilhelmshaven in Germany. The pallia were not particularly finely woven cloths, but were especially strong and water-resistant (Peek and Siegmüller 2007). Brandenburgh has studied over four hundred textile-fragments from the dwelling mounds of Friesland and Groningen and Peek and Siegmüller those from dwelling mounds in adjacent North Germany (Brandenburgh 2010; 2016; Peek and Siegmüller 2007; 2018a; 2018b). These provide a wealth of evidence about the range of weaving techniques. Besides a lot of fairly coarse weaves, some very refined, beautiful textiles

Fig. 2.3  Leather shoe from Hallum-Jousemaburen, dated to the fifth century AD. Fries Museum. After Van Driel and Van der Plicht 2016.

18

Z01 Hines Book B.indb 18

07/05/2021 13:29

Material Culture in Frisia have been found. The recovered textiles were often fragments that seem to have ended up in the dung heap as cleaning rags. Unfortunately, they therefore seldom allow reconstructions of garments. In two instances a mitten was found. The six recovered woollen caps are all somewhat different; some were decorated with stitching in contrasting colours. One is believed to have been worn by a woman (Zimmermann 2010; Brandenburgh 2012; 2016). A small but notable category of finds is that of some distinctive leather shoes from the Oostergo district of Fryslân (Fig. 2.3). These have been dated to the fifth and early sixth centuries. The clear-cut dating and limited distribution suggest that these shoes, like the distinctive pottery and brooches, were associated with the typical Anglo-Saxon lifestyle (Van Driel and Van der Plicht 2016). The shoes, given their size, seem to have been worn especially by men. There is simpler footwear too. Presumably many mound dwellers, especially women and children, often went barefoot. Perhaps in wintertime woollen footwraps were used or clogs. On the Halligen of Schleswig-Holstein, where people still live on mounds open to the sea, the custom of going barefoot survived into the twentieth century. Leather shoes were for formal occasions only. No footwear is known from Early-medieval cemeteries in Frisia. In peat, textiles will also be well preserved, which means that a bog body may be a good source of evidence. Much peat near the coastal zone has now been lost as a result of geographical changes and human intervention before there was much professional attention to bog bodies. Fortunately, just across the border, at Bernuthsfeld near Tannenhausen (Kreis Aurich), a male body has been found, dating from the eighth century AD. He wore a knee-length tunic with sleeves, composed of fifty-three patches of different fabrics, and a shoulder cape. Three-metre long puttees clothed his legs, and his feet were wrapped in triangular footwraps. This type of leg covering is quite obsolete nowadays, but until the 1940s was still part of Dutch army uniform. A leather belt and a leather knife sheath were also found on the body (Farke 1998; Bär 2014; Lehmann et al. 2018; Jahn 2019). As just one Early-medieval bog body is known, it is still uncertain how conventionally the man was dressed. The patchwork tunic seems especially curious. Bog bodies are nowadays regarded as sacrifices, which might imply a ceremonial garment. On the other hand, the outfit does seem quite practical.

Personal adornment This category comprises jewellery, but also toiletries such as combs and tweezers. We may assume that they served to express personal identity. The copper-alloy brooches and antler combs – with help from parallels found elsewhere – are particularly good for establishing a typochronological framework. In inhumation graves, jewellery is found more or less in situ, affording us a glimpse of people’s style of dress. At Oosterbeintum and Hogebeintum, pairs of fifth- and early sixth-century brooches of the small long type were found at the collarbones, with a larger cruciform brooch below the chin. The former may have fastened an undergarment, and the latter a cloak. At Oosterbeintum indeed some – not yet studied – traces of fabric were spotted in the corrosion. At Hogebeintum one grave contained two cruciform brooches of different ages (Bos and Brouwer 2005; Knol 2011, fig. 2). The older is likely to have been an heirloom. On the chest beads were often found, numbering from just a few to several dozen (Knol et al. 2019). Presumably the beads were attached in a way so as to display them to best effect. 19

Z01 Hines Book B.indb 19

07/05/2021 13:29

Frisians of the Early Middle Ages Especially in sixth- and seventh-century Friesland, annular brooches were worn, which are extremely rare towards the interior of the continent. At the same time, there also were the typical Domburg brooches, typical of Friesland. Graves of the seventh and eighth centuries often have just strings of beads, and brooches apparently became less important. Christian symbols (crosses or saints) appear on many brooches of the ninth and tenth centuries. These are mainly detector finds and do not appear in graves. They identified the wearer as a Christian. We do not know how intensively Christianity was actually practised in these parts.

Fig. 2.4  A semi-manufactured (unfinished) comb; the tooth plates have not yet been cut. From Wirdum, Groningen. Groninger Museum. After Miedema 2000, fig. 132.

Large numbers of combs, sometimes still with their cases, have been recovered from the mounds (Roes 1963; Tempel 1969). Almost always they were made of antler. Deer did visit the salt marshes, but mostly lived in the interior. Antlers may have been gathered by the mound dwellers, or were obtained from the interior through barter. As deer cast their antlers every year, antler will not have been a rare commodity. Semi-manufactured combs are known too (Fig. 2.4); the combs were therefore manufactured in the dwelling-mound settlements. They are often decorated with dot-in-circle motifs. The combs of the fifth and sixth centuries fit in well with the range of forms known from Anglo-Saxon England. These include the so-called barred zoomorphic combs. Like Anglo-Saxon style pottery, and the small long brooches and annular brooches mentioned before, these combs point to overseas contact with England (Knol 1993; Hills 1996). Personal ornaments also include pendants that can be interpreted as amulets. They are slices of antler, as well as obelisk-shaped pendants which may be Thor amulets. Then we have teeth of boar, dogs and bears. Tweezers are usually regarded as beard trimmers, used for extracting hairs, but may also have served to extract splinters.

Domestic articles Wood was of course an important material for all sorts of artefacts including domestic articles, but also tools. Wood may be well preserved in the dwelling mounds, but after long use usually will have ended up in the kitchen fire. A major inventory of wooden artefacts recovered in modern excavations between 1997 and 2017 comprised forty-seven items from the northern Netherlands, including fragments that were hard to 20

Z01 Hines Book B.indb 20

07/05/2021 13:29

Fig. 2.5  A wooden stool found in grave 44 in the Early-medieval cemetery of Hogebeintum. After Knol et al. 2019b.

Fig. 2.6  Two bowls of alder wood from two Early-medieval inhumation graves at Hogebeintum. Left: grave 23/27; right: grave 44. After Knol et al. 2019.

Z01 Hines Book B.indb 21

07/05/2021 13:29

Frisians of the Early Middle Ages interpret (Lange 2017, 30). Earlier excavations also yielded wooden items on a regular basis, but the scope for conserving them was rather limited. There is very little evidence about furniture. In the Netherlands it is limited to one small, undecorated stool from a grave at Hogebeintum (Fig. 2.5; Knol et al. 2019: grave 44). Graves at Fallward, however, produced several benches and even a chair decorated with chip-carving, made from a hollowed tree trunk (Schön 1995; 1999; 2015). This was an exceptional and fortuitous find; but for the Dutch coastal zone too, such items should not be ruled out. The Fallward furniture shows that wood-turning on a lathe was competently done. The technique of coopering, making small and large barrels, was also known in the Early Middle Ages. In the northern Netherlands we know of many wells lined with stacked-up barrels. Some of these will have entered this region as imports, containing wine, but coopering was presumably practised on the dwelling mounds as well. At Sneek remains were found of a treen box made from curved, thin slices of wood. By lashing the overlaps together an oval box was created, which was given a thin wooden base and lid. In Scandinavia such boxes have never gone out of use and in recent centuries they were brought home and used by Dutch sailors. In two graves at Hogebeintum, lathe-turned wooden bowls were uncovered. Unfortunately these have been poorly conserved (Fig. 2.6; Knol et al. 2019: graves 23–27 and 44). Some trough-shaped wooden dishes have fared better. A fine example from the fifth century was found in the Benserwatt salt marsh, north of Esens. The trough contained the burial of a young child (Fig. 2.7; Bärenfänger et al. 1997). In a riverside house on the Hoge der Aa in Groningen town centre, the side plank of a ninth-century bed was recovered (Lange 2017, 118). Small fragments these may all be, but they all point to effective woodworking skills. Trees did not grow in the marsh area, so wood was imported from the hinterland. Perhaps some furniture arrived complete or half-finished. Of course domestic crockery is a far more familiar subject. The dwelling mounds have produced a wealth of potsherds, which, in part, have been intensively analysed. Plenty of vessels were even complete. As a result, much has been learnt about the stylistic development of the native, hand-made pottery and about the imported ceramics. Much of the pottery is in the Anglo-Saxon style, which may point to immigrants from the Elbe-Weser area, but in the nearby interior of Drenthe this style is found to have been adopted by the local population as well. Maybe we should think of polycausal population increase, a growing remnant population, immigrants from Drenthe, and especially in Friesland, immigrants from the Elbe-Weser region. Annet Nieuwhof ’s contribution (this vol.) reviews this question in detail. It is not just a marked similarity in the pottery, but also in the bone and antler working, and among the metal items the typical brooches linking northern Germany and eastern England, that offer evidence of such a migration. From the late fifth century, the pottery becomes simpler and more sparsely decorated. In the course of the eighth century this evolves into the Kugeltopf, the globular pot, in a range of sizes. The crockery served many purposes: as kitchen ware, storage vessels, water pots; and might eventually be reused as cremation urns. The urns seem to have been simply taken off the kitchen shelf, but possibly there was a personal link between the deceased and the vessel, untraceable to us. Both large and small pots were used as urns. The cremation grave with the golden and silver mountings of an annular brooch held the remains in a simple, undecorated urn (Knol 1993, 214 ; Knol et al. 2019, grave 112). 22

Z01 Hines Book B.indb 22

07/05/2021 13:29

Z01 Hines Book B.indb 23

07/05/2021 13:29

Fig.  2.7 A wooden trough of the fifth century AD re-used as a coffin for a young child, excavated outside the dike at the Benserwatt. Exhibited in the exhibition ‘Het verdronken land is vruchtbaar’ [The drowned land is fertile] at the Groninger Museum in 2013 (photograph: Marten de Leeuw). Normally on display at Museum Mensch und Meer, Essens, Germany.

Frisians of the Early Middle Ages Most of the hand-made pottery was produced on or close to the dwelling mound. Besides this home-made ware, pottery was imported from the Rhineland. The Rhineland produced ceramics on an industrial scale. Remarkably, this wheel-thrown pottery in the sixth and seventh centuries occurred mainly along the Dutch west coast and in the Westergo district of Frisia and far less in the district of Oostergo; in Groningen it is rare, and in German Ostfriesland it is virtually absent (Knol 1993, 192; Kaspers and Sibma 2017). Apparently the ships carrying pottery did not venture much beyond Westergo, or its distribution was controlled to such an extent that its further exportation was hampered. However, by Carolingian times imported pottery was well represented in the eastern part of Frisia as well.

Tools The mound dwellers of course employed numerous kinds of tool, but very few of these have survived. Worn-out wooden tools were probably burnt as fuel, and iron melted down and reused. Some undated wooden spades are the only evidence of digging tools. The 1,500 dwelling mounds and innumerable ditches must of course have been made with spades. Here we see a parallel with the famous Dutch wooden shoe, the clog. Clogs are virtually never recovered in excavations of Early-modern sites. A large survey of wooden items from modern excavations of 1997 to 2017 lists just six wooden shoes – one Early-medieval, five Early-modern (Lange 2017, 33). A worn-out clog would, and in my experience still does, end up in the kitchen stove.

Fig. 2.8  Two beams of a harrow, re-used for reinforcement at the base of a well at Hattersum, northern Germany. After Bärenfänger 1993, fig. 4.

24

Z01 Hines Book B.indb 24

07/05/2021 13:29

Material Culture in Frisia The opportunities for arable farming in the coastal zone have long been underrated by researchers, but by now it is clear that this was practised on a considerable scale. Ploughmarks were found at the base of the mound of Oosterbeintum. This means that in the earliest phase of settlement on the salt marsh – these traces date from 500 BC – arable farming was practised here (De Langen et al. 2016). No actual ploughs have turned up in the coastal zone, though they have elsewhere in the country (Lange 2017, 128–30). At Hattersum near Wittmund (Ostfriesland) some beams belonging to a harrow were found to have been reused in the base of a well: it is dated to the first half of the ninth century (Fig. 2.8; Bärefänger 1993). From Ezinge we have a yoke that has been dated c. 200 BC (Van der Laan 2016). Certain antler fragments have been related to breaking up the soil; they are labelled as hoes or grubber cultivators. Several specimens were found at Ezinge, dating from the Late Roman or the Migration Period (Prummel et al., 2016, 227; Miedema 1983, 243, fig. 198). The harvested grain, and of course any imported grain, was ground locally. Rotary querns, usually in the form of quernstone fragments, have been found with great frequency. Stone is well preserved, and is a good indicator of long-distance trade. These querns were made from tephrite, a volcanic rock from the Eifel region in Germany. They arrived in these parts by ship down the Rhine, possibly passing through Dorestad. Excavations at Dorestad in the 1970s brought to light over seven hundred fragments of quernstones, as well as four hundred potential fragments (Kars 1980). Almost every excavation in a dwelling mound will produce one or more broken pieces of quernstone. Basketry was a technique that was well understood. In Ostfriesland as well as in Englum (Groningen) a well was found to have at its base a basket that reinforced its base (Bärenfänger 1999, 94; Bottema-Mac Gillavry 2008, 181). Basketwork is also useful for constructing fish traps. Many of these have appeared in excavations elsewhere, but not yet in the northern provinces. This may be due to the strong focus on the dwelling mounds: the vast space around them is only rarely investigated. The common tool above all others is the knife. Graves in the cemetery of Oosterbeintum produced twelve knives, but they are less common in settlement excavations (Knol et al. 1996). There is a wide variety of whetstones, however, which offer circumstantial evidence of metal blades. Besides plain knives, some folding knives too have been documented. They could have served many different uses.

Artisans The dwelling-mound communities were probably self-sufficient in terms of various crafts apart from weaving. People presumably made their own wooden, antler and bone implements. Semi-manufactures and waste from antler working are well documented. But there will also have been specialists. Traces of precious metalworking and bronzeworking have been found. The items of precious metal and bronze bear great similarity to similar pieces recovered elsewhere, but with typically ‘Frisian’ traits. This is indicative of local manufacture taking into account Frisian preferences. Whether these metalworkers, like the blacksmiths, travelled around, or settled on the larger mounds, has not yet been decided. At any rate they had plenty of work in this densely populated part of the world. Unfortunately during the quarrying of the terps slags and other debris of metal working were not considered as interesting objects. No doubt blacksmiths were working in the area. The iron came from the hinterland. In the Netherlands iron 25

Z01 Hines Book B.indb 25

07/05/2021 13:29

Frisians of the Early Middle Ages was produced from klapperstenen, nodules of loam stone with an iron ore core, in the Veluwe. It could be transported via the harbour of Dorestad (Joosten 2004; Van Nie 2007). Another specialist was the beadmaker, who would create a wide variety of beads from pre-manufactured glass. The glass was imported from the south in the form of glass rods and tesserae. The Wijnaldum excavations uncovered traces that point to glassworking (Sablerolles 1999a; 1999b). Glass beakers were imported from the Rhineland (Maul 2002).

Vehicles and boats Wheels have been found in several instances, both disc and spoked wheels. In summer, when the clay is dry, the land is perfectly passable. For overland transport, however, bridges will be needed for crossing the many watercourses. In historical times, the region saw a wide variety of bridge-types, ranging from simple beams to complex structures. Around the Halligen of northern Germany, such bridges may still be observed. The chance of such bridges being recovered archaeologically is small, but we should not doubt that they were there. Of course the region is highly suitable for waterborne transport. Our knowledge about the boats is limited, but expanding. For instance, at Jemgum, just across the German border, half a dugout canoe dating from the sixth century has been recovered (Fig. 2.9a; Thieman and Kegler 2013). A boat grave was reportedly uncovered in the cemetery of Hogebeintum, but the description is very brief and the surviving photograph shows too little for us to be certain. To make the boat more visible for the photograph, it has been daubed with limewash. Unfortunately there is no comparable photograph without limewash. But given the fact that the excavators had already uncovered several tree-trunk coffins and were also familiar with river craft, their observations probably were correct (Knol 2019b; Knol et al. 2019: grave 88). The interpretation is backed up by a boat grave also having been found at Early-medieval Dunum in Ostfriesland, and a further example documented at Solleveld near The Hague (Siegmüller and Peek 2015; Waasdorp and Eimermann 2008). In various Frisian dwelling mounds, but also in South-Holland along the coast and the Rhine, rivets have been found, pointing to clinker-built vessels (Knol et al. 1996; Reinders and Aalders 2007; Knol 2008b, 308; Brouwers et al. 2015). The coastal zone was poor in timber, so that old beams, driftwood and any other wood that could not be reused will have been burnt as fuel. Old wheels have been found to reinforce wells, and traces of ships’ timbers and rivets turn up in graves, reused as coffins or pyre wood. In the terp of Pogum, just across the German border on the banks of the River Ems, a small ceramic model of a boat was found (Fig. 2.9b; Bärenfänger 1997).

Weaponry Like tools, weapons are a rare finds in the terpen. Iron weapons generally would be recycled. An exception were the weapons that ended up as grave goods. In the fifth, sixth and seventh centuries this was not a common custom. Angos (angones) are known from graves at Hogebeintum and Hitzum (Knol 1993, fig. 49; Knol et al. 2019, grave 132 ). At Oosterbeintum a male inhumation was accompanied by a narrow seax and a 26

Z01 Hines Book B.indb 26

07/05/2021 13:29

Fig. 2.9  (a) The half dug-out canoe from Jemgum in the exhibition ‘Het verdronken land is vruchtbaar’ [The drowned land is fertile] at the Groninger Museum in 2013 (photograph: Marten de Leeuw). Collection Archäologische Forschungsinstitut der Ostfriesischen Landschaft, Aurich; (b) a ceramic boat found in the terp of Pogum, c. ninth/tenth century. After Bärenfänger 1997.

Z01 Hines Book B.indb 27

07/05/2021 13:29

Fig. 2.10  Two yew-wood bows. Right: from Aalsum in Groningen; left: from Hoogterp (Heechterp) near Leeuwarden, Friesland. The front of the bow is at the top: as a result of drying out, the Aalsum bow has curved the wrong way. Groninger Museum (Aalsum), Fries Museum Hoogterp, after Junkmans 2001, fig. 29.

Z01 Hines Book B.indb 28

07/05/2021 13:29

Material Culture in Frisia spearhead (Knol et al. 1996: grave 335). Some other weapons, including a throwing axe, were recovered as stray finds, but may also have come from graves (Knol 1993, 178–80). In the eighth century, the time when the Frankish kings successfully conquered the northern Netherlands, large numbers of weapon graves suddenly appear. They were mainly burials containing a sword, or multiple weapons including a sword, and sometimes also a long seax or a spear. Stirrups too may be found in these graves, as may a glass tumbler. Horse graves can at least in some part be assigned to this period. The battles that accompanied the subjection of the Frisians apparently encouraged a martial emphasis in the burial rite. The inhumation graves with weaponry may have been those of sympathisers or opponents of the Franks; the cremation graves with weaponry must have belonged to their opponents, as the Christian Church was strongly opposed to cremation of the dead. Several of the cemeteries contain multiple weapon graves. Although it is tempting to regard these as the burials of a local elite, the political power of these individuals need not have been very great. They certainly were not a supra-regional elite (Knol and Bardet 1999; Knol 2001). From later times we know that farm tools, and dogs too, could be fearsome weapons. Several dog graves were found in the cemeteries of Hogebeintum and Oosterbeintum. The dogs stood around 65 cm at the withers. Well-trained dogs of this size were a formidable defence against predators, but also against human enemies (Prummel 1992; Knol et al. 1996, 315–27). So there was at least some historical basis for the cartoon figure of Obélix’s dog Idéfix! The weapons show that the mound dwellers did not always live in peace. Inevitably there were local feuds to fight over. Besides, money or goods had to be paid to outside rulers. For long periods these are likely to have been rulers from the north, then Frankish kings, then again the Vikings for a while (IJssennagger 2017, 129–40, 154–9). Unfortunately there are no chronicles or other documentary sources that tell us about these developments in detail. Special finds are two archers’ bows of yew. The specimen from Aalsum in Groningen has been radiocarbon-dated to cal AD 680–900. The one from Heechterp near Leeuwarden has a far broader dating covering cal AD 210–515 (Fig. 2.10). The bows were presumably used in hunting, but might be deployed in warfare as well (Junkmanns 2001; Lanting et al. 1999). Like most other wooden artefacts, bows will usually have ended up ignominiously as fuel.

Coinage A particular form of ‘implement’ is money. A great deal of exchange will have taken the form of barter. Rent too was often paid in kind, but the Frisians, especially those who traded for a living, did know the value of cash. This currency derived from the Roman standards with the solidus coin as the highest gold denomination. As well as the Carolingian coinage of the eighth and ninth centuries, in the eighth century quite a quantity of sceattas were circulating in Frisia. The economic connections with England were strong (Op den Velde and Metcalf 2007). They would also need silver change. Coins themselves are generally closely datable, but because they could remain in circulation for a considerable time their value in context-dating is limited. With stray finds especially, it is hard to tell when they ended up in the ground. Westergo in Friesland has produced a remarkable number of gold hoards from the seventh century (Nicolay 2014). By the eighth century the gold and silver finds are more 29

Z01 Hines Book B.indb 29

07/05/2021 13:29

Frisians of the Early Middle Ages evenly spread around the coastal region of the northern Netherlands (Knol 1993, 222; 2005). In the ninth century the distribution is still even but there is a strong increase in the number of hoards. This is also the era of the imitation solidi of Louis the Pious. He had minted gold solidi in 816, which evidently were popular in the northern Netherlands, and frequently copied there. Coupland (2016) surmises that they were made on the instigation of Viking chiefs who had parts of Frisia in fief. This fits in with a considerable number of phenomena that point to a northerly influence in the ninth century. The typical ‘guide fossils’ of Vikings include Islamic silver coins (Besteman 2004). The hoards and the solidi suggest a restless time with foreign rulers. This has remained unacknowledged because the Vikings provided themselves with Frankish items that ended up in the northern Netherlands even without interference of the Vikings. In this context, the hoard of Marum is interesting. Here two brooches were found, both consisting of a silver coin set into a silver beaded rim – a well-known type of disc brooch. Of the brooches from Marum, one contained a Carolingian coin and the other an Islamic dirham (Groenendijk and Van den Bosch 2016). The latter makes the find into one linked to the Vikings. Yet it also begs the question of other examples of such brooches with a Carolingian coin or imitation solidus of Louis the Pious. If Coupland is right in believing these coins to be Viking-related, then the jewellery may certainly be so too. For instance, there is a finger ring with a gold imitation solidus with a beaded rim from Herbrum near Aschendorf on the River Ems, just outside Ostfriesland (Berghaus 1959). But there also are examples in baser metals, which are known as ‘pseudo-coin brooches’ and made after a solidus of Louis the Pious (e.g. Rasquert: Groninger Museum 1935-02-004).

Ideological objects Little is known in detail about the religion of the Germanic peoples, including the Frisians (cf. Hines, this vol.). No unequivocal cultic sites have been identified. Of course we have cemeteries with both inhumations and cremations. The dead were buried wearing their personal ornaments and sometimes with a few grave goods such as a pot. Garments were in evidence only as impressions in the corrosion of certain pieces of jewellery. It seems that the dead were expected to benefit from their possession in the afterlife. We have a somewhat better idea of amulets. Teeth of dogs, wolves and bears, slices of antler with or without decoration, obelisk-shaped pendants, all point to the assignment of powers to certain objects, be they prophylactic powers or in support of one’s own powers. In the case of teeth, we may think of warding off toothache, or conferring on the wearer the strength of a large dog, wolf or bear. A cowrie shell is presumably meant to enhance fertility. These shells came from the Red Sea area. One was found at Ferwerd in the grave of a woman, two stray finds are known from Lekkum and Adorp, and from Holwerd we have a fragment with a perforation (Knol 2006; Prummel and Olivier 2008). No cultic idols have been found.1 It should be noted that our archaeological knowledge is based mainly upon the settlements, in the dwelling mounds. Yet the wide open landscape was also part of the inhabited world. Any ritual activities in the salt marsh, 1

Very newly revealed, in fact, is a halved anthropomorphic wooden figure that appears to have been an idol from the site of Heiloo in North-Holland (ed.). See Hines, this vol.

30

Z01 Hines Book B.indb 30

07/05/2021 13:29

Fig. 2.11  A pectoral cross of the ninth century. Found before 1884 in one of the three villages in Groningen called Oostwold. Groninger Museum (photograph: Marten de Leeuw).

Z01 Hines Book B.indb 31

07/05/2021 13:29

Frisians of the Early Middle Ages at watercourses or on bogs are hidden from our observation. However, any marked accumulation of ritual objects might well have been noticed at some point. In the eighth and ninth centuries, objects adorned with crosses form a reference to Christianity, which was introduced in Frisia in the eighth century. Apart from brooches with crosses, there are bronze keys with crosses. A very special object is a bronze pectoral cross, found in one of the three villages in Groningen bearing the name of Oostwold – at the time people failed to specify which of them. All three settlements were founded in the tenth century at the earliest. The cross is heavily worn. It must have been made in the ninth century in Frankish court circles (Fig. 2.11; Brandt and Knol 2007). The importance of hunting as a food source was not very great – as is evident from the faunal remains. Taking part in a hunt possibly had an ideological significance above all. As regards status symbols, our attention is drawn mainly by the precious items of gold, silver and garnet (Nicolay 2014). But apart from these obvious items, there are others which may equally point to elite status. For instance, there is a copper neck-collar from Cornjum which in fact is a replica of a costly gold neck-ring (Knol 1993, 235). At Tzum a small bronze vase of the seventh century was found that used to be regarded as a Coptic vessel from Egypt. It turns out that such vessels were made much closer to home, presumably in northern Italy, but still it belongs among the guide fossils for the elite (Knol 1993, 207; Willemsen 2018). Another special find is a staff of whale bone with a Frisian runic inscription, presumably dating from around 800 (Fig. 2.12; Knol and Looijenga 1990; IJssennagger 2012; Knol 2018). It must have been an emblem of dignity belonging to a priest, a judge or even a ruler. It is quite unique. The material, the jawbone of a whale, may well have been found on the Wadden Sea coast. At the time, whale strandings probably occurred no less often than they do nowadays; the whale population had not yet been decimated by overfishing.

Fig. 2.12  The staff from Bernsterburen, made of whale bone and inscribed with runes. Fries Museum. After Knol and Looijenga 1990.

32

Z01 Hines Book B.indb 32

07/05/2021 13:29

Material Culture in Frisia The remains of the material culture of Frisia offer many opportunities for study, of which just some examples have been given here. They show that the coastal area – seemingly a desolate zone where storms would bring the threat or the reality of inundation – was a first-rate settlement area for well-adapted people. Arable and livestock farming could feed a considerable population, and also supported flourishing trades and commerce. Small wonder that in the ninth century this part of the world proved highly attractive to the Vikings.

Discussion NIJDAM  Nelleke [IJssennagger] has written on the Bernsterburen Tau staff, suggesting it may be a bishop’s staff. IJSSENNAGGER-VAN DER PLUIJM  Yes, there is a short article (IJssennagger 2012). It was very speculative, trying to think about the name tuda, which is a known bishop’s name, and trying to connect it with other finds from the Early Medieval Period. The staff is difficult to date. Its decoration suggests the ninth century. Radiocarbon-dating failed, but it would have been problematic because of the marine reservoir effect anyway. KNOL  I wonder if the many runic inscriptions with personal names all record the makers or the owners/possessors. Those would refer to people. But could it be that the names are those of the objects themselves? We know of that in the case of weapons of course. It would be difficult to prove, though some combs are called ‘comb’ (see Looijenga, this vol.). Bishops did have Tau-staffs, but perhaps other officials could carry such a staff of office and authority. LOOIJENGA  There is a picture from a manuscript in the Pierpoint Morgan library where a bishop has just such a staff. KNOL  But there could be others with such staffs too. IJSSENNAGGER-VAN DER PLUIJM  Tuda is a known bishop isn’t he? WOOD  Yes, of Lindisfarne, in the seventh century (Bede, HE, III.25–7). NIJDAM  My second question is about the clay model of the ship. A long time ago, in the old Frisian Museum, there were similar models of small pots. Has any research been done on these? They seem to be symbolic. IJSSENNAGGER-VAN DER PLUIJM  These may be different things. There is a whole series of these miniature vessels, and they seem to have a different function. There are further miniature ships as well. I believe there’s one from Dorestad. They can be seen as toys. KNOL  You can never be certain. From the fifth century BC we have a unique human mask in Groningen. We immediately think such an item must be ritual, but a school class asked me if clay was rare, and couldn’t it just have been made for fun. You couldn’t argue with that. IJSSENNAGGER-VAN DER PLUIJM  People will make models of things they are familiar with. NIJDAM  It was meant as a question for archaeologists to tell me what the state of research currently is. NIEUWHOF  Miniature pots are found in all periods from the Pre-Roman to the Middle Ages. The medieval ones are usually little cups for drinking. But they seem always (in Ezinge) to end up in ritual deposits, even if they were practical objects. 33

Z01 Hines Book B.indb 33

07/05/2021 13:29

Frisians of the Early Middle Ages MAJCHCZAK  We have also found them in North Frisia, where they appear in ritual deposits. But some are very poorly made indeed. NIEUWHOF  Perhaps they were containers for offerings of butter or the like? HINES  The range of topics you have reviewed may appear mundane, but the range of reflection that goes with them is extremely valuable. Looking at much more humble objects which may require very little interpretation brings home to us how much of the material culture of these people was based on organic materials, which will be preserved much better in the terp area. An especially interesting point is that with the combs: large amounts of antler were used, but that won’t have been collected locally; it must have been transported and traded in some way. Likewise the barrels: the timber will have come from further inland. Has dendroprovenancing produced any pattern in that respect? And a final question concerns the iron: is this an area in which large amounts of bog iron could be collected? KNOL  There isn’t a lot of iron in the coastal region, but further inland in the Netherlands, close to Dorestad and so not far away, there was a lot of bog iron and iron production. As for the antler, deer will occasionally have made it into the marsh area, and probably aurochs too. But there certainly was exchange with the Pleistocene hinterland; and of course in stone too. From much further afield came the cowrie shells from the Red Sea, and garnets from even further away. HINES  You might want a cowrie shell, but you don’t need one in the same way as you will need the iron, wood, stone, and maybe antler too. This builds up a picture of a regionally complex economy in which exchange is essential. KNOL  The marsh area is extremely fertile; a huge meadow. That is what made it productive. The density of habitation in the Early Middle Ages was massive. In good years there would have been an agricultural surplus to exchange. That is the wealth that made these areas highly viable in medieval and early modern times too. VERSLOOT  When you have that sort of exchange with the hinterland, one will expect similarities with the hinterland in material culture. What is additionally interesting in cultural terms is if there are material similarities throughout the geographical exchange zone. Do we have a situation whereby what is found in Oostergo and Rustringen is more similar within itself than what is found in Groningen and Drenthe, for example? Was there a coastal material culture, or was it similar across this range? KNOL  I would say it was the latter: in Drenthe you find more or less the same pottery, although of course you cannot find the same organic materials there. In Westergo you have a predominance of Merovingian wheel-thrown pottery which does not appear further east. I believe that more complex objects were distributed via Dorestad. All of these areas needed millstones, which all came from the Eifel area, along the Rhine. There must have been a lot of small-scale exchange. VERSLOOT  We can say there must be something common between the inland hinterland and the coastal area, but the differences within the latter area will have been comparable to those within the wider area. KNOL  I think so. Johan [Nicolay]’s view of the brooches does identify some specifically coastal items. I am focussing on basic subsistence matters: food and heat. This is not the level of luxury.

34

Z01 Hines Book B.indb 34

07/05/2021 13:29

Material Culture in Frisia NICOLAY  Perhaps we use the term ‘Early Medieval’ too easily. Each century is different, and chronological differences have to be considered alongside geographical difference. FLIERMAN  I was struck by the fact that you hinted that the shoes may have been a social marker of some kind, but that they have a limited chronological framework. KNOL  They do not turn up in pairs. We have about seven. They were found during the quarrying period of excavations; most of them from wells. I do not know why they were not found in Groningen as well. Van Giffen was an avid collector. It is puzzling. There is a coincidence in distribution with Anglo-Saxon pottery and certain brooches. NIEUWHOF  Why not in Groningen though? FLIERMAN  Shoes may be a display matter, but could also be a very functional choice. KNOL  The record only shows us shoes found in this area, from this period. This was almost a forgotten collection of finds that was re-discovered by an expert on the history of shoes who undertook the research and discovered these aspects. MAJCHCZAK  How many glass objects are there in the terp region; is it common? KNOL  Six of the ‘bell-shaped tumbler’ (Glockentummler) have been found in Early-medieval graves, at Wons Hayumertille, Pingjum, Ferwerd Burmania I, Westergeest Weerdeburen, Amtum I and Oldenburg Klunderborg (Knol 1993, 78–9, 181–7). NICOLAY  But there were dozens of sherds from Wijnaldum. MAJCHCZAK  In North Frisia, if you can apply the appropriate methods you find that these sherds are fairly numerous, as quite a common import. NICOLAY  However the glassworking sites are quite exceptional. Wijnaldum is one. MAJCHCZAK  Do you think these vessels were made here? NICOLAY  No, not these vessels; this is a matter of beadmaking. We would still link that to important sites. MAJCHCZAK  We have done a lot of research into glass in collaboration with colleagues in Cologne, who have copious information from the harbour area, where funnel beakers were found. Chemical studies were able to show that at a certain date those were being manufactured in Cologne but later were made in Hedeby. NICOLAY  The details of the chemistry are in the Wijnaldum volume (Henderson 1999), but the comparative studies haven’t really been done. KNOL  These are grave finds. The graves often contain both a sword and the glass beaker. So they are part of the weapon grave milieu, along with spears, long seaxes, stirrups etc, specific to the eighth century. NIJDAM  To return to the shoes – did people used to go around barefoot? How common were shoes? KNOL  Very few shoes have been found archaeologically. There is a peat body called Bernie who had something on his feet (Pünschel et al. 2020), a kind of woollen footwear, which is possible in the terp area too. The nineteenth-century population went around barefoot in the summer. This was regarded as an unusual area then. NIJDAM  One reason I ask is because of references in the Old Frisian laws to the barefooted champion (bereskinzes kempa) who would fight a duel. There is an image in a thirteenth-century mural in the church of Westerwijtwerd in Groningen (Nijdam 2008: 170, 296–300). 35

Z01 Hines Book B.indb 35

07/05/2021 13:29

Frisians of the Early Middle Ages KNOL  Sumo wrestlers are also barefooted. It is a special situation of course. At least in the summer time people were usually barefooted. We, for instance, are usually barehanded, wearing gloves only in the winter. IJSSENNAGGER-VAN DER PLUIJM  Actually, a woollen mitten from Aalsum in the terp area may be the oldest known in the Netherlands, possibly from the seventh century, and there is one of the eighth or ninth century from Dorestad (Willemsen 2015). KNOL  Being barefoot is no surprise. Nor is it to use dung for heating, which was also used in the Halligen up to the nineteenth or the beginning of the twentieth century. In such a fertile area, you don’t need it to manure the soil. HINES  There is a nice anecdote in a book about Appledore in North Devon in the inter-war period, where one of a group of boys in a photograph is deduced to have just come from home, because he still has his shoes on (Slade 1980, pl. 5). This has interesting implications: that wearing footwear or not may be age-related, and it’s governed by context – it seems to have been considered proper to wear footwear in the house, but not, amongst the peer-group, where you’d think it would have been most wanted for comfort and protection.    To ask about the coinage as well: I am interested in the considerable number of finds of sceattas. Mark Blackburn’s detailed study of stray finds in England suggests these coins represent small change that had been lost and can now be retrieved by metal-detectors (Blackburn 2003). There is a huge peak of finds over a large area in the eighth century that implies intense monetization, with coinage used for regular, day-to-day exchange. Is there any comparable analysis of the pattern of coins finds in Frisia which might or might not match with this? This again is relevant to the question of economic relations with the hinterland. KNOL  I know there are some studies, but I don’t know a distributional study. The Netherlands’ PAS [portable antiquities] scheme is a new system, but the numismatic evidence has been recorded over a long period. There have, however, been recent changes. IJSSENNAGGER-VAN DER PLUIJM  Frans Theuws may be looking at these now. DE LANGEN  There are hoards, but there are also stray finds. What is striking is that the sceattas I have seen appear quite new, barely worn. Wybrand Op den Velde has studied them together with Michael Metcalf (2003). They note the possible role of Radbod in striking sceattas. Local minting by itinerant moneyers is possible too. HINES  One of the interesting corollaries of Mark Blackburn’s views is that the coin reforms of the later eighth century effectively demonetized the economy. They brought an aspect of economic exchange much more under the control of a social elite. Yet Northumbria in England, and Friesland, continue with a low-value currency. DE LANGEN  You also get continuity in Frisia of an old way of counting. NIJDAM  Henstra did a detailed study, which has hardly been noted (Henstra 2000; 2010). It deserves more attention. DE LANGEN  …and sceattas continue until the end of the eighth century in Denmark. So that coinage moves into a marginal area. NIJDAM  The Old Frisian laws refer to payments in Cologne pennies, and a preference for the old Frisian form of money (Henstra 2000, 64–6, 92–3). 36

Z01 Hines Book B.indb 36

07/05/2021 13:29

Material Culture in Frisia IJSSENNAGGER-VAN DER PLUIJM  How do we know where the sceattas were minted? HINES  It can be inferred from the distributions in some cases, particularly with the so-called ‘Secondary sceattas’. That can be risky, but there are, for instance, types which predominate in Southampton and cluster in the area around it (Metcalf 1994, esp. 297–302). Some sceattas are ‘Frisian’ in distribution, while another type is predominant in Ribe but there is strong disagreement about whether they were locally made or not (e.g. Feveile 2008). IJSSENNAGGER-VAN DER PLUIJM  The coins were intended for circulation. HINES  That is one of the odd things about the Ribe coins, that they don’t seem to circulate outside the trading site to any great extent. There are few very locally; it’s very different from England. But you shouldn’t impose expectations from one area on another. DE LANGEN  The Continental runic types are linked to Frisia.

References Primary sources Bede, Historia Ecclesiastica Gentis Anglorum. Ed. and trans. B. Colgrave and R. A. B. Mynors 1969, Bede’s Ecclesiastical History of the English People (Oxford). Secondary sources Bär, J. 2014, ‘“Bernie” wiederentdeckt – Die Moorleiche von Bernuthsfeld im Blickppunkt interdiziplinärer Wissenschaft’, Berichte zur Denkmalpflege in Niedersachsen 34 (4), 146–52. Bärenfänger, R. 1993, ‘Frühmittelalterliche Eggenbalken und weitere Holzfunde aus Hattersum, Kreis Wittmund/Ostfriesland’, Archäologisches Korrespondenzblatt 23, 127–39. — 1997, ‘Ein vergessenes Keramik Schiffchen aus Pogum, Landkreis Leer’, Nachrichten aus Niedersachsens Urgeschichte 66(1), 299–302. — 1999, ‘Mittelalter’. In Führer zu archeologischen Denkmälern in Deutschlang Band 35 Ostfriesland (Stuttgart), 90–106. — 2002, ‘Befunde einer frühmittelalterlichen Siedlung bei Esens Ldkr Witmund (Ostfriesland)’, Probleme der küstenforschung im südlichen Nordseegebiet 27, 249–300. Bärenfänger, R., A. Burkhardt, P. Carli-Thiele, H. Freund, S. Grefen-Peters and M. Schultz 1997, ‘Völkerwanderungszeitliche Körpergräber aus dem ostfriesischen Wattenmeer – archäologischer, anthropologischer, paläopathologischer und paläoethnobotanischer Befund’, Studien zur Sachsenforschung 10, 1–47. Behre, K.-E. and P. Schmid 1998, Das Niedersächsisches Instuitut für historische Küstenforschung 60 Jahre Forschungstätigkeit im Küstengebiet (Wilhelmshaven). Berghaus, P. 1959, ‘Ein Karolingischer Münzring von Herbrum (Kreis Asschendorf-Hümmling)’, Die Kunde 10, 90–7. Besteman, J. C. 2004, ‘Scandinavisch gewichtsgeld in Nederland in de Vikingperiode’. In Van Solidus tot Euro. Geld in Nederland in economisch-historisch en politiek perspectief, eds. E.H.P. Cordfunke and H. Sarfatij (Hilversum), 21–42. Blackburn, M. 2003, ‘“Productive sites” and the pattern of coin-loss in England, 600–1180’. In Markets in Early Medieval Europe: Trading and ‘Productive’ Sites, 650–850, eds. T.

37

Z01 Hines Book B.indb 37

07/05/2021 13:29

Frisians of the Early Middle Ages Pestell and K. Ulmschneider (Macclesfield), 20–36. Bos, J. M. and E. W. Brouwer 2005, ‘De kruisvormige fibulae van Friesland’, De Vrije Fries 85, 9–36. Bottema-Mac Gillavry, N. 2008, ‘Hout’. In De Leege Wier van Englum. Archeologisch onderzoek in het Reitdiepgebied, ed. A. Nieuwhof, Jaarverslagen van de Vereniging voor Terpenonderzoek 91 (Groningen), 176–86. Brandenburgh, C. R. 2010, ‘Early medieval textile remains from settlements in the Netherlands. An evaluation of textile production’, Journal of Archaeology of the Low Countries 2 (1), 41–78. — 2012, ‘Old finds rediscovered: Two early medieval headdresses from the National Museum of Antiquities, Leiden, the Netherlands’. In Medieval Clothing and Textiles 8, eds. R. Netherton and G.R. Owen-Crocker (Woodbridge), 25–47. — 2016, Clothes Make the Man. Early Medieval Textiles from the Netherlands, Archaeological Studies, Leiden University 30 (Leiden). Brandt, M. and E. Knol 2007, ‘Het Karolingische kruis van Oostwold’, Groninger Kerken 24 (3), 655–71. Brouwers, W., E. Jansma and M. Manders 2015, ‘Middeleeuwse scheepsresten in Nederland: de Vroege Middeleeuwen (500–1050)’, ArcheoBrief 19 (3), 6–24. Coupland, S. 2016, ‘Recent finds of imitation gold solidi in the Netherlands’, The Numismatic Chronicle 176, 261–9, plate 30. De Langen, G. J. 2011, ‘De gang naar een ander landschap’. In Gevormd en omgevormd landschap van prehistorie tot middeleeuwen, ed. M. Niekus, 70–97. De Langen, G. J., J. A. W. Nicolay, M. Schepers, A. Nieuwhof, D. Postma, M. Bakker, T. W. Varwijk, W. Prummel, G. Aalbersberg and P. C. Vos 2016, ‘Terpenlandschap als actueel onderzoeksthema’, Tijdschrift voor landschapsgeschiedenis 34 (1), 50–8. Dijkstra, J. and J. A. W. Nicolay eds. 2008, Een terp op de schop. Archeologisch onderzoek op het Oldehoofsterkerkhof te Leeuwarden (Amersfoort), 25–42. Dijkstra, M. 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’ (Ph.D. thesis, University of Amsterdam). Farke, H. 1998, ‘Der Männerkittel aus Bernuthsfeld. Beobachtungen während einer Restaurierung’. In Textiles in European Archaeology: Report from the 6th NESAT Symposium, 7–11th May 1996 in Borås, ed. L. Bender Jørgensen and C. Rinaldo (Götenborg), 99–106. Feveile, C. 2008, ‘Series X and coin circulation in Ribe’. In Studies in Early Medieval Coinage 1: Two Decades of Discovery (Woodbridge), 53–67. Fischer, A., C. Peek and A. Siegmüller 2012, ‘Feinstratigraphische Untersuchungen an Eisenobjekten des Gräberfeldes von Dunum, Ldkr. Wittmund, Ostfriesland. Zur Funktion und Deutung organischer Funde und Befunde’, Siedlungs- und Küstenforschung im südlichen Nordseegebiet 35, 317–51. Groenendijk, H. and M. van den Bosch 2016, ‘Depot Jonkersvaart (Gr.): twee 9de eeuwse zilveren muntfibula’s en een Romeinse bronzen munt’, Paleo-Aktueel 26, 55–63. Groenendijk, H. A. 2006, ‘Dorfwurt Ulrum (De Marne, prov. Groningen) Eine Fundbergung im Jahre 1995 als anregung zur Benutzung hydrologischer messdaten bei der Erhaltung von Grosswurten’, Palaeohistoria 47/48, 529–53. Henderson, J. 1999. ‘Scientific analysis of the glass and the glass-bearing artefacts: technique, raw materials used and archaeological interpretation’. In J. C. Besteman et al., The Excavations at Wijnaldum: Reports on Frisia in Roman and Medieval Times, Vol. 1

38

Z01 Hines Book B.indb 38

07/05/2021 13:29

Material Culture in Frisia (Rotterdam), 287–97. Henstra, D. J. 2000, ‘The Evolution of the Money Standard in Medieval Frisia. A Treatise on the History of the Systems of Money of Account in the Former Frisia (c. 600–c. 1500)’ (Ph.D. thesis, University of Groningen) — 2010 (ed. A. T. Popkema), Fon Jelde. Opstellen van D.J. Henstra over Middeleeuws Frisia (Groningen). Hielkema, J. B. 2017, Dames van Midlum’Opgraving terprestant Midlum-zuid; N31 Traverse te Harlingen, Gemeente Harlingen, Archeologisch onderzoek: een opgraving. Report 3299 – version 23-11-2017 (Weesp). Hills, C. 1996, ‘Frisia and England: the archaeological evidence for connections’. In Frisian Runes and Neighbouring Traditions. Proceedings from the First International Symposium on Frisian Runes at the Fries Museum, Leeuwarden, 26–29 January 1994, eds. T. Looijenga, A. Quak, Amsterdammer Beiträge zur älteren Germanistik 45 (Amsterdam–Atlanta), 35–46. IJssennagger, N. L. 2012, ‘Cold case: de runenstaf van Bernsterburen’, Historisch Tijdschrift Fryslân 18 (5), 4–5. — 2017, ‘Central Because Liminal. Frisia in a Viking Age North Sea World’ (Ph.D. thesis, University of Groningen). Jahn, W. ed. 2019, Die Moorleiche von Bernuthsfeld. Kurzführer (Emden). Jensma, G. and E. Knol 2005, ‘Op zoek naar de juiste methode. Wierdenarcheologie in de negentiende eeuw’. In Professor Van Giffen en het geheim van de wierden, eds. E. Knol, W. Prummel and A. C. Bardet (Groningen–Veendam), 34–49. Joosten, I. 2004, Technology of Early Historical Iron Production in the Netherlands, Free University Amsterdam, Geoarcheological and Bioarcheological Studies 2 (Amsterdam). Junkmanns, J. 2001, Pfeil und Bogen. Herstellung und Gebrauch in der Jungsteinzeit (Biel). Kars, H. 1982, ‘Early-medieval Dorestad, an archaeo-petrological study. Part I: General introduction – The trephrite querns’, Berichten van de Rijksdienst voor het Oudheidkundig Bodemonderzoek 30, 393–422. Knol, E. 1993, ‘De Noordnederlandse Kustlanden in de Vroege Middeleeuwen’ (Ph.D. thesis, Vrije Universiteit Amsterdam). — 2001, ‘Carolingian weapons from the northern Netherlands, particularly from the cemetery of Godlinze’. In Kingdoms and Regionality, Transactions from the 49th Sachsensymposium 1998 in Uppsala, ed. B. Arrhenius (Stockholm), 115–20. — 2005, ‘Gold und Silber aus Marsum – karolingische Schatzfunde in den Niederlanden’. In Die Macht des Silbers, Karolingische Schätze im Norden, eds. E. Wamers and M. Brandt (Regensburg), 119–24. — 2006, ‘Kaurischelpen in wierdenland’, Stad & Lande 15 (1), 6–7. — 2008a, ‘Het Angelsaksische grafveld Beetgum Besseburen’. In Diggelgoud. 25 jaar Argeologysk Wurkferbân: archeologisch onderzoek in Fryslân, eds. K. Huisman and R. Salverda (Leeuwarden), 148–57. — 2008b, ‘Metaal uit de vroege middeleeuwen in Katwijk-De Zanderij’. In Cananefaten en Friezen aan de monding van de Rijn, Tien jaar archeologisch onderzoek op de Zanderij-Westerbaan te Katwijk (1996–2006), ed. H. van der Velde, ADC monografie 5 (Amersfoort), 295–310. — 2011, ‘The first early medieval cemeteries along the northern Dutch coasts and their significance for the Anglo-Saxon migration’. In Transformations in North-Western Europe (AD 300–1000), eds. T. A. S. M. Panhuysen and B. Ludowici, Neue Studien zur

39

Z01 Hines Book B.indb 39

07/05/2021 13:29

Frisians of the Early Middle Ages Sachsenforschung 3 (Hannover–Stuttgart), 218–27. — 2014, ‘Ezinge: metaal uit een opgraving zonder detector’. In En dan in hun geheel. De vondsten uit de opgravingen in de wierde Ezinge, ed. A. Nieuwhof, Jaarverslagen van de Vereniging voor Terpenonderzoek 96 (Groningen), 187–206. — 2018, ‘750 Schrift op de kwelder’. In De Wereldgeschiedenis van Nederland, eds. L. Heerma van Voss, M. ’t Hart, K. Davids, K. Fatah-Black, L. Lucassen and J. Touwen (Amsterdam), 93–7. — 2019a, ‘Living near the sea: the organisation of Frisia in early medieval times’. In Power and Place in Europe in the Early Middle Ages, eds. J. Carroll, A. Reynolds and B. Yorke, Proceedings of the British Academy 224 (Oxford), 369–91. — 2019b, ‘Het vroegmiddeleeuwse grafveld Hogebeintum’. In De Hoogste Terp van Friesland. Nieuw en Oud Onderzoek in Hogebeintum, eds. A. Nieuwhof, E. Knol and J. Nicolay, Jaarverslagen Vereniging voor Terpenonderzoek 101 (Groningen), 159–80. Knol, E. and X. Bardet 1999, ‘Carolingian weapons from the cemetery of Godlinze, the Netherlands’. In In Discussion with the Past, Archaeological Studies Presented to W. A. van Es, eds. H. Sarfatij, W. J. H. Verwers and P. J. Woltering (Zwolle), 213–25. Knol, E., M. P. L. Hoogland, H. T. Uytterschaut and W. A. Casparie 2019, ‘Catalogus van het grafveld Hogebeintum, 400–730 n. Chr.’. In De Hoogste Terp van Friesland. Nieuw en Oud Onderzoek in Hogebeintum, eds. A. Nieuwhof, E. Knol and J. Nicolay, Jaarverslagen Vereniging voor Terpenonderzoek 101 (Groningen), 181–231. Knol, E. and T. Looijenga 1990, ‘A tau staff with runic inscriptions from Bernsterburen (Friesland)’. In Aspects of Old Frisian Philology, eds. R. H. Bremmer, G. van der Meer and O. Vries, Amsterdamer Beiträge zur älteren Germanistik 31–2 (Amsterdam), 226–41, pl. IV–V, map 3. Knol, E., W. Prummel, H. T. Uytterschaut, M. L. P. Hoogland, W. A. Casparie, G. J. de Langen, E. Kramer and J. Schelvis 1996, ‘The early medieval cemetery of Oosterbeintum (Friesland)’, Palaeohistoria 37/38, 245–416. Kötschke, R. ed. 1906, Die Urbare der Abtei Werden a.d. Ruhr A: Die Urbare vom 9. – 13. Jahrhundert (Bonn). Lange, S. 2017, Uit het juiste hout gesneden. Houten gebruiksvoorwerpen uit archeologische context tot 1300 n.Chr., Nederlandse Archeologische Rapporten 54 (Amersfoort). Lanting, J. N., B. W. Kooi, W. A. Casparie and R. van Hinte 1999, ‘Bows from the Netherlands’, Journal of the Society of Archer-Antiquaries 42, 7–10. Lehmann, R., M. Viebrock, C. W. Karl, H.-J. Schmidt and D. Wengerowsky 2018, ‘HPLCESI-MS, GC-MS, LC-MS und ATR-IR-Farbstoffenanalysen an der Flickentunika der Moorleiche von Bernuthsfeld’, FAN-Post, 36. Looijenga, T. 2003, Texts & Contexts of the Oldest Runic Inscriptions (Leiden). Maul, B. 2002, Frühmittelalterliche Gläser des 5.–7./8. Jahrhundert n.Chr.: Sturzbecher, Glockenförmige Becker, Tummler und Glockentummler (Bonn). Metcalf, D. M. 1994, Thrymsas and Sceattas in the Ashmolean Museum Oxford, vol. 3 (Oxford). Miedema, M. 1983, ‘Vijfentwintig eeuwen bewoning in het terpenland ten noordwesten van Groningen’ (Ph.D. thesis, Vrije Universiteit Amsterdam). — 1989, ‘Het archeologisch materiaal uit de terp Wierhuizen’, Groningse Volksalmanak 1989, 74–164. — 1990, ‘Oost-Fivelingo 250 v.Chr.–1850 n.Chr.; archeologische kartering en beschrijving van 2100 jaar bewoning in Noordoost-Groningen’, Palaeohistoria 32, 111–245.

40

Z01 Hines Book B.indb 40

07/05/2021 13:29

Material Culture in Frisia — 2000, ‘West-Fivelingo 600 v.Chr.–1900 n.Chr. Archeologische kartering en beschrijving van 2500 jaar bewoning in Midden-Groningen’, Palaeohistoria 41/42, 237–445. Nicolay, J. A. W. ed. 2010, Terpbewoning in oostelijk Friesland. Twee opgravingen in het voormalige kweldergebied van Oostergo, Groningen Archaeological Studies 10, 94–131. — 2014, The Splendour of Power: Early Medieval Kingship and the Use of Gold and Silver in the Southern North Sea Area (5th to 7th century AD), Groningen Archaeological Studies 28 (Groningen). Nieuwhof, A. 2014, En dan in hun geheel. De vondsten uit de opgravingen in de wierde Ezinge, Jaarverslagen van de Vereniging voor Terpenonderzoek 96 (Groningen). — 2017, ‘Vlas in Ezinge – de herkenbaarheid van linnenproductie in het terpengebied’. In Paleo-Palfenier met Rita van Egypte tot Ezinge, eds. G. Aalbersberg, S. Boersma and M. Schepers (Groningen), 121–31. Nijdam, H. 2008, Lichaam, eer en recht in middeleeuws Friesland: Een studie naar de Oudfriese boeteregisters (Hilversum). Nüsse, H.-J. 2004a, ‘Upleward FStNr 2508/7:2–5, Gde Krummhörn, Ldkr. Aurich, Reg. Bez. W-E’, Nachrichten aus Niedersachsens Urgeschichte Beiheft 10, Fundchronik Niedersachsen 2003, 103–5. — 2004b, ‘Konserviertes bäuerliches Leben – frühmittelalterliche Häuser aus Ostfriesland’, Mitteilungen der Berliner Gesellschaft für Anthropologie, Ethnologie und Urgeschichte 25, 71–6. Op den Velde, W. and M. Metcalf 2003 (2007), ‘The Monetary Economy of the Netherlands, c. 690 – c. 715 and the Trade with England: A Study of the Sceattas of Series D’, Jaarboek voor Munt- en Penningkunde 90. Peek, C. and A. Siegmüller 2007, ‘Kostbarkeiten aus dem Norden? Neue Überlegungen zur Identifizierung Friesischer Tuche’, Archäologisches Korrespondenzblatt Heft 2, 283–96. — 2016, ‘Nadelröhrchen – ein praktisches Accessoire der Frauen in karolingischer Zeit. Zur Funktion und Trageweise der Nadelröhrchen aus dem Gräberfeld Dunum, Ostfriesland’, Archäologie in Niedersachsen 19, 62–4. — 2018a, ‘Ausgewählte Textilien der Wurt Niens, Ldkr. Wesermarsch, und der Feddersen Wierde, Ldkr. Cuxhaven – Ergebnisse neuer Untersuchungen’. In Siedlungs- und Küstenforschung im südlichen Nordseegebiet im südlichen Nordseegebiet 41 (Rahden), 173–88. — 2018b, ‘Die Gewebe aus der Wurt Oldorf, Landkreis Friesland. Ein Textilkomplex aus dem 7. Jh.’. In Von Hammaburg nach Herimundesheim. Festschrift für Ursula Koch, eds. A. Wieczorek and K. Wirth. Mannheimer Geschichtsblätter Sonderveröffentlichung 11, Publikationen des Reiss-Engelhorn-Museen 85 (Mannheim), 209–22. Prummel, W. 1992, ‘Early medieval dog burials among the Germanic tribes’, Helinium 32, 132–94. — 1993, ‘Paarden en honden uit vroeg-middeleeuwse grafvelden’. In Het Tweede Leven van Onze Doden, eds. E. Drent, W. A. M. Hessing and E. Knol (Ammersfoort), 53–60. Prummel, W. and E. Knol 1991, ‘Strandlopers op de brandstapel’, Paleo-aktueel 2, 92–6. Prummel, W. and L. Olivier 2008, ‘Twee bijzondere terpvondsten uit Holwerd’, Van wierden en terpen: Nieuwsbrief van de Vereniging voor Terpenonderzoek 11, 5. Pünschel, K., E. Jopp-van Well, W. Jahn, H. Haβmann, M. Schultz and A. Bauerochse 2020, “Bernie”- Die Moorleiche von Bernuthsfeld. Ergebnisse der interdisziplinären Erforschung und Rekonstruktion eines frühmittelalterlichen Fundkomplexes aus Ostfriesland, Materialhefte zur Ur- und Frühgeschichte Niedersachsens 57 (Rahden). Reinders, R. and Y. Aalders 2007, ‘Friese klinkerschepen in de vroege Middeleeuwen’, De

41

Z01 Hines Book B.indb 41

07/05/2021 13:29

Frisians of the Early Middle Ages Vrije Fries 87, 9–28. Roes, A. 1963, Bone and Antler Objects from the Frisian-Terpmounds (Haarlem). Sablerolles, Y. 1999a, ‘The glass vessel finds’. In The Excavations at Wijnaldum. Reports on Frisia in Roman and Medieval Times. Volume I, eds. J. C. Besteman, J. M. Bos, D. A. Gerrets, H. A. Heidinga and J. de Koning (Rotterdam–Brookfield), 229–52. — 1999b, ‘Beads of glass, faience, amber, baked clay and metal including production waste from glass and amber bead making’. In The Excavations at Wijnaldum. Reports on Frisia in Roman and Medieval Times. Volume I, eds. J. C. Besteman, J. M. Bos, D. A. Gerrets, H. A. Heidinga and J. de Koning (Rotterdam–Brookfield), 253–85. Schlabow, K. 1953, ‘Leichtvergängliche Stoffe aus der Wurtengrabung Hessens’, Probleme der Küstenforschung m südlichen Nordseegebiet 5, 26–43. — 1974, ‘Vor- und Frühgeschichtliche Textilfunde aus den nördlichen Niederlanden’, Palaeohistoria 16, 169–221. Schmid, P. 1969, ‘Zum heidnischen und frühchristlichen Bestattungsbrauch auf dem frühmittelalterlichen Gräberfeld von Dunum, Ostfriesland’, Frühmittelalterliche Studien 3, 258–76. — 1970a, ‘Das frühmittelalterliche Gräberfeld von Dunum, Kreis Wittmund (Ostfriesland) (Grabung 1967–1968)’, Neue Ausgrabungen und Forschungen in Niedersachsen 5, 39–62 and Taf. 19–24. — 1970b, ‘Die Keramik aus dem frühmittelalterlichen Gräberfeld von Dunum, Kr. Wittmund’, Probleme der küstenforschung im südlichen Nordseegebiet 9, 59–76. — 1972, ‘Zum Datierung und Gliederung der Grabanlagen von Dunum, Kreis Wittmund’­, Neue Ausgrabungen und Forschungen in Niedersachsen 7, 211–40. — 1994, ‘Oldorf – eine frühmittelalterliche friesiche Wurtsiedlung’, Germania 72, 231–67. — 1995, ‘Zum mittelalterlichen Besiedlung der Dorfwurt Feddersen Wierde, Land Wursten, Ldkr Cuxhaven’, Probleme der küstenforschung im südlichen Nordseegebiet 23, 243–63. Schön, M. D. 1995, Der Thron aus der Marsch (Bederkesa). — 1999, Feddersen Wierde, Fallward, Flögeln (Bederkesa). — 2015, ‘Möbel aus Gräbern des 4/5 Jahrhunderts an der Fallward bei Wremen’, Archäologie in Niedersachsen 2015, 43–7. Schwarz, W. 1990, Besiedlung Ostfrieslands in ur- und frühgeschichtlicher Zeit (Aurich). — 2016, Archäologische Funde aus dem Reiderland Ldkr. Leer (Rahden). Siegmüller, A. 2011, ‘Leichentücher und Federstreuungen. Das frühmittelalterliche Gräberfeld von Dunum, Ostfriesland, als Spiegel politisch-religiöser Wandlungen des 7.–10. Jahrhunderts im Küstenraum’. In Transformations in North-Western Europe (AD 300– 1000), eds. T. A. S. M. Panhuysen and B. Ludowici, Neue Studien zur Sachsenforschung 3 (Hannover–Stuttgart), 239–50. — 2012, ‘Leichentücher und Federunterlagen als Zeichen für religiösen Wandel und Fernkontakte am Beispiel des Gräberfelds von Dunum, Ostfriesland’. In Wechsel der Religionen – Religion des Wechsels. Tagungsbeiträge der AG SFM-Tagungen in Nürnberg 2010, Studien zu Spätantike und Frühmittelalter 4 (Hamburg), 289–93. — 2013, ‘Schafhaltung und Schafwäsche – Überlegungen zur Produktion und Funktion von Geweben im frühen Mittelalter’. In The North European Symposium for Archaeological Textiles XI, eds. J. Banck-Burgess and C. Nübold (Rhaden), 241–6. Siegmüller, A. and C. Peek 2015, ‘Das Bootgrab aus dem frühmittelalterlichen Gräberfeld von Dunum, Ldkr. Wittmund’, Nachrichten Marschenrat zur Förderung der Forschung im

42

Z01 Hines Book B.indb 42

07/05/2021 13:29

Material Culture in Frisia Küstengebied der Nordsee 52, 46–8. Slade, W.J. 1980, Out of Appledore, ed. B. Greenhill, 4th edition (Greenwich). Tempel, W.-D. 1969, Die Dreilagenkämme aus Haithabu, Studien zu den Kämmen der Wikingerzeit im Nordseeküstengebiet und Skandinavien (Göttingen). Theuws, F. C. J. W. 2018, ‘Reversed directions. Rethinking sceattas in the Netherlands and England’, Zeitschrift für Archäologie des Mittelalters 46, 27–84. Thiemann, B. and J. F. Kegler 2013, ‘Das Boot im Damm – ein frühmittelalterlicher Einbaum aus Jemgum, Ldkr. Leer (Ostfriesland)’, Siedlungs- und Küstenforschung im südlichen Nordseegebiet 36, 235–47. Tidow, K. and P. Schmid 1979, ‘Frühmittelaterliche Textilfunde aus der Wurt Hessens (Stadt Wilhelmshaven) und dem Gräberfeld vond Dunum (Kreis Friesland) und ihre Bedeutung’, Probleme der Küstenforschung im südlichen Nordseegebiet 13, 123–53. Tuinstra, S. J., J. R. Veldhuis and J. A. W. Nicolay 2011, Hallum, een welvarend dorp aan de monding van de Middelzee. Een archeologische opgraving te Hallum, gemeente Ferwerderadeel, (Fr), ARC-publicaties 205 (Groningen). Van der Laan, J. 2016, ‘De bijzondere houten voorwerpen uit de opgravingen in Ezinge’. In Van Wierhuizen tot Achlum. Honderd Jaar Archeologisch Onderzoek in Terpen en Wierden, ed. A. Nieuwhof, Jaarverslagen van de Vereniging voor Terpenonderzoek 98 (Groningen), 153–64. Van der Velde, H. ed., Cananefaten en Friezen aan de monding van de Rijn. Tien jaar archeologisch onderzoek op de Zanderij-Westerbaan te Katwijk, ADC monografie 5 (Amersfoort). Van Driel, C. and H. van der Plicht 2016, ‘Het gelijk van Boeles: schoenvondsten uit de Friese terpen’. In Van Wierhuizen tot Achlum. Honderd Jaar Archeologisch Onderzoek in Terpen en Wierden, ed. A Nieuwhof, Jaarverslagen van de Vereniging voor Terpenonderzoek 98 (Groningen), 171–80. Van Giffen, A. E. 1920, ‘Een Karolingisch grafveld bij Godlinze’, Jaarverslagen van de Vereniging voor Terpenonderzoek 3–4, 39–96 and pl. 0–XVI. — 1940, ‘Een systematisch onderzoek in een der Tuinster wierden te Leens’, Jaarverslagen Vereniging voor Terpenonderzoek 20–24, 26–115 and afb. 1–28. Van Nie, M. 2007, ‘Early medieval iron production and its organisation in the Veluwe area, the Netherlands’. In Material Culture in Medieval Europe. Papers of the ‘Medieval Europe Brugge 1997’ conference, volume 7, eds. G. De Boe and F. Verhaeghe (Zellik), 33–41. Waasdorp, J. A. and E. Eimermann 2008, Solleveld. Een opgraving naar een Merovingisch grafveld aan de rand van Den Haag, Haagse Oudheidkundige Publicaties 10 (Den Haag). Willemsen, A. 2015, ‘Handschoenen in Nederland voor 1700’, Archeologie Magazine 4, 58–61. — 2018, ‘A “Coptic” bowl from Ewijk and its cemetery context’. In Rural Riches & Royal Rags? Studies on Medieval and Modern Archaeology Presented to Frans Theuws, eds. M. Kars, R. van Oosten, M. A. Roxburgh and A. Verhoeven (Zwolle), 56–9. Zimmerman, H. 2010, ‘Two early medieval caps from the dwelling mounds Rasquert and Leens in Groningen province, the Netherlands’. In North European Symposium for Archaeological Textiles X, eds. E. Andersson Strand, M. Gleba, U. Mannering, C. Munkholt, and M. Ringgaard (Oxford), 288–90. Zimmerman, J. A. 2008, ‘Ein textilfund aus der Dorfwurt Ulrum’, Palaeohistoria 47/48, 552–3. Zimmermann, W. H. 2015, ‘Miszellen zu einer Archäologie des Wohnens’, Archäologie in Niedersachsen 15, 8–25.

43

Z01 Hines Book B.indb 43

07/05/2021 13:29

Z01 Hines Book B.indb 44

07/05/2021 13:29

• 3 • THE FRISIANS AND THEIR POTTERY: SOCIAL RELATIONS BEFORE AND AFTER THE FOURTH CENTURY AD Annet Nieuwhof

T

Discontinuity and connections

he beginning of the present-day Frisians lies in the fifth century, when new inhabitants, conventionally called ‘Anglo-Saxons’, arrived in the virtually abandoned terp region of the northern Netherlands. The Frisians that are mentioned by Roman authors such as Tacitus (Germania, 34–5) and Pliny (Naturalis Historia, 16.2), and who inhabited the coastal areas of the present provinces North-Holland, Friesland and Groningen, had largely abandoned the area in the third century, moving probably to the south, where they may have joined Frankish groups (Taayke 2000). In areas bordering the coastal salt-marsh area, in particular in Pleistocene northern Drenthe, inhabitation was continuous in this period, although maybe diminished (Nicolay and Den Hengst 2008, 584; Nieuwhof 2008b, 291–2). The correlation between continuity or not and these different landscapes is suggestive of the causes of the abandonment of the coastal region: changes in the landscape, especially increasing problems with drainage, may have been the stimulus. The salt marshes had developed over the centuries under the influence of the relative sea-level rise that the coastal areas from northern Germany to Flanders experienced from the beginning of the Holocene (Kiden et al. 2002; Vos and Knol 2015). The younger, northern parts were therefore much higher in elevation than the older salt marshes bordering the inland peat bogs. That caused increasing drainage problems in these older areas during the second and third centuries AD. While the terp settlements were well adapted to marine inundations that went as quickly as they came, long-term drainage problems were much harder to live with; they may have been the primary push-factor. The collapse of the Roman Empire was probably a pull-factor; it did not cause abandonment (in that case, inland areas would also have been abandoned), but it did provide new opportunities for the terp dwellers whose homelands had become too wet. Once the older and lower areas were abandoned, the well-drained parts of the salt marsh in the north also became less attractive because communities there lost a large part of their social network, and the social fabric disintegrated (Nieuwhof 2011). The remaining inhabitants, with some exceptions, left the area no later than AD 325 (e.g. at Wijnaldum: Gerrets and De Koning 1999, 99–106). 45

Z01 Hines Book B.indb 45

07/05/2021 13:29

Frisians of the Early Middle Ages There were regional differences. Some settlements in the present province of Groningen remained inhabited, although in a diminished form. That is certain for the settlement of Ezinge (Nieuwhof 2013; 2014), which is therefore often referred to here, and possibly some others (Taayke 1996, III, 68). There are also some inconclusive indications of continuity along the Middelzee in Friesland (Bakker and Varwijk 2016; Bosma 2015 [but cf. Taayke 2020]; Nieuwhof 2016). It is likely that these differences are related to the different social networks in which these areas participated (Nieuwhof 2011), as we will see. The cause of this abandonment probably was not warfare and raiding of the depleted and vulnerable population in the open coastal landscape, as Bazelmans (2000, 61) has suggested. Groups from northern Germany and Denmark would have been responsible for these raids. Traditionally, the Chauci, neighbours of the Frisians, and their Saxon successors are portrayed as pirates for whom raiding was a way of life, and who were involved in enlarging their territory to the cost of their neighbours (e.g. Taayke 1996, V, 191–2; Dhaeze 2011, 59). The accusation of Chaucian piracy is based on a very small number of mentions in the work of Roman authors (Tacitus, Annales XI, 18; Scriptores Historiae Augustae, Didius Julianus I, 7–8), which are often interpreted as the proverbial tip of the iceberg. I have argued elsewhere (Nieuwhof 2017a) that the raids by the Chauci may well have been induced by Roman aggression and policy of expansion, and should not be considered as a sign of aggression and raiding as a general social characteristic of the Chauci. It is by no means certain that they also raided their neighbours. Apart from the third-century abandonment itself, we have no indications of raiding or tribal warfare, such as defensive structures, reinforcements to settlements, weaponry, or trauma on human remains in the northern coastal region. We may also ask how the marine landscape of the terp region, one of the more extreme natural landscapes in the world, could have developed into a densely populated area,1 if communities had been subjected to raiding all the time. The evidence suggests rather that this was a relatively peaceful area, where cooperation and good relations with people nearby and further off contributed to successful existence. The open, maritime landscape does not seem to have been considered a safety risk; it rather facilitated the maintenance of contacts and participation in socio-cultural networks. Technological know-how was exchanged during visits. It is likely, for instance, that the knowledge of water management with the aid of dams equipped with culverts with tidal flapper valves, which were in use already in the Late Iron Age in such distant places as Vlaardingen in South-Holland (De Ridder 2005), Jemgumkloster on the Ems (Prison 2011) and Huntebrück near Bremen (Siegmüller and Mückenberger 2017), was exchanged during encounters between coastal population groups. The boats that were available until around the third century AD were paddled canoes: simple or expanded logboats (Van de Moortel 2011; Nieuwhof 2017a), some of which, according to Pliny (Naturalis historia, 16), could accommodate as many as thirty people. Logboats remained in use well into the Middle Ages. Clinker-built rowing boats appeared in northern Germany in the third century AD (Rieck 2003). In the Netherlands, rivets belonging to such boats are found in settlements and cemeteries from the fifth century to the ninth (Reinders and Aalders 2007, 18–23; Van de Moortel 2011, 96), 1

Taayke (1996, V, 185) made an estimate of around one settlement per square kilometre.

46

Z01 Hines Book B.indb 46

07/05/2021 13:29

Pottery and Social Relations for instance in the settlement of Wijnaldum and in the cemetery of Oosterbeintum. Morris (2015) argues that the introduction of these boats, which were better suited to long sea voyages, was a major factor in the cross-North Sea population movements from around AD 400. Earlier contacts probably went via Roman-controlled routes across the southern North Sea. The use of traditional ‘tribal’ names such as Frisians, Chauci or (Anglo-)Saxons ignores the fact that socio-cultural networks overlapped and boundaries between these groups were fluid. We do not know how these groups defined themselves or how important ethnic identity was to them. Other identities, for instance family relationships, may have been far more important. These ethnic names are therefore not used here to denote ethnic identity, but rather to describe different regions of origin. This chapter is about social relationships between these groups. The same social mechanisms, such as visiting, gift exchange and intermarriage functioned before and after the fourth century, and the same social networks and regions were involved. To understand the ‘New Frisians’ and their descent from migrating ‘Anglo-Saxons’, we cannot ignore the ‘Old Frisians’ and their contacts with the ‘Chaucians’. The focus is on pottery, since that is the most widespread material category and a telling one when social relationships are concerned.

Pottery and exchange The common pottery of the coastal area was hand-made until well into the Middle Ages. During the Roman Period, a very small amount of wheel-made pottery was imported, but it hardly played a role in daily use. The percentage of imported pottery during the Early Middle Ages was larger, especially in the Merovingian Period. At the Groningen terp settlement of Ezinge, which is one of the most extensively studied terp settlements in the northern Netherlands and therefore often serves as an example here, 30% of the total amount of pottery was wheel-made in this period (Thasing and Nieuwhof 2014), and at Wijnaldum even 60% (Gerrets and De Koning 1999, 97); the latter high percentage is undoubtedly related to the function of Wijnaldum and its surroundings as a central place in this period (Nicolay 2014, 346–66). Merovingian wheel-made pottery was probably obtained from a largely commercial trading network that connected the Frankish world to the north. The hand-made pottery, both before and after the fourth-century habitation hiatus, does not have the typical traits of mass-produced wares. It was not standardized, although a limited number of different types can be identified (for the widely used typology of the pottery of the Iron Age and the Roman Period in the northern Netherlands, see Taayke 1996). Each pot is different and hand-made cooking and storage vessels are large and heavy. Small pots are less heavy, but often decorated and far from homogeneous. That indicates that hand-made pottery was not commercially made in workshops, but at home, for the personal use of a household. While workshop-produced pottery is often made by men, household-produced pottery in a subsistence economy is usually made by women, as ethnographic studies show (Abbink 1999, 39; Sinopoli 1991, 100; Rice 1987, 184). That gender-related production mode may apply to the past as well; in the Iron Age and Roman Period subsistence economy of the terp region, hand-made pottery was probably made by women for their 47

Z01 Hines Book B.indb 47

07/05/2021 13:29

Frisians of the Early Middle Ages own households. If that assumption is accepted as a working hypothesis, it becomes possible to explain various features of the distribution of pots and pottery types. If household production was indeed normal procedure, and potters learned the craft from older generations and adopted new ideas (techniques and shapes) from their social environment, we may expect local and regional similarities, gradually changing towards other regions. That is indeed the case. The typological study by Taayke (1996) has shown that there are considerable similarities but also identifiable differences between his four sample areas: Westergo and Oostergo in Friesland, central Groningen and northern Drenthe (Fig. 3.1). However, we do not find only locally produced pottery in settlements. In every excavated settlement, a small percentage of the excavated pottery comes from elsewhere (Nieuwhof 2017b). The pottery assemblage of a terp settlement such as Ezinge (Groningen) includes a small amount of pottery from the western coastal area of the Netherlands, from Friesland, and from Drenthe or further to the south, the area of the so-called Rhine-Weser-Germanic (RWG) Ware. There is also pottery from the east, from areas in Lower Saxony. Such foreign pottery is the tangible remnant of interregional relationships. Foreign pottery is found in various forms: as pots and as potsherds, and as locally made but foreign types. Complete, non-locally made foreign pots are usually of the

Fig. 3.1  Areas with different pottery styles, with Taayke’s sample areas: 1. Westergo; 2. Oostergo; 3. central Groningen; 4. northern Drenthe. purple: Pottery of Friesland and North-Holland (‘Frisian pottery’); orange: Nordseeküstennahe Fundgruppe (formerly ‘Chaucian’); green: Rhine-Weser-Germanic pottery; blue: Stade-Harburger Gruppe. Lefthand map: northern Netherlands, after Taayke 1996, V, Abb. 8; right-hand map: northwestern Germany, after Schmid 2006, Abb. 5.

48

Z01 Hines Book B.indb 48

07/05/2021 13:29

Pottery and Social Relations smaller types, often decorated drinking cups. In contrast to the larger cooking and storage vessels, these were easy to transport and may have been obtained as gifts from visitors, or when paying visits. In the first century AD, the pottery of the north-west German coastal area (‘Chaucian’ pottery) belongs to the most easily recognizable pottery. Small pots in this typical style are found in many places that are accessible over water, for instance in the castellum of Valkenburg and in the settlement of Rijswijk, both in the present province of South-Holland (Taayke 1996, V, footnote 59). Such finds have been associated with Chaucian piracy (e.g. Halbertsma 1953, 244, Afb. 39a.2), but mutual visits during which gifts were exchanged is a far more likely explanation of the occurrence of complete small pots in faraway places. Foreign pottery is often also found as sherds, just like indigenous settlement ceramics. That will often be the result of accidental breakage and the following dispersal of sherds. However, there are indications that pottery was sometimes broken deliberately in a ritual context, to divide the fragments among those present, as memorabilia of certain events or agreements (Nieuwhof 2015, 100, 227). This practice of fragmentation and enchainment also connected the participants symbolically (Chapman 2000; Chapman and Gaydarska 2007). One indication of this practice is that single foreign sherds are often found in much younger contexts, in finds assemblages that can be interpreted as ritual deposits. That also applies to objects that may have been obtained as gifts. That implies that these objects were kept for a long time in what I would like to call a family archive prior to being deposited, often as part of household rituals that were related to family identity (Nieuwhof 2017b). Such family archives can be envisaged as collections of objects (e.g. heirlooms, gifts, even ancestral bones) that were related to a household’s history and identity and that served as memorabilia (Nieuwhof 2015, 292). Story-telling must have brought the events during which objects were obtained and the people who were involved back into memory. A third social mechanism that results in foreign-looking pottery in settlements is the dispersal of the potters themselves, who made pottery in their own style but in local raw materials in a new place of residence. That means that the fabrics are similar to local fabrics, but shapes or decorations are different. If women made the pottery for their own use and households, such finds represent women from elsewhere who married local men and moved in with them, and thus indicate a virilocal marriage system. In their new homes, women probably kept producing pottery in their own style for some time, but also adopted new elements from their new environment and at the same time influenced the local style. An intensive marriage network with virilocality, combined with home production of pottery by women, explains the often quick spread of stylistic features of pottery. A virilocal marriage system also may explain the very rare cremation graves that are found in the terp region during the Roman Period, a period in which mortuary practices were variable but did not normally include cremation. That suggests that the origins of women influenced the way they were treated after death (Nieuwhof 2015, 262). It also implies a degree of respect for the individual identity of these women. Relationships that are accompanied by the exchange of technological knowledge (e.g. the culverts with tidal flapper valves), gifts and marriage partners are indicative of socio-cultural networks, which influenced the material culture. The origin of different styles of foreign pottery suggests that each settlement could participate in several, partly overlapping networks, but had stronger ties and more frequent exchange with 49

Z01 Hines Book B.indb 49

07/05/2021 13:29

Frisians of the Early Middle Ages some than with others. Settlements within networks shared their material culture to some degree. Similarities in material culture are indicative of social networks, but not necessarily of ethnic identity. Similarities must to a high degree be related to the frequency of the exchange and the strength of the relationships.

The Roman Period Changing networks This article focuses especially on eastern and western connections, since these play an important role in the habitation history of the Frisian area. That is not to say that there were no contacts with southern neighbours. An influence from the southern RWG area on the pottery of Oostergo/Groningen/northern Drenthe and the coastal areas of Germany is unmistakable (Taayke 1996, V, 175–7; 2013). Conversely, northern pottery (from Groningen/northern Drenthe/Lower Saxony) seems to have influenced RWG pottery, and northern beakers have been found in the eastern Netherlands (Taayke 2006, 210). ‘Frisian’ pottery occurs in Roman-period settlements in the central part of the Netherlands (Van den Broeke 2018), but the social mechanisms as described above, which involved the exchange of pottery, may not have played a role in this case. Frisian pottery was possibly taken to the central Netherlands by Frisian soldiers and their families, or perhaps, as Van den Broeke (ibid.) has suggested, as containers of Frisian products. The location of the boundary between the pre-Migration Period Frisians and their eastern neighbours, the Chauci, is not certain. Ptolemy mentions the River Ems (Geographia 2.11.7) but it was perhaps the River Lauwers, between the present provinces of Friesland and Groningen, that was in fact the boundary. That is because considerable changes occurred around the beginning of the Roman Period. During the Iron Age, pottery styles were so similar in the northern and western parts of the Netherlands that it is only rarely possible to identify pottery of a different

Fig. 3.2  Small pots of the first century AD from terps in the province of Groningen, all in the new, ‘Chaucian’ style. The pot from Englum is typical of the regional Wierum style (type K3, Taayke 1996, III). Both pots from Ezinge are probably of north-west German origin. Collection Noordelijk Archeologisch Depot, Nuis. Drawings and photograph by the author.

50

Z01 Hines Book B.indb 50

07/05/2021 13:29

Pottery and Social Relations geographic origin. Around the beginning of the first century AD, however, a new pottery style was introduced in Groningen and northern Drenthe, highly similar to that of the neighbouring coastal areas of Lower Saxony. Avoiding the ethnic name Chaucian, Taayke called this new style the Wierum-style (Fig. 3.2), after one of the terp settlements in Groningen (Taayke 1996, V, 175). The pottery was not the only thing that changed. The number of settlements in the sparsely populated north-eastern part of Groningen increased from eight to forty-three; in the western part it doubled (Taayke 1996, III, 68). Taayke argues for Chaucian immigration to explain these changes (Taayke 1996, V, 191). If that was indeed the case, the influx must have been peaceful. The new style was accepted quickly in existing settlements, but the change was not abrupt, as the pottery in terp settlements such as Englum and Ezinge indicates (Nieuwhof 2008a; 2014). Pottery fabrics remained the same; older pottery remained in use until well into the first century AD and is often associated with the new Wierum-style pottery. Older linear (streepband) decoration was also applied to Wierum-style pottery. And while most of the Wierum-style pottery was undoubtedly made locally, there were still gifts of nice small pots, which stand out because they are slightly different from locally made small pots and must come from the German coastal area (Fig. 3.2). Whether the changes were induced by Chaucian immigration or by cultural influences combined with natural population growth (another possible scenario), in both cases the result was that Groningen and northern Drenthe seem from the first century to be focused on their eastern neighbours. Friesland and North-Holland kept sharing their pottery tradition, which suggests that these regions were still closely connected culturally and socially. Still, finds of ‘Frisian’ pottery in Groningen and Lower Saxony (Nieuwhof 2017b; Schmid 2006, 76 and 78) show that relationships between communities in these areas did not come to an end. Whether the inhabitants of Groningen still considered themselves Frisians after this change is a question that cannot be answered. During the second century AD, the pottery of the ‘Frisian’ and the ‘Chaucian’ traditions developed separately. The Groningen and northern Drenthe areas became part of the nordseeküstennahe Fundgruppe, a term introduced by Von Uslar (1977), with similar or related forms and decorations in the entire coastal area up to the Eider in Schleswig-Holstein (Fig. 3.1). This must have been a strong socio-cultural network that was maintained through intensive contacts. In the third century, when many terp settlements in Friesland and Groningen had already been abandoned, the remaining settlements in Friesland and North-Holland seem to have joined the eastern network. They adopted the pottery with the characteristic third-century long rims from the nordseeküstennahe Fundgruppe, a style that is known as Driesum Ware in the northern Netherlands (Figs. 3.3–4; Taayke 1996, V, 180). Contacts with the eastern network may also be inferred from a find from Wijnaldum. A very early small pot in the so-called Anglo-Saxon style2 at Wijnaldum-Tjitsma, which had been used as a cremation urn for a woman and a child (Cuijpers et al. 1999; Nieuwhof 2015, 239–40 and App. C. 95.b), must date from this period, before the habitation hiatus.3 This may be one of the cases 2 I prefer to refer to ‘Anglo-Saxon style’ rather than ‘Anglo-Saxon’ pottery, because it avoids ethnic ascription. 3 Two radiocarbon dates of cremated bone (GrA-44595 and GrA-45845) resulted in statistically consistent ages of 1780±35BP and 1795±35BP, producing a combined age of 1787 ± 25 BP (Lanting and Van der Plicht 2012, 305), which gives a date of cal AD 136–329 (2 σ: 95% probability).

51

Z01 Hines Book B.indb 51

07/05/2021 13:29

Frisians of the Early Middle Ages

Fig. 3.3  Part of one of the excavated levels in the terp of Ezinge, excavated in the 1920s and 1930s. Only one, incompletely excavated, building plan dates to the fourth century (grey). The small rectangular features are sunken huts, most of which belong to higher levels. Numbers refer in some cases to elevation (m +NAP, Dutch Ordnance Datum), although some (with +) are finds numbers, including pottery. The pottery is depicted in Figure 3.4.

in which origin influenced the type of mortuary ritual, so just like the pot, the woman may have come from north-western or northern Germany. Changing society The Roman Period is not only a period of changing contacts but also of social change, caused by internal processes as well as the influence of the Roman Empire. During the Iron Age, settlements consisted of households of more or less equal status and wealth, as can be inferred from settlement finds. Society was composed of families and households whose status was defined by their ancestry and family identity, and leadership was probably chosen, based on merit (Nieuwhof 2015, 288). The ancestors were an important extension to the social environment, providing depth to family identities. They were considered to have become supernatural beings, as can be inferred from offerings near deposited human remains (Nieuwhof 2015, 142, 226). Mortuary rituals were associated with these beliefs and practices; they consisted of rare inhumation burials near and in houses and, more often, of excarnation with the aid of scavengers, and subsequent collection of the remaining bones (Nieuwhof 2015, 229–84). 52

Z01 Hines Book B.indb 52

07/05/2021 13:29

Fig. 3.4  Pottery from the fourth-century transition period in Ezinge. Driesum-style pots (Dr) occur together with pottery derived from that style (G7 in Taayke’s 1996 typology of northern Drenthe) and with Anglo-Saxon-style pottery (AS). Collection Noordelijk Archeologisch Depot, Nuis. Pottery drawings by the author.

Z01 Hines Book B.indb 53

07/05/2021 13:29

Frisians of the Early Middle Ages The growing population of the terp region, culminating in a peak around the beginning of the first century AD in many settlements,4 resulted in a degree of competition that was expressed in ritual practice. Household or family rituals, in which pottery was destroyed and objects from the family archive, especially worked or unworked human bones, were deposited near the houses or in boundary ditches, emphasized family identity and territorial claims (Nieuwhof 2015, 295–6). Some families must have been more successful in this competition than others and it thus laid the foundation for the emergence of elite families. We may assume that, from then on, leaders often came from these families. These were the leaders that Roman diplomacy was aimed at and who thus acquired Roman luxury goods (Erdrich 2001, 34), resulting in a further increase of social inequality. The presence of returned veterans of the Roman army may have contributed to the development of hereditary leadership. They undoubtedly brought extra prestige to their families, and they could boast an experience and knowledge of the wider world that was unmatched by their co-residents. Rituals now turned literally inwards, inside the houses, and the practice of depositing human remains seems to have ended in the course of the Roman Period. Ritual deposits in houses often consisted of small pots, which must have served as containers for food. An increasing standardization in ritual deposits suggests that the identity of individual families became less important in this period. The veneration of family ancestors may have given way to the veneration of ancestors with a more collective position, possibly those of the leading families (Nieuwhof 2015, 223–6, 296). Mortuary customs also changed. Inhumation and secondary deposition of human remains was associated with individual households until into the second century. From around AD 200, however, there seems to be a tendency towards clustering of graves, though still near individual houses. The sparse evidence suggests that small household cemeteries, with inhumations and/or cremations, came into existence in the third century. Such early cemeteries have been found in North-Holland at Castricum-Oosterbuurt (Hagers and Sier 1999, 85, 187–97) and Schagen-Muggenburg (Hagers and Sier 1999, 86), in Drenthe at Wijster (Van Es 1967, 409–521) and Midlaren-De Bloemert (Tuin 2008, 531–9) and in Lower Saxony at Dingen and Barward (Plettke 1940; Genrich 1941). In Ezinge, a pair of graves in the farmyard of one of the houses perhaps forebodes this new trend (Nieuwhof 2015, 405–6). This pair of graves dates from the third century AD. One of these burials is one of the very few graves with a grave gift in the terp region: a small, decorated, non-local pot from the first century AD, probably once received as a gift and kept in the family archive for a long time, had been placed near the feet of the deceased (Fig. 3.2, 1176).

The fourth-century transitional phase Joining the eastern network in the third century could not prevent the full abandonment of the western part of Friesland, Westergo, and the virtual abandonment of Oostergo. North-Holland was also partly abandoned (Dijkstra and De Koning 2017). We may assume the ties with the eastern network were not yet strong enough to provide sufficient social cohesion when natural circumstances became less favourable. In 4

Based on the amount of pottery per period, e.g. Bakker and Varwijk 2016, 200–1; Nieuwhof 2008a, 63; 2014, 114–16.

54

Z01 Hines Book B.indb 54

07/05/2021 13:29

Pottery and Social Relations

Fig. 3.5  Annular brooches with backward-looking animal heads from the northern Netherlands and north-western Germany dated to the fourth or early fifth century. 1. Midlum-Northum (Kr. Wesermünde); 2. Oostum; 3. Unknown terp in Friesland; 4. Ezinge (only drawing available). Scale 1:1. Photographs after Zimmermann 1972; drawing: University of Groningen, Groningen Institute of Archaeology.

Groningen, at least in the settlement of Ezinge that was not abandoned, Driesum-style pottery developed under the strong influence of the eastern network into the expressively decorated pottery in ‘Anglo-Saxon’ style (Figs. 3–4), in the same way as in, for instance, the Feddersen Wierde in the coastal area between Weser and Elbe, and inland Midlaren in northern Drenthe (Nieuwhof 2008b; 2011; 2013). Midlaren as well as Ezinge were accessible via the River Hunze. As before, relationships may have entailed regular bilateral visits, the exchange of marriage partners and gift exchange of small pots. 55

Z01 Hines Book B.indb 55

07/05/2021 13:29

Frisians of the Early Middle Ages Other fourth-century objects such as brooches are rare but perhaps not completely absent. A small group of annular brooches with opposed animal heads, of which we know only four specimens (Fig. 3.5), may belong to this period or to the first phase of immigration. One of these comes from Midlum-Northum in Lower Saxony, one from a layer with third- to fifth-century pottery in Ezinge (Knol 2014, 201), one from an unknown terp in Friesland and one from Oostum, a terp not far from Ezinge that may also have had continuous habitation. The typical animal heads, which are thought to be inspired by Roman design (Zimmermann 1972), are also known from equal-armed brooches of the Sahlenburg type from the Elbe-Weser region (Böhme 2003, 256–9). Zimmermann dates this type of brooch (his Form VII) to the fourth or early fifth century.5 It is curious that the indigenous, but eastern-inspired pottery development seems to have passed Ostfriesland, a region with coastal Wurten as well as inland Geest-settlements. It was certainly part of the eastern network during the Roman Period, as its pottery from the nordseeküstennahe Fundgruppe clearly shows (Fig. 3.1). The question of continuity in Ostfriesland is concerned with the fifth rather than with the fourth century. Bärenfänger (2001) has already pointed out that there is a considerable research gap, and that previous claims of discontinuity are not to be trusted. Most Roman-period settlements in Ostfriesland seem to have remained inhabited into the fifth century, but were then abandoned (Knol 1993, tab. 6). Anglo-Saxon style pottery (in Germany: völkerwanderungszeitliche verzierte Ware), which is relatively abundant in the fourth- and fifth-century settlements of the northern Netherlands, seems to be less common in Ostfriesland, and an indigenous development towards the Anglo-Saxon style cannot be established. Saxon-style beakers (as they are called in Ostfriesland), which sometimes occur,6 are interpreted as indications of a growing influence from the Elbe-Weser region (Bärenfänger et al. 2013, 166). For now, we cannot but conclude that contacts between the settlements in the northern Netherlands and the eastern network in this period did not go via Ostfriesland, gradually spreading from a centre in the east to the periphery in the northern Netherlands. There rather seems to have been a direct connection with the Elbe-Weser region and possibly areas even more to the north.

New Frisians Repopulation New inhabitants arrived around AD 400. They chose deserted terps for their new settlements (Fig. 3.6). The earliest radiocarbon dates are from cemeteries in Oostergo along the Middelzee and the northern coast (Beetgum-Besseburen, Hogebeintum, Oosterbeintum: Knol 1993, Tab. 9; Lanting and Van der Plicht 2010, 142; 2012, 376; Nieuwhof 2015, 234–43). Archaeologically dated finds indicate that immigrants, at the 5

6

The annular brooch from Hallum mentioned by Fowler (1963, fig. 5) is slightly different in type, although Fowler classifies this brooch and the one from an unknown terp in Friesland as both belonging to the same type, H4 (ibid. 144). The animal heads on the Hallum brooch are less naturalistic, so this specimen may be later. A small inventory that H. Prison (Ostfrisieschische Landschaft) was kind enough to send me includes only a very small number of fourth-century Trichterpokale and Anglo-Saxon style pots (pers. comm., 13 January 2019). The latter are represented by the dots along the Ems in Figure 3.6.

56

Z01 Hines Book B.indb 56

07/05/2021 13:29

Pottery and Social Relations

Fig. 3.6  The distribution of Anglo-Saxon style pottery in the northern Netherlands, against the background of a palaeogeographic map (the status quo c. AD 800) and the contours of the modern geography. 1. Wijnaldum; 2. Beetgum; 3. Hogebeintum; 4. Oosterbeintum; 5. Aalsum; 6. Englum; 7. Ezinge; 8. Oostum; 9. Eelde; 10. Midlaren. Map by P. C. Vos and S. de Vries, Deltares; data from Knol 1993, fig. 57, updated.

same time or soon after, also settled on empty terps near Ezinge, such as Aalsum (Knol 1993, Tabel 9) and Englum (Nieuwhof 2016, fig. 4). As before, watercourses were still the major transport routes. Vos and Knol (2015) suspect that a canal was dug in this region between the Hunze and the developing Lauwerszee in the west to solve problems with drainage; it may have been these new settlers, together with the people from still inhabited terps such as Ezinge, who dug this canal. The attempt eventually and probably unintentionally resulted in the development of a wide, meandering river, the present Reitdiep.7 The locations along the Middelzee and near Ezinge were the regions that had remained inhabited, possibly or certainly, in the fourth century. Since the remaining population had stayed in contact with the eastern network, as can be inferred from the developing pottery style, these may already have been familiar areas to the newcomers, known from previous visits and possibly connected by marriages. Terps in Westergo seem to have been repopulated slightly later, in the first half of the fifth century (e.g. Nicolay 2014, fig. 4.21). The terp of Wijnaldum-Tjitsma, for instance, is thought to be occupied again around AD 425 (Gerrets and De Koning 1999, 103). The new population of the terp region grew slowly. The number of settlements from the Migration Period is smaller than settlements from the Merovingian Period, and the number of settlements from the Carolingian Period is still larger (Fig. 3.6; Knol 1993, 107–9). 7

Several more canals may be at the basis of present rivers. A similar canal was probably dug between the southern part of the Middelzee and the Marne in the west, resulting in considerable erosion and enlargement of the Middelzee. Digging of these canals cannot be precisely dated (Vos and Knol 2015).

57

Z01 Hines Book B.indb 57

07/05/2021 13:29

Frisians of the Early Middle Ages It is surprising that North-Holland seems to have been shunned by the migrants of the first part of the fifth century, despite the short distance between Westergo and North-Holland that had never hindered the people of the Iron Age and the Roman Period. Anglo-Saxon style pottery is entirely missing in North-Holland. We know of only a very small number of late Anglo-Saxon style sherds, dated no earlier than around AD 500, from Texel (De Koning 2018, 150–1). The reason may be that the migrants were not looking for just any deserted lands to settle. Their priority must have been to build a new, tight-knit society based on existing connections of old. The elite networks and regional kingdoms that developed in the Early Middle Ages (Nicolay 2014, 346–66; 2017) are thus rooted in conglomerations of early new settlements. Origins The new population of the coastal areas must have been diverse. It combined immigrants from the northeast (probably the majority) with the descendants of the very small, remaining population of the fourth century and possibly of people from inland Drenthe, where a settlement like Eelde-Grote Veen was abandoned around AD 400 (Jelsma 2015, 148). That the new settlers chose deserted terps for their new settlements indicates that they were familiar with living on terps and came from a coastal region in the German counterpart of the terp region: the coastal areas of Lower Saxony and of Schleswig-Holstein. A significant population decline seems to have occurred in these areas, possibly first in Schleswig-Holstein and then, from the fifth century, also in the Elbe-Weser triangle (Nösler and Wolters 2009; Gerrets 2010, 184). Feddersen Wierde, for instance, was abandoned around the middle of the fifth century (Schmid 2006); in settlements on the North Frisian islands, finds from the period between the early sixth century and the end of the seventh century appear to be missing (Majchczack, this vol., ‘Discussion’). During the fifth and sixth centuries, there may have been several phases of immigration into the northern Netherlands, with different areas of origin also further to the north, into Jutland. It is nearly impossible to disentangle these phases and to pinpoint a specific area of origin for the first or for later settlers. Moreover, all these areas had participated in socio-cultural networks long before migration began around AD 400, just like the northern Netherlands, and the material culture the migrants took with them when they left their homelands was thus already an amalgam of many different influences. That is certainly the case with pottery. The pottery of the newcomers often cannot be distinguished from the locally developed pottery, not even on the basis of fabrics (Krol et al. 2018). Stylistic elements might provide, at least in theory, a clue to their origins. Close parallels of shapes and decorative motifs in the northern Netherlands can be found in any of the large cemeteries and settlements in the Elbe-Weser triangle or further to the north, in Schleswig-Holstein: the cemeteries of, e.g., Westerwanna (Von Quillfeldt and Roggenbuck 1985), Flögeln (Schön 1988), and Issendorf (Häßler 1994; Weber 1998); and settlements such as the Feddersen Wierde (Schmid 2006), Loxstedt (Nösler 2017), and Tinnum on the island of Sylt (Scholtz 2015). That means that the origins of immigrants cannot easily be established from stylistic parallels. Quantitative and statistical comparison of pottery assemblages (see Weber 1998) might be helpful, but the assemblages from the large cemeteries and few settlements in the areas of origin cannot simply be compared with the settlement material and much smaller cemeteries of the northern Netherlands. A further complication is that the 58

Z01 Hines Book B.indb 58

07/05/2021 13:29

Pottery and Social Relations development and chronology of Anglo-Saxon style pottery during the two centuries of its existence is not very well established, and some of the variation between geographically different groups undoubtedly relates to chronological factors (Nielsen 2003, 203). Also on the basis of other material categories, a one-to-one relationship with a location or an area of origin of the newcomers in the terp region of the northern Netherlands is impossible to establish. Brooches point in the direction of the Elbe-Weser triangle (Nicolay, this vol.). However, the turf houses that are associated with the new Frisians do not come from that area; timber buildings were the rule there (Gerrets 2010, 174–5). A Migration-period turf house and structures with turf walls were excavated on the North Frisian island of Sylt (Scholz 2015; Majchczack and Segschneider 2015, 128), but what the relationship is between these houses and the turf houses in the coastal area of the northern Netherlands is as yet unknown. This conflicting evidence might represent two fifth-century phases, an early one from the Elbe-Weser triangle and a later one from the coastal area of north-western Schleswig or Jutland. The houses that belong to the early phase are perhaps yet to be uncovered.

The new society As we have seen, society became more stratified already during the Roman Period. The new communities continued this development, with a more stratified social hierarchy, leaders that were in a strong patron-client relationship with their retinues (Nicolay 2014, 6–11), and an increasing division of labour. More products were made by specialists, no longer for their own households, but for an employer or a (small) market. The direct relationship between burials and households had disappeared. Cemeteries were no longer associated with households, but with settlements as a whole (Knol 2011). That process is undoubtedly linked to the changes in social structure. Family archives probably no longer included ancestral bones, but a new element may have become part of these archives: objects that were obtained as gifts, with runic inscriptions. Looijenga (this vol.) mentions several objects with inscriptions that might belong to the domain of gift exchange, such as a comb from Oostum with an inscription that translates as ‘Habuku made the comb for Aib’, or a comb case from Ferwerd, with the inscription me uræ or em uræ; in translation: ‘I am from/of Ura’. These combs may have been used as combs, but also were part of the family archive as described above. These runic inscriptions were a reminder of the occasion and the giver and served as clues when telling the associated stories. Brides still moved in with their husbands in this period, possibly also across the North Sea; nice pots still served as gifts, decorative motifs spread over large areas, and the heterogeneous Anglo-Saxon style pottery was probably still made by women (Nieuwhof 2017b). That may have changed at the end of the fifth century, when the often abundantly decorated and carefully finished pottery in Anglo-Saxon style was replaced by the coarse and plain Hessens-Schortens Ware that would be the common hand-made pottery of the following centuries.8 This new pottery is no longer useful for 8

‘Hessens-Schortens Ware’ is the designation introduced by Tischler (1956, 87). This coarse ware developed around AD 500 from undecorated types of the fourth and fifth centuries (Drenthe-G7; Taayke (1996, II)). Bayesian modelling of a large number of radiocarbon dates confirms this date (Krol et al. 2020). In Germany, this ware, although dated later (Stilke 2001), is known as weiche Grauware. A

59

Z01 Hines Book B.indb 59

07/05/2021 13:29

Frisians of the Early Middle Ages

Fig. 3.7  Silver ‘hand pin’ of British origin, found in a terp near Hallum and dated to the second part of the fourth or the fifth century. Scale 1:1. Drawing from Knol 1993, fig. 62.

tracing contacts between communities and areas. While many of the Anglo-Saxon style decorations may have had symbolical meaning, the Hessens-Schortens Ware, which is found in all the new residential areas of the Anglo-Saxons from around AD 500, did not communicate much. Pottery was clearly no longer a medium to express oneself, or to express symbolic meaning. The new population was an amalgam of people of different origins, but they were no total strangers since they had been participating in the same socio-cultural plant-tempered variety of Hessens-Schortens Ware in the northern Netherlands is known as Tritsum Ware (Taayke and Knol 1992), in England as chaff or grass-tempered ware. This pottery (and possibly Tritsum Ware too) is actually dung-tempered (Hamerow, Hollevoet and Vince 1994).

60

Z01 Hines Book B.indb 60

07/05/2021 13:29

Pottery and Social Relations

Fig. 3.8  Three sherds, allegedly of Oxfordshire Ware (Fulford 1977), dated to the late third and fourth centuries, from Groningen terps. Top: Ezinge 1925-VIII.132 (the number is incorrect); middle: Oostum 1913-VI.11; bottom: Brillerij 1926-VI.96 (length 85 mm). The hole may have been a repair hole before the sherd was used as a pendant. Collection Noordelijk Archeologisch Depot, Nuis. Photographs by the author.

networks for centuries. The new Frisians kept in contact with their areas of origin, and also with settlers in England. That is not only apparent from the very similar Anglo-Saxon style pottery in all of these regions (for England, see, for instance, the cemetery of Spong Hill: Hills and Lucy 2013), but also from several unusual finds. The most conspicuous is a silver hand pin from a terp near Hallum (Fig. 3.7), the most eastern find of this type, which originated in northern Britain and is currently dated from the mid-fourth to the fifth centuries AD (Fowler 1963, 126–7; IJssennagger 2014; Nicolay 2014, 93). There is also a horizon of late Romano-British sherds in the northern Netherlands and along the coast further to the east, which, according to Morris (2015, 428), may point to contacts around AD 400 (Fig. 3.8).9 There may have been some exchange with the British population during visits to Britain, or these sherds were pick-ups, taken as souvenirs. Samian Ware and similar red fabrics had been popular as a raw material for centuries (Nieuwhof and Volkers 2015). Another pick-up might be a peculiar find from Ezinge: a late first- or second-century copper-alloy cockerel with enamelled chest of a type that was probably made in Britain (Knol 2014; 9

A quick scan of these sherds in the Northern Archaeological Depot in Nuis with terra sigillata (Samian) expert Mrs T. B. Volkers showed that identification is difficult; this category merits further research.

61

Z01 Hines Book B.indb 61

07/05/2021 13:29

Frisians of the Early Middle Ages

Fig. 3.9  Finds from a fifth-century sunken hut in the terp of Ezinge, probably a ritual deposit (finds no. 698). The assemblage consists of a copper-alloy cockerel (inv. no. 1931–05–0698, originally enamelled, of British origin, and dated to the first or second century), a bone needle and Anglo-Saxon style pottery. Pottery: Collection Noordelijk Archeologisch Depot, Nuis, drawings and photographs by the author; photograph of the bone needle: W. Prummel; photograph of the cockerel: M. de Leeuw, Groningen Museum.

Morris 2015, 422; Hoss et al. 2015). This small object does not come from a contemporaneous context but from a fifth-century sunken hut, probably as part of a ritual deposit together with Anglo-Saxon style pottery (Fig. 3.9). Following the studies of Nicolay (2014; 2017; this vol.), we know that also luxury metal objects were exchanged among North Sea elites, and that there were ideological influences which show from the decorations on metalwork. The new population came to be known as Frisians again, just like in the Roman Period, but we hear of that name only in seventh century written sources. Flierman (this vol.) notes that this coincides with the appearance of supraregional identities. Did the new Frisians adopt that name themselves, or was it ascribed to them by others? Bazelmans (2000, 54–61) argues that it was the Frankish elite that introduced the old name, which was still known to Frankish scholars who had access to the works of classical authors. Alternatively, the names of populations may have been connected to regions rather than to ethnic identity. If that is the case, a contemporary equivalent of Friesland may have been remembered within the entire North Sea network, and the new population of this land naturally called themselves, and were called, Frisians.

Conclusions Pottery is the most common material category in excavations of later prehistoric to medieval sites. Most of the pottery in the northern Netherlands and north-western Germany was hand-made, and homemade for private use. Hand-made pottery is heterogeneous; it shows an enormous variety of shapes, fabrics and decorations. Geographic and chronological differences and similarities allow us to distinguish different styles, types and developments. Changes are usually phrased in terms of one style influencing another, but in reality, of course, it was the potters who were influenced by other potters and by pots within their reach. Areas with similar styles must represent socio-cultural networks, in which technological knowledge and pots were exchanged, and in which people moved. If we accept the hypothesis that pottery was made by women for their 62

Z01 Hines Book B.indb 62

07/05/2021 13:29

Pottery and Social Relations own households (a model that is supported by ethnological evidence from pre-modern societies), stylistic similarities must indicate contacts and socio-cultural networks. Within these networks, it was common practice for women to move in with their husbands when they married, not only in their immediate environment, but also over longer distances. That explains the often quick spread of stylistic elements over large areas, and foreign types of pottery in local fabrics within settlements. Following the pottery, the inhabitants of the present provinces of Friesland and Groningen, the ‘old Frisians’, participated in networks that extended to the west, the south and the east already before the Roman Period. Friesland was oriented more to the west, to North-Holland, while Groningen was from the beginning of the first century AD part of a network that also included northern Drenthe and north-west Germany. Friesland and North-Holland seem to have joined the eastern network in the third century, but the new ties were not strong enough to prevent the abandonment of these regions. During the fourth century, this was a virtually deserted area, with the exception of a very small number of places in Oostergo and North-Holland. The population of Groningen also decreased considerably, but here, the strong eastern socio-cultural network served as a safety net when problems with drainage became intolerable in many areas. Some settlements in Groningen remained inhabited across the fourth century. Pottery developed during that period in the same way as elsewhere in the eastern network, into what is traditionally known as Anglo-Saxon pottery, but is better called Anglo-Saxon style pottery. The area was re-inhabited from around AD 400 onwards. The new population was composed of people of different origins: remaining inhabitants from Groningen and perhaps Oostergo, small groups from abandoned settlements in northern Drenthe and along the Ems, and newcomers from more eastern areas, probably coastal areas of Lower Saxony, especially between the Weser and Ems, and further north, of ­Schleswig-Holstein or even Jutland. Although of different origins, these people were no total strangers. They had been part of the same socio-cultural networks for centuries, had been visiting each other frequently, were related through marriages, and shared more or less the same material culture and way of life. They came to be known as the Frisians, just like their predecessors of the Roman Period, possibly because the name was connected to this area, and had survived within the eastern network. The ‘new Frisians’ filled their new land gradually, but they did not forget their background. They kept in touch with other migrants from their homelands. The old eastern network did not disintegrate, but widened eventually to encompass the entire southern North Sea area.

Acknowledgements I am grateful to prof. dr. Haio Zimmermann for the use of his photographs, to Amy Kuiper and Jelle Schokker of the Northern Archaeological Depot in Nuis for looking up late-Romano-British sherds, and to Mrs Tineke Volkers for examining these sherds. I also thank the editors and authors of this volume for a very fruitful exchange and discussion of ideas and data.

63

Z01 Hines Book B.indb 63

07/05/2021 13:29

Frisians of the Early Middle Ages

Discussion KNOL I have a question about the terra sigillata. I understood that in Ezinge too some terra sigillata has been found that originates in Tunisia. Is that different? NIEUWHOF Yes, it’s African Red Slip Ware, from around AD 400. I think that may have come here with mercenary soldiers who had served in the Roman army in the early fifth century. So it’s younger than Oxfordshire Ware. KNOL That is also very unusual. NIEUWHOF The whole range of finds from the fourth century in Ezinge is a strange collection of both very familiar and very unusual types. The population was very sparse at the time across the terp region, so there’s little to compare with. KNOL What do you know about the terp layers above the house? Was that one of the first structures to be found, or does it belong to a deeper level? NIEUWHOF There were two excavated levels above the one that had the fourthcentury house, and then the topsoil. Above this house, several younger sunken huts, with Anglo-Saxon style pottery, were excavated. That is why I am quite certain that the pottery that I have shown (Fig. 3.4) belongs to the fourth century. IJSSENNAGGER-VAN DER PLUIJM I was struck by the penannular ring-brooches you showed, as I once wrote a very short article trying to list similar finds from Friesland (IJssennagger 2014). There is quite a nice typology of these finds which can be helpful to also see how they relate to one another. I think there are related finds from Hallum. They may be a bit later, from a different stage in the chronology, but they belong to the same typology and I thought they were actually quite similar, probably deriving from the type you showed. NIEUWHOF Fowler (1963) classifies both of these as her type H4, but I think the one from Hallum is more stylized and probably slightly later. Hayo Zimmermann (1972) thinks the brooches with naturalistic, opposed animal heads – and so not the Hallum brooch – are the earliest finds, and that is why he dates them to the fourth and early fifth centuries. IJSSENNAGGER-VAN DER PLUIJM There are also two finds from Ferwerd; again a bit different, but they feature in Fowler’s publication of Celtic metalwork (1963). Fowler lists them in terms of typology. The two examples from Ferwerd, which are more stylized, are also dated to the fourth and fifth centuries there. I don’t know if you are familiar with those examples and what your thoughts are? NIEUWHOF The brooches from Ferwerd belong to Fowler’s type F and she thinks they are from the same workshop as similar British ones. I didn’t realize that they have similar dates to the ones with the animal heads. Of course, these brooches and the one from Hallum all point to contacts between Britain and the northern Netherlands, either in the fourth century or the fifth. NICOLAY The animals are most closely related to the equal-armed brooches. These are now dated to the fifth century. They are certainly related to Saxon-style metalwork from the Elbe-Weser region, so this is another link to that area. KNOL On Anglo-Saxon style pottery: as far as I know, some of the cremations found at Beetgum and at Hogebeintum date from the fifth to the beginning of the sixth century (Knol 1993, fig. 13; Lanting and Van der Plicht 2010, 142). I got the impression that you were assuming an earlier date for Anglo-Saxon pottery.

64

Z01 Hines Book B.indb 64

07/05/2021 13:29

Pottery and Social Relations NIEUWHOF The earliest dates from Beetgum as well as from Hogebeintum are actually from AD 400, or the early fifth century. There are several cremations with two or three radiocarbon dates taken because dates on cremated material are often rather confusing. If you compare them you cannot but conclude that some of these must be very early (Nieuwhof 2015, 238–40). I think these dates represent the earliest repopulation of the terp region. KNOL Radiocarbon dates cannot be precise, they often have long ranges. HINES … and unfortunately that is especially the case in the fifth century, because of a plateau in the calibration curve covering a large part of that century.10 NIEUWHOF That is true, but the multiple sampling can make the dates more reliable and precise. HINES For dating the pottery, it will be interesting to compare in detail the design elements of that pottery in the northern Netherlands with the details used in the seriation recently produced from Spong Hill in Norfolk (Hakenbeck, in Hills and Lucy 2013). This showed that not only could the cremation cemetery be divided into three phases, but the most intense periods of burial were those of the mid-fifth century, c. AD 425–475. The third phase at Spong Hill had many fewer cremation burials, and both chronologically and in terms of location these are directly associated with the earliest inhumations there. After that I have a couple more questions and/or comments. To start with the problem of the term ‘Angelsaksisch’: the Buckelurnen are indeed what we would describe as ‘Saxon’ in character rather than ‘Anglian’. It is generally accepted that we can distinguish those cultural traditions to an extent: the highly moulded, bossed style is typically ‘Saxon’ while forms with horizontal corrugated necks are ‘Anglian’, with parallels in and around Schleswig-Holstein rather than Lower Saxony. Fabric is a key component in ceramic studies too. With the very simple pots that seem to be late ‘Angelsaksisch’, we would say, I think, that such pots could be found practically at any time and anywhere, but date and/or provenance can sometimes be indicated by what has been put into the clay as temper. The major difference in England is that such simple and plain pottery appears in higher proportions as grave goods in inhumation graves and on settlement sites, while decorated pottery is more typically used for cremation urns. That distinction is real, although far from absolute, of course. There is for instance the chaff-tempered ware, with chopped up grass from dung mixed into the clay, to which special attention has been drawn because of the strong parallels with potting in the area of Flanders. NIEUWHOF That is a lot to comment upon. Firstly, the Buckelurnen may well be Saxon, but these account only for a small part of the decorative range, and there are also many stylistic elements that are common in more northern regions, and so may be Anglian or even Jutish. About the fabrics: thin-sectioning has been done (Krol et al. 2018), but there seems to be no difference whatsoever between fabrics from the third and fifth centuries from the terp region, fourth- and fifth-century Ezinge, even inland Midlaren, which was located in a Pleistocene area but had access to marine clay, and the Feddersen Wierde in Lower Saxony; so from the third century to the fifth, the same raw materials were used. The fabrics are not all 10 A new international standard calibration curve issued in August 2020, IntCal20, has reduced this plateau to a significant degree – Eds.

65

Z01 Hines Book B.indb 65

07/05/2021 13:29

Frisians of the Early Middle Ages the same of course, but they show about the same variation within each settlement in this entire period. Daniel Nösler from Stade has looked at the transition from decorated, what we call Anglo-Saxon style pottery to the coarser ware of the younger weiche Grauware, and although finishing and shapes are quite different, and the grit particles get larger, the fabrics remain basically the same again (Nösler and Wolters 2009). So it is very hard to draw any conclusions on origins and relations, continuity even, on the basis of fabrics. NIJDAM How important is this? How many ways are there to make pottery? NIEUWHOF There are quite some ways to do that. You have different clay sources, and you have different tempering materials, materials that were added to the clay for various reasons. HINES These tempers affect the technical capacity of the pot, and particularly how well it can withstand heat. NIEUWHOF A potter can mix the clay, for example, with stone grit or pottery grit (grog), or with some organic material such as chopped grass or dung or chaff. NIJDAM Is this continuity of technique, which must have been handed down? Or how otherwise should I interpret this? HINES One idea is that this is something that just keeps on being done; there is always the very pragmatic view that you just go to your nearest clay source and use it, which I think is too reductive. These folk knew what they were doing for what they wanted to achieve. NIEUWHOF Yes I agree, they were selective. But the same techniques keep recurring in the history of pottery making; they do not necessarily point to continuity. Then to your comment on the undecorfated ware. There was undecorated ware alongside the decorated ware, and these developed from third-century pottery (the Driesum style mentioned in the text: see Fig. 3.4). The shapes are somewhat different from the decorated Anglo-Saxon style ware, but it is also, just like the decorated ware, very nicely finished. That changes, though, at the end of the fifth century. I think that the fifth-century undecorated pottery develops into the later coarser wares (Hessens-Schortens Ware or weiche Grauware). But the fabrics are still highly similar, apart from the coarser grit. And, as you say John, in cemeteries there is a lot of decorated pottery, but we find that too in the settlements, alongside undecorated pottery. KNOL And there is a lot of undecorated pottery in cemeteries. Certainly in Hogebeintum: many of the sixty or so urns there were undecorated, including the earlier ones. MAJCHCZACK I would like to comment on the Anglian and Saxon pottery styles for the North Frisian islands. Especially in the fourth and fifth centuries AD, we find elements of both styles on the islands. Both the Saxon bossed Buckelurnen and the typical Anglian horizontal corrugated styles appear on the local pottery. The pottery from the North Frisian islands shows strong connections to various regions in southern Jutland and the southern North Sea coast throughout the Roman Iron Age and Migration Period, reflecting the islands’ position in a North Sea coastal network. Close connections to the southern North Sea coast, specifically the ElbeWeser triangle, can be seen in the pottery from the third century AD onwards. It seems to me that some sort of migration started in the fourth century, while the habitation of the North Frisian islands seems to end with an emigration in the 66

Z01 Hines Book B.indb 66

07/05/2021 13:29

Pottery and Social Relations fifth to early sixth century AD.11 I think it is highly likely that the North Frisian population of that date took part in the southward migrations along the North Sea coast. NIEUWHOF What about the people in the house at Tinnum. Was that abandoned in the fifth century? MAJCHCZACK The burned-down house at Tinnum on the island of Sylt was abandoned in the fifth century AD. The house contains the complete pottery ensemble of one household (Scholz 2015). It includes both decorated and undecorated pottery, and touches upon two major problems in our research: the question of continuity of occupation of the island in the sixth–seventh centuries as well as that of continuity in development from the Anglo-Saxon style pottery towards the Early-medieval weiche/frühe Grauware, also known as ‘HessensSchortens’ Ware. At this point, it is difficult to determine the development of the pottery during the sixth and seventh centuries. For the North Frisian islands, a settlement hiatus is very likely, since there is no datable find material from this period. But this perception might be skewed by the dating of the pottery. Recent research from the Elbe-Weser triangle (Nösler 2017) presumes continuous development from the decorated Anglo-Saxon pottery to the undecorated and coarse frühe Grauware. In this case it is possible that some North Frisian find material that has previously been dated into the late seventh or eighth centuries, might be somewhat older. This might especially be true for the settlement at Tinnum/Sylt. The excavations revealed settlement phases of the third–fifth centuries and of the eighth–ninth/tenth centuries. But some finds from the younger phase seem to be rather older than the eighth century. NIEUWHOF As early as the sixth century? Is it possible that you have continuity? MAJCHCZACK Yes, there may be younger find material, but it is hard to prove. There are no metal finds or other finds which could be securely dated before the end of the seventh century or after the first quarter of the sixth century. It is not possible to close the gap of this hiatus of roughly 150 years. This situation is still true not only for Tinnum/Sylt, but for all sites on the islands. They indicate a drastic change over the course of the possible hiatus. Furthermore, there are no cemeteries with a combination of Anglo-Saxon style pottery and what is deemed to be Earlymedieval frühe Grauware. NIEUWHOF John, I’m surprised that you can distinguish Anglian and Saxon pottery. HINES Not in the case of every single pot, by any means. But there are certain features that seem to belong to quite distinct traditions (Myres 1969, 21–35). VERSLOOT Are they geographically distinct? HINES Again by no means completely so. But most of the large cremation cemeteries are within the Anglian area; and if we think in terms of culture and identity those could in fact be seen as having a surprisingly large admixture of ‘Saxon’ pottery. But of course we do not apply simply a naïve and direct reading of Bede’s history and what he thought he could claim had been the pattern of settlement in the fifth century. We also know there’s a lot of overlap in terms of what we would consider to be culturally distinctive metalwork. Nevertheless I think it is true that we get 11 For further comments on the stylistic connections of North Frisian pottery in the first–sixth centuries AD, see Majchczack 2018.

67

Z01 Hines Book B.indb 67

07/05/2021 13:29

Frisians of the Early Middle Ages ‘Saxon’-type material occurring further north than we would expect, in contrast to ‘Anglian’-type material occurring in any great density further south. KNOL You are talking about ‘Anglian’ in England, not in Schleswig-Holstein? HINES Certainly; but the prototypes of these styles of pottery lie back in the Schleswig-Holstein area, in the large cremation cemeteries there. KNOL One of the differences between Germany and England and the northern Netherlands is the size of the cemeteries; you have thousands of cremation urns in a cemetery, but we have a few hundred at most. DE LANGEN You talked about marriages and memorabilia in relation to foreign pottery. Can you elaborate a bit more on the distances you think these bridged? NIEUWHOF In Ezinge, for instance, you find foreign sherds from Friesland, from North- and South-Holland (they have similar pottery), from the south, what we call Rhine-Weser-Germanic pottery, and also from areas in Germany although it is not always certain what specific area, from as early as in the Roman Period. So the distances are quite large. VERSLOOT So we are talking distances of 40–50 km or more? DE LANGEN The south of Holland? NIEUWHOF I mean the present province of Zuid-Holland (South-Holland). Firstcentury Chaucian pottery, for instance, is also found in South-Holland in the Roman castellum at Valkenburg and in the native settlement of Rijswijk (Bloemers 1978). When Rijswijk was published, the differences between the early Romanperiod pottery styles from Friesland and Groningen were not yet known (Taayke published them in his 1996 thesis), so Bloemers calls the northern influence ‘Frisian’, referring to Paddepoel in Groningen and Leeuwarden in Friesland (Bloemers 1978, 99–101). His illustrations of the pottery, however, show several clearly Chaucian- or Wierum-style shapes. DE LANGEN  Do you see that as evidence of marriage exchange? NIEUWHOF That is entirely possible for Rijswijk. The find from Valkenburg is not a set of pottery, but just a small, well-finished decorated Chaucian- or Wierum-style pot. I think such small pots were often exchanged as gifts during visits. So I would interpret such finds indications of exchange and friendly visits rather than of marriage – or of raiding as they have also been explained. DE LANGEN  Can you limit the marriage exchanges to certain periods? NIEUWHOF No, it just continues into the Early Middle Ages. Johan’s research on the Early Middle Ages also points to marriage exchange. NICOLAY The brooches support this, especially during the fifth to seventh centuries (e.g. small-long brooches of Domburg type). NIEUWHOF Marriage exchange is continuous from the Iron Age, all along the coast. There are many indications that people along the entire coast from Germany to Flanders maintained contacts and paid each other visits in their dugout boats. DE LANGEN  …to get a wife. NIEUWHOF …to get a wife, or simply to visit other people. For example the sluices of similar type that I describe in my paper, that are found in Vlaardingen as well as in coastal Lower Saxony: these are evidence that these people were in contact and exchanged information and technology. NICOLAY I think that, when you look at the material culture, there are two types of evidence for exchange. There are areas with the same material culture, which 68

Z01 Hines Book B.indb 68

07/05/2021 13:29

Pottery and Social Relations indicates exchange of people and ideas within specific geographical zones. And then there are these ‘exotic’ objects, which come from different areas. Both are evidence of exchange, including marriages, but the distances differ. MOL About these marriage relationships: is there a pattern for such long-distance links from non-western societies? Crafting pots is not, in my view, something you would link with elites, but those are the people you would expect to make longdistance connections. I would not expect that at a lower level of social status. NIEUWHOF From ethnography we know many examples of such marriage relationships. There is a choice between virilocality and uxorilocality: wives moving in with husbands, or the other way around. Both are well attested. MOL But do they occur over long distances? NIEUWHOF You would indeed expect higher-class contacts over longer distances; but we cannot be sure these people would not have made their own pottery. It is an assumption that high-status people have other people to make the things that they need. IJSSENNAGGER-VAN DER PLUIJM That only the elite would travel over the longest distances is also an assumption. NIEUWHOF Indeed. I think that along these coastal areas, visits and exchanges took place at different levels. These were well-connected populations, of exploratory people who were acquainted with the maritime landscape and who liked to visit each other and bring each other gifts. So it may not be limited to people with high social status. VERSLOOT I can imagine that the upper class would also bring their servants with them when they married. And that’s who would make the pottery. Another parallel would appear in the Icelandic sagas, where marriages are recorded over long distances. There was no real aristocracy on Iceland. Linguistically, no dialects emerged because of the level of mobility. NIJDAM You were talking about the interesting family archives, which include things that may be as old as 200 years. Can you tell me more about this phenomenon? NIEUWHOF This is a hypothesis, of course, although I think there is quite some evidence to support it. I first came to think about family archives when I was working on a finds assemblage with several human skulls, that had been deposited together (Nieuwhof 2015). These showed signs of being kept above-ground somewhere: they had lost their teeth and radiocarbon dating showed that they were probably not of the same dates. This leads to the idea of family archives. Not only human, ancestral bones must have been part of such archives, but all kinds of memorabilia, including gifts, that were connected to the identity of the family. These objects must also have been connected to stories, and were the basis for story-telling and for keeping the family history alive. I believe that these objects were curated within families, and that some of them were deposited in family territory, while others, for instance jewellery or beads, were inherited. That practice continues into the Early Middle Ages. NIJDAM How formalized would this have been? The practice that you mention of pieces of a broken pot being handed out, that sounds like a contract where sections are cut up and handed out.

69

Z01 Hines Book B.indb 69

07/05/2021 13:29

Frisians of the Early Middle Ages NIEUWHOF Yes, I agree; that is what I have in mind too. When people have parts of the same object, they are connected, and it reminds them of a specific event or an agreement. NIJDAM But, thinking on legal lines, how standard would the practice be? NIEUWHOF I don’t know that, but I do think there is a lot of evidence. NIJDAM It must have been rather gruesome, to have bones of your ancestors in your family archive. And to use dogs for excarnation as you suggest. NIEUWHOF This whole practice of excarnation with the aid of scavengers, and the subsequent collecting of the remaining bones, and storing these in your family archive, is rather gruesome in our modern view. But attitudes towards what is gruesome vary over time. KNOL It is also gruesome to put your loved ones in the ground to rot away. NIJDAM What about the sunken huts: were these the closets where they kept these family archives? NIEUWHOF No: deposits were sometimes made in sunken huts, but sunken huts are a relatively late development, originating in the Middle or Late Roman Period. Family archives are much older than that. They were probably stored in some corner of the house, in bags or on shelves. But they must have been kept together somehow. MAJCHCZACK There are similar cases known from North Frisia, where older objects appear in Early-medieval settlements. We thought these were objets trouvés, found by new inhabitants on the old sites and used as a connection to those who lived there earlier. A particularly fine example is a single sherd with nice AngloSaxon style motifs deposited at the bottom of a posthole of a ninth-century sunken hut in Goting/Föhr. VERSLOOT But perhaps that could have been an inherited object, deliberately placed? To mark possession where a new household had settled. MAJCHCZACK If so, it would cover a very long time from the fourth/fifth to the ninth century. KNOL Such finds can have several causes and explanations. One of them is that finds may be mixed up, which can be expected since terps were in continuous use for over two thousand years. The older excavations of Van Giffen and others may well be affected by mistakes. Finds may have been mixed up during the habitation period because of digging in the mound, or after the excavation during storage. NICOLAY I think that is a too negative way of explaining the patterns that we observe. I find this same pattern with Early-medieval brooches too; there is more structure and purpose to this phenomenon. I think this family archive is a very nice idea that also fits with the idea of the treasure hoard of the Early-medieval elite. Those also include objects of different ages with very high symbolic value. Of course we might explain it all away through confusion, but the family archive is a promising new explanation that fits many different phenomena. IJSSENNAGGER-VAN DER PLUIJM Someone who has done research on this, and on heirlooms in graves, for this period is Mirjam Kars (2011). NICOLAY She is working on bead necklaces. Her idea is that bead necklaces in graves often consist of many different beads, which have different stories, each related to different occasions and people. That also fits the idea of a family archive, which, in case of a bead necklace, was visible to everybody. 70

Z01 Hines Book B.indb 70

07/05/2021 13:29

Pottery and Social Relations NIJDAM I am currently undertaking research on Frisian birth spoons, and how they function as anchors for memories. NICOLAY Those are highly individualized though, with the name of a specific person written on them. NIJDAM The parallel lies in the fact that they remained in families for generations, thus becoming heirlooms closely connected to the family identity — a family archive in silver so to speak. FLIERMAN In your paper (and more extensively in your book) you describe a Frisian society that gradually becomes more hierarchical, but in the first century AD is still rather egalitarian, or at least characterized by limited social stratification. Reading your archaeological perspective I kept thinking about Tacitus, who famously describes two Frisian kings (Verritus and Malorix) who appear on the Romans’ radar in the first century AD. I’m wondering: how would you fit these Tacitean ‘kings’ into your model? Is Tacitus overestimating their importance out of ignorance or for literary purposes? Are these kings some sort of exception that confirms the rule: they were rather modest local strongmen at first who were able to exploit their Roman connections to significantly increase their social status and power at home? Or do these kings represent a different geographical region, a society closer to the Rhine which was perhaps more socially stratified to begin with? NIEUWHOF The story entails that they were Frisian kings, who wanted to move from the north with their people and settle in the empty buffer zone north of the Rhine. They asked the Emperor in Rome, Nero, for approval, but failed to get that. LOOIJENGA This is about the ancient Frisians. NIEUWHOF Those Frisians had leaders, but I don’t think there was a real aristocracy in that period. Tacitus describes them as kings, but what is a king? I think that the story describes them as regional leaders. Your ‘modest local strongmen’ might not be far from the truth.

References Primary sources Pliny, Naturalis Historia. Ed. and trans. H. Rackham 1945, Natural History, Vol. IV: Books 12–16 (London–Cambridge, MA). Ptolemy, Geographia. Trans. M. C. Galestin 2008a, ‘Frisii and Frisiavones’, Palaeohistoria 49/50, 687–708. Scriptores Historiae Augusta, Didius Julianus. Trans. D. Magie 1921, Historia Augusta, Volume I: Hadrian. Aelius. Antoninus Pius. Marcus Aurelius. L. Verus. Avidius Cassius. Commodus. Pertinax. Didius Julianus. Septimius Severus. Pescennius Niger. Clodius Albinus, Loeb Classical Library 139 (Cambridge, MA), 349–69. Tacitus, Germania. Trans. J. B. Rives 1999 (Oxford). — Annales. Trans. J. W. Meijer 1955, Publius Cornelius Tacitus: Kronieken (Haarlem). Secondary sources Abbink, A.A. 1999, Make It and Break It: The Cycles of Pottery. A Study of The Technology, Form, Function, and Use of Pottery from the Settlements at Uitgeest-Groot-Dorregeest and

71

Z01 Hines Book B.indb 71

07/05/2021 13:29

Frisians of the Early Middle Ages Schagen-Muggenburg 1, Roman Period, North-Holland, The Netherlands (Ph.D. thesis, Leiden University) Archaeological Studies Leiden University 5 (Leiden). Bakker, M. and T. Varwijk 2016, ‘Een verhaal over trends en continuïteit: aardewerkonderzoek nieuwe stijl toegepast op het aardewerk van Jelsum’. In Van Wierhuizen tot Achlum. Honderd Jaar Archeologisch Onderzoek in Terpen en Wierden, ed. A Nieuwhof, Jaarver­ slagen van de Vereniging voor Terpenonderzoek 98 (Groningen), 81–206. Bärenfänger, R. 2001, ‘Befunde einer frühmittelalterlichen Siedlung bei Esens, Ldkr. Wittmund (Ostfriesland)’, Probleme der Küstenforschung im südlichen Nordseegebiet 27, 249–300. Bärenfänger, R., E. Taayke and J. F. Kegler 2013, ‘Wir waren schon immer hier! Brüche und Kontinuitäten in der Besiedlung des Küstenraums/We waren er altijd al! Breuken en continuïteiten in de bewoning van het kustgebied’. In Land der Entdeckungen. Die Archäologie des friesischen Küstenraums/Land van ontdekkingen. De archeologie van het Friese kustgebied, eds. J. F. Kegler, A. Nieuwhof, K. Nowak-Klimscha and H. Reimann (Aurich), 161–70. Bazelmans, J. 2000, ‘Een laat-Romeins bewoningshiaat in het Nederlandse kustgebied en het voortbestaan van de Friezennaam’, Jaarverslagen van de Vereniging voor Terpenonderzoek 76–82, 14–75. Bloemers, J. H. F. 1978, Rijswijk (Z.H.), ‘De Bult’: eine Siedlung der ananefaten, Nederlandse oudheden 8 (Amersfoort). Böhme, H. W. 2003, ‘Das nördliche Niedersachsen zwischen Spätantike und frühen Mittelalter. Zur Ethnogenese der Sachsen aus archäologischer Sicht’, Probleme der Küstenforschung im südlichen Nordseegebiet 28, 251–70. Bosma, K. L. B. 2015, ‘Keramisch vondstmateriaal’. In Nederzettingssporen op de Kwelder. Haak Noord, Vindplaats 1: Marssum-It Aldlân, Gemeente Menaldumadeel, Deel 1, ed. J. B. Hielkema, RAAP-Rapport 2997 (Weesp), 78–144. Chapman, J. 2000, Fragmentation in Archaeology. People, Places and Broken Objects in the Prehistory of South Eastern Europe (London–New York). Chapman, J. and B. Gaydarska 2007, Parts and Wholes. Fragmentation in Prehistoric Context (Oxford). Cuijpers, A. G. F. M., C. M. Haverkort, J. M. Pasveer and W. Prummel 1999, ‘The human burials’. In The Excavations at Wijnaldum. Reports on Frisia in Roman and Medieval Times. Volume I, eds. J. C. Besteman, J. M. Bos, D. A. Gerrets, H. A. Heidinga and J. de Koning (Rotterdam–Brookfield), 305–22. De Koning, J. 2018, ‘Trans Flehum. Wijnaldum, Den Burg, Texel, Westergo: het Vlie als verbinder en grens’. In Fragmenten uit de Rijke Wereld van de Archeologie, eds. A. Nieuw­ hof, E. Knol and J. Schokker, Jaarverslagen van de Vereniging voor Terpenonderzoek 99 (Groningen), 146–56. De Ridder, T. 2005, ‘Wassermanagement in römischer Zeit: Die ältesten Deltawerke in Westeuropa’. In Kulturlandschaft Marsch. Natur-Geschichte-Gegenwart, ed. M. Fansa, Schriftenreihe des Landesmuseums für Natur und Mensch 33 (Oldenburg), 60–7. Dhaeze, W. 2011, ‘De Romeinse kustverdediging langs de Noordzee en het Kanaal van 120 tot 410 na Chr. Een onderzoek naar de rol van de militaire sites in de kustverdediging en drie casestudies over de militaire versterkingen van Maldegem-Vake, Aardenburg en Boulogne-sur-Mer’ (Ph.D. thesis, Ghent University). Dijkstra, M. and J. de Koning 2017, ‘All quiet on the western front?’ The western Netherlands and the ‘North Sea Culture’ in the Migration Period. In Frisians and their North Sea

72

Z01 Hines Book B.indb 72

07/05/2021 13:29

Pottery and Social Relations Neighbours. From the Fifth Century to the Viking Age, eds. J. Hines and N. IJssennagger (Woodbridge), 53–74. Erdrich, M. 2001, Rom und die Barbaren. Das Verhältnis zwischen dem Imperium Romanum und den germanischen Stämmen vor seiner Nordwestgrenze von der späten römischen Republik bis zum Gallischen Sonderreich, Römisch-Germanische Forschungen 58 (Mainz am Rhein). Fowler, E. 1963, ‘Celtic metalwork of the fifth and sixth centuries A.D. A re-appraisal’, Archaeological Journal 120, 98–160. Fulford, M. G. 1977, ‘Pottery and Britain’s foreign trade in the later Roman period’. In Pottery and Early Commerce. Characterization and Trade in Roman and Later Ceramics, ed. D. P. S. Peacock (London), 35–84. Galestin, M. C. 2007/2008, ‘Frisii and Frisiavones’, Palaeohistoria 49/50, 687–708. Genrich, A. 1941, ‘Bericht über die Untersuchungen auf der Barward (Gemarkung Imsum, Kreis Wesermünde)’, Probleme der Küstenforschung im südlichen Nordseegebiet 2, 157–70. Gerrets, D. A. 2010, Op de grens van land en water. Dynamiek van landschap en samenleving in Frisia gedurende de Romeinse tijd en de volksverhuizingstijd, Groningen Archaeological Studies 13 (Groningen). Gerrets, D. A. and J. de Koning 1999, ‘Settlement development on the Wijnaldum-Tjitsma terp’. In The Excavations at Wijnaldum. Reports on Frisia in Roman and Medieval Times. Volume I, eds. J. C. Besteman, J. M. Bos, D. A. Gerrets, H. A. Heidinga and J. de Koning (Rotterdam–Brookfield), 73–124. Hagers, J.-K. A. and M. M. Sier 1999, Castricum-Oosterbuurt, bewoningssporen uit de Romeinse tijd en middeleeuwen, ROB-Rapportage Archeologische Monumentenzorg 53 (Amersfoort). Halbertsma, H. 1953, ‘Enkele aantekeningen bij een verzameling oudheden, afkomstig uit een terpje bij Deinum’, Jaarverslagen van de Vereniging voor Terpenonderzoek 33–37, 239–56. Hamerow, H., Y. Hollevoet and A. Vince 1994, ‘Migration Period settlements and “AngloSaxon” pottery from Flanders’, Medieval Archaeology 38, 1–18. Häßler, H.-J. 1994, Neue Ausgrabungen in Issendorf, Niedersachsen. Ein Beitrag zur Erforschung der Kulturgeschichte des Sächsischen Stammes auf dem Kontinent, Studien zur Sachsenforchung 9 (Hannover). Hills, C. and S. Lucy 2013, Spong Hill. Part IX: Chronology and Synthesis (Cambridge). Hoss, S., J. Kempkens and T. Lupak 2015, ‘Bronzen beeldje van een haan met voetstuk’. In Een Romeins heiligdom en een vroegmiddeleeuws grafveld bij Buchten (L.), eds. T. Derks and B. de Fraiture, Rapportage Archeologische Monumentenzorg 226 (Amersfoort), 159–71. IJssennagger, N. L. 2014, ‘“Keltische” objecten langs de Friese kust?’, Kelten 62, 8–12. Jelsma, J. 2015, ‘Synthese: de pre- en protohistorische bewoning van Eelde, Groote Veen’. In Eelde-Groote Veen. De archeologie van steentijd tot Romeinse tijd, ed. C. Tulp (Groningen), 143–52. Kars, M. 2011, ‘A Cultural Perspective on Merovingian Burial Chronology and the Grave Goods from the Vrijthof and Pandhof Cemeteries in Maastricht’ (Ph.D. thesis, University of Amsterdam). Kiden, P., L. Denys and P. Johnston 2002, ‘Late Quaternary sea-level change and isostatic and tectonic land movements along the Belgian-Dutch North Sea coast: geological data and model results’, Journal of Quaternary Science 17, 535–46.

73

Z01 Hines Book B.indb 73

07/05/2021 13:29

Frisians of the Early Middle Ages Knol, E. 1993, ‘De Noordnederlandse Kustlanden in de Vroege Middeleeuwen’ (Ph.D. thesis, Vrije Universiteit Amsterdam). — 2011, ‘The first early medieval cemeteries along the northern Dutch coasts and their significance for Anglo-Saxon migration’. In Transformations in North-Western Europe (AD 300–1000), eds. T. A. S. M. Panhuysen and B. Ludowici, Neue Studien zur Sachsenforschung 3 (Hannover–Stuttgart), 218–27. — 2014, ‘Ezinge: metaal uit een opgraving zonder detector’. In En dan in hun geheel. De vondsten uit de opgravingen in de wierde Ezinge, ed. A. Nieuwhof, Jaarverslagen van de Vereniging voor Terpenonderzoek 96 (Groningen), 187–206. Krol, T. N., M. Dee and A. Nieuwhof 2020, ‘The chronology of Anglo-Saxon style pottery of the fourth to eighth century in radiocarbon dates: improving the typo-chronology’, Oxford Journal of Archaeology, on-line (Sept. 2020), https://doi.org/10.1111/ojoa.12202 Krol, T. N., K. Struckmeyer and A. Nieuwhof 2018, ‘Anglo-Saxon style pottery from the northern Netherlands and north-western Germany: Fabrics and finish, regional and chronological patterns, and their implications’, Archaeometry 60, 713–30. Lanting, J. N. and J. van der Plicht 2010, ‘De 14C-chronologie van de Nederlandse pre- en protohistorie VI: Romeinse tijd en Merovingische periode, deel A: historische bronnen en chronologische thema’s’, Palaeohistoria 51/52, 27–168. — 2012, ‘De 14C-chronologie van de Nederlandse pre- en protohistorie VI: Romeinse tijd en Merovingische periode, deel B: aanvullingen, toelichtingen en 14C-dateringen’, Palaeohistoria 53/54, 283–391. Majchczack, B. S. 2018, ‘Die Insel Föhr und ihr römisch-kaiserzeitliches Netzwerk im Spiegel des Urnengräberfeldes von Nieblum LA 41, Kr. Nordfriesland‘, Siedlungs-und Küstenforschung im südlichen Nordseegebiet 41, 133–72. Majchczack, B. S. and M. Segschneider 2015, ‘Eine Siedlung der Jungeren Romischen Kaiserzeit und Volkerwanderungszeit sowie des Frühmittelalters bei Tinnum auf Sylt’. In Archaologische Siedlungsforschung auf den nordfriesischen Inseln, eds. C. von Carnap-Bornheim and M. Segschneider, Offa-Bücher 89 (Neumünster), 119–34. Morris, F. M. 2015, ‘Cross-North Sea contacts in the Roman period’, Oxford Journal of Archaeology 34, 415–38. Nicolay, J. A. W. 2014, The Splendour of Power: Early Medieval Kingship and the Use of Gold and Silver in the Southern North Sea Area (5th to 7th century AD), Groningen Archaeological Studies 28 (Groningen). — 2017, ‘Power and identity in the southern North Sea area: The Migration and Merovingian periods’. In Frisians and their North Sea Neighbours. From the Fifth Century to the Viking Age, eds. J. Hines and N. IJssennagger (Woodbridge), 75–92. Nicolay, J. A. W. and P. den Hengst 2008, ‘Synthese: De Bloemert in een regionale context’. In Opgravingen bij Midlaren. 5000 jaar wonen tussen Hondsrug en Hunzedal, ed. J. A. W. Nicolay, Groningen Archaeological Studies 7 (Groningen), 577–622. Nielsen, K. H. 2003, ‘Saxon art between interpretation and imitation: the influence of Roman, Scandinavian, Frankish and Christian art on the material culture of the Continental Saxons, AD 400–1000’. In The Continental Saxons from the Migration Period to the Tenth Century: An Ethnographic Perspective, eds. D. Green and F. Siegmund, Studies in Historical Archaeoethnology (Woodbridge), 193–233. Nieuwhof, A. 2008a, ‘Aardewerk’. In De Leege Wier van Englum. Archeologisch onderzoek in het Reitdiepgebied, ed. A. Nieuwhof, Jaarverslagen van de Vereniging voor Terpenonderzoek 91 (Groningen), 47–75.

74

Z01 Hines Book B.indb 74

07/05/2021 13:29

Pottery and Social Relations — 2008b, ‘Het handgemaakte aardewerk, ijzertijd tot vroege middeleeuwen’. In Opgravingen bij Midlaren. 5000 jaar wonen tussen Hondsrug en Hunzedal, ed. J. A. W. Nicolay, Groningen Archaeological Studies 7 (Groningen), 261–304. — 2011, ‘Discontinuity in the Northern-Netherlands coastal area at the end of the Roman Period’. In Transformations in North-Western Europe (AD 300–1000), eds. T. A. S. M. Panhuysen and B. Ludowici, Neue Studien zur Sachsenforschung 3 (Hannover–Stuttgart), 55–66. — 2013, ‘Anglo-Saxon immigration or continuity? Ezinge and the coastal area of the northern Netherlands in the Migration Period’, Journal of Archaeology in the Low Countries 5, 53–83. — 2014, ‘De geschiedenis van Ezinge in scherven. Handgevormd aardewerk van 500 v.C. tot 1500 n.C.’ In En dan in hun geheel. De vondsten uit de opgravingen in de wierde Ezinge, ed. A. Nieuwhof, Jaarverslagen van de Vereniging voor Terpenonderzoek 96 (Groningen), 30–128. — 2015, Eight Human Skulls in a Dung Heap and More. Ritual Practice in the Terp Region of the Northern Netherlands. 600 BC–AD 300, Groningen Archaeological Studies 29 (Groningen). — 2016, ‘De lege vierde eeuw’. In Van Wierhuizen tot Achlum. Honderd Jaar Archeologisch Onderzoek in Terpen en Wierden, ed. A Nieuwhof, Jaarverslagen van de Vereniging voor Terpenonderzoek 98 (Groningen), 83–98. — 2017a, ‘Het imago van de Chauken. Plunderende zeerovers of ondernemende terpbewoners?’, Archeologie in Nederland 1, 26–33. — 2017b, ‘Potters and pottery from afar: Some observations on long-distance contacts’. In Life on the Edge: Social, Political and Religious Frontiers in Early Medieval Europe, eds. S. Semple, C. Orsini and S. Mui, Neue Studien zur Sachsenforschung 6 (Braunschweig), 295–303. Nieuwhof, A. and T. B. Volkers 2015, ‘Luxe servies? Terra sigillata ten noorden van de limes’, Archeobrief 19, 26–32. Nösler, D. 2017, ‘Ein Jahrtausend in Scherben – Ein Beitrag zur Typochronologie frühgeschichtlicher Siedlungsware aus Loxstedt, Ldkr. Cuxhaven’, Siedlungs- und Küstenforschung im südlichen Nordseegebiet 40, 217–319. Nösler, D. and S. Wolters 2009, ‘Kontinuität und Wandel –zur Frage der spätvölkerwanderungszeitlichen Siedlungslücke im Elbe-Weser-Dreieck’. In Dunkle Jahrhunderte in Mitteleuropa?, eds. O. Heinrich-Tamaska, N. Krohn and S. Ristow, Studien zu Spätantike und Frühmittelalter Band 1 (Hamburg), 367–88. Plettke, A. 1940, Der Urnenfriedhof Dingen, Kr. Wesermunde (Hildesheim). Prison, H. 2011, ‘Ausgrabungen im Umfeld der Wurt Jemgumkloster, Gde. Jemgum, Ldkr. Leer (Ostfriesland)- Ein Vorbericht’, Nachrichten aus Niedersachsens Urgeschichte 80, 117–36. Reinders, H. R. and Y. Aalders 2007, ‘Friese klinkerschepen in de vroege middeleeuwen’, De Vrije Fries 87, 9–28. Rice, P. M. 1987, Pottery Analysis. A Sourcebook (Chicago–London). Rieck, F. 2003, ‘The ships from Nydam bog’. In The Spoils of Victory. The North in the Shadow of the Roman Empire, eds. L. Jørgensen, B. Storgaard and L. Gebauer Thomsen (Copenhagen), 296–309. Schmid, P. 2006, Die Keramikfunde der Grabung Feddersen Wierde (1. Jh. v. bis 5 Jh. n. Chr.), Probleme der Küstenforschung 29/Feddersen Wierde Bd. (Oldenburg).

75

Z01 Hines Book B.indb 75

07/05/2021 13:29

Frisians of the Early Middle Ages Scholtz, T. 2015, ‘Ein Sodenwandhaus der Volkerwanderungszeit bei Tinnum auf Sylt’. In Archaologische Siedlungsforschung auf den nordfriesischen Inseln, eds. C. von Carnap-Bornheim and M. Segschneider, Offa-Bücher 89 (Neumünster), 135–297. Schön, M. D. 1988, ‘Gräberfelder der römischen Kaiserzeit und frühen Völkerwanderungszeit aus dem Zentralteil der Siedlungskammer von Flögeln, Landkreis Cuxhaven’, Neue Ausgrabungen und Forschungen in Niedersachsen 18, 181–297. Siegmüller, A. and K. Mückenberger 2017, ‘Structure and function of landing places and riverside markets along the Lower Weser in the Roman Iron Age’. In Life on the Edge: Social, Political and Religious Frontiers in Early Medieval Europe, eds. S. Semple, C. Orsini and S. Mui, Neue Studien zur Sachsenforschung 6 (Braunschweig), 273–283. Sinopoli, C. M. 1991, Approaches to Archaeological Ceramics (New York–London). Stilke, H. 2001, ‘Grauware des 8. bis 11. Jahrhunderts’. In Handbuch zur mittelalterlichen Keramik in Nordeuropa, eds. H. Lüdtke and K. Schietzel, Schriften des Archäologischen Landesmuseums S-H 6 (Neumünster), 23–82. Taayke, E. 1996, ‘Die einheimische Keramik der nördlichen Niederlande. 600 v.Chr. bis 300 n. Chr.’ (Ph.D. thesis, University of Groningen). — 2000, ‘Onder Franken en Saksen. Friesland in de laat-Romeinse tijd’, De Vrije Fries 80, 9–28. — 2006, ‘Uslarien? Rijn-Wezer-Germaans aardewerk op Nederlandse bodem’. In Vakken in Vlakken. Archeologische Kennis in Lagen, eds. O. Brinkkemper, J. Deeben, J. van Doesburg, D. P. Hallewas, E. M. Theunissen and A. D. Verlinde, Nederlandse Archeologische Rapporten 32 (Amersfoort), 199–214. — 2013, ‘Rhein-weser-germanische Keramik in den Niederlanden im 1.–5. Jahrhundert’. In Westgermanische Bodenfunde. Akten des Kolloquiums anlässlich des 100. Geburtstages von Rafael von Uslar am 5. und 6. Dezember 2008, ed. G. Rasbach, Kolloquien zur Vorund Frügeschichte Band 18 (Bonn), 191–8. — 2020, ‘Hand-made pottery of the Migration Period and the Merovingian Period’. In The Excavations at Wijnaldum II: Hand-made and Wheel-thrown Pottery, eds. A. Nieuw­ hof, J. de Koning, G. de Langen and E. Taayke, Groningen Archaeological Studies 38 (Groningen), 39–98. Taayke, E. and E. Knol 1992, ‘Het vroeg-middeleeuwse aardewerk van Tritsum, gem. Franekeradeel (Fr.)’, Paleo-aktueel 3, 84–8. Thasing, S. and A. Nieuwhof 2014, ‘Importaardewerk in Ezinge. Uitwisseling en sociaal-­ politieke structuur in de Romeinse tijd en de vroege middeleeuwen’. In En dan in hun geheel. De vondsten uit de opgravingen in de wierde Ezinge, ed. A. Nieuwhof, Jaarverslagen van de Vereniging voor Terpenonderzoek 96 (Groningen), 129–46. Tischler, F. 1956, ‘Der Stand der Sachsenforschung, archäologisch gesehen’, Bericht der Römisch-Germanischen Kommission 35, 21–215. Tuin, B. 2008, ‘Graven aan de rand. Onderzoek van de akkers grenzend aan De Bloemert’. In Opgravingen bij Midlaren. 5000 jaar wonen tussen Hondsrug en Hunzedal, ed. J. A. W. Nicolay, Groningen Archaeological Studies 7 (Groningen), 521–43. Van de Moortel, A. 2011, ‘Medieval boat and ship-finds of Germany, the Low Countries, and Northeast France. Archaeological evidence for shipbuilding traditions, shipbuilding resources, trade and communication’, Siedlungs- und Küstenforschung im südlichen Nordseegebiet 34, 67–105. Van den Broeke, P. W. 2018, ‘Pierenpaté? Fries aardewerk ten zuiden van de Nederrijn’. In Fragmenten uit de rijke wereld van de archeologie, eds. A. Nieuwhof, E. Knol and

76

Z01 Hines Book B.indb 76

07/05/2021 13:29

Pottery and Social Relations J. Schokker, Jaarverslagen van de Vereniging voor Terpenonderzoek 99 (Groningen), 77–87. Van Es, W. A. 1967, ‘Wijster, a native village beyond the Imperial Frontier 150–425 A.D.’, Palaeohistoria 11, 1–595. Von Quillfeldt, I. and P. Roggenbuck 1985. Westerwanna II. Die Funde des völkerwanderungszeitlichen Gräberfeldes im Helms-Museum, Hamburgisches Museum für Vor- und Frühgeschichte. Katalog und Tafeln, Die Urnenfriedhöfe in Niedersachsen (Hildesheim). Von Uslar, R. 1977, ‘Zu einer Fundkarte der jüngsten Kaiserzeit in der westlichen Germania libera’, Praehistorische Zeitschrift 52, 121–47. Vos, P. C. and E. Knol 2015, ‘Holocene landscape reconstruction of the Wadden Sea area between Marsdiep and Weser. Explanation of the coastal evolution and visualisation of the landscape development of the northern Netherlands and Niedersachsen in five palaeogeographical maps from 500 BC to present’, Netherlands Journal of Geosciences 94, 157–83. Weber, M. 1998, ‘Das Gräberfeld von Issendorf, Niedersachsen. Ausgangspunkt für Wanderungen nach Britannien?’, Studien zur Sachsenforschung 11, 199–212. Zimmermann, W.H. 1972, ‘Eine Ringfibel mit auswärts gewendeten Tierköpfen aus Midlum-Northum, Kreis Wesermünde’, Neue Ausgrabungen und Forschungen in Niedersachsen 7, 185–202.

77

Z01 Hines Book B.indb 77

07/05/2021 13:29

Z01 Hines Book B.indb 78

07/05/2021 13:29

• 4 • LANDSCAPE, TRADE AND POWER IN EARLY-MEDIEVAL FRISIA Gilles de Langen and J. A. Mol

I

n general text books, Early-medieval Frisia is presented as a rich, long stretched out but in itself coherent, region along the North Sea coast boasting much trading activity. Its landscape between Bruges and Bremen is thought to have been highly maritime, with many local trade settlements having access to the sea. In this paper, we venture to question both its coherency and its maritime character for the period AD 650–850 by adopting a spatial approach. This approach is threefold. First we reconstruct Frisia in the geographical sense. We describe the main types of landscape and define the boundaries of the districts or pagi that together constituted Frisia. Against this background, we then take a look at the supposed trade settlements, to discover that outside the well-known emporia in border positions along major rivers and coasts, no such ever existed. Up to the tenth century, trade transactions that fed the day-to-day economy apparently did not need permanent separate trade centres. This absence of a real hierarchy of settlements strongly suggests that Frisia in its early days was a rural society building its welfare on agricultural production. Our third objective is to get some grip on the political coherency of the Frisian lands and their population groups in the seventh and early eighth century. We do so by focusing on the conditions for the development of large landownership, which are found to have been more favourable in the sandy coastal area of the later Holland than in the clay or terpen districts east of the Vlie. This then gives at the end room to develop a few thoughts on the possible West Frisian or ‘Holland’ power base of the heathen Frisian king Radbod (Redbad) and his predecessor, Aldgisl.

Introduction Early-medieval Frisia and the phenomenon of Early-medieval Frisian trade are often mentioned in the same breath. What they have to do with each other, however, is not easy to determine and it is therefore no wonder that this relationship has been under discussion for a long time. In research on Anglo-Saxon England, the literature shows that it is not problematic to speak of visiting traders. For the Frisian regions this seems somewhat more difficult, precisely because the trade is known as ‘Frisian’. An answer to this question can only be found when we take into account the different stages of the history of the trade and also when we look at the trading activities within the Frisian 79

Z01 Hines Book B.indb 79

07/05/2021 13:29

Frisians of the Early Middle Ages regions themselves and not only at the important trade of Dorestad, Domburg and the other well-known ‘Frisian’ emporia. Trade is a constant, and for that reason alone we may assume that the inhabitants of the Frisian regions traded in some way at an early stage just as the inhabitants of the neighbouring non-Frisian regions will have done. They will have searched and travelled themselves to maintain contacts and exchange goods. The question now is whether Frisian trade at some point became separated from the agricultural communities that in an earlier phase mainly kept the trade going themselves. Before we try to answer this question, it is worth emphasizing that Frisian trade was not (exclusively) conducted by ethnic Frisians, whoever such Frisians might be. In any case, by Frisian trade we mean the long-distance trade-flows that between AD 650 and 850 passed through the Low Countries. In a certain sense, this means that we first separate the trade from the Frisian regions and then look at what interference the Frisians living along the coasts might have had with that distant trade. Broadly speaking, three forms of local involvement can be assumed, which do not have to be mutually exclusive: 1. Frisians in the regions were sought out by the (passing) trade; the levying of tolls could provide extra revenue. 2. Local leaders led the trade; they could increase their revenues by being successful in larger political and social contexts. 3. Frisians living in the regions were independently active in trade; they traded within the region, inter-regionally and/or internationally. In this contribution we shall discuss these three forms in more detail. The historio­ graphy of the views on Frisian trade will often be left aside. Nevertheless, it is useful to make an exception here in the case of the view that the Frisian regions have been particularly active in trade because the majority of their inhabitants were livestock farmers who had to trade and who, thanks to their extensive agrarian operations, had plenty of time to do so. This thesis has sometimes been substantiated with the opinion that socio-economic and political differences between these cattle breeders were small and that they therefore were individually free and autonomous in their movements. The basis for this model was laid in the nineteenth century by, among others, the numismatist Jacob Dirks, who attributed a large share in the Early-medieval trade to the inhabitants of the terpen areas (Dirks 1846). In the early twentieth century, Dirks’s view was criticized, by inter alios the historian H. A. Poelman (1908) and the Frisian archaeologist P. C. J. A. Boeles. The latter clearly set apart the trade of Carolingian Dorestad, at the expense of the trade activity of the former inhabitants of his own province and their alleged early trading places (Boeles 1927, 152–7). Nevertheless, he still kept an eye on the role of these Frisians in an older large-scale trading network with both the Frankish and Scandinavian worlds, a network that explained the wealth of sixth- and seventh-century gold discoveries and continued to function during the period of Dorestad’s trade. Nonetheless, around the Second World War, the old idea that the Frisians between Vlie and Weser had a large share in the Frisian trade made a real comeback, to the displeasure of Boeles who stuck to his phased and indeed layered trading model. In 1951 he criticized the ‘the tendency, which has reappeared in recent times, already known from Dirks’ time, but considered overcome, to bring the active trade relations of the Frisians with surrounding countries back to very early times, to overestimate their significance 80

Z01 Hines Book B.indb 80

07/05/2021 13:29

Landscape, Trade and Power and to predate, as it were, their later generally acknowledged achievements’ (Boeles 1951, 360–2). And as far as this later trade was concerned: in Dorestad’s trade of the eighth and ninth centuries, as before, non-Frisian traders were also active, while the provinces of Friesland and Groningen were highly dependent on Dorestad. According to him it was only in the late stages of Dorestad trade that middlemen established themselves in Friesland and Groningen; middlemen, who bought locally produced export products such as woollen cloth, dairy products, horses and animal skins in order to sell them in Dorestad. This view was followed by the German archaeologist Werner Haarnagel when in 1955 he identified the oldest Emden in East Frisia as a wīc settlement in the form of an Einstrassensiedlung on a Langwarf, a settlement of foreign (presumably West Frisian) merchants in an otherwise agrarian environment. According to Haarnagel, there would have been more of such Langwarf trade settlements in East Frisia (Haarnagel 1955, 76). As already noted, around 1940 resistance to Poelman’s vision began. This reaction was led by the historians J. F. Niermeijer (1937; 1977a; 1977b) and B. H. Slicher van Bath (1948; 1949; 1968). In their view, the Frisian trade was not at all separate from the Frisian regions. It would have been the responsibility of the Frisians living there. They saw the Central Frisia of the seventh century and the first half of the eighth as the centre of a traffic area around the North Sea or rather ‘the country of birth of most of the merchants who by then had found Dorestad as their actual meeting point’ (Niermeijer 1977a, 24–5). Niermeijer and Slicher van Bath believed that the inhabitants of the salt marshes east of the Vlie were almost exclusively engaged in livestock farming; arable farming would only have been possible on a limited scale (cf. also Postma 1934, 148). Thus they came to the opinion that the Frisians could export livestock products and had to import agricultural products, especially grain. Livestock farmers with extensive farming activities would have had a good starting position as potential traders: during the summer months they could release manpower for trade much easier than arable farmers. Especially their younger sons, for whom there was no room on the parental farm, would have become traders. Slicher van Bath believed that there must have been a lot of them available, since the salt marsh area was considered to have been densely populated. In addition, livestock farmers would have been well acquainted with the handling of capital, which was advantageous for their own continuous trade – a large herd of livestock constituted wealth, cattle could generate money and thus make trade (better) possible. Trade then in turn yielded money, which brought the Frisians into contact with Scandinavia, where a lot of money was circulating at that time, a contact that enabled the Frisians to maintain their money economy even better. Slicher van Bath, in particular, expressed the opinion that cattle breeding and the resulting wealth of accumulated money prevented them from ending up in hereditary serfdom. He even took it further: these farmers had long used their own land and were relatively equal socially and politically and were therefore individually free in their movements. This explanation was apparently convincing, because the opinion soon gained ground that farmers in the northern Netherlands were cattle breeders in the Early Middle Ages and (thus) had a large share in the Early-medieval trade.1 Haarnagel’s research into 1 Thus, according to Van Buijtenen 1953, 92–3 (following Niermeijer and Slicher van Bath), after the Second World War the historian’s gaze shifted from agricultural life on the terps to the distant Mare Frisicum: ‘the image of the primitive merchant fleet breaks through the rigidity of agricultural seclusion’;

81

Z01 Hines Book B.indb 81

07/05/2021 13:29

Frisians of the Early Middle Ages the earliest Emden also proved influential. Already in 1956, Herre Halbertsma took over the idea that trading villages had developed on the Frisian salt marsh, although he dated their first stages no earlier that the late Carolingian Period (Halbertsma 1956, 245–6) – more about this further on. It is important to note for now that both hypotheses were forged into a unity. In the 1980s some criticisms were uttered, but the idea that merchants of the Carolingian Period had taken over the Merovingian trading function from the other Frisian settlements remained. As recently as 1992 Lebecq wrote about the Frankish/Frisian trade of Dorestad that the terpen area surely was ‘the start point of the first merchantile expansion of the Frisians towards Scandinavia’, and that as ‘a good departure point’ the terps were soon followed by ‘trade terps’ (Lebecq 1992). And thus the idea that the Frisian coastal residents were ‘free guys’ par excellence, who during the summer months were to be found in many places on their ships full of merchandise still affects the image of the daily economy of the individual Frisian regions. Meanwhile, hardly any of these compelling reasons that would have incited the Frisian farmers of the salt marshes to an independent and intensive distant trade has withstood close scrutiny, as will be shown below. Nevertheless, it is wise not to wipe away the Frisian farmers as traders altogether, but to see the Frisians as a coastal people along the North Sea and, on the one side, to explain their possible trading activities with more general functions of contact2 and, on the other, to give the Early-medieval ‘Frisian’ trading ships also non-Frisian agents, skippers and crews (Lebecq 1992). In order to study the daily economy of the Frisian regions in more detail, we carry out a threefold spatial analysis. First, we reconstruct Frisia in the geographical sense. We describe the main types of landscape and establish the boundaries of the districts or pagi that together constituted Frisia. After a short introductory paragraph on the landscape of the Frisian coastal area up to about 900, we try to define the economic and political position of the several Frisian districts and give the historically known Frisia Citerior new borders. Our next objective then is to analyse the structure of Frisian trade. As mentioned above, the discussion about this Frisian trade was partly determined by the idea that the Early-medieval Frisian trade not only needed emporia but also trading places on a local level. In recent years, however, we have come to realize that for a long time there may have been no need for such local trading settlements, as the conditions for their development were not met: a large volume of continuous trade, a considerable number of permanently settled traders and craftsmen, an invariably favourable geographical location and a powerful and stable protective power. These considerations induce us to study the link between trade and power in Early-medieval Frisia and to analyse its landscape as a whole rather than trying to explore separate trade centres. Therefore, we Keuning 1968, 581: a highly fragmented agricultural society in which the majority of the population lived as livestock farmers; ibid. 581–2: highly developed trade; Van Es 1971, 29–30 and 31: the inhabitants of the terpen area had an important share in the trade; Jansen 1976, 123–6: mainly animal husbandry, too little arable land: grain imports necessary, densely populated and an important necessary trade; Jansen 1982, 148: Carolingian economy fairly closed, Frisians offer a window on the world; Blok 1979, 106: extensive trade, substantial money circulation, extensive animal husbandry; ibid. 112: wealthy farmers. See also Verhulst and De Bock-Doehaerd 1981, 195; Lebecq 1983. 2 For contacts between the Frisian regions, see Nieuwhof, this vol.; for maintaining contacts between the various Early-medieval communities of the North Sea coasts and the significance of the concept of connectivity for Frisian archaeology in particular, IJssennagger 2017; for the political necessity for the exchange of goods, Nicolay 2014.

82

Z01 Hines Book B.indb 82

07/05/2021 13:29

Landscape, Trade and Power look for anchor points concerning the trade activities at the regional and local level, more specifically within the communities of the two regions that formed central Frisia, namely the pagi of Westergo and Oostergo. Our method is ‘retrogressive’; departing from the spatial patterns in the Late Middle Ages, we try to find our way back in time discerning how High-medieval settlements were organized and to unveil the shaping forces under the found patterns. After discussing the separate elements of reconstruction we attempt to draw some central hypotheses on the development of Early-medieval Frisia in general, by comparing Central Frisia with the immediately adjacent parts, especially with the (West Frisian) areas between the Maas and Vlie. In the third part, we shall discuss the consequences of the predominantly agrarian character of the Frisian lands for the power developments in Frisia as a whole, that is to say, the coastal zone between the Sincfal and Weser, from the period around 650 until about 850. This is necessary because the literature on trade pays much attention to the concept of Magna Frisia or Greater Friesland as a political construction that would have provided a breeding ground or basis for Frisian trade. Central to this is the political coherency of the Frisian lands and their population groups in the seventh and early eight century, the key to which seems to lie in the power constellations surrounding the Frisian king Radbod and his predecessor Aldgisl.

The heterogeneous Frisian landscape Around AD 800 all southern coastal regions of the North Sea east of Britain were together called Frisia. According to Lex Frisionum, these habitation areas were globally located between the stream Sincfal, near the current Westerschelde, and the mouth of the Weser, north of Bremen. This Frisia did not form a contiguous area, as it was split up by wider rivers. Inland the sub-regions did not connect with neighbouring residential areas because the coastal zone was separated from the hinterland by extensive peat bogs. By facilitating shipping, the streams and the sea promoted mutual contact between the regions, while land traffic with the hinterland was difficult. Contact was only possible by means of shipping traffic on the rivers that crossed Frisia in a limited number of places. It is clear that early in the Middle Ages these coastal districts developed a shared material culture, language and identity: the Frisian.3 From a landscape and economic point of view however, Frisia was and remained a heterogeneous set of lands, since each district had its own specific set of geographical characteristics and, in a broader context, its own geopolitical and economic position. Because in the existing literature Early-medieval Frisia differs in size and meaning, the aim of this first part of our contribution is to determine the boundaries of Frisia, to examine which landscape differences can be discerned and also to determine which non-Frisian areas were at some point considered to belong to Frisia. Central to this are the scenic differences between West Frisia on the one hand and Central and East Frisia on the other.

3 This Frisian cultural zone developed within a closely linked environment but at the same time was characterized by regional differences. See Dijkstra 2011, 341–70; Nicolay 2014; and other contributions in this volume.

83

Z01 Hines Book B.indb 83

07/05/2021 13:29

Frisians of the Early Middle Ages Frisia west of the Vlie: continuous sand ridges suitable for agriculture in a manorial setting In terms of their main structure, the Frisian districts resembled one another to the extent that each consisted of a relatively thin strip of habitable soils situated between sea and peat bogs (Fig. 4.1).4 Only in Friesland between Vlie and Boorne, that is to say in the Westergo district, was there a significantly wider zone suited for habitation.

Fig. 4.1  Early-medieval Frisia and its districts c. AD 700. Map S. de Bruijn, G. J. de Langen and J. A. Mol, Province of Friesland/Fryske Akademy.

West of the Vlie, people lived on a series of sand ridges parallel to the coast, and on the narrow clay zones along the Oude Rijn and Meuse rivers, apart then from the Pleistocene moraine structures of Texel and Wieringen in the far north. The north– south orientation of these sand ridges, consisting of old beach barriers and dunes, together with the most recently formed beaches, meant that the pagi or districts of Texel, Kennemerland and the northern part of Rijnland were interconnected and thus highly accessible by land to each other’s inhabitants. Perhaps the southern part of the Rhineland and the northern part of Maasland can also be included in this complex, as the Oude Rijn was easy to cross. What the extension of this complex of old beach barriers south of the Hook of Holland looked like in the Early Middle Ages is still largely unknown because of the large-scale erosion that has taken place in the estuaries of the Meuse and Scheldt. It is difficult to estimate how densely this part of Frisia was inhabited in the Early Middle Ages, although we do not want to underestimate the importance of these residential areas, since they are known to have included some large royal domains in Frankish times. Nor do we know whether the mouths of the Meuse and the Scheldt were as divisive then as they were later. Possibly their widening only 4

Our reconstruction of Frisia’s landscape is mainly based on Vos 2015 (for the Dutch part) and Vos and Knol 2015 (for the German part), although we disagree with these authors on some important points, inter alia, on the basis of Zomer 2016; Aalbersberg 2019; Molenaar et al. 2009; Worst 2020, in prep.; Hoek 1979 and Henderikx 2014a.

84

Z01 Hines Book B.indb 84

07/05/2021 13:29

Landscape, Trade and Power started in the course of the eighth century, when on both sides of the Scheldt the sea began to erode the beach-wall complex, which would eventually lead to the loss of, among others, the emporium near Domburg. At the same time, however, new deposits of clay layers made the salt marshes inland increasingly inhabitable, which triggered their colonisation during the late eighth century and beyond (Van Dierendonck 2009). (Our map of Frisia in Figure 4.1 may give too rosy a picture of the scope for habitation on both sides of the Scheldt for the period around AD 700.) On the whole, West Frisia was an agricultural area where arable farming was highly possible on the easy-to-treat top layer of the sandy ridges. It is important to observe that in the High Medieval Period, for which the sources reveal more information, many manorial structures can be identified precisely here. A good impression of this can be obtained when we place the manors (curtes or villae in Latin) in Holland known from literature on a map (Fig. 4.2: based in part on data provided by Gosses 1926; De

Fig. 4.2  High-medieval manors in the county of Holland projected on the landscape of c. AD 700. Map S. de Bruijn, G. J. de Langen and J. A. Mol, Province of Friesland/Fryske Akademy.

85

Z01 Hines Book B.indb 85

07/05/2021 13:29

Frisians of the Early Middle Ages Monté Verloren 1942; Meilink 1951; Kort 1981, Introduction). Nearly all the manors then appear to have been situated on the backbone of Holland: on the sandy ridges from Monster to Petten, on the boulder clay embankments of Texel and Wieringen, or on the early reclaimed edges of the adjoining peat bogs covered with a thin layer of clay. Most reports on these estates date from after the Viking Period, when it becomes clear that they were then for the most part in the hands of either the king or the comital dynasty of Gerulfingen, who had taken over authority in West Frisia from the Vikings on behalf of the king. However, the available data also suggest that at that time quite a lot of these manors must have been continuations of estate units going back as early as to the period before the Franks (De Langen and Mol 2020). These units had apparently been taken by the Franks from earlier Frisian rulers, temporarily given in fief to Viking rulers, and then been transferred to the Gerulfingen. If this picture is correct, there was a strong path dependency in the landscape stretching back to Merovingian times. At that time, this domanial power system was already very different from the more egalitarian social structures in the Frisian clay areas. Since the end of Roman times, the interconnecting element of the West Frisian districts, the often mentioned parallel ridges of old beach barriers and dunes, was apparently an excellent ‘highway’ with stations where the count could stay temporarily and along which he could supply men and means to exercise power and influence. Zooming in on the salt marsh regions of Central and East Frisia To the east of the Vlie, Friesland showed a different landscape, with different agricultural possibilities and also different power structures. Here, up to the mouth of the Weser, people lived on open, unembanked salt marshes, in various pagi that were best reached by water. Of these salt marsh regions, the area up to the Lauwers consisting of the pagi of Westergo and Oostergo was the largest. It was also the most densely populated and the richest one. It cannot be a coincidence that it is referred to in Lex Frisionum as Frisia per se, in the sense that it is distinguished from West and East Frisia. This meant that with regard to the laws that were observed it counted as Frisia’s centre or core region; what in West and East Frisia was found to be divergent in legislative terms had to be indicated separately. The interpretation of both this historical information and the archaeological data from this core area has had a major influence on the way the Frisian coastal regions as a whole are viewed. That is why it makes sense to pay some more attention to Central Frisia and its landscape. Until the middle of the eleventh century, the people of Central Frisia (and East Frisia) mainly lived in the clay zone bordering the sea, although from the eighth century or perhaps even slightly earlier pioneers settled in transitional zones along the still high-lying peat bogs to the south in order to start the first reclamations. Several large and small tidal streams kept the area open and made it easily accessible from the sea. Situated behind the Wadden isles, which functioned as a sort of a buffer, the salt marshes lay to some extent protected against excessive storm surges.5 Thanks to their characteristic elevated house platforms or dwelling mounds – the so-called terpen – the 5

That is, on a daily basis. Due to large-scale developments, for example, moving the inlets between the islands and thereby shifting the largest tidal channels, the salt marsh could locally become eroded and disappear in part. Such developments, however, did not occur in most of the salt marsh for most of the time.

86

Z01 Hines Book B.indb 86

07/05/2021 13:29

Landscape, Trade and Power inhabitants could cope with floods, as long as they assembled their livestock and property within or next to their high-lying farmhouses in good time. Storm surges would only cause major damage after about 1100 when the building of dikes allowed a new pattern of occupation that was no longer restricted to living on terpen, but which was risky as well. While systematic embankment prevented the sea from flooding the salt marsh on a broader front, it led to high water levels outside of the dikes and, during storm surges, to higher risks and dramatic consequences in case of breaches. In general, the marine clay soil was extremely fertile, depending on the altitude and drainage possibilities, much more in fact than the cultivation layers on the sand ridges in West Frisia. In respect of the habitat of the people living here, for a long time the view prevailed in the literature that before the embankments the terp dwellers had to cope with a saline, water-rich environment in which only the conditions for cattle breeding were optimal, which made them livestock farmers in the first place. We have already touched on that in our Introduction. That perception has now been completely revised. At least since the early 1980s, terp researchers are aware of the fact that certain agricultural crops can thrive on the salt marsh outside the dikes. For the time being, however, this experimentally proven insight has had little effect on the predominant view that terp dwellers were mainly dependent on cattle breeding. Though scepticism was sometimes expressed, serious criticism had to wait for more fieldwork and an active contribution from archaeobotany, which in the end has brought new insights. During the past ten years, excavations have repeatedly established that on or around the terps, both on artificially raised fields next to the terps (Varwijk and De Langen 2018; 2019) and on the highest natural salt marsh ridges, substantial arable farming had indeed been carried out. This leads to the idea that arable farming on the salt marsh may have been the rule rather than the exception (Schepers 2014; 2016). In addition, it has become evident that the salt marsh farmers not only dug ditches on and along their terps, but also far beyond, and that they developed numerous activities in the peripheral zones of their settlements (Nicolay and De Langen 2015). This could mean that in large parts of the open salt marsh area, farming and nature were symbiotically linked (Nieuwhof and Schepers 2016). Recent experiments on the present salt marsh have shown that by making minimal adjustments to their landscape, such as digging (extra) ditches and erecting low dams, the terp farmers must have been able to increase their yields significantly. It can be assumed that the terp residents did indeed make these adjustments and practised mixed farming in a way that in many respects did not differ much from that of the Late Middle Ages,6 just like the West Frisians, who similarly did not have to cope with restrictions in farming that could have encouraged them to engage in distant trade (Dijkstra 2011). There is another indication that the Frisians east of the Vlie were not driven to active long-distance trade by a specialization in cattle breeding. That is the fact that, while in the regions to the east of the Vlie the conditions for farming were the same, Frankish gold coins and Frankish wheel-thrown pottery are not found to have been evenly spread over Central and Eastern Frisia. These finds, which can surely be associated 6 Schepers and De Langen 2019 (noting that further research is needed); see also Nicolay et al. 2018, 190–4. It is still to be investigated for what reason the inhabitants of the salt marshes started to use the terps more and more as crop fields after about 850 and also expanded them for that purpose: De Langen and Mol 2016.

87

Z01 Hines Book B.indb 87

07/05/2021 13:29

Frisians of the Early Middle Ages with trade, occur in significantly lower numbers in East Frisia than in Oostergo and especially Westergo (Nicolay 2014; on imported Merovingian pottery, Kaspers 2019). At first glance this seems strange; in the case of specialization in cattle breeding, the East Frisians would have been able to participate in trade as well as their neighbours to the West. So why did they not have so many imports? We shall come back to this later. Finally, it should be noted that for the entire Frisian area east of the Vlie few or no manorial structures of the High Middle Ages are found. Large royal landholdings, holdings of one or more Frisian abbeys, and high aristocratic families were also lacking there. This leads us to infer that there were no foundations for this in pre-Frankish times either. The establishment of the Church and the boundaries of Frisia Useful as it is to get a grip on the variation in the Frisian landscape by using a retrospective perspective for the manorial estates – by examining in which numbers they occurred where in both the High and Early Middle Ages – it is also worth the effort of examining the ecclesiastical-spatial structures from around 1000, and to see what they tell us, not about differences between the three Frisias but about their common border with the hinterland. Since it would take too much space to analyse the building up of the Church throughout Frisia between Sincfal and Weser, our focus is on the district of Oostergo, the eastern part of Central Frisia, to reveal the principles. These can then be tested, first for the Emsgouw in East Frisia, and then for Kennemerland and the Rijnland district, both in West Frisia. The history of the Church in Frisia in general shows clearly that in the Early Middle Ages the residential areas mainly coincided with the above sketched clay and sandy zones along the coast and the main rivers. In the tenth century this would change rapidly when the adjacent peatlands were systematically taken into exploitation on a large scale. The layout and demarcation of the oldest territorial parishes – an innovation from the late tenth century – allow us to determine the main boundaries of the pre-reclamation pagi as well, and subsequently those of Frisia as a whole (De Langen and Mol 2017; 2018). This section will illustrate this. The first churches in Central Frisia date back to the late eighth century, when the definitive establishment of Frankish power opened up the way for Christianization. It is plausible that with the foundation of places of worship at Staveren, Franeker and Dokkum, the Westergo and Oostergo regions received their first missionary posts (Fig. 4.3). Somewhat later, in the early ninth century, other churches were founded, in Westergo at Tzum and Bolsward, as well as on the island of Terschelling, and in Oostergo at Holwerd, Ferwerd, Leeuwarden and on the island of Ameland. These early church foundations may have benefited from the introduction of ecclesiastical tithes in the later Carolingian Period. However, the Church as an institution will not have stood strong yet. Twenty years after Charles Martel’s victory in the Battle at the Boorne in 734, there was still strong opposition to the new religion in Central Frisia, as witnessed by the fact that Boniface could not visit Westergo and Oostergo in 753 and 754 without an armed escort, which did not prevent him from being killed near Dokkum. In the following decades missionaries met a lot of resistance in the districts east of the Lauwers; this region only became part of the Carolingian Empire in 784 after Charlemagne had crushed the Saxon revolt. Viewed in this light, the Church had little time to build up its structure properly, because already in the course of the ninth century the political situation grew unstable 88

Z01 Hines Book B.indb 88

07/05/2021 13:29

Landscape, Trade and Power

Fig. 4.3  Territorial parishes and their churches c. AD 1000 in Westergo and Oostergo, their boundaries projected on the landscape situation c. AD 700. Map S. de Bruijn, G. J. de Langen and J. A. Mol, Province of Friesland/Fryske Akademy.

as a result of an internal dynastic power struggle in the centre of the empire and an increasing number of raids by Norsemen. These raids must have been disastrous for the churches in West Frisia, the first of which had already been founded in the first half of the eighth century (De Langen and Mol 2019). In the period when West Frisia was dominated by Danish warlords, the whole of Central and East Frisia seems to have become detached from its Continental centre. According to an overview of the erstwhile possessions of the church of Utrecht in West Frisia and the surroundings of Utrecht, compiled sometime after the Danes had gone, many churches and goods had been lost there. It can be assumed that this was also the case in Central and East Frisia. The Carolingian abbeys that held possessions in all of these areas will also have suffered considerable losses. In the second half of the tenth century, however, there was a recovery of the Church and a renewed and planned expansion, with the result that around the year 1000 churches were placed in an overarching diocesan system within which all arrangements with regard to pastoral care and the duties of the faithful could be organized territorially. Every four years, the bishop of Utrecht or his vicar visited the various regional church meetings; in other years, the management per district was in the hands of a dean. During these meetings the local or regional elite was present as standard, 89

Z01 Hines Book B.indb 89

07/05/2021 13:29

Frisians of the Early Middle Ages making it clear that Central Frisia was an important part of the diocese. Central Frisia was not exceptional in this; the system was implemented in other parts of the dioceses of Utrecht and Münster as well, in the same way and at the same time. The establishment of synodal jurisdiction or seend in Central Frisia was a vital part of this planned expansion of the church system. Recent study of the size and boundaries of the parishes in the sixteenth century, of the possessions and patron saints of the individual primeval churches and their daughter churches, as well as of their function within synodal jurisdiction, made it possible to ‘filter’ the oldest centres out of the more than 360 churches of Westergo and Oostergo in AD 1500 and to reconstruct their early parish boundaries (Fig. 4.3; De Langen and Mol 2017). If we analyse those earliest contiguous parish areas in relation to what is now known about the dynamics of the landscape, it becomes clear that they all correspond to, or fit into, river basins or parts of them. It also turns out that rivers provided connection in the hinterland but created division at their mouths. In Central Frisia we can best observe this at the Boorne, which upstream, in the peat bogs, brings all the reclamation areas on both sides of the main stream and on either side of its tributaries together in one ecclesiastical district under

Fig. 4.4  Territorial parishes and their churches c. AD 1000 in Eemsgo, their boundaries projected on the landscape situation c. AD 700. Map S. de Bruijn, G. J. de Langen and J. A. Mol, Province of Friesland/Fryske Akademy.

90

Z01 Hines Book B.indb 90

07/05/2021 13:30

Landscape, Trade and Power the earliest mother parish of Oldeboorn. Downstream, however, it turns into a tidal channel separating the pagus or district of Westergo from that of Oostergo, flowing in between a number of separate parishes with centres near small tributary waters which flow together into the Boorne. Also instructive is the situation in eastern Oostergo, where the River Lauwers at its downstream outlet into the Wadden Sea functioned as the boundary dividing the dioceses of Utrecht and Münster, but was connective in the hinterland, with the result that the later district of Achtkarspelen here became the sole part of Late-medieval Friesland that fell ecclesiastically under the diocese of Münster. The fact that hydrology determined the ecclesiastical administrative division that was created at the time is logical. Ensuring a good water flow was so important for agriculture that it brought the people together administratively and politically, and then, as a matter of course, on the church scene. If we test this ‘catchment area’ principle for the district of Emsgo (Fig. 4.4), which was incorporated by Liudger as part of a Frisian enclave into his Münster diocese, then it proves to be true. The church of Leer, situated on the spot where the large peat river Leda flows into an already wide, tidal Ems, could still serve as a mission cult centre for a very wide region before the Viking Period. With the further expansion and territorialization of the Church in the late tenth century, it was assigned most of the catchment areas to the east of the estuary, albeit that the Ehe system further north came to fall under the new church of Emden. At the same time, south of Emden and across the river Ems, the new primeval parish of Nesse was established to cover the residential areas on the western bank of the Ems and on either side of the River A system. Where the main watershed could not function as a border – in this case the continuous Ems upstream – the boundary had to be drawn halfway between the coastal clay area on the one side and the inhabited Pleistocene sandy soils on the other. In the case of the district of Emsgo, this global area boundary also coincided immediately with a diocese boundary (i.e. the boundary between Münster, Frisia and the diocese of Osnabrück). In the central part of West Frisia (Kennemerland and Rijnland; Fig. 4.5) it can be observed that the rear boundaries of the parish areas under the mother churches of Kennemerland, such as those of Velsen and Heiloo, located on the old sand barriers, coincide with the watershed over a number of north–south connected peat bogs on the central axis of the current province of North-Holland. Further south, this also applies to the eastern border of the Rijnland district. Remarkably, the Rhine separates parishes on both sides at the mouth of the North Sea but connects the territories behind it. This is proved by the shape of the late tenth-century parish of Leiderdorp, which stretched on both sides of the river. The pagus of Rijnland was thus a territorial entity on its own and certainly not an area that in the seventh and eighth centuries was still divided in two by a limes. This clashes with the image, still prevalent in literature but never convincingly based, that the Rhine as an ancient Roman border river would have formed the dominant boundary between Frisia Citerior and Frisia (propria) or Frisia Ulterior (see Halbertsma 2000, 77; cf. Rombaut 1990, 181).7 On closer inspection, we have to conclude that there is no indication of this. On the contrary, it may now be clear that Frisia, viewed from the Frankish point of view, was the North Sea land behind the moors, and Frisia Citerior, 7

The name Frisia Citerior is first used by Bede, Historia Ecclesiastica V, 10, where he describes the crossing of Willibrord to Frisia in 690.

91

Z01 Hines Book B.indb 91

07/05/2021 13:30

Frisians of the Early Middle Ages

Fig. 4.5  Territorial parishes and their churches c. AD 1000 in Kennemerland and Rijnland, their boundaries projected on the landscape situation c. AD 900. Map S. de Bruijn, G. J. de Langen and J. A. Mol, Province of Friesland/Fryske Akademy.

as the Dutch central river area temporarily or permanently controlled by Frisians, lay at this side of the peat wilderness. Internal differences within Frisia proper The Frisian residential areas along the North Sea were able to maintain good maritime contact with each other. Or, to put it more cautiously, they were dependent on each other via water connections, precisely because land traffic with inland areas was severely hampered by the peat wilderness. In this way, Frisia maintained contacts not only with continental Europe, by travelling the rivers, but also by crossing the sea with the inhabitants of other coastal regions around the North Sea, a situation that facilitated trade opportunities and will have had influence on the political position of the Frisian rulers. However, this does not alter the fact that the constituent elements of this coastal Frisia were diverse and that each region or pagus had a geopolitical value of its own, thanks to its specific location and accessibility. 92

Z01 Hines Book B.indb 92

07/05/2021 13:30

Landscape, Trade and Power A major factor was their particular location, stretching at both sides of the Scheldt estuary, which was of great importance for controlling the waterway traffic to and from Flanders and western Brabant. The pagus of Maasland, situated south and north of the mouth of the Meuse, had an analogous function, since all traffic up and downstream of the rivers Meuse and Waal, coming from and going to the German Rhineland and the Belgian Maasland, had to pass its shores. The (Holland) Rijnland district was of a similar importance for dominating the Oude Rijn as that other shipping route between the German Rhineland and England, while the Vecht region functioned as the gateway between the German Rhineland and the district of Kennemerland (via the river IJ), Central Frisia and Scandinavia (via Lake Almere and the Vlie). It follows that the districts of Texel, Wieringen and Westergo lying alongside the Vlie had their special geopolitical significance as well. The Frisian pagi east of the Lauwers could be reached from their hinterland, and vice versa, via the rivers Ems and Weser. This situation meant that Frisian trade, when seen as trade through (and along) the Frisian districts, was practically impossible to control from one specific centre. From Dorestad, a large part of the shipping could be monitored, which is also evidenced by the presence of a (Carolingian) central toll there, but by no means all traffic. From Dorestad it was possible to follow the movements on the Lek, but the traffic on the Waal and Meuse by-passed it. This separation of trade-flows will not have promoted central authority within Frisia, not to mention trade conducted via the Ems and the Weser. The observed internal landscape differences may also have had a negative influence on Frisia’s economic and political coherency. We have already noted that land traffic between the districts west of the Vlie was much easier than between those in Central and East Frisia. This meant, among other things, that West Frisia, and certainly the parts north of the Meuse estuary, had the potential to develop strong internal cohesion. And then of course there was the difference in the structure of the soil: West Frisia with its mainly sandy soil against the salt marshes of Central and East Frisia. The end of Frisia Ulterior and Citerior The originally heterogeneous character of Frisia, both of the actual coastal strip and in the combination with Frisia Citerior, conceived as the Dutch central river area, has become blurred by the incorporation of the whole into the Frankish empire. In Carolingian times, Frisia no longer existed as an autonomous political entity. It nevertheless remained culturally and legally recognizable, as indicated by Lex Frisionum. It can even be argued that with the flourishing of the Dorestad trade and the large share the Frisians had in it, this Frisia retained and further developed its name and significance between 750 and 850, especially as an economic zone. Considered in this way, the end did not come until the late ninth century, when a territorial economic and administrative division emerged and the old geographical features gradually faded. Whereas before 900 the Frisians mainly lived on the old beach barriers and salt marshes as well as along the rivers, they afterwards colonized the peat areas and by so doing rapidly and nearly automatically dissolved the old barriers between the Frisian districts and the hinterland. Finally, with the progress of the embankments after the turn of the millennium, Frisia, which was already divided over several Frisian ‘lands’ (or Frieslanden), would largely lose its maritime character.

93

Z01 Hines Book B.indb 93

07/05/2021 13:30

Frisians of the Early Middle Ages

Frisians and the emporia of Frisian trade Because they were settled along the North Sea coast, the Early-medieval Frisians are considered to have been a maritime people, travelling mainly by boat, both for short and long distances. This basic characteristic has led many historians to assume that these Frisians participated in long-distance trade and that this activity must have been intertwined with their daily economy. The ideal Frisian was a farmer and a seafaring merchant at the same time: the Frisians living east of the Vlie, especially, portrayed as cattle breeders needing imports, were thought to have plenty of time to go on trading journeys to peddle wares themselves. However, this picture of a long-distance trade economy linked to an agricultural economy based on livestock farming has been drastically altered in recent years, as just noted. In this section of our contribution we aim to defend the idea that the Early-medieval emporia – pre-eminently as hubs for long-distance trade – were not situated within, but rather outside the Frisian districts, at best near their borders, and that the daily economy of the coastal dwellers, both in West and East Frisia, was first and foremost agricultural in character. Long-distance trade via Frisia In the literature, the exchange of goods by and through Frisia in the period AD 650–900 is called ‘Frisian trade’. Given its long duration and the fact that this trade also included that of the Dutch central river area, it is a catch-all term that ignores the economic and political developments that took place during this time. It also disregards the possibility that not all Frisian coastal regions were equally involved in this trade. To get some grip on this issue we propose to examine the development of Frisian trade more closely and see whether its phasing was linked to political developments. On the basis of this we then would like to analyse to what extent the Frisians played a role in the trade that went through their districts, and whether we should speak of a ‘Frisian trade’ at all. On the basis of the political framework described above (and further explained in the last part of our contribution), the following timeline – useful in our discussion of the Frisian economy – can be given: -_

Before 650: Frisia stretches along the shores of the North Sea between the Sincfal and Weser and is then heterogeneous in landscape, economic and political terms. c. 630–650: Franks are dominant in the Dutch central river area; a church in Utrecht; coinage in Dorestad. c. 650–695: Frisians dominant in the Dutch central river area: Frisia is extended with Frisia Citerior; Kings Aldgisl and Radbod in Utrecht; Dorestad south of the Lek gets an extension north of the Lek; introduction of the early penny (= sceatt: see below). 697–719: Franco-Frisian alliance: Radbod is vassal in Frisia Citerior but retains his royal position in West Frisia; church in Utrecht restored, but the mission to the Frisians is unsuccessful: Willibrord focuses on areas of Frisia outside of Radbod’s control; continuation of trade and use of the early penny. 719–734: After the death of Radbod Frankish power only slowly penetrates into West Frisia: 722 war against the North; after 719 Utrecht strengthened, with a monastery founded by Willibrord; 719–721: mission of Boniface in Woerden, Attingahem, Nederhorst ten Berg or Breukelen (River Vecht area) and finally in Velsen ‘because this place was closer to the pagans’; continuation of trade and the early penny. 94

Z01 Hines Book B.indb 94

07/05/2021 13:30

Landscape, Trade and Power 733/4–754/5: Battle at the Boorne in 734, perhaps as a punitive expedition following a first encounter in 733; coin hoards c. 730; mission among the West Frisians is only now really promising; start of the heyday of Dorestad’s trade; end of the early penny due to the Carolingian coin reform of 754/5. 754/755–c. 850: Frankish conquest of East Frisia; Frankish consolidation of power in Frisia as a whole, expansion of the Church in in West Frisia, the first churches are founded in Central Frisia and after 785 also in East Frisia. c. 850–c. 922/950: The Viking Period; the foundation laid for the power of the Gerulfingen family in Rijnland. c. 922/950–c. 1000: The secure establishment of the Church: the introduction of a territorial parish system and ecclesiastical jurisdiction; extension of the power of the Gerulfingen as counts of West Frisia. -

According to all historical and archaeological sources, the following goods changed hands in the Early Middle Ages: slaves, metals (iron/iron ore, lead, tin, gold and silver), metal products (e.g. weapons), natural stone (precious stones, grinding stones, whetstones), wine, wheel-thrown pottery, glass, amber, pelts, ivory, woollen and linen cloth, salt, building timber, grain, fish, dairy products, meat, skins and leather and the products of artisanal activities for which wood, leather and antler were the raw materials used (Niermeijer 1977a; Verhulst and De Bock-Doehaerd 1981, 196; Lebecq 1983, I, 126–31; Van Es 1994, 106–7; Dijkstra 2011, 308–16; Henderikx 2014a, 74–7). The Frisian export products par excellence are thought to have been: wool, hides, cloth and salt (Henderikx 2014a, 76; on wool and cloth: below; on medieval saltmaking in Central Frisia: Nicolay and Vos 2010). Within Frisia there was a market for building timber (Dijkstra 2011, 308; Knol, this vol.). It is unlikely that all of these products were traded over long distances from the earliest Middle Ages. However, it requires further examination to determine when which product was in transit and in what quantities.8 It is generally accepted that, at least at the outset, long-distance trade focussed on easily transported luxury goods (Verhulst and De Bock-Doehaerd, 190–3; Van Es 1990, 169). This assumption seems to be confirmed by archaeological data from Central Frisia. What has been found suggests that in the treeless salt-marsh area to the east of the Vlie, house building still aimed at re-use and self-sufficiency for a long time. Until the tenth century the necessary timber seems to have been obtained from nearby regions at the most (Postma 2015). Self-sufficiency also applied to daily food; remains of imported food have hardly been found in Central Frisia (Schepers 2014). Nevertheless, complete self-sufficiency was not possible: metals had to be imported because these do not occur naturally in Frisia. It is therefore striking that in the Merovingian Period, Central Frisia was above averagely rich in gold items. There is no consensus on the explanation for this wealth and the fact that it did not spread evenly across Central and Eastern Frisia. Could it be interpreted as a sign that Westergo’s elite was more successful in participating in interregional and international political networks (Nicolay 2014)? Or was Westergo simply richer because it lay alongside the Vlie, which was the main traffic route between the Dutch central river area and Scandinavia 8 For example, no evidence has yet been provided that wine and grinding stones were already traded during the third phase of Dorestad (690/695–719/725): Van Es et al. 2015, 381.

95

Z01 Hines Book B.indb 95

07/05/2021 13:30

Frisians of the Early Middle Ages (Theuws 2018; contra Nicolay 2014)? Without immediately neglecting the role of the regional elite, we tend to think that the latter was the case. Powerful men from Wes­ tergo may have been in a good position to tap the passing trade by means of levying tolls and, at the same time, to trade agricultural surpluses more easily, in which case we think of livestock products, in particular wool and the cloth that was made from it. This Frisian cloth, as it is known from sources from the Carolingian Period, would then not have been called Frisian because it was traded by Frisians but because it was produced by them (on wool and cloth from Scheldeland: Lebecq 1983 I, 131–4; II, 382–3; followed by Verhulst 2002, 75–6, 110–11; Henderikx 2014a, 76; on wool from East Frisia as payments to the monastery of Werden: Noomen 1996b; cf. also Dirks 1846, in fn. 9 infra). Such a configuration could also explain why, around 800, wool processing was legally protected in Lex Frisionum, albeit possibly at the instigation of the Franks.9 Coincidentally, Westergo in particular, with its relatively extensive salt marshes, was ideally suited to produce wool: sheep thrive on salty soils, because these protect them from certain diseases. Recent excavations in Westergo have demonstrated the i­mportance of ­Early-medieval sheep farming (Çakirlar et al. 2019; Hullegie and Prummel 2015; Prummel 2014).10 A favourable economic location that provided a good connection to long-distance trade in combination with favourable conditions for the production of a surplus or a luxury product such as cloth can explain why Westergo in the Merovingian period boasted more imported wheel-thrown pottery than Oostergo and the other districts further east (Kaspers 2019). In the Carolingian Period such internal differences in the Frisian area east of the Vlie became less pronounced (Knol 1993). It is possible that this levelling process already started around 700. This suggests that the Frisian long-distance trade must have known different stages and that the impact of long-distance trade may have fluctuated per district. We must bear in mind that finds of imports do not necessarily mean that the people who used them actively participated in long-distance trade themselves. Westergo for instance may also have been visited by long-distance traders on their way to Scandinavia. Later on, when Dorestad developed regional market functions, the region may also have constituted a special target for merchants then living permanently in Dorestad. German archaeologists in particular have repeatedly underlined the important role that visiting traders may have played in exchange (Brandt 1977; 1984; Jankuhn 1980; 1984; Haarnagel 1984; Segschneider 2002; Siegmüller and Jöns 2012). Lex Frisionum, addition by Wlemar, de Pace faidosi I, 3: the penalty for destroying a screona. A legal provision interpreted as a sign of Frankish influence with the screona as a sunken hut, and specifically a weaving hut: Hilker-Suckrau 1980, 62–3. She considers the law text to be consistent with a screona interpreted as being a women’s weaving hut by, e.g., Gysseling 1976, 96. Also Lex Frisionum, addition by Wlemar erroneously published in 1557 as part of Lex Thuringorum VI, 24: the fine on hitting the hand of a woman who makes ‘fresum’: Feminae fresum facienti similiter. Fresum was already identified in the nineteenth century as a woollen cloth: Dirks 1846, 101, 133–7, where he also discussed the higher punishment for the violation of a bortmagad or house virgin, i.e., a handmaid who does not milk or grind but who – in Dirks’s explanation – was engaged in spinning, weaving or sewing (Tit. XIII de Stupro ancillarum). See further Nijdam, this vol. 10 With the exception of locations in the vicinity of the Vlie, West Frisia north of the Meuse may have produced less wool per settlement: cf. figures presented by Dijkstra 2011, 314–15 and appendix 8. In Scheldeland, conditions for sheep farming became increasingly favourable after the eighth century according to Van Dierendonck 2009; for Scheldeland see also Henderikx 2014, 74–6, who reports that cereal cultivation was already possible there in the ninth century. 9

96

Z01 Hines Book B.indb 96

07/05/2021 13:30

Landscape, Trade and Power These visiting merchants may have interconnected with local and (inter-)regional traders who on their own could also have taken care of the supply of foreign goods. These latter forms of exchange are constant. They served different purposes and had different forms depending on the distances covered and the status of the commodity, varying from maintaining the existing social and political order and satisfying basic needs (cf. Nicolay 2014; IJssennagger 2017). In Frisia, this traffic will mostly have been achieved by certain forms of redistribution or simply during short-term markets, at which also the more routine necessities from remote extraction areas or production centres were offered for sale (Heidinga and Verhoeven 1992). It is clear that this trade was still of low volume11 and could function without permanent trading settlements. So the question is not so much where in the Frisian districts during the Early Middle Ages the trading settlements were located, but whether they were there at all. Trade and the early penny This ‘through-Frisia’ trade, then, appears to have been scaling up strongly in the period after 650. The growth in trade volume required, and was subsequently stimulated by, the introduction of a silver currency. From c. 670 on, deniers (denarii) were struck in the Frankish Empire and early pennies, widely known as sceattas, in Anglo-Saxon England.12 It is generally assumed that early pennies were struck in Frisia as well, although this has recently been questioned (Theuws 2018 contra Naismith 2017, esp. 87–92, and Henstra 2000). The early silver penny is a small silver coin with a fluctuating average weight, albeit with a wide spread around that average (Metcalf and Op den Velde 2010, 87). Their widely varying weight suggests that these coins were weighed during transactions (Breternitz et al. 2017, 12–14). The early penny was minted in many variations and successive generations for about a century. Different production centres are known from both the official coins and their imitations, although in the case of the early penny it is not at all clear what should be labelled as ‘official’. It is assumed that the currency circulated in very large quantities and that these numbers may be seen as an indication of a highly developed monetary economy, in which it was used for daily purchases (Op den Velde and Metcalf 2003; Metcalf and Op den Velde 2010 cf. Blackburn 2003). Some historians regard the early penny as a high value currency, curiously enough by pointing out that up to twelve loaves of bread could be bought with only one penny.13 This argument seems anachronistic: why should people in a well-functioning exchange economy use money in daily transactions? It seems better to assume that the early penny was introduced as an alternative to the expensive gold coin. As such it meant an important step towards a more monetized economy. In fact, the new currency facilitated the exchange of goods over long distances, the purchase of slightly more expensive goods than the twelve loaves of bread mentioned (for instance cattle) and the storage of capital, to name but a few improvements (for further advantages, see Henstra 11 Even for a trading post such as Dorestad it is assumed that the volume of trade remained low until 719/725: Van Es et al. 2015, 283. 12 With regard to the general development of the early penny, we mainly follow Naismith 2017, also in respect of discarding the long-established term ‘sceatta’ for the early silver penny (ibid., 66–7). 13 Naismith 2017, 108 (too valuable to be used for day-to-day payments); Naylor 2012, 251 (high-value coins); Naylor 2007, 51 (a restricted, high-value currency) and 59 (no cash currency); the practicality of sceattas for daily purchases is also contested by Breternitz et al. 2017, 11. However Theuws 2018, 76, denies the high numbers, but refers to the early penny as a low-value denomination.

97

Z01 Hines Book B.indb 97

07/05/2021 13:30

Frisians of the Early Middle Ages 2000, 56–7). Reasons enough in any case to explain why the Early-medieval penny was almost immediately a great success. The role and position of the Frisians in Early-medieval international (trade) relations are clearly shown by the distribution of the early penny, more in particular of the early types of the Series D (Continental Runic Type c. 690 – 710/715) and E (Porcupine Type c. 690 – c. 755), whether the Frisians stuck these types themselves or not (Op den Velde and Metcalf 2003; Metcalf and Op den Velde 2010; dates from Naismith 2017, 86). At the moment it is still difficult to interpret the distribution of the early penny finds throughout the whole of Frisia, because in general the find circumstances vary considerably from one district to another and even within districts (cf. Bazelmans, Gerrets and Pol 2002). The fact, however, that in this case the circumstances in Central and Groningen East Frisia are very similar strongly suggests that the coin distributions there may well reflect Early-medieval coin traffic. The pattern resembles that of the gold objects from the sixth and early seventh centuries, in the sense that fewer coins are found the further one moves away from the Vlie. It should be noted, however, that over time the areas to the east of the Lauwers received more gold coins (cf. Nicolay 2014, 99). It is not impossible that this improvement continued during the period of the early penny.14 Add to this the fact that large quantities of early pennies have been found in the Dutch central river area, the estuary of the Rhine, and near Domburg in Scheldeland, and it can be concluded that, as in Anglo-Saxon England, also in Frisia, the early penny functioned mainly for, or in relation to long-distance trade (Series D: Op de Velde and Metcalf 2003, 109–12; Series E: Metcalf and Op de Velde 2010, 269; Theuws 2018; for comparison with England: Naylor 2007, 47 and 58). As in England, the pennies of the so-called Primary Phase (c. AD 680–715) cannot be or are not exclusively linked to emporia.15 Series D has even been explicitly associated with Central Frisia, where such centres were absent (Op de Velde and Metcalf 2003, 90: the majority of Type 2C produced in the present province of Friesland/Fryslân, and 94 : Type 8 possibly produced in Friesland). This may be different for Series E from the Secondary Phase, as that type may have been produced at Domburg (Metcalf and Op de Velde 2010, 156; for comparison with England: Naismith 2017, 109; Naylor 2012, 250 and 254–61). It is difficult to link this development of the early penny and its distribution patterns to politics, if only because the political developments in themselves are not easy to interpret either. In an attempt to make some progress we follow our own phasing of Frisian-Frankish relations. Until 697, as we will argue, the Frisians controlled the Dutch central river area. After that year the Franks were in control, although Radbod stayed in power, not only in his Frisian districts but also in the region of Utrecht and Dorestad, as discussed below. This transition took place during the continued use of Type D. In 719 Radbod died, after which the Franks gradually took control in Frisia west of the Vlie. This happened during the continued use of Type E. However, it is quite possible that the Frankish influence in West Frisia only became completely dominant in 733/734, when 14 This is our primary impression of the preliminary data kindly provided by Portable Antiquities of the Netherlands. 15 In relation to Series D: Op de Velde and Metcalf 2003, 82–96. With Series E it may be different, although no link with Domburg or Dorestad can yet be demonstrated: Metcalf and Op de Velde 2010, 145. For comparison with England: Naismith 2017, 107–19; Naylor 2012, 248–9.

98

Z01 Hines Book B.indb 98

07/05/2021 13:30

Landscape, Trade and Power the first successful military actions were organised to penetrate into Central Frisia and secure the incorporation of West Frisia into the Frankish realm. It still not very clear whether the Frankish victory in the Battle at the Boorne in 734 caused the ending of Frisian minting.16 The hoards of the 730s may reflect the political tensions concurrent with these events, but in all probability Type E remained in use even then. In West and Central Frisia the coin lost its importance only around the middle of the eighth century, about two decades after 733/734. It is possible that the decline of the early penny from the 740s was caused by a decline in international trade (Naismith 2017, 108–9; Naylor 2012, 47–50) and that the devaluation of Type E was one of the reasons for King Pippin III to decide on the currency reform of 754/755. These signs of discrepancy between coinage and political developments may indicate that the early penny was not minted in Frisia or that the production of the coin in Frisia was free of royal interference.17 Indeed, it may very well be that, until the middle of the eighth century, the Frisian and later Frankish leaders hardly had any direct influence on the coinage or even on the Frisian trading as a whole and were satisfied with the revenues from the toll. For example, Carolingian coinage at Dorestad was not resumed until around 754/755. In other words the production of the early penny in Frisia may have been organized at a local level and more ad hoc, also in view of its variability and anonymity (cf. Theuws 2018, e.g. 56). This possible local coinage may not have been entirely new, given the assumed earlier Frisian imitation of gold coins (Henstra 2000, 45–57; Pol 2008). Of course, it is not yet exactly clear who commissioned the production of coins: merchants, local lords, or even both. This possibly autonomous character of the Frisian long-distance trade, also known from both sources and literature (Lebecq 1983; Loveluck and Tys 2006), deserves further study. The emporia Dorestad is considered to have been the main centre of Frisian trade. Located at the junction of the Kromme Rijn and Lek, next to the modern town of Wijk bij Duurstede, it was one of the most important emporia of northwestern Europe at the time, with a key role within the trade network around the southern North Sea and the English Channel. Dorestad connected this network with the German Rhineland. In the period 650–695 it was situated on the Frankish border with Frisia and the rest of the non-Christian world. The Frisians had probably taken it over from Frankish-orientated local lords, although it is also possible that its trading section only came into existence after 650, as a sort of innovation. In that case, Frisian rulers would have founded it, next to the old nucleus, in or near the old Roman castrum Levefanum which was also situated on the southern bank of the Rhine but on the other side of the River Lek. Within this old centre, around 640, the moneyer Madelinus minted his coins (Van Es et al. 2015, 379–80). Although Dorestad changed hands around 697, the development of its economy remained stable. This is not surprising: both the Franks and the Frisians – or should 16 Breternitz 2018 is of the opinion that the coinage in West and Central Frisia came to an end around 730, but does not consider a link with the events of 733/734 to be proven. 17 Henstra 2000, 46: ‘…we do not know whether they [the Frisian kings] assumed anything like control over the money in Frisia’; Theuws 2018, 46 and 48; cf. Naismith 2017, 109, and Naylor 2012, 265–6; contra Metcalf and Op de Velde 2010, 280, who argued that Series D was the coinage of the Frisian king Redbad.

99

Z01 Hines Book B.indb 99

07/05/2021 13:30

Frisians of the Early Middle Ages we say the traders concerned in the first place – had the greatest interest in the continuation of its trading function. Nevertheless, trade via Dorestad experienced its heyday between (720) 750/775 and 830/840 (850), only after the greater part of Frisia was incorporated into the Frankish empire. After 695 and even more after 719 the site was no longer located in a border zone. By then, it also assumed regional market functions, which could explain the diminution of the internal differences in Central and East Frisia, although its international trade function remained important. Dorestad certainly did not focus only on transit trade but also fostered industrial activities. Its artisans and traders found important outlets in the western and northern Netherlands (Van Es 1990). From then on, extra checkpoints or trans-shipment centres may have been needed at the new borders of the Frankish area, which in this case meant new posts near the mouths of the rivers Oude Rijn, Vecht and (or along) the Vlie. Although such a centre may well already have been functioning at this time at Rijnmond, its name has not been handed down and its location – possibly on the bank zone of Katwijk-Valkenburg (Dijkstra 2011, 326–9) – is difficult to pinpoint. Whether the historically known ninth-century portus on the former estuary of the IJ near Egmond in Kennemerland was actually such a trans-shipment centre with eighth-century roots is not known. It could very well have been one (Vita Adalberti, ch. 10). After the Frankish incorporation of Westflinge (the pagi of Kennemerland, Texel and Wieringen and the region of Medemblik combined) in 719 (or when it became effective in 733/734), there may have been a need for a new trading post in Medem­ blik, at the spot where the salty tidal stream of the Vlie connected with the freshwater drainage system of the Almere. Staveren, as a counterpart of Medemblik at the other side of the Vlie, could only start as a satellite Frankish trading settlement after the Franks had pacified the western part of Central Frisia in 734. At the point where the Vecht flowed into Lake Almere, the forerunner of Muiden (Almere) is also counted as a Carolingian trading post (Verkerk 1992, 43–4). It must be noted, though, that the early trading function of Medemblik and Almere-Muiden is deduced from a report in a later Carolingian document hinting at the levying of toll in these places which is not confirmed in other sources.18 Unfortunately, Almere-Muiden and Early-medieval Sta­ veren were washed away during the High Middle Ages, so that archaeological research can no longer give a definitive answer about their earlier trade function. However, the significance of Medemblik in this respect has been proved archaeologically. Its take-off as a trading post is placed around the middle of the eighth century, based on the fact that most Early-medieval finds date back to the period 750–900 (Besteman 1989, 16).19 18 In 753 Pippin III reconfirmed the earlier donations of one-tenth of all royal possessions within the catchment area of the church of Utrecht to this church (Blok 1979, 49–50). That this donation related to, amongst other sites, a domain in Medemblik can be deduced from the property list of the church of Utrecht: ‘In Medemolac. regalis decima’ (Henderikx 1986, 551), at least if we can read this tithing as the proceeds of a domain. However, it is a step too far to deduce from the property list that it would (also) be a tenth on a toll at Medemblik, let alone that it would follow that Medemblik started out as a trading post (contra Henderikx 1986, 545, and especially Besteman 1989, 8–9, and Van Leeuwen 2014, 319; cf. Dijkstra 2011, 324). This also applies to Almere-Muiden. 19 The oldest layer rich in imports dates from around the middle of the eighth century. This layer consists of a man-made deposit of clay that covers the oldest traces of habitation (trenches, pits and wells) dated to 675–725/750 (Van Leeuwen, 82 and 131). It is therefore quite possible that Medemblik was set up as a trading settlement as late as 750, within or close to an extant agricultural settlement system.

100

Z01 Hines Book B.indb 100

07/05/2021 13:30

Landscape, Trade and Power As with Medemblik, Staveren and Almere/Muiden, there are no reports on the toll functions of Witla in Maasland and of Domburg in Scheldeland, possibly because the income from these tolls was collected via the central toll in Dorestad (Verkerk 1992, 45–6, following [on Domburg] Blok 1979, 85). Although we do not know a lot about these more southern trading centres since they too have been washed away, their importance should not be underestimated. Even Witla’s exact location is unclear, but it is suspected to have been directly west of the place where the local peat bog river Widele flowed into the Maas (Hoek 1979, 117 and 130; cf. Dijkstra 2011, 324). Thanks to the Annales Fuldenses we at least know that it was indeed an emporium, rich enough to be plundered by the Vikings in 836. As to the significance of Domburg, or Walacria as it was called, it can be assumed on the basis of a series of finds made on the beach of Domburg, including a large number of coins (Op den Velde and Klaassen 2004) and the remains of houses and three burial sites (Van Heeringen 1995, 42–5), that it was a large emporium, situated next to an important royal estate (Henderikx 2014a, 74 and 76). It possibly had a counterpart on the other side of the Scheldt, on Schouwen, in the ninth century near the villa Scaltheim, which disappeared into the sea (Henderikx 2014a, 76–7). In all likelihood these posts served trade-flows that could not be controlled directly from Dorestad. Although Witla was connected to Dorestad via the Lek, it was also a passage point for merchandise shipped directly from the Frankish Rhine and Meuse area via the Maas and the Waal along the Maas estuary to North Sea destinations and vice versa, while Domburg and Scaltheim were centres for trade between England and the Scheldt river basin, a traffic stream that was logistically entirely separate from the one controlled by Dorestad.20 That these places are reported to have flourished in Carolingian times does not mean that they did not have older roots. As the oldest medieval finds of Domburg date from the sixth century, it is quite possible that its trade function also went that far back. The presumed trading settlement next to Scaltheim on Schouwen may, given the finds of early pennies found, go back up to about 700, although this is disputed (Henderikx 2014a, 76–7, contra Van Heeringen 1995, 48). Witla will also have functioned as an emporium, but due to the early erosion of the site this can no longer be proved. A late-seventh-century mention refers to a donation of a farm and therefore says nothing about the suspected trading function (Dijkstra 2011, 324). However, not all trading settlements from the Carolingian era may have had predecessors on the spot. Medemblik, for example, only started about 675 and was at that time only an agricultural reclamation settlement.21 These emporia all seem to have emerged next to an agrarian domain. This brings us to the commercial activity carried out near agrarian complexes and single settlements and to the question of whether the owners or inhabitants of these places actually 20 Dorestad itself seems to have been mainly focused on the Rhine trade: cf. Blok 1986, 170, noting that in the Ottonian era, the bishop of Utrecht covered the entire Rhine trade (but no more than that) by having tolls in Muiden, Medemblik and Zaltbommel. Blok seems to suggest that these tolls replaced former tolls collected at Dorestad. 21 Contra Besteman 1989. Given the early donation of a domain, there was probably a Frankish confiscation or takeover of an older domain that, if it is to be sought in or near Medemblik, may date from the late seventh century given the archaeological discoveries (ibid., 16). It seems that the reclamations started in the late seventh century, and Medemblik then became the central settlement (ibid., 9 and 21; cf. Van Leeuwen 2014, 170–1, 207).

101

Z01 Hines Book B.indb 101

07/05/2021 13:30

Frisians of the Early Middle Ages participated in trade themselves; or employed traders; or acted more passively by receiving (professional) traders. If we leave the agrarian economy of the large domains aside, we can imagine that the commercial activity by the average agrarian settlement was negatively affected by the rise of the emporia. It is assumed that (most) traders who used to travel to Dorestad during the spring and summer season began to settle permanently there after 695 (Van Es et al. 2015, 379–84). This would imply that from then onwards the Frisian regions received more goods passively. Nevertheless, it can by no means be ruled out that the absolute decline in the number of traders who remained in the Frisian regions was actually small, precisely because their number had always been low. And also because the merchants of Dorestad were certainly not all ethnic Frisians – some of them were Franks (Lebecq 1992, 11), which opens up the possibility that the inhabitants of the Frisian regions were far less active in ‘Frisian’ trade than has always been assumed. In this context, reference can be made to recent research into the function of Early-medieval settlements near Valkenburg (Dijkstra 2011, 140–55) and Oegstgeest (Dijkstra 2011, 134–9), at Leiderdorp (Dijkstra et al. 2016; Dijkstra and Verhoeven 2019) and in Leidsche Rijn near Utrecht (Nokkert et al. 2009), settlements located along the main Rhine route for long-distance trade: these must have come into contact with the trade, but are found to have remained agricultural in character. Trading patterns changed again in the ninth century. Dorestad’s trade fell sharply after 830/840 (850), not only because of devastating attacks by Vikings, but also as a result of hydrological alterations and changes on the economic and administrative level. New, resilient and more locally controlled trading places emerged, such as Tiel on the Waal and Deventer along the IJssel. In particular, the traffic on the increasingly important River IJssel would become of growing importance for Central Frisia, making Deventer a central storage and trans-shipment location. In a number of cases, local landscape changes made an end to emporia. Places like Domburg, Witla and ­Almere-Muiden were washed away around 1000; their (possible) trading functions had to be transferred to Middelburg, Vlaardingen and Muiden. Dorestad suffered as a result of the diminishing water flow in the Oude Rijn and ultimately lost its function as a trading post as well (Van Es et al. 2015, 377, 379–80, explicitly noting discontinuity between Dorestad and Wijk bij Duurstede).

Frisian trade and Frisian farmers To what extent is this redefined image of Frisian trade as a stand-alone structure confirmed by data from the Frisian regions themselves? What traces of trade can actually be expected, given the probability that the annual volume of Early-medieval trade within the Frisian pagi was initially low?22 To temper our expectations we also have to take into account the point that the local branches of long-distance trade were controlled in a certain way by local leaders. The latter could offer protection to traders and (also) participate in trade activities. The more stable their power was, the more likely that exchange of goods and commodities would flourish under their rule. Nevertheless, such involvement of local leaders would not directly determine the main spatial structure of the settlement of the leader in question. Thus we could expect trading activities 22 The total number of annual shipping movements needed to supply the Frisian districts with wheelthrown pottery were surprisingly low, according to Van Es and Verwers 1994, 188.

102

Z01 Hines Book B.indb 102

07/05/2021 13:30

Landscape, Trade and Power to have taken place at several places within the Frisian territories, but not necessarily trading settlements as such. In a search for local trade we also have to realize that small Early-medieval trading places are difficult to trace because of the fact that the distribution system was layered and, for some products, also multi-staged. It should also be kept in mind that a lot of merchandise did not stay at the place of transit or sale but found its way into the dispersed settlements. Indications of the production of luxury goods point to the high status of a local leader or simply to the wealth of the residents. At best they inform us indirectly about the trading function of the place, since, for example, the production of precious jewellery and coins may normally have been undertaken by itinerant specialists. So, although traces of specialized activities of a non-agricultural nature are found within most Early-medieval agricultural settlements, such traces cannot simply be used as evidence for one or more special economic functions of the settlement measured at the supra-local level. This means that we must take a closer look at the agricultural landscape as a whole. We do this by analysing the way in which the landscape and agricultural settlements were structured, and subsequently by determining the period or phase in which local and regional professional traders obtained their own separated dwelling areas. Our main sources consist of historical maps and recent archaeological information about how terpen developed following the arrival of new inhabitants in Central and Groningen East Frisia in the fourth and fifth centuries. The agrarian structure of the settlements within the Frisian district In the clay area east of the Vlie the pattern of land parcellation has traditionally been adjusted to the natural relief. Because this relief has been fixed in the clay area since the large-scale embankment works of the late eleventh or the early twelfth century, it can be assumed that the main structure of parcellation in both the clay and the peat areas dates back to the High Middle Ages at least. What also makes the clay area suitable for analysis is that the settlements have hardly moved, for the simple reason that the clay soil remained fertile and people were not compelled to leave their original homesteads as in the sandy regions elsewhere. In addition, the terpen kept many settlements in place. Thanks to the digitization of the oldest cadastral plans (1812–32) for the processing of maps and register data in a GIS (www.hisgis.nl), it is much easier than before to carry out spatial research into various economic and administrative developments. When the spatially processed property and boundary reconstructions for the Early-modern period are tested against medieval texts and archaeological sources, older, Late-­medieval patterns can be uncovered. Then it also becomes clear that some of their roots reach deep into the Middle Ages. For example, the spatial layout of the landscape as reflected in early nineteenth-century parcel boundaries, provides insight into both the habitation history and the causes and factors behind the development of the specific parcel patterns (De Langen and Mol 2016). By ignoring all kinds of intermediate forms, two main versions can be distinguished: regular strip-field parcellation in the former peat areas and block-field parcellation in the clay districts (Fig. 4.6). This block-field parcellation can then be divided into a regular and an irregular version, in which ‘regularity’ is mainly determined by the extent to which plots with right angles occur within an certain area. By presenting these versions as separate types, some historical geographers have suggested that these forms of parcellation were applied or 103

Z01 Hines Book B.indb 103

07/05/2021 13:30

Frisians of the Early Middle Ages

Fig. 4.6  Distribution of different forms of field parcellation in Central Frisia and Groningen East Frisia. Reconstruction on the basis of the oldest cadastral plans (1812– 1832). Map S. de Bruijn, G. J. de Langen and T. Sibma, Province of Friesland.

even sought for their own sake. However, when the different versions are plotted on modern elevation maps, it becomes clear that in most cases the patterns can be traced back to the natural relief: where large and straight higher ridges dominate the terrain, the division is rectangular (Fig. 4.7), where such ridges are lacking and traces of former winding tidal streams in the form of small-scale depressions are predominant, an irregular field pattern is most common (Fig. 4.8). Just as, for a long time, irregular block-field parcellation was seen as the oldest and most common type of division of the entire salt marsh, the radially subdivided circular dwelling mound (like that of Hogebeintum in Oostergo: Fig. 4.9) was until recently regarded as the ideal type of terp settlement on the unembanked clay soils of the northern Netherlands. The world-famous excavation of the mound of Ezinge in the years 1923–34 (Van Giffen 1936; followed by the impact of the excavations of the mound of Feddersen Wierde in Land Wursten: Haarnagel 1979) greatly influenced twentieth-century thinking on landscape patterns in the salt marshes. However, further analysis of the historical data shows that this type of settlement with its pie-shaped parcels originated only in areas with irregular block-field parcellation and then only when the settlement counted five farms or more. If the number was less than five, then a rectangular layout was applied. The boundaries that subdivided the settlement radially ran as far as the surrounding salt marsh. Thus the farms were given the same spatial set-up, each owning a proportional share in both the fields on the mound and the adjacent higher and lower lands, so that the possibilities for arable farming and animal husbandry were 104

Z01 Hines Book B.indb 104

07/05/2021 13:30

Fig. 4.7  Modern elevation map (www.ahn.nl) of a part of Westergo in 1832 according to www.hisgis.nl showing regular block-field parcellation. Terps in red. Map S. de Bruijn, Province of Friesland.

Fig. 4.8  Modern elevation map (www.ahn.nl) of a part of Westergo in 1832 according to www.hisgis.nl showing irregular block-field parcellation. Terps in red. Map S. de Bruijn, Province of Friesland.

Z01 Hines Book B.indb 105

07/05/2021 13:30

Frisians of the Early Middle Ages

Fig. 4.9  Hogebeintum in Oostergo in 1832 according to www.hisgis.nl . Red line: size of the terp (on the basis of www.ahn.nl). Map S. de Bruijn, Province of Friesland.

Fig. 4.10  Firdgum in Westergo in 1832 according to www.hisgis.nl . Red line: size of the terp (on the basis of www.ahn.nl). Map S. de Bruijn, Province of Friesland.

the same for all the farmers concerned. On the higher and larger salt marsh ridges the parcel structure took a different shape, following the same principles but resulting in straight layouts, not only of the land but of the settlements as well (Fig. 4.10). On the whole it appears that the layout of the landscape did not result from the layout of the settlements, but that, conversely, the layout of the settlements was determined by the natural form and cultural interpretation and following subdivision of the broader landscape grid (De Langen and Mol 2016, 108–10). Like the general parcel structure, the settlement patterns can be traced back far in time. But just as ditches could be added or removed without violating the laws of nature, 106

Z01 Hines Book B.indb 106

07/05/2021 13:30

Landscape, Trade and Power

Fig. 4.11  Biessum in Fivelingo in 1832 according to www.hisgis.nl . Red line: size of the terp (on the basis of www.ahn.nl). Map S. de Bruijn, Province of Friesland.

settlements were subject to change also. One of the best known developments resulted from the embankment works. From the twelfth century, when sea-resistant dikes were considered reliable, more and more farmers decided to leave the old dwelling mounds and move their houses just beyond the edge of the terp or even further, to a spot amidst their fields at some distance from the old settlement. In this way a new settlement pattern was created. Some of the older dwelling mounds were even completely abandoned. Along the banks of the dangerous Ems estuary, though, the risk of settling at ground level was considered too great and the mounds kept all their inhabitants, as was the case at Biessum, a terp settlement that still around 1832 housed no fewer than nine farms (Fig. 4.11). In 1832, similar radially subdivided mounds, like those of Foudgum and Hogebeintum, only had four or five farms. It seems that some of the historical farms that belonged to their village area but were located outside the terp since Late-medieval times, were still standing on the mound in the tenth century. What is more striking about the Biessum terp is that it covered a much larger area than was strictly necessary for the protection of body and property. It also counted small fields and gardens. The dwelling mound that also offered space for fields and gardens even at full occupation appears to have been an innovation that dates back to the middle of the ninth century. In the tenth century a lot of effort was put into the transformation of separate living platforms into one communal complex of raised arable fields, the co-called akkerterp (crop-field terp). Around the middle of the eleventh century this movement passed its peak, although the terps were still being enlarged for agricultural use. Following the edge of the ever growing terps, in the course of time the successive farms moved ever further from the centre of the old dwelling mound (De Langen and Mol 2016, 102–8). Before 850, the farms were in general located significantly closer to or even at the top of a terp or on separate house platforms when they were built on the edge of an older terp or just beyond it. Like their Early-modern successors, these Early-medieval 107

Z01 Hines Book B.indb 107

07/05/2021 13:30

Frisians of the Early Middle Ages farms seem to have been spatially arranged according to the layout of the surrounding landscape. At Foudgum (De Langen 1992; Fig. 4.12) and Wijnaldum (Gerrets and De Koning, 1999; Fig. 4.13), for example, the Early-medieval houses at least had the same orientation as the later ones. We can therefore conclude that in the terpen area of the northern Netherlands the main structure of the settlements was determined by agricultural factors. The excavation data do not contradict that this was also the case in West Frisia, where even settlements along the Rhine – although equipped with quays or moorings – were mainly agrarian (or should we say remained so), as noted above. This is a warning not to interpret similar indications for traffic on water, which were indeed found next to some terps, as evidence of active long-distance trade by the Frisians living east of the Vlie. The boat-owners in question were most likely farmers who used their vessels primarily for their own agricultural operations: for example, for the transport of hay

Fig. 4.12  Foudgum in Oostergo in 1832 according to www.hisgis.nl. Inner red line: size of the Roman-period terp; red rectangle: house from the Carolingian Period; outer red line: size of the medieval terp (on the basis of www.ahn.nl). Map S. de Bruijn, Province of Friesland.

Fig. 4.13  Wijnaldum in Westergo in 1832 according to www.hisgis.nl . Red line: size of the terp (on the basis of www.ahn.nl). Red rectangles: possible Early-medieval houses: the younger the darker. Map S. de Bruijn, Province of Friesland.

108

Z01 Hines Book B.indb 108

07/05/2021 13:30

Landscape, Trade and Power and livestock and for local and regional exchange and communication. The most practical means of transport for that traffic were small vessels, which were not suitable for navigating the main sea routes, but were easy to manoeuvre in the shallow and narrow waters close to home. The introduction of villages and towns Within this fully agricultural environment, the traders were eventually provided with their own centres. By dating the origins of these centres, we try to find out when and how non-agricultural units really did appear in the terpen area. In doing so, we also attempt to get an impression of their size in order to model the importance of active trade by the inhabitants of the Early-medieval Frisian districts between Vlie and Ems and ultimately the whole of Frisia. The rise of most towns in former Frisia is generally dated only to the thirteenth century. Of the places ranked as the oldest cities, the nearest ones lie on the edge of Frisia or just beyond. Usually Vlaardingen, Dordrecht, Utrecht, Muiden, Medemblik, Staveren and/or Groningen in the present Netherlands, and Ghent, Bruges and Antwerp in Belgium are classified as such (Fig. 4.14). The question is not so much whether and to what extent pre-urban status is justified for each of these places as whether it can be stated that in the Frisian districts no (pre-)urban centres functioned before 1200. To avoid getting caught up in a dispute over the definition of the concept of city, we would simply point out which Frisian settlements are found among the mint centres of northwestern Europe in the tenth and eleventh centuries. The picture looks different then. In Frisia we encounter, besides Medemblik and Staveren, the eleventh-century mint sites of Bolsward in Westergo, Leeuwarden, Dokkum and Oldeboorn in Oostergo, Winsum in Hunsingo and Garrelsweer near Loppersum in Fivelingo (Henstra 2000, 81–106; Albrecht 1959). In view of their location these pre-urban centres were able to mediate between the long-inhabited salt marshes and the new settlements in the peat districts.

Fig. 4.14  Eleventh- and twelfth-century (pre-)urban centres in or near Frisia projected on to the landscape of c. AD 700. Map S. de Bruijn, G. J. de Langen and J. A. Mol, Province of Friesland/Fryske Akademy.

109

Z01 Hines Book B.indb 109

07/05/2021 13:30

Frisians of the Early Middle Ages The choice of founding a mint in these places obviously followed the development of new economic conditions in the regions concerned from the tenth century onwards. In this context it is natural to mention the West Frisian cities of Alkmaar, Haarlem, Leiden (the successor of the eleventh-century mint site of Rijnsburg) and Delft, the emergence of which as embryonic economic centres under the protection of the count of Holland can also be placed before 1200. Some of them even date back to the eleventh century (Alkmaar: Bitter 2014; 2016; Haarlem: Numan 2014; Leiden: Brandenburgh and Orsel 2014; Rijnsburg: Dijkstra 2011, 114–33, 326–8; Delft: Bult 2014, 126–59). They had a similar location to the aforementioned Central Frisian mint sites. In the eleventh century, Middelburg had become the central trading post (portus) in Scheldeland south of the Scheldt. Its trading function may go back to the tenth century. North of the Scheldt Zierikzee followed with some delay; it was not until the twelfth century that it was referred to as a portus (Henderikx 2014b, 99–100). In the present German area of East Frisia the cities of Emden, Leer and Jever may be counted among the oldest towns, acting as mint places in the eleventh century (Henstra 2000, 81–106). The birth of the eleventh-century mint places in Central Frisia and Groningen East Frisia not only followed the reclamations but also the formation of the first parishes. The named mint places facilitated both long-distance trade and exchanges within their own region. As market places each also served an area that included several primeval parishes. Nevertheless, not all local merchants and craftsmen established themselves in or near these new regional centres. Small non-agricultural nuclei also came into being under the protection of early churches elsewhere in the districts, resulting in a large number of villages in Central and East Frisia which together accounted for an important part of the regional economy. To model their earliest functioning we have to rely on church-historical data and archaeological information about the terpen on which the villages are located, since most of these villages have not been researched archaeologically. On this basis it can be assumed that real village formation only occurred when trade and industry established themselves next to or near a church, that is, where the craftsmen were given a place and legal protection on a stable basis. The houses of

Fig. 4.15  Sexbierum in Westergo in 1832 according to www.hisgis.nl . In red line and on the basis of www.ahn.nl: size of the church terp with its extension for the oldest settlement (on the right) and the agrarian terp on the left. In the middle the secondary settlement on the non-elevated, natural level. Map S. de Bruijn, Province of Friesland.

110

Z01 Hines Book B.indb 110

07/05/2021 13:30

Landscape, Trade and Power traders, craftsmen and also workers, which were constructed on relatively small yards, together formed small and densely built-up neighbourhoods (De Langen and Mol 2016, 116–24). The fact that most of the new churches of the tenth and early eleventh centuries were not erected on top of the akkerterpen but on separate platforms next to them (Fig. 4.15), strongly suggests that at that time raised arable land was still considered to be of high value. By then the majority of the mounds were still essential for the cultivation of oats and barley, as also follows from their continued growth. It was not until the middle of the eleventh century that churches – the older missionary centres are left out of consideration, since their place had been determined earlier – were placed centrally on the akkerterpen. It is striking that not only the new churches of the tenth and early eleventh centuries but also their non-agrarian neighbourhoods came into being on (new) terp layers. Although these neighbourhoods were younger than their churches, they originated either on an extension of the church terp or on the very edge of the older agrarian terp, but never on top of an akkerterp. The fact that they still needed protection against flooding induces us to conclude that these old neighbourhoods also date from the High Middle Ages. Most of the younger ones though were developed on unraised or slightly raised grounds: for example, alongside a road or a newly dug canal, a process that mirrors the effects of the ongoing embankment works. The concentration of trade and industry within a separate space is a development that left its historical traces in Central and Groningen East Frisia only in neighbourhoods near a church. This and the notion that the Church could provide stable protection only after the Viking Age, that is to say after c. AD 950 (De Langen and Mol 2019), seem to indicate that these historically known non-agricultural neighbourhoods did not develop earlier than the tenth century, when the old mission centres were rebuilt and a lot of new churches were founded. Only a few of the non-agricultural nuclei have been archaeologically researched, but the first results seem to confirm this. The best example is offered by the village of Oldeboorn. The development of its neighbourhood, situated on its own elongated terp (Fig. 4.16), did not precede but followed the

Fig. 4.16  Oldeboorn in Oostergo in 1832 according to www.hisgis.nl . In red line: size of the church terp and the terp of the settlement (on the basis of www.ahn.nl). Map S. de Bruijn, Province of Friesland.

111

Z01 Hines Book B.indb 111

07/05/2021 13:30

Frisians of the Early Middle Ages foundation of the church in the late tenth century. Though at present it cannot be ruled out that neighbourhoods came into existence near the missionary churches as early as the ninth or even the eighth century, such a great age has not been demonstrated in any case yet. In this respect it is very important in that even the settlements of Bolsward, Dokkum and Leeuwarden most likely originated in the tenth century (De Langen and Mol 2019; Bolsward: Varwijk et al. 2019; Dokkum: De Langen 2007; Leeuwarden and Oldeboorn: De Langen 1992 and 1999). Exit the Carolingian ‘trading terp’ These observations have important consequences for how we discuss the trade conducted by the Early-medieval Frisian districts. Supplemented with other data they call for the rejection of the theory of the Early-medieval ‘trading terps’ (Dutch: handelsterpen; German: Langwurten), according to which trading settlements were already active in the Frisian regions at the time of the Dorestad trade. As mentioned above, the excavations in post-war Emden led Haarnagel to suppose that Emden had begun in Carolingian times as a trading centre in the form of a non-agrarian street settlement situated on an elongated terp. The simultaneous establishment of church and trade, as determined by him, prompted him to present the settlement as a Christian Fremdkörper in an otherwise non-Christian agrarian setting (Haarnagel 1955). He also saw some other elongated mounds in this light and assumed that these too had once housed trading settlements. Soon the combination of street village and elongated mound was seen as an independent indication of Early-medieval trade and crafts, not only in Germany (Reinhardt 1959, 1965; 1970; Brandt 1977; 1979; 1984; 1986; Haarnagel 1984; Krämer 1984; Schmid 1988, 136–7) but also in the Netherlands (Halbertsma 1956; Van Es 1971; Klok 1974–75; Schuur 1979; Elzinga 1981). However, as more regional research was done into this in the Netherlands, the theory of the trading terps quickly came under pressure (De Langen 1992 and 1999, following Miedema 1983, who still assumed the existence of trading terps but wondered why she could not find them in her study area in Groningen; see also Blok 1979, 112). It has now been established that elongated mounds either accommodated a non-agricultural neighbourhood that was no older than the church, or were Early-medieval (or even older) but if so and notwithstanding their elongated form housed mainly farms (Siegmüller and Jöns 2012). In fact, this conclusion implies that there are no indications of independent trading settlements in the Frisian regions before the tenth century. If some of the merchants and skippers who were active in the Early-medieval trade were resident in Central Frisia, they thus must have found accommodation elsewhere. So far, it is not yet known where, but it is obvious to assume that they were housed in the existing agrarian settlements, where before the arrival of the church they could enjoy the protection of local lords, who in turn took their main income from farming, as was still customary for the sixteenth-century Central Frisian nobility (Bos et al. 2021). Here the traders may have been part of the lord’s household, in the sense that they were employed throughout the year and then, if needed, went on commercial shipping in the summer, instead of the lord, because it does not seem likely to us that a farmer – even if he was a lord – went trading himself.23 When the traders 23 In regional markets they could well trade on their own, but long-distance trading was not an option: Lebecq 1992, 8.

112

Z01 Hines Book B.indb 112

07/05/2021 13:30

Landscape, Trade and Power enjoyed a more independent position and were hired on a temporary basis, they may have been guests, living in the lord’s house or outside the settlement, for example, on or near their boats. However, before the tenth century the number of locally settled traders could not have been high, at least if we consider the relatively limited total size of the approximately thirty neighbourhoods that may have lain next to the churches present at that time.24 On the other hand, the withdrawal of trade and crafts from agrarian settlements by the High-medieval neighbourhoods and later cities may have been a lengthy process, which has not yet been sufficiently researched in Frisia. So it is entirely possible that in the tenth and early eleventh centuries some of the local merchants and craftsmen were still living in agricultural settlements, or were still travelling around to visit farmers. Given the state of research, it is currently impossible to come up with a definite view on the active trade conducted by the Early-medieval inhabitants of Frisia east of the Vlie. This also has to do with the fact that many physical expressions of centrality in Early-medieval Friesland possessed a certain degree of mobility. It is in this respect striking that on or near a number of terp settlements which must have been of importance in Early-medieval Westergo, such as Midlum, Wijnaldum, Dongjum, Berlikum and Dronrijp, no early church came into being. The complex of Wijnaldum, during Merovingian times the presumed residence of a substantial lord, only gained a parish church in the twelfth or thirteenth century, attracting no more than a very modest neighbourhood, a situation which does not do justice to its Early-medieval status. This can indicate both a shift and a change. In general, one can speak of a shift when the economic and landscape conditions change to the degree that the non-agricultural element of a settlement does not grow any further, but slowly declines or even disappears. There are enough High-medieval examples of such a process – in Jorwerd, a terp with a tenth-century mother church, for instance. Little remained of its central function when its daughter settlements, situated near the Middelzee tidal inlet that separated Westergo from Oostergo, came to bloom (Noomen 1996a). The High-medieval mint site of Garrelsweer left no trace at all. A neighbourhood could also be relocated within a settlement, as was probably the case at Middelstum (Fig. 4.17), situated, like Garrelsweer, in Fivelingo. Its size could also become fixed, in the case of the development of a second neighbourhood at another spot within the settlement, as happened in Sexbierum in Westergo and Leeuwarden in Oostergo. Such adjustments are constant and it is therefore quite possible that the status of the Wijnaldum complex was already declining in the late seventh or early eighth century, in favour of the younger terpen of Westerbierum (the villages Pietersbierum and Sexbierum) and Oosterbierum north of Wijnaldum (Fig. 4.18). It is, of course, impossible to say whether the assumed ‘lord’ of Wijnaldum moved his seat during this transition or whether this northern complex came into being as the actual, permanently populated trading area of Wijnaldum. The last option does not seem very likely, given all that has been said above, which leads us to presume a change instead of a shift. We believe that Early-medieval Wijnaldum could have functioned perfectly without a trading quarter and also that the development of the

24 As measured by the estimated size of their terps and taking into account the fact that these neighbourhoods must have benefited from the tenth-century economic boom.

113

Z01 Hines Book B.indb 113

07/05/2021 13:30

Fig. 4.17  Middelstum in Hunsingo in 1832 according to www.hisgis. nl . In red line: size of the terp (on the basis of www.ahn.nl). Dotted line: southern boundary of the presumed oldest settlement. Map S. de Bruijn, Province of Friesland.

Fig. 4.18 The Wijnaldum and Dongjum terp complexes (to the south) and the terps of Westerbierum (Pietersbierum and Sexbierum) and Oosterbierum (to the north) and their surroundings in 1832 according to www.hisgis. nl . Red lines: size of the terps (on the basis of www. ahn.nl). Map S. de Bruijn, Province of Friesland.

Z01 Hines Book B.indb 114

07/05/2021 13:30

Landscape, Trade and Power neighbourhood of Sexbierum, the oldest part of which developed on a new elevated site next to the church, was completely new in the area in the late tenth or eleventh century. We do not, however, think it likely that Wijnaldum would have ascended to any significance without a leader living there. In fact, we even are inclined to assume that at the local level it was the Frisian leaders who were decisive in the execution of long-distance trade, regardless of the location or size of their residence. In this respect, we should not place too much value on the physical appearance of agricultural complexes and should also expect trading activities and important players outside these complexes. A gold disc-on-bow brooch from c. 620–640, found in Hogebeintum in Oostergo, suggests that there too an important lord or nobleman was resident (Nicolay 2014, 77–9). Possibly these men were large landowners who owned most of the entire settlement as well as some farms elsewhere.25 Although living in an agricultural settlement themselves, they could well have protected markets in the vicinity. If we take the situation in sixteenth-century Friesland as a comparison again, such noble positions were not always stable, which will have discouraged the crystallization of permanent trading quarters even more than the entire economic situation already did. All this is of course highly hypothetical, but it does illustrate why we are not surprised to find that Wijnaldum, despite the large number of coins and the objects of high status, plus indications of goldsmithing, being found there, in the end was only an agricultural settlement (Gerrets and De Koning 1999). It would thus appear that before the Carolingian Period there were no real trade settlements in Central Frisia as places where the main structure was determined by trading activities. It is likely that the trading activity itself barely had a permanent physical expression that differed from what was already needed locally for the production of certain goods and waterborne traffic. There are no indications that this does not also apply to Early-medieval East Frisia, given, among other things, the similarities in landscape and settlement structures. In West Frisia excavations have shown that in this part of Frisia, too, locally the most important settlements were all agrarian in character. Good examples are the previously mentioned agricultural settlements around the mouth of the Oude Rijn, some of which, like Wijnaldum, were of high status but still agrarian, despite the mooring places along the banks of the river (Dijkstra and Verhoe­ ven 2019). There and elsewhere within the West Frisian regions, up to now real trading settlements have remained invisible. So, what we see when we zoom in on the Frisian regions is an agricultural landscape with mixed farms basing their existence mainly on agricultural production. This puts the maritime and commercial character of Frisia into perspective when it comes to its daily business. In the next section we want to explain that this structure will also have determined the way in which power was based in the individual Frisian districts.

25 We base our opinion on what we know about the possessions of the most important nobles of sixteenth-century Central Frisia. However, there are Early-medieval indications of extensive land ownership too; see the donation to the abbey of Werden by the single owner of seven farms located on terp Arwerd near Krewerd in Fivelingo: Noomen 1996b, 60.

115

Z01 Hines Book B.indb 115

07/05/2021 13:30

Frisians of the Early Middle Ages

Magna Frisia, Radbod and the Pippinids In this final section of analysis, the focus turns to the concentration and distribution of power in Frisia. We will argue that the power of the rulers of West Frisia, although they facilitated and controlled the trade-flows when they exercised authority in Frisia Citerior, more specifically in Utrecht and Dorestad, was based, in the end, on the possession of rural estates in their homeland. Their authority certainly did not extend over all Frisia, but it may very well be that both Aldgisl and Radbod did maintain federal ties of a military nature with elites in Central and East Frisia. ‘Elites’, because there are strong indications that society in the whole of Frisia was stratified, with groups of powerful noblemen dominating at the top. The idea of Early-medieval Frisia being composed of egalitarian ‘republics’ of farmers and seafaring traders can be dismissed. However, before presenting our model of space and power in Frisia, we must first elaborate on the position of the Frisian kings in the Dutch central river area. Magna Frisia and the Frisian kings In the literature we often encounter the notion of Magna Frisia or Greater Friesland (Lebecq 1992, I, 101–5). This is not a historic name but one forged by scholars to denote a political-military intra-Frisian connection that implicitly would emerge from the written sources. Most experts allow the formation of this entity to coincide with the aforementioned seventh-century extension of coastal Frisia with the Dutch central river area (Heidinga 1986; 1997). How long, or to what extent, this Greater Frisia was under the central leadership of one or more Frisian kings has long been a topic of discussion. It is plausible that the various sub-areas had and kept at least individual characteristics, as is evident from the step-by-step incorporation of Frisia into the Frankish realm (AD 719, 734 and 785). The first question in this regard is whether it is justifiable to denote the Frisian rulers Aldgisl and Radbod as kings. And by kings we do not mean temporarily ruling Heerkönige or warlords (Schlesinger 1956; Meeder and Goosmann 2018, 31), elected in circumstances of threat or expansion, but permanent autonomous rulers advised and militarily supported by a circle of wealthy magnates, who in return were rewarded by them. It has often been noticed that in most Frankish sources Aldgisl and especially Radbod are referred to as dux (duke) or, but less often, princeps (prince), while AngloSaxon authors designate them as reges (kings) (Van Egmond 2005). The Anglo-Saxons then, would have recognized them as the type of independent rulers they were familiar with in their own country. The Franks, from their position and tradition, would, using dux, have wanted to emphasize the subordination of the Frisian rulers to their own king. The latter may be true in general, but we should bear in mind that the Frankish empire in the seventh century was characterized by a special constellation of power, in which the actual authority was not exercised by the king but by majores palatii, whose functions were then fulfilled by high aristocrats who were always striving to make their offices hereditary. The Austrasian majordomo (‘Mayor of the Palace’) Pippin could therefore, in Frankish texts, only be presented as princeps or dux. This also applied to his son Charles Martel. Since it was the intention of the Frankish chronicles and annals to praise the Pippinides and present their opponents as predestined losers, their authors were not expected to place Pippin’s or Charles Martel’s opponents on the highest level of rex. Therefore, in these sources, the Frisian rulers may thus only have been 116

Z01 Hines Book B.indb 116

07/05/2021 13:30

Landscape, Trade and Power designated as duces because Pippin and Charles could still not be presented as kings themselves. In this context, the conflict between Radbod and Pippin, and the later one between Radbod and Charles Martel, has been described as a struggle between aristocrats (Van Egmond 2005). This however, does not seem appropriate to us, for two reasons. Firstly, Pippin and Charles Martel, being majores palatii, were more than just powerful Frankish magnates with private armies. Secondly, though Radbod was most probably interested in getting access to networks of powerful Frankish noblemen, his starting position was that of an independent ruler. He exerted power in a (Frisian, not Frankish) territory of his own. And he had a group of aristocrats of his own around him, as can clearly be deduced from Vita Liudgeri, in which the history of the Frisian missionary Liudger and his family is explained (Vita Liudgeri, I). Its author portrays Liudger’s grandfather Ado or Atto, alias Wrssing (son of Wurso), as a Frisian nobilis and vassal of Radbod who possessed landed estates in the Vecht region. What is more, in this unsuspecting Frankish source, Radbod is really called king (rex). Its author Altfrid, who was related to Liudger and therefore is thought to have been well informed on the situation, presents him as a militarily and politically independent lord exerting power on a permanent base – a ruler also who was able to hunt down and confiscate the heritage of disloyal vassals: exactly the position one would assign to a king. The next question then is where the original power base of Radbod (and Aldgisl) has to be located. Did the Vecht region traditionally form part of it – implying that Wrssing and his predecessors owned estates there for a very long time – or had it been conquered by Aldgisl or one of his predecessors only a few decades before? In the latter case, which seems the most likely one to us – given that the Vecht region was off the coast, on the ‘Frankish’ side of the peat bogs – Wrssing’s father or grandfather may have been placed there by the king as a vassal and rewarded with confiscated property of a Frankish landowner. The answer to this question is not easy to give. In recent literature, it is only generally assumed that Aldgisl and Radbod had their main bases of power both in the strategically located Dutch central river area and in a chain of separate pagi along the rivers and coasts of Frisia (cf. Blok 1979, 42–5). According to Halbertsma, for instance, their entire territory would then have included both Frisia Citerior and a part of the original Frisian coastal area, with the connotation that Frisia Citerior as a geopolitical unit will first and foremost have covered the area south of the Rhine Limes (Halbertsma 2000, 76). As we argued in discussing the landscape, the Rhine can at that time hardly have functioned as an all-determining border river. Apart from the fact that, due to the growth of the River Waal, the flow of water in the seventh-century Oude Rijn was already much less than in Roman times, the Rhine by then seems to have been rather a linking than a separating stream, connecting the banks on both sides to each other. That has been demonstrated for the estuary area, where north and south were united in the same Rijnland district and the late tenth-century parish of Leiderdorp covered both banks of the river. However, it also applies to the regions to the north (the Vecht region) and south of Utrecht. The Roman castella had been ruins for centuries, with nobody there to guard the river crossings. It is therefore wise to consider the inhabited areas along the rivers, including the Vecht region, as units involved with each other, albeit units that together were separated from coastal Frisia by high and widely expanded peat bogs. 117

Z01 Hines Book B.indb 117

07/05/2021 13:30

Frisians of the Early Middle Ages Radbod’s position and his struggle with Pippin of Herstal to 695 After Aldgisl or his predecessor acquired this fertile area with its important trade centres, some of the native nobility will have come to an agreement with him to retain their positions. But this group and the circle of aristocrats around the king will also have participated or have continued to participate in existing Frankish networks. Their positions may therefore also have been sensitive as well to the unrest that was caused by the regularly occurring Frankish changes in power and the tensions between the West and East Franks. Possibly some of them were also interested in entering into alliances with Frankish aristocrats who wanted to expand their power base to the North. The most powerful man amongst these aristocrats was Pippin of Herstal, whose family owned huge complexes of land in the Ardennes and the area between Cologne, Maastricht and Liège. His victory over the Neustrians in the battle of Tertry in 687 gave him the means to develop his position as de facto ruler of the Frankish empire further. One of the first directions to which his attention will have turned was to the west, to clear and safeguard the trade routes from Austrasia and northern Neustria to the North Sea, that is via the Rhine, the Meuse and the Scheldt. Though there are no well-dated reports on it, it does not take much strategic insight to see that the one route Pippin wanted to get hold of first was the Scheldt route, with Antwerp and the emporium of Domburg/Walcheren as its main centres. The Frisian Scheldt district may have been easy to seize since it is unlikely that it was dominated by Radbod at that time. The same could go for the adjoining region covering both sides of the Meuse estuary. Creating a passageway here must have meant an enormous profit in any case, since not only the Meuse shipping but also the Rhine’s cargoes from Cologne bound for England could be steered through here, via the Waal as the increasingly navigable outflow of the Rhine. If, in the years 688–694, Pippin also developed his power in this direction, he was going to collide with Radbod. The reporting in the various Carolingian sources about the confrontation between both rulers is not unambiguous, partly due to these sources’ positive bias towards Pippin and the Pippinids. The youngest and most partisan text, the Annals of Metz, mentions two battles won by Pippin near Dorestad. Had there really been two of those combats at the same spot, Radbod would have recovered miraculously quickly from the first and Pippin would have had to gather forces again in a relatively short time to reach the same goal. Because the majordomo had to be active on other borders of his empire at the same time, this is very unlikely. It makes therefore more sense to follow the two other, older texts, namely the Continuatio Fredegarii and the Liber historiae Francorum, which record only one battle (Halbertsma 2000, 80–1; Volquartz 2018, 13–14). Their dating is not entirely certain: some historians opt for the year 695, others for 697. Like Van Egmond, we choose 695 because of the connection with Pippin’s direction of Willibrord to Rome to be consecrated archbishop of the Frisians, a task that only made sense when the intended missionary area had been brought under Frankish rule (Van Egmond 2005, 28). And we know for sure that it was not until 696 that Willibrord undertook this consecration journey to Pope Sergius. Radbod remaining in power: cooperation and alliance with Pippin, 695–714 The triumphant words in the aforementioned sources about Pippin’s victory suggest that Radbod was driven completely out of Frisia Citerior in 695. Whether this was really the case cannot be verified in the same texts because they no longer mention 118

Z01 Hines Book B.indb 118

07/05/2021 13:30

Landscape, Trade and Power the Frisian king until 711. Precisely that next report, however, suggests that Radbod had been able to maintain his dominant position to a large extent. For in 711, although still pagan, he gave his daughter Theudesinde in marriage to Pippin’s rightful heir, the majordomo Grimoald (Liber Historiae Francorum, 49; Fredegar, cont. 7): an act which in our opinion can only be interpreted as the sealing of an alliance between two autonomous rulers (cf. Wood 1994; this vol.). How then should we imagine the developments leading to this outcome, between 695 and 711? Although official Frankish historiography ignores the events in Frisia around 700, there are some hagiographical texts that shed light on them. These are reports on Willibrord’s early missionary activities, from Bede’s Historia Ecclesiastica and the Vita Willibrordi of Alcuin, as well as from the Vita Liudgeri of Altfrid. Willibrord arrived with eleven or twelve companions in the estuary of the Rhine in 690, according to Alcuin, and apparently continued his journey from there to Utrecht (Volquartz 2018, 20–1). Since Radbod, residing there, did not intend to facilitate his conversion work, the missionary travelled on with his team to Pippin who then deployed him within the boundaries of his own territory (Van Egmond 2005, 42). This is in line with our observation that at that time Pippin had not yet succeeded in driving Radbod out of the central river area. It is also consistent with the fact that Willibrord’s companion, Suidbert, who after temporarily returning to England to be ordained bishop in 692 while Willibrord went to Rome to collect relics, was sent by Pippin and his wife Plectrudis to the east of Cologne to evangelize there (Vis 1990, 130–1). Where exactly Willibrord was active between 690/692 and 696, after having returned from his second trip to Rome as archbishop, sources do not say. It has long been thought that this was the area of Antwerp, because of the mention of a donation of the church of Antwerp to Willibrord, dated 692/693, in the Liber Aureus Epternacensis (Wampach 1930 I, 78–82, nos. 34 and 35). Although the reliability of the donation notice is problematic (Rombaut 1990; Bijsterveld et al. 1999), we can well imagine from a geopolitical point of view that Pippin induced or suggested the Anglo-Saxon missionary should start in the already conquered Scheldt area in this early phase. After all, Antwerp was an excellent springboard to reach the island of Walcheren, with its important trading settlement and toll centre near Domburg, downstream. And it is certain that Willibrord was active in Walcheren, although no date can be connected with that (Weiler 1989, 167; Hende­ rikx 2014a, 78). The victory at Dorestad made it possible for Pippin to install Willibrord in Utrecht in 697 and let him establish a bishop’s see in the castellum. At least that is what Bede reports on the basis of hearsay by Bishop Acca of Hexham, who visited Willibrord in Utrecht in 703 (HE, III, 13). Willibrord is said to have had much success in preaching and the founding of churches and monasteries. Precise information about this first missionary phase within Frisia Citerior, however, appears nowhere. Could it be that this information is lacking because there was hardly any progress to mention?26 More than forty years ago, in his book on the Franks in the Netherlands, Dirk Blok signalled that Radbod must have succeeded in staying in power in the Vecht region (Niftarlake) north of Utrecht (Blok 1979, 45). His main argument in this is that the repudiation of Liudger’s grandfather Atto or Wrssing by the Frisian king – described in Vita Liudgeri – can only 26 Weiler 1989, 125, found it strange ‘that so little, in fact nothing, has been left over from Willibrord’s missionary ventures in the Frisian-Frankish river area’.

119

Z01 Hines Book B.indb 119

07/05/2021 13:30

Frisians of the Early Middle Ages have taken place in the period after the battle of Dorestad, between 698 and 711. The year 698 is the terminus post quem because the Christian Wrssing is said to have fled to Pippin’s son, the majordomo Grimoald, who was only allowed to call himself major palatii (of Neustria) from that year. This gave Blok, who was thinking from the Limes perspective, reason to argue that Frankish power at that time did not extend beyond the former Roman bases along the rivers. Whether the latter is a necessary deduction can be doubted. What remains, however, is that after 698 Radbod at least was able to exercise compelling ‘pagan’ power under the walls of the castellum of Utrecht. At the same time it is striking that from 698 Willibrord developed a great deal of activity in the Ardennes, the Eifel, the Meuse area and Taxandria. So few traces of his activity in Frisia are left, so many more are transmitted regarding his missions and interests within these Frankish regions (Weiler 1989, 112–25; Honée 1996, 113–23). The first steps were taken with the initial donation of a small monastery in Echternach by Irmina van Oeren, the mother of Plectrudis in 697/698, which was supplemented by Pippin and his wife in 706 with a large property complex. The abbey of Echternach was intended to constitute a sort of base camp to which the missionaries could regularly withdraw and where new monks could be trained for the conversion work. And then there was the transfer of another religious house, at Susteren on the Meuse in 714, also made possible by Pippin and Plectrudis. Remarkably, many other goods and churches were transferred to Willibord in the present Dutch province of Noord-Brabant in these years (Werner 1980). Among these, the donations at Ruimel (699), Waalre (703/704), Alphen (near Breda, 709), Hoksent (710), Luisel (712) and Diesen (713) can be dated fairly precisely (Bijsterveld et al. 1999, 205–6). About Willibrord’s associated apostolate in Taxandria we need not elaborate further here. The point is to underline that after 700/703 Pippin deployed Willibrord in a completely different mission area than was his intention before, and that this can only have been inspired by a new political reality. That new reality must have been that Radbod had been able to maintain a significant part of his power in parts of Frisia Citerior and that it was not a priority for Pippin to try to change this militarily. Radbod must have been able to continue to control the traffic to and from Dorestad and Utrecht along the Vecht and the Vlie and the Oude Rijn. The only solution Pippin had for this problem was to acknowledge Radbod’s strength and come to an agreement with him – an agreement that included freezing missionary activity from Utrecht for the time being, and which eventually found its seal in the marriage between Theudesinde and Grimoald. But it was also an agreement that had to end by itself after the almost simultaneous murder of Grimoald and the death of Pippin by disease in the autumn of 714. Charles Martel and the slow pacification of West Frisia up to 734 After the death of Pippin II in 714, a battle for the throne followed, known as the Frankish civil war, won in the end by Charles Martel, half-brother of Grimoald. In the autumn of 715 Radbod, who apparently did not have to remain bound to Pippin’s widow Plectrudis, joined the West Frankish camp of the new Neustrian majordomo Raganfred. As part of this alliance, in 716 he sailed with a fleet along the Rhine to Cologne to defeat Charles Martel, who had taken over the rule of Plectrudis. But the coalition as a whole did not hold up, and in that same year the cards appeared to have fallen in favour of Charles Martel. Once again, however, Radbod did not disappear from the scene (Fritze 1971, 115). It can hardly be otherwise, as he soon reached an agreement with the new 120

Z01 Hines Book B.indb 120

07/05/2021 13:30

Landscape, Trade and Power Frankish man in power. Only after the death of Radbod in 719 did Charles see sufficient reason to regain Frisia Citerior, bringing in the missionaries anew, promoting church building in Utrecht and finally also penetrating into coastal Frisia. Many authors assume that Charles Martel conquered West Frisia all the way to the Vlie in one go in 719 and was able to seize all the heritage of Radbod and other Frisian opponents. There is, however, no reason to overestimate the speed of this takeover. An indication of its slow progress is the news about the deployment of the still young Boniface as assistant of Willibrord between 719 and 722. From the Vita Gregorii, written by Liudger, we know that Boniface was active in Woerden (at the Oude Rijn), Attingahem (= Breukelen or Nederhorst) at the Vecht, and finally in Velzen (Blok 1979, 51). This means that he first had to work on two central places in Frisia Citerior before he could cross over to one of the former power bases of Radbod in South Kennemerland. In Vita Gregorii it is stated of Velzen that it was ‘close[r] to the pagans’. We are inclined to assume rather slow Christianization and an equally slow build-up of Frankish government. Radbod’s death would probably have cleared the western Rhine route for Charles Martel, but certainly not the northern one via the Vlie. As long as the Frisian elite of Westergo was not yet bound to the empire, they could control the sea-going traffic along its coast and, with or without short-term raids, keep a grip on the region around Medemblik as well as Wieringen and Texel. It was only in 733 and 734 that Charles Martel found the opportunity and means to solve this problem with two or three large amphibious expeditions.27 We are talking about the voyages that in 734 resulted in the Battle of the Boorne, where, according to Continuatio Fredegarii (ch. 17), Charles slaughtered the pagan dux Bubo with his now rebellious Frisian supporters, after which he had their pagan temples burned and could return home loaded with loot. It is not probable that Charles immediately conquered Friesland between Vlie and Lauwers and made it accessible for the mission. The story of Boniface’s expeditions accompanied by men-at-arms in 753 and 754 clearly proves that twenty years later great resistance had still to be overcome. The first goal of Charles’s expeditions in 733 and 734 must therefore have been definitively to control the trade route to the north and secure the incorporation of West Frisia into the Empire.

Conclusion Frisia and the Frisian trade In the literature, Frisian trade and Early-medieval Frisia are always closely linked. In order to get a grip on their mutual relationship we have first separated the two concepts and then focused on the Early-medieval trade-flows passing Frisia in order to test the resulting insights against two other concepts: landscape and power. This leads to the conclusion that Frisia (proper) consisted of the series of narrow interlinked residential areas along the North Sea coast between the Sincfal and the Weser, with the observation that these coastal areas were definitively separated from their hinterland by vast peat bogs. Frisia Citerior, the temporary extension of Frisia under Kings Aldgisl and 27 Volquartz 2018, 20, points out that, according to short reports in the Annales Petaviani, Annales Sancti Amandi and Annales Tiliani, Charles Martel went to Westergo with an army in 733 and that, according to the Annales Sancti Amandi, he did so for a second time (iterum) in 734.

121

Z01 Hines Book B.indb 121

07/05/2021 13:30

Frisians of the Early Middle Ages Radbod, was found to have included only the Dutch central river area including the Vecht region. Frisia itself, as a coastal ribbon, was intersected by various watercourses in such a way that each sub-area had its own economic and geopolitical weight. Another landscape observation is that the group of areas west of the Vlie with its continuous sand ridges was fundamentally different from the northern group of lands between Vlie and Weser. The latter consisted of salt-marsh areas that were not only separated from each other by running water but also divided internally by larger and smaller channels. Another conclusion is that the various trading places associated in the literature with the long-distance Frisian trade are not found within Frisia itself; they were all located in border zones. The Frisian regions, east and west of the Vlie, turn out to have had a purely agricultural economy. Frisian trade may have been partly instigated by the Frisian coastal inhabitants, but it seems to have been transformed into an autonomous superstructure soon after 650, taking over old trading functions from the regions. It was not ethnic Frisians alone who were active in this long-distance trade. The name ‘Frisian’ for the long-distance trade will indicate that it passed through the Frisian lands, but it was not principally organized by the inhabitants of the Frisian districts. The development of long-distance trade does not seem to be directly linked to changes in power, which suggests that to a large extent it had its own autonomous dynamics. We cannot subscribe to the view defended in the literature that the Frisian trade was based on the freedom of economic movement enjoyed by the coastal inhabitants because they were predominantly livestock farmers with enough spare time and capital to go on trading voyages by ship in the summer season. Most Frisians did not even specialize in cattle breeding and simply worked all year round on their farmsteads in their own regions. If we relate the Frisia so defined and the ‘Frisian’ trade to what is known about the distribution of power in Frisia, we conclude that power in the context of the various regions was strongly determined by the landscape and subsequent economic circumstances. In light of their urge for expansion, which gained ground after 687 when Pippin of Herstal seized power in Neustria, we assume that the Franks quickly gained access to the Scheldt and Meuse estuaries. The area controlled by Radbod, with his strategic holdings in the central river area and West Frisia, was less easy to take over. Although sources report that Radbod was crushed by Pippin near Dorestad in 695, the subsequent events prove that Pippin must soon have come to an agreement with the Frisian king. This agreement allowed Radbod to retain authority over Frisia Citerior, albeit in the name and as a vassal of Pippin, while his home rule over (coastal) Frisia was not challenged. Given that Radbod remained a man of high status, he and his predecessor Aldgisl must have owed their kingship first and foremost to their position as the largest landowners in their Frisian homeland. Of course, their authority over Frisia Citerior undoubtedly enabled them to cream off long-distance trade more efficiently by using the Rhine and Vlie routes to the west and the north, levying tolls and so increasing their power even further. The rural power base of the Frisian kings In view of the present state of archaeological research and the fragmentary nature of the historical sources, we have repeatedly used a retrospective method whereby later historical information is projected back in time for testing purposes. This mainly involves the spatial analysis of ecclesiastical and economic data, more specifically the mapping 122

Z01 Hines Book B.indb 122

07/05/2021 13:30

Landscape, Trade and Power of borders and property complexes. With the help of this, we believe we have developed a number of new hypotheses that may be useful for further research. In order to further stimulate this new research, we take the liberty of expressing a few speculations here. These concern first of all the rural power base of Aldgisl and Radbod. If we arrange all the data logically, we arrive at the following for the reconstruction of the Frisian kingdom in West Frisia. The southern Frisian districts of Scheldeland and Maasland did not form an indivisible block with the rest of West Frisia, and certainly not with Frisia Citerior. They probably came under Frankish authority earlier, between 687 and 695, and will not have belonged to the core area of Radbod. Rhineland, on the other hand, and the other West Frisian regions up to the Vlie, did. This can be deduced, with some reservations, from the demarcation of the ‘margraviate’ that some Danish magnates received in fief from the Frankish king in the ninth century to prevent invasions by other Vikings. In view of the later developments, it can be assumed that the territorial base of Radbod, and its roots, must be sought in Kennemerland, and the districts of Texel and Wieringen (the area of Medemblik included). It included the Dutch central river area (Frisia Citerior) and the district of Rijnland as well, given the fact that Gerulf, an ancestor of the Gerulfingen, who in the Viking era, together with his brother Gardulf, operated as a count under a Danish feudal lord, later appears to have had the nucleus of his later group of counties in this very district. As stated, after the loss of Utrecht and Dorestad Radbod remained an independent ruler in the coastal region because it will have been clear to the majordomo Pippin that complete control of the long-distance trade to the north and west via Dorestad and Utrecht was not easy to achieve. For that would entail defeating Radbod beyond the peat bogs as well. And not only Radbod. After all, Radbod (just like Aldgisl before him) will have had an agreement with the leaders of Westergo and possibly also those of other Frisian districts east of the Vlie, since they too participated in tapping of the trade to and from the north. For his part, Radbod was able to control this trade-flow partly through the Vecht region and partly from his own territory in northern West Frisia. Pippin, whose military presence was also needed elsewhere on the borders of the Frankish empire, will for the time being have been satisfied with his grip on the trading traffic via the Meuse and the Waal. This meant that he had to grant Radbod a continuation of his influence in Utrecht and the surrounding area, with the consequence that the missionary activity from there had to be temporarily suspended. By 700, Radbod’s property in West Frisia will have consisted of several large estates that together strengthened the landscape backbone of West Frisia with a political one. He must have inherited these estates, directly or indirectly, from Aldgisl, for he, as the first known Frisian rex, will have based his kingship on his position as an important landowner. That West Frisia around 700 boasted some large landholdings can be deduced from the subsequent presence of royal fiscs from which the Frankish kings transferred substantial parts to Willibrord’s abbey of Echternach and the church of Utrecht in the course of the eighth century. The later counts of Holland also managed to acquire part of this royal property and built their domain up with it, with manors as nuclei of exploitation. These properties enabled them to act as church lords and to make important donations to monasteries and parish churches.

123

Z01 Hines Book B.indb 123

07/05/2021 13:30

Frisians of the Early Middle Ages No domain-formation east of the Vlie After the (natural) death of Radbod, West Frisia came into Frankish hands. Central Frisia, however, remained independent for several decades, which can be interpreted as an indication that it never belonged to Radbod’s core area. We therefore rank the Westergo and Oostergo regions, just like Scheldeland and Maasland, among the (presumed) allies of the West Frisian king. In the decades following the campaigns of 733 and 734, the Franks gradually gained control of Central Frisia, so that here too the mission could begin and the foundations could be laid for the establishment of the Church, a process that continued until the Norman invasions made it stagnate. Nevertheless, this area would administratively follow a completely different route than West Frisia, which incidentally indirectly suggests that at the time of the Christianization it already had gone through a different development all together. This development can be traced back to the specific way in which land ownership in Frisia east of the Vlie was divided up and how it was exploited. We may assume that, following the events of 733 and 734, there was confiscation of landed property by the king, just like in West Frisia after 719. But we can hardly find traces of large complexes here, with some exceptions, such as at Franeker. Land ownership did exist, though as far as we can see the largest nuclei of such property did not exceed the extent of the small landscape unit of a terp settlement. Above, we saw that the design of these terp settlements was aimed at forming separate farms that were structurally equivalent to each other. This principle also applies to the farms that shared a terp. We can now add that before the construction of continuous winter dikes began, the possibilities for internal colonization were limited and that most of the clay area was shared among the terp dwellers, probably at a very early stage. In addition, the land was already fertile by nature and the existing farms will have had high efficiency. Together these factors must have discouraged changes in management such as scaling up. We therefore suspect, and even dare to say, that in Central Frisia massive large land holdings were already absent before 850, so that there were few starting points for the Carolingian rulers (and too little time) to introduce the manorial system here and expand it. In fact, we can say that the possibilities for domain-formation in the clay area were limited from the start. This meant that in the Frisian coastal strip to the east of the Vlie large land ownership consisted of a collection of independent farm-units. However, this did not mean that there was no large-scale land ownership, any more than that there were no nobles. After all, Lex Frisionum mentions the nobiles, most of whom will have been responsible for the administration of justice within and the defence of the regions. In the Early Middle Ages, more than in the later period when intra-regionalization took place, it was also possible to derive prestige and authority from marital relations with Frisian families in other districts and even with noble families from outside Frisia. From this it follows that with these economic and political internal differences even at local level Early-medieval Frisia did not consist of a federation of egalitarian peasant republics. This model is supported by the High- and Late-medieval data. In the Late Middle Ages, from about the middle of the thirteenth century, the inhabitants of the Frisian lands east of the Vlie enjoyed freedom from direct sovereign authority. They met their obligations to church and king, but kept any count out of their territory and chose their judges themselves. Given the above, clearly the exercise of communal autonomy at a regional and local level must have had older roots. Although during the eleventh 124

Z01 Hines Book B.indb 124

07/05/2021 13:30

Landscape, Trade and Power and twelfth centuries Central and East Frisia were still ruled by counts, some of whom are known to have minted coins in Frisia, none of these was successful in expanding his power. This must be due to the fact that they themselves – at least the Brunswick counts – never personally resided in the region and pursued higher interests elsewhere, but to all probability also because the domanial infrastructure was lacking. Not even side-branches of the comital families were challenged to establish themselves in Frisia and to start a new dynasty. Apparently no large former royal domain was available, parts of which could be used to enfief vassals or in the patronage of monasteries, which also meant that the Frisian lands to the east of the Vlie, unlike those to the west, never became feudalized. From this it can be deduced that even in the late Carolingian Period here, the position of the counts was weak. In the tenth and eleventh centuries they hardly participated in the foundation of parish churches. The places of worship that existed before 1050 were either the old mission churches of Utrecht, Münster and a number of abbeys in the Carolingian heartland, once founded by missionaries, or new churches established on the initiative or with support of the bishop, who therefore had to rely on the material input of lesser or local noblemen. The heathen king and his allies The observation that the perspectives for building up large domains and thus for increasing sustainable power on both sides of the Vlie were not the same brings us back to the subject we started with, namely the heterogeneous nature of Frisia. That much is clear: this area along the southern North Sea coast was already called Frisian by the Franks in the late seventh century. Evidently there was something that, in the eyes of neighbouring peoples, tied the different population groups of the coastal strip between the Sincfal and the Weser to each other. Perhaps the Frisians themselves also experienced a mutual connection at an early stage. But our overview shows that there were some major differences between the constituent parts as well. These differences were possibly very old and stemmed from geopolitical characteristics or the internal structure and partitioning of the landscape. This internal differentiation would fit in well with the role we want to attribute to Aldgisl and Radbod. The function they fulfilled in the Frisian world may well have been that of leader who was a king in his own territory but could not hesitate to call in support of the other Frisian pagi as his allies in case his position was threatened and a domino effect was to be feared. At certain moments these other pagi will have been represented by the most influential nobiles of the time. As far as this is concerned, it is particularly illustrative that, according to Vita Liudgeri, there were not only principes but also consiliarii, members of a council, among whom we may seek among others, a number of representatives from the elites of Central and possibly also East Frisia. King Radbod was in close contact with the Frankish world because of economic and political interests. Even though he profited from long-distance trade, his position as an independent ruler was ultimately based on his possessions in the Frisian coastal region. That origin of Frisian kingship placed him in the world of the other areas around the North Sea, of which an important part was non-Christian, starting with the Frisians directly east of Vlie, with whom he shared economic interests and on whom he had to rely militarily. The historical sources bear witness to the fact that it was precisely the Frisians living east of the Vlie who had an especial aversion to the coming of the Christian faith. Radbod’s mythical refusal to be baptised thus becomes understandable: 125

Z01 Hines Book B.indb 125

07/05/2021 13:30

Frisians of the Early Middle Ages he simply could not afford to alienate his support by becoming a Christian. That he would not convert must have been part of the deal when Radbod and Pippin came to an agreement shortly after 695. Radbod will not have renounced baptism at the last moment because of the meeting with his ancestors as the legend would have it, but will always have remained well clear of any baptismal font.

Discussion HINES  This is a very good overview of major aspects of development. It is surely an interesting point in itself, even if it can be counted as over-idealized, that especially in the Early Middle Ages the external view of the Frisians is precisely that of the ‘Frisian trader’; if any sites are internationally recognized as Frisian they are Dorestad, or even Ribe which is always described as Frisian-style; and also Emden. I’m happy to have an account of a system which explains those sites (and Medemblik) more in external terms than in terms of factors internal to Frisia. This is paralleled in the ‘Viking world’: that expansion requires some sort of urban site that is not typical of the homeland. In a way, urban sites are models which are taken ‘off the peg’ and created in contexts of expansion. I find the information on agrarian productive areas and the church eye-opening as well. KNOL  I too would like to begin with a compliment; but also pick up on the point that your paper does not cross the German border to include Ostfriesland or even further east. Trading terps are a key point introduced to archaeology from the Wilhelmshaven school, even if based on limited evidence. Gilles and Hans may reject the idea of trading terps. Alternatively, could such places only have been found in East Frisia? MAJCHCZACK  Actually the idea of trading terps on Langwurten is now being dismantled. Settlements that were involved in some trading activities are known, but it was only on the cusp of the High Middle Ages that trading centres with typical Langwurten features appeared. KNOL  Would it be possible to expand the palaeogeographical map to cross over the Dutch-German border? DE LANGEN  To include Holland was important, and indeed Domburg down to Flanders. Of course, Frisia should be taken as a whole but we were not able yet to expand our view to the area east of the Ems. VERSLOOT  Because the data are not there? DE LANGEN/MOL  It is more a matter of time and energy. Base maps have to be made for every sub-region, both for the palaeography and the development of the Church. KNOL  We have a map that has solved the problems of differences between German and Dutch maps. MOL  Georeferencing is not a problem; rather, the information on the use and the value of land per parcel is lacking for Ostfriesland. So far we have only been able only to go into detail for Central Friesland and Groningen. One of our main questions is, working from a central Frisian perspective, whether the terp region, including Ostfriesland, has some parallel in Holland. We tend to conclude that there were fundamental differences between the landscapes east and west of the 126

Z01 Hines Book B.indb 126

07/05/2021 13:30

Landscape, Trade and Power Vlie, and that the power structures, based upon the landscape, were different as well. KNOL  It would be best to have the whole terp region in one view. DE LANGEN  This is our first draft. We promise to deliver Frisia as a whole. KNOL  …and this is a discussion to support a better draft! DE LANGEN  I could add that when we reach the north of the province of Friesland the situation changes because the Pleistocene landscape becomes dominant and we get a different power structure. IJSSENNAGGER-VAN DER PLUIJM  A point made by Frans Theuws (2004) is that national and regional scales differ significantly, especially when Dorestad declines. He argues also that the trading sites were there because they were neutral, on boundaries between different spheres, including the religious one. Dorestad is not a church centre; maybe this is reflected in the terp region. DE LANGEN  Dorestad obtained a new function as soon as it was no longer located on the border. This has already been underlined by Van Es (1990). Yet the emporium continued to flourish in the eighth and ninth centuries. I do not see this as a problem. MOL  Dorestad boasted an old church, the so-called upchirica, which was the first daughter church of Utrecht. IJSSENNAGGER-VAN DER PLUIJM  …but Utrecht was the centre. NICOLAY  Would you explain once more why no large compact manorial estates are found in the terp area? I don’t see the relationship with the landscape. DE LANGEN  We don’t see any traces of them historically, so their existence is improbable. We find them in relatively large numbers in Holland, and only a few in a much more densely inhabited Frisia, where they are handed over to the Church at an early date. We do see large landowners in Frisia, but these do not hold compact estates with dependent clusters of mansi, held by serfs: the landowners in Central and East Frisia hold farms, and these farms are scattered. The landscape in some areas could be changed easily, but not in Friesland province because of the watery topography. Terp settlements seem not to have been taken over by the incoming Carolingian leadership either, unlike a series of villae on the geest ridges in Holland. MOL  Manorial complexes are not totally absent from Central Frisia, for instance, in Baflo and Leer. In Groningen, on the border with Drenthe, there were large royal curtes that were handed over to the Church. The contrast with Holland remains strong. DE LANGEN  …especially if you look at the many large estates owned by the count of Holland in the tenth and eleventh centuries. NICOLAY  What does this say about the way these areas were ruled just before Frankish annexation? DE LANGEN  We hypothesize that Radbod and his like were already building up or managing larger estates, especially in Holland and around the Rhine. NICOLAY  How was the terp region ruled then? DE LANGEN and MOL  That’s the big question! When we look at the landscape, we see how Frisian nobiles in the Late Middle Ages were struggling to keep their property viable, because it had continually to be subdivided amongst their successors. 127

Z01 Hines Book B.indb 127

07/05/2021 13:30

Frisians of the Early Middle Ages VERSLOOT  I see the clear contrast in landscape between Holland and the Fryslân/ Groningen provinces, but looking at the kwelder (marshland) ridges: they extend over considerable distances and are very fertile; possibly enough to provide a solid basis for political power based on landed property. I do not understand exactly why the landscape excludes this development. Perhaps it was rather the rule of inheritance that made the difference. Why was that it different in the terp region and in Holland? KNOL  Was it different? DE LANGEN  We should keep in mind that with the emergence of Radbod and his ancestors, something was developing. It is also feasible that the Vlie was a dividing line, and things were different in Holland. At the moment we do not know and cannot compare the inheritance rules. VERSLOOT  What was the situation in Holland later in the Middle Ages? DE LANGEN  By that time the count had many domains at his disposal. FLIERMAN  How do the Danes fit into this picture? Did they pass through without affecting things? The victory of the counts of Holland over the Danes is supposed to have enhanced their prestige. MOL  The Danes took over a Carolingian position of authority which already existed, and was then passed on from them to Gerulf, who had been count, most probably in the Rhineland district, under Rorik and Godfrey, who were licensed by successive Carolingian kings to rule over a group of counties in a sort of Frisian March. After the murder of Godfrey, Gerulf and his sons stepped into their place and gradually took control over the whole coastal area of Holland. The Danes fit in between the state of affairs under Charles Martel (for instance) and that under the Gerulfingians. But did their Carolingian power base go back to the year 719 and Radbod? We have ideas in that direction, but they are still rather speculative. IJSSENNAGGER-VAN DER PLUIJM  A final point about the Danes: there are different factors in play. We know where the Danish fiefdoms were. It is important to emphasize that they were granted by the Frankish overlord. Where we find archaeological evidence for them it appears to be in different locations from those fiefdoms, suggesting different spheres of interaction. FLIERMAN  They were being granted significant areas. IJSSENNAGGER-VAN DER PLUIJM  … but to operate under Frankish overlords.

References (including Discussion) Primary sources Altfrid, Vita Liudgeri. Ed. W. Diekamp 1881, Die Vitae Sancti Liudgeri, Geschichtsquellen des Bistums Münster 4 (Münster). Annales Mettenses Priores. Ed. B. von Simson 1905, MGH Scriptores, SS rer. Germ. 10 (Hannover). Bede, Historia Ecclesiastica Gentis Anglorum. Eds. and trans. B. Colgrave and R. A. B. Mynors 1969, Bede’s Ecclesiastical History of the English People (Oxford). Fredegar Continuator. Ed. J. M. Wallace-Hadrill 1960, The Fourth Book of the Chronicle of Fredegar (London).

128

Z01 Hines Book B.indb 128

07/05/2021 13:30

Landscape, Trade and Power Fredegarii et aliorum Chronica. Ed. Bruno Krusch 1888, MGH Scriptores, SS rer. Merov. 2. (Hannover). Lex Frisionum. Ed. K. A. Eckhardt and A. Eckhardt 1982, MGH Leges, Fontes iuris 12 (Hannover). Liber Historiae Francorum. Ed. B. S. Bacharach 1973 (Lawrence). Vita Sancti Adalberti confessoris. Ed. G. N. M. Vis 1987. In Egmond en Berne. Twee verhalende historische bronnen uit de middeleeuwen, ed. P. Leupen, Nederlandse Historische Bronnen 7 (s-Gravenhage–Leiden), 1–86. Willibald, Vita Bonifatii. Ed. W. Levison 1905, Vitae sancti Bonifatii, MGH Scriptores, SS rer. Germ. 57 (Hannover). Secondary sources Aalbersberg, G. 2019, Geogenese van het gebied rond Abbega (Friesland), AGEA-rapport 4 (Pieterburen). Albrecht, G. 1959, Das Münzwesen im niederlothringischen und friesischen Raum vom 10. bis zum beginnenden 12. Jahrhundert (Hamburg). Bazelmans, J., D. Gerrets and A. Pol 2002, ‘Metal detection and the Frisian kingdom. Questions about the central place of northern Westergo in the Early Middle Ages’, Berichten van de Rijksdienst voor het Oudheidkundig Bodemonderzoek 45, 219–41. Besteman, J. C. 1989, ‘The pre-urban development of Medemblik: from an early medieval trading centre to a medieval town’. In Medemblik and Monnickendam. Aspects of Medieval Urbanisation in Northern Holland, eds. H. A. Heidinga and H. H. van Regteren Altena, Cingula 13 (Amsterdam), 1–30. Bijsterveld, A. J., P. Noomen and B. Thissen 1999, ‘Der niederländische Fernbesitz der Abtei Echternach im Früh- und Hochmittelalter’. In Die Abtei Echternach 698–1998, ed. M. C. Ferrari (Luxembourg), 203–28. Bitter, P. 2014, ‘Alkmaar en de graaf van Holland. Enkele archeologische en archivalische gegevens nader beschouwd’, Westerheem 2014 (3), 32–57. — 2016, ‘Alkmaar: Langestraat 58, 60 en 62’, Archeologische kroniek van Noord-Holland 2015, 24–34. Blok, D. P. 1979, De Franken in Nederland, 3rd edition (Bussum). — 1986, ‘Het mondingsgebied van de Oude Rijn in de vroege middeleeuwen’. In Teksten van lezingen gehouden tijdens het Symposium ‘Landschap en bewoning rond de mondingen van Rijn, Maas en Schelde’, eds. M. C. van Trierum and H. E. Henkes, Rotterdam Papers. Contributions to Medieval Archaeology 5 (Rotterdam), 169–72. Boeles, P. C. J. A. 1927, Friesland tot de elfde eeuw. Zijn oudste beschaving en geschiedenis (Den Haag). — 1951, Friesland tot de elfde eeuw. Zijn oudste beschaving en geschiedenis, 2nd revised and extended edition (Den Haag). Bos, J., G. J. de Langen, and J. A. Mol in prep., ‘Frisian noblemen, their capital and their farms in the sixteenth century: the case of Rienk Hemmema’. Brandt, K. 1977, ‘Handelsplätze des frühen und hohen Mittelalters in der Marsch zwischen Ems- und Wesermündung, Ein Vorbericht über archäologisch-historische Untersuchungen’, Zeitschrift für Archäologie des Mittelalters 5, 121–44. — 1979, ‘Siedlung und Wirtschaft in der Emsmarsch während des frühen und hohen Mittelalters’, Probleme der Küstenforschung im südlichen Nordseegebiet 13, 155–85. — 1984, ‘Langwurten, ihre Topographie und ihre Funktion’. In Archäologische und

129

Z01 Hines Book B.indb 129

07/05/2021 13:30

Frisians of the Early Middle Ages naturwissenschaftliche Untersuchungen an ländlichen und frühstädtischen Siedlungen im deutschen Küstengebiet vom 5. Jahrhundert v. Chr. bis zum 11. Jahrhundert n. Chr. 2: Handelsplätze des frühen und hohen Mittelalters, eds. H. Jankuhn, K. Schietzel, H. Reichstein (Weinheim), 100–13. — 1986, ‘Archäologische Untersuchungen in einem mittelalterlichen Marktort an der Nordseeküste, Ergebnisse der Ausgrabungen in Langwarden, Ldkrs. Wesermarsch’, Probleme der Küstenforschung im südlichen Nordseegebiet 16, 127–69. Breternitz, P. 2018, ‘Das Ende der eigenständigen friesischen Münzprägung im 8. Jahrhundert. Beobachtungen zur Chronologie der Porcupinevarianten B, E und F’, Jaarboek voor Munt- en Penningkunde 105, 83–102. Breternitz, P., D. Stangier and W. Tillmann 2017, ‘Friesische Halbsceattas Werkstoffkundliche und historische Überlegungen’, Jaarboek voor Munt- en Penningkunde 104, 1–22. Brandenburgh, C. and E. Orsel 2014, ‘Leiden en de graaf ’, Westerheem 2014 (3), 91–104. Bult, E. J. 2014, ‘De Hof van Delft’, Westerheem 2014 (3), 126–59. Çakirlar, C., Y. van den Hurk, I. van der Jagt, Y. van Amerongen, J. Bakker, R. Breider et al. 2019, ‘Animals and people in the Netherlands’ past: >50 years of archaeozoology in the Netherlands’, Open Quaternary 5 (3), 1–30. De Langen, G. J. 1992, Middeleeuws Friesland. De economische ontwikkeling van het gewest Oostergo in de vroege en volle middeleeuwen (Groningen). — 1999, ‘De archeologie van vroeg- en vol-middeleeuws Leeuwarden (750–1250)’. In Leeuwarden 750–2000. Hoofdstad van Friesland, eds. Y. B. Kuiper and G. T. Jensma (Franeker), 18–45. — 2007, ‘Nog steeds op zoek naar het oudste Dokkum’, De Vrije Fries 87, 200–2. De Langen, G. J. and J. A. Mol 2016, ‘Terpenbouw en dorpsvorming in het Friese kustgebied tussen Vlie en Eems in de volle middeleeuwen’. In Van Wierhuizen tot Achlum. Honderd Jaar Archeologisch Onderzoek in Terpen en Wierden, ed. A. Nieuwhof, Jaarverslagen van de Vereniging voor Terpenonderzoek 98 (Groningen), 99–128. — 2017, ‘Church foundation and parish formation in Frisia in the tenth and eleventh centuries: a planned development?’, The Medieval Low Countries 4, 1–55. — 2018, ‘Kerk, macht en ruimte in Holland tot het midden van de elfde eeuw. De uitbouw van het parochiewezen tussen Maas en Vlie bezien vanuit Fries perspectief ’, Holland. Historisch Tijdschrift 50, 264–73; 309–10. — 2019, ‘Vikings and the Church in the Frisian lands’. In We Vikings. Frisians and Vikings in the Coastal Area of the Low Countries, eds. M. Stoter and D. Spiekhout (Zwolle), 22–41. De Monté Verloren, J. Ph. 1942, ‘Hoven in Holland’. In Opstellen aangeboden aan prof. jhr. Dr. D.G. Rengers Hora Siccama (Utrecht), 107–57. Dijkstra, M. 2011. Rondom de mondingen van Rijn en Maas: landschap en bewoning tussen de 3e en 9e eeuw in Zuid-Holland, in het bijzonder de Oude Rijnstreek (Leiden). Dijkstra, M. F. P. and A. A. A. Verhoeven 2019, ‘Karolingisch Leiderdorp. Een nederzetting aan een geul’, Archeologie in Nederland 3, 2019 (1), 16–25. Dijkstra, M. F. P., A. A. A. Verhoeven and K. C. J. van Straten 2016, Nieuw licht op Leithon. Archeologisch Onderzoek Naar De Vroegmiddeleeuwse Bewoning in Het Plangebied Leiderdorp-Plantage, Themata 8 (Amsterdam). Dirks, J. 1846, Geschiedkundig Onderzoek van Den Koophandel der Friezen van De Vroegste Tijden tot aan Den Dood van Karel den Grooten (Utrecht). Elzinga, G. 1981, ‘De groeiende heuvels van Friesland, De bewoningsgeschiedenis van de terpen’. In Spectrum atlas van historische plaatsen in de Lage Landen, eds. A. F. Manning

130

Z01 Hines Book B.indb 130

07/05/2021 13:30

Landscape, Trade and Power and M. de Vroede (Utrecht), 18–21. Fritze, W. 1971, ‘Zur Entstehungsgeschichte des Bistums Utrecht. Franken und Friesen 690–734’, Rheinische Vierteljahrsblätter 35, 107–51. Gerrets, D. A. and J. de Koning 1999, ‘Settlement development on the Wijnaldum-Tjitsma terp’. In The Excavations at Wijnaldum. Reports on Frisia in Roman and Medieval Times. Volume I, eds. J. C. Besteman, J. M. Bos, D. A. Gerrets, H. A. Heidinga and J. de Koning (Rotterdam–Brookfield), 73–124. Gosses, I. H. 1926, Welgeborenen en Huislieden. Onderzoekingen over Standen en Staat in Het Graafschap Holland (Groningen–Den Haag). Gysseling, M. 1976, ‘De Germaanse woorden in de Lex Salica’, Verslagen en Mededelingen van de Koninklijke Academie voor Nederlandse Taal- en Letterkunde, Nieuwe Reeks 1, 60–109. Haarnagel, W. 1955, ‘Die frühgeschichtliche Handelssiedlung Emden und ihre Entwicklung bis ins Mittelalter’, Jahrbuch der Gesellschaft für Bildende Kunst und Vaterländische Altertümer zu Emden 35, 9–78. — 1979, Die Grabung Feddersen Wierde. Methode, Hausbau, Siedlungs- und Wirtschaftsformen sowie Sozialstruktur, Feddersen Wierde 2 (Wiesbaden). — 1984, ‘Die frühgeschichtliche Handelssiedlung Emden und ihre Entwicklung im Mittelalter’. In Archäologische und naturwissenschaftliche Untersuchungen an ländlichen und frühstädtischen Siedlungen im deutschen Küstengebiet vom 5. Jahrhundert v. Chr. bis zum 11. Jahrhundert n. Chr. 2: Handelsplätze des frühen und hohen Mittelalters, eds. H. Jankuhn, K. Schietzel, H. Reichstein (Weinheim), 114–35. Halbertsma, H. 1956, ‘Oudheidkundig bodemonderzoek in de Sint-Maartenskerk te Bolsward, Een bijdrage tot de oudste geschiedenis van Bolsward’, Nieuwsbulletin van de Koninklijke Nederlandse Oudheidkundige Bond 6 (9), 217–48. — 2000, Frieslands Oudheid: Het Rijk van de Friese Koningen, Opkomst en Ondergang (Utrecht). Heidinga, H. A. 1986, ‘Midden-Nederland tussen Friezen, Franken en Saksen’, Westerheem 35, 128–40. — 1997, Frisia in the First Millenium. An Outline (Utrecht). Heidinga, H. A. and A. A. Verhoeven 1992, ‘Handel en wandel in de vroege middeleeuwen’. In Teksten van lezingen gehouden tijdens het symposium ‘Handel, handelsplaatsen en handelswaar vanaf de Vroege Middeleeuwen in de Lage Landen’ te Rotterdam van 2 t/m 3 november1990, ed. A. Carmiggelt, Rotterdam Papers. Contributions to Medieval Archaeology (Rotterdam), 71–5. Henderikx, P. A. 1986, ‘The lower delta of the Rhine and the Maas: Landscape and habitation from the Roman period to ca 1000’, Berichten van de Rijksdienst voor het Oudheidkundig Bodemonderzoek 36, 447–599. — 2014a, ‘Vroege middeleeuwen’. In Geschiedenis van Zeeland deel I Prehistorie–1550. eds. P. Brusse and P. Henderikx (Zwolle), 61–79. — 2014b, ‘Landschap, bewoning, sociale structuren’. In Geschiedenis van Zeeland deel I Prehistorie–1550. eds. P. Brusse and P. Henderikx (Zwolle), 91–106. Henstra, D. J. 2000, ‘The Evolution of the Money Standard in Medieval Frisia. A Treatise on the History of the Systems of Money of Account in the Former Frisia (c. 600–c. 1500)’ (Ph.D. thesis, University of Groningen). Hilker-Suckrau, M. 1980, ‘Sachbezeichnungen aus dem Bereich der Hausbaus im Frühmittelalter’. In Wörter und Sachen im Lichte der Bezeichnungsforschung, ed. R.

131

Z01 Hines Book B.indb 131

07/05/2021 13:30

Frisians of the Early Middle Ages Schmidt-Wiegard (Berlin–New York), 58–73. Hullegie, A. G. J. and W. Prummel 2015, ‘Dieren op en rond de Achlumer terp’. In Graven aan de voet van de Achlumer dorpsterp. Archeologische sporen rondom een terpnederzetting, ed. J. A. W. Nicolay and G. J. de Langen, Jaarverslagen van de Vereniging voor Terpenonderzoek 97 (Groningen), 134–59. Hoek, C. 1979, ‘De heren van Voorne en hun heerlijkheid’. In Van Westvoorne tot St. Adolfs­ land. Historische verkenningen op Goeree-Overflakkee, ed. H. C. R. Ariese (Ouddorp), 115–45. Honée, E. 1996, ‘Das Apostolat Willibrords in Toxandrien. Ablauf, Ergebnisse und Voraussetzungen einer langjährigen Diskussion’, Nederlands Archief voor Kerkgeschiedenis/ Dutch Review of Church History 76, 101–27. IJssennagger, N. L. 2017, ‘Central Because Liminal. Frisia in a Viking Age North Sea World’ (Ph.D. thesis, University of Groningen). Jankuhn, H. 1980, ‘Beobachtungen und Ueberlegungen zur “Infrastruktur” des wikingerzeitlichen Seehandels’, Offa 37, 146–53. — 1984, ‘Die historische und sozialgeschichtliche Bedeutung der Handelsplätze’. In Archäologische und naturwissenschaftliche Untersuchungen an ländlichen und frühstädtischen Siedlungen im deutschen Küstengebiet vom 5. Jahrhundert v. Chr. bis zum 11. Jahrhundert n. Chr. 2: Handelsplätze des frühen und hohen Mittelalters, eds. H. Jankuhn, K. Schietzel, H. Reichstein (Weinheim), 441–9. Jansen, H. P. H. 1976, ‘Sociaal-economische geschiedenis’. In Historie van Groningen. Stad en Lande, eds. W. J. Formsma et al. (Groningen), 123–46. — 1982, ‘Handel en nijverheid 1100–1300’. In Algemene Geschiedenis der Nederlanden II (Haarlem), 148–86. Kaspers, A. 2019, ‘Wat scherven vertellen over terpen en handel’, De Vrije Fries 99, 169–74. Keuning, H. J. 1968, ‘Het geografisch milieu’. In Geschiedenis van Friesland, eds. J. J. Kalma, J. J. Spahr van den Hoek and K. de Vries (Drachten), 577–600. Klok, R. H. J. 1974–75, ‘Terpen zullen ons een zorg zijn’, Groningse volksalmanak. Historisch jaarboek voor Groningen 1974–75, 129–66. Knol, E. 1993, ‘De Noordnederlandse Kustlanden in de Vroege Middeleeuwen’ (Ph.D. thesis, Vrije Universiteit Amsterdam). Kort, J. C. 1981, Het archief van de graven van Holland 889–1581 (’s-Gravenhage). Krämer, R. 1984, ‘Historisch-geographische Untersuchungen zur Kulturlandschaftsentwicklung in Butjadiingen. Mit besondere Berücksichtigung des milttelalterlichen Marktortes Langewarden’, Probleme der Küstenforschung im südlichen Nordseegebiet 15, 65–126. Lebecq, S. 1992, ‘The Frisian trade in the Dark Ages; a Frisian or a Frankish/Frisian trade?’ In Teksten van lezingen gehouden tijdens het symposium ‘Handel, handelsplaatsen en handelswaar vanaf de Vroege Middeleeuwen in de Lage Landen’ te Rotterdam van 2 t/m 3 november 1990, ed. A. Carmiggelt, Rotterdam Papers 7 (Rotterdam), 7–15. — 1983, Marchands et navigateurs frisons du Haut Moyen Âge. 2 vols. (Lille). Loveluck, C. and D. Tys 2006, ‘Coastal societies, exchange and identity along the Channel and Southern North Sea shores of Europe, AD 600–1000’, Journal of Maritime Archaeology, 140–69. Meeder, S. and E. Goosmann 2018, Redbad. Koning in de marge van de geschiedenis (Houten). Meilink, P. A. 1951, Het archief van de abdij van Egmond, vol. I (Den Haag). Metcalf, M. and W. op den Velde 2010, ‘The monetary economy of the Netherlands, c. 690–c. 760 and the trade with England: A study of the “porcupine” sceattas of Series

132

Z01 Hines Book B.indb 132

07/05/2021 13:30

Landscape, Trade and Power E – vol. II’, Jaarboek voor Munt- en Penningkunde 97, 96–7. Miedema, M. 1990, ‘Oost-Fivelingo 250 v.Chr.–1850 n.Chr.; archeologische kartering en be­schrijving van 2100 jaar bewoning in Noordoost-Groningen’, Palaeohistoria 32, 111–245. — 1983, ‘Vijfentwintig eeuwen bewoning in het terpenland ten noordwesten van Groningen’ (Ph.D. thesis, Vrije Universiteit Amsterdam). Molenaar, S., C. M. Soonius and D. Bekius 2009, Noord-Holland Laagland. De archeologie en het landschap in 7 lagen. RAAP-rapport 1838 (Weesp). Naismith, R. 2017, Medieval European Coinage 8. Britain and Ireland c. 400–1066 (Cambridge). Naylor, J. 2007, ‘The circulation of early-medieval European coinage: a case study from Yorkshire, c. 650–c. 867’, Medieval Archaeology 51, 41–61. — 2012, ‘Coinage, trade and the origins of the English emporia, ca. AD 650–750’. In From One Sea to Another, eds. S. Gelichi and R. Hodges, SCISAM 3 (Turnhout), 237–66. Nicolay, J. A. W. 2014, The Splendour of Power: Early Medieval Kingship and the Use of Gold and Silver in the Southern North Sea Area (5th to 7th century AD), Groningen Archaeological Studies 28 (Groningen). Nicolay, J. A. W. and G. J. de Langen 2015, Graven aan de voet van de Achlumer dorpsterp. Archeologische sporen rondom een terpnederzetting, Jaarverslagen van de Vereniging voor Terpenonderzoek 75 (Groningen). Nicolay, J., M. Schepers and A. Nieuwhof 2018, ‘Ulrum: dubbelwierde op een markante kwelderwal’. In De geschiedenis van terpen- en wierdenland. Een verhaal in ontwikkeling, ed. A. Nieuwhof, J. Nicolay and J. Wiersma, Jaarverslagen van de Vereniging voor Terpen­onderzoek 100 (Groningen), 173–96. Nicolay, J. A. W. and P. C. Vos 2010, ‘De bewoningsgeschiedenis van Dongeradeel en het belang van de middeleeuwse zoutwinning in Friesland’. In Terpbewoning in oostelijk Friesland. Twee opgravingen in het voormalige kweldergebied van Oostergo, ed. J. A. W. Nicolay, Groningen Archaeological Studies 10 (Groningen), 173–215. Nieuwhof, A. and M. Schepers 2016, ‘Synanthropic salt marshes in the coastal area of the northern Netherlands from around 600 BC’, Archaeological Review from Cambridge 31 (2), 48–74. Niermeijer, J. F. 1937, ‘Over het staatsgezag in Midden-Friesland, voornamelijk in de twaalfde eeuw. Een diplomatiek onderzoek’, Bijdragen voor Vaderlandsche geschiedenis en Oudheidkunde 7 (8), 1–33. — 1977a, ‘De Friese handel’. In Welvaart in wording, Sociaal-economische geschiedenis van Nederland van de vroegste tijden tot het einde van de middeleeuwen, eds. W. J. Alberts and H. P. H. Jansen, 2nd edition (Den Haag), 20–38. — 1977b, ‘De herleving van de handel en de opkomst van de steden tot het midden van de 13e eeuw’. In Welvaart in wording, Sociaal-economische geschiedenis van Nederland van de vroegste tijden tot het einde van de middeleeuwen, eds. W. J. Alberts and H. P. H. Jansen, 2nd edition (Den Haag), 39–56. Nokkert, M., A. C. Aarts and H. L. Wynia 2009, Vroegmiddeleeuwse bewoning langs de A2. Een nederzetting uit de zevende en achtste eeuw in Leidsche Rijn, Gemeente Utrecht, Stads­ontwikkeling, basisrapportage archeologie 26 (Utrecht). Noomen, P. N. 1996a, ‘Jorwerd’. In Verborgen verleden belicht. Introductie tot het historische en archeologische archief van Friesland, eds. G. J. de Langen. J. A. Mol, P. N. Noomen and L. Oldersma (Sneek), 30–3. — 1996b, ‘Middeleeuwse bezitsverhoudingen in Noordoost-Fivelgo’. In Het Bierumer

133

Z01 Hines Book B.indb 133

07/05/2021 13:30

Frisians of the Early Middle Ages Boerderijenboek. Een bijdrage tot de geschiedenis van Noordoost-Fivelgo, ed. E. de Boer (Scheemda), 59–82. Numan, A. 2014, ‘Een verblijfplaats van de graven van Holland in Haarlem’, Westerheem 2014 (3), 58–71. Op den Velde, W. and C. J. F. Klaassen 2004, Sceattas and Merovingian Deniers from Domburg and Westenschouwen, Werken uitgegeven door het Koninklijk Zeeuwsch Genootschap der Wetenschappen 15 (Middelburg). Op den Velde, W. and M. Metcalf 2003 (2007), ‘The monetary economy of the Netherlands, c. 690–c. 715 and the trade with England: a study of the sceattas of Series D’, Jaarboek voor Munt- en Penningkunde 90. Pol, A. 2008, ‘Boeles en het mysterie van het Dronrijp-type’. In Diggelgoud. 25 jaar Argeologysk Wurkferbân: Archeologisch Onderzoek in Fryslân, eds. K. Huisman, K. Bekkema, J. M. Bos, H. de Jong, E. Kramer, R. Salverda (Leeuwarden), 162–9. Poelman, H. A. 1908, Geschiedenis van den handel van Noord-Nederland gedurende het Mero­vingische en Karolingische tijdperk (Amsterdam). Postma, D. 2015, Het zodenhuis van Firdgum. Middeleeuwse boerderijbouw in het Friese kust­ gebied tussen 400 en 1300 (Groningen). Postma, O. 1934, De Friesche kleihoeve. Bijdrage tot de geschiedenis van den cultuurgrond vooral in Friesland en Groningen (Leeuwarden). Prummel, W. 2014, ‘Schapenteelt in terpengebied belangrijker dan gedacht’. In Van Wierden en Terpen, Nieuwsbrief van de Vereniging voor Terpenonderzoek 19, 19–21. Reinhardt, W. 1959, ‘Die Grabungen auf der Dorfwarf von Groothusen, Krs. Nord, und ihre Ergebnisse’, Jahrbuch der Gesellschaft für bildende Kunst und vaterländische Altertümer zu Emden 39, 20–36. — 1965, ‘Studien zur Entwicklung des ländlichen Siedlungsbildes in den Seemarschen der ostfriesischen Westküste’, Probleme der Küstenforschung im südlichen Nordseegebiet 8, 73–148. — 1970, ‘Untersuchungen zur Stadtkernforschung in Emden’, Probleme der Küstenforschung im südlichen Nordseegebiet 9, 101–12. Rombaut, H. 1990, ‘Het grondbezit van de abdij van Echternach’. In Willibrord. Zijn wereld en zijn werk. Voordrachten gehouden tijdens het Willibrordcongres Nijmegen, 28–30 september 1989, eds. P. Bange and A. G. Weiler (Nijmegen), 177–83. Schepers, M. 2014, Reconstructing Vegetation Diversity in Coastal Landscapes, Advances in Archaeobotany 1 (Eelde). — 2016, ‘Gebruiksplanten in het terpen- en wierdengebied’. In Van Wierhuizen tot Achlum. Honderd jaar archeologisch terpenonderzoek in terpen en wierden, ed. A. Nieuw­ hof, Jaarverslagen van de Vereniging voor Terpenonderzoek 98 (Groningen), 114–32. Schepers, M. and G. J. de Langen 2019, ‘Efficiënte kwelderboeren: resultaten van een buitendijks experiment’, De Vrije Fries 99, 161–3. Schlesinger, W. 1956, ‘Über germanisches Heerkönigtum’. In Das Königtum. Seine geistigen und rechtlichen Grundlagen. Vorträge und Forschungen 3, ed. Th. Mayer (Lindau–Konstanz), 105–41. Schmid, P. 1988, ‘Die Mittelalterliche Neubesiedlung der Niedersäsischen Marsch’. In Archeologie en Landschap, eds. M. Bierma, O. H. Harsema and W. van Zeist (Groningen), 133–64. Schuur, J. R. G. 1979, Leeuwarden tot 1435. Een poging tot reconstructie van de oudste stadsgeschiedenis (Zutphen).

134

Z01 Hines Book B.indb 134

07/05/2021 13:30

Landscape, Trade and Power Segschneider, M. 2002, ‘Fränkisches Glas im Dünensand. Ein Strandmarkt des 5. Jahrhunderts auf der nordfriesischen Insel Amrum und die völkerwanderungszeitliche Handelsroute zwischen Rhein und Limfjord’, Archäologisches Korrespondenzblatt 32, 117–36. Siegmüller, A. and H. Jöns 2012, ‘Ufermärkte, Wurten, Geestrandburgen. Herausbildung differenter Siedlungstypen im Küstengebiet in Abhängigkeit von der Paläotopographie im 1. Jahrtausend’, Archäologisches Korrespondenzblatt 42, 573–90. Slicher van Bath, B. H. 1948, Boerenvrijheid (inaugural lecture, University of Groningen). — 1949, ‘Problemen rond de Friese Middeleeuwse geschiedenis’. In Herschreven Historie, Schetsen en studien op het gebied der middeleeuwse geschiedenis (Leiden), 259–80. — 1968, ‘Middeleeuwse welvaart’. In Geschiedenis van Friesland, eds. J. J. Kalma, J. J. Spahr van den Hoek and K. de Vries (Drachten), 201–28. Theuws, F. C. J. W. 2004, ‘Exchange, religion, identity and central places in the Early Middle Ages’, Archaeological Dialogues10, 121–38. — 2018, ‘Reversed directions. Re-thinking sceattas in the Netherlands and England’, Zeitschrift für Archäologie des Mittelalters 46, 27–84. Van Buijtenen, M. P. 1953, De grondslag van de Friese vrijheid (Assen). Van Dierendonck, R. M. 2009, ‘The early medieval circular fortresses in the province of Zeeland, the Netherlands: ten years after’. In Ringwälle und verwandte Strukturen des ersten Jahrtausends n. Chr. an Nord- und Ostsee. Symposium Utersum 2005, ed. M. Segschneider, Schriften des Archäologischen Landesmuseums Ergänzungsreihe 5 (Neumünster), 249–74. Van Egmond, W. S. 2005, ‘Radbod van de Friezen, een aristocraat in de periferie’, Millennium 19, 24–44. Van Es, W. A. 1971, Terpen (Kampen). — 1990, ‘Dorestad centred’. In Medieval Archaeology in the Netherlands. Studies Presented to H.H. van Regteren Altena, eds. J. C. Besteman, J. M. Bos and H. A. Heidinga (Assen– Maastricht), 151–282. — 1994, ‘Friezen, Franken en Vikingen’. In Romeinen, Friezen en Franken in het hart van Nederland. Van Trajectum tot Dorestad 50 v.Chr. – 900 n. Chr, eds. by W. A. van Es and W. A. M. Hessing (Utrecht–Amersfoort), 82–119. Van Es, W. A. and W. J. H. Verwers 1994, ‘Handel in Karolingische potten’. In Romeinen, Friezen en Franken in het hart van Nederland. Van Trajectum tot Dorestad 50 v.Chr. – 900 n. Chr, eds. by W. A. van Es and W. A. M. Hessing (Utrecht–Amersfoort), 184–8. Van Es, W. A., W. J. H. Verwers, and C. Isings 2015, Excavations at Dorestad 4. The Settlement on the River Bank Area, Nederlandse Oudheden 18 (Amersfoort). Van Giffen, A. E. 1936, ‘Der Warf in Ezinge, Provinz Groningen. Holland, und seine westgermanische Häuser‘, Germania 20, 40–7. Van Heeringen, R. M. 1995, ‘Kolonisatie en bewoning in het mondingsgebied van de Schelde in de vroege Middeleeuwen vanuit archeologisch perspectief ’. In Vroeg-Middeleeuwse ringwalburgen in Zeeland, eds. R. M. van Heeringen, P. A. Henderikx and A. Mars (Goes–Amersfoort), 41–69. Van Leeuwen, J. 2014, Middeleeuws Medemblik. Een centrum in de periferie. Archeologisch onderzoek naar de (vroeg)middeleeuwse handelsnederzetting en het oudste regionale centrum van West-Friesland in de periode 675–1289, Archeologie West-Friesland, WestFriese Archeologische Rapporten 61 (Hoorn). Varwijk, T. W. and G. J. de Langen 2018, Opgraving Wommels-Stapert. Terug na 20 jaar: Nieuw archeologisch onderzoek aan de commercieel afgegraven terp Stapert bij Wommels

135

Z01 Hines Book B.indb 135

07/05/2021 13:30

Frisians of the Early Middle Ages in het hart van Westergo (Friesland), Grondsporen 35 (Groningen). — 2019, Opgraving Archeologisch onderzoek aan de zool van de commercieel afgegraven terp Saksenoord bij Lollum in Westergo (Friesland), Grondsporen 37 (Groningen). Varwijk, T. W., Y. Boonstra and G. J. de Langen 2019, ‘Een opgraving achter het Stadshuis: een nieuwe blik op middeleeuws Bolsward’, De Vrije Fries 99, 180–5. Verkerk, C. L. 1992, ‘Het tolsysteem in het mondingsgebied van Rijn, Maas en Schelde tot de elfde eeuw?’. In Teksten van lezingen gehouden tijdens het symposium ‘Handel, handels­ plaatsen en handelswaar vanaf de Vroege Middeleeuwen in de Lage Landen’ te Rotterdam van 2 t/m 3 november1990, ed. A. Carmiggelt, Rotterdam Papers. Contributions to Medieval Archaeology 7 (Rotterdam), 39–49. Verhulst, A. 2002, The Carolingian Economy (Cambridge). Verhulst, A. and R. de Bock-Doehaerd 1981, ‘Handel en nijverheid’. In Algemene geschiedenis der Nederlanden I, eds. D. P. Blok et al. (Bussum), 183–215. Vis, G. N. M. 1990, ‘De twaalf gezellen van Willibrord’. In Willibrord, zijn wereld en zijn werk. Voordrachten gehouden tijdens het Willibrordcongres Nijmegen, 28–30 september 1989, eds. P. Bange and A. G. Weiler (Nijmegen), 128–49. Volquartz, J. B. 2018, ‘Der fränkisch-friesische Konflikt (690–734) aus Sicht mittelalterlicher Quellen’, Emder Jahrbuch 98, 9–32. Vos, P. C. 2015, Origin of the Dutch Coastal Landscape. Long-Term Landscape Evolution of the Netherlands during the Holocene, Described and Visualized in National, Regional and Local Palaeogeographical Map Series (Groningen). Vos, P. C. and E. Knol 2015, ‘Holocene landscape reconstruction of the Wadden Sea area between Marsdiep and Weser’, Netherlands Journal of Geosciences 94 (2), 157–83. Vos, P. and S. de Vries 2013, 2e generatie paleogeografische kaarten van Nederland (versie 2.0) (Utrecht). (Consulted on 13 May 2018, via www.archeologieinnederland.nl). Wampach, C. 1930, Geschichte der Grundherrschaft Echternach, 2 vols. (Luxembourg). Weiler, A. G. 1989, Willibrords missie. Christendom en cultuur in de zevende en achtste eeuw (Hilversum). Werner, K. F. 1980, Der Lütticher Raum im frühkarolingischer Zeit. Untersuchungen zur Geschichte einer karolingischer Stammlandschaft, Veröffentlichungen des Max-Planck-Instituts für Geschichte 62 (Göttingen.) Wood, I. 1994, The Merovingian Kingdoms 450–751 (London–New York). Worst, D. in prep., ‘Middeleeuwse veenontginningen in het land van Kuinder en Linde’ (Ph.D. thesis, University of Leiden). Zomer, J. 2016, ‘Middeleeuwse veenontginningen in het getijdebekken van de Hunze. Een interdisciplinair landschapshistorisch onderzoek naar de paleogeografie, ontginning en waterhuishouding (ca 800 – ca 1500)’ (Ph.D. thesis, University of Groningen).

136

Z01 Hines Book B.indb 136

07/05/2021 13:30

• 5 • LAW AND POLITICAL ORGANIZATION OF THE EARLY MEDIEVAL FRISIANS (c. AD 600–800) Han Nijdam

T

he newcomers who settled the Frisian coastal area after c. 400 CE brought their own law and political organization with them, which in later centuries developed organically, this new population meanwhile becoming ‘Frisians’. The drawing up of Lex Frisionum (c. 785) enables us to say something about Frisian law in that early period (c. 400–800), using the later Old Frisian material (c. 1050–1500) to fill some of the gaps. The political organization of the Early-medieval Frisians underwent considerable change during the Medieval Period, especially at the highest level. At the ground level, however, the institution of the assembly or thing, where law, politics, military affairs and religion came together was the constant factor that remained in place almost everywhere. There is also the enigmatic phenomenon of the Frisian kings who are mentioned in the Early-medieval sources, of which Radbod (Redbad), who died in 719, was the most famous. What was the status of these Frisian kings? How many kings were there at one time among the Frisians? What did Frisian kingship look like in this period? In this contribution, I will focus on the triad ‘law – thing – king’ during the period between c. 600 and c. 800 CE. At the end of this period, the Frisians had been conquered by the Carolingians and the Viking raids were about to begin. After the Frisians had been conquered by the Carolingians, the latter appointed counts to govern Frisia. These counts in turn were hampered in establishing a firm grip on the Frisians by the Viking raids. I subscribe to the view put forward by Gilles de Langen and Hans Mol (2017; this vol.), that Frisia experienced a Viking Period between c. 850 and 950, during which the grip of the Franks on the region was loosened. When the Carolingian Empire was able to get a foothold again, the counts who ruled over Frisia between Vlie and Weser during the tenth to twelfth centuries never permanently resided in Frisia itself. After 950, a network of parishes and churches was quickly built up by the bishops of Utrecht and Münster. For this to succeed, the bishops had to negotiate with local elites in order to get the churches which they needed for their closely knit parish network funded and raised. But that, as they say, is another story. My focus will be on the most archaic period, as well as on the tricky phenomenon of Frisian kingship.

137

Z01 Hines Book B.indb 137

07/05/2021 13:30

Frisians of the Early Middle Ages

Law Old Frisian law is the medieval Frisian law tradition of the Frisian regions between the Rivers Vlie and Weser that was for the most part written down in the Frisian vernacular. Even though the oldest complete manuscripts stem from the end of the thirteenth century, it has long been recognized as being extremely archaic. Although the discussion has gone back and forth, there is now a general consensus that parts of Old Frisian law are very old indeed, and that the reason for the fact that the Frisians by and large were able to hold on to their own indigenous law throughout the Middle Ages must be sought in the famous Frisian Freedom, which started in the tenth or eleventh century and lasted until the end of the Middle Ages (Vries 2015). This political autonomy enabled the Frisians to retain their own law and develop it at their own desire. Lately a debate has arisen among scholars of Old Frisian which revolves around the question of when Old Frisian law was first written down. An older generation of scholars dated the oldest texts to the eleventh century and assumed that they had first been written down in that period as well, even though no such early manuscripts have survived. In 2004, Rolf Bremmer posited the view that even though Old Frisian law does indeed show a number of archaic features, it was committed to writing only after 1200 (Bremmer 2004; 2014). Before that, he claims, there was no infrastructure to accommodate the writing down of Frisian law, let alone in the Frisian vernacular. It was only in the period after 1175 that a number of monastic orders settled in Frisia, creating that infrastructure. This hypothesis has a number of weak points. First, Bremmer overlooks the role the churches and their associated schools as well the early monasteries of Staveren and Reepsholt might have played in this respect.1 Second, following the Reformation in 1580, a tremendous amount of medieval material was destroyed (Nijdam and Savelkouls 2017; Bremmer 2004, 11–17). A few scraps of Old Frisian which predate the twelfth century testify to this. In 2015 two tiny fragments of parchment came to light with the remnants of interlinear psalm glosses in Old Frisian. The language has been dated 1100–1125 (Langbroek 2015). Until that moment another fragment of psalm glosses dated around 1200 had been the oldest piece of Old Frisian (Langbroek 1990). Third, the two oldest law texts, The Seventeen Statutes and The Synodical Law of Central Frisia, show features which point to a date of composition (i.e. as written texts) in the eleventh century (Henstra 2010, also discussed below; Hallebeek 2019). In the light of this contribution it is not necessary to go into this matter too deeply. It suffices to make two observations. First, after the recording of Frisian law in Lex Frisionum, a few centuries elapsed before Frisian law was written down again. Second, Old Frisian law is very archaic and especially for the compensation tariffs (on which more below), a direct continuity has been shown for the legal tradition of compensating injuries. This is important because this continuity shows that the Frisian legal tradition of the seventh and eighth centuries was not hampered by the political turmoils of the period between c. 800 and 1100. This in itself is indirect corroboration of the Frisian Viking Period 850–950 identified by De Langen and Mol and its effect on the grip of the Carolingians on the Frisians. Another way of saying this is that the Frisians from c.

1

Dirk Jan Henstra (2010, 109) suggests The Seventeen Statutes might have been written down with the assistance of the monastery of Reepsholt.

138

Z01 Hines Book B.indb 138

07/05/2021 13:30

Law and Political Organization 800 onwards were able to evade all incoming foreign political threats and by and large go about their legal business as they had always done.

Lex Frisionum: the source and its historiography Lex Frisionum is the most important source for our knowledge of Early-medieval Frisian law. Johannes Herold, who edited the text in 1557, had a medieval manuscript at his disposal that probably contained Lex Saxonum, Lex Thuringorum and Lex Frisionum as a group of three texts (Siems 1980, 118–28). That manuscript is now lost. Since it is now also clear that Lex Frisionum was never officially accepted as law by the Carolingian king and must be viewed as a draft, it is no surprise that only a few copies circulated in the Middle Ages. No copy survived and no other manuscripts were used by later editors, making the Herold text the sole primary source. The absence of medieval manuscripts made some scholars of the past suspicious of the authenticity of Lex Frisionum. Siems, however, pointed to medieval references to Lex Frisionum and manuscripts containing the text (Siems 1980, 49–53) and to the fact that Lex Frisionum contains details (e.g. concerning the borders of the three Frisian areas, on which more below) which Herold could not have known. Moreover, recent articles on the compensation tariffs in Lex Frisionum, comparing it with the Kentish laws (Nijdam 2017) and with the younger Old Frisian tradition (Nijdam 2019) make it absolutely clear that Lex Frisionum is exactly what it claims to be: a late eighth-century result of writing down Frisian customary law, with Frankish material added. The fact that only Herold’s text survived does not make matters easier, for Lex Frisionum is a very complicated text (see Appendix). It consists of various, diverse blocks of information. On top of that, what were probably marginal glosses in the original manuscript have been inserted into the text by Herold, adding to the intricacy of the witness. Moreover, it was probably Herold who divided the text into titles and paragraphs (or at least partly so). Lastly, again probably Herold moved a paragraph into the (in his edition) adjacent Lex Thuringorum. In all, Lex Frisionum presents a very complicated puzzle on which much has been written during the last three centuries. Before moving to the description of the text and a discussion of its contents, I would like to make some historiographical observations. If Siems is correct in his analysis of a correspondence between the Frisian scholar Joachim Hopperus (1523–1576) and the Flemish theologians George Cassander2 (1513–1566) and Cornelis Wouters3 (1500– 1577), then Hopperus acquired a copy of Lex Frisionum in 1555 from these two men, two years before the Herold edition (Siems 1980, 50).4 The Herold edition itself was received rather quickly in Frisia: amongst the manuscripts of the first state historian of Friesland, Suffridus Petrus (1527–1597) is a complete transcription of the Herold text, dated c. 1590.5 A few decades later, in 1617, the humanist scholar Sibrandus Siccama (1571–1622)

2 Alias Cassant. 3 Alias Cornelius Walterius or Cornelius Gualtherus. 4 The antiquarian activities of Joachim Hopperus and Viglius van Aytta (1507–1577), both serving the Habsburgs and living outside Frisia, require more study. Both acquired large collections of Frisian antiquities during their lifetimes, but both collections have largely been destroyed as a result of various calamities. 5 Tresoar Leeuwarden, MS. 65 Hs B.

139

Z01 Hines Book B.indb 139

07/05/2021 13:30

Frisians of the Early Middle Ages delivered a new, heavily annotated edition of Lex Frisionum. For these notae, he made use of an Old Frisian law manuscript that is now lost (Nijdam and Versloot 2012). Since the classic study by Siems (1980), the most important new insights into Lex Frisionum were made by the economic historian Dirk Jan Henstra (1927–2016), who studied the money standard in medieval Frisia 600–1500 (Henstra 2000 and 2010). He also published a pioneering monograph on the counts of medieval Frisia between 700 and 1200 (Henstra 2012). His work deserves more study and corroboration, since his wergild hypothesis – i.e., that the standard wergild or blood money (more on this below) of a Frisian freeman was kept constant at 1,664 grams or 64 German ounces of silver throughout the Middle Ages – is both daring and potentially a powerful tool to analyse sources with.6 Apart from the important studies by Henstra, not much original research has been done on Lex Frisionum. A few other recent studies should also be mentioned though. In 2000, legal historian Nico (N. E.) Algra (1927–2002) published his study of Old Frisian law, 800–1256 (Oudfries recht 800–1256), in which he paid considerable attention to Lex Frisionum. The study is not easy to get to grips with: Algra had a very specific view on the origins of the Frisian Freedom (Algra 2000) and the role of the oath of fidelity the free Frisians had to swear to the Carolingian king. Apart from this, he presents some useful analyses of parts of Lex Frisionum, often elaborating on Siems’s work. Rijk Timmer (2000) searches for vestiges of ‘Old Germanic law’ in Lex Frisionum. As has been observed by previous scholars, these are the paragraphs on killing a newborn child (Tit. V,1), punishments for temple robbers (AS Tit. XI), and lot casting (Tit XIV). Problematic is the fact that he discards the rich tariff material because it would have been ‘heavily influenced’ by the Leges Alamannorum (Timmer 2000, 26–7). This view had already been put forward by Siems, but, as we shall see below, can now be proved wrong. The tariffs are Frisian and an important source of information. Lex Frisionum: a description of its contents The text consists of twenty-two titles (Tituli, abbrev. ‘Tit.’), each containing a number of paragraphs, which yields a total of 194 paragraphs (of which 89 make up Tit. XXII De Dolg ‘on injuries’). To this core text 114 paragraphs were added in a section called Additiones Sapientum (‘AS’) or ‘additions of the wise men’, where two legal experts by the names of Wlemar and Saxmund give additional material. This then adds up to a total sum of 308 paragraphs. Next to the Additiones Sapientum, another important feature of Lex Frisionum is that it distinguishes three main Frisian areas: _

1. Inter Flehi et Sincfalam – between the rivers Sincfal and Vlie (i.e., the modern provinces of Zeeland, South- and North-Holland in the Netherlands); 2. Inter Laubachi et Flehum – between the rivers Vlie and Lauwers (i.e. the modern province of Fryslân/Friesland in the Netherlands)

6 One example is Henstra’s article, ‘De eerste optekening van de algemeen-Friese Keuren’ [The First Recording of the pan-Frisian Statutes] (Henstra 2010, 97–126), in which he shows that the wergild mentioned in the 15th statute of the pan-Frisian law text The Seventeen Statutes most likely dates from the period 1015–1040, thereby corroborating the view that the oldest Old Frisian law texts were written down in the eleventh century.

140

Z01 Hines Book B.indb 140

07/05/2021 13:30

Law and Political Organization 3. Inter Laubachi et Wisaram – between the rivers Lauwers and Weser (i.e. the modern provinces of Groningen in the Netherlands and Ostfriesland in Germany).

-

Of these regions, the Central Frisian area, coinciding with the modern province of Friesland, is the default of the law text. The main text is written for that region, and the situations in the other regions are given. Note too, that the orientation varies between from east to west (cf. inter Flehi et Sincfalam) and from the perspective of Central Frisia (cf. inter Laubachi et Wisaram). Furthermore, it is thought that Wlemar and Saxmund came from East and West Frisia respectively, on which more below. The identification of these three regions – Central, West and East – fits well with other and later information concerning the Frisians. Although each of these areas in turn was built up of even smaller entities – pagi or shires, on which more below – the rivers mentioned were indeed well-known borders. The River Lauwers, for instance, is an old dialect border between what is known as Old East Frisian and Old West Frisian (i.e. the dialects of East Frisia and Central Frisia) (Bremmer 2009; Versloot, this vol.).7 Also, the river Vlie was an important border between Central and West Frisia (De Koning 2018). In short, this division of Frisia into its three parts seems to reflect an indigenous situation and was not something that was imposed by the Carolingians. Lex Frisionum presents a mixture of indigenous and Frankish legal material; it is a compilation which has been worked on for a long period of time. Dirk Jan Henstra has shown that the money of account in Lex Frisionum reflects a change from an older system based on the Roman gold solidus to a newer system based on new silver pennies, minted by the Anglo-Saxons and the Frisians (the so-called sceattas) of 1.3 grams of silver (Henstra 2010, 47–70). The Carolingian monetary reform of AD 793, which entailed the introduction of a new, heavier silver penny of 1.7 grams, has not been taken into consideration in the text, hence Henstra’s dating of the text to before that date. In this light, it is interesting to note that Henstra observes an even older layer, pointing to the Frisians using the older and heavier Byzantine gold solidus (the wergild calculated in this unit of account was the round number of 50 solidi), which had been replaced by the slightly lighter Roman solidus after 575 (yielding the wergild we find in Lex Frisionum of 53⅓ solidi) (Henstra 2000, 289–90). In all then, the indigenous material of Lex Frisionum might be of a considerable age, reaching back into the seventh, possibly even the sixth, century. Simultaneously, Lex Frisionum also contains material that testifies to Carolingian rule. This Frankish layer also comprises regulations referring to the Church and the christianization of the Frisians. The vernacular words in Lex Frisionum are not always Frisian in form. This had already been observed for the vernacular words in the titles, but Arjen Versloot has recently shown (2015) that even the vernacular words in the text itself do not present the Frisian forms we would expect, even if the text undoubtedly concerns indigenous technical terms. The most likely explanation for this would be that Lex Frisionum was written down by Frankish scribes, who rendered the Frisian vernacular words in the text Frankish. The Carolingian king asserts himself in the text in a number of ways. At various points, we find the royal ban or bannus, accompanied by a fine for breaking the king’s 7

The nomenclature has to do with the fact that the area between Sincfal and Vlie became the county of Holland and Zeeland during the High Middle Ages, from which no Old Frisian sources are known.

141

Z01 Hines Book B.indb 141

07/05/2021 13:30

Frisians of the Early Middle Ages peace, the fredus (Lambert and Rollason 2009). The ban was a Germanic phenomenon, most often used at the assembly or thing meeting, where peace was ordered. The PIE root *bhe- of ProtoGermanic *banna- means ‘to speak’, so the original meaning must have been ‘to order, to command’ (Boutkan and Siebinga 2005, s.v. BON 1). If the ban or order had been disobeyed, the peace or fredus had been broken, and a fine was exacted from the perpetrator. It has been observed by scholars that the king tried to fine especially those cases that could lead to a feud or were part of the hostilities that arose during a feud (Algra 2000, 117–18, 185–99).8 Another category is addressed in Tit. XVII Hic bannus est ‘this is the ban’. It consists of five paragraphs in which the king puts high fines on fighting amongst each other in the army (in exercitio), and killing someone in a church or in curte ducis ‘at the court of the dux’. There has been ample discussion on what was meant by dux in this context. It must have been a special official appointed by the king. It probably was a margrave (Lat. marchio), a military commander assigned to maintain the defence of one of the border provinces, in this case Frisia (Siems 1980, 324–5; Algra 2000, 107–8). Also under this Titulus, the envoys of the king and the dux are protected by high fines: 2. Qui in curte ducis, in ecclesia, aut in atrio ecclesiae hominem occiderit, novies weregildum eius componat, et novies fredam ad partem dominicam. 3. Si quis legatum regis vel ducis occiderit, similiter novies illum componat, et fredam similiter novies ad partem dominicam. 2. If someone kills a man at the court of the margrave, in the church or in the hall of the church, he has to pay nine times his wergild and a ninefold fine for breaking the peace to the king. 3. If someone kills an envoy of the king or the margrave, he also has to compensate him ninefold, and also has to pay a ninefold fine for breaking the peace to the king.

Another new feature is that the king uses the concept of wergild as an instrument of penalization: a perpetrator has to pay his own wergild to the king and on top of that a fine of 30 solidi for breaching the peace. In some extreme cases, a ninefold fine is exacted (Siems 1980, 321–6). In a few rare instances we can discern the Frankish layer and see it contradict indigenous Frisian law. This happens in Tit. XIV,7, where a case is described in which a man has been killed in a crowd and it is difficult to identify the actual killer. The heir can accuse someone and this defendant is in turn allowed to point out someone else who is then accused of the homicide. The matter can then be resolved by means of a judicial duel. Both parties are allowed to hire a professional champion. Surprisingly, Tit. XIV,7 stipulates that if one of the champions is killed, the one who hired him has to pay 60 solidi to the king and on top of that has to pay the wergild of the deceased champion. An amount of 60 solidi can be found twice more in the entire Lex Frisionum. In Tit. III,8–9 it is presented as a wergild that has to be paid. But the normal Frisian wergild was 53⅓ solidi. This 60-solidi wergild was a new, Frankish one. To find it in Tit. XIV,7, 8

Tit. VII,2 (arson); Tit. XX,1–2 (killing a hostage; homicide and hiding the corpse); AS Tit. I,1–2 (killing someone in a feud setting who is in the church, in his house, going to or from the church, or to or from the thing meeting).

142

Z01 Hines Book B.indb 142

07/05/2021 13:30

Law and Political Organization where it also seems to be a wergild is a corroboration of the suspicion that this is a Frankish addition to original Frisian legal material, for in Tit. V a number of people are enumerated who can be killed without having to pay a wergild for them. Among these are professional champions. That this is indigenous Frisian law is moreover confirmed by the fact that we also find this provision in Old Frisian law (His 1901, 257–8). The Carolingian influence in Lex Frisionum is specially concentrated in Tit. XVI– XXI, where we also find Tit. XVII Hic est bannus, which, as we have seen, commanded a peace inside the church. Here we also find Tit. XVIII, De die dominico, regulating the observance of Sunday rest. Further Christian elements are the fact that oaths are to be sworn on holy relics, and the prohibition on selling a slave to pagan peoples (Tit. XVII,5). Note too, that the ritual of casting lots to find out the killer (Tit. XIV,1–2) of a man who was killed in a crowd must have been of pre-Christian origin but was christianized: one of the two initial lots used to determine whether the guilty person is amongst the seven who have been accused is marked with a cross. Also, the ritual is performed in a church by a priest, and the lots are placed on the altar. This ritual of casting lots brings us to the pre-Christian elements in Lex Frisionum, on which much has been written (Siems 1980, 334–49; Hines, this vol.). Apart from this christianized ritual, two features can be found that most scholars agree on would have been left out in the definitive redaction of Lex Frisionum. It concerns the right of a mother to kill her newborn child without being prosecuted for it (Tit. V,1) and the description of the punishment for those who have desecrated Frisian pagan temples. These have to be ‘taken to the sea, and on the sand, which will be covered by the flood, his ears will be cleft, and he will be castrated and sacrified to the gods, whose temple he dishonoured.’9 The right to kill a newborn child is corroborated by other sources, such as Vita Liudgeri (Siems 1980, 334–5). Below we will see that in Vita Vulframni, human sacricifes are described that bear resemblance to the ritual described here for the desecrators of temples. Finally, the younger Old Frisian legal material contains no explicit references to Lex Frisionum as a text. In other words, no medieval evidence can be found of Lex Frisionum circulating as a text in Frisia. Yet, in his review of Siems (1980), Gerbenzon (1983) points to the fact that if what is said in Tit. V concerning people who can be killed without having to pay compensation is Frankish, or rather that if this material was indeed conceived in the post-Frankish period, it is remarkable to find it again, almost literally, in Old Frisian law. This applies to the description of a thief being apprehended in the tunnel he was digging (but which is also an Old Testament provision: see Exodus 22.2) but especially to that of the person who is trying to perform arson on someone’s house while the torch he is holding is touching the wall or the roof of the house (Gerbenzon 1983, 173). Since no literal traces of Lex Frisionum – i.e. as a text – have been found in Old Frisian law, the only way to explain this is that this stipulation became part of Frisian law, which was then transmitted orally until it was written down again in Old Frisian law. This needs to be investigated further.

9

ducitur ad mare, et in sabulo, quod accessus maris operire solte, finduntur aures eius, et castratur, et immolatur Diis quorum templa violavit. AS Tit. XI,1. Translation by Kees Nieuwenhuijsen. http://www. keesn.nl/lex/lex_en_text.htm

143

Z01 Hines Book B.indb 143

07/05/2021 13:30

Frisians of the Early Middle Ages The core of Lex Frisionum: wergild, injuries and compensation The bulk of Lex Frisionum deals with homicide and injuries and their compensations.10 If we add all these titles up we arrive at 248 paragraphs which deal with violence and associated compensation, or more than five out of every six of the total sum of 308 paragraphs. Clearly, this was the most important thing to settle in law. I will treat the implications of this below, when I address the topic of honour. The system of blood money or wergild is fully present in Lex Frisionum. It was a mechanism used by all Germanic peoples as well as various societies all over the world to buy off revenge for a killing by handing over a number of valuables by the killer and his kin to the closest relatives of the deceased person (Boehm 2007). This system was a way of ending or warding off an endless feud where one killing led to the next. Among the Germanic peoples, and certainly among the Frisians, not only the direct relatives or heirs were entitled to a part of the wergild, but a large circle of relatives was also entitled to a share. In Lex Frisionum the standard distribution of two-thirds for the next of kin and one-third for the wider group of relatives is already in place (Tit. I,1; Siems 1980, 283–7; Henstra 2000: 263–7). In Old Frisian law the next of kin were called the ‘six hands’ (i.e. father and mother, brother and sister, son and daughter). This group of direct heirs can also be found in Tit. XIX,2, which deals with fratricide: Si quis fratrem suum occiderit, solvat eum proximo heredi sive filium aut filiam habuerit, aut si neuter horum fuerit, solvat patri suo vel matri suae vel fratri vel etiam sorori suae; quod si nec una de his personis fuerit, solvat eum ad partem regis. If someone kills his brother, he has to pay for him to his closest heirs if he had a son or daughter. If he [the brother] did not have these, he [the killer] has to pay his father or his mother or his brother or also his sister; if none of these persons are alive, he has to pay for him to the king.

Henstra (2000, 265–6) pointed out that in the later medieval sources it is not always clear when a wergild is mentioned whether only the share of the heirs is meant (the two-thirds), or the so-called ‘full wergild’, i.e. the share of the heirs plus the share of the relatives. Several references are made to the wergild of a free or noble woman, which was the same as that of a man (AS Tit. V). The relatives who could claim a right to a share of the one-third part, which was called meitele in Old Frisian, also had to aid in a feud and pay a share in the meitele if one of their relatives committed a homicide. In Old Frisian law, this group of relatives reached back genealogically to the eight great-grandparents of the subject. We can see a web of interrelations here. If these relatives refused to help in a feud, they forfeited their right to a share of the meitele of the subject (Hoppenbrouwers 1985; Hulscher 2002; Nijdam 2008, 135). From the paragraphs on wergild and those on fratricide and parricide (Tit. XIX), we can conclude that there was a solid system of inheritance present in Early-medieval Frisia. It might therefore be seen as remarkable that other provisions concerning inheritance law are lacking from Lex Frisionum.

10 Tit. I, II, IV, V, XIV, XV, XVI, XIX, XX, XXII of the main text and Tit. II, III, V, of the Additiones Sapientum, plus the misplaced section Lex Thuringorum Tit. VI.

144

Z01 Hines Book B.indb 144

07/05/2021 13:30

Law and Political Organization The compensation tariffs in Lex Frisionum are unmatched by any of the other Leges Barbarorum as far as their length and extent are concerned (Wormald 2003, 53). Compensation tariffs were an offshoot of wergild. Once a price for a whole person was established, the various bodyparts and injuries could be valued. Again, we see this not only in the Germanic world but in many other cultures (Nijdam 2014). This rich tradition was continued in Frisia and could do so unhampered by kings and counts because of the Frisian Freedom. In the Old Frisian corpus there is an abundance of compensation tariffs from all areas between the Rivers Vlie and Weser (Nijdam 2008). It is also safe to state that the tariffs go back a long time. They must have been brought to Frisia by the ‘new Frisians’ around the year 400. A comparison with the tariffs in the Law of Aethelberht (c. AD 600) shows that these two traditions, i.e. the Frisian and the Kentish, are very much alike (Nijdam 2017). They do, however, show structural differences, making it unlikely that they stem from a common proto-tariff. They rather show cultural similarities, in line with so many other Anglo-Frisian parallels. Furthermore, as has been said earlier, Dirk Jan Henstra observed that the solidi mentioned in Lex Frisionum are those that were used at the beginning of the seventh century (Henstra 2000, 51–2) and moreover that there are traces of the use of the Byzantine solidus as the unit of account, dating back to the late sixth century. So it would seem that the indigenous material in Lex Frisionum, and especially the tariff material, was already old when it was written down. For Frisia this means a continuity in the period 400–1500 for the textual genre and the legal practice of the compensation tariffs. I recently compared the Old Frisian compensation tariffs with those in Lex Frisionum to see how much of the old material could be found in the Old Frisian traditions (Nijdam 2019). The results were as follows. First, there are two more or less complete tariffs in Lex Frisionum, one in Tit. XXII (De dolg), the other in the Additiones Sapientum, given by Wlemar. Both contain close to ninety paragraphs. The material by Saxmund comprises only a few paragraphs, so this cannot be seen as a complete tariff text (Nijdam 2019). Second, comparing these two complete tariffs with the various Old Frisian traditions, it turns out that the closest match with Tit. XXII were the tariffs stemming from Central Frisia. I had not expected this, because the Old West Frisian (= Central Frisia between the Vlie and Lauwers) tradition is the youngest of the two, with manuscripts no older than the fifteenth century, even though I had already been able to connect the tariffs in this tradition to the thirteenth-century political units in Frisia and had already observed they were relatively archaic in nature. The Wlemar tariff can – with caution – be matched with the tariffs stemming from the eastern region between Lauwers and Weser. These observations also contradict the alleged influence from the Leges Alamannorum on Lex Frisionum in the tariff section. The examples Siems gives (1980, 356–60) show only superficial similarities and if anything attest more to the common Germanic practice of compensating injuries, as analysed most recently by Lisi Oliver (2011). Embodied honour Medieval Frisia was an honour society (Nijdam 2008; 2014). This can also be concluded for Early-medieval Frisia. Having law is having honour. Being allowed to participate on the thing meeting is one of the aspects of being a man of honour. The model in Figure 5.1 was drawn up for Old Frisian law. The means of compensation encountered 145

Z01 Hines Book B.indb 145

07/05/2021 13:30

Frisians of the Early Middle Ages

Fig. 5.1  A diagrammatic model representing the nested circles of the embodied honour of an individual under Old Frisian law, centred upon the person and extended through social and political bodies, with the most important components related to his power and status.

in that corpus are slightly different from those in Lex Frisionum, where only money is mentioned and we find no instances of paying with land, houses etc. This means the idea behind the model stands, but would look different in detail when plotted out for Lex Frisionum. This model of embodied honour means that a man of honour has guardianship (OFris. mund and were, ‘protection, guardianship, responsibility’) over the objects, persons and livestock which fall under his power and which make up the second and third ring of the model. He is liable for all these elements, can be damaged by damaging these elements, but can also be compensated in the form of these elements. This means that a man’s embodied honour consists of all the elements in the three circles. It is interesting to see clearly in Lex Frisionum that both the litus and the servus fell under the mund of their dominus. The servus could do nothing autonomously; he was 146

Z01 Hines Book B.indb 146

07/05/2021 13:30

Law and Political Organization not allowed to appear in court, nor to take oaths etc. The litus could take oaths, but only to a certain extent. His lord was entitled to his wergild if he were killed. In my book on the Old Frisian compensation tariffs (Nijdam 2008), I compared the Old Frisian law texts to bones, or a skeleton. They reveal the structure of the society beautifully, but we miss the juicy meat of the Icelandic sagas for medieval Frisia. In a similar way, Lex Frisionum allows us a few glimpses of the Frisian mead hall but nothing more than that, while for a fuller picture we have to rely on Old Icelandic and Old English sources, the poem Beowulf presenting a perfect example. In Lex Frisionum, we encounter a table maid (bortmagad; Tit. XIII) who apparently had no other chores such as milking or grinding grain but only served drinks. Furthermore a harpist (AS: LT Tit. VI,23) and a lap dog (barmbraccus; Tit. IV,4) – for which the highest compensation of all types of dogs is to be paid – also point to a hall setting and splendour. Actually, no fewer than five types of dog are mentioned in Lex Frisionum: -

1. the lap dog or barmbraccus (4 solidi); 2. those that can kill wolves (3 solidi); 3. those that can wound wolves (2 solidi); 4. those that guard the cattle (1 solidus); 5. that which does nothing but just hangs around in curte aut in villa (in the hall or on the farm?) (1 tremissis).

-

Some of these types of dogs can be matched with the archaeological and historical data available. First, the lap dog or barmbraccus probably was a spaniel-type small hunting dog, not unlike the King Charles spaniel. The term bracco ‘hunting dog’ (cf. Dutch brak, German Bracke) seems to confirm this: spaniels (although the etymology of the word refers to Spain) have been in use as hunting dogs for a long time in Europe. The small types of these hunting dogs were kept as lap dogs, as has been attested for other regions (Kerrigan 2020).11 Skeletons of lap dogs have not been found in Early-medieval graves in Frisia, but those of large dogs (with heights at the withers of c. 60–70 cm), probably of the types meant by categories 2 and 3, i.e. those that could kill or wound wolves, have been found buried besides persons in graves from especially the seventh and eighth centuries (Prummel 1992; 1993; Knol 2012). At the excavation of Oosterbeintum, skeletons of these large dogs have been found that were buried on their own, i.e. not accompanying a human, which is a very rare phenomenon (Knol et al. 1996, 324–7). Just like the lap dogs, these large dogs were also typically owned by members of an elite. They could be used not only to ward off wolves, but also as battle dogs, taken into war. In the Old Frisian corpus, by contrast, we encounter only one type of dog, which seems to belong in either category 4 or 5 in the list in Lex Frisionum – it is described in the context of biting wounds brought about by dogs at the farm (Nijdam 2008, 307–8). The list of various classes of dog in Lex Frisionum is indeed then a strong indicator of a very self-conscious elite.

11 I would like to thank Jorieke Savelkouls for guiding me through the maze of ancient dog breeds.

147

Z01 Hines Book B.indb 147

07/05/2021 13:30

Frisians of the Early Middle Ages Social stratification: estates Lex Frisionum distinguishes four separate groups or estates: nobiles (noblemen), liberi (freemen), liti (half-free men/serfs), and servi (slaves). Of these groups, the slaves had no rights and were treated as property and livestock (Tit. IV). The other groups were comparable with each other in worth, and were able to act as persons at a thing meeting. An interesting group is that of the litus, the serf or half-free man. There are all kinds of paragraphs in Lex Frisionum which attest to the fact that the position of the litus is derived from that of a liber. A litus’ wergild is half that of a liber. Unlike a slave, however, who has almost no rights at all, a litus is allowed to appear before court, to defend his case and to swear an oath of innocence and produce compurgators of the same estate (i.e. also liti). His lord takes over the role of the next of kin and is entitled to the kin’s share of the wergild of his litus (Siems 1980, 296). The litus is unlike the nobilis and liber with respect to having to endure corporal punishment. The first two groups seem to be exempt from this – much like the situation we encounter in Old Frisian law, where it is explicitly stated that Free Frisians can buy off all forms of corporal punishment by paying compensation (Algra 2000, 192–3) – but a litus who has murdered his lord is to be tortured to death (Tit. XX,3). Apparently the difference between a litus and a liber was sometimes hard to tell. Tit. VI and XI deal specifically with the legal status of the litus. If a free woman marries a litus but was unaware of this until he is summoned to appear at the thing meeting because of his estate, she can declare she was unaware of the fact that he was not a freeman and refrained from sleeping with him since she heard. In this way, she herself can remain free as well as their children. In Tit. XI, a number of cases are described in which a liber/litus has to establish his status at the thing meeting, either swearing that he has not entered into the servitude of another person (Tit. XI,1) or that he has bought himself free (Tit. XI,2). A topic of long debate has been the apparent vanishing of the estates in medieval Frisia, under the Frisian Freedom period. It seems as though only freemen were left, called fria Fresa, ‘free Frisian’ in the sources. Only at the end, in the fifteenth century, the ‘hoofdeling’ capitaneus, chieftain (OFris. havedling) appears. In a number of publications, however, Hans Mol and especially Paul Noomen have shown that this was not a new nobility, but that some families can be traced back to at least the twelfth century (Noomen 2009, 28–49). More recently (2012), Noomen has argued that there is a continuity of the estates between those mentioned in Lex Frisionum and the later period. We even find a hapax legomenon letslachta (i.e., litus) in the Riustring Old Frisian law tradition (Noomen 2012, 268). The view is now that the ‘free Frisians’ consisted of both noblemen and freemen, the freemen being freeholders, with their own farm and lands, having access to the thing meeting, etc: in short ‘having law/rights’ as it is put in Old Frisian (Nijdam 2008, 284–7). Last comes the slave. In Tit. I no wergild for a slave is mentioned, and both there (Tit. I,11), as well as in Tit. IV it is said that a slave is to be compensated according to his worth (Tit. IV,1). In Tit. XV however, a wergild – or rather a compensation – is laid down for a slave, giving a ratio 1:2:4:8 from slave to nobleman. Tit. XV is said to apply to the region between Sincfal and Lauwers (i.e., West and Central Frisia). Also, this fixed price does not imply we are talking about a wergild here. Rather, it can be seen as a compensation to be paid to the lord of the killed slave. Although it is not mentioned 148

Z01 Hines Book B.indb 148

07/05/2021 13:30

Law and Political Organization explicitly, we may assume that the family of the slave was not involved and was not allowed to take revenge if the compensation was not paid. Rather, the slave was seen as a piece of property of his lord, and thus it was the lord who had the right to take revenge if compensation was not paid. Siems and others (Siems 1980, 270–1; Timmer 2000, 29) have pointed out that Tit. I, which deals with wergilds for members of the various estates and the numbers of oaths of innocence required if the defendant wished to claim he was innocent, is unique within the corpus of Leges Barbarorum. I have plotted these numbers of oaths in Table 5.1. Notice that twelve is the number of oaths one has to swear in case of a homicide where the victim is of the same estate as the defendant. Note too that whereas the wergild ratio is always (whether the servus is seen as a piece of property and compensated according to his value as in Tit. I or alloted a separate wergild as in Tit. XV) doubled if you go up an estate, this is not matched in the number of oaths. There is a doubling from litus to liber, but then always a 2:3 ratio for the transition from liber to nobilis; or, from a different perspective, there is a doubling (liber) or tripling (nobilis) from the base number of oaths starting at the litus. More on performing oaths below, under ‘Legal procedure’. Table 5.1  Wergild ratios in Lex Frisionum and numbers of oaths of innocence per estate. Estate

Wergild Title I

Ratio Wergild Title I

Wergild Title XV

Ratio Wergild Title XV

Number of oaths if defendant = nobilis

Number of oaths if defendant = liber

Number of oaths if defendant = litus

Nobilis

80 solidi = 240 shillings

4

11 lb

8

12

18

36

Liber

53 solidi, 1 2 tremissis = 160 shillings

5.5 lb

4

8

12

24

Litus

27 solidi, 1 1 tremissis = 80 shillings

2 lb 9 oz

2

4

6

12

Servus

1 lb 4.5 oz 1

Other legal material in Lex Frisionum In this section I shall briefly describe the other topics that are addressed in Lex Frisionum and then draw some conclusions on its overall focus. First, there are a number of paragraphs on women. The common thread in these paragraphs is female chastity and reproduction. We have already encountered the paragraphs concerning a free woman who had unknowingly married a litus. Tit. IX contains thirteen paragraphs on all types of adultery.12 A special case is the bortmagad, the table maid (Tit. XIII). If someone has 12 Paragraphs 13–17 deal with robbery: one of the examples of the state in which Lex Frisionum has reached us.

149

Z01 Hines Book B.indb 149

07/05/2021 13:30

Frisians of the Early Middle Ages intercourse with her, he has to pay 12 solidi, which is three times the compensation for having sex with a normal slave (Tit. IX,3). Second, a number of paragraphs deal with the domain of possession: theft (III), loans (AS Tit. X), pawns (AS Tit. VIII), robbery/taking something with force (Tit. IX,13–17). Since there is not much that is out of the ordinary here, I will content myself with a number of observations on damage done to cattle, especially horses. Tit. IV begins with compensation for killing a slave and continues with compensations for the rest of a man’s livestock. Besides all other livestock, liability for a horse is separately addressed (AS Tit. X,2).13 If a horse is lent to someone, and during that period the horse injures or kills the person who has borrowed the horse, no compensation needs to be paid by the owner. In other words, the owner is not liable; the horse has temporarily gone over to the mund of the borrower. In Old Frisian law, liability for livestock is treated in a little more detail, but as Old Frisian law as a whole is much more extensive than Lex Frisionum that need not surprise us (Nijdam 2008, 307–9). Third, we turn to stipulations on various types of violence. I shall first address the types that might or might not refer to a feud setting, and then address the clauses that explicitly deal with feuding. Close to each other (AS Tit. III and VI) are two clauses on blocking the road. The first of these is very simple and easily recognizable as what is called weiwendene in Old Frisian law: stopping someone on the road and/or robbing him of some of his possessions (Nijdam 2008, 156). The second clause is not entirely clear (AS Tit. VI): ‘if someone closes off the public road in a river’. This seems to refer to a ford, a place where a river can be crossed, but what exactly is meant here? Is it a form of one-off obstruction, or has the perpetrator systematically closed the ford? Several instances of taking hostages are mentioned (Tit. XX, XXI). We know from the later period that this was common practice during a feud: people were taken hostage and released if ransom had been paid. It is not entirely clear if this is meant here. Especially not in Tit. XXII, where it is said that if someone (nobleman or freeman) sells someone (nobleman or freeman) abroad, he has to pay his wergild, i.e. this is treated the same as if he had murdered the victim. Or the perpetrator has to do his best to buy the victim free again. Once returned, the victim can claim twice what the perpetrator fetched for the victim as compensation. This seems to refer to a context of slave trading. Finally, Tit. I of the Additiones Sapientum explicitly addresses feuding. A peace is decreed by the king for persons being involved in a feud to be safe in church, in their house and at the thing. AS Tit. I,3 decrees that within a feud situation, a death penalty is put on stealing a horse or a cow, or breaking down a screona (weaver’s hut), or the defendant has to pay his wergild to buy off the death sentence. This seems to be one of the cases in which the influence of the Frankish kings and new legislation is visible, as was addressed earlier. In concluding this section, then, we can say something about the scope and focus of Lex Frisionum. The keywords for the entire law are: honour, violence, compensation and liability, i.e. mund, which is an important aspect of someone’s honour. The law was clearly written for the social groups which dominated society: the nobility and the freemen. Paragraphs on important matters such as inheritance law and on property of land are absent, but these are lacking in most Leges Barbarorum as well. 13 For the history of the Frisian horse, see Savelkouls 2015; 2016.

150

Z01 Hines Book B.indb 150

07/05/2021 13:30

Law and Political Organization

The assembly or thing All Germanic peoples gathered at thing meetings at certain times during the year (Green 1998, 35–9; Iversen 2013); so did the Frisians. The term thing has been translated in a number of ways: ‘court’, ‘court meeting’, ‘assembly’ (Vogt and Nijdam 2018, 38–53), but in various contexts it could also mean ‘time’, ‘place where an assembly is held’, ‘judgment’, ‘agreement’ (Green 1998, 36). The etymology of the word can be retraced to ‘time appointed for a particular purpose’ (Green 1998, 35). The thing fulfilled several purposes at the same time: it was a meeting which was held at specific times during a year, during which all kinds of matters concerning the people in a certain district (gau or pagus) were discussed. These included legal, military, political and religious affairs. This mixture of functions is reflected in an early votive stone found near Hadrian’s Wall dedicated to Mars Thingsus, i.e. ‘Tiwaz of the thing’ by a number of Tuihanti, the name referring to the region of Twente in the Netherlands. These Tuihanti have been interpreted as Frisians (Green 1998, 34; Iversen 2013, 10). It is generally agreed upon that Mars is the interpretatio Germanica for the Germanic god Tiwaz, OIcel. Týr, the god of law. Where a number of Germanic languages use the name referring to the god, English Tuesday and Frisian Tiisdei, for the second day of the week, Dutch and German have dinsdag and Dienstag referring to the thing. Hence, the inscription Mars thingsus fits well in the context of other pieces of information. At the thing, law and politics met. In that sense the thing is the linking pin between law and political organization (including the problem of Frisian kingship). In turn, a thing was linked to a certain district or pagus. In his Vita Bonifatii, Willibald observed that the Frisians saw themselves as one people, even though they lived in a large number of very diverse pagi which are separated from each other by rivers (Siems 1980, 177). For Frisia the oldest pagi or districts are now firmly established, as can be seen on Figure 4.1 in the contribution by Gilles de Langen and Hans Mol in this volume. The identification of these pagi is affirmed by various types of sources. Most of the pagi have rivers as natural boundaries, making their identification even firmer. Moreover, the pagi in the regions that became part of the Frisian Freedom (between Vlie and Weser) remained in place as political units. It was only in the course of the Middle Ages that the sub-districts became more and more important, but even then meetings were held at the level of the pagus. If we assume that the Frisian pagi were the oldest building blocks we can discern, then how are we to interpret the threefold division of Frisia in Lex Frisionum, each part consisting of a number of pagi? Even though the pagi are identifiable as administrative units, this does not solve all the problems, because there has been much debate among legal historians about how many administrative levels a pagus had (Iversen 2013, 6–10). Usually, three levels are proposed, of which the pagus is the middle. The top level has been named the civitas. In the case of the situation described in Lex Frisionum this level would be represented by West, Central and East Frisia, each block consisting of several pagi. Below the level of the pagus, there was the centena or hundred. For (Central) Frisia a toponym Cammingehunderi is attested in a charter issued by Louis the Pious in 839, but there has been a lot of debate about the exact meaning of this term (Schuur 1976). In short, we simply lack source material to obtain a firmer picture. We know from the Old Frisian legal codes that there were three normal thing meetings a year which each Free Frisian was obliged to attend (Algra 2000, 78, 203). But 151

Z01 Hines Book B.indb 151

07/05/2021 13:30

Frisians of the Early Middle Ages the actual thing locations have not been firmly established for medieval Frisia. Hans Mol has identified the locations of the late medieval gallows in Central Frisia (Mol 2006). His analysis however reflects a situation in which the original pagi Oostergo and Westergo had split up into a large number of sub-districts with corresponding thing locations (Nijdam 2008: 91; Vries 1986: 68). But what were the original thing locations for Westergo or Oostergo? The latest attempt at an overview for all Frisian high medieval pagi has been carried out by Hajo van Lengen in 2003 but he has identified locations where historical sources say people gathered. These need not have been the original thing sites (van Lengen 2003). Menno Dijkstra was able to combine historical and archaeological research for the South-Holland region between the rivers Rhine and Maas to identify a number of thing locations for those areas (Dijkstra 2011, 298–300). Returning to the question of the original thing locations of Westergo and Oostergo, these were most likely Franeker and Dokkum respectively (Nicolay 2014, 41; Noomen 2009, 81). Another possibility to use archaeology to locate thing sites is to look for a concentration of pit houses, serving as temporary dwellings for those attending the thing (Nørgård Jørgensen, Jørgensen and Gebauer Thomsen 2011). In a Frisian context, these pit houses have all been interpreted as weaving huts, probably partly inspired by the explicit mention of a screona ‘weaving hut’ in Lex Frisionum (AS Tit. I,3). The last major question to be addressed here concerning thing sites and levels of administration is whether there was a pan-Frisian assembly prior to the Carolingian conquest, comparable to the Icelandic Althing and the general assembly the Saxons were said to hold annually at Marklô before their conquest by Charlemagne (Anderson 1999, 28–34; Flierman, this vol., Discussion). One immediately thinks of the Upstalsbam assembly near Aurich in East Frisia, which for a certain time in the period of the Frisian Freedom did function as a pan-Frisian assembly for the Frisians between Vlie and Weser (Meijering 1974). The problem with the Upstalsbam as a candidate for the ancient pan-Frisian thing site is that the location is wrong in two respects. First, it is not where we would expect it, namely in Central Frisia. Second, the Aurich region is not ancient, but was settled only after the colonization of the peat bog lands.14 Also, there are indications that Upstalsbam started as a central assembly for the Frisians between the Lauwers and Weser and that the Frisians between the Vlie and Lauwers joined later (Nijdam 2008, 110). With some effort the Upstalsbam as an institution, i.e. an overarching consultative body or union, can be traced back to the eleventh century, even though the earliest specific reference is from 1216 in the chronicle of the monastery of Wittewierum (Ehbrecht 2003, 134–40). In the fourteenth century, it was these Central Frisians who tried to bring the languishing Upstalsbam confederation back to life, probably because of a major threat from the counts of Holland and Zeeland who tried to conquer Frisia and they needed all the help they could get. The Old Frisian Statutes of the Upstalsbam from 1323 were thus probably drafted in Central Frisia (Fig. 5.2; Meijering 1974). So, to conclude this section, it seems safe to assume that the Frisians gathered three times a year at the thing meetings of their pagus. The level above that, what has been called the civitas in the literature, is probably reflected in the East–Central–West 14 Information from Hans Mol and Gilles de Langen.

152

Z01 Hines Book B.indb 152

07/05/2021 13:30

Law and Political Organization

Fig. 5.2  The Totius Frisiae (‘of all Frisia’) seal of 1323. New impression from the original seal kept by the Groninger Museum. Groninger Museum (Photograph: Marten de Leeuw).

division found in Lex Frisionum. And overarching those three levels was an all-embracing notion of all being Frisians, a notion we encounter in later medieval sources as well. But how these levels functioned eludes us because of the lack of sufficient source material. Legal procedure at the thing Firmer ground is reached once we turn to the topic of legal procedure. Not only does Lex Frisionum yield a firm quantity of information, but much of it is supported and supplemented by the later Old Frisian law material. The sapientes Wlemar and Saxmund have often been compared to the legal expert in the most archaic material in Old Frisian law, the asega (a compound of ā ‘law’ and sega ‘speak’: ‘lawspeaker’) (Algra 2000, 99–100). There has been some debate on what exactly the function of the asega was: whether he would actively recite the law once 153

Z01 Hines Book B.indb 153

07/05/2021 13:30

Frisians of the Early Middle Ages every year just as the Icelandic lögsögumaðr did, or whether he knew the appropriate rules for specific cases and would give them when asked to do so (Gerbenzon 1973). This latter variant is what we actually find in the Old Frisian legal texts. In Lex Frisionum (Tit. IV,3) a iudex ‘judge’ is mentioned once, who seems to perform the same task an asega did in Old Frisian law: Aut si negaverit, iuxta quod iudex dictaverit, iuret ‘or if he [the defendant who is accused of slaying someone’s cattle] denies this, he is to swear according to what the judge dictates to him’. But even though the sapientes are not called iudex explicitly, they do ‘dictate’: Haec iudicia Saxmundus dictavit ‘Saxmund dictated these decrees’ (AS Tit. III,59). In other words, the sapientes and the iudex in Lex Frisionum as well as the asega were all legal experts who knew the rules of the law and of legal procedure: which oaths to take, which formulas to utter, etc. The tariffs too attest to the fact that there was a body of specialized legal knowledge, including technical legal terms for types of injury and sets of rules on how to claim compensations before court (Munske 1973). Lex Frisionum shows that the legal procedure we know from Old Frisian law and from other Germanic people was in place in Frisia too. A claimant brought a case before court, and the defendant had the right to deny the charges and to swear inncocence. If more oaths were required, the defendant had to find compurgators to take an oath in his favour. They would swear under oath that the defendant was a man of honour who always spoke the truth. Their function was thus that of a character witness rather than to bear witness of what had actually happened concerning the case at hand. This also underpins how important honour was, and to what extent it was the basis of many aspects of this society, in this case the law and legal procedure. A number of different oaths are mentioned. The most common oath was probably that on the holy relics (in reliquiis sanctorum; Tit. XII,1 etc.), which we also find in Old Frisian law (OFris. wīthēth) and which of course must have been post-conversion and probably replaced an older custom (plausibly by swearing on an oath ring). A second type of oath is that on clothing (in vestimento; Tit. XII,1 etc), which meant the person took his own garment in his hands and pledged upon those clothes. This form is also found in Old Frisian law (Nijdam 2008: 179). A third type is that called in pecunia ‘on money’ (Tit. XII,1 etc). This probably reflects the Old Frisian fiāēth, which meant one swore on his own possessions (Popkema 2007). If a matter could not be resolved, a solution could be found in a so-called ordeal, in which the sacred powers (‘God’) would decide a case. Two types are mentioned in Lex Frisionum: the ordeal by hot water (III,6–9; XIIII,3) and the judicial duel. Again these ordeals are also found in Old Frisian law (and are widely attested throughout medieval Europe and beyond). The ordeals slowly died out after the Fourth Lateran Council in 1215 (Buma 1949).

Kingship Finally, the enigma of Frisian kingship needs to be addressed in the context of law and political organization. What was the relationship between king and thing? What did kingship entail in Frisia until c. 750? It will quickly become clear that even the king we learn most about in the sources, Radbod, remains elusive. Firstly, sources about Frisian kings are very scarce. There are three possible and two firmly attested Frisian kings. The first one is the mythical king Finn whom we know 154

Z01 Hines Book B.indb 154

07/05/2021 13:30

Law and Political Organization about thanks to the Old English poems Beowulf, the Finnsburh fragment (Klaeber 1950) and Widsith. Here, a Frisian king, Finn Folcwalding or Folcwaldan sunu ‘son of Folc­ walda’, is mentioned, whose sister was married to a Danish prince Hnæf. It has been observed by various scholars that the proper name Folcwalda, which literally means ‘ruler of the people’, seems more like a title than a proper name, more so since it has also been attested as one of the epithets of the god Freyr (Laur 1996). If Finn was a historical figure, he would have lived somewhere in the fifth century, it is believed. Chronologically, the next possible Frisian king is Audulfus, whose name can be found on a number of Frisian gold tremisses from the late sixth century, although some scholars believe this to be the name of the moneyer (Nicolay 2014: 61 note 29). Next come the firmly attested kings, Aldgisl and Radbod. Aldgisl is mentioned only in the Vita Wilfridi. Bishop Wilfrid of York spent the winter of 678/679 at the court of Aldgisl where he received a warm welcome. The king was not interested in being converted to Christianity, however, but he refused to extradite Wilfrid to the Frankish maior domus Ebroin after receiving a letter with this demand. Apparently Aldgisl was powerful enough to be able to ignore this request. Radbod is encountered in sources from c. 688 onwards. He is the most famous Frisian king of all times, and thanks to his reception as a character in the chansons de geste (the cycle on Charlemagne) he remained so throughout the Middle Ages (Nijdam and Knottnerus 2019). By analysing Radbod, we can get some idea of Frisian kingship. After Radbod had died in 719, we encounter a dux Bubo, who was defeated by Charles Martel in 734 (Wood, this vol.; De Langen and Mol, this vol.). Again, there is too little information on Bubo: where he resided, over which territory he ruled, what his relationship to Radbod was. Secondly, let us take a look at the way kingship developed in the surrounding areas, especially Anglo-Saxon England and Scandinavia. And of these two, Anglo-Saxon England in particular, for both Frisia and Anglo-Saxon England came into being as a consequence of the same migration wave by roughly the same people. The distinction Tacitus makes between two types of leader, i.e. duces and reges, is reflected in the Germanic terminologies that have been handed down (Green 1998, 121–40). That means there were war leaders (duces) on the one hand and ‘kings’ (reges) with a religious function on the other. Tacitus also says that the people who were eligible for kingship came from noble families (ex nobilitate), whereas the war leaders were chosen on the basis of their martial courage (ex virtute). When we look at the historical sources, though, it has proved to be very difficult to tell the difference between the two types of leaders for the various Germanic peoples of the Early Middle Ages. The oldest word for king seems to have been *þiudans ‘leader of the people’ being followed by druhtin ‘army leader’ (comparable to dux) and then lastly by *kuning-, each word being tricky to interpret (Green 1998, 121–40; Seebold and Schneider 2001). The most probable scenario is that kingship was strongly influenced by the aspect of military leadership as a consequence of the upheavals of the Migration Period and the contacts there had been with the Roman Empire, especially with the Roman army. Germanic kingship developed more or less simultaneously in the fifth century throughout Europe: from the Merovingians on the continent all the way up to Scandinavia (Wolfram 2006). In the course of the seventh century, Scandinavia was no longer in line with the European developments and fell back, whilst the Continental and Insular kingdoms flourished (Skre 2020; Nicolay 2014). 155

Z01 Hines Book B.indb 155

07/05/2021 13:30

Frisians of the Early Middle Ages The way kingship must have developed in the first centuries CE is succinctly described by Dagfinn Skre in a recent article on rulership in Scandinavia in the first millennium (Skre 2020, 197): Kings were members of royal lineages that were associated with lands and peoples, but their authority and polity type varied. The diverse meanings of the term konungr suggests that no uniform idea of kingship existed among Germanic peoples at the time, and that the emergence through late fifth–ninth-century Europe of widespread monarchy was not a linear and uniform development. Under shifting conditions, kings as other types of rulers before them, will have navigated between personal ambitions, acute constraints and opportunities, their polity’s legal tradition, interests among the aristocracy, popular consensus expressed at assemblies, possible rivals within royal lineages, and the like; thus constantly modelling and remodelling the institution of kingship.

The meandering ways in which various kingships developed in Northwest Europe between c. 500 and 1200 is further described by Yorke (1990) for Anglo-Saxon England and Hybel (2017) for Denmark. These accounts indirectly explain some of the uncertainties we encounter in the Frisian case, the understanding of which is further hampered by the scarceness of the sources. Archaeology can help only to a certain extent here, for although Johan Nicolay has brought research a big step further by his analysis of the gold and silver objects from Frisia and their implications for Frisian kingship, no ruler’s site has been found. The important excavations of Wijnaldum, where the famous brooch was found, has not yielded a king’s grave comparable to Sutton Hoo (Nicolay 2014, 346–66). Let us therefore, thirdly, turn to an investigation of Radbod’s kingship. In their contribution to this volume, Gilles de Langen and Hans Mol have already described much of Radbod’s activities during his lifetime. They moreover identify the territorial base of Radbod in West Frisia – in the regions of Kennemerland and the districts of Texel and Wieringen. John Hines (this vol.) points to a number of instances where Radbod is involved in religious affairs of the Frisians. Also, he observes that Radbod probably was forced by internal politics to adhere to the non-Christian religion of the Frisians. Suffice it to say that keeping in mind the rex/dux distinction that Tacitus makes, Radbod is clearly not only a military leader, but plays a significant role in religious affairs as well. The Vita Vulframni describes two cases of human sacrifice (chs. 6 and 8) where Radbod oversees the ritual. In the Vita Willibrordi, Willibrord is brought before Radbod for having broken the religious rules concerning the sacred well and cattle of the god Fosite. In all three instances, the casting of lots is mentioned. This casting of lots was described in detail in Lex Frisionum as well. The religious and the legal sphere meet in ch. 6 of the Vita Vulframni in yet another respect. When the boy Ovo – who after having been saved by Wulfram was later to be trained in the monastery of Fontenelle – had been chosen by lot casting to be sacrificed on the gallows, Wulfram implored Radbod to let the boy go. Radbod replied, however, that that was impossible: Respondebat autem dux patrio sermone, decretum esse lege perhenni olim a praedecessoribus suis omnique Fresionum gente, ut, quemcumque sors elegisset, in eorum sollemniis diis offerendum sine mora.

156

Z01 Hines Book B.indb 156

07/05/2021 13:30

Law and Political Organization But the duke answered in his native language that it was traditionally laid down in the unchanging law by his predecessors and by the entire Frisian people that anyone who had been chosen by fate had to be sacrificed to the gods in their ceremonies without delay.

It is interesting to observe that religion is described as law, as custom. Also, once again in this quotation, we see Radbod declaring how much he is bound by the laws and customs of his people and predecessors (he also explicitly referred to these two groups when he refused baptism, later in the same Vita). This brings us to the fourth point I want to consider: was Radbod a ‘Freswalda’, a ruler of all Frisians? There are a few things to consider here. It has been observed that in most sources, Radbod is active in Frisia citerior or West Frisia. If De Langen and Mol are correct, then his power base is indeed to be located there. But there is the famous episode in the Vita Willibrordi, in which the island ‘Fositeland’, which has been identified as Helgoland, apparently falls under Radbod’s jurisdiction. Moreover, no other Frisian king is mentioned in the sources. Radbod apparently is the only Frisian king during the period 680/688–719. Also, as already mentioned, Radbod had been such a formidable antagonist to the Carolingians that he became a character in the chansons de geste. Finally, there is the later Vita Willibrordi, written by Thiofrid of Echternach c. 1100. In ch. 11 it is stated that Radbod did not want the Frisian laws to be infringed upon because he had been the first to collect them and have them laid down.15 If this were to be true (and in view of the late date of the source, this is debatable), then Radbod would be comparable to Æthelberht of Kent, who also was the first Kentish king to lay down the law of his people (Oliver 2002; Wormald 1999, 93–101). Radbod certainly seems to have had the stature and might have felt the need in the light of his contacts with the Franks and his Anglo-Saxon neighbours: laying down the laws certainly was in fashion around 700. All this evidence makes it likely that Radbod either inherited the title of ‘Freswalda’ or developed into one himself (Nicolay 2014, 22 for the historiography of this notion). This question needs to be investigated further, in combination with the last point I want to consider here. For fifthly and finally we need to bring law, thing and king together. In the recent literature on Early-medieval kingship, much attention has been given to the relationship between the king and his retainers. Johan Nicolay’s important recent study on Frisian Early-medieval elite networks is entirely based on a hierarchical model where the king forms the pinnacle of social relationships (and even society) (Nicolay 2014, 2–19). But how about the relationship between the king and the assembly of free men, the thing? In this respect it is important to observe that the thing did and could function without a king. Indeed, most information, whether from Tacitus or from the already mentioned Icelandic Althing and the thing of the Saxons, shows that those assemblies functioned autonomously, without a king as primary mover of the institution (Anderson 1999, 28–34). A priest led the thing meeting, as Tacitus tells us, and this is in line with the Icelandic goðar, who seem to have been priests primarily. This observation explains 15 Noluit a quoquam deprehendi leges a se infringi, quas ipse primus primum ex gentilitate legislatorem secutus Lacedaemonium Ligurgum, suffectum fratri suo Polibitae, Spartanorum regi, invenit ac instituit Fresonum genti ‘He did not want the laws to be infringed upon by anyone, which he himself first thought out and instituted for the Frisian people, following in the footsteps of the first legislator of pagan peoples Licurgus of Lacedaemon, who had succeeded his brother Polybitas, the king of the Spartans.’

157

Z01 Hines Book B.indb 157

07/05/2021 13:30

Frisians of the Early Middle Ages how both the Frisian Freedom and the Icelandic Commonwealth could function very well without a king. Also, for Frisia, the structure of the pagi and the three Frisian areas (West, Central, East) were in place since the Early Middle Ages and for the most part were stable unities throughout the Middle Ages even though Frisian kings, Frankish kings and Frankish counts came and went. This must then imply that the thing assemblies were very persistent factors that should not be overlooked. As Frode Iversen (2020, 251) rightly observes for early medieval Scandinavia: The thing was instrumental to the acceptance of new kings and the regulation of political and economic relations between the king and the people. The thing meetings were cyclical and contributed to shaping and maintaining identities on various geographical levels. The meeting was held at given times and places and created social bonds beyond its formal legal role. One imperative of the ethnogenesis theory, as both Reinhard Wenskus (1961) and Herwig Wolfram (1970, 1990) argue is that tribal identity is formed by origin myths told among the tribe’s leading strata (for further discussions, see also Pohl 1994; Goffart 2006; Heather 2009). This chapter holds that geobound identity is also formed through human meetings at the thing, in particular among an upper and a mid-stratum of landowners meeting regularly, while upper elite identity also was rooted in supra-regional mobility and a wider economic and political network.

Iversen thus rightly observes that in the recent literature on kingship and Gefolgschaft, the thing community and the law have been somewhat overlooked. Fortunately, the problem has been addressed, both more recently (Weitzler 2006) and by legal historians in the past (Krause 1971). Jürgen Weitzler (2006, 361) explicitly states that the thing community hindered kings becoming absolute rulers too fast: Das dinggenossenschaftliche Denken war ein Schutzwall gegen die schnelle Rezeption spätantiker Herrschafts- und Rechtsvorstellungen. … Spätantike Maximen wie princeps legibus solutus und quod principi placuit, legis habet vigorem konnten in Mittel- und Nordeuropa nur allmählich und erst dann eindringen, als die Rezeption römischer und stärker noch kanonistischer Vorstellungen von einem hierarchisch gestuften, selbsturteilenden Einzelrichtertum dafür den Boden bereitet hatte.

In short, then, the Frisian king (if indeed there was only one Frisian king at one time) must have negotiated at the thing with the communities of the various pagi, which he probably visited at certain times (almost certainly by ship). The relationship between the Frisian kings and the thing has remained largely understudied through a combination of factors, one of these of course being that the sources offer almost no information.

Conclusion In conclusion, we are left with mixed feelings after having looked at Early-medieval Frisian law, thing and king. The legal matter we encounter in Lex Frisionum is very recognizable for someone versed in Old Frisian law. There are differences, to be sure, but overall the material in Lex Frisionum is perfectly explicable as an earlier phase of what can be found in Old Frisian law. This points to the fact that the Frisians by and large have been able to uphold their own system of law. The Viking Period may have 158

Z01 Hines Book B.indb 158

07/05/2021 13:30

Law and Political Organization been crucial in that the Frankish grip on the Frisians never became tight enough to make far reaching changes. As a result, Frisian law evolved on its own, by internal dynamics. The legal aspects of the thing are not difficult to grasp either. The thing was a Germanic institution so there is ample comparative material. The legal procedure in Lex Frisionum matches that of the most archaic phase of Old Frisian law, where one man of honour brought a claim to court against another man of honour and the defendant could exonerate himself by swearing an oath of innocence. In more serious cases, he needed to acquire compurgators to swear on his behalf. There are legal experts at work in Lex Frisionum, and these resemble the Old Frisian asega to a great extent. Things get exponentially more unclear once we get to the political side of things. What was the relationship between the pagi as Frisian administrative nuclei and the overarching regions and even to Frisia as a whole? And what was the role of the Frisian king? Could Radbod really muster all the armed forces in all the Frisian lands? How did he go about that? The combination of king and law needs to be researched further for Early-medieval Frisia. It is where legal history, anthropology, history and archaeology should meet to shed more light on these questions.

Appendix: An overview of the contents of the Lex Frisionum (Number of paragraphs per Titulus in brackets) Main text Title I Title II Title III Title IV Title V Title VI Title VII Title VIII Title IX Title X Title XI Title XII

De homicidiis Homicide (21) Forresni Instigation (10) Thiubda Theft (9) De servo aut iumento alieno occiso Compensations for killing someone’s slave or animal (9) De hominibus qui sine compositione occidi possunt People who can be killed without compensation (2) De coniugiis ignoratisi Free woman unknowingly marrying a litus (2) De brand Arson (2) De notnumfti Robbery (2) De farlegani Adultery (17) De testibus Witnesses / perjury (1) De lito Serfs (3) De delicto servorum Crimes by slaves (2) 159

Z01 Hines Book B.indb 159

07/05/2021 13:30

Frisians of the Early Middle Ages Title XIII Title XIV Title XV Title XVI Title XVII Title XVIII Title XIX Title XX Title XXI Title XXII

De stupro ancillarum Fornication with female slaves (1) De homine in turba occiso Homicide in a crowd (7) De compositionibus wergildi Wergilds (4) De fredo Penalty for breaking the peace (1) Hic bannus est Royal bans (5) De die dominico Sunday rest (1) De parricidiis Parricide and fratricide (2) De mordrito Homicide (3) De plagio Kidnapping (1) De dolg On wounds (89)

Additiones Sapientum Title I Title II Title III Title IIIIb Title IV Title V Title VI Title VII Title VIII Title IX Title X Title XI

De pace faidosi Feuding (3) Compositio vulnerum Wounds (10) Hoc totum in triplo componatur Wounds (78) De eo qui alteri viam contradixerit Obstructing the road (1) De eo qui alterum de caballo iactaverit Throwing someone off his horse (1) De muliere occisa Killing a woman (1) De flumine obtruso Closing off a ford in a river (1) De rebus fugitivis Harbouring / concealing runaway slaves or cattle (1) De pignoribus Collateral (1) De compositione Compensations (1) De re praestita Lending an object (2) De honore templum Desecrating a pagan temple (1) 160

Z01 Hines Book B.indb 160

07/05/2021 13:30

Law and Political Organization Lex Thuringorum Title VI

Haec iuditia Wlemarus dictavit Wounds (11)

Discussion IJSSENNAGGER-VAN DER PLUIJM   You mentioned sacral kingship, and it has been mentioned that there appear to be differences between Anglo-Saxon and Frankish perspectives (cf. Wood, Ch. 7, this vol.). But there could also be a difference as the Frankish sources are political, they are annals, whereas the AngloSaxon sources are saints’ lives. If they also embody those different perspectives, could this relate to your point on sacral kingship? There is a difference in text types – a political and a hagiographical tradition – Ricky Broome has written about that (2014). FLIERMAN   To back up Nelleke’s point: it has been argued that the terminology used by Frankish sources to describe external leaders varied according to the aim of the author. Saints’ lives were all about missionaries converting kings, whereas the Frankish annals typically referred to external leaders as duces in order to suggest a subservient status to the Frankish kings. NIJDAM   At this point it is interesting to look at the Germanic equivalents to dux and rex. Gmc *kuning (rex) is considered more sacral than *druhtin (dux). This brings me to a question to Ian and John, who have been present at many of the symposia in this series. I saw that Professor Dennis Green attended a number of these symposia. I have read his book, Language and History in the Early Germanic World (Green 1998) and I was wondering how the philological perspective he brought in was perceived by the historians. Was there agreement or discussion? HINES   There was always a lot of discussion, certainly; often leading to consensus on a lot of points, while in other cases there was agreement up to a certain point and then some clearly recognized differences of view or judgement as well. The philological dimension was very useful; it did not explain everything but could control some more speculative and the freer interpretative approaches by standing very firmly on matters of what we actually know words meant and what their actual derivation was. WOOD   There are two key books by Dennis Green – The Carolingian Lord (1965) and Language and History. The analysis and interpretation varies a bit between them in terms of putting things into a historical context. It would be rare to disagree with him on a point of language, although the relevance of alternative contexts could be put to him. NIJDAM   What I gathered from reading the discussions in the previous volumes was that he brought into the discussions the dimension of a Germanic worldview; a bit static, maybe, but useful. HINES   He was very interested in seeing how words came to mean different things and to have different uses in different linguistic contexts. He did tend to think in what we would probably now see as rather traditional Stammbaum terms. But it was the branches and the leaves on the branches that were interesting to him in

161

Z01 Hines Book B.indb 161

07/05/2021 13:30

Frisians of the Early Middle Ages the end, rather than the trunk and the roots; however, the latter helped explain the former. To follow up on the word for king, *kuning-: this is a very consistent term across the Germanic-language populations. That is noteworthy in light of sometimes rather Utopian but not completely baseless ideas of there having been an early, almost primeval, egalitarianism in Anglo-Saxon society which in fact is quite similar to the Free Frisian ideal. That is hard to reconcile with the idea that the Anglo-Saxon cultural domination of southern Britain was a matter of elite social dominance. The term cyning must have been there all the time in pre-Old English and Early Old English, so that cyningas existed all the time. It appears to me that this situation would fit with the Tacitean concept of the rex as a relatively ceremonial primus inter pares. In terms of your interest in matters of the social body and the physical body, the lexical connexion of this term with the root cynn (in Old English), ‘kindred’, must mean that those very early kings somehow embodied the social identity of a group. I am cautious of using the term ‘sacral’ – or would want to know exactly what it is supposed to mean – but the idea that there is a mystical and intangible significance that is almost magical is a very interesting one. Altogether this may help us to understand the position of kings in what is otherwise a relatively flat society in hierarchical terms. NIJDAM   Indeed, this model of Germanic *kuning- and its implications allows for scalarity. Mysticism and magic makes it both intangible and real at the same time. FLIERMAN   The Saxon evidence concerning assemblies is interesting and controversial but may be useful to pinpoint certain organizational features of the trans-Rhine societies. We have Bede (Historia Ecclesiastica, V.10), who states that the Saxons near the Rhine don’t have a king, but many independent satrapes (an Old Testament term). These satraps are said to convene in times of war to select a warleader. They did so by casting lots. Then there is the tenth-century Vita Lebuini antiqua, which offers the same story based on Bede, but more elaborate: the Saxons are said to convene at a place called Marklô, near the Weser. Each satrap brings 10 nobles, 10 freemen, and 10 liti to aid him in the decision-making process, reflecting the social layers of Saxon society. Now, historians working on pre-conquest Saxon society tread carefully when using this material: both stories served to glorify the evangelizing efforts of AngloSaxon missionaries and there are several literary layers that need to be peeled away. Nevertheless, I would say that the notion of assemblies as an organizational feature of pre-conquest Saxon society is very persistent; it also suits the historiographical and legal material produced during the Saxon Wars (e.g. Capitulatio de partibus Saxoniae, ch. 34), which portrays assemblies as a key platform of Franco-Saxon negotiation, suggesting that this was a well-entrenched political mechanism for the Saxons. NIJDAM   Part of me thinks in Stammbaum terms about thing assemblies. But such assemblies are universal phenomena too. LOOIJENGA   The romantic, nineteenth-century idea of a central place where Frisians came together has been very influential. Did the Frisian Laws contain traces of Roman Law? 162

Z01 Hines Book B.indb 162

07/05/2021 13:30

Law and Political Organization NIJDAM   That’s an interesting point. If we go back to the historiography of the Historical Archaeoethnology series, the volume on the Scandinavians has Brink discussing the situation in Scandinavia, and the dominant view of sources in the Ancient World (Brink 2002). There is a lot of literature on Roman influences in the Leges Barbarorum. Patrick Wormald (2003) had identified the compensation tariffs for injury as the one potentially truly Germanic component. Most of what can be seen in Lex Frisionum is indigenous, albeit with Frankish influence. VERSLOOT   Back to the kings again! On the one hand we have a sort of democratic society, which at times of need appoints a practical dux; this doesn’t seem to fit with the idea of a sacral king, which has something to do with the ‘kin’. Do these go together? NIJDAM   I am struggling with this; hence the model of honour as having a sacral aspect where you can end up with sacral kingship in the highest spectrum. If we go back to Radbod, you can say he is a war-leader. At the same time, in the saints’ lives, he is doing sacral things: punishing those who desecrate Fositeland; presiding over sacrifice in Vita Vulframni (see Hines, this vol.). VERSLOOT   Another remark relating to an article I recently read from a Slavic perspective, concerning the ninth century (Lübke and Urbańczyk 2009). This showed very similar political organizations in the Slavic and Saxon territories before the Carolingian conquest. It seems that political organization was less a matter of ethnicity and more one of cultural difference, with a non-centralized form of organization. This was an interesting way of looking at things; the pattern is thus not really ‘Germanic’ in itself. WOOD   I want to come back to the issue of Roman Law. The problem is not the Leges Barbarorum but ‘Roman Law’: no one knows what the latter was. We only have the Theodosian Code to refer to; we don’t know what the provincial laws were; we don’t know how judgments were made; oath-taking existed, but does not appear in any of the codes. Wergild has been associated with Latin ‘viri gildus’ as a Roman concept. Compensation is a Roman practice but does not appear in Codex Theodosianus. Patrick Wormald did not go as far as many would now do, but did lead in hacking away at this idea of ‘what is Roman Law?’ (Wormald 1977). We can use the example of the eighth-century Farmers’ Law: was that Visigothic, or Roman provincial law? NIJDAM   I thought that source was very problematic. WOOD   It has now been placed pretty clearly in the context of early eighth-century Byzantine legislation (Humphreys 2017). NIJDAM   I’ve talked about human universals before: taking revenge is a universal, but accepting compensation in place of vengeance is not. The alternative is longlasting blood feuds. WOOD   The key point is the problem of whether or not we can identify something as ‘non-Roman’. NIJDAM   Yes, the Romans knew about accepting compensation. A contributor to a conference on wergild held in Berlin in 2014 quoted the story of a Roman nobleman who walked through town slapping everyone he disliked, with a slave following in his footsteps who would immediately pay the fine. A Germanic society with a strong sense of honour makes a real difference in that situation: they would not accept the fine because of the honour and shame that are involved. 163

Z01 Hines Book B.indb 163

07/05/2021 13:30

Frisians of the Early Middle Ages WOOD   Roman society was very highly stratified. NIJDAM   I have not read anything on the influence of Roman Law on the Leges Barbarorum that is not highly problematic. Lisi Oliver’s contribution The Body Legal in Barbarian Law (2011) was very important. WOOD   We don’t have the tariffs for the Roman provincial laws. Lex Salica is by far the least ‘Roman’ code, but the Burgundian Code has in fact got a known Roman model, with very tight correlations. Likewise the Visigothic Code, down to the interpretationes. This gives a clue as to how you move from a Roman model to the Visigothic code. There is less direct evidence with the Ostrogoths and Langobards in Italy; Lex Salica and the Anglo-Saxon laws are even further away. MOL   There seems to have been no centre or central place serving the whole Frisian territory around AD 800. The Frisian central places are likely to have functioned within the separate pagi. There was no gathering centre like the Icelandic Alþing. Oostergo and Westergo became united as a result of a common threat only in the thirteenth. NIJDAM   I agree that each pagus should be expected to have a thing-assembly. If Radbod had his powerbase in one pagus but was able to dominate a larger area, how did that work: did he have to go to each regional assembly, or could he bring them together somewhere? Some Old Frisian law texts do represent a pan-Frisian perspective. IJSSENNAGGER-VAN DER PLUIJM   And they also explicitly mention solidarity between the areas. If one area is attacked, the others come to help (e.g. in the Hunsinger Recht). So even if they are separate areas, there is this sense of belonging together one way or the other, of supporting each other and of coming together. There must have been a way of getting that support and of having that communication. NIJDAM   The question then is when did that start? NIEUWHOF   In what way is the Lex Frisionum different from the Laws of the Saxons or other laws? Are there different fines, or is the system different? NIJDAM   One thing is there is a lot of tariff material in Lex Frisionum, probably because it was a draft, not a final edited version. Only a copy of 1557 survives. In his classic study Siems (1980) mentions references to a medieval manuscript that contained a copy of the text. The Lex Frisionum code appears to have been unknown in Frisia in the later Middle Ages. Coming back to your question: there is a lot of material that is also found in other Leges Barbarorum. WOOD   Most of the Leges overlap considerably; a lot has to do with local landscapes, for instance, water or hedges etc. Different societies are described; but the basic idea of a compensation culture is clear. This is how very small communities operate. Some have more information on how law suits should be pursued. I think the main point is that the major cases will go to the king, not to the local courts. NIEUWHOF   The same law is applied in those circumstances? WOOD   I think the king does what he likes; there is no law book for the top cases, and we have to turn to narratives … e.g. bishops’ disputes with counts. NIEUWHOF   I wonder why these laws are so similar? Do they have similar sources? WOOD   One view emphasizes the similarity between Scandinavian and Langobardic law codes (Sjöholm 1988); are those matters of borrowing, or are there similar solutions to similar problems? 164

Z01 Hines Book B.indb 164

07/05/2021 13:30

Law and Political Organization NIJDAM   Even borrowings can be significant. FLIERMAN   I was fascinated by what you said about the unfinished state of Lex Frisionum. It was an attempt to codify the distinct legal traditions of several Frisian regions, whilst adding certain Frankish regulations. However, as you said, the project was probably never completed. So I was wondering: what then did the legal situation in ninth-century Frisia look like? Did the old, pre-conquest legal situation simply continue into Carolingian Frisia: so each separate Frisian region living under its own customary law, without much Frankish intervention? NIJDAM   Do you mean what would the text have looked like had it been edited, or does it reflect what was actually done in the ninth century? FLIERMAN   I think both questions are worth asking. Firstly, if the Lex Frisionum had been implemented, how drastically would the legal situation in Frisia have changed? How radically were the compilers trying to intervene in existing Frisian legal traditions? Or was it on the whole a rather conservative text? Secondly, seeing as the Lex Frisionum was not finished or implemented, how much did the legal situation in ninth-century Frisia actually change in the end? What was the practical impact of Frisia’s (temporary) incorporation into the Carolingian Empire on the legal situation of someone living in, say, Westergo? NIJDAM   If I look at what I’ve got, I see that a lot of what is in Lex Frisionum also appears in the later Old Frisian Laws: for instance a legal procedure of claiming innocence by swearing oaths. Only later does synodical law (ecclesiastical law) bring in other investigative procedures for determining the truth. There are two instances in Lex Frisionum: one of a thief being found in the tunnel that is being dug, and one of the arsonist in the act of setting fire. They are said to be Frankish by scholars working on Lex Frisionum but they turn up in the Old Frisian Laws too. If they were indeed Frankish, they have also been adopted in later Old Frisian law and so would be interesting evidence of Lex Frisionum actually having been implemented in Frisia. IJSSENNAGGER-VAN DER PLUIJM   We cannot, of course, be sure it was never implemented, because we only have the one unfinished Lex Frisionum text. And the copy that was handed down to us is of a very late date. Rob, I think you were wondering if it would not have been implemented, would any of the Frankish elements ever have been present? FLIERMAN   Yes: so Carolingian authorities, in cooperation with people on the ground, tried to bring about a ‘new’ legal situation building on existing material and customs, but they may never have succeeded. How much changed then? NIJDAM   That is hard to say. There was an existing legal situation. One would have to look at those instances in Lex Frisionum that are considered Frankish and see whether they survived in Old Frisian Law. A complication is that shortly after 800 the Vikings started their raids and this had an effect on Frankish rule in Frisia in the period c. 850–925. NICOLAY   Do you think the Lex Frisionum was ever implemented? NIJDAM   In my view, ‘implementation’ could only apply to new Frankish paragraphs in Lex Frisionum. The pre-existing indigenous material in Lex Frisionum did not need to be ‘implemented’: it was law the Frisians already adhered to.

165

Z01 Hines Book B.indb 165

07/05/2021 13:30

Frisians of the Early Middle Ages WOOD   The clause on temples surely shows that it would never have been implemented under the Carolingian authorities. There is a parallel in the sacrificial pigs in Lex Salica (see Hines, this vol.). NIJDAM   I think Rob’s question is what new elements would Frankish influence have brought into Frisian law. There is a major disruption and hiatus in the Viking Period, and after that far more Church Law comes into play. MAJCHCZACK   Do we know if Lex Frisionum was applicable in North Frisia? NIJDAM   Unfortunately we know nothing about the legal situation in North Frisia before 1425. The region is not included in Lex Frisionum, which defines three areas between the Sincfal, Vlie, Lauwers and Weser. North Frisia is beyond the River Weser and hence not included. And this situation does not change in the High Middle Ages, since there are no Frisian medieval law texts from North Frisia. We do have Low German law texts from 1425 onward. Large parts of the North Frisian mainland also fell under Danish law (the Jyske Lov of Jutland).

References Primary sources Bede, Historia Ecclesiastica Gentis Anglorum. Eds. and trans. B. Colgrave and R. A. B. Mynors 1969, Bede’s Ecclesiastical History of the English People (Oxford). Beowulf. F. Klaeber 1950, Beowulf and the fight at Finnsburg (Boston). Capitulatio de partibus Saxoniae. Ed. C. von Schwerin 1918, Leges Saxonum und Lex Thuringorum, MGH Leges, Fontes iuris 4 (Hannover), 37–44. Lex Frisionum. Ed. K. A. Eckhardt and A. Eckhardt 1982, MGH Leges, Fontes iuris 12 (Hannover). Overkeuren, Hunsinger recht. Eds. W. J. Buma and W. Ebel 1969, Das Hunsingoer Recht, Altfriesische Rechtsquellen. Texte und Übersetzungen Band 4 (Göttingen). Vita Vulframni. Ed. W. Levison 1910, MGH Scriptores, SS Rer. Merov. 5 (Hannover). Secondary sources Algra, N. E. 2000, Oudfries Recht 800–1256 (Leeuwarden). Anderson, C. E. 1999, ‘Formation and Resolution of Ideological Contrast in the Early History of Scandinavia’ (Ph.D. thesis, University of Cambridge). Boehm, C. 2007, ‘The natural history of blood revenge’. In Feud in Medieval and Early Modern Europe, eds. J. Büchert Netterström and B. Poulsen (Aarhus), 189–203. Boutkan, D. and S. M. Siebinga 2005, Old Frisian Etymological Dictionary, Leiden Indo-European Etymological Dictionaries 1 (Leiden). Bremmer, R. H. 2004, Hir is eskriven. Lezen en schrijven in de Friese landen rond 1300 (Hilversum). — 2009, An Introduction to Old Frisian. History, Grammar, Reader, Glossary (Amsterdam–Philadelphia). — 2014, ‘The orality of Old Frisian law texts’. In Aspects of Old Frisian Philology, eds. R. H. Bremmer, G. van der Meer and O. Vries, Amsterdamer Beiträge zur älteren Germanistik 31–2 (Amsterdam), 1–48. Brink, S. 2002, ‘Law and legal customs in Viking Age Scandinavia’. In Scandinavians from the Vendel Period to the Tenth Century, ed. J. Jesch, Studies in Historical Archaeoethnology

166

Z01 Hines Book B.indb 166

07/05/2021 13:30

Law and Political Organization (Woodbridge), 87–117. Buma, W. J. 1949, Het Godsoordeel in de Oud-Friese literatuur. Rede uitgesproken bij de aanvaarding van het ambt van bijzonder hoogleraar in de Friese taal- en letterkunde aan de Rijksuniversiteit te Utrecht (Groningen–Batavia). Broome, R. C. 2014, ‘Approaches to Community and Otherness in the Late Merovingian and Early Carolingian Periods’ (Ph.D. thesis, University of Leeds). De Koning, J. 2018, ‘Trans Flehum. Wijnaldum, Den Burg, Texel, Westergo: het Vlie als verbinder en grens’. In Fragmenten uit de Rijke Wereld van de Archeologie, eds. A. Nieuwhof, E. Knol and J. Schokker, Jaarverslagen van de Vereniging voor Terpenonderzoek 99 (Groningen), 146–56. De Langen, G. J. and J. A. Mol 2017, ‘Church foundation and parish formation in Frisia in the tenth and eleventh centuries: a planned development?’, The Medieval Low Countries 4, 1–55. 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 (Leiden). Ehbrecht, W. 2003, ‘Gemeinschaft, Land und Bund im Friesland des 12. bis 14. Jahrhunderts’. In Die Friesische Freiheit des Mittelalters – Leben und Legende, ed. H. van Lengen (Aurich), 134–93. Gerbenzon, P. 1973, ‘Der altfriesische asega, der altsächsische eosago und der althochdeutsche esago’, Tijdschrift voor Rechtsgeschiedenis 41, 75–91. — 1983, ‘Review of Siems, Studien’, Tijdschrift voor Rechtsgeschiedenis 51, 169–73. Green, D. H. 1965, The Carolingian Lord (Cambridge). — 1998, Language and History in the Early Germanic World (Cambridge). Hallebeek, J. 2019, Layci erant coiudices. Over de rol van leken in de Westerlauwerse seendprocedure. Rede in verkorte vorm uitgesproken bij het afscheid als hoogleraar Europese rechtsgeschiedenis aan de Vrije Universiteit te Amsterdam op 26 april 2019 (Amsterdam, Vrije Universiteit). Henstra, D. J. 2000, ‘The Evolution of the Money Standard in Medieval Frisia. A Treatise on the History of the Systems of Money of Account in the Former Frisia (c. 600–c. 1500)’ (Ph.D. thesis, University of Groningen). — 2012, Friese graafschappen tussen Zwin en Wezer. Een overzicht van de grafelijkheid in middeleeuws Frisia (ca. 700–1200) (Assen). — 2010 (ed. Anne T. Popkema), Fon Jelde. Opstellen van D.J. Henstra over Middeleeuws Frisia (Groningen). His, R. 1910, Das Strafrecht der Friesen im Mittelalter (Leipzig). Hoppenbrouwers, P. C. M. 1985, ‘Maagschap en vriendschap. Een beschouwing over de structuur en functies van verwantschapsbetrekkingen in het laat-middeleeuwse Holland’, Holland. Regionaal-historisch tijdschrift 17, 69–108. Hulscher, H. 2002, ‘De omslag van zoengeld over de verwanten van een doodslager volgens het Rechtsboek van Den Briel’, Pro Memorie. Bijdragen tot de rechtsgeschiedenis der Nederlanden 4.1, 68–86. Humphreys, M. 2017, The Laws of the Isaurian Era: the Ecloga and its Appendices (Liverpool). Hybel, N. 2017, The Nature of Kingship c. 800–1300: The Danish Incident, The Northern World 83 (Leiden). Iversen, F. 2013, ‘Concilium and pagus – revisiting the Early Germanic thing system of northern Europe’, Journal of the North Atlantic, Special Volume 5, 5–17. — 2020, ‘Between tribe and kingdom – people, land, and law in Scandza AD 500–1350’.

167

Z01 Hines Book B.indb 167

07/05/2021 13:30

Frisians of the Early Middle Ages In Rulership in 1st to 14th Century Scandinavia. Royal Graves and Sites at Avaldsnes and Beyond, ed. D. Skre, Ergänzungsbände zur Reallexikon der Germanischen Altertumskunde (Berlin–Boston), 245–304. Kerrigan, J. 2020, Brehon Laws. The Ancient Wisdom of Ireland (Dublin). Knol, E. 2012, ‘Dr. W. K. J. Schoor and the Frisian terp dog’. In A Bouquet of Archaeozoological Studies. Essays in Honour of Wietske Prummel, eds. D. C. M. Raemaekers, E. Esser, R. C. G. M. Lauwerier and J. T. Zeiler (Groningen), 21–9. Knol, E., W. Prummel, H. T. Uytterschaut, M. L. P. Hoogland, W. A. Casparie, G. J. de Langen, E. Kramer and J. Schelvis 1996, ‘The early medieval cemetery of Oosterbeintum (Friesland)’, Palaeohistoria 37/38, 245–416. Krause, H. 1971, ‘Gesetzgebung’. In Handwörterbuch zur deutschen Rechtsgeschichte, eds. A. Erler and E. Kaufmann (Berlin), 1606–20. Lambert, T. B. and D. Rollason eds. 2009, Peace and Protection in the Middle Ages (Toronto). Langbroek, E. 1990, ‘Condensa atque Tenebrosa; die altfriesischen Psalmen: Neulesung und Rekonstruktion (UB Groningen Hs 404)’. In Aspects of Old Frisian Philology. Eds. R. H. Bremmer, G. van der Meer and O. Vries, Amsterdamer Beiträge zur älteren Germanistik 31–32 (Amsterdam–Groningen), 255–84. Langbroek, E. in collaboration with F. Brands, 2015, ‘So viel geschrieben, so wenig geblieben. Eine neue Entdeckung: unbekannte altfriesische Psalmglossen’, Amsterdamer Beiträge zur älteren Germanistik 74, 135–46. Laur, W. 1996, ‘Zum Namen des sagenhaften Friesenkönigs Folcwalda: Überlegungen zum Verhältnis Eigenname – Appellativ’. In Friezen. In bondel stúdzjes oer persoansnammen, eds. R. A. Ebeling, K. F. Gildemacher and J. A. Mol, Fryske Nammen 10 (Leeuwarden), 171–4. Lübke, C. and P. Urbańczyk 2009, ‘Ests, Slavs and Saxons: Ethnic Groups and Political Structures’. In Wulfstan’s Voyage: The Baltic Sea Region in the Early Viking Age as Seen from Shipboard, eds. A. Englert and A. Trakadas, Maritime Culture of the North 2 (Roskilde), 50–7. Mol, J. A. 2006, ‘Galgen in laatmiddeleeuws Friesland’, De Vrije Fries 86, 95–140. Meijering, H. D. 1974, De Willekeuren van de Opstalsboom (1323). Een Filologisch-Historische Monografie (Groningen). Munske, H. H. 1973, Der germanische Rechtswortschatz im Bereich der Missetaten. Philologische und sprachgeografische Untersuchungen. I. Die Terminologie der älteren westgermanischen Rechtsquellen (Berlin–New York). Nicolay, J. A. W. 2014, The Splendour of Power: Early Medieval Kingship and the Use of Gold and Silver in the Southern North Sea Area (5th to 7th century AD), Groningen Archaeological Studies 28 (Groningen). Nijdam, H. 2008, Lichaam, eer en recht in middeleeuws Friesland. Een studie naar de Oudfriese boeteregisters (Hilversum). — 2014, ‘Indigenous or universal? A comparative perspective on medieval (Frisian) compensation law’. In How Nordic Are the Nordic Laws? Ten Years after. Proceedings of the Tenth Carlsberg Conference on Medieval Legal History 2013, eds P. Andersen, K. Salonen, H. Sigh and H. Vogt (Copenhagen), 161–81. — 2017, ‘A comparison of the injury tariffs in the early Kentish and the Frisian law codes’. In Frisians and their North Sea Neighbours. From the Fifth Century to the Viking Age, eds. J. Hines and N. IJssennagger (Woodbridge), 223–42. — 2019, ‘The body legal in Frisian law. Bridging the gap between the Lex Frisionum and

168

Z01 Hines Book B.indb 168

07/05/2021 13:30

Law and Political Organization the Old Frisian compensation tariffs’. In Languages of the Law in Early Medieval England. Essays in Memory of Lisi Oliver, eds. A. Rabin and S. Jurasinski (Leuven–Paris–Bristol), 101–26. Nijdam, H. and O. S. Knottnerus 2019, ‘Redbad, the Once and Future King of the Frisians’. In Northern Myths, Modern Identities. The Nationalisation of Northern Mythologies since 1800, eds. S. Halink and M. Baár (Leiden–Boston, Brill), 87–114. Nijdam, H. and J. Savelkous in collaboration with P. Breuker 2017, ‘The manuscript collection of the Frisian State Historian Simon Abbes Gabbema (1628–1688) from an Old Frisian perspective’. In Frisian Through the Ages. Festschrift for Rolf H. Bremmer Jr., eds. M. de Vaan and S. Laker, Amsterdamer Beiträge zur älteren Germanistik 77 (Leiden), 303–32. Nijdam, H. and A. Versloot 2012, ‘Kodeks Siccama. Spoaren fan in ferdwûn Aldwestfrysk rjochtshânskrift’, Us Wurk 61, 1–56. Noomen, P. N. 2009, De Stinzen in Middeleeuws Friesland en hun Bewoners (Hilversum). — 2012, ‘Eigenerfd of edel? Naar aanleiding van de af komst van de Aytta’s’, It Beaken 74, 257–301. Nørgård Jørgensen, A., L. Jørgensen and L. Gebauer Thomsen 2011, ‘Assembly sites for cult, markets, jurisdiction and social relations: historic-ethnological analogy between North Scandinavian church towns, Old Norse assembly sites and pit house sites of the Late Iron Age and Viking Period’. In Det 61. Internationale Sachsensymposion 2010 Haderslev, Danmark, ed. L. Boye (Neumünster), 95–112. Oliver, L. 2002, The Beginnings of English Law (Toronto–Buffalo–London). — 2011, The Body Legal in Barbarian Law (Toronto–Buffalo–London). Popkema, A. T. 2007, ‘Die altfriesischen Eidesbezeichnungen’. In Advances in Old Frisian Philology, eds. R. H. Bremmer Jr., S. Laker and O. Vries, Amsterdamer Beiträge zur älteren Germanistik 64 (Amsterdam – New York), 325–56. Prummel, W. 1992, ‘Early medieval dog burials among the Germanic tribes’. Helinium 32, 132–94. — 1993, ‘Paarden en honden uit vroeg-middeleeuwse grafvelden’. In Het Tweede Leven van Onze Doden, eds. E. Drent, W. A. M. Hessing and E. Knol (Ammersfoort), 53–60. Savelkouls, J. 2015, ‘The Friesian horse and the Frisian horse, The (re)invention and the historicity of an iconic breed’, De Vrije Fries 95, 9–42. — 2016, Het Friese Paard (Gorredijk). Schuur, J. R. G. 1976, ‘De betekenis van de naam “Cammingehunderi”’, De Vrije Fries 56, 31–41. Seebold, E. and R. Schneider 2001, ‘König und Königtum’. In Reallexikon der Germanischen Altertumskunde 17, eds. Heinrich Beck et al. (Berlin–New York), 102–9. Siems, H. 1980, Studien zur Lex Frisionum (Ebelsbach). Sjöholm, E. 1988, Sveriges medeltidslaga: europeisk rättstradition i politisk omvandling (Stockholm). Skre, D. 2020, ‘Rulership and ruler’s sites in 1st–10th-century Scandinavia’. In Rulership in 1st to 14th century Scandinavia. Royal Graves and Sites at Avaldsnes and beyond, ed. D. Skre, Ergänzungsbände zur Reallexikon der Germanischen Altertumskunde (Berlin–Boston), 193–234. Timmer, R., 2000, ‘Restanten van Oud-Germaans recht in de Lex Frisionum’, Pro Memorie. Bijdragen tot de rechtsgeschiedenis der Nederlanden 2.1 (2000), 6–45. Van Lengen, H. 2003a, ‘Tota Frisia: Sieben Seelande und mehr. Die territoriale Gliederung

169

Z01 Hines Book B.indb 169

07/05/2021 13:30

Frisians of the Early Middle Ages des freien Frieslands im Mittelalter: ein Überblick mit einer Karte’. In Die Friesische Freiheit des Mittelalters – Leben und Legende, ed. H. van Lengen (Aurich), 56–89. — 2003b, ‘Karl der Große, Jungfrau Maria und andere Heilsbringer als Garanten und Patrone Friesischer Freiheit. Zu den Siegeln der Landesgemeinden Frieslands im Mittelalter’. In Die Friesische Freiheit des Mittelalters – Leben und Legende, ed. H. van Lengen (Aurich), 90–133. Versloot, A. P. 2015, ‘Die “friesischen” Wörter in der Lex Frisionum’. Us Wurk 64, 1–10. Vogt, H. and H. Nijdam, ‘Translating a medieval legal system into modern English’. In Translation and Medieval Documents. Voices of Law: Language, Text and Practice (Digital Online publication Voices of Law 2018: https://voicesoflaw.files.wordpress.com/2018/06/ vol-translation-booklet.pdf), 38–53. Vries, O. 1986, Het Heilige Roomse Rijk en de Friese Vrijheid (Leeuwarden). — 2015, ‘Frisonica Libertas. Frisian Freedom as an Instance of Medieval Liberty’, Journal of Medieval History 1, 1–20. Weitzel, J. 2006, ‘Die Bedeutung der Dinggenossenschaft für die Herrschaftsformung’. In Leges – Gentes – Regna. Zur Rolle von germanischen Rechtsgewohnheiten und lateinischer Schrifttradition bei der Ausbildung der frühmittelalterlichen Rechtskultur, eds. G. Dilcher and E. M. Distler (Berlin), 351–66. Wolfram, H. 2006. ‘Grundlagen und Ursprünge des europäischen Königtums’. In Leges – Gentes – Regna. Zur Rolle von germanischen Rechtsgewohnheiten und lateinischer Schrifttradition bei der Ausbildung der frühmittelalterlichen Rechtskultur, eds. G. Dilcher and E. M. Distler (Berlin), 185–96. Wood, I. 1983, ‘Roman Law in the Barbarian Kingdoms’. In Rome and the North, eds. A. Ellegärd and G. Åkerström-Hougen (Jonsered), 5–14. Wormald, P. 1977, ‘Lex scripta and verbum regis: legislation and Germanic kingship, from Euric to Canute’, in Early Medieval Kingship, eds. P. Sawyer and I. Wood (Leeds), 105–38. — 1999, The Making of English Law: King Alfred to the Twelfth Century. Volume I. Legislation and Its Limits (Oxford). — 2003, ‘The Leges Barbarorum: law and ethnicity in the post-Roman West’. In Regna and Gentes. The Relationship between Late Antique and Early Medieval Peoples and Kingdoms in the Transformation of the Roman World, eds. H.-W. Goetz, J. Jarnut and W. Pohl, The Transformation of the Roman World 1 (Leiden–Boston), 21–53. Yorke, B. 1990, Kings and Kingdoms of early Anglo-Saxon England (London–New York).

170

Z01 Hines Book B.indb 170

07/05/2021 13:30

• 6 • RECENT DEVELOPMENTS IN EARLY-MEDIEVAL SETTLEMENT ARCHAEOLOGY: THE NORTH FRISIAN POINT OF VIEW Bente Sven Majchczack

T

he German region of North Frisia (Kreis Nordfriesland) is the northernmost part of the ‘Frisian’ settlement area and some of the population still speak several local Frisian dialects today. Being somewhat removed from the Frisian core area in the Netherlands and Lower Saxony, it is often out of focus when dealing with Frisian archaeology. In the past decade, new archaeological discoveries and subsequent investigations have provided new insights into the Early-medieval settlements of North Frisia. Their focus lies on the three northern islands of Sylt, Föhr and Amrum, which possess an especially rich archaeological heritage and may be regarded as a foremost important settlement area for the first settlers in the period. The paper will give a short overview in the recent developments and aims to draw a picture of the Early-medieval inhabitants of North Frisia.

The state of settlement research The development of the North Frisian population and identity can only be traced back to the High Middle Ages by means of historical accounts; the earlier phases must be examined through archaeological records. The beginning of Frisian settlement on the North Frisian islands is traditionally considered to have commenced no earlier than the mid-seventh century AD, moving into a landscape void of settlement activity due to the fourth- and fifth-century migrations in the North Sea area (Jankuhn 1960; Århammar 1995). The early settlement activities concentrated mostly on the North Frisian islands of Sylt, Föhr and Amrum, the marsh island Wiedingharde, parts of Eiderstedt and the edges of the mainland’s moraines, the ‘geest’. From the seventh century onwards, lively settlement activity unfolded, leading to a settlement pattern which can still be seen today, since most of the island villages’ place-names date back to the Early Medieval Period (Laur 1960). Archaeological research concentrates on the little known period between the seventh and eleventh centuries AD (Fig. 6.1). In the research history of the late nineteenth and twentieth centuries, the objects of excavation were the monuments which left their marks in the landscape – hundreds of burial mounds and impressive defensive enclosures in the form of ring-shaped, earthen ramparts (Ringwallburgen: here referred to simply as ‘ringforts’). While numerous graves were more or less scientifically excavated 171

Z01 Hines Book B.indb 171

07/05/2021 13:30

Frisians of the Early Middle Ages

Fig.  6.1 Map of North Frisia with sites mentioned in the text: 1 Tinnum, 2 Tinnumburg, 3 Archsum, 4 Wrixum, 5 Midlum, 6 Witsum, 7 Goting, 8 Nieblum, 9 Borgsumburg, 10 Hooge, 11 Pellworm, 12 Elisenhof. Black circles: ringforts. Dotted white line: boundary of the Nordfriesland region. Map: B. Majchczack; map data: GÜK200 © Bundesanstalt für Geowissenschaften und Rohstoffe, Hannover, 2007.

to make way for intensified agriculture, only a little knowledge was gained about the settlements. This changed somewhat in the early 1950s, with small-scale excavations in the ringfort of Borgsum/Föhr (Segschneider 2009). Much more was learned from the Eiderstedt peninsula south of the islands. The large-scale excavation of the Early-medieval Wurt (= terp or wierde) settlement Elisenhof by the River Eider in 1957–64 provided deep insights into the settlements in the marsh areas. The Elisenhof excavations results are well published (e.g. Bantelmann 1975; Westphalen 2014). Another large-scale project of 1963–78 explored the settlement history of the Archsum moraine area on the island of Sylt (Kossack et al. 1980). The Early-medieval settlement of Alt-Archsum 172

Z01 Hines Book B.indb 172

07/05/2021 13:30

Early-Medieval Settlement Archaeology was excavated in large parts, and revealing three multi-phase farmsteads, but the site still awaits full publication (Kossack et al. 1974, 355–64; Reichstein 1986, preliminary reports). The Wadden Sea area was investigated in the years 1975–1981 by the Norderhever Project, leading to insights into Early-medieval settlements on level ground in the marshes (Higelke et al. 1988). In the 1970s and in 1986, parts of an Early-medieval settlement on the mainland were excavated at Rantrum (Jöns 1996, 201–22). During the 1990s, some excavations were conducted at wurts in the marsh areas of the Wiedingharde (today part of the mainland) and Föhr (Segschneider 2018). Up until that time, no major Early-medieval settlement had been surveyed or excavated on a large scale on the North Frisian islands. This changed dramatically from 2006, with development-driven excavations in Tinnum/Sylt (Segschneider 2008; Majchczack and Segschneider 2015) and with results from aerial archaeology, leading to the discovery of several major settlements on Föhr and Amrum (Schlosser Mauritsen et al. 2009; Segschneider et al. 2015). Combining aerial photography and geophysical methods such as magnetometry, ground-penetrating radar and geoelectrics, the settlements at Witsum, Nieblum, Goting and Wrixum (Föhr) were investigated (Majchczack 2015). It was possible to gain extensive insight into the settlement layouts, buildings and the overall settlement structure on the island. In 2013–18, the North Sea Harbour Project within the German Research Foundation’s Special Priority Programme 1630 on Harbours conducted further geophysical research and several archaeological excavations on the sites to secure the dating and learn about the economic activities (Majchczack et al. 2018; Majchczack and Offermann 2018; full publication: Majchczack 2020).

Remarks on the landscape The North Frisian area forms a natural landscape with diverse settlement conditions. The coastal landscape was shaped by still active geologic processes and human activity, especially the construction of dikes and the cultivation of fenlands. Therefore, the modern landscape differs much from the conditions of the Early Medieval Period. No precise reconstruction of the coastline and the extent of the marshlands fit for habitation and agriculture is possible, but the geological and archaeological observations reveal a conclusive picture (Hoffmann 1988; Behre 2003). In the Early Middle Ages, the islands of Sylt, Föhr and Amrum were the most favourable locations for settlements. The elevated moraine cores and adjacent marshlands were perfectly suited for settlements and agricultural activities – dry land for farming and housing, and marshland for grazing sheep and cattle. It is very likely that the islands were then surrounded by larger marshlands, embedded in a maritime Wadden Sea landscape of mudflats and large tidal inlets. The islands possess the richest archaeological record in North Frisia. In the southern Wadden Sea area, around the Hallig Hooge and island of Pellworm, only a few remains of Early-medieval settlements are known. It seems that only small areas of arable marshland at the back of sand spits were accessible in the period, with large areas of what is today the Wadden Sea area covered by bogs and uninhabitable wet marshlands. The same seems to be true for the North Frisian mainland. Most of the now fertile marshlands were fenlands and not cultivated before the High and Late Medieval Period. Only a few settlement sites are known along the elevated moraine boundary of the mainland, usually close to larger rivers with access to the sea. The southernmost area is the Eiderstedt peninsula, which formed another 173

Z01 Hines Book B.indb 173

07/05/2021 13:30

Frisians of the Early Middle Ages hotspot of Early-medieval settlement activity. Bordering on fossil sand spits, an arable sea-marsh had formed along the estuary of the Eider. The raised river embankments especially provided flood-safe settlement grounds, leading to a chain of settlements on level ground (Flachsiedlungen), which developed into wurts like Tofting and Elisenhof. In the entire Wadden Sea area, sedimentation was low during the first millennium AD and we can assume a fairly stable coastline. Beginning around AD 1000, a marine transgression led to further development of marshlands and, connected to extensive dike-building, later to the catastrophic storm surges of the fourteenth and seventeenth centuries, which led to drastic changes in the shape of the Wadden Sea area.

Preceding the ‘Frisian immigration’: settlement activities of the fourth–seventh centuries Before discussing the Early-medieval settlement and trade structures of North Frisia, some comments on the preceding period are necessary. The existing settlement landscape of the fourth to early sixth century and the settlement hiatus of the late Migration Period are important aspects in the conditions for the Frisian immigration of the Early Medieval Period. During the fourth and fifth centuries, the archaeological record is quite dense and provides a picture of a flourishing population. Settlement activities are mostly focused along the boundaries of the elevated geest. The North Frisian islands possess an especially rich record due to the excavations of the Archsum Project at the settlement mound of Melenknob on the island of Sylt, which yielded a multiple-phased large farmstead (Kossack et al. 1974; Harck et al. 1992), and due to large-scale aerial and geophysical settlement surveys on the island of Föhr (Majchczack 2015b). The rural settlement sites are dispersed along the geest-marsh boundaries in the form of single farmsteads and small villages of multiple farmsteads, in a position to use the elevated geest for farming and the adjacent marshes for husbandry. The favourable conditions for archaeological prospections on Föhr are best illustrated at the Wrixum site (Fig. 6.2). The aerial images and geomagnetic measurements show a detailed settlement layout of large farmsteads of more than 4,000 sq m, containing three-aisled longhouses of up to 60 m, wells and SFBs (Grubenhäuser). The farmsteads are surrounded by ditch-systems and several farmsteads combine to form large enclosed groups. The longhouses have clear parallels in similar settlements in the Elbe-Weser Triangle (e.g. at Flögeln-Eekhöltjen: Dübner 2015) as well as in southern Jutland (e.g. Vorbasse: Hvass 1986) and can be dated to the fourth–fifth centuries AD. The settlement pattern is also comparable to those areas: single farmsteads as well as clusters of farmsteads can be found to shift over time within a settlement area on the geest. The houses move with each generation, creating large archaeological sites. Similar settlements are also found in the southern North Frisian mainland, as shown by excavations in Rantrum (Jöns 1996). In Eiderstedt, settlements are found on the ridges of fossil sand spits (Bokelmann 1988) and in the marshes on the riverbanks of the river Eider in the form of large village wurts (for example, at Tofting, Bantelmann 1955). Similar marsh settlements are also found in the Wiedingharde (Segschneider 2018). Apart from the rural settlements, traces of a specialized beach-market site of the fourth/fifth centuries have been uncovered in the dunes at Norddorf on the island of Amrum. Located at a natural harbour, a seasonal settlement with SFBs served as a local site for production and trade. A large number of Frankish glass-vessel sherds and glass 174

Z01 Hines Book B.indb 174

07/05/2021 13:30

Early-Medieval Settlement Archaeology

Fig. 6.2  Aerial view of the Migration-period settlement at Wrixum/Föhr. After Majchczack 2015, 29, fig. 7 (Aerial photograph: E. S. Mauritsen, Holstebro Museum).

beads testify to an influx of high-status luxury goods to the islands and trade connections along the North Sea coast (Segschneider 2002). These and other imported goods such as coins, quernstones and high-quality metal objects are distributed in the settlement network and have been found in settlements on the islands, the Wiedingharde and the mainland (Jöns 1996; Majchczack and Segschneider 2015; Segschneider 2018). The prehistoric archaeological record on the North Frisian islands ends with the beginning of the sixth century. Not a single archaeological feature is known which can be securely dated much after AD 500. The burial sites end in the fifth century with few known finds, mostly Saxon-style pottery and cruciform brooches (Majchczack 2018). The same seems true for settlement sites, but this inference is conditional because of the very limited excavation of Migration-period settlements. But some observations are possible: the Archsum Project noticed an end of the settlements in the fourth century, and only some secondarily deposited pottery sherds are believed to date into the sixth century (Harck et al. 1992, 35). More recent excavations yield datable metal finds from the settlements of Tinnum/Sylt and Goting/Föhr which may belong to the first quarter of the sixth century. They provide the latest dating before the hiatus (Majchczack 2015a; Majchczack and Segschneider 2015). Although a complete settlement hiatus can be neither proven nor rejected, a general decline in settlement activity is obvious and it must be assumed, that large parts of the North Frisian population emigrate from the region in the course of the fifth and early sixth centuries. 175

Z01 Hines Book B.indb 175

07/05/2021 13:30

Frisians of the Early Middle Ages

The Early-medieval settlements: seventh to mid-ninth centuries The earliest group of Early-medieval finds are continental gold coins found on Föhr and Sylt. One Merovingian tremissis found in Alkersum/Föhr and two imitation Madelinus tremisses of the Dronrijp type from List/Sylt and Wenningstedt/Sylt form the well-known and often cited evidence for the beginning of the Early Medieval phase on the islands, and for immediate connections with the Frisian-Frankish area from the mid- to late seventh century AD (Kersten and La Baume 1958, 84–5), although these finds were all single finds and not connected to settlement sites. The recent excavations of trade settlements on Föhr and Sylt led to insights into the early phases of the re-settlement in the late seventh century. The earliest secure dates came from Tinnum/ Sylt with well-timbers dated around AD 659, and in Witsum/Föhr a house was built around AD 697; its timbers were subsequently re-used in a well. Several investigated settlements will be discussed in detail. We can distinguish between settlements in the marshes and on the geest. The marsh settlements are mostly agrarian, but wurts located along major trade routes also profited from craft production and trade to the degree that they can be considered to be ‘commercial wurts’ and to have served as part of the network of trading centres along

Fig. 6.3  Schematic model of the system of trade and communication in the southern North Sea area. After Siegmüller / Jöns 2012, fig. 7, revised.

176

Z01 Hines Book B.indb 176

07/05/2021 13:30

Early-Medieval Settlement Archaeology the North Sea coast (as defined by Siegmüller and Jöns 2012). In those areas where the Pleistocene geest boundary connects with the North Sea via tidal channels and rivers, another settlement type with a high emphasis on craft production and trade can be found, often in the vicinity of fortifications in the form of ringforts (defined as Geestrandhafen and Geestrandburgen: Siegmüller and Jöns 2012; Fig. 6.3). The fortifications as well as the associated craftsmen’s settlements are located in the transitional zone between the geest and the marshland and it can be assumed that their inhabitants had full access to the supraregional trade routes and could thus also sell their locally produced wares (Jöns 2015, 249–50). The sites of Witsum and Goting/Föhr and Tinnum/Sylt belong to this settlement type and will be presented in detail together with some remarks on the marsh and wurt settlements. Marsh settlements An example of a small agrarian marsh settlement was excavated at Midlum/Föhr (Segschneider 2018, 25–33). This started as a settlement on level ground in the late seventh or mid-eighth century in a favourable, slightly elevated location. The single farmstead consisted of a three-aisled longhouse with clay-sod walls. The site was continuously raised while the house was renewed in up to four phases with application of fresh clay layers, thus forming a small settlement mound. After four phases, the house was abandoned and the mound covered with two thick layers of clay, dating to the ninth/tenth century. A new longhouse was built in the same spot and the settlement lived on into the second half of the eleventh century. The finds from the houses include simple household items such as pottery, whetstones, simple tools made of flint and amber pieces. The zoological and botanical remains reveal the agricultural activities of a single farmstead in the marsh: husbandry of cattle and sheep and the cultivation of barley. Finds of rye and emmer wheat indicate a supply of grain from the geest, since it would have not been possible to grow these crops in the marshes. The wurt of Elisenhof in Eiderstedt is an example of an extensively excavated wurt settlement with many households and long continuity from the eighth to the eleventh century (Westphalen 2014). The first longhouses were built in the early eighth century on the edge of a high embankment of the Eider, providing a flood-safe location and a navigable access to the river. The surrounding sea marshes were suitable for grazing. The settlement grew constantly through the eighth and ninth centuries. The settlement layout initially followed the topography of the river embankment. After the eighth century, the layout became more systematic. The farmsteads were built in two rows of long and narrow parallel yards. The longhouses were of different designs, mostly three-aisled with walls of clay-sods or wattle-and-daub. They reached lengths of up to 32 m and the byre sections held twelve to thirty-two head of cattle. In a total of ten settlement phases, sixty-eight longhouses were built, forming up to ten simultaneous farms. The permanent rebuilding of houses and application of clay layers formed a considerable wurt with a height of 4 m. The people of Elisenhof kept sheep for wool production (c. 50% of the livestock), cattle and pigs for meat (27% and 18%) and horses, using the salt marshes along the Eider as pasture. The more elevated marsh areas to the north were used for the cultivation of beans, barley, oats and flax (Behre 1976). The diet was complemented with intensive fishing in the Eider as well as on the open sea (Reichstein 1984). In addition to their agriculture, the people of Elisenhof were engaged in craft production and trade. Textile production from wool can be found in nearly every 177

Z01 Hines Book B.indb 177

07/05/2021 13:30

Frisians of the Early Middle Ages household. Amber, bone, leather, iron and bronze were used for production. It seems that the settlement profited strongly from its location along the trade route between the North Sea and Haithabu via the Eider, Treene and the land route along the Dannevirke. Many imported goods from Frisia, the Frankish Empire and Scandinavia were found in Elisenhof, such as quernstones of basaltic lava, whetstones, soapstone vessels, glass vessels and glass beads (Westphalen 1999). It has been argued that the economic activities and the access to imported goods in Elisenhof far exceed a subsistence economy and the wurt inhabitants were heavily engaged in trade, being located in a favourable spot on the shipping route towards Hedeby. The Elisenhof wurt has therefore been labelled a ‘commercial wurt’ (Jöns 2015, 249). In the research tradition of the Early-medieval wurt settlements in the East Frisian marsh belt, Haarnagel (1955; 1984) and Reinhard (1959) defined the ‘long wurts’ as a particular, trade-related settlement type. Based on the excavations of Emden and Groothusen, wurt settlements with an elongated shape and a settlement layout along a central road axis were seen as specialized trade settlements founded by craftsmen and traders in the Early Medieval Period, highlighted by finds of imported goods and specialized crafts (most recently Brandt 1994; Stilke 1995). This model of a settlement type is problematic, since it could not be applied to the neighbouring marsh areas in the Netherlands, the Elbe-Weser Triangle or Schleswig-Holstein (see also de Lange and Mol, this vol.). Moreover, re-evaluation and recent excavations in the ‘long wurts’ of Emden, Groothusen and Grimersum have shown that the initial settlements are agrarian and the elongated layouts developed later on. Therefore, the model of ‘long wurts’ has been abandoned in favour of a more neutral term ‘commercial wurts’, which fits to a wider array of wurt settlements that served as stations for trade and transhipment in the wider marsh belts along the southern North Sea coast (Eichfeld 2015; Jöns 2015, 249; Siegmüller and Jöns 2012). Geest settlements The Early-medieval settlement of Witsum is located close to the southern coast of Föhr, at the foot of the geest (Fig. 6.4). The mouth of the small, tidally influenced River Godel joins the North Sea directly south of the settlement, draining a wetland with salt marshes. The river touches the southern edge of the settlement site, providing a possible navigable connection to the sea. The settlement was first discovered in the summer of 2006 by aerial survey through crop marks, indicating the presence of SFBs and ditch systems. In the following years, the site was fully surveyed using magnetometry. Further geophysical measurements were conducted on selected magnetic anomalies using ground-penetrating radar and electric resistivity tomography. The general settlement layout, extending over an area of c. 10 hectares, is composed of large yards surrounded by ditches and located along both sides of a central east–west roadway. Inside the yards, up to 105 strong rectangular magnetic anomalies indicate SFBs (Majchczack 2015, 87–8). In several cases SFB anomalies form groups of two to six features close together, possibly indicating several phases of the same building. The impression of multiple phases of the settlement is further strengthened by intersecting linear structures, presumably further ditches. Only a few clusters of anomalies indicate longhouses, giving the impression that the settlement structure is strongly dominated by SFBs. 178

Z01 Hines Book B.indb 178

07/05/2021 13:30

Early-Medieval Settlement Archaeology

Fig. 6.4  Magnetic map (±9 nT) of the settlement at Witsum with excavation trenches. After Majchczack and Offermann 2018, fig. 2 (Coordinates DHDN/Gauss-Krüger Zone 3; magnetic map: T. Wunderlich; aerial view © Google Maps; cartography: B. Majchczack).

The overall settlement layout is comparable to Early-medieval rural settlements which have been excavated in Jutland like Sædding, Vorbasse, Omgård or Åparken (Stoumann 1979; Hvass 1979; Nielsen 1979; Eriksen et al. 2009, 267–70). In any event, the dominance of the SFBs in the settlement structure, the close vicinity to the North Sea as well as the presence of craft and trade-related finds, suggests an interpretation as a specialized trading site with an emphasis on craft production. A systematic field survey and metal detector prospection have confirmed the Early-medieval dating of the site, yielding typical goods such as local handmade pottery, loomweights, quernstone fragments of basaltic lava and a German silver denier of Heinrich II, minted in Duisburg between 1039 and 1046 (Majchczack 2015, 85–7). Four archaeological excavation campaigns were conducted in 2013–16 (Majchczack et al. 2018; Majchczack and Offermann 2018). Small trenches in the southern area sought to locate structures connected with harbour activities, while a trial trench in the north-eastern part of the settlement uncovered dense remains of longhouses. One entirely excavated longhouse with a trapezoid outline and a length of 19 m contained only living quarters and no signs of a byre. Excavation in the central settlement area investigated the yard enclosures, composed of multiple ditches and lined with palisades, as well as the central road with cart tracks, several SFBs, one well and parts of a longhouse. In total, 700 sq m were excavated, which is about 0.7% of the entire settlement. Simultaneously, an extensive geoarchaeological coring campaign was able to establish a better understanding of the maritime connection of the settlement. It was possible to identify a maritime influence in the Godel wetlands from the seventh/ eighth century, forming a navigable natural harbour. 179

Z01 Hines Book B.indb 179

07/05/2021 13:30

Frisians of the Early Middle Ages

Fig. 6.5  Example of an SFB from Witsum/Föhr with traces of a standing loom: photogrammetric picture of Planum 1 and schematic drawing. After Majchczack and Offermann 2018, fig. 3.

The excavations proved all the rectangular magnetic anomalies to be indeed SFBs, with few differences between the anomaly and the actual archaeological feature. The SFBs are rather large, 4.5–6.2 m in length by 3.2–4.0 m wide, and up to a a metre in depth (Fig. 6.5). Their strong signature in the magnetic map derives from their filling with settlement refuse at the end of their use, containing large amounts of burnt clay. 180

Z01 Hines Book B.indb 180

07/05/2021 13:30

Early-Medieval Settlement Archaeology Some SFBs were found with a sandy filling, and these do not appear in the magnetic map. Therefore, the actual number of buildings is greater than it appears in the survey results. In all cases, the excavated SFBs show signs of craft production. They are usually constructed with two roof-bearing posts along the central axis of the rectangular pits. Another post hole in the southern area might belong to an entrance situation. In the northern part, a long shallow pit lies between two small post holes next to the roof-bearing posts. More small post holes are found behind the long pit. In the pit and next to it, we usually find loomweights or their remains. The entire construction clearly belongs to a standing loom, which occupies the northern half of the SFBs (compare Zimmermann 1982). This pattern is found in every SFB, indicating their main purpose was that of weaving huts. But there are other activities as well. At least in some SFBs, amber pieces were worked into beads. We found several unfinished beads which broke apart during the boring of the central hole. The many iron objects from the SFBs are mostly small knives, but also needles, nails and comb rivets. The conditions for the preservation of organic material are unfortunately not provided by the sandy soils, and only a few pieces of bone survived, including parts of an ornamented comb. The refuse layers which were dumped into the pits contained lots of traces of ironworking, mostly pieces of smithing slag. The best finds related to trade are the numerous glass objects (Fig. 6.6). We found sherds of high-quality glass vessels, including a blue claw beaker, globular beakers with reticella decoration and optical-blown funnel beakers. Several pieces of the same vessels indicate that they were used and broken on site rather than being imported as glass cullet for beadworking. Although no direct evidence of bead making was found, many beads from Witsum are either broken or not very well made, so they might be considered to be production refuse. Moreover, they form a perfectly datable group of finds, belonging to a mid-eighth-century group of blue beads with red-and-white decoration (Callmer 2007). These beads are the main group of product from the market place in Ribe during the eighth century and indicate strong trade ties between Witsum and Ribe. This is further emphasized by a small bead of wound gold wire, which finds a perfect parallel in the Ribe market place in the late eighth century.

Fig. 6.6  Examples of vessel-glass sherds and glass beads from Witsum/Föhr. After Majchczack and Offermann 2018, fig. 6.

181

Z01 Hines Book B.indb 181

07/05/2021 13:30

Frisians of the Early Middle Ages The settlement of Goting, about 2 km from and in sight of Witsum, is of a different structure. It is located on an elevated moraine core which encompasses a small inland depression of 8 ha. This wetland, called ‘Bruk’, has a small connection to the North Sea in the form of a tidally influenced inlet near the western end of the cliff. On the southern border of the depression, the geest rises gently to a height of 7 m and falls off to the beach, thereby forming the Goting Cliff. The coastline runs in a smooth bend from east to north-west, forming a slightly protruding position of the cliff and a bay-like situation at the mouth of the tidal creek towards the Bruk. The topography of the depression and the inlet is comparable to the Godel wetland in Witsum, making it plausible that bay and watercourse may have served as harbour sites. During the twentieth century, many archaeological finds were reported from the Goting Cliff due to erosion after storm surges. Find material gathered by private collectors and documented archaeological features in the cliff walls revealed a settlement dating to the third–fifth centuries AD. A strong storm surge in 1976 eroded the cliff at

Fig. 6.7  Magnetic map (±9 nT) of the settlement at Goting. A: Rural settlement of the Migration Period. B: Early-medieval SFB settlement at the cliff. C: Early-medieval settlement east of the Bruk. D: Coin hoard. E: Digital elevation model of Goting. After Majchczack et al. 2018, fig. 8 (Coordinates DHDN/Gauss-Krüger Zone 3; magnetic map T. Wunderlich; aerial picture © Google Maps; digital elevation model (DGM2) GeoBasis-DE/LVermGeo SH; cartography B. Majchczack).

182

Z01 Hines Book B.indb 182

07/05/2021 13:30

Early-Medieval Settlement Archaeology its western end and exposed Early-medieval SFBs and wooden wells in the cliffside, along with a coin hoard of silver sceattas (Majchczack 2015b). In the following years, the cliff walls were covered by colluvium and dunes, preventing further erosion. In the years 2013–15, the North Sea Harbour Project conducted a survey using magnetometry in the entire area around the Bruk wetland to assess the extent and structure of the archaeological site, to distinguish settlement areas, and to search for the location of the Early-medieval settlement and a possible harbour (Fig. 6.7; Majchczack et al. 2018). The magnetic map implies a large multi-phased site with three main settlement areas, each with its own characteristics. The south-eastern area towards the cliff is occupied by a rural settlement, characterized by a loose distribution of longhouses within farmyards. The settlement layout is characteristic of the later Roman Iron Age and Migration Period. The south-west, between the end of the cliff and the Bruk, is occupied by rectangular anomalies of Early-medieval SFBs, as they were found in that area after the 1976 storm surge. A third centre of anomalies is found on the eastern edge of the wetland, where a dense pattern of strong anomalies of rectangular plots indicates an area of extensive activities, a parcelled-out settlement structure and again many SFBs. The orientation of the plots towards the edge of the wetland suggests a functional relationship between this settlement and the possibly navigable Bruk. Further to the north-east, dense anomalies of farmsteads with longhouses and also possibly a younger settlement area can be found. An analysis of all the archived finds from Goting Cliff has shown that the Early-medieval inventory contains many imported goods. The local pottery comprises mostly sherds of egg-shaped and globular pots. Imported Scandinavian goods, such as whetstones of light and dark grey shist and a sherd of a soapstone vessel were also found. Another find of Scandinavian provenance might be a fragment of a rectangular plate brooch with an animal ornament. On the contrary, many finds are connected with the south, mainly with the Frisian and Frankish realms. The only imported pottery found so far is Frisian shell-tempered ware. An enamelled disc brooch of the ninth–tenth century originated in the Frankish area, as well as quernstone fragments of basaltic lava. Several glass beads in simple annular and barrel shapes of green, turquoise and light blue colour, as well as an oriental segmented bead, might also be imported from the south (Majchczack 2015b). The strongest evidence of trade activities in Goting is the silver coin hoard found in 1976. Of the eighty-seven coins, ten are Frankish deniers and fifty-eight belong to the group of Anglo-Frisian sceattas, mostly of the porcupine type. Another nineteen coins are transitional types minted in the contact zone between Frankish deniers and Frisian sceattas. The closing coin of the hoard dates to around 755; the hoard is therefore likely to have been brought to Goting early in the second half of the eighth century by a merchant from the Frisian area. It is highly unlikely that the coins were assembled from the local circulation of money, since no coins of the Wodan-monster type, which is believed to have been minted in Ribe, are included (Hatz 2001). A metal detector survey at the Goting Cliff recently yielded an equalarmed bronze brooch of the eighth- to ninth-century Domburg type, highlighting another link with Frisia. Excavations were conducted in Goting in 2014 and 2017–18 over a total of fifteen weeks. The excavation in the southern part, on the elevated geest moraine, resulted in a vertical stratigraphy with settlement features from the Neolithic to the Early Medieval Period. As early as during the Pre-Roman Iron Age, the local population had begun to 183

Z01 Hines Book B.indb 183

07/05/2021 13:30

Frisians of the Early Middle Ages deposit humus on the site, creating more than a metre’s thickness of plaggic anthrosoils (Plaggeneschboden) throughout the first millennium AD. The uppermost layers contained the Early-medieval settlement. Three excavated SFBs of the late eighth and ninth centuries were very similar in size and construction to those at Witsum, each featuring a standing loom and remains of loomweights. They contained a wide array of imported goods: glass beads (again completely similar to the beads from Ribe) with a higher proportion of imported Islamic beads from the ninth century, Scandinavian whetstones, ‘Frisian’ shell-tempered pottery, Rhenish pottery, basaltic lava and glass vessels, mostly funnel beakers. Excavations in the lower area alongside the Bruk wetland resulted in a complex stratigraphy. At the beginning of the site in the late seventh or early eighth century, the gentle slope towards the low wetland was occupied with SFBs with signs of textile production. After a short time, a parcelled system of small plots, separated by small ditches and possibly fences, replaced the SFBs. The structures are reminiscent of a market place as in Ribe, but less clear. There are traces of craft production, but far less than the amounts known from emporia. Again some Ribe-type beads occur, but mostly remains from ironworking with lots of smithing debris. Boat rivets indicate use of the area for the repair or building of boats. The site flourished for a while in the eighth century but changed completely afterwards. In the ninth century, a quite homogeneous and findpoor layer of humus soil was deposited on the site. In the late ninth or tenth century, settlement features reappear but are unclear in their structure and meaning as the site lived on into the High Medieval Period. The overall settlement structure differs much from Witsum. In Goting, no village-like settlement with designated yards is found. A loose SFB settlement was laid out on the higher elevated areas as a possible production area, while a parcelled workshop area was located on the edge to the wetland with its navigable creek. A living area with longhouses and possibly agrarian settlements lay a bit further to the east. In any event, Goting seems to be a settlement which was genuinely focused on craft and trade. Another settlement specializing in production and trade is found at Tinnum on the island of Sylt. Close to a Migration-period settlement, an SFB settlement of the seventh–tenth centuries was excavated in 2006–7 (Majchczack and Segschneider 2015; Segschneider 2008, preliminary report). The settlement has much more the characteristics of a market place or a workshop settlement. Besides the SFBs and wells, no settlement structures such as longhouses or clear divisions into plots were observed. The SFBs were rich in remains of craft production. One SFB group contained many fragments of imported glass vessels in unusually bright colours. The vessels had been brought to Tinnum as cullet and directly worked into glass beads – chemical analysis show the direct link between raw material and product. The bead assemblage is somewhat unusual in relation to the known spectrum of eighth-century beadmaking in the southern Scandinavian region and is believed to be relatively early. The beads do not resemble the typical dark blue glass or polychrome decoration, but form predecessors to the later products. The dendro-dating of the well to the mid-seventh century supports this early date for craft production on the island. The glass vessels might partly be from the Rhineland, while the very vivid blue glass might even be an Anglo-Saxon product. The most striking find material was found in a refuse layer in a single SFB – more than 3 kg of raw amber in small and large pieces, together with cutting debris, and broken and complete beads and pendants. The amber must have been so easily 184

Z01 Hines Book B.indb 184

07/05/2021 13:30

Early-Medieval Settlement Archaeology available on the island’s beaches that the beadmaker could discard any piece which was not perfect for his needs (Kordowski 2013). The discarding of such a great amount of amber might also be seen as ritually motivated, as there are several indications of ritual practices and depositions within the SFBs (Segschneider 2012/13). Apart from the glassand amberworking, the SFBs were used for textile production. The imported goods are a bit scarcer: only a single piece of imported Rhenish pottery of the late seventh century. All the other pottery is local. Additionally, there are lots of basaltic lava and some Scandinavian whetstones. Smithing slags testify to ironworking and boat rivets might again indicate the repair of boats. Most of the archaeological features are likely to date into the seventh–eighth centuries, but the deposition of a small necklace of later lead-glass beads and the finding of a possibly Borre-style equal-armed brooch indicate a longer duration of the settlement.

Signs of power When looking for traces and manifestations of power and control, the search is quickly satisfied on the North Frisian islands. The most impressive ringforts at Borgsum/Föhr and Tinnum/Sylt are the largest and most dominant monuments, with dimensions of c. 140 m in diameter and a wall height of up to 10 m (Fig. 6.8). From the top of the walls of the Borgsumburg, the entire island of Föhr can be viewed with the described settlements in the marsh and on the southern coast in sight. Small excavations inside the Borgsumburg in the 1950s and geophysical surveys in 2001/2003 found a systematic radial arrangement of small longhouses with clay-sod walls, their doors facing the centre and connected by a circular wooden causeway. The buildings are all constructed in the same way at the same size. The central area is occupied by a well or waterhole. The original idea of the 1950s, that the ringforts might belong to the tenth-century Trelleborg type, was quickly abandoned due to the completely different arrangement and type of the buildings and the much earlier finds: the pottery finds comprise local egg-shaped pots, wheel-thrown pottery from Ribe and Haithabu, and Pingsdorf Ware. Especially the presence of Ribe wheel-thrown pottery puts the beginning of the ringfort back at least into the second quarter of the eighth century (Segschneider 2009). The main question is: who built the ringfort and for what purpose? It is most unlikely to have served as a secure refuge for the local population in times of threat. The systematic building arrangement is more reminiscent of a military garrison. The find material provides further evidence: boat rivets and wooden toggles from a boat’s rigging indicate seafaring inhabitants. Their diet contained a lot of seafood, including large mammals like harbour seal (Phoca vitulina), grey seal (Halichoerus grypus) and harbour porpoise (Phocoena phocoena). Furthermore birds like different types of duck and goose but also many remains of the common guillemot (Uria aalge, Trottellumme), which can only be found on the rocky coast of Helgoland (Schmölcke 2009). It is evident that the inhabitants of the Borgsumburg were not part of the local, permanent population but rather were posted on the island. They were not very much involved in agrarian production, but relied on gathering food from elsewhere. The ringforts were obviously an instrument for controlling and protecting the islands. It is remarkable that the same structure occurs on both islands Föhr and Sylt (small-scale excavations have also been carried out at the Tinnumburg, with comparable results to the Borgsumburg: Harck 1990; Segschneider 2009). In both cases, the ringforts had 185

Z01 Hines Book B.indb 185

07/05/2021 13:30

Frisians of the Early Middle Ages

Fig. 6.8  Aerial photograph with a view of the Borgsumburg in the foreground by the southern coast of Föhr. The settlement site at Witsum can be seen to the back, the dotted line indicating the connecting rampart. After Majchczack et al. 2018, fig. 13 (aerial photograph: E. S. Mauritsen, Holstebro Museum; drawing: B. Majchczack).

direct and navigable access (at least with boats) to the North Sea via tidal creeks and they are built in the direct vicinity of important settlements with an emphasis on trade. The Tinnumburg controlled the access to the close-by marketplace of Tinnum; the Borgsumburg was connected to the settlement at Witsum via a rampart. Therefore, all indications concerning the ringforts’ inhabitants hint at outsiders, who acted as a controlling power over the islands. In this light, the ringforts should be regarded as a possible part of the Danish Dannevirke system, which forms the border along the Eider-Treene-Schlei 186

Z01 Hines Book B.indb 186

07/05/2021 13:30

Early-Medieval Settlement Archaeology between the Danish realm to the north and the Saxon and Slavonic areas to the south. The North Frisian coastal zone must be seen as a border area north of the Dannevirke, so that a Danish interest in control is conceivable (as discussed by Kühl and Hardt 1999 and Eisenschmidt 2004). Whether Scandinavian/Danish sovereignty was exercised over the North Frisian islands and its inhabitants or if it was more a partnership has yet to be discussed and verified by further archaeological evidence.

Later on: thoughts on the tenth/eleventh centuries and Christianization Beyond the ninth century, the archaeological record grows thin on the North Frisian islands. The settlement excavations have yielded little or no finds from that period. It is most likely that the large geest settlements at the coast moved a little into the locations of the present villages. A similar settlement movement can be observed in the survey data from Goting. Unfortunately, no excavations have been done within the modern villages to shed light on their earliest dating. In Eiderstedt, the large wurts such as Elisenhof continued with little change in settlement organization into the eleventh century. Another limiting aspect is the burial sites. The burial habit of cremation with urn burials under small mounds was phased out sometime in the ninth century and the large sites with burial mounds in Nebel/Amrum, Morsum/Sylt, Goting and Hedehusum/Föhr seem to end. The problem is that the excavation methods applied at the burial sites provide no evidence for flat graves and the shift to inhumation graves cannot be investigated. Despite the missing settlement and burial finds, the North Frisian islands still played an important role in the region. In the tenth century, a new class of high-grade finds appears: large hoards of silver objects. Hoards of coins, hacksilver and jewellery from List/Sylt, Westerland/Sylt and Utersum/Föhr illustrate an inflow of wealth, and connections with the Scandinavian power (Wiechmann 1996). The recently found tenth-century hoard of Morsum/Sylt contained a silver and gold ring brooch of highest quality in Scandinavian style (Siegloff and Tummuscheit 2017), probably belonging to a high-ranking person with connections to the Danish kingdom (M. Segschneider, pers. comm.). In this light it is noteworthy that the inhabitation of the ringforts and the power they asserted continued into the eleventh century. The Christianization of the North Frisian islands is a topic without much reliable information. The end of the urn burials in the ninth century may be related, but there are no archaeological data to characterize and date the transition. The Christianization of nearby Ribe began in 860 with Ansgar’s mission; his impact on the islands is unknown. Evidence for early North Frisian churches dates to the late eleventh and early twelfth centuries (Panten 2010, 71); the churches on the islands were built from the late twelfth century onwards. The oldest church is St Laurentii in Süderende/Föhr, its first phase constructed in granite blocks (Granitquaderkirche). Since no archaeological excavations have been carried out in any of the island’s churches, we have no know­ ledge about earlier buildings.

North Frisian trade The recent excavations have greatly expanded our knowledge about the trade in North Frisia and also about the connections between the islands and the large emporia of the time. It is clear that the inhabitants of Early-medieval North Frisia were a maritime, 187

Z01 Hines Book B.indb 187

07/05/2021 13:30

Frisians of the Early Middle Ages seagoing people. The natural landscape just demands it: the islands, being the most important settlement area, are only accessible by boat. The North Frisian mainland is a much less densely settled hinterland, access is difficult due to extensive fenlands bordering the geest. Evidence for boats can be found on all sites with access to waterways: rivets in Tinnum, Goting and Borgsumburg; wooden parts in Alt-Archsum and Borgsumburg. The settlements were connected by waterways, and natural harbours have been confirmed in Witsum and Goting, while all other sites have similar natural features which would have allowed shipping. The find material provides a wealth of evidence for the trade in the late seventh– ninth centuries. During this time, the North Frisians were mainly involved in seagoing trade along the North Sea coast. The islands can be regarded as a stop on the way between the southern coastal areas and further north to and from the emporium of Ribe. The islands’ trading sites act as hubs for local trade and production, channelling the trade towards the major emporia and distributing goods to and from the hinterland and minor sites. Wheel-thrown pottery, quernstones of basaltic lava and intricate glass vessels came from the Rhineland, most certainly via the emporium at Dorestad. This is the best evidence for the often discussed ‘Frisian trade’. Also the sceattas/denarii hoard from Goting is believed to have been the possession of a trader from the Frisian-Frankish coastal area and not that of a local islander. Another commodity from the southern North Sea coast is the shell-tempered pottery. The main distribution area of this ware is the East Frisian peninsula where it accounts for just about all of pottery in the ninth century, and lower percentages in the West Frisian areas (Stilke 2001, 195). Local production on the North Frisian islands is highly unlikely. Scandinavia is highlighted by ornamental metalwork and whetstones of Norwegian shist. The ties with the emporium of Ribe are especially strong, as shown by the eighth-century bead material from Witsum. It is most likely that Ribe was a preferred place to buy foreign goods and to sell products from the North Frisian islands: most likely the woollen textiles, which could represent the historically known pallia fresonica, the Frisian cloth (Ellmers 1972; Siegmüller 2010). From the ninth century onwards, the importance of Ribe fades, and the west–east connection along the Rivers Eider and Treene to the emporium of Haithabu dominates the trade. Especially the settlements on Eiderstedt like Elisenhof profit from their location along the shipping route, but the North Frisian islands also start to show more imported goods from the Baltic. The silver hoards of the tenth–eleventh centuries illustrate far-reaching Frisian contacts across different regions in the North Sea and Baltic (Wiechmann 1996).

Frisian identity? The results of the settlement excavations offer little reliable information about the question of whether the people inhabiting the North Frisian islands and the coastal zone were indeed Frisians or felt part of a Frisian identity. The structures of the geest settlements show many parallels with settlements in the Lower Saxon coastal region, but their elements are not exclusively found there. The settlement layout of Witsum shows strong parallels with Early-medieval villages all over southern Denmark. At the same time, similar combinations of ring fortresses and SFB-dominated trade settlements appear in the ‘Saxon’ Elbe-Weser Triangle, as shown by excavations in Cuxhaven-Altenwalde and Cuxhaven-Sahlenburg (Eichfeld 2017; Offermann et al. 2019). The North 188

Z01 Hines Book B.indb 188

07/05/2021 13:30

Early-Medieval Settlement Archaeology Frisian wurt settlements such as Elisenhof follow a general pattern which is largely designated by their natural landscape – they start out as level-ground settlements on elevated embankments and are continuously raised with each new house generation and thus adapt to changes in the environment as for rising sea-levels. Similar wurts are found all along the ‘Frisian’ marshlands of Lower Saxony. Since most of the existing marshland was cultivated from the High Medieval Period onwards, only a few wurt settlements in Eiderstedt are predecessors of modern villages. Interestingly, the special pattern of settlement organization in Elisenhof, consisting of rows of parallel, narrow farmyards and separated by roads and fences, shows strong parallels to the later village layouts on the island of Föhr (Bantelmann 2003, 72) and might reflect a North Frisian habit of organizing agrarian settlements later in the Early Medieval Period. When tracing the connections of trade and communication, the find material in the settlements puts the North Frisian area strongly into the network of ‘Frisian trade’ between Dorestad and Ribe, all along the North Sea coast. The imported goods from the Rhineland and maybe even from England and the strong ties with the emporium in Ribe in the eighth century illustrate this trade route and the connections with the ‘Frisian’ coastal zone, but cannot serve as a proof for the identity of the North Frisian traders. The burial practices provide the best way to distinguish the ‘Frisian’ population from their neighbours. In an analysis of all Early-medieval burials in the historical region of Schleswig between the rivers Eider and Kongeå, the North Frisian coastal zone clearly stands out from the Scandinavian areas to the east (Eisenschmidt 2004, 293–300 with an ethnic analysis of the North Frisian burials). The habit of cremations with urn burials is only found along the coast on the North Frisian islands, Eiderstedt, the North Frisian mainland and in smaller numbers along the southern Danish coast up to Esbjerg. The urn graves show parallels to the burial sites in Lower Saxony, the most important combining elements are the egg-shaped pot, urn graves with surrounding post holes and ring ditches, as well as a certain array of grave goods, including needles, combs, keys, needle cases and knives. The small equal-armed brooches of the Domburg type are numerous in the West Frisian areas and can also be found in North Frisian urn graves, as well as on the settlement sites of Witsum and Goting. Similar urn graves are unknown in the Saxon settlement areas in Lower Saxony and Holstein and in Denmark, further afield from the western Wadden Sea coast. It might therefore be valid to speak of a Frisian population on the North Frisian islands. From the ninth century onwards, grave goods of Scandinavian origin are found in urn graves, mostly tortoise brooches. At the same time, cremation graves without urns (Brandschüttungsgräber) appear which might also show a Scandinavian influence. It is conceivable that the North Frisian population was mostly Frisian at first and more mixed with Scandinavians from the end of the eighth century onwards. The same can be observed in the emporium of Ribe, which shows some Frisian influence in the eighth century, including urn burials. In the second part of the eighth century, the egg-shaped pot disappears and is replaced with Jutish forms and the artisans produce mostly objects in Scandinavian style.

Conclusions Recent discoveries have shed a new light on the Early-medieval settlements in North Frisia. While knowledge of the settlement history from the late seventh century 189

Z01 Hines Book B.indb 189

07/05/2021 13:30

Frisians of the Early Middle Ages onwards was very limited and mostly depended on the burial sites and the excavation results of the wurt settlement of Elisenhof in Eiderstedt, the geophysical prospections and excavations on the islands of Föhr and Sylt have added a new type of trading sites and a better understanding of the early phase of the Frisian re-settlement of the North Frisian islands. Following a 150-year-long hiatus after the migrations of the fourth and fifth century, new settlements appear in the second half of the seventh century. Located on the edges of the islands’ high moraine cores and with access to navigable waterways, trading sites at Tinnum/Sylt and Witsum and Goting/Föhr are founded. The sites are occupied by local craftsmen and traders, who engage in the trade networks along the North Sea coast. A surplus production of wool textiles and worked amber was presumably the basis for the exchange for foreign goods such as silver coins, vessel glass, glass beads, quernstones, wheel-thrown pottery or whetstones. The trading sites were thriving in the eighth and ninth centuries and established strong ties with the emporium at Ribe in southern Denmark as well as with the Frisian areas in the south. The ringforts of Borgsumbug and Tinnumburg protected the trading sites and might be seen as Danish military garrisons, asserting power over the islands. The settlement pattern changes during the tenth/eleventh centuries and the archaeological record becomes less visible, but the emergence of large silver hoards demonstrate the continuous involvement in the trade networks. Throughout the Early Medieval Period, the inhabitants of the North Frisian islands were a seafaring people who were embedded in the North Sea coastal area. Trade connections, burial habits and settlement layouts connect the islands both with the core ‘Frisian’ southern North Sea coast and with southern Jutland, probably resulting in a local identity founded by immigrating Frisians and fusing cultural elements from the neighbouring areas.

Discussion NIEUWHOF   I wonder about time-depth at the site of Goting/Föhr. How old may the habitation have been? You showed us the oldest houses from the Roman Period along the cliff. Is there a shift in habitation inland and do you think there was an older phase that was eroded away? MAJCHCZACK   The habitation at Goting covers all periods as far back as the Bronze Age, but the important question concerns the continuity from the Migration Period into what we know of as the Early Medieval Period. It is hard to answer, but the Migration-period single farmyard structures seem to be dispersed along the geest edge and further inland. The Early-medieval settlement layout is quite different, with a clustered area of SFBs and no sign of longhouses. I wouldn’t be sure if there was continuity from the fifth–sixth centuries into the eighth century onwards. But the archaeological preservation in Goting is remarkable because of the thick layers of plaggic anthrosols. Large-scale excavations would probably show connexions, but maybe not continuity. NIEUWHOF   Do you have eroded-out sherds from earlier periods on the beach? MAJCHCZACK   Yes, this is what has come in from earlier private collectors. There are significant quantities of third- to fifth-century pottery, and occasionally also from even earlier in prehistory.

190

Z01 Hines Book B.indb 190

07/05/2021 13:30

Early-Medieval Settlement Archaeology HINES   A factual question to start with. On the geophysical survey plot of Witsum there seemed to be some narrow strips running between the lane and the sea, the waterfront. Are those a feature that is recognizable, and anything of significance? MAJCHCZACK   No, they are nothing. The stripes are a most striking element in the plot, but the excavations proved them to be near-surface remains, probably from High-medieval ploughing. HINES   It is, though, of some interest in its own right that the area became part of a laid-out field system by the High Middle Ages.    As a second question, or rather an observation or suggestion: finds from both Witsum and Goting are very similar to finds from Ribe. In light of the apparently seasonal character of the earliest occupation at Ribe, and the persistent idea that Ribe is somehow ‘Frisian’ in layout, shouldn’t we see the link between these areas as something stronger than just ‘contact’ and exchange of materials? For instance, that the people who lived at these settlements were also those who, seasonally, were in Ribe? MAJCHCZACK   Yes, indeed. It is only a short hop of one or two days by boat to get there. There are several sites with very strong connexions with the eighth-century marketplace at Ribe. I do believe that the people from the North Frisian islands are directly involved in Ribe. HINES   What makes Ribe different is its considerable agricultural hinterland; while Ribe also stands out because it was a production centre for a lot of Scandinavianstyle material. It is possible to see how that was being distributed in the hinterland. WOOD   I am interested in Helgoland. Is it Frisian or Danish? This is an important historical question, because Vita Willibrordi is very unclear. Do you think it was under Danish control? MAJCHCZACK   That would take it too far, but we do see the ringforts on the North Frisian islands as having been Danish. The people or troops stationed at these ringforts were definitely travelling to Helgoland on a regular basis, as shown by the find material. This could indicate either that Helgoland was Danish or just that the relationships in respect of Helgoland were good. Unfortunately, there is no Early-medieval archaeological evidence from Helgoland itself, due to the massive destruction caused by twentieth-century fortification works and the bombing during and after World War II. IJSSENNAGGER-VAN DER PLUIJM   There may also not necessarily be such a distinction between Frisian and Danish. There is a sense that in some of these periods they could be within a similar power spheres or even a shared sphere between Frisian and Danish to a certain extent. DE LANGEN    When would that have been? IJSSENNAGGER-VAN DER PLUIJM   I would say at the start of the Viking Age, so around AD 800, but also in a new light in the late ninth or tenth century. DE LANGEN    Then, of course, we ask, which Frisians? NIJDAM   A response to Ian, because Vita Willibrordi is very problematic. You say it’s not clear whether Willibrord is brought before Ongendus or Radbod. WOOD   The text isn’t clear. NIJDAM   But he goes to the border between the Danes and the Frisians. WOOD   Yes, but doesn’t say which side of it he is on. The previous chapter has been about Ongendus, but Willibrord could have crossed the border. 191

Z01 Hines Book B.indb 191

07/05/2021 13:30

Frisians of the Early Middle Ages NIJDAM   To have Radbod’s power stretching all the way to Helgoland, and even to North Frisia, is problematic. It would be the sole evidence for that. IJSSENNAGGER-VAN DER PLUIJM   Though there are much later sources that make Radbod a Danish king. NIJDAM   Another question to Bente: I thought that Frisians in Ribe were characterized by more Frankish material culture, and signs of Christianity. Or is that more in the ninth century? MAJCHCZACK   It is rather earlier. The Frisian presence in Ribe is mostly indicated by graves. The very early material produced in Ribe is connected more with the southern North Sea area, and then it becomes more Scandinavian. DE LANGEN    What is the date of that? MAJCHCZACK   Already in the eighth century. NIJDAM   Is this material (referring to the geomagnetic image of Witsum) from the seventh century? MAJCHCZACK   The settlement at Witsum also starts in the late seventh century. We have two dendro-dates from around AD 697. Most of the excavated features date into the eighth century and some are ninth century, but nothing later than that. There is one detector find, though, of an eleventh-century German silver coin. The general settlement layout with plots can be dated within the eighth century. IJSSENNAGGER-VAN DER PLUIJM   Just to quickly add, recent discussions with Danish colleagues who came to visit Egge (Knol) and myself also showed that what often has been thought of as typically Frisian in material culture in Denmark, such as folding knives, are not necessarily seen as typically Frisian here, as far as we were able to tell. HINES   The case for the ‘Frisian’ character of Ribe in the eighth century is very much based on the layout and structure of the site, and that in turn is strongly dependent on the old interpretations of Emden in particular. There’s a possibly circular argument. Skre argues for a Frisian household amongst the excavated plots in Kaupang in the early ninth century, but that actually means having a close parallel in Dorestad, and the terms Frankish and Frisian are almost interchangeable (Skre 2010). NIJDAM   What about the frisa kiltar, Frisian guilds? When does the runestone that appears on date from? IJSSENNAGGER-VAN DER PLUIJM   It is of the eleventh century (U379: Sigtuna, Uppland, Sweden; cf. also U391, an incomplete inscription from the same town). DE LANGEN   You have been careful in using the term market place. You also talked about craft activity, and so talked about both a workshop area and a market area. Why the latter? MAJCHCZACK   I think we were stuck with the old idea about a market zone. The settlement layout did seem comparable to Ribe, which is characterized that way. But it can just as well be a workshop area. DE LANGEN   We could agree on a workshop area then. A further question on the start of the ringforts: initially you dated them to the eighth century, but then you said possibly also as early as the seventh century. MAJCHCZACK   Yes, we don’t really know exactly when they start, since the excavations were very limited. The earliest datable material found at the Borgsumburg/Föhr is Ribe wheel-thrown pottery from the second quarter of the 192

Z01 Hines Book B.indb 192

07/05/2021 13:30

Early-Medieval Settlement Archaeology eighth century. The seventh century is a question mark. At Tinnum/Sylt, though, there seems to be a direct relationship between the settlement and the ringfort, so it’s conceivable they go back to the seventh century. DE LANGEN   You also mentioned a switch to the Danish sphere in the eighth century. You see these ringforts, I think, as firstly oriented to the southern North Sea area and subsequently to the Danish power. MAJCHCZACK   I think what might have existed is a Danish sphere of power and influence which is represented in the defensive system of the Dannevirke. We are not sure how well established that system was by the seventh century. The main rampart of the Dannevirke near Hedeby does have much earlier phases than previously thought, but these early phases are not well understood. It is possible that at first people had been moving into the North Frisian islands in the seventh century, whereupon the Danes decided they should be within their sphere of power and looked to assert that in some way – possibly by agreement. Or perhaps that was a later development, and Danish rule was already present. I think the ringforts, though, were under Danish control from the beginning. DE LANGEN    But there were other people living there? MAJCHCZACK   I think at least those people in the ringforts should have been Danes. DE LANGEN    So where are the Frisians living? MAJCHCZACK   They would form the larger part of the overall population of the islands, living in the agrarian and also in the trade settlements. NICOLAY   Were these Frisians from the south? MAJCHCZACK   I suppose so. NICOLAY   A current Masters thesis is deconstructing the idea of an early Frisian settlement because of lack of evidence, and then looking at the traditional idea of two waves of settlement: one in the seventh century and a later one in the tenth/ eleventh century. The first is deconstructed, but the second appears true. MAJCHCZACK   So is it arguing that there was no immigration in the seventh century? NICOLAY   The pottery shows no evidence for a migration wave in the seventh century from the southern part of the North Sea area to the east, especially because no regional differentiation can be made for ‘Hessens-Schortens Ware’, which is found in similar shapes all along the North Sea coast. The only ‘truly Frisian’ find is a tremissis of Dronrijp type from Klappholtal (Sylt), which, of course, is no evidence for migration on its own. MAJCHCZACK   It is difficult for me to talk in terms of Frisians and Danes. We don’t normally do that. What we do find is a local population which appears in the seventh century after a 150-year period without any datable finds. The people in the ringforts are obviously different and could be identified with the Danes. VERLSOOT   Is this difference shown by different ways of living? MAJCHCZACK   It is the fact that they live in the ringforts and are not supplied by local farming. It seems to be a troop or garrison that is posted there. VERLSOOT   A thought on this from linguistic evidence: the dialects in the area and on Helgoland are one group. They are very similar to the rest of Frisian but differ on certain points. There are various hypotheses historically; the present consensus is that these are early Frisian dialects from the south, which must have come in 193

Z01 Hines Book B.indb 193

07/05/2021 13:30

Frisians of the Early Middle Ages from about the seventh/early eighth century. There is no evidence that suggests any earlier branching from West Germanic. The older linguistic evidence does suggest that the new population of the seventh century is made up of people from the south. There is early evidence of contact with Scandinavian, but it is not strong. Helgoland is closely related to Sylt, dialectally, but these may be Late-medieval innovations and relationships. The second wave too is absolutely Frisian, with evidence of developments that had taken place in the Ems region.    One more question about the fort. There are also forts on Texel, Zeeburg etc. Are those completely independent? KNOL   One of them is under a local lord. But Texel could have a Viking/Norse origin. That doesn’t account for Deventer. IJSSENNAGGER-VAN DER PLUIJM   I think there are also differences in date, shape and size. And there has been a lot of debate around whether they could be Danish, Frankish or Frisian. KNOL   Letty ten Harkel (2013) has written on these in Zeeland. IJSSENNAGGER-VAN DER PLUIJM   And then there are also possibly related forts in the Flemish border areas. The latest interpretation that Dries Tys, Pieterjan Deckers and Barbora Wouters have put forward is that these may be more continuous ways of expressing local power in that particular area (Tys et al. 2016). The concept of a ringfort as such is quite a general one, I suppose, and there are various schools of thought on how to interpret them. KNOL   Looking at the map of ringforts on the edge of the geest and the marsh: you also have a symbol at Jever – is that a ringfort? MAJCHCZACK   No, the Woltersberg at Jever probably developed from a terp and has the form of a motte, but it dates from earlier, most likely in the ninth or tenth century. KNOL   Is it a castle? MAJCHCZACK   It is not well understood and no excavations have been carried out yet, but we expect there was some sort of fortification there. KNOL   I am struck by the similarity in the locations of Groningen and Jever. Groningen is also at a distinct point where the geest of Drenthe meets the marsh area, and there is a river on either side. This is the explanation for why the city grew up here. From the eighth century onwards, Groningen was royal domain. I think this is probably very similar. Later development has obliterated archaeological traces. MAJCHCZACK   In Lower Saxony the forts are all located on crossroads. They are situated on the geest edge with access to the marshlands, the open sea and main roads into the inland. This gives access to resources as well as to strategic transport routes. KNOL   I was also wondering about the erosion cliff: do you have any idea what has been lost? You showed us the area that has been studied on Sylt and Föhr. Is there another area to the south? MAJCHCZACK   I don’t believe that much of the geest of Föhr has been lost. The Goting site lies on a geest outcrop that has eroded somewhat, but there are preserved Bronze-age surfaces with peat and forest remains in the mudflats close by that prove a limited extent of the Goting geest area. I don’t think that more than maybe 200 m has been lost to erosion. Since land survey began, less than 100 m 194

Z01 Hines Book B.indb 194

07/05/2021 13:30

Early-Medieval Settlement Archaeology has been lost. Apart from that, there is no geological evidence for any larger geest areas to the south of Föhr and Sylt that have been lost since the Early Middle Ages. KNOL   The cartographic situation in North Frisia is very different from that of our terpen area, and looks much more like North-Holland. MAJCHCZACK   For the geest islands, though, we are very certain that these were really islands surrounded by water and the Wadden Sea. Towards the south, in the area of the marsh islands of Pellworm, Nordstrand and the Halligen, there were large peat areas that were uninhabitable apart from the highest marsh ridges close to the sea. South of that, the area of Eiderstedt was partly inhabitable, especially along the Eider estuary. NICOLAY   I would like to return to the issue of how Frisian North Frisia is. It is outside the area where Lex Frisionum was operating. I think Arjen’s comment on the language evidence is important because I had suspected this was a circular argument, fitted to the narrative of a seventh-century Frisian arrival. VERLSOOT   That date comes from the archaeology; the linguistic arguments are independent. NICOLAY   But the archaeology is moving towards the eleventh/twelfth century, the traditional second wave of ‘Frisian’ migration. VERLSOOT   That probably applies to the second wave of settlement, doesn’t it? There is a first wave in the seventh/eighth century. These two waves are linguistically secure. NICOLAY   The question is, was there a new population, and if so were these people ‘Frisians’? And even where they must have come from. KNOL   Why should they be Frisians? If you review the archaeological evidence, you can identify at most some of the pottery as Frisian. DE LANGEN   …and a hoard of sceattas. NICOLAY   There is the one Frisian tremissis from Sylt that I mentioned before. MAJCHCZACK   As Gilles said, there is also the hoard of sceattas from Goting. The numismatic analysis demonstrates that it was not taken from local currency circulation, but must have been brought from the Frisian-Frankish contact area. KNOL   If I look at the pottery I find the evidential basis thin. I am not clear why this pottery should be Frisian. It makes more sense that when the peat area was reclaimed, then practices that were already going on for some time in the northern Netherlands spread with people to North Frisia. I wonder how closely you can date the changes in dialect? VERLSOOT   The dates of the linguistic evidence (e.g. runic inscriptions) come from the archaeology. What is secure is that the dialects of the North Frisian islands can only be Frisian; they are not closely related to other North Sea West Germanic languages. And they are more archaic than the later settlement on the North Frisian mainland. KNOL   Is it like comparing Afrikaans to Dutch? I see the former as rather old-fashioned. VERLSOOT   The dialects show that there was a split in the Frisian language family earlier than the second wave of settlement. KNOL   But why earlier than the second wave? VERLSOOT   Because there are linguistic features that can only be earlier. As in archaeology, we work with sequences of change which fall into an order, and that 195

Z01 Hines Book B.indb 195

07/05/2021 13:30

Frisians of the Early Middle Ages can be reconstructed. This gives us a very early split. An article by Nils Århammar (1995) can be recommended. KNOL   I wonder how secure the dating of the branching is. VERLSOOT   The dating is relative. KNOL   Do you have enough evidence of the Frisian spoken in Dutch Frisia and North Frisia in the eleventh century? VERLSOOT   It is largely based on reconstruction because the earliest written evidence is late thirteenth century. But the comparative method enables you to go back in time to reconstruct relative chronologies. There are typical features of the North Frisian coastal dialects, from the second wave, that are paralleled in the Ems region. These are later changes to other ones, which are found throughout Frisian. Wider changes are usually earlier. MAJCHCZACK   To go back to the archaeological evidence for the first and second waves, we can definitely see great discontinuity between the fifth and the seventh century. It doesn’t mean there must be complete abandonment. The material culture is probably best looked at from the point of view of the burials. In the late seventh and eighth centuries, the burials show very little Scandinavian influence, but rather have pottery and metalwork which links with the southern North Sea zone. KNOL   But the southern North Sea area is much larger than Frisia alone. VERLSOOT   In the Early Middle Ages, the entire southern North Sea area on the Continent is Frisia at that time, from the Sincfal to the Weser. MAJCHCZACK   It is hard to grasp these things of course. But it is clear that there was discontinuity, and a new population moved in, mostly, apparently, not from Scandinavia. NIJDAM   …and they speak a Frisian language, so where else can they come from? MAJCHCZACK   I don’t know how many people we are talking about. Parts of North Frisia were not heavily settled in the early phases: only the geest islands and parts of Eiderstedt, and there are some settlements on the mainland. It doesn’t have to be a large population to start with. And something new happens in the tenth/eleventh century, when the population became much larger. VERLSOOT   Where is the problem? This sounds clear; and then we have the linguistic evidence too, which points to a source in the southern North Sea, which is not contested by the archaeology. As a further remark about the conclusion that the island dialects are southern Frisian, which is only a couple of decades old: in fact, the previous idée fixe was that the languages there had been there all the time, as an independent branch of West Germanic. WOOD   To return to the previous discussion on the island of Helgoland: it is clear in the case of Vita Willibrordi, that no rex is named in the chapter, but the logic of the story must be that the king he was brought before was Radbod. It is not clear who controls the island. IJSSENNAGGER-VAN DER PLUIJM   Two things related to that point: there might be historical or literary sources related to the point that can give a bit more context, although it’s not easy to pinpoint, that could help understand how the area could have been both Frisian and Danish. Perhaps we can talk more about that later. The other point is that you mentioned that North Frisia was not subject to the Lex Frisionum. I wonder, does that matter? If we assume people moving from core 196

Z01 Hines Book B.indb 196

07/05/2021 13:30

Early-Medieval Settlement Archaeology Frisia to North Frisia and mingling with a Danish population, then I don’t see why it would have to be included in the Lex Frisionum in order to have a Frisian character of some form. HINES   We may make sometimes naïve assumptions about what the term ‘Frisian’ even means. But one thing we can be very sure about is that from some time in the eighth century onwards there was a kingdom of the Danes, and it had a territory. One of the points about Lex Frisionum as a Carolingian document is that the Carolingian government would not be issuing legislation for use within the kingdom of the Danes; you could actually read that point positively within this context. FLIERMAN   Returning to your archaeological research on the North Frisian communities: is it possible to estimate the population numbers of these settlements? MAJCHCZACK   No. I can’t make up a number which I haven’t really thought about. There’s a problem in that the geophysical plots show an overview of all of the phases, and you would have to excavate and date everything to be able to phase the site. At Witsum at least we can see that the eighth-century settlement does look like a village with yards and houses, but I can’t put a number to that. IJSSENNAGGER-VAN DER PLUIJM   Could it have been partly seasonal? MAJCHCZACK   I don’t think so. The settlement is definitely permanent. In Goting I would not be so sure. Witsum is a partly farming, craftworking and to some degree trading settlement where people live and work. FLIERMAN   Building on the previous discussion regarding the incongruities between the archaeological and the linguistic evidence: is this necessarily problematic? Do they need to align? Would it not have been possible that people who were distinguishable through their language were indistinguishable in their material culture? VERLSOOT   But as I see it, the material culture does fit at least in some way with the southern North Sea area; while later on this area becomes more Scandinavian. This fits with settlement and acculturation. The dialects are full of Danish loanwords and shared phonological developments etc. But they did not give up their language. FLIERMAN   I was concerned that, archaeologically speaking, there is a North Sea material culture, but nothing more specific than that. NICOLAY   The problem then is to say that these people came from the northern Netherlands and specifically from Friesland, and that is not necessarily the case. I do find the term ‘Frisian’ problematic. KNOL   In East Frisia, in the marsh area and especially around Ems, there is also a gap in the sixth century, and they believe there was depopulation of this area. I am not confident about this, but it is still something that is proposed. NICOLAY   The sixth-century discontinuity in East Frisia and the Elbe-Weser area is also disputed now. There are, for instance, the Sievern finds in a supposedly empty zone. Your extensive settlements in North Frisia look quite exceptional in relation to Dutch evidence. I wonder how these settlements are related to the wider settlement pattern. Presumably there were smaller settlements here too? I could see similar sites as centres.

197

Z01 Hines Book B.indb 197

07/05/2021 13:30

Frisians of the Early Middle Ages

References (including Discussion) Århammar, N. 1995, ‘Zur Vor- und Frühgeschichte der Nordfriesen und des Nordfriesischen. In Friesische Studien II. Beiträge des Föhrer Symposiums zur Friesischen Philologie vom 7. –8. April 1994, eds.V. F. Faltings, A. G. H. Walker and O. Wilts (Odense), 63–96. Bantelmann, A. 1955, Tofting. Eine vorgeschichtliche Warft an der Eidermündung. OffaBücher 12 (Neumünster). — 1975, Die frühgeschichtliche Marschensiedlung beim Elisenhof in Eiderstedt. Landschaftsgeschichte und Baubefunde. Elisenhof Band 1 (Bern–Frankfurt am Main). — 2003, Nordfriesland in vorgeschichtlicher Zeit. Durchgesehen und ergänzt von Martin Segschneider. Geschichte Nordfrieslands 1 (Bredstedt). Behre, K.-E. 1976, Die Pflanzenreste aus der frühgeschichtlichen Wurt Elisenhof. Elisenhof Band 2 (Bern–Frankfurt am Main). — 2003, ‘Eine neue Meeresspiegelkurve für die südliche Nordsee. Transgressionen und Regressionen in den letzten 10.000 Jahren’, Probleme der Küstenforschung im südlichen Nordseegebiet 28, 9–63. Bokelmann, K. 1988, ‘Wurten und Flachsiedlungen der römischen Kaiserzeit. Ergebnisse einer Prospektion in Norddithmarschen und Eiderstedt’. In Norderheverprojekt 1. Landschaftsentwicklung und Siedlungsgeschichte im Einzugsgebiet der Norderhever (Nordfriesland). Offa-Bücher 66, eds. B. Higelke, D. Hoffmann, H. J. Kühn and M. Müller-Wille (Neumünster), 149–62. Brandt, K. 1994, ‘Archäologische Quellen zur frühen Geschichte von Emden’. In Geschichte der Stadt Emden von den Anfängen bis 1611. Geschichte der Stadt Emden 1. Ostfriesland im Schutze des Deiches 10, K. Brandt, H. van Lengen, H. Schmidt and W. Deeters (Leer), 2–57. Callmer, J. 2007, ‘Blue, white and red’, Archaeologia Polona 45, 85–98. Dübner, D. 2015, Untersuchungen zur Entwicklung und Struktur der frühgeschichtlichen Siedlung Flögeln im Elbe-Weser-Dreieck. Studien zur Landschafts- und Siedlungsgeschichte im südlichen Nordseegebiet 6 (Rahden). Eichfeld, I. 2015, ‘Groothusen und Grimersum – Siedlung, Wirtschaft und Wasserwege im frühmittelalterlichen Ostfriesland’, Siedlungs- und Küstenforschung im südlichen Nordseegebiet 38, 217–37. — 2017, ‘Grubenhaus und Königshof. Historisch-archäologische Untersuchungen zu frühmittelalterlichen Handelsplätzen auf der Hohen Lieth bei Cuxhaven’. In Häfen im 1. Millenium AD. Standortbedingungen, Entwicklungsmodelle und ökonomische Vernetzung, eds. S. Kalmrin and. L. Werther, RGZM-Tagungen 31 (Mainz). Eisenschmidt, S. 2004, Grabfunde des 8. bis 11. Jahrhunderts zwischen Kongeå und Eider. Zur Bestattungssitte der Wikingerzeit im südlichen Altdänemark. Studien zur Siedlungsgeschichte und Archäologie der Ostseegebiete 5 (Neumünster). Ellmers, D. 1972, Frühmittelalterliche Handelsschiffahrt in Mittel- und Nordeuropa (Neumünster). Eriksen, P., T. Egeberg, L. H. Olesen and H. Rostholm 2009, Vikinger I vest. Vikingetiden i Vestjylland, Jysk Arkæologisk Selskab Skrifter 70 (Aarhus). Haarnagel, W. 1955, ‘Die frühgeschichtliche Handelssiedlung Emden und ihre Entwicklung bis ins Mittelalter’, Jahrbuch der Gesellschaft für Bildende Kunst und Vaterländische Altertümer zu Emden 35, 9–78.

198

Z01 Hines Book B.indb 198

07/05/2021 13:30

Early-Medieval Settlement Archaeology — 1984, ‘Die frühgeschichtliche Handelssiedlung Emden und ihre Entwicklung im Mittelalter’. In Archäologische und naturwissenschaftliche Untersuchungen an ländlichen und frühstädtischen Siedlungen im deutschen Küstengebiet vom 5. Jahrhundert v. Chr. bis zum 11. Jahrhundert n. Chr. 2: Handelsplätze des frühen und hohen Mittelalters, eds. H. Jankuhn, K. Schietzel, H. Reichstein (Weinheim), 114–35. Harck, O. 1990, Archsum auf Sylt Teil 3. Die Ausgrabungen in den römerzeitlichen Erdwerken Archsumburg, Tinnumburg und Trælbanken an der Westküste Schleswigs, Römisch-Germanische Forschungen 30 (Mainz). Harck, O., G. Kossack and J. Reichstein 1992, ‘Frühe Bauern auf Sylt. Ausgrabungen in Archsum’. In Der Vergangenheit auf der Spur. Archäologische Siedlungsforschung in Schleswig-Holstein, eds. M. Müller-Wille and D. Hoffmann, (Neumünster), 11–38. Hatz, G. 2001, Der Münzfund vom Goting-Kliff/Föhr, Numismatische Studien 14 (Regenstauf). Higelke, B., D. Hoffmann, H. J. Kühn and M. Müller-Wille 1988, Norderheverprojekt 1. Landschaftsentwicklung und Siedlungsgeschichte im Einzugsgebiet der Norderhever (Nordfriesland), Offa-Bücher 66 (Neumünster). Hoffmann, D. 1988, ‘Das Küstenholozän im Einzugsbereich der Norderhever, Nordfriesland’. In Norderheverprojekt 1. Landschaftsentwicklung und Siedlungsgeschichte im Einzugsgebiet der Norderhever (Nordfriesland), Offa-Bücher 66, eds. B. Higelke, D. Hoffmann, H. J. Kühn and M. Müller-Wille 1988 (Neumünster), 51–115. Hvass, S. 1979, ‘Vorbasse. The Viking-age settlement at Vorbasse, central Jutland’, Acta Archaeologica 50, 137–72. — 1986, ‘Vorbasse. Eine Dorfsiedlung während des 1. Jahrtausends nach Christus in Mitteljütland, Dänemark’, Berichte der Römisch-Germanischen Kommission 67, 529–42. Jankuhn, H. 1960, ‘Zur Frage der friesischen Einwanderung in Nordfriesland’. In Philologia Frisica anno 1959. Lezingen en Debatten. Frysk Filologekongres Augustus 1959 (Groningen), 43–7. Jöns, H. 1996, ‘Zum Stand der eisenzeitlichen Siedlungsarchäologie in den Altmoränenlandschaften Nordfrieslands’, Offa 53, 193–226. — 2015: ‘Early medieval trading centres and transport systems between Dorestad, Ribe and Wolin’. In Small Things – Wide Horizons. Studies in Honour of Brigitta Hårdh, eds. L. Larsson, F. Ekengren, B. Helgesson and B. Söderberg (Oxford), 245–52. Kersten, K. and P. La Baume 1958, Vorgeschichte der nordfriesischen Inseln. Vor- und Frühgeschichtliche Denkmäler und Funde in Schleswig-Holstein 4 (Neumünster). Kordowski, J. 2013, ‘Bernsteinschnitzer an der Nordsee – in Tinnum auf Sylt’, Archäologische Nachrichten aus Schleswig-Holstein 19, 78–81. Kossack, G., O. Harck and J. Reichstein 1974, ‘Zehn Jahre Siedlungsforschung in Archsum auf Sylt’, Berichte RGK 55, 261–428. — eds. 1980, Archsum auf Sylt Teil 1. Einführung in den Forschungsverlauf und Landschaftsgeschichte, Römisch-Germanische Forschungen 39 (Mainz). Kühl, J. and N. Hardt 1999, Dannevirke. Nordens største fortidsminde (Herning). Laur, W. 1960, ‘Das Siedlungsgebiet der Nordfriesen in seiner geschichtlichen Entwicklung im Spiegel der Ortsnamen’. In Philologia Frisica anno 1959. Lezingen en Debatten. Frysk Filologekongres Augustus 1959 (Groningen), 11–19. Majchczack, B. 2015a, ‘Siedlungen aus dem Nichts. Die Zusammenführung zerstörungsfreier Prospektionsmethoden als Mittel der Siedlungsforschung auf der nordfriesischen Insel Föhr’. In Archaologische Siedlungsforschung auf den nordfriesischen Inseln, eds. C.

199

Z01 Hines Book B.indb 199

07/05/2021 13:30

Frisians of the Early Middle Ages von Carnap-Bornheim and M. Segschneider, Offa-Bücher 89 (Neumünster.), 15–117. — 2015b, ‘Neues vom Goting-Kliff auf Föhr. Eine Siedlung von der Jüngeren Römischen Kaiserzeit bis ins Frühmittelalter im Spiegel alter Sammlungen und aktueller Prospektion’. Arkæologi i Slesvig/Archäologie in Schleswig 15, 139–52. — 2018, ‘Die Insel Föhr und ihr römisch-kaiserzeitliches Netzwerk im Spiegel des Urnengräberfeldes von Nieblum LA 41, Kr. Nordfriesland. Mit einem Beitrag von B. Großkopf ’, Siedlungs- und Küstenforschung im südlichen Nordseegebiet 41, 133–72. — 2020, Die Rolle der nordfriesischen Inseln im frühmittelalterlichen Kommunikationsnetzwerk. PhD-Dissertation University of Rostock, 2020. Studien zur Landschafts- und Siedlungsgeschichte im südlichen Nordseegebiet 11 (Rahden/Westf.). Majchczack, B. and K. Offermann 2018, ‘Frühmittelalterliche Handelsplätze an der deutschen Nordseeküste. Aktuelle Grabungsergebnisse aus Witsum auf Föhr und Altenwalde bei Cuxhaven’. In Auf archäologischer Spurensuche. Gedenkschrift für Dr. Wolf-Dieter Tempel,15.08.1937–07.03.2017, ed. S. Hesse, Archäologische Berichte des Landkreises Rotenburg (Wümme) 21 (Oldenburg), 239–54. Majchczack, B., S. Schneider, T. Wunderlich, D. Wilken, W. Rabbel and M. Segschneider 2018, ‘Early medieval trading sites on the North-Frisian island of Föhr. First results of fieldwork in Witsum and Goting’. In Harbours as Objects of Interdisciplinary Research – Archaeology + History + Geosciences, eds. C. von Carnap-Bornheim, F. Daim, P. Ettel, and U. Warnke, RGZM-Tagungen 34 (Mainz.), 311–28. Majchczack, B. and M. Segschneider, M. 2015, ‘Eine Siedlung der Jüngeren Römischen Kaiserzeit und Völkerwanderungszeit sowie des Frühmittelalters in Tinnum auf Sylt’. In Archaologische Siedlungsforschung auf den nordfriesischen Inseln, eds. C. von Carnap-Bornheim and M. Segschneider, Offa-Bücher 89 (Neumünster), 119–34. Nielsen, L. C. 1979, ‘Omgård. A settlement from the Late Iron Age and the Viking Period in West Jutland’, Acta Archaeologica 50, 172‒208. Offermann, K. J., S. Schneider, F. Schlütz, and A. Fediuk 2019, ‘Archäologische und naturwissenschaftliche Untersuchungen auf dem frühmittelalterlichen Siedlungs- und Handelsplatz Kamp-Seeburg, Cuxhaven-Altenwalde, Ldkr. Cuxhaven’, Siedlungs- und Küstenforschung im südlichen Nordseegebiet 42, 203–36. Panten, A. 2010, Die Nordfriesen im Mittelalter, Geschichte Nordfrieslands 2 (Bredstedt). Reichstein, H. 1984, ‘Haustiere’. In Archäologische und naturwissenschaftliche Untersuchungen an ländlichen und frühstädtischen Siedlungen im deutschen Küstengebiet vom 5. Jahrhundert v. Chr. bis zum 11. Jahrhundert n. Chr. 1. Ländliche Siedlungen, eds. G. Kossack, K.-E. Behre and P. Schmid (Weinheim). Reichstein, J. 1986, ‘Ausgrabungen in Alt-Archsum auf Sylt’, Ber. RGK 67, 373–84. Reinhardt, W. 1959, Die Grabungen auf der Dorfwarf von Groothusen, Kreis Norden, und ihre Ergebnisse, Jahrbuch der Gesellschaft für bildende Kunst und vaterländische Altertümer zu Emden 39 (Emden). Schlosser Mauritsen, E., M. Segschneider and T. Wunderlich 2009, ‘Siedlungsprospektion mit Luftfotografien und Geomagnetik auf der nordfriesischen Insel Föhr’, Archäologische Nachrichten aus Schleswig-Holstein 15, 20–3. Schmölcke, U. 2009, ‘Neue Bemerkungen zu alten Funden — Die Tierknochen aus der Borgsumburg („Lembecksburg’) auf Föhr’. In Ringwälle und verwandte Strukturen des ersten Jahrtausends n. Chr. an Nord- und Ostsee. Symposium Utersum 2005, ed. M. Segschneider, Schriften des Archäologischen Landesmuseums Ergänzungsreihe 5 (Neumünster), 123–36.

200

Z01 Hines Book B.indb 200

07/05/2021 13:30

Early-Medieval Settlement Archaeology Segschneider, M. 2002, ‘Trade and centrality between the Rhine and the Limfjord around 500 AD. The beachmarket on the Northfrisian island Amrum and its context’. In Central Places in the Migration and Merovingian Periods. Papers from the 52nd Sachsensymposium Lund, August 2001, eds. B. Hårdt and L. Larsson, Uppåkrastudier 6. Acta Archaeologica Lundensia 39 (Lund), 247–56. — 2008, ‘Wikingerzeitliche Bernsteinschnitzerei und Glasperlenherstellung in einer neu entdeckten Grubenhaussiedlung bei Tinnum, Sylt-Ost (LA 128), Kr. Nordfriesland’, Archäologische Nachrichten aus Schleswig-Holstein 14, 62–5. — 2009, ‘Die Ringwälle auf den nordfriesischen Inseln’. In Ringwälle und verwandte Strukturen des ersten Jahrtausends n.  Chr. an Nord- und Ostsee. Symposium Utersum 2005, ed. M. Segschneider, Schriften des Archäologischen Landesmuseums Ergänzungsreihe 5 (Neumünster), 99–111. — 2012/13, ‘Ausgesiebtes vom Roggen. Ein archäobotanischer Fund aus der wikingerzeitlichen Grubenhaussiedlung bei Tinnum auf Sylt, LA 128’, Offa 69/70, 21–5. — 2018, Die Marschen der Insel Föhr und der Wiedingharde, Kreis Nordfriesland. Eine siedlungsarchäologische Studie, Studien zur Landschafts- und Siedlungsgeschichte im südlichen Nordseegebiet 10 (Rahden). Segschneider, M., B. Majchczack and E. S. Mauritsen 2015, ‘Luftfotoarkæologiske og geofisiske undersøgelser på Föhr.’ In Luftfotoarkæologi i Danmark, L. Helles Olesen and E. S. Mauritsen (Holstebro), 246–63. Siegloff, E. and A. Tummuscheit 2017, ‘Es findet zusammen, was zusammen gehört... Ein Vorbericht zum wikingerzeitlichen Hortfund von Morsum auf Sylt, Kreis Nordfriesland’, Archäologische Nachrichten aus Schleswig-Holstein 23, 74–83. Siegmüller, A. 2010, Die Ausgrabungen auf der frühmittelalterlichen Wurt Hessens in Wilhelmshaven. Siedlungs- und Wirtschaftsweise in der Marsch. Studien zur Landschaftsund Siedlungsgeschichte im südlichen Nordseegebiet 1 (Rahden/Westf.). Siegmüller, A. and H. Jöns 2012, ‘Ufermärkte, Wurten, Geestrandburgen — Herausbildung differenter Siedlungstypen im Küstengebiet in Abhängigkeit von der Paläotopographie im 1. Jahrtausend’, Archäologisches Korrespondenzblatt 42/4, 573–90. Skre, D. 2010, ‘From Dorestad to Kaupang: Frankish traders and settlers in a 9th-century Scandinavian town’. In Dorestad in an International Framework: New Research on Centres of Trade and Coinage in Carolingian Times, eds. A. Willemsen and H. Kik (Turnhout), 137–41. Stilke, H. 1995, ‘Die früh- bis spätmittelalterliche Keramik von Emden’, Probleme der Küstenforschung im südlichen Nordseegebiet 22, 9–200. — 2001, ‘Muschelgrusware’. In Handbuch zur mittelalterlichen Keramik in Nordeuropa, eds. H. Lüdtke and K. Schietzel, Schriften des archäologischen Landesmuseums 6 (Neumünster), 175–208. Stoumann, I. 1979, ‘Sædding. A Viking-Age village near Esbjerg’, Acta Archaeologica 50, 95‒118. Ten Harkel, Letty 2013, ‘A Viking Age landscape of defence in the Low Countries? The Ringwalburgen in the Dutch province of Zeeland’. In Landscapes of Defence in Early Medieval Europe, eds. J. Baker, S. Brookes and A. Reynolds, Studies in the Early Middle Ages 28 (Turnhout). Tys, D., P. Deckers and B. Wouters 2016, ‘Circular, D-shaped and other fortifications in 9th- and 10th-century Flanders and Zeeland’. In Fortified Settlements in Early Medieval Europe, eds. N. Christie and H. Herold (Oxford–Philadelphia), 175–91.

201

Z01 Hines Book B.indb 201

07/05/2021 13:30

Frisians of the Early Middle Ages Westphalen, P. 1999, ‘Die Kleinfunde aus der frühgeschichtlichen Wurt Elisenhof ’. In Elisenhof 7, Offa-Bücher 80 (Neumünster), 1–231. — 2014, Die Häuser von der frühgeschichtlichen Warft Elisenhof, Offa-Bücher 87 (Neumünster). Wiechmann, R. 1996, Edelmetalldepots der Wikingerzeit in Schleswig-Holstein. Vom „Ringbrecher’ zur Münzwirtschaft, Offa-Bücher 77 (Neumünster). Zimmermann, W. H. 1982, ‘Archäologische Befunde frühmittelalterlicher Webhäuser. Ein Beitrag zum Gewichtswebstuhl’, Jahrbuch der Männer vom Morgenstern 61, 111–44.

202

Z01 Hines Book B.indb 202

07/05/2021 13:30

• 7 • FRANKS AND FRISIANS Ian Wood

O

ur knowledge of relations between the Franks and Frisians in the Merovingian Period largely concerns the region that Bede calls Frisia Citerior (Historia Ecclesiastica, V, 10). Unfortunately the Anglo-Saxon monk nowhere offers a definition of the region, but Alcuin, in his Vita Willibrordi, states that when Willibrord first visited Radbod, which is normally thought to have been in 690, he went to the king’s castrum at Utrecht (Vita Willibrordi 5). This and the revelatory excavations of Wim van Es at the site of Wijk bij Duurstede have not surprisingly led historians to concentrate their attention on the region of Utrecht and Dorestad. The very word citerior itself implies that it is more central than the other part of Frisia defined as ulterior. This, however, is only Bede’s nomenclature (Bazelmans 2009, 331). We do not know where it came from, or whether it reflected an established view. It has arguably led to a rather blinkered understanding of Frankish-Frisian relations. For a start, it is worth noting that Utrecht and Dorestad do not appear to have been the home territory of those Frisian kings about whom we know anything. Stephen of Ripon (Vita Willibrordi 27) tells us nothing about the whereabouts of Aldgisl’s centre of power – but he implies that it was close to fishing grounds. As we have seen, Radbod had a fortress at Utrecht, but this is unlikely to have been ancestral land, since the castellum had been in Frankish hands earlier in the seventh century (Boniface, ep. 109). Current views have placed the centre of his power in West Frisia, perhaps in Texel or Kennemerland (van Egmond 2005; Nijdam and Knottnerus 2019). Alcuin seems to imply that his authority stretched as far as Helgoland – but unfortunately the chapters of the Life of Willibrord describing the island shrine of Fosite (Vita Willibrordi, 10–11) do not make it entirely clear whether Helgoland was part of Radbod’s kingdom, or whether his reaction to the saint’s sacrilege was merely a response to demands made on him by the inhabitants of the island. The only other named leader, Bubo, was killed at the Battle at the Boorne (Fredegar, Cont. 17), and appears to have had his powerbase in Westergo or Oostergo. Oostergo was, of course, the focus of Boniface’s mission at the time of his martyrdom at Dokkum (see Fig. 7.1 for the principal sites referred to in this chapter). Equally, however, Utrecht was obviously peripheral for the Franks. Boniface, in a letter to Pope Stephen (ep. 109), states that Dagobert conveyed the castellum of Utrecht to the church of Cologne as a missionary base. The Dagobert in question is usually thought to have been Dagobert I (623–638/9), but Wolfert van Egmond (2010) has argued that a more likely candidate would be Dagobert II (676–9). There seems, however, to be no reason to reject the traditional identification. It fits well with what is known of the missionary activity of Amandus, who was certainly active in the region 203

Z01 Hines Book B.indb 203

07/05/2021 13:30

Z01 Hines Book B.indb 204

07/05/2021 13:30

Fig. 7.1  Map of the Frisian region showing the principal sites referred to. Note the orientation of the North point.

Franks and Frisians of the Scheldt and lower Rhine in the time of the earlier king (Wood 2001, 38–42), and while he did survive into the later reign, he was around ninety years old at the time. After the death of Dagobert I the Merovingians seem to have shown little or no interest in the region, although the Neustrian maior palatii Ebroin attempted to put pressure on Aldgisl to send the Anglo-Saxon bishop Wilfrid to him, dead or alive (‘Eddius’ Stephanus, Vita Wilfridi, 27). Nor is there any evidence for Pippinid or Arnulfing interest in Frisia before the 690s. Their power base was much further south, in Metz, Liège and Nivelles, and it was by no means assured even as late as 687. The image that we have of Pippinid power before the 730s is dependent on chronicle accounts, almost all of which were compiled after the successes of Charles Martel and of Pippin III in the mid-eighth century. It would seem that Utrecht and Dorestad were peripheral places for the Franks as for the Frisians, even if the leaders of both groups wished to control them. In other words, although most of what we know about Frankish-Frisian relations concerns the area around the Kromme Rijn and the Vecht, we should not assume that this was the heart of Frisian power. That there were alternative centres has been graphically illustrated by Johan Nicolay (2014). Nevertheless in the years before 719 it was over Dorestad and Utrecht that Franks and Frisians came into conflict. Our evidence for relations between Franks and Frisians falls neatly into three areas of study: political, economic and religious. When discussing the economy it is tempting to emphasise the evidence of the sceatt coinage and of the archaeology of Dorestad, along with what little can be learnt of Domburg, although the project of Frans Theuws on the rural population of Brabant is shedding significant light on the non-urban economy. But, as Johan Nicolay (2014) has demonstrated, more important for the Frisians themselves may have been the economy of the coastal zone, which is being steadily revealed by the archaeology of the terpen and other settlement sites. For religious relations we have the hagiographical narratives relating to the Frisian missions, above all those of Willibrord, Boniface and Liudger, together with the highly problematic Life of Wulfram of Sens (cf. Hines, this vol.). Unlike the evidence for trade and for politics, these sources do direct us away from the Utrecht-Dorestad region, most obviously to Walcheren, Dokkum and Helgoland. For Toxandria, the area immediately to the south, we have a few precious charters in the Liber Aureus of Echternach (Costambeys 1994). I will concentrate, however, on the political aspects of Frankish-Frisian relations, and that means focusing on the Utrecht-Dorestad region. I shall begin with a brief survey of our written evidence, before providing an outline of the narrative of political relations between Franks and Frisians in the Merovingian Period. In the course of that outline, however, it will be necessary to pause on certain specific episodes and some of the evidential problems associated with them. Although there are Roman references to Frisia and Frisians, there is a caesura between the third and the sixth centuries which coincides with the archaeological evidence for desertion and depopulation (cf. Knol and IJssennagger 2017, 11–12), and which suggests that the situation in our early medieval sources was a new one. There is an intriguing cluster of material from the middle of the sixth century. Procopius (Wars VI, xx, 6–10) mentions Frissones as being among the peoples of Brittia, alongside Angiloi and Brittones, and he states that there were Angiloi present in a legation sent to Constantinople to make claims over the island of Brittia by a king of the Franks. If the anecdote is correctly placed in Procopius’ Histories it should relate to c. AD 550, and 205

Z01 Hines Book B.indb 205

07/05/2021 13:30

Frisians of the Early Middle Ages the king in question ought therefore to have been Chlothar I (511–61), in which case this was a different legation from that sent by Theudebert I (534–47), when we know from the Epistolae Austrasicae (20) that the Merovingian king claimed dominance over the Saxons and Jutes, and thus presumably the North Sea littoral between the Scheldt and Jutland. Whether we can make any more use of this section of Procopius’ Histories, which includes a famous story of the betrothal of Radigis, son of Hermegisl, king of the Warni, to an Anglian princess, is open to question. According to Procopius, Hermegisl fell ill before the marriage had taken place, and on his deathbed he instructed his son to marry his step-mother, the sister of the Merovingian Theudebert – an act which prompted an invasion by the Angiloi, to assert the rights of their jilted princess. Procopius, of course, does not state that Radigis was a Frisian, and the whole narrative looks like a semi-legendary tale. But the elements in Radigis’ name recur in those of two known seventh-century rulers of the Frisians, Aldgisl and Radbod, which makes one wonder whether this belongs to Frisian prehistory. If this is the case, there were marriage ties between Merovingian royalty and the rulers of Frisia before the mid-sixth century. If Hermegisl’s wife really was the sister of Theudebert, this may shed some light on the Merovingian king’s claims over the Saxons and Jutes – a claim which may have been inherited by Chlothar after Theudebert’s death. The supposed actions of the Angiloi, however, are a reminder that we have also to factor in relations with peoples in England. Equally crossing the boundary between history and legend is the story of Hygelac’s death, as recounted in Beowulf, where we also have Gregory of Tours’ account (LH III, 3) of the victory of Theudebert over the Danish invader Chlochilaichus somewhere on the lower Rhine. But Gregory does not name Frisia or Frisians once in his works: for him the Chlochilaichus story is a Franco-Danish one. Gregory’s contemporary and friend, the late sixth-century poet Venantius Fortunatus, is a little more forthcoming, and in his panegyric on Chilperic (561–84) he refers in a single line (carm. IX, 1, l. 75) to the king’s response to the threat of Frisians. This, however, is not enough to allow us to build a picture of relations between Frisians and Franks. Nor does the solitary coin of Audulfus help, not least because we cannot be certain of its place of origin or its date. Our sixth-century evidence, in other words, does not allow us to postulate the existence of a Frisian kingdom, but merely to state that Frisians were to be found amongst the rather fluid groupings of the peoples of the Lower Rhine and around the coastal zone between the Scheldt and the Weser (Wood 1990). Our evidence for the seventh century is stronger. The Merovingian coins that have been found, especially in Dorestad, show that the Lower Rhine was well integrated into the Frankish economy (Wood 1994, 296–9), while the written evidence does allow us to explore relations between Franks and Frisians, although it should be stated immediately that almost all of it was written down in the eighth century. The main exception is a reference to the penalty for killing a Frisian in Lex Ribvaria (40, 4), which is usually dated to the reign of Dagobert I, at which time, as we have seen, Utrecht at least was in Frankish hands. Otherwise we have to wait for the composition of the Vita Wilfridi by Stephen of Ripon (alias Eddius Stephanus) in c. 712–14, where we are told (ch. 27) that in c. 678 Wilfrid passed through Frisia, which at the time was under the control of King Aldgisl, of whom we can only say that he was firmly independent of the Merovingians and the aggressive Neustrian maior palatii Ebroin, because he refused to kill or hand over the 206

Z01 Hines Book B.indb 206

07/05/2021 13:30

Franks and Frisians Anglo-Saxon saint when asked to do so. Although this information was set down forty years after the events in question, we should remember that when Wilfrid visited Willibrord, presumably in Utrecht, in c. 703, he was accompanied by Acca, who Stephen, the author of the Vita Wilfridi, knew personally (Bede, Historia Ecclesiastica, III, 13). The hagiographer’s comments on Frisia are those of someone with near-inside information. For the last decade of the seventh century and the first two decades of the eighth we have a good deal more evidence, although the earliest text, the Liber Historiae Francorum, which was written in the Soissons region, dates to c. 727. This was elaborated in the middle of the century by the Fredegar continuator. There are also important entries in other annals, most notably in a group of texts known as the Minor Frankish Annals. Unfortunately, although the later entries in these annals are relatively well understood, as deriving from the Annales Regni Francorum, the origin of the information in the entries before 741 is unknown, and they contain the most important information we have on Radbod. As Rosamond McKitterick (2000) has demonstrated in no uncertain terms, the eighth-century annals (and this includes both the Liber Historiae Francorum and the continuations of Fredegar) are deceptive in their presentation of Pippinid/Carolingian power in the first half of the century. In addition to the Carolingian narrative sources we have some important information in Bede’s Ecclesiastical History, which the author claims to have written in 731. Bede provides important evidence on Willibrord’s activity in the region of Utrecht, and indeed he supplies us with our fullest narrative account. Unlike the Frankish authors he had no Carolingian bias, and he is also likely to have had a good source of information, since like Stephen of Ripon he knew Acca, who had visited Willibrord on the Continent (Bede, Historia Ecclesiastica, III, 13). On the other hand, it is not always easy to integrate his information with what we find in other sources, and on at least one point he is marginally incorrect. Otherwise we are dependent on the hagiography, most obviously the Vita Willibrordi by Alcuin, which was only written in 796 (Wood 2001, 80–3), and is therefore a good deal later than the evidence that we have for Boniface’s comparatively limited activity in Frisia, to be found in his letters, and in his earliest vita, the Vita Bonifatii by Willibald, written in 763–8. Alcuin did, however, have family connections with Willibrord, and may have had access to genuine early traditions. Nevertheless the earliest hagiographical text of relevance is the Life of Ermin, the second abbot of Lobbes, who died in c. 737. The Vita Erminonis of Anso of Lobbes was written between 751 and 768 (Dierkens 1985, 103) – and therefore perhaps earlier than any of the annal compilations. The hagiographical texts are obviously not simple historical narrations, but religious works which address ecclesiastical issues of the early Carolingian Period. Alcuin’s Life of Willibrord, for instance, belongs to a complex debate about mission that was taking place in the mid-790s (Wood 2001, 80–3). Even so, Anso and Alcuin, like Stephen of Ripon and Bede, may have had access to good information. More problematic is the Life of Wulfram of Sens. The work has a dedication to Bainus, who was abbot of St Wandrille, on the Lower Seine, from 701 to 710. From the Gesta Abbatum Fontanellensium (GAF, II, 4) we know that Bainus had the body of Wulfram translated, along with those of Wandregisl and Ansbert, which gives us a terminus ante quem for the saint’s death. Yet the author of the Vita Vulframni (ch. 10) claims that Wulfram outlived Radbod, who died in 719. That the author was not a contemporary of Bainus is also clear from the fact that he cites Alcuin’s Vita Willibrordi, which was 207

Z01 Hines Book B.indb 207

07/05/2021 13:30

Frisians of the Early Middle Ages written in c. 796. In fact the composition of the Life of Willibrord may well have inspired that of Wulfram, which was certainly in existence before the death of Abbot Gervold in 807 (GAF XII, 3: Wood 2001, 92–4). But even if we regard the text as making fraudulent claims, it is likely, as Stéphane Lebecq has demonstrated (2000; 2018), that Wulfram genuinely did work in Frisia, and it is possible that the Frisians Ovo, Eurinus, Ingomar, Fresio and Chuningus, who the hagiographer states returned to St Wandrille with the missionary, were real individuals whose memories were preserved at the monastery on the Seine. We should, therefore, consider seriously the statement made by the author of the Life of Wulfram (ch. 4) that Radbod allowed Frisians to convert to Christianity, and that one of the converts was his own son, another Radbod. It is worth noting that the Vita Willibrordi of 796 and the Vita Vulframni of the following decade are almost exactly contemporary with Lex Frisionum, which appears to date between 785 and 793. This was a period in which Frisia and its past were attracting considerable attention. The period itself is perhaps best illustrated by the Vita Liudgeri of Altfrid, written between 825 and 849 (Wood 2001, 52, 113–15), when Liudger’s nephew was deeply concerned about his uncle’s legacy as a missionary in Frisia and Saxony, and as a result set down a history of his saintly ancestor, and indeed of the family, which was Frisian in origin, from the days of Radbod onwards. This is remarkable evidence, but like all our texts it has its biases and its inaccuracies, which need to be noted. Bearing in mind the problems posed by individual sources, we can revisit the narrative of relations between Frisians and Franks in the late seventh and early eighth centuries. That Frisia had been independent of the Franks before Radbod took control of the region is clear from the Vita Wilfridi (chs 26–7) of Stephen of Ripon, written in 712–14. There we are told that the Frisian leader Aldgisl, who is described by Stephen as rex, was prepared to oppose the demands of the Frankish maior palatii Ebroin, at a date that can be calculated as c. 678. Unfortunately Stephen does not provide an exact location for Aldgisl’s centre of authority, merely placing his power to the east of Ebroin’s, that is to say, to the east of the Neustrian coast, which stretched from Brittany up into the wetlands of the Maas, and north of Strasbourg. Whether Aldgisl controlled territory as far south as Utrecht and Dorestad is unclear. If his power extended into what Bede calls Frisia Citerior (Historia Ecclesiastica V, 10), we might wonder whether he should be compared to the rulers of the so-called peripheral duchies (Alamannia, Thuringia and Bavaria: Werner 1973), who took advantage of the internal divisions that plagued the Merovingian kingdom from the 660s onwards to increase their independence from their more powerful neighbour. Alternatively, and perhaps more significantly, Aldgisl came from Frisia Ulterior, and he took advantage of the divisions within the Merovingian world to take control of Utrecht and Dorestad, and their trade. The two readings might, of course, overlap. Wolfert van Egmond has argued that Radbod should be seen as a major aristocrat living on the periphery of the Merovingian world – and that in this he was not unlike Pippin II or Charles Martel (van Egmond 2005; cf. Theuws 2018, 46–9). Aldgisl may not have been so different. Given the evidence for Utrecht in the days of Dagobert provided by Boniface, it would seem that the Frisians controlled more territory in the Rhineland in c. 680 than they had fifty years earlier. Even so the Franks, at least in retrospect – as we have seen most of our evidence comes from the mid-eighth century or later – regarded the Frisians as a dependent people. In the Merovingian and Carolingian sources Radbod is usually described as dux (implying subordination to a Frankish ruler). It is the Anglo-Saxons, 208

Z01 Hines Book B.indb 208

07/05/2021 13:30

Franks and Frisians Bede (Historia Ecclesiastica V, 10), Willibald (Vita Bonifatii 4–5) and Alcuin (Vita Willibrordi 5 and 9) who describe him as rex, as Stephen described Aldgisl. (On Radbod see most recently van Egmond 2005; Nijdam in Schoorstra 2018, 7–17; Meeder and Goosmann 2018; Theuws 2018, 46–9.) The Anglo-Saxons also provide us with the next piece of information. In 690 Willibrord was sent from Rathmelsigi in Ireland by Ecgbert to Frisia: the date is fixed by an entry in the Calendar of Willibrord. However, according to Bede, who provides us with our fullest narrative, he was not the first missionary who Ecgbert had sent to the region: his predecessor Wihtbert spent two years preaching in Frisia in the time of rex Radbod, but he achieved nothing. In Bede’s account (Historia Ecclesiastica, V, 9–10) Willibrord went straight to Pippin, the dux Francorum, who had just driven Radbod out of Frisia Citerior. Then (Historia Ecclesiastica V, 11), having been given permission to preach by the princeps (presumably Pippin), he went to Rome to secure the approval of Pope Sergius. While he was in Rome those who were preaching in Frisia elected Swithbert as bishop, and sent him to England, where he was consecrated by Wilfrid. On Swithbert’s return he went to preach to the Boructuarii. Pippin then sent Willibrord for a second time to Rome, where he was consecrated archbishop of the Frisians by Sergius in 696 (the date is Bede’s, not Willibrord’s). Back in Frisia Willibrord was given the fortress of Wiltaburg, or Traiectum (Utrecht), as his episcopal base. This does not dovetail exactly with the Frankish evidence that we have for the history of Utrecht and Dorestad. That begins with the defeat of Radbod. The earliest account of much of this comes in the Liber Historiae Francorum, written probably in the Soissons region in 727, which tells us that Pippin II launched numerous attacks on the Frisians (LHF 49). This information is expanded a little by the first continuator of the Fredegar Chronicle, writing in the middle of the eighth century (Cont. 6). The continuator certainly had access to the Liber Historiae Francorum, but he adds chronological and geographical precision to imply that it was at some point in or after 695 that Pippin defeated Radbod in the vicinity of Dorestad, and that the Frisian dux fled. This is presumably the expulsion of Radbod from Frisia Citerior mentioned by Bede in his Historia Ecclesiastica of c. 731 (V, 10). The Annales Mettenses Priores, probably composed in 806, perhaps at Chelles at the request of Charlemagne’s sister Gisela (Nelson 1991, 192–5), provide the fullest information on Pippin’s campaigns against Radbod, listing two: the first in 692; and the second which it claims culminated in a battle at Dorestad in 697. The first campaign supposedly led to the payment of a heavy tribute, the second to the taking of large amounts of booty. The Metz Annals are certainly a highly partisan account of early Carolingian history, and it is unclear whether their dates, or indeed these claims, are to be trusted. In fact the chronology in general is problematic. It is not easy to see how accounts of Pippin’s initial victories fit with Willibrord’s certain arrival in 690. One possibility, stated by Hans-Joachim Reischman (1989, 54, n. 122) is that Radbod was driven out of Utrecht in 689. This obviously depends on accepting Bede’s account without question and can be straightforwardly dismissed. Alcuin (Vita Willibrordi 5–6) is clear that when Willibrord arrived Radbod held the castellum of Utrecht, and that the Anglo-Saxon therefore went to get help from Pippin, who sent him to Rome to be consecrated. Alcuin’s narrative, however, raises problems with the date given for Pippin’s later victory in 697, since the Calendar of Willibrord states that the Anglo-Saxon missionary was consecrated as archbishop of the Frisians by Pope Sergius in November 695. Bede 209

Z01 Hines Book B.indb 209

07/05/2021 13:30

Frisians of the Early Middle Ages states that Willibrord made two visits to Rome, and that it was on the second visit, which he dates to 696, that he was consecrated archbishop of the Frisians by the pope at the request of Pippin (Historia Ecclesiastica V, 11): the date given by the Calendar of Willibrord clearly trumps that provided by Bede, but the statement that there were two visits to Rome is probably correct. Bede goes on to say that, on his return, Willibrord was immediately established in Utrecht. Given Fredegar’s date of 695 or later for Pippin’s victory at Dorestad, we should either envisage a great deal of activity in that year, with the defeat of Radbod being followed swiftly with the creation of the archdiocese of Utrecht, or we should see Radbod reacting against the new ecclesiastical establishment but immediately being defeated by Pippin. At the very least we can reconstruct a history of antagonism between Pippin and Radbod in the early 690s. Since Pippin had been caught up in internal Frankish conflicts prior to his victory at Tertry in 687, when he defeated the Neustrians under Theuderic III and his maior palatii Berchar (LHF 48; Fred. Cont. 5), we may guess that Radbod, like his predecessor Aldgisl, had been ruling without much interference from the Franks before that date. Following Pippin’s victory or victories over Radbod, Frisia Citerior, and Utrecht and Dorestad in particular, came into the Frankish sphere of influence. The most important outcome of Pippin’s victory or victories over Radbod would seem to have been religious, with the establishment of Willibrord at Utrecht. The early development of the Anglo-Saxon’s holdings in the region can be seen in the foundation of his monastery at Susteren on land given by Pippin and Plectrudis in 714 (Heidrich, cart. 6). But perhaps more interesting are the grants of Alphen on the Rhine and Eersel in Toxandria made by Engelbert in 709 and 712 (Costambeys 1994, 44, 56–7). Pippin’s victories in the 690s seem to have ushered in a period in which Radbod was subordinate to Frankish authority. But we have some reason for thinking that there was continuing tension. The best evidence for this comes in the remarkable opening chapters of the Vita Liudgeri written by Altfrid in the second quarter of the ninth century. Liudger, who was born in 742, and who was bishop of Münster from 805–9, was the grandson of Wrssing, a leading Frisian, who, according to the saint’s nephew Altfrid (Vita Liudgeri, I, 2), fell foul of Radbod, and along with other Frisians fled to Francia, where he joined Pippin’s son Grimoald. The latter came to the fore in c. 695 when he was appointed maior of the Neustrian palace. This might suggest that Wrrsing fled from Radbod either at the time of the war with Pippin or immediately after. According to Altfrid (Vita Liudgeri, I, 3), after a while Radbod asked Wrrsing to return, which he refused to do, although he did send his younger son Thiadgrim back to Frisia. He himself followed after Radbod’s death. Although Altfrid certainly had access to fascinating family information, there are aspects of his account that are clearly erroneous – we will come to a problem relating to Radbod’s actions in his last years. However, we can probably accept the presence of Wrrsing in Grimoald’s household, and this may be of some relevance to the next significant development in relations between Radbod and the Pippinids. According to the Liber Historiae Francorum, Grimoald married Radbod’s daughter (LHF 50). The Fredegar continuator (ch. 7) places the event in the reign of Dagobert III; that is, at some point between 711 and 715. The Metz Annals also provide a date of 711 for the marriage. One later chronicle, that of Moissac, apparently compiled at Ripoll in Spain at the end of the tenth century but using earlier materials (Collins 1998, 6), dates it to 713, 210

Z01 Hines Book B.indb 210

07/05/2021 13:30

Franks and Frisians and adds the interesting detail that the daughter of Radbod who married Grimoald was called Thudsinda. This would seem to have been the high point in relations between Radbod and the Pippinid family. It is worth noting here that Grimoald had a son named Theudoald who, according to the Annales Mettenses Priores (s.a. 714), was the son of a concubine. Given Germanic naming patterns the mother’s name might well have included the Theud- element. This raises the possibility that Theudoald was actually Thudsinda’s son, and that the claim that the boy’s mother was a concubine was a slur – not impossible given that the Annals championed Grimoald’s half-brother, Charles Martel, and his offspring. The problem with this is that Theudoald, while still only a child, is unlikely to have been an infant in 714. The period after the marriage of Grimoald to Thudsinda saw considerable change. The Liber Historiae Francorum (ch. 50), the Fredegar continuator (ch. 7) and the Annales Mettenses Priores (s.a. 714) state that Grimoald was killed while praying at the shrine of Lambert in Liège, when he was en route to visit his sick father at Jupille. It is worth pausing on this, because it may shed light on what happened subsequently. Bishop Lambert of Liège had been murdered six years previously. The man behind his murder is said to have been the domesticus of Pippin, Dodo, who we are also told was related to the maior domus. The earliest version of the Life of Lambert (the Vita Landiberti vetustissima) says that Dodo was exacting vengeance for the murder of his relatives Gallus and Rivaldus, who had been killed by Lambert’s supporters. When Sigibert of Gembloux wrote his version of the Life of Lambert at the end of the eleventh century he claimed that Grimoald visited the shrine of Lambert because the saint had defended the marriage of Plectrudis against the supporters of Alpaida (Sigibert, Vita Landiberti, 16 and 27), a point Sigibert repeated in his Chronicle under the year 713. It is not at all clear what should be made of this late evidence. Sigibert was apparently the first person to link Lambert’s murder with his criticism of Pippin’s liaison with Alpaida, and also the first to accuse Radbod of responsibility for Grimoald’s murder. The Liber Historiae Francorum (ch. 50) simply notes the murderer’s name, Rantgar, and claims that he was a pagan (Fouracre 2000, 58). The Fredegar continuator (ch. 7) repeats the name. It is likely that Sigibert invented the idea that Radbod was behind the murder to suit his own agenda, but he may have had access to some early tradition. It is, therefore, just possible that Radbod was not happy at the marriage of his daughter to Grimoald. The possibility is a reminder that our evidence is not strong enough to allow much certainty. The Liber Historiae Francorum, the Fredegar continuator and the Annales Mettenses Priores tell us that Pippin appointed Theudoald, the young son of Grimoald and a concubine, in his father’s place as maior of the Neustrian palace, and the Metz Annals add that Pippin also avenged the death of Grimoald, before himself dying. This meant that in 714 the sole maior palatii was a child, Theudoald, who was supported by Plectrudis. Pippin’s only surviving adult son was Charles Martel, the child of his mistress Alpaida. Not surprisingly Theudoald and Plectrudis were quickly challenged, although Pippin’s widow succeeded in incarcerating Charles. The sources vary slightly in the order of narrative they relate, but the Neustrians appointed a maior of their own, Ragamfred, who defeated the young Theudoald. The Neustrian leader also formed an alliance with Radbod. Here it is worth remembering that with the death of Grimoald, the family ties between Radbod and the Pippinids had been broken. The Frisian dux was presumably in search of a new reliable Frankish ally – and he may well have regarded Theudoald 211

Z01 Hines Book B.indb 211

07/05/2021 13:30

Frisians of the Early Middle Ages as too young to be effective. Soon after, the Merovingian king Dagobert himself died, and the Neustrians appointed Chilperic II in his place. Chilperic and Ragamfred then marched across the Ardennes towards Cologne, where Plectrudis was residing, while Radbod approached the city from the north. At this moment Charles escaped (LHF 51; Fredegar Cont. 8–9). What followed is variously related in our sources. According to the Liber Historiae Francorum (chs 52–3), the Frisians defeated Charles, who was lucky to survive (Fouracre 2000, 61). Meanwhile Chilperic and Ragamfred forced Plectrudis to hand over her treasure, but they were ambushed at Amblève by Charles, who went on to defeat them again at Vinchy, which in retrospect marked the end of the Neustrian challenge to the Pippinids. The Fredegar continuator (chs 9–10) also recounts a battle between Charles and Radbod, but avoids saying who was the victor. He then recounts the arrival of Chilperic and Ragamfred at Cologne, their taking of Plectrudis’ treasure and their defeat at both Amblève and Vinchy. The Annales Mettenses Priores (s.a. 716) tell the same story, also leaving it unclear as to whether Charles or Radbod was defeated, by stating that both sides suffered heavy losses. The Minor Frankish Annals, however, all place emphasis on Radbod: the Annales sancti Amandi (s.a. 716) and the Annales Tiliani (s.a. 716) record his presence in Cologne. The Annales Petaviani (s.a. 716) relate his arrival in Cologne, and his battle with Charles without naming a victor. The Annals of Lorsch (s.a.716), the Annales Alamannici (s.a.716), the Annales Nazariani (s.a. 716) and the Short St Gall Annals (s.a. 716) all mention the battle but say nothing of the outcome. The Annales sancti Columbae (s.a.716) are more specific and place the battle at Cologne. The Moissac Chronicle (s.a. 716) states that Charles suffered huge losses. The Annals of Fulda (s.a.715) talk of the harm suffered by the army, although the grammar does not make it clear who was the loser. Among the so-called minor annals only the Annales Laurissenses minores ignore Radbod and say (s.a. 716) simply that Chilperic and Ragamfred went to Cologne. That Radbod was the victor, as claimed by the Liber Historiae Francorum (52), and that the victory was significant is confirmed by Anso of Lobbes, writing between 751 and 768 – and therefore perhaps earlier than any of the annal compilations. The Vita Erminonis (7), talking of the period immediately before Radbod’s death, states that the Franks were terrified by the threat of Frisian attacks, because they remembered that the Frisian princeps had once caused them to flee, with significant losses. In other words our earliest two accounts (which one might note are accounts least influenced by Carolingian propaganda) both assert that Charles was defeated. Although Radbod’s victory may seem unlikely, it is worth remembering that at this moment in time Frankish armies may still have been small – essentially the warbands of competing aristocrats. There is no contemporary evidence for the Frankish Marchfield between the early sixth century and the early eighth: the Fredegar continuator first talks of a Campus Madius in the year 761, and again in 763, 766 and 767. It would seem that the 730s and 740s saw a significant development in Frankish military organisation, perhaps in part due to the Arab threat. And here one might note that in his 737 campaign against the Muslim invaders Charles Martel relied on Lombard help, according to Paul the Deacon (Historia Langobardorum, VI, 54 and 58). In other words, Radbod did not have to defeat a large Frankish army in 715/16. As Wolfert van Egmond has noted (2005), there is not as much difference between Pippin II, Charles Martel and Radbod as the Frankish sources would have us believe. 212

Z01 Hines Book B.indb 212

07/05/2021 13:30

Franks and Frisians Despite the weight of references to what appears to have been a victory of Radbod, the normal narrative of the Frankish world in this period ignores his success, and instead concentrates on Charles’ victories at Amblève and Vinchy. In so doing it follows the Neustrian view of events expressed in the Liber Historiae Francorum and adapted by the Fredegar continuator. Of course, in the long run Amblève and Vinchy were decisive, whereas the victory of Radbod had no major long-term effect. One reason for this is, no doubt, Radbod’s death in 719, apparently from natural causes. That this was an event of huge importance is clear from the number of sources that refer to it (Annales Sancti Amandi, Annales Tiliani, Annales Petaviani, Annales Laureshamenses, Annales Alamannici, all with a date of 719, while the Chronicle of Moissac places it in 716) – though significantly both the Liber Historiae Francorum and the Fredegar continuations are silent on the matter. This silence surely reflects their Neustrian bias. It is, therefore, worth revisiting the years from 716 to 719. During this period Charles was occupied in the west, dealing with the challenge posed by Chilperic, Ragamfred and their ally Eudo of Aquitaine. It would seem that Radbod was left very much to his own devices. We can certainly dismiss the assertion of Alcuin, who states in the Life of Willibrord (ch. 13) that Charles defeated Radbod and took over Frisia. The absence of any comparable statement in the annalistic record makes this impossible. And in any case Willibald’s Life of Boniface (ch. 4) makes it clear that when Boniface first travelled to the Continent in 716 Radbod was installed in Utrecht, apparently blocking any chance of further Christian missions. Why Charles should have concentrated his attention on Neustria and Aquitaine is unclear, although it suggests that his real concern was with control of the Merovingian kingdom rather than with the security of Austrasia. He must surely have been aware that Radbod did present a threat: here the evidence of Anso, writing in Lobbes on the Sambre, is unequivocal. The Austrasians were terrified of the Frisian leader down until his death. Ermino’s prophesy of his death was a miracle. On the other hand we should probably take note of the fact that by this time Radbod was ageing – he had been an active warrior for a quarter of a century, and he would die in 719. According to the Life of Liudger (Altfrid, Vita Liudgeri I, 3), he was ill for the last six years of his life. Charles, in other words, may have been content to leave Radbod in control of the Lower Rhine after 716 in the knowledge that the Frisian leader was likely to die in the near future. It is also probable that Radbod had no obvious male heir. Theudoald, who may or may not have been his grandson, was out of the picture. It is also worth recalling the statement in the Vita Vulframni (ch. 4), that Radbod allowed Wulfram to baptise his son (also named Radbod), who died shortly after. Despite the evidence of the Vita Erminonis, what Radbod did between his defeat of Charles outside Cologne and his death in 719 is unclear, even with regard to the Church. He may have turned against Willibrord and driven him out of Utrecht, which would not have been surprising given that the Anglo-Saxon, having been very much a supporter of Plectrudis, transferred his allegiance during this period to Charles Martel. On the other hand, when, at exactly this moment, Boniface arrived in Frisia, hoping to support the Frisian mission, he was able to approach Radbod in Utrecht. Although he concluded that the time was not propitious for missionary activity and returned to England (Willibald, Vita Bonifatii 4), we are not told that he was driven out. What would have happened had Radbod lived to confront Charles when he turned 213

Z01 Hines Book B.indb 213

07/05/2021 13:30

Frisians of the Early Middle Ages his attention back to the Rhineland can only be a matter of guesswork. We should not, however, ignore Radbod’s brief period of success. Following the death of Radbod, Charles seems to have seized Frisia Citerior, although there is not much direct evidence. Altfrid (Vita Liudgeri I, 4) talks of Charles adding Frisia to his territory after the death of Radbod. Presumably Alcuin’s statement (Vita Willibrordi 13) that Charles took over the lands of the Frisians after his defeat of Radbod really refers to events following Radbod’s death. Otherwise we are dependent on information relating to the restoration of the Frisian mission. Willibald, in his Life of Boniface (ch. 5), states that the saint returned to Frisia as soon as he heard of the death of the Frisian king, which led directly to the ending of the persecution of the Christians – although we should note that Alcuin has probably exaggerated this persecution, as we have seen from Willibald’s Life of Boniface. The period after 719 was the highpoint of Willibrord’s missionary activity in Frisia. We do, however, hear of one further challenge to Frankish secular authority in the region. According to the Fredegar continuator (ch. 17), in 734 Charles sent a fleet to deal with a Frisian rebellion in Westergo and Oostergo, and he authorised the construction of a fortress on the River Boorne. The Frisian leader Bubo was killed, and the pagan temples of the Frisians were plundered and destroyed. This same campaign is confirmed by a number of minor annals, including the Annales Laureshamenses, the Annales Alamannici, the Annales Nazariani, and the Annales Sangallenses Breves, all referring to the year 735. The Annales Laurissenses minores are surely talking of the same event when they state, under the year 729, that Charles sent a naval force, against the Frisians, which killed their leader Poppo, destroyed their sacred groves and temples, and returned with great booty. What is most striking about this campaign is its naval nature – it is perhaps the only known maritime campaign in Merovingian history. Although Charles destroyed Bubo or Poppo, the region of Oostergo remained hostile to the Franks at least down to the moment of Boniface’s martyrdom at Dokkum in 754. By the 730s, however, Frisian opposition to Charles seems to have been confined to Frisia Ulterior, away from the Rhine basin. Not long after, the main secular leadership of Frisia was collaborating with the Franks: in 748, according to the Fredegar continuator (ch. 31), Pippin was supported by Frisians and Wends when he dealt with a Saxon uprising. Frisian-Frankish political history is, therefore, very largely a history of the relations between first Aldgisl and then Radbod on the one hand and the Frankish ruling class on the other. This was frequently an antagonistic relationship, as in the case of Aldgisl’s opposition to Ebroin, or Radbod’s hostility to Pippin before the mid-690s and to Charles Martel after 716. But between c. 695 and 714 Radbod seems to have come to a modus vivendi with the Frankish maior palatii. This seems to have culminated in the marriage of his daughter to Grimoald, although Sigibert of Gembloux later thought that the Frisian was opposed to the match, and that he was determined to have Pippin’s son killed – an unlikely reading of events, but one that cannot be disproved. During this period of apparent cooperation Radbod had no option but to accept the institution of a Christian mission, something that he may have allowed, but did not actively encourage, even before 690. After 714, however, he seems to have regarded Willibrord and the Christians in general as a threat to his authority. Yet even during this period, which is described by Willibald (Vita Bonifatii 4–5) as a time of persecution, Radbod was in contact with members of the Frankish governing class, being allied to Ragamfred and 214

Z01 Hines Book B.indb 214

07/05/2021 13:30

Franks and Frisians Chilperic II. Whether he liked it or not, he was operating within a Frankish world, and he knew it. Nevertheless, at least between 716 and 719 Radbod was able to maintain his independence, no doubt in part because Charles Martel was occupied in the west. But we should not ignore the reality of his defeat of Charles at Cologne, nor of the threat that this posed to Austrasia. At the same time we should not reduce Frankish-Frisian relations simply to the dealings of Aldgisl and Radbod with the maiores palatii, Ebroin, Pippin, Grimoald and Charles Martel, and with the missionaries they supported. Here the family history that begins the Life of Liudger is instructive. Altfrid provides us with the example of Wrrsing and his family, but he implies that they were not alone. Members of the elite of Frisia Citerior could and did cross over into Francia, and they also returned. Altfrid tells us that Wrrsing and his family regained their family estates at Zuilen, with the backing of Charles Martel, after the death of Radbod (Vita Liudgeri, I, 4). But the hagiographer also makes it plain that there was communication between Wrrsing and Frisia even during the lifetime of Radbod, who attempted to persuade the self-imposed exiles to return in the years before he died (Vita Liudgeri, I, 3). From the other side of the border we can point to the Engelbert who donated Alphen to Willibrord in 709, and Eersel three years later (Costambeys 1994). Despite the political divisions, the lower Rhineland was a single socio-economic zone in which Franks and Frisians had to co-exist.

Discussion MOL   I am interested in the relationship between Radbod and Charles Martel, and what happens after the death of Pippin; there appears to be a determination for the relationship to continue. But nothing is said about succession. The question is what is happening there? WOOD  Vita Vulframni records that Radbod had a son who converted to Christianity, but died shortly after. It would seem that Radbod had no other heir, except his daughter, who married Grimoald. MOL  Was Charles expecting to succeed Radbod after the latter’s death? WOOD  Probably, but we do not know enough about the fate of Radbod’s daughter, who may have been under the control of Charles. MOL  In North-Holland can see a lot of property that becomes royal domain, by succession or confiscation. WOOD  It is tempting to think that Radbod’s estates passed to Charles. FLIERMAN  A point regarding the political manoeuvrings after Pippin II’s death: we know that in 716 Charles Martel moved to Neustria, leaving Austrasia, the region bordering on Frisia, wide open. We also know that Radbod was very strategic in picking his alliances. So could we deduce from this that there was some sort of alliance between the two at that point? WOOD  My guess is that Charles, who had been defeated by Radbod, decided to abandon the Frisian border region, to concentrate on problems elsewhere, but in the knowledge that Radbod had no heir. DE LANGEN  I suggest an alliance; should we talk about peace between the Frisians and Franks rather than conflict? Do the sources shift the emphasis too much on to the latter? 215

Z01 Hines Book B.indb 215

07/05/2021 13:30

Frisians of the Early Middle Ages WOOD  To my mind, although there was a considerable amount of low-level violence, in terms of major military conflicts the seventh century was actually the most peaceful century in European history. DE LANGEN  I would certainly emphasize negotiations, and a possible alliance; Radbod came to an agreement with Pippin quickly. IJSSENNAGGER-VAN DER PLUIJM  We need to remember that how Radbod is depicted and the role he plays in Frankish politics represents just one level. It involves one person, and political and family ties become very much interrelated. These high-level social ties may not necessarily reflect the same level as the archaeological evidence that you would find throughout Frisia. I think these separately represent different elements of one society. Radbod of course was a very special character: the only one we know in this detailed way. WOOD  The archaeology clearly concerns different aspects of Frisian history from those recorded in our written sources. IJSSENNAGGER-VAN DER PLUIJM  What I found really interesting is what you highlight in the Minor Frankish Annals: the defeat of Charles and the naval expedition. Do you think this tradition was deliberately suppressed elsewhere? Or how do you see this? WOOD  That is hard to explain or understand. My own view, however, is that the Carolingians edited the history of this period very substantially. IJSSENNAGGER-VAN DER PLUIJM  I can see how that may have worked in the case of the depiction of Radbod. But the naval expedition is a different story. Do you think the naval element would not fit with the Frankish framework – was it un-Frankish? WOOD  I wonder if the decision to write about a naval attack may have been related in some way to the story of Boniface’s martyrdom. FLIERMAN  Could you attack Frisia by sea without Frisian aid? NIJDAM  There are later sources, more literary sources, which present Radbod as a character in rather different way: not least chansons de geste, in which he is turned into an ally of the Franks. We can link these up with the eleventh- to early twelfthcentury historian Sigebert of Gembloux in date. WOOD  In Radbod’s last years there was clearly no conflict with Charles Martel: whether or not that implies an alliance I don’t know. Perhaps there were relatives of Radbod who found it diplomatic to claim that there had been an alliance. MOL  The naval campaign makes practical sense. North-Holland has to be held by land forces, but the other side of the Vlie is very different. Was the campaign all that successful in fact? WOOD  I am just struck by how little discussion of the naval campaign there has been. VERSLOOT  Some battles seem to be decisive, and others simply are not. What does it actually mean that Radbod won one battle? And how big were the armies? WOOD  I would reckon that the armies of seventh-century Francia were quite small. We hear mainly of rival factions fighting each other. The arrival of the Islamic invaders changes things, but initially the Franks have to rely on Lombard help (something we only hear of in Lombard sources). Charles Martel really needed to go in for military mobilization. 216

Z01 Hines Book B.indb 216

07/05/2021 13:30

Franks and Frisians MOL  Every region had a call-up system. Even a naval campaign would require this in a way. WOOD  Agreed, but calling-up the army was possible from the sixth century onwards. It didn’t happen very often, though. The evidence we have for raising troops from church lands relates to the 730s onwards. MOL  I would suggest that some sort of thing or juridical organization would coordinate responses when danger arose. We can see this in action in Later Middle Ages, but it should have been there at this early date too. WOOD  The first reference to the Carolingian Marchfield dates to the mid-eighth century … it just wasn’t called upon. DE LANGEN  It’s also a question of level, isn’t it? There is a new level of activity, and innovation becomes a factor. WOOD  Innovation is important, but the Carolingians wanted to present things as traditional. LOOIJENGA  What was the language that the Frisians and Franks used? Would it have been Latin? WOOD  Further away from the Rhine, less Latin would have been used. Charlemagne knew a variety of Germanic language, and I would suspect that Radbod would have known some Latin. The merchant class in Utrecht must have been able to use both languages. FLIERMAN  Old High German was used in poetry. WOOD  There is actually little to show that the Merovingians could speak any Germanic language. VERSLOOT  Frisian and Frankish variants of Germanic should have been mutually intelligible up to around the year 700. May I ask a question about sources? The experience of scholarly philology is, that when new fragments of material appear, we tend to get yet another copy of known texts rather than something entirely new. Does that extend to the historical sources as well? Were there dozens of chronicles or one or two traditions that are copied over and over again? What about possible alternative sources, such as oral tradition? WOOD  This is a fair point. Most of our information fits into an intellectual genealogy. Individual centres add bits of local knowledge to the Annals. There may have been more traditions, which have not survived: the memory of each power group must have been preserved in centres such as monasteries. You do get alternative traditions, for instance in Aquitania and Neustria. FLIERMAN  You mentioned that Gregory of Tours – our main history for the sixth century – does not mention Frisians, which is noteworthy in itself. I would add to this that Fredegar’s Chronicle – our main history for the seventh century – doesn’t mention them either, which is all the more striking when we consider that Fredegar was very interested in gentes: Romans, Saxons, Burgundians, Suevi, Wends – they’re all there, but no Frisians. So what do we make of this? WOOD  There is indeed a problem, because we have no evidence apart from the Audulfus coin, if it is genuine. IJSSENNAGGER-VAN DER PLUIJM  I believe there have been doubts about that coin? DE LANGEN  Yes, there are discussions about the tradition it represents (Grierson 1974, 153–5; Faber 1998; Bazelmans 2000, 57–8). 217

Z01 Hines Book B.indb 217

07/05/2021 13:30

Frisians of the Early Middle Ages WOOD  As far as we have it, the history of Frisia in this period comes into focus with the Vitae of the missionary saints. FLIERMAN  But Fredegar does talk extensively about what is going on across the Rhine, and he seems to have a consistent view of the ethnic landscape of the trans-Rhine regions. Frisians were not part of that landscape, as far as Fredegar was concerned. WOOD  This is not so different from the picture we find in the fourth-century Roman sources. IJSSENNAGGER-VAN DER PLUIJM  Anglo-Saxon sources do mention Frisians. WOOD  Not before late seventh century; Vita Willibrordi is the earliest. NIJDAM  Returning specifically to Arjen’s question: Saints’ Lives are the genre in which new information appears. Vita Vulframni remains very interesting as a source. I note what Lebecq suggested (2007), that the information is genuinely from Frisia: especially chapters 9 and 10, on Radbod’s aborted baptism and his dream. There could have been a local song with this tradition. It fits perfectly into a late eighth-century setting. WOOD  I do agree with Lebecq that Wulfram’s mission took place in the 690s. I also think that Wulfram did bring people back with him from Frisia. I cannot be sure that connections with Frisia continued, although it’s likely. But why does Vita Vulframni make its hero later than Willibrord? Maybe Alcuin’s Vita carried so much authority. NIJDAM  The change made the story more favourable from a Christian perspective. There are some strange details in the lists of names, and what they mean. Isn’t it odd too for Radbod’s son to have the same name as his father? WOOD  I agree. Kuningus may indeed be the king’s son. NIEUWHOF  I would note that Frisians do not appear historically until the seventh century. We are then actually dealing with a new population, who were in reality ‘Anglo-Saxons’. It may have been as Bazelmans argues, that the name was given by the Franks from historical sources; or that it came from the name of the area. WOOD  That’s possible. HINES  Procopius does refer to Φρισσόνες in this area in the sixth century, but we probably shouldn’t go into the question of what authority that reference may or may not have. He uses the same root form with an i, Fris-, as the early Latin sources, unlike the Germanic form with a long ē, which appears to be original: ē to ī is a general early Celtic sound change. I’d draw particular attention to the distinction of Frankish sources using the term dux while Anglo-Saxon sources use the term rex. Those Anglo-Saxon sources do not use that terminology in a careless or random manner. Similarly, Radbod’s acceptance of his son converting to Christianity: there is a strong parallel with Penda in Mercia allowing Peada to do that and marry a Christian princess (Bede, HE, III, 21). The Anglo-Saxon dimension in the thinking of Aldgisl and Radbod is an issue of interest. The Franks were the threat, but the Anglo-Saxons must have offered a model. To what extent might the missionary activities from England represent the English powers seeking to increase influence there, to establish the Frisians as a useful ally while keeping the border to the Franks at a greater distance? 218

Z01 Hines Book B.indb 218

07/05/2021 13:30

Franks and Frisians WOOD  Certainly the Anglo-Saxons need to be borne in mind when considering the conversion of Frisia. The contacts went back a long way. Yet the Franks were also a factor. The Anglo-Saxons must have been thinking about a complex set of relations across the southern North Sea. IJSSENNAGGER-VAN DER PLUIJM  I see that this relationship continues right up and into the Viking Age, with Frisia being significantly situated between the east and the west as well as the south and the north. VERSLOOT  From an English perspective, where there were small kingdoms, Frisia was a kingdom, but in the Frankish perspective it was too small. WOOD  There are, however, powerful duces. HINES  There’s a well structured semantic field within which the titles are hierarchical. A dux could be more powerful, yet still of lower rank or status, than a king. In the Anglo-Saxon/Insular context we see some playing around with higher titles when uses of Imperator come in. NIEUWHOF  Could one person be dux and rex at the same time? HINES  In England sub-regulus is very frequent; there are grades of kingship. WOOD  It is worth consulting James Campbell’s comments on the consistent use of terms of kingship in Anglo-Saxon England (1979). IJSSENNAGGER-VAN DER PLUIJM  In the case of religion, I wonder, when the first missions came and missionaries were invited to come and preach, was the religion itself not seen as a threat? Was it perhaps the political friction, and the connection of religion with politics, that were particularly problematic? WOOD  The nature of Germanic ‘paganism’ is a massive problem. Our evidence for the Rhineland religion of the pre-sixth century actually involves the Deae Matres. Traditions relating to the so-called Germanic pantheon that becomes so clear in later Scandinavian sources do not survive from this period, and the earliest all come from considerably to the east of the Rhine. But there are lots of Rhineland inscriptions which tell us about other deities. Also there is the evidence of Bede. NIJDAM  It is a bold suggestion that that pantheon did not exist then! WOOD  I would think that there were numerous pagan traditions. NIJDAM  Who the main gods is the problem for Frisians. WOOD  There is Fosite, and there is the slightly earlier goddess Nehalennia (Stuart and Boegars 2001). MAJCHCZACK  It would be interesting to look into the nature of the networks before and after Christianization. Prior to conversion we would suppose more local Gefolgschaften; what the Church brings is something wider. DE LANGEN …something more stable. WOOD  I would emphasize the importance of the transfer of landed property to the Church. Roman temples did not have land; only treasure. A religious organization supported by property implies a very new form of stability.

219

Z01 Hines Book B.indb 219

07/05/2021 13:30

Frisians of the Early Middle Ages

References Primary sources Alcuin, Vita Willibrordi. Ed. W. Levison 1920, MGH Scriptores, SS Rer. Merov. 7 (Hannover). — Vita Willibrordi. Ed. H.-J. Reischmann 1989, Willibrord Apostel der Friesen (Sigmaringendorf). Altfrid, Vita Liudgeri. Ed. W. Diekamp 1881, Die Vitae Sancti Liudgeri, Geschichtsquellen des Bistums Münster 4 (Münster). Annales Alamannici. Ed. G. H. Pertz 1826, MGH Scriptores, SS 1 (Hannover). Annales Petaviani. Ed. G. H. Pertz 1826, MGH Scriptores, SS 1 (Hannover). Annales Laureshamenses. Ed. G. H. Pertz 1826, MGH Scriptores, SS 1 (Hannover). Annales Laurissenses minores. Ed. G. H. Pertz 1826, MGH Scriptores, SS 1 (Hannover). Annales Mettenses Priores. Ed. B. von Simson 1905, MGH Scriptores, SS rer. Germ. 10 (Hannover). Annales Nazariani. Ed. G. H. Pertz 1826, MGH Scriptores, SS 1 (Hannover). Annales sanctae Columbae senonensis. Ed. G. H. Pertz 1826, MGH, SS 1 (Hannover, 1826) Annales sancti Amandi. Ed. G. H. Pertz 1826, MGH Scriptores, SS 1 (Hannover). Annales Sangallenses Baluzii. Ed. G. H. Pertz 1826, MGH Scriptores, SS 1 (Hannover). Annales Sangallenses Breves. Ed. G. H. Pertz 1826, MGH Scriptores, SS 1 (Hannover). Annales Tiliani. Ed. G. H. Pertz 1826, MGH Scriptores, SS 1 (Hannover). Anso of Lobbes, Vita Erminonis. Ed. W. Levison 1913, MGH Scriptores, SS rer. Merov. 6 (Hannover). Bede, Historia Ecclesiastica Gentis Anglorum. Ed. C. Plummer 1896, Bedae Opera Historic. vol. 1 (Oxford). — Historia Ecclesiastica Gentis Anglorum. Eds. and trans. B. Colgrave and R. A. B. Mynors 1969, Bede’s Ecclesiastical History of the English People (Oxford). Boniface, Epistolae. Ed. M. Tangl 1916, S. Bonifatii et Lulli Epistolae. MGH Epistolae, Epp. 3, VI (Berlin). Calendar of St Willibrord. Ed. H. A. Wilson 1918 (London). Chronicon Moissiacense/Chronicle of Moissac. Ed. G. H. Pertz 1826, MGH Scriptores, SS 1 (Hannover). Epistolae Austrasicae. Ed. W. B. Gundlach 1892, MGH Epistolae, Epp. III (Berlin). Fredegar Continuator. Ed. J. M. Wallace-Hadrill 1960, The Fourth Book of the Chronicle of Fredegar (London). Gesta Abbatum Fontanellensium. Ed. P. Pradié 1999, Chronique des abbés de Fontanelle (Paris) Gregory of Tours, LH (Decem Libri Historiarum). Eds. B. Krusch and W. Levison 1951, MGH Scriptores, SS rer. Merov. 1 (Hannover). Heidrich, I. ed. 2011, Die Urkunden der Arnulfinger, MGH Diplomata (Hannover). Lex Ribvaria. Eds. F. Beyerle and R. Buchner 1954, MGH Leges, LL nat. Germ. 3,2 (Hannover). Liber Historiae Francorum. Ed. B. Krusch 1885, MGH Scriptores, SS rer. Merov. 2 (Hannover). Paul the Deacon, Historia Langobardorum. Ed. G. Waitz 1878, MGH Scriptores, SS rer. Germ. 48 (Hannover, 1878). Procopius, History of the Wars. Ed. H. B. Dewing 1914–28 (Cambridge, MA). Sigibert, Vita Landiberti. Ed. B. Krusch 1913, MGH Scriptores, SS rer. Merov. 6 (Hannover). Stephen of Ripon, Vita Wilfridi. Ed. B. Colgrave 1927, The Life of Bishop Wilfrid by Eddius

220

Z01 Hines Book B.indb 220

07/05/2021 13:30

Franks and Frisians Stephanus (Cambridge). Venantius Fortunatus, Poèmes. Ed. M. Reydellet 2002–4, 3 vols. (Paris). Vita Landiberti vetustissima. Ed. B. Krusch 1913, MGH Scriptores, SS rer. Merov. 6 (Hannover). Vita Vulfhramni. Ed. W. Levison 1910, MGH Scriptores, SS Rer. Merov. 5 (Hannover). Willibald, Vita Bonifatii. Ed. W. Levison 1905, Vitae sancti Bonifatii, MGH Scriptores, SS rer. Germ. 57 (Hannover). Secondary sources Bazelmans, J. 2000, ‘Het laat-Romeinse bewoningshiaat in het Nederlandse kustgebied en het voortbestaan van de Friezennaam’, Jaarverslagen van de Vereniging voor Terpenonderzoek 76–82, 14–75. — 2009, ‘The early-medieval use of ethnic names from Classical Antiquity. The case of the Frisians’. In Ethnic Constructs in Antiquity: The Role of Power and Tradition, eds. T. Derks and N. Roymans (Amsterdam), 321–38. Campbell, J. 1979, Bede’s Reges and Principes, Jarrow Lecture (Jarrow). Collins, R. 1998, Charlemagne (London). Costambeys, M. 1994, ‘An aristocratic community on the northern Frankish frontier 690– 726’, Early Medieval Europe 3, 39–62. Dierkens, A. 1985, Abbayes et chapitres entre Sambre et Meuse (VII–XI siècles) (Sigmaringen). Faber, K. P. H. 1998, ‘Audulfus, een Friese koning’, Fryslân 4 (4), 29–30. Fouracre, P. 2000, The Age of Charles Martel (London). Grierson, P. 1973–1974, ‘A new Audulfus Frisia triens’, Jaarboek voor munt- en penningkunde 60–61, 153–5. Knol, E. and N. IJssennagger 2017, ‘Palaeogeography and people’. In Frisians and their North Sea Neighbours. From the Fifth Century to the Viking Age, eds. J. Hines and N. IJssennagger (Woodbridge), 5–24. Lebecq, S. 2000, ‘Vulfram, Willibrord et la mission de Frise: pour une relecture de la Vita Vulframni’. In L’évangélisation des régions entre Meuse et Moselle et la fondation de l’abbaye d’Echternach, ed. M. Polfer (Luxembourg), 429–51. — 2007, ‘Paganisme et rites sacrificiels chez les Frisons des VIIe-VIIIe siècles’. In Bonifatius – Leben und Nachwirken. Die Gestaltung des christlichen Europa im Frühmittelalter, actes du colloque de Mayence, eds. F. J. Felten, J. Jarnut and L. E. von Padberg (Mainz), 111–20. — 2018, ‘Vulfran, clerc palatin, moine de Fontanelle, évêque métropolitain de Sens’. In Religion, animaux et quotidien au Moyen Âge: études offertes à Alain Dierkens à l’occasion de son soixante-cinquième anniversaire. Part 2, eds. M. de Waha, J.-M. Sansterre, J.-M. Duvosquel, A. Wilkin and N. Schroeder (Brussels), 555–68. McKitterick, R. 2000, ‘The illusion of royal power in the Carolingian annals’, English Historical Review 115, 1–20. Meeder, S. and E. Goosmann 2018, Redbad, Koning in de marge van de Geschiedenis (Houten). Nelson, J. L. 1991, ‘Gender and genre in women historians of the early Middle Ages’. In L’historiographie médiévale en Europe, ed. J.-P. Genet (Paris), 149–63. Nicolay, J. A. W. 2014, The Splendour of Power: Early Medieval Kingship and the Use of Gold and Silver in the Southern North Sea Area (5th to 7th century AD), Groningen Archaeological Studies 28 (Groningen).

221

Z01 Hines Book B.indb 221

07/05/2021 13:30

Frisians of the Early Middle Ages Nijdam, H. 2018, ‘Redbad. koning van de Friezen voor eens en altijd’. In Redbad Koning van de Friezen, W. Schoorstra (Leeuwarden), 7–17. Nijdam, H. and O. S. Knottnerus 2019, ‘Redbad: the once and future king of the Frisians’. In Northern Myths, Modern Identities: the Nationalisation of Northern Mythologies since 1800, eds. M. Baár and S. Halink (Leiden), 87–114. Reischmann, H.-J. 1989, Willibrord Apostel der Friesen (Sigmaringendorf). Stuart, P. and J. E. Boegars 2001, Nehalennia: Römische Steindenkmäler aus der Oosterschelde bei Colijnsplaat (Leiden). Van Egmond, W. 2005, ‘Radbod van de Friezen, een aristocraat in der periferie’, Millennium 19, 24–44. — 2010, ‘Utrechts oudste kerk en Dagobert: Vraagtekens bij een brief van Bonifatius’, Millennium 24, 95–112. Werner, K.-F. 1973, ‘Les principautés périphériques dans le monde franc du VIIIe siècle’, Settimane di Studio 20, I problemi dell’Occidente nel secolo VIII, 483–514. Wood, I. 1990, ‘The Channel from the 4th to the 7th centuries AD’. In Maritime Celts, Frisians and Saxons, ed. by S. McGrail, Council for British Archaeology Research Reports 71 (London), 93–9. — 1994, The Merovingian Kingdoms 450–751 (Harlow). — 2001, The Missionary Life: Saints and the Evangelisation of Europe, 400–1050 (Harlow).

222

Z01 Hines Book B.indb 222

07/05/2021 13:30

• 8 • MIRROR HISTORIES: FRISIANS AND SAXONS FROM THE FIRST TO THE NINTH CENTURY AD Robert Flierman

F

or the first centuries of their documented existence, neither Frisians nor Saxons are particularly well defined. They appear in Roman writing as two among a host of barbarian groups active at the northern periphery of the empire, sometimes as enemies of the Romans, sometimes as allies or soldiers in the Roman army. Descriptions are short and circumstantial, and tend to conform to deeply ingrained ideas about the uncivilized inhabitants of the North, who were expected to be, by turn, strong, brave, stubborn, tempestuous, cruel and prone to excess. Few educated Romans who picked up Tacitus’ Historiae would have travelled beyond the northern frontier, but they would have understood a story about the inhabitants of Cologne, who were able to overcome a formidable fighting force made up of Chauci and Frisians by plying these Germani with copious amounts of food and drink (Historiae, 4.79). Some centuries later, the Gallic bishop Sidonius Apollinaris was still upholding a venerable literary tradition when he treated one of his sea-bound friends to an unsolicited exposé on the religious customs of the Saxon pirates presently threatening the coasts of Gaul: among other unsavoury practices, these Saxons were said to sacrifice every tenth one of their human prisoners (Epistulae, 8.6.14). In so far as Roman observers located them in space, Frisians and Saxons were typically linked to the coastal areas of the North Sea, with both groups being associated with seafaring and piracy. Tacitus’ Germania (c. AD 98) offers a slightly more precise area of habitation: the Roman historian mentions two Frisian nationes, whom he specifies as the minor and the major Frisians and locates in the wetlands between the Rhine and the ocean (ch. 34). Significantly, Tacitus’ extensive ethnography of the peoples of Germania failed to mention Saxons. Some fifty years after Tacitus, the Alexandrian cartographer Ptolemy did mention them, assigning the Saxons to the Jutland Peninsula above the Elbe, with the Frisians being once again located on the Dutch coast left of the river Ems. The two groups were separated, in Ptolemy’s view, by the Chauci, who inhabited the coastal regions between the Ems and the Elbe (Geographia, 2.11.11). The absence of Saxons from Tacitus’ Germania fits a more general pattern. From a Roman perspective, Frisii and Saxones belonged to different timeframes: the former are attested in Roman writing from the first to the third centuries AD (Seebold 2001), whereas the latter become a recurring presence only from the fourth century onwards (Flierman 2017). Ptolemy is in fact the only surviving Roman author to mention Frisians and 223

Z01 Hines Book B.indb 223

07/05/2021 13:30

Frisians of the Early Middle Ages Saxons in the same work, and the very earliness of his reference to the Saxons has led some modern commentators to designate it a scribal error (Springer 2004; Kahrstedt 1934), though this palaeographical argument collapses under close scrutiny (Seebold 2012). Looking at Frisians and Saxons through the lens of Roman writing, two questions present themselves. First, what sort of historical entities were behind the labels ‘Frisian’ and ‘Saxon’ in such writing? Can we speak about distinct Frisian and Saxon peoples in this period, and if not, at what point can we do so? Second, what can be said about the relationship between these two groups, considering that few Roman sources explicitly tie them together? Variations on these two questions remain relevant well into the Middle Ages. They constitute the backbone of this paper, which roughly covers the period AD 1–900.

Frisii and Saxones (AD 1–600) It is difficult to make far-reaching claims about the self-perception of those labelled Frisii and Saxones by Roman observers. On the whole, the Frisian case looks slightly more promising. Extensive archaeological research in areas of alleged Frisian habitation, above all the terp region of modern Friesland and Groningen, has left little doubt that the fertile salt marshes along the Dutch coasts were widely inhabited, at least up to the third century AD (e.g. Knol and IJssennagger 2017). ‘Frisia’ in this period was a patchwork of not-quite-egalitarian communities of farmers, mostly self-sufficient but connected to each other by bonds of trade, social competition and ritual (Nieuwhof 2015). An explicit textual reference to Frisian political organization can be found in Tacitus’ Annales (13.54), which describes the barbarian antics of two Frisian ‘kings’ who went to Rome to enter into negotiations with the emperor Nero (d. AD 64). Their names – Verritus and Malorix – would suggest both hailed from a Celtic-speaking community (Schrijver 2017). The case for a distinct Frisian self-awareness may well have its strongest prop in the epigraphical evidence, i.e., the half a dozen or so inscriptions, found in Rome as well as along the German and British Limes, that were commissioned by, or for, soldiers who were natione Frisius or natione Frisiaeo (Galestin 2007/2008). These inscriptions testify to the existence of individuals in the first and second centuries AD who not only claimed to be ‘of Frisian origin’, but also considered such an affiliation meaningful enough for it to be commemorated, be it in a Roman social context and alongside other identities (e.g., that of a citizen or soldier). Of course, the big problem with any diachronic study of the Frisians is the question of continuity. The ethnonym Frisii disappears from the textual record by the end of the third century, to reappear only in the sixth (Bazelmans 2009). This disappearance has been linked to marked discontinuities in the archaeological record of the northern Netherlands for this period, hinting at a strong decline in habitation during the fourth century, followed by an influx of immigrants from northern Germany in the fifth (Nicolay 2005; Knol 2009; Nieuwhof 2011; 2013). Population decline was not equally extreme everywhere: Friesland is thought to have become virtually deserted, whereas Drenthe and Holland, and to a lesser extent Groningen, seem to have retained some degree of habitation (van Es 1990; Taayke 1996; Dijkstra and de Koning 2017). Nevertheless, it is assumed nowadays that the Frisii or Fresones that eventually turn up in the Early-medieval sources in relation to the northern Netherlands were the result of a new 224

Z01 Hines Book B.indb 224

07/05/2021 13:30

Frisians and Saxons process of ethnogenesis, in which those labelled Frisians in the Roman period played only a minor role. The ancient Saxons come with their own set of complications. Certainly, the coastal marshes of the trans-Elbe region were densely populated in the second century AD, when Ptolemy came to designate it Saxon land (Capelle 1998). The same goes for the coasts and Pleistocene hinterland of the Elbe-Weser triangle (Meier 2003), which Ptolemy had associated with the Chauci, but which are thought to have become Saxon territory in the century thereafter (Udolph 1999). This Saxon ‘expansion’ into the ElbeWeser triangle has traditionally been inferred from changes in the material culture of the region. Between the third and the fifth centuries AD, its hinterland witnessed the rise of new large-scale cemetery fields with distinctive urn-types (Böhme 2003). The deposition of prestigious gold and silver ornaments, as grave goods (late fourth century) and subsequently also as part of hoards (fifth century), suggests the presence of well-connected elites eager to distinguish themselves (Nicolay 2014, 140–68). Part of their social standing appears to have derived from contact with the Roman world and service in the Roman army, as is attested by a substantial amount of Roman-type weapons, brooches and belts found in fourth-century graves (Böhme 1999). The question is what such burial rites and other material developments have to do with the ethnic background of the inhabitants, with their claiming a specific Saxon identity. Most archaeologists today would say very little (Ludowici 2019). The textual record is not much help either when it comes to locating ‘the Saxons’ in space. Apart from Ptolemy, few ancient authors explicitly associate the ethnonym Saxones with Germania. A rare fourth-century reference to Franks and Saxons being ‘the most warlike of the peoples who live beyond the Rhine and on the shores of the western sea’ (Julian, Orations 1:34D; Zosimos, 3.6.1–2) is as precise as it gets. It is only at the end of the sixth century that Saxons start to be attested near the Weser (Gregory, Historiae, 4.16). There is, at any rate, very little written evidence for a specific Saxon self-awareness in the Roman Period. We have no inscriptions of individuals claiming a Saxon identity, no references to Saxon personal names, no descriptions of Saxon political organization. What we have, starting in the middle of the fourth century, is a series of historiographical and panegyric references to Saxon (naval) attacks on the Roman frontiers. Gaul is the initial target (Eutropius, Breuiarium, 9.21) but from c. 380 onwards we also hear about Saxon raids on Britain (Ammianus, Res Gestae, 26.4.5; Claudian, De consulatu Stilichonis, 2.250–1). Around the middle of the fifth century, Continental commentators start to allude to Saxon settlement on the Isles (Chronicle of 452, p. 80; Constantius, Vita Germani, ch. 18), culminating in Gildas’ tearful lamentations on the invasion of the Saxons, ‘hateful to God and man’ (De excidio, ch. 23). While Roman authors typically refer to Saxons as a gens or ethnos, the chronological and geographical variety of Saxon activity in this period, coupled with the sheer ubiquity of the term Saxones in relation to any sort of barbarian activity around the Channel, would suggest that these were not the doings of a single ethnic group or political unit. Rather, as Matthias Springer has argued, Late-antique usage of the ethnonym Saxones was not unlike the ninth-century label Vikings: an umbrella term of sorts, used to denote the actions of highly situational Raubscharen or raiding parties (Springer 2004, 46). Bluntly put, any barbarian on the Channel in the fifth century could find himself labelled a Saxon, regardless of his actual ethnic background. But this is not to say that the Late-antique Saxons were a Roman literary fiction. Indeed, the continued attestation of the Saxon name in Early-medieval 225

Z01 Hines Book B.indb 225

07/05/2021 13:30

Frisians of the Early Middle Ages Britain as well on the coasts of the Loire and Seine suggests that quite a few migrant communities claimed this name for their own: they considered themselves Saxons or eventually came to perceive themselves as such (Flierman 2017, 53–87). The point is rather that such ‘Saxonness’ did not necessarily originate in bonds of ancestry, be they real or perceived, but could also be the result of shared military activity (pirates, mercenaries) or conscious association (following a ‘Saxon’ leader or elite). Considering the complex nature of the categories ‘Frisian’ and ‘Saxon’ in Antiquity, what can be said about the relationship between the two in this period? At first glance, this question seems to run into issues of chronology: Saxons become a frequent presence only in the fourth century, when Frisians have vanished from the textual record, and to a large extent, from the material record. On closer consideration, one might well ask whether the disappearance of the Frisian name, followed by the increasingly widespread use of the label Saxones, was not on some level related. It is evident that Late-antique conceptualizations of the world beyond the Rhine were patchy and preoccupied with groups that were of immediate interest. The ‘Franks’ and ‘Saxons’ threatening the Rhine frontier and coasts of Gaul in the fourth century were of great interest, the dwindling communities of the terp region less so. When we witness the Franks and later the Saxons replace the earlier Frisians and Chauci as typical maritime peoples and coast-dwellers in Late Roman panegyric and historiography (Hiddink 1999, 223–6), we are at least partly witnessing a recalibration of Roman priorities (Wood 1990). The historian Orosius (d. 420) lived in a world where ‘Saxon’ attacks posed a very real threat, so for him ‘the coasts and inaccessible swamps of the Ocean’ were inhabited by the Saxons (Aduersum paganos, 7.32.10). In reality, the ethnic landscape of the North Sea coasts was probably more diverse. We don’t know how those few remaining inhabitants of Drenthe and Groningen in the fourth century identified themselves – they could have continued to think of themselves as Frisians – but they were occluded from the textual record by a Roman preoccupation with Franks and Saxons. Frisian involvement in the raids on the Channel or the migrations to Britain could likewise have been hidden from view by generic references to Saxones (Taayke 2000, 20–1), though, admittedly, archaeological and toponymical support for a Frisian participation in the adventus Saxonum has proved inconclusive so far (Hines 2001; Bremmer 1981). We are in fact better served looking for Frisio-Saxon interaction in the northern Netherlands, where starting at the end of the fourth century we see a number of marked changes in the material record. Not only are there increasing signs of renewed habitation in previously abandoned regions, but such habitation is accompanied by the adoption of new pottery styles and ornaments, the introduction of new burial practices, and changes to the lay-out of settlements. These developments and their regional implications have been studied intensely over the past two decades, so I will content myself here with four short observations. First, that most would agree that these changes were at least partly due to the arrival of migrants. Second, that materially speaking, many of the new objects and practices attested in the northern Netherlands show a close connection to the Elbe-Weser region (Schleswig-Holstein, Jutland and Scandinavia being other parallels), indicating that at least some of the migrants hailed from that region. Third, that if we assume the Elbe-Weser region was indeed Saxon territory in Late Antiquity, then at least some of the new inhabitants of what are now the provinces of Friesland and Groningen were Saxons. Finally, that around AD 700 the inhabitants of Friesland and Groningen were nevertheless known as Frisians, not as Saxons (or 226

Z01 Hines Book B.indb 226

07/05/2021 13:30

Frisians and Saxons Angles, Jutes or Danes, for that matter). What are we to make of this? It is bound to remain undecided whether the Frisian name stayed in continued use from Antiquity onwards or whether it was retrieved from obscurity at some point by the new inhabitants of the Netherlands, possibly with the Franks acting as cultural brokers (Bazelmans 2001; 2009). Yet it can be said that the timing of the reappearance of the Frisian name in the written sources does not appear random. It coincides, at least approximately, with the rise of several regional or even supra-regional power-blocks in the Netherlands c. AD 600, who had not just each other to contend with, but would soon also enter into prolonged competition with their Frankish neighbours (Nicolay 2014, 355–9). There was thus a political elite in this period for whom affirming, or re-affirming, a distinct ethnic affiliation would have been useful and salient, as means to validate their newfound claims to power as well to set themselves apart from an external rival.

Neighbours and allies (AD 600–800) The histories of the Frisians and Saxons enter a new phase around the end of the seventh century, when the lands across the Rhine become the target of Frankish military expansion. The Frankish conquest of Frisia was enacted for the most part under the Carolingian mayors of the palace Pippin II (d. 714) and Charles Martel (d. 741), though Carolingian control over the lands between the Sincfal and Weser was not fully established until the end of the eighth century (Reuter 1995) and would become contested once more with the ninth-century Viking invasions (IJssennagger 2013). Carolingian subjugation of the Saxon lands between the Rhine and Elbe was equally protracted: Franco-Saxon hostilities continued intermittently for most of the eighth century, until Charlemagne finalized the conquest of his north-eastern neighbours through three decades of violent campaigning (772–804; Rembold 2017, 39–84). In both cases, conquest was accompanied by missionary activity and set in motion a process of Christianization (Wood 2001; Büttner 1965). The above sketch would make it appear as if by the start of the eighth century Frisia and Saxony had become well-defined territorial entities, inhabited by unified gentes who were duly incorporated into the Frankish and Christian world. This is indeed how their Carolingian conquerors liked to picture things, and with the Saxons, at least, the Frankish administrative insistence on Saxon unity made it easier for the region’s leading families to see themselves as members of a single gens Saxonum (Becher 1996, 26–66; Shuler 2010; Flierman 2017, 119–62). But such unity was the end-result of a longterm process, which in the seventh and eighth centuries was still in full swing. Indeed, the very duration of Frankish military activity in Frisia and Saxony seems to belie the existence of a unified adversary in those regions. These were not centralized regna like Bavaria or the Lombard kingdom, which Charlemagne was able to incorporate in one or two years by means of a dynastic take-over. There is considerable debate about what the political organization of pre-conquest Frisia would in fact have looked like. A significant increase c. AD 600 in the deposition of gold coins, jewellery and other prestigious items has been tied to the emergence of powerful regional elites, with central places developing in northern Westergo, around the Rhine estuary and possibly the central Netherlands (Knol 1993; Dijkstra 2011; Nicolay 2014). Frankish and Anglo-Saxon sources refer to two king-like figures active in the late seventh century, Aldgisl and Radbod, who had enough clout to challenge the 227

Z01 Hines Book B.indb 227

07/05/2021 13:30

Frisians of the Early Middle Ages North-Frankish aristocracy for control of the economically crucial Rhine and Meuse deltas (Halbertsma 2000). The question is whether they did so as regional strongmen, whose core territory was situated either around Utrecht (Bazelmans et al. 2004), North-Holland (van Egmond 2005) or Westergo (Nicolay 2015), or whether their rulership would have had a more supra-regional character (Heidinga 1999). A comparison with the Saxon situation is useful here, not because the situations were necessarily the same, but because it underlines that we are dealing with dynamic societies, whose political and social organization could be subject to rapid change. By the end of the seventh century, the Saxon sphere of influence had extended across the Weser towards the Ems and Rhine. Describing the earliest Anglo-Saxon missionary efforts in this western section of Saxony, Bede (d. 735) famously claimed that the ‘Old Saxons’ did not have a king, but were governed instead by ‘a great many satraps’ (satrapas plurimos), who each ruled in their own locality but could unite under a warleader at times of peril (HE V, 10). A Bible scholar by heart, Bede adopted the term satrapa [sic] from the independent Philistine leaders of 1 Samuel, a clever association that could have signalled both the otherness of the Continental Saxons and their status as potential converts (Becher 1999; Wood 2003; 2015). Late eighth-century (Frankish) annals paint a somewhat different picture of Saxon organization, distinguishing three major Saxon groups in the area between the Rhine and Elbe (Westphalians, Angrarii and Eastphalians), with a fourth group (Nordliudi) being occasionally attested in the trans-Elbe region (ARF, s.a. 775, 779, 780, 784, 798, 799, 810). In all likelihood, these were recent conglomerates that had formed over the course of the eighth century in the face of mounting Franco-Saxon hostilities (Becher 1999, 24–5). The Westphalians, Angrarii and Eastphalians were still meaningful enough sub-groups around the turn of the century to appear in a Saxon capitulary issued in 797 (Capitulare Saxonicum, praefatio, ch. 9) as well as in the Lex Saxonum (chs. 47–8) codified in 802. Significantly, the Lex Saxonum mentions only one legal point on which the three groups differentiated: when it came to dowries, the Westphalians followed the East Frankish Lex Ribvaria (Faulkner 2016, 76–7); a marked contrast with the recurring territorial differences in Lex Frisionum. While the three sub-groups continue to be mentioned on occasion in the succeeding centuries, they play no role in the Saxon hagiographies and family histories that started to circulate around the middle of the ninth century. As said, the main ethnic identification of the post-conquest elite was with the gens Saxonum (Flierman 2017, 119–62). The Frankish annals also mention several Saxon leaders during the Saxon Wars: Hessi for the Eastphalians, Bruno for the Angrarii, and Widukind as ‘one among the leading men of the Westphalians’ (ARF, s.a. 775; Reviser, s.a. 777). Between them, these figures exemplify the various strategies pursued by Saxon leaders in the face of the social disruption brought on by the Saxon Wars. Hessi, for instance, submitted to Charlemagne as early as 775. He subsequently converted to Christianity and was appointed a count in Saxony, before retiring to the monastery of Fulda to ‘soldier for the Lord’ (Vita Liutbirgae, ch. 1; Krüger 1950, 84–9). Widukind, by contrast, chose to resist the Franks and their religion, and this decision in itself may have turned him into a leader of supra-regional standing. Like Radbod, Widukind remains a slippery figure. He was certainly a convenient scapegoat for court-related annalists (Rembold 2018, 66–9), who managed to implicate ‘that perfidious Widukind’ in two of the Franks’ most painful military debacles in Saxony without ever specifying the Westphalian’s precise 228

Z01 Hines Book B.indb 228

07/05/2021 13:30

Frisians and Saxons involvement (ARF, s.a. 778, 782; Flierman 2015). But even on a minimalist reading, there are aspects of Widukind’s career that point to a man of wide influence. In 777 he stayed with the Danish king Sigifrid in Jutland (Reviser, s.a. 777). In 784, he is said to have moved the Frisian communities right of the Vlie to rebellion by urging them to ‘relinquish the Christian faith and sacrifice to the idols according to their former error’ (ARF, s.a. 784; Altfrid, Vita Liudgeri, ch. 21). While the reference is late and the image typical – several saintly careers were built on Frisians returning to their idols – it is not out of place in the early 780s, when the Saxon Wars were taking on an increasingly grim character and conversion was becoming less and less optional (Becher 2013). In such a context, paganism may well have become a rallying point for resistance against the Franks (Effros 1993; Wood 2013, 7), which Widukind was able to play into with the Frisians. The fact that the Widukinden owned property near Arnhem (DB, no. 181) could have facilitated his move into Frisia. In the end, the 784 rebellion proved of short duration, for a year later Widukind negotiated to be baptized at the royal palace of Attigny, with Charlemagne acting as his baptismal parent (Lorsch Annals, s.a. 785). This was the sort of staged public ritual that was often made to accompany the pacification of the Carolingians’ most high-profile opponents, yet another hint at Widukind’s elevated status in the 780s. Nevertheless, this status ultimately proved temporary. Widukind’s descendants retained positions of prominence in Carolingian Saxony, but did so alongside several other aristocratic families (Althoff 1983; Schmid 1964). Widukind’s appeal to the Frisians in 784 fits within a wider pattern of shared military action, not just against the Franks but also alongside them. In the years following Widukind’s baptism, Frisian and Saxon contingents joined two of Charlemagne’s campaigns, against the Slavs in 789 and against the Avars in 791 (ARF, s.a. 789, 791). The latter campaign took them along the Danube into modern-day Hungary. The Avars avoided a major confrontation, however, forcing Charlemagne to try again next season. It was not to happen, for the Saxon and Frisian levies never showed up. Instead, news arrived that they had ambushed and killed the Frankish force that had come to collect them. This included the Frankish leader, Count Theoderic, who had earlier led the Frisians and Saxons against the Avars. The incident took place either at the mouth of the Elbe (St Amand Annals, s.a. 792) or, more likely, at Rüstringen, near the mouth of the Weser (Reviser, s.a. 793). Either way, it set off another decade of violent warfare in the Elbe-Weser and trans-Elbian regions. Quite possibly, it is to these years also that we should assign the undated reference to an uprising by the ‘East Frisians’ spearheaded by their enigmatic principes Unno and Eilrat (Vita Liudgeri, ch. 22). While the Frankish annals offer no further clues on this rebellion, continued Frisian resistance can be deduced from the events of 797, when Charlemagne is said to have crossed the ‘swamps and impenetrable places’ of the Elbe-Weser region to get to the ocean (ARF, s.a. 797; Rembold 2017, 42–4). There, according to the contemporary Lorsch Annals, the king was met by Saxons ‘from all regions and corners’ to receive hostages, ‘and [he received] likewise from the Frisians’ (Lorsch Annals, s.a. 797). Looking at such close interaction between Frisians and Saxons in the late eighth century, especially near coasts and major waterways of the north-west of Germany, the question arises what can be said about the borders between these two groups. For the coastal area, the annalistic evidence can be placed alongside Lex Frisionum, now dated to c. 787–93, which distinguishes three Frisian regions by their legal customs: a western region between the Sincfal and Vlie, a middle region between the Vlie and 229

Z01 Hines Book B.indb 229

07/05/2021 13:30

Frisians of the Early Middle Ages Lauwers, and an eastern region between the Lauwers and Weser (Dijkstra 2011, fig. 8.6). The Lex does not specify how far these regions extended southwards, but it should be noted that the densely inhabited salt marshes near the coasts would at this point still have been separated from the Pleistocene hinterlands by peat bogs (Vos and Knol 2015, 164–5, 172–4). It is tempting to see these bogs as a natural border between the Frisian coasts and the Saxon territories inland, at least for the Ems-Weser region. Incidentally, archaeological and linguistic evidence suggests that in c. 800 Frisian habitation did not stop at the Weser, but extended further eastwards towards the Elbe mouth (Kleemann 2002, 350–1; Versloot and Adamczyk 2017). A Frisian presence along the coasts of Schleswig-Holstein and Jutland is similarly attested (Panten 2001). This fits with the Frankish annals which, as we saw above, referred to Frisians sailing up the Elbe against the Slavs (ARF, s.a. 789) and submitting themselves to Charlemagne at the ocean between the Elbe and Weser (Lorsch Annals, s.a. 797). As for the western borders between Frisia and Saxony, the ninth-century vitae of the Frisian missionary Liudger (d. 809) suggest the IJssel (Vita Liudgeri, ch. 13) and the Ems (Vita Liudgeri secunda, ch. 27) functioned as natural boundaries. Paradoxically, these and related hagiographical texts are also among our best evidence for the flexible nature of such borders (Wood 2000). Missionaries like Liudger and Willehad (d. 789) did not limit their evangelizing efforts to either Frisians or Saxons, and much the same could be said for the religious institutions they founded or helped establish. We will take a closer look at some of these institutions in the next section.

Gens commixta (AD 800–900) While ‘Carolingian Saxony’ has a self-evident ring to it, ‘Carolingian Frisia’ is a more contestable appellation. By the second half of the ninth century, Saxony had taken on many of the organizational features of the Frankish heartland: a comital system relying on local Saxon families (Krüger 1950), eight episcopal seats at least partly held by Saxon bishops (Honselmanns 1984), two heavily patronized royal monasteries alongside a whole string of aristocratic monastic foundations (Ehlers 2007), and a religious landscape rapidly filling up with the imported remains of Roman and Gallic saints (Röckelein 2002). Frisia, too, had its counts and duces, but such offices were typically held by Frankish and Saxon outsiders, or by Viking princes with their own autonomous agendas (IJssennagger 2013; Henstra 2012). As for Frisia’s ecclesiastical landscape, there were plenty of missionary churches as well as a fair number of aristocratic donors, but such churches were administered by outside bishoprics and monasteries like Utrecht, Echternach, Werden, Münster, Bremen and Corvey (de Langen and Mol 2017, 7–12). The contrast is no doubt exacerbated by unequal spread of the evidence. There survives a substantial Saxon textual corpus from the ninth century, made up of charters, letters, vernacular and Latin poetry, and above all hagiographies (Löwe 1986). Such a corpus is missing for ninth-century Frisia. Interestingly, Saxon texts from this period have plenty to say about Frisians. Most importantly, they signal that the divide between Saxony and Frisia was frequently bridged by ties of land-ownership, trade, marriage, ecclesiastical organization and sworn loyalty. Even the Carolingian rulers found the divide problematic on occasion. After the Treaty of Verdun (843), which effectively divided the Carolingian realm into Western, Middle and Eastern Kingdoms, the Emperor Lothar I (d. 855) found himself ruling in Frisia but not in Saxony; still, he 230

Z01 Hines Book B.indb 230

07/05/2021 13:30

Frisians and Saxons could count several Saxon families among his aristocratic supporters. Writing a letter of introduction for one of these supporters in 850, he coined the unique phrase gens Saxonum et Fresonum commixta (‘a mixed people of Frisans and Saxons’) to denote the northernmost inhabitants of his realm (Translatio Alexandri, ch. 4). This might have been a shrewd circumlocution on Lothar’s part, which allowed him to claim Saxon allegiance without defying Verdun, but it can just as easily be taken for a simple statement of fact: there was a part of Lothar’s regnum, around the modern border between Drenthe and Westphalia, where Saxons and Frisians, and indeed Franks, were living in close proximity to each other (Kleemann 2002, 351; Wood 2000, 160–2). It is worthwhile, in this regard, to take a brief look at the Saxon supporter receiving the emperor’s letter of recommendation: this was Count Waltbert, who was in fact a grandson of Widukind. Waltbert is first mentioned in a diploma issued in 834 for the church of St Martin in Utrecht (DB, no. 181). We read how Waltbert had ordered his local agents to donate family possessions in Oosterbeek and the Praal (near Arnhem) to St Martin, hoping thereby to obtain salvation for his recently deceased father Wikbert. The diploma, which was drawn up in Oosterbeek, lists both Frankish and Saxon witnesses, as well as an agent named Knut, suggesting a social network that extended into various ethnic groups. In 850, Waltbert’s piety led him to Rome, where he acquired the relics of the Roman saint Alexander for a church he had recently founded on his lands in Wildeshausen, to the south-west of Bremen (Röckelein 2002, 241–59). Upon his return, Waltbert approached the Frankish monastery of Fulda to produce a written account of the relics’ transfer and the miraculous events that had surrounded it. This Translatio s. Alexandri offers a striking overview of the range of people who came to forge a spiritual connection with Wildeshausen’s new saint: a crippled boy from the Elbe-Weser region (ch. 7), a deaf-blind man from Deventer (ch. 8), and a Frisian woman named Femburg, who was said to live near Wildeshausen as an ‘exile’ (ch. 13). On the one hand, Femburg’s presence in a Westphalian community seems to confirm the permeability of the Frisio-Saxon border (Wood 2003, 280). On the other hand, the details of her case also underline the limits of this permeability: as a ‘stranger’ (peregrina) without a male guardian, she apparently was at risk of being ‘sold’; until the saint’s timely intervention, that is. Another cluster of relevant texts is centred around the extended family of the Frisian missionary Liudger (Schmid 1978; Angenendt 2005). The first Vita Liudgeri, written c. 839–49 by the saint’s nephew Altfrid, traces the family’s history back to Liudger’s paternal grandfather Wrssing, a Frisian magnate at the time of Radbod. According to Altfrid, the family held possessions around Loenen and Muiderberg, both in Frisia at the time, but also across the border in Westphalia (Vita Liudgeri I, chs. 4, 21, 27, 32; Wood 2000, 159–60). In the early 790s, Liudger came to use these Saxon lands to found monastic communities at Werden and Münster. When he was appointed bishop of Münster soon thereafter, his diocese covered not just Saxon territory but also five Frisian coastal pagi between the Ems and Lauwers (Vita Liudgeri I, chs. 22, 24; Prinz 1948). It has been pointed out that Altfrid, who was bishop of Münster and rector of Werden at the time, was far from disinterested when he styled his saintly uncle a ‘bishop of Saxons and Frisians’ (Vita Liudgeri II, ch. 20) with family roots among both peoples (Palmer 2011, 148–9; Rembold 2016, 369). With the Carolingian civil war of AD 840–3 and the subsequent threefold division of the empire, Münster would suddenly have found its diocese split between two Carolingian regna: Münster itself belonged to the Eastern 231

Z01 Hines Book B.indb 231

07/05/2021 13:30

Frisians of the Early Middle Ages Kingdom of Louis the German, but its Frisian pagi were situated in Lothar’s Middle Kingdom. Werden faced a similar predicament: it was part of Saxony but held possessions in Frisia, which by the 840s had become quite substantial through a rich stream of aristocratic donations (Bleiber 1965, 142–3, 156–8). In such a context, it was in Altfrid’s interests to show that his communities’ claims to power and property in Frisia could be traced back deep into the past. There are, to be sure, no explicit signs that Lothar ever encroached on Werden’s Frisian possessions. In fact, in November 855 the community received its most lavish donation up to that point, when the Frisian aristocrat Folcer took up the habit and endowed Werden with lands in the Betuwe, Hameland, Veluwe, Kennemerland and Westergo (Die Urbare, pp. 9–15). The timing of this acquisition is nevertheless noteworthy – it was arranged two months after Lothar’s death. Being part of the Carolingian empire brought Frisia and Saxony closer together, but also put up new political boundaries.

Conclusion The subject set for this paper was Frisians and Saxons. I have sought to explore this subject by pursuing two closely-related issues. One concerned the long-term development of the Frisians and Saxons as, or into, distinct peoples. The assumption was that, seeing as the Frisian and Saxon trajectories show many similarities as well as some conspicuous differences, it might add to our understanding of each of the two by considering them alongside each other. With regard to Late Antiquity, it has been postulated that the major issues associated with the Frisians and Saxons in this period – discontinuity for the former, flexible use of the ethnonym Saxones for the latter – were on several levels related. On a textual level, Roman observers started to see Franks and Saxons where they had previously seen Frisians and Chauci; a radical reconceptualization of the ethnic landscape beyond the Rhine that was mostly an extrapolation from experiences in Gaul or near the Rhine frontier and could thus well have served to obscure some of the remaining Frisians. On a material level, the fifth century saw North German groups migrating to Britain, the western coasts of Gaul and the northern Netherlands. Contemporary witnesses typically referred to these migrants as Saxones, and this name continued to be used by communities in Britain and Gaul. In the northern Netherlands, by contrast, the Frisian ethnonym ultimately prevailed. With regard to the Early-medieval period, a long-standing issue for both groups is their political organization, which is inevitably tied up also with the question whether there existed coherent Frisian and Saxon peoples at this time. Focusing on the slightly better-documented Saxon case, I have tried to emphasize the changing and flexible nature of such political organization. A good example is the rise of the Westphalian leader Widukind, whose authority among the Saxons in the late 770s and early 780s seems to have derived not from any fixed ‘ducal’ dignity, but from his careful manoeuvring during a time of great social upheaval. The Franks played the role of facilitators in this picture, first as an external threat, then as conquerors forcibly imposing the ideological and administrative ideal of a single gens Saxonum. An important point of discussion is to what extent the Saxon model can be mapped on to the Frisians, who may have been more politically centralized c. AD 700, with several Frisian leaders being considered ‘kings’ by contemporary observers. 232

Z01 Hines Book B.indb 232

07/05/2021 13:30

Frisians and Saxons With the development of more coherent Frisian and Saxons gentes in the eighth and ninth centuries, the second angle of this paper moved to the forefront: what could be said about the interaction between these two peoples? My approach here was inevitably selective. More might have been said on issues like trade relations (Steuer 2003), pre-Christian beliefs (Glatthaar 2004) and intermarriage (Wenskus 1976). As it stands, this paper has focused first on the close military cooperation between Saxons and Frisians during the 780s and 790s. Such cooperation was built on shared resistance against the Franks and their Church, but facilitated by geographical proximity, easily navigable waterways and cross-border landholding. Close interaction continued into the post-conquest period, as became evident from looking at two well-documented families active on the border between Frisia and Westphalia. These families testify to the various economic, social, religious and political ties between Frisians and Saxons in the ninth century, but also show that these ties did not always easily align. The dynastic divisions of the Carolingians could cut right through the ecclesiastical networks connecting Frisia and Saxony, as well as through the landed possessions sustaining these networks.

Discussion VERSLOOT  I have a linguistic point. There was a linguistic continuum between Saxonia and Frisia at least up to about AD 800. The isoglosses that can be reconstructed – denoting the geographical boundaries of linguistic developments – run through and across Saxonia rather than around it. Before 800, there was very little difference between what was spoken in Frisia and what was spoken in Saxony. Nils Århammar’s study of 1990 on ‘Friesisch und Sachsisch’ is useful here; he argues – rightly in my opinion – that the linguistic boundary between Frisia and Saxony arose mainly as a result of political circumstances brought about by Frankish expansion. Old Saxon came under strong Frankish influence, whereas Frisian remained an archaic and different language. FLIERMAN  That’s an important point. It fits the historical evidence of close FrisioSaxon cooperation in the eighth century and would explain also how someone like Widukind was able to gather support across the Ems or the IJssel. He would have been able to communicate relatively easily with the inhabitants of those regions. VERSLOOT  Yes, there was no sharp linguistic boundary separating those regions at that time; probably only gradual differences in accent. An important piece of evidence, in this regard, is the Straubing fragment of the Old Saxon Heliand, which shows such affinity to Old Frisian that it has (incorrectly) been interpreted as Proto-Old Frisian (Klein 1990). Even after 800, there were still many morphological details in which Old Frisian remained closer to Old Saxon than to Old English (its ‘default’ relative). IJSSENNAGGER-VAN DER PLUIJM  Thinking about the border region and imports, as comes up in a number of our papers, I wonder about imports across the border. Arjen, you mentioned the continuum in linguistic traits, and I would be quite interested in discussing whether the material culture should also be thought of in terms of a continuum rather than importation. As a more general discussion, how do we tell when something forms part of a continuum and when something is imported from a neighbouring area? Perhaps that would help in thinking about 233

Z01 Hines Book B.indb 233

07/05/2021 13:30

Frisians of the Early Middle Ages some of these issues. It is hard to distinguish, and often is about how the question or perception is framed. These issues about ‘Frisian’ versus ‘Saxon’ are about how these things are framed in the written sources, but again it may have been different on the ground and in the material culture. I would be really interested in the analogy with linguistics here and a wider discussion about how we can look at this. FLIERMAN  An important issue at stake here is the question of interdisciplinarity: how can we combine and integrate the distinct stories offered by historical, archaeological and linguistic research? What struck me, especially for the Roman Period, is how little overlap there appears to be between material and textual sources. I don’t mean that they are contradictory, but that they tend to be concerned with different aspects of life and society. The scope of material culture ranges from habitation patterns and economic circumstances to social and family relations (to name but a few things). Historical references to Frisians and Saxons, by contrast, are nearly always of a military nature: we hear about Frisians or even the Frisians rebelling against Roman rule or fighting in Roman armies. How do the Frisians depicted by Tacitus or identified in the epigraphical material measure up to the rather egalitarian first-century coastal settlements excavated by archaeologists? Was ethnicity – ‘being a Frisian’ – even a relevant form of identification for those labelled ‘Frisians’ by Roman authors? IJSSENNAGGER-VAN DER PLUIJM  A very interesting observation you made is that Frisians do occur in Saxon sources, if I am not mistaken. FLIERMAN  Yes, in ninth-century Saxon hagiographies, which rank among our earliest surviving Saxon texts, Frisians appear with some frequency and in a variety of contexts: as a neighbouring people, as donors to Saxon monastic houses, and even as members of Saxon communities. Such references are often quite casual; the presence of Frisians in Saxon society was not seen as something completely out of the ordinary. IJSSENNAGGER-VAN DER PLUIJM  You also made the observation that Chauci were eventually replaced by Franks and Saxons in the Roman tradition? FLIERMAN  Indeed, though I would stress that this change in Roman perceptions did not necessarily reflect the actual ethnic landscape at the time, which must have been far more complex. What the Roman sources were documenting was borderland activity: barbarian groups attacking the Roman frontiers or migrating into Roman territory. They projected what they saw happening around their frontiers into the hinterlands: the fourth and fifth centuries witnessed Saxon raids on the Roman frontiers, hence the trans-Rhine area and coastal areas of Germania were generally characterized as Saxon land. This was almost certainly a generalization; we know that small Frisian groups continued to inhabit Drenthe and Groningen. IJSSENNAGGER-VAN DER PLUIJM  It did make me wonder if we need to think more about how long the Roman framework could have remained influential. It has at least made me aware that it played a bigger role than I had realized. So thanks for that. WOOD  I agree there was ethnic fluidity around the Channel region in Late Antiquity. We are dealing with Roman outside observers who determine who is to be identified as Franks and Saxons. Wenskus (1961) is rather good on this. The point I wanted to raise pertains to boundaries. I agree that the boundary between 234

Z01 Hines Book B.indb 234

07/05/2021 13:30

Frisians and Saxons Frisians and Saxons in the Early Middle Ages was fluid; I found the same looking at the region from a missionary perspective. What struck me then, though, is that you can’t really define a boundary in this region until we have diocesan boundaries. The history of post-conquest Saxony quickly turns into a history of bishoprics. FLIERMAN  That’s an important point. Under the Carolingians, the regions between the Rhine and Elbe changed radically, turning into a religious landscape dominated by bishoprics, monasteries and the religious foundations of aristocratic families. I also have the feeling, however, that the evidence for seeing ninth-century Saxony as a religious landscape may be somewhat biased. The textual record for this time is dominated by charters and hagiography, both of which emphasize monastic and episcopal agency. The activities of the Saxon counts, for instance, receive comparatively little attention. WOOD  The hagiographical evidence does not simply support a solely episcopal view. The case of Liudger is interesting, in that it shows episcopal and royal boundaries that cut directly across the family’s interests. FLIERMAN  I agree, and this was a big issue for the family. In a way, what Vita Liudgeri shows us is that boundaries in ninth-century Saxony could be conceptualized variously: we see different parties – Carolingian rulers, bishops, aristocratic families – setting their own boundaries. I would read the text as an attempt by the family and the community of Werden to show that its landed claims in the various Carolingian kingdoms go way back into the past. Their eagerness to defend this claim could further suggest that there were parties out there – Lothar I, for one – who were disputing it. WOOD  Vita Lebuini antiqua is also important when it comes to Frisio-Saxon boundaries, and deeply problematic too. FLIERMAN  Yes, we find Saxons showing up as far west as Deventer in this text. In fact, the Translatio Sancti Alexandri is all about setting boundaries as well: right from the start, the Saxons are presented as a distinct gens, hesitant to intermarry with other peoples. Their territory is sharply demarcated by a list of neighbouring peoples: the Frisians in the west, the Franks in the south, the Slavs in the east and the Northmen in the north. This recurring insistence on boundaries in the hagiographical sources hints at a society in which boundaries were important, contested and in need of constant confirmation. WOOD  And it probably indicates that these boundaries are something new: a landscape definition of community being imposed as opposed to previous tribal or kindred communities. FLIERMAN  Yes, though I would stress that the Carolingians were not the sole agents behind this development; Saxon aristocratic families were actively involved as well. NIEUWHOF  You mentioned a ‘Frankish gaze’ on the North and the map of gentes they created, and how that contrasts with the Late-antique ‘Roman gaze’. But there is actually a very interesting parallel with the earlier Roman Period, when the Romans did quite the same. They also created a ‘map of gentes’ in the north, applying ethnic labels to all the different groups they encountered there. I think that is probably something that colonial powers habitually do. FLIERMAN  That’s a good point. My impression of the Roman ‘ethnic gaze’ of the first and second centuries AD is that it was more expansionist: emperors going on naval expeditions beyond the frontier, Ptolemy trying to map the known world. In 235

Z01 Hines Book B.indb 235

07/05/2021 13:30

Frisians of the Early Middle Ages the Later Roman Period, this outward gaze retreats inwards: what happens at the frontier becomes crucially important. The Franks subsequently return to a more expansionist view of external gentes, especially under the Carolingians. NIEUWHOF  The use of ethnic labels does not confirm that people thought in similar terms about themselves. These labels may not have been that important to them. FLIERMAN  I agree wholeheartedly. There is no guarantee that those labelled Saxons or Frisians by Roman observers thought of themselves in those terms. For the Saxons, the problem is that before the ninth century, we have no Saxon texts: nothing. With the Frisian name, at least, we have a number of inscriptions from the first and second centuries AD: soldiers in Britain, Gaul and Italy identifying themselves as natione Frisius. Such identification occurred in a specific military context, of course, but it does suggest there were people out there who considered the label Frisian a meaningful form of self-identification. NIEUWHOF  Nico Roymans has written about ethnic identity with regard to the Batavians (2004). He thinks that these ethnic labels play a role especially in contacts with the Romans, and that the Batavians and Germans and other groups came to define themselves as such in that context. FLIERMAN  That mechanism applies to the ninth-century Saxons as well. When the Saxons start to write their own histories from the ninth century onwards, they use Frankish and sometimes even Roman sources to do so. On their own terms of course. MAJCHCZACK  I would like to comment on the question of whether the people saw themselves as different ethnic groups. I think when we look at the Lower Saxon areas, there are huge similarities amongst marsh and geest settlements and in the material culture. But it is primarily the coastal landscape that determines how people live and how their settlements are organized. Possibly the label Frisian could be adopted by new settlers in marsh areas actually because it didn’t matter that much. I think the culture was surely mixed, and defined by the coastal landscape. FLIERMAN  It is interesting to note that migrants from those same regions also ended up in England and northern France in the fifth century. In this case, however, some did come to think of themselves – or continued to think of themselves – as a distinct ethnic community, e.g. the Saxons of Bayeux mentioned by Gregory of Tours. MAJCHCZACK  I suggest that especially when people live in a diaspora amongst foreigners or in a foreign context, they think of themselves as belonging to a different group. And they might demonstrate their origins by holding on to certain cultural traditions. A good example is the mixed-rite cemeteries at the emporia on the North [Sea] and Baltic Sea coasts, for example at Ribe and Groß Strömkendorf/ Reric. FLIERMAN  I find that very compelling. It could also help explain why and how the name ‘Frisian’ eventually became important again in the seventh century. Jos Bazelmans has argued that the name was imposed on the inhabitants of the northern Netherlands by the Franks, but this seems to me only half an explanation. Why did people subsequently accept this identification? What made the label stick? One possible answer, turning back to what you just said, would be increased 236

Z01 Hines Book B.indb 236

07/05/2021 13:30

Frisians and Saxons pressure by, and competition with, outside groups like the Franks and Saxons, which pushed elites of the region to emphasize their distinctiveness: i.e. we are a distinct group called the Frisians. IJSSENNAGGER-VAN DER PLUIJM  The question is also when the label became important again for the Frisians to use, as it does not happen at that very moment, at least we don’t actually know that it did. It seems later, so that again we have multiple, different, layers in time. FLIERMAN  Yes: coming up with an exact chronology for the disappearance and reappearance of the Frisian name is tricky. For example: there is a late sixthcentury reference to Frisians in the poetry of Venantius Fortunatus, who praises a Merovingian ruler for keeping even ‘the furthest Frisians’ subdued. Is the use of the ethnonym a literary relic from the Roman Period, as Bazelmans would argue? Or does it mean that, for the Merovingian rulers, the Frisian name was already becoming important again? I lean towards the latter. IJSSENNAGGER-VAN DER PLUIJM  I absolutely agree, but that doesn’t mean the Frisians themselves used that term, that is a different angle. NIJDAM  I detect a great deal of influence from the powerful ‘ethnogenesis’ paradigm. That certainly brings some insights, but we also lose things when we see everything as fluid, and focus so strongly on the written sources and on elite perspectives. So my question is: how do you think these people viewed themselves at lower levels? What infrastructures would have existed that helped them to define ethnicity? I would immediately think of thing assemblies and legal boundaries. FLIERMAN  Yes, I was wondering about this myself. How often would, say, a sixthcentury woman in Frisia have encountered the law? How and when would it have regulated her life? NIJDAM  I think they would have had to go a thing assembly regularly. In the High Middle Ages, this was three times a year. VERSLOOT  We could use the Icelandic situation as a close analogy. Of course it is later and from a different region, but it gives some sort if impression of how these assemblies worked. It was an important event. The annual calendar revolved around it. It was not just a political platform, but involved trade and other social dealings as well. FLIERMAN  That mechanism is feasible. I can easily imagine an annual assembly acting as a vehicle for the creation for a shared sense of identity and social cohesion. The question for me is mainly one of methodology: can we use later material as evidence for earlier practices, especially when the society in question is known fundamentally to have changed in the meantime? Take for instance the tenth-century Vita Lebuini antiqua, which offers an elaborate description of a pre-conquest annual Saxon assembly. Combined with earlier Frankish and AngloSaxon material, I would be willing to argue from this that assemblies were part of pre-conquest pagan Saxon society. But I am less certain about their form and scale. The suggestion of Vita Lebuini that this was an assembly that involved all the Saxons makes sense in a tenth-century context with an Ottonian king ruling over a united Saxony. But we know from earlier sources that pre-conquest Saxony was far less centralized, consisting of various more or less independent Saxon groups. NIJDAM  But I think the accent has been too much on fluidity. What does the anthropological literature say about self-perception? 237

Z01 Hines Book B.indb 237

07/05/2021 13:30

Frisians of the Early Middle Ages HINES  One of the problems, I would argue, is that anthropological literature is no less influenced by political currents than any other academic subject or aspect of contemporary life; therefore answers are found that tend to conform with what the intellectual world prefers to think at any time. You cannot just go to ethnographic literature and a stock of perfectly observed answers; what is clear is that there is no one answer. I can give an example from the eastern Polish border area where villages are clearly categorized and recognized as ‘Polish’, ‘Belorussian’ or ‘Ukrainian’, on the basis of various cultural markers: for instance, the denomination of the church (Roman Catholic or Orthodox), but also very practical variables such as the orientation of buildings: gable-on or side-on to the street. You can identify what the expectations in terms of visible marking are straightaway; but you can also appreciate how much of a threat non-conformity, or an idiosyncratic choice to be different, could be to the community’s sense of itself: its coherence and its identity. People would be likely to feel very threatened even if they might struggle to articulate why they were so upset by such divergency from the norm. NIJDAM  Hence the importance of us versus them. This fits with research into human universals, which has been off the agenda after the 1930s but is now returning because of insights within the cognitive sciences: historians are now increasingly amenable to the idea that human beings do something universally in a certain way, and then culture comes in. Think of revenge, violence, etc. Such human behaviour is informed by several layers. Returning to John’s point about conformity and identity, Patrick Wormald (2003) pointed out the importance of injury tariffs in the barbarian law codes. Every code offered different tariffs, suggesting it was an ethnic marker to have your own list of tariffs. FLIERMAN  I completely agree that ethnic identity was very meaningful in Earlymedieval societies. But I would also argue that its meaning and salience derived from context. When paying injury tariffs, yes; during assembly politics, certainly; in military contexts, absolutely. But I don’t think we should approach ethnic identity as an essential quality or determinant: someone was a Saxon or Frisian, ergo they behaved in such and such fashion. This is the wrong way around. The question, in my opinion, should not be was ethnic identity important, but when was it important. NIJDAM  So you suggest that from day to day it is unmarked, until you come into the relevant context? FLIERMAN  There are spaces where identity is highly marked and contexts where it is less relevant. I would agree also that there are very refined and often unconscious mechanisms at work when it comes to (ethnic) identification: think of our uncanny ability to identify compatriots on a holiday. But that too constitutes a specific context. KNOL  I was thinking about the Carolingian empire and its impact on the northern Netherlands. It is striking that the cemeteries of the eighth century show a high number of weapon graves and horse-burials, which are almost absent in the sixth and seventh centuries. Moreover, the eighth-century weapon graves are both inhumation and cremation burials. I believe that cremation must mark opposition to the Frankish world, but inhumation could be both Frankish and anti-Frankish. My second remark is about the border between the Saxon and the Frisian worlds. On one level, it was very simple: Frisians have the sea and the outer ring 238

Z01 Hines Book B.indb 238

07/05/2021 13:30

Frisians and Saxons of tidal flats and salt marshes; then come the uninhabitable peat bogs; then the Saxons on the Pleistocene hinterland. The problem is that, here and there, the Pleistocene hinterland was very close to the sea. More importantly, there were myriad waterways through the peat bogs. Apart from the big rivers, there were many smaller waterways that came into the sea, and these were useful for making connections with the hinterland. The people living in the coastal area needed those connections to get wood, an essential hinterland resource, and also stone. This belies the notion of clear boundaries. FLIERMAN  On your first point, about the eighth-century weapon burials, I agree that these must represent a period of social upheaval. I am wondering, though, who was responsible for these graves? I can imagine two types of groups. One would consist of the people who, as you mentioned, resisted Frankish influence. This fits with the eighth-century Carolingian legislation against deviant burial rites. A second group who could potentially benefit from lavish burial rites would be members of a new post-conquest Frisian elite who were falling in line with the Franks, but who needed to establish and display their authority on the ground. KNOL  We may raise that question but at the time people must have been well aware of who was whom; both parties could use the same form of burial expression. That said, I wouldn’t expect a ‘new’ leader – or an old leader who got in line with Charlemagne’s programme – to have practised cremation. A common use of weapon burial is plausible, however. FLIERMAN  I think a late eighth-century Saxon had to make a choice. The Carolingian laws advocated burial in an ecclesiastical context, and we know of some eighth-century Saxon leaders (e.g. Hessi) who were buried in a monastery. But weapon burial would also have been possible; there was no hard regulation against it. KNOL  There was no ecclesiastical legislation against grave goods; the Church doesn’t like it but it is permitted; cremation, however, is strictly forbidden. FLIERMAN  Agreed, though Charlemagne’s Saxon Capitulary (c. 782–95) is among the first known pieces of Christian legislation explicitly to ban cremation. I think the choice of burial form was more about social strategy than about legal pressure. KNOL  There are examples of Christian grave goods, but they quickly became less and less common. Weaponry, however, expresses power. Archaeology associates weapon burials with founder graves particularly; this seems to have a good parallel in the House of Orange in Delft. The founder – William of Orange – has an exceptional monument; but not his successors. Further markers were not needed. IJSSENNAGGER-VAN DER PLUIJM  To go back to what we were talking about earlier around assemblies, thing and law as vehicles for identity, is that territorial or personal law, and does it matter? These would presumably be different as vehicles for identity? WOOD  They are the same. Personal law is determined by where you were born, not who you were born to. So personal law is always geographical, but you carry it with you from your birthplace. IJSSENNAGGER-VAN DER PLUIJM  Yes it is geographical, but then is it the law for Frisia or for Frisians? WOOD  It wouldn’t normally make a difference, because most of the people who are subject to these laws aren’t moving very far. 239

Z01 Hines Book B.indb 239

07/05/2021 13:30

Frisians of the Early Middle Ages IJSSENNAGGER-VAN DER PLUIJM  With so much mobility, I think for the Frisians it would actually make a difference. WOOD  Mobility is within the Frisian area. The reason why a person is bound by the law of his birthplace rather than that of his parents is that he knows what he’s carrying with him. IJSSENNAGGER-VAN DER PLUIJM  How would this then work between Frisia and Saxony? FLIERMAN  There is a beautiful example from the Werden cartulary that ties in with the question of personal versus territorial law and also shows the sensitivity of contemporaries to acting ‘within the law’. In 855 a Frisian aristocrat named Folcher donated a substantial amount of property to the monastery. He is said to have held and donated this land according to three laws: the Lex Ribuaria, the Lex Salica and the Ewa Frisonum (the reference to this last law may be a later interpolation). So what does this mean? Did different pieces of the land he donated fall under different laws? Or were multiple legal procedures involved in donating one and the same piece of land? NIJDAM  I didn’t know this. I wonder what the procedures were. FLIERMAN  As you said, there is nothing on landholding in Frisian law. NIJDAM  Indeed, it’s absent. We would have to take a look at the other Leges Barbarorum. MOL  Returning to the question of landscape and the differences between Frisia and Saxony: I concur with Egge’s view of a Frisian landscape that’s a thin strip between the sea and the peat, with many streams running through. It’s also a very densely populated area. The Saxon landscape is very different, in this regard: far more extensive, with vast wastes. The contrast applies also to the political landscape. In the High Middle Ages, Saxony is a land of aristocrats, magnates, powerful bishops. I cannot see how that system was constructed anew around the year 1000; it must have had very old roots. Along the coast, meanwhile, the landscape in terms of land-ownership is much more fragmented, with Folcher being a bit of an outlier in my opinion. Around 1000 Frisia is split into just four dioceses, yet Saxony has many ecclesiastical principalities, with large landed properties. There are, in short, significant differences between the two entities, especially when we are looking for or thinking in terms of continuity. FLIERMAN  What struck me, agreeing with your point about continuity, is that Charlemagne appears to have been very aware of pre-existing Saxon structures and powerful pre-conquest families. He actively tried to exploit those when giving administrative and ecclesiastical shape to post-conquest Saxony. While this approach was largely successful in Saxony, it did not really come off the ground in Frisia. This surprised me at first, because I took pre-conquest Frisia in the eighth century to have been more centralized than pre-conquest Saxony. MOL  Well, that’s the big question. We only really know about Aldgisl and Radbod as ‘centralized’ rulers. I would locate them in North-Holland. Yet the coastal terp landscape was very different, possibly with a very different society. The Carolingian conquest of Saxony focused on replacing or winning over the big land-owning families. I don’t see that happening in the coastal area of Frisia and Groningen. You were dealing with a smaller landholders there, who you had to win over. FLIERMAN  Landscape is the determinative factor here. 240

Z01 Hines Book B.indb 240

07/05/2021 13:30

Frisians and Saxons MOL  Exactly. Wherever the ‘new’ inhabitants of Frisia had come from, they had to adapt to the landscape. Pole-jumping over ditches, for instance, was an ancient cultural practice that the newcomers took over to negotiate their surroundings. KNOL The terp area is very densely populated, much more so than the Pleistocene hinterland. MOL  You can’t compare it with other areas of the Netherlands. FLIERMAN  I find it feasible to look at landscape as a crucial factor. It would imply that Radbod’s kingly authority had to be fundamentally different from Frankish kingship. Would this explain why Frisia was never fully incorporated into the Carolingian framework? The Carolingians did try to impose their usual system of counts, for instance, but these were introduced from outside. NIJDAM  Yes, that’s what goes wrong. FLIERMAN  But someone like Folcher did have land. And there must have been others like him. So why didn’t the Carolingian system stick? Landscape might be the answer. DE LANGEN  The distribution of land was also different; the estates were scattered. This goes for Folcher’s donation as well. IJSSENNAGGER-VAN DER PLUIJM  We need to think about water and waterways as well as land. HINES  To use the Icelandic parallel again to see if it can shed light: there the centralizing thing-system serves to counteract or even to overcome a completely dispersed settlement pattern of isolated farmsteads. That suggests a further but different question of analogy: in this case the remarkable contrast in the trajectories followed in the ninth century between Saxony and Frisia. The failure of an Old Frisian Christian literary tradition to develop in the way it does in every other Germanic language is remarkable. Is that just to be explained in terms of anarchy and breakdown in Frisia, with sheer pragmatic survival strategies taking precedence over everything else, so that there was no need or place to bother with anything sophisticated in the way of religion? Or does something more positive come in its place, which analogies with what happened in adjacent areas cannot identify for us? It is probably much easier to formulate a question like that than to answer it. FLIERMAN  Could it partly be an issue of transmission? All of our Saxon material comes from just a handful of monasteries and bishoprics. It didn’t necessarily originate there, but these centres allowed these traditions to be passed on. With Early-medieval Frisia, are we dealing with an actual absence of Christian literary traditions, or with the absence of an ecclesiastical system that documented and transmitted those traditions? HINES   But the Old Saxon does survive and leave footprints: not just surviving fragments; one of the Old Saxon texts (Genesis) was first identified hypothetically as the source of an Old English/West Saxon ‘translation’, and subsequently a fragment of the original was indeed discovered in the Vatican Library (Doane 1991). We also have a couple of ghostly shadows of ancient Old Frisian literature in Old English: for instance in the Maxims and above all in the Finnsburh fragment. It is a puzzle. MOL  Going back to the differences between Saxony and Frisia: Saxony witnessed the foundation of many large and wealthy monasteries already in the late Carolingian 241

Z01 Hines Book B.indb 241

07/05/2021 13:30

Frisians of the Early Middle Ages Period. It continued in the Ottonian Period. This could happen because there were families who could afford large donations. In the coastal area, the first monasteries were founded in the twelfth century; that shows how the social structures were different. FLIERMAN  Are you saying Frisia was generally poorer than Saxony, or more egalitarian? MOL  Not necessarily poor, but the aristocracy was small scale. Local lords had small estates. And even if those estates were considerably more fertile than those in the hinterland, they had to pass them on to the next generation. Of course, most of what we know of Frisian landholding comes from the High Middle Ages or later. The question then is if this system of small-scale landholding is old or new. Using the evidence of twelfth-century foundations, I would suggest that it goes a long way back. NIJDAM  Frisian law confirms this picture. A freeholder is someone with one or two farms, a nobleman has three or four farms. This matches neatly with wergild ratios. It’s very consistent.

References Primary sources Altfrid, Vita Liudgeri. Ed. W. Diekamp 1881, Die Vitae Sancti Liudgeri, Geschichtsquellen des Bistums Münster 4 (Münster). Ammianus Marcellinus, Res Gestae. Ed. W. Seyfarth 1978, 2 vols. (Leipzig). Annales regni Francorum. Ed. F. Kurze 1895, MGH Scriptores, SS rer. Germ. 6 (Hannover). Annales sancti Amandi. Ed. G. H. Pertz 1826, MGH Scriptores, SS 1 (Hannover), 6–14. Bede, Historia Ecclesiastica Gentis Anglorum. Ed. C. Plummer 1896, Bedae Opera Historia. vol. 1 (Oxford). Capitulare Saxonicum. Ed. C. von Schwerin 1918, MGH Leges, Fontes iuris 4 (Hannover). Chronicle of 452. Ed. R. W. Burgess 2001, ‘The Gallic Chronicle of 452: a new critical edition with a brief introduction’. In Society and Culture in Late Antique Gaul, eds. R. Mathisen and D. Shanzer (Aldershot), 52–84. Claudian, Carmina. Ed. T. Birt 1892, MGH Scriptores, Auct. ant. 10 (Berlin). Constantius of Lyon, Vita Germani. Eds. B. Krusch and W. Levison 1920, MGH Scriptores, SS rer. Merov. 7 (Hannover and Leipzig). Diplomata Belgica ante annum millesimum centesimum scripta, Vol. 1 Teksten. Eds. M. Gysseling and A. C. F. Koch 1950 (Brussel). Die Urbare der Abtei Werden an der Ruhr: Die Urbare vom 9.–13. Jahrhundert. Ed. Rudolf Kötzschke 1906 (Bonn). Eutropius, Breviarium ab Urbe Condita. Ed. C. Santini 1979 (Leipzig). Gildas, De Excidio et Conquestu Britanniae. Ed. T. Mommsen 1898, MGH Scriptores, Auct. ant. 13 (Berlin). Gregory of Tours, Historiae. Eds. B. Krusch and W. Levison 1951, MGH Scriptores, SS rer. Merov. 1.1 (Hannover). Julian, Orations. Ed. T. E. Page and trans. W. C. Wright 1918, 2 vols. (New York). Lex Frisionum. Eds. K. A. Eckhardt and A. Eckhardt 1982, MGH Leges, Fontes iuris 12 (Hannover).

242

Z01 Hines Book B.indb 242

07/05/2021 13:30

Frisians and Saxons Lex Saxonum. Ed. C. von Schwerin 1918, MGH Leges, Fontes iuris 4 (Hannover). Lorsch Annals (Annales Laureshamenses). Ed. E. Katz 1889, Annalium Laureshamensium editio emendata secundum codicem St. Paulensem, Seperatabdruck vom Jahresberichtdes öffentlichen Stifts-Untergymnasiums der Benedictiner zu St. Paul (St Paul). Orosius, Historiarum aduersum paganos libri. Ed. and trans. M.-P. Arnaud-Lindet 1990–91, 3 vols. (Paris). Ptolemy, Geographia. Eds. and trans. A. Stückelberger and G. Graßhoff 2006, Ptolemaios. Handbuch der Geographie, 2 vols. (Basel). Revised Annales regni Francorum (Annales qui dicuntur Einhardi). Ed. F. Kurze 1895, MGH Scriptores, SS rer. Germ. 6. (Hannover). Sidonius Apollinaris, Epistulae. Ed. C. Lütjohann 1887 (Berlin). Tacitus, Germania. Ed. J. G. C. Anderson 1961 (Oxford). — Historiae. Ed. H. Heubner 1978 (Leipzig). Translatio s. Alexandri. Ed. B. Krusch 1933, ‘Die Übertragung des H. Alexander von Rom nach Wildeshausen durch den Enkel Widukinds. Das älteste niedersächsische Geschichtsdenkmal’. In Nachrichten von der Akademie der Wissenschaften in Göttingen. Philol.-hist. Klasse 4 (Göttingen), 405–36. Vita Liudgeri Secunda. Ed. W. Diekamp 1881, Die Vitae Sancti Liudgeri, Geschichtsquellen des Bistums Münster 4 (Münster). Vita Liutbirgae. Ed. O. Menzel 1937, Das Leben der Liutbirg. Eine Quelle zur Geschichte der Sachsen in karolingischer Zeit. MGH Weiteren Reihe, Dt. MA 3. 1937 (Leipzig). Zosimos, Historia Nova. Ed. F. Paschoud 1979 (Paris). Secondary sources Althoff, G. 1983, ‘Der Sachsenherzog Widukind als Mönch auf der Reichenau. Ein Beitrag zur Kritik des Widukind-Mythos’, Frühmittelalterliche Studien 17, 251–79. Angenendt, A. 2005, Liudger: Missionar - Abt - Bischof im frühen Mittelalter (Münster). Bazelmans, J. 2001, ‘Die spätrömerzeitliche Besiedlungslücke im niederländischen Küstengebiet und das Fortbestehen des Friesennamens’, Emder Jahrbuch für historische Landeskunde Ostfrieslands 81, 7–61. — 2009, ‘The early-medieval use of ethnic names from Classical Antiquity. The case of the Frisians’. In Ethnic Constructs in Antiquity: The Role of Power and Tradition, eds. T. Derks and N. Roymans (Amsterdam), 321–38. Bazelmans, J., M. Dijkstra and J. de Koning 2004, ‘Holland during the first millenium’. In Bruc Ealles Well. Archaeological Essays Concerning the Peoples of North-West Europe in the First Millennium AD, ed. M. Lodewijckx (Leuven), 3–36. Becher, M. 1996, Rex, Dux und Gens. Untersuchungen zur Entstehung des sächsischen Herzogtums im 9. und 10. Jahrhundert (Husum). — 1999, ‘“Non enim habent regem idem Antiqui Saxones ...” Verfassung und Ethnogenese in Sachsen während des 8. Jahrhunderts’. In Sachsen und Franken in Westfalen, ed. Hans-Jürgen Häßler, Studien zur Sachsenforschung 12 (Oldenburg), 1–31. — 2013, ‘Gewaltsmission. Karl der Grosse und die Sachsen’. In Credo: Christianisierung Europas im Mittelalter, vol. 1, eds. C. Stiegemann, M. Kroker and W. Walter (Petersberg), 321–9. Bleiber, W. 1965, ‘Fränkisch-karolingische Klöster als Grundherren in Friesland’, Jahrbuch für Wirtschaftsgeschichte 3, 127–75. Bremmer, R. 198, ‘Frisians in Anglo-Saxon England: A historical and toponymical

243

Z01 Hines Book B.indb 243

07/05/2021 13:30

Frisians of the Early Middle Ages investigation’, Fryske Nammen 3, 45–95. Böhme, H. W. 1999, ‘Sächsische Söldner im römischen Heer: das Land zwischen Ems und Niederelbe während des 4. und 5. Jahrhunderts’. In Über allen Fronten. Nordwestdeutschland zwischen Augustus und Karl dem Großen, eds. F. Both and H. Aouni (Oldenburg), 49–73. — 2003, ‘Das nördliche Niedersachsen zwischen Spätantike und frühem Mittelalter. Zur Ethnogenese der Sachsen aus archäologischer Sicht’, Probleme der Küstenforschung im südlichen Nordseegebiet 28, 251–70. Büttner, H. 1965, ‘Mission und Kirchenorganisation des Frankenreiches bis zum Tode Karls des Großen’. In Karl der Grosse. Lebenswerk und Nachleben, vol 1, eds. H. Beumann and W. Braunfels (Düsseldorf), 454–87. Capelle, T. 1998, Die Sachsen des frühen Mittelalters (Darmstadt). De Langen, G. and J. A. Mol 2017, ‘Church foundation and parish formation in Frisia in the tenth and eleventh centuries: a planned development?’, The Medieval Low Countries 4, 1–55. Dijkstra, M. 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 (Leiden). Dijkstra, M. and J. de Koning 2017, ‘All quiet on the western front. The western Netherlands and the “North Sea Culture” in the Migration Period’. In Frisians and their North Sea Neighbours. From the Fifth Century to the Viking Age, eds. J. Hines and N. IJssennagger (Woodbridge), 53–75. Doane, A. N. 1991, The Saxon Genesis: An Edition of the West Saxon Genesis B and the Old Saxon Vatican Genesis (Madison). Effros, B. 1997, ‘De partibus Saxoniae and the regulation of mortuary custom: a Carolingian campaign of christianization or the suppression of Saxon identity?’, Revue Belge de Philologie et d’Histoire 75, 267–86. Ehlers, C. 2007, Die Integration Sachsens in das fränkische Reich: 751–1024 (Göttingen). Faulkner, T. 2016, Law and Authority in the Early Middle Ages. The Frankish leges in the Carolingian Period (Cambridge). Flierman, R. 2015, ‘Gens perfida or populus Christianus? Saxon (in)fidelity in Frankish historical writing’. In The Resources of the Past in Early Medieval Europe, eds. C. Gantner, R. McKitterick and S. Meeder (Cambridge), 188–205. — 2017, Saxon Identities, AD 150–900 (London). Galestin, M. C. 2007/2008, ‘Frisii and Frisiavones’, Palaeohistoria 49/50, 687–708. Glatthaar, M. 2004, Bonifatius und das Sakrileg. Zur politischen Dimension eines Rechtsbegriffs (Frankfurt am Main). Halbertsma, H. 2000, Frieslands oudheid: het rijk van de Friese koningen, opkomst en ondergang (Utrecht). Heidinga, A. 1999, ‘The Frisian achievement in the first millenium AD’. In The Making of Kingdoms, eds. T. Dickinson and D. Griffiths (Oxford), 11–16. Henstra, D. 2012, Friese graafschappen tussen Zwin en Wezer: een overzicht van de grafelijkheid in middeleeuws Frisia (c. 700–1200) (Assen). Hiddink, H. A. 1999, ‘Germaanse Samenlevingen Tussen Rijn en Weser: 1e eEuw voor –4e Eeuw na Chr.’ (Ph.D. thesis, University of Amsterdam). Hines, J. 2001, ‘The role of the Frisians during the settlement of the British Isles’. In Handbuch des Friesischen / Handbook of Frisian Studies, eds. H. H. Munske, N. Århammer, V. F. Faltings, J. F. Hoekstra, O. Vries, A. G. H. Walker and O. Witts (Tübingen), 503–11.

244

Z01 Hines Book B.indb 244

07/05/2021 13:30

Frisians and Saxons Honselmann, K. 1984, ‘Die Bistumsgründungen in Sachsen unter Karl dem Großen, mit einem Ausblick auf spätere Bistumsgründungen und einem Exkurs zur Übernahme der christlichen Zeitrechnung im frühmittelalterlichen Sachsen’, Archiv für Diplomatik 30, 1–50. IJssennager, N. L. 2013, ‘Between Frankish and Viking: Frisia and Frisians in the Viking Age’, Viking and Medieval Scandinavia 9, 69–98. Kahrstedt, U. 1934, ‘Die politische Geschichte Niedersachsens in der Römerzeit’, Nachrichten aus Niedersachsens Urgeschichte 8, 1–20. Kleemann, J. 2002, Sachsen und Friesen im 8. und 9. Jahrhundert: eine Archäologisch-Historische Analyse der Grabfunde (Oldenburg). Klein, T. 1990, ‘Die Straubinger Heliand-Fragemente: Altfriesisch oder Altsächsisch?’ In Aspects of Old Frisian Philology, eds. R. H. Bremmer Jr, O. Vries and G. van der Meer, Amsterdamer Beiträge zur älteren Germanistik 31–32 (Groningen–Amsterdam– Atlanta), 197–225. Knol, E. 1993, ‘De Noordnederlandse Kustlanden in de Vroege Middeleeuwen’ (Ph.D. thesis, Vrije Universiteit Amsterdam). — 2009, ‘Anglo-Saxon migration reflected in cemeteries in the northern Netherlands’. In Foreigners in Early Medieval Europe. Thirteen International Studies on Early Medieval Mobility, ed. D. Quast (Mainz), 113–30. Knol, E. and N. IJssennagger 2017, ‘Palaeogeography and people’. In Frisians and their North Sea Neighbours. From the Fifth Century to the Viking Age, eds. J. Hines and N. IJssennagger (Woodbridge), 5–24. Kötzschke, R. 1901, Studien zur Verwaltungsgeschichte der Großgrundherrschaft Werden an der Ruhr (Leipzig). Krüger, S. 1950, Studien zur Sächsischen Grafschaftsverfassung im 9. Jahrhundert (Göttingen). Krusch, B. 1933, ‘Die Übertragung des H. Alexander von Rom nach Wildeshausen durch den Enkel Widukinds. Das älteste niedersächsische Geschichtsdenkmal’. In Nachrichten von der Akademie der Wissenschaften in Göttingen. Philol.-hist. Klasse 4 (Göttingen), 405–36. Lebecq, S. 1983, Marchands et navigateurs Frisons du Haut Moyen Âge. 2 vols. (Lille). Löwe, Heinz 1986, ‘Lateinisch-christliche Kultur im karolingischen Sachsen’. In Angli e Sassoni al di qua e al di là del mare (Spoleto), 491–536. Ludowici, B. ed. 2019, Saxones. Eine neue Geschichte der alten Sachsen (Darmstadt). Meier, D. 2003, ‘The North Sea coastal area: settlement history from Roman to early medieval times’. In The Continental Saxons from the Migration Period to the Tenth Century: An Ethnographic Perspective, eds. D. Green and F. Siegmund, Studies in Historical Archaeoethnology (Woodbridge), 37–76. Nicolay, J. A. W. 2005, ‘Nieuwe bewoners van het terpengebied en hun rol bij de opkomst van Fries koningschap. De betekenis van gouden bracteaten en bracteaatachtige hangers uit Friesland (vijfde-zevende eeuw na Chr.)’, De Vrije Fries 85, 37–103. — 2014, The Splendour of Power: Early Medieval Kingship and the Use of Gold and Silver in the Southern North Sea Area (5th to 7th century AD), Groningen Archaeological Studies 28 (Groningen). — 2015, ‘Kingship in early-medieval “Frisia”: historical versus archaeological sources’. In Golden Middle Ages in Europe. New Research into Early-Medieval Communities and Identities, eds. A. Willemsen and H. Kik (Turnhout), 25–32. Nieuwhof, A. 2011, ‘Anglo-Saxon immigration or continuity? Ezinge and the coastal area

245

Z01 Hines Book B.indb 245

07/05/2021 13:30

Frisians of the Early Middle Ages of the northern Netherlands in the Migration Period’, Journal of Archaeology in the Low Countries 5 (1), 53–84. — 2013, ‘Discontinuity in the northern-Netherlands coastal area at the end of the Roman Period’. In Transformations in North-Western Europe (AD 300–1000), eds. T. A. S. M. Panhuysen and B. Ludowici, Neue Studien zur Sachsenforschung 3 (Hannover–Stuttgart), 55–66. — 2015, Eight Human Skulls in a Dung Heap and More. Ritual Practice in the Terp Region of the Northern Netherlands. 600 BC–AD 300, Groningen Archaeological Studies 29 (Groningen). Palmer, J. 2011, ‘Beyond Frankish authority? Frisia and Saxony between the Anglo-Saxons and Carolingians’. In Anglo-Saxon England and the Continent, eds. H. Sauer and J. Story (Tempe), 139–62. Panten, A. 2001, ‘Geschichte der Friesen im Mittelalter: Nordfriesland’. In Handbuch des Friesischen / Handbook of Frisian Studies, eds. H. H. Munske, N. Århammer, V. F. Faltings, J. F. Hoekstra, O. Vries, A. G. H. Walker and O. Witts (Tübingen), 551–5. Prinz, J. 1948, ‘Die Parochia des heiligen Liudger. Die räumlichen Grundlagen des Bistums Münster’. In Liudger und sein Erbe, vol. 1, eds. M. Bierbaum, H. Börsting, K. G. Fellerer, B. Peus, J. Prinz, H. Rademacher, E. Reinhard, G. Schreiber, A. Schröer, R. Schulze and W. Stüwer (Münster), 1–38. Rembold, I. 2015, ‘Rewriting the founder: Werden on the Ruhr and the uses of hagiography’, Journal of Medieval History 41, 363–87. — 2018, Conquest and Christianization: Saxony and the Carolingian world, 772–888 (Cambridge). Reuter, T. 2005, ‘Charlemagne and the world beyond the Rhine’. In Charlemagne. Empire and Society, ed. J. Story (Manchester), 183–94. Röckelein, H. 2002, Reliquientranslationen nach Sachsen im 9. Jahrhundert. Über Kommunikation, Mobilität und Öffentlichkeit im Frühmittelalter (Stuttgart). Roymans, N. 2004, Ethnic Identity and Imperial Power. The Batavians in the Early Roman Empire, Amsterdam Archaeological Studies 10 (Amsterdam). Schmid, K. 1964, ‘Die Nachfahren Widukinds’, Deutsches Archiv 20, 1–47. — 1978, ‘Die “Liudgeriden”. Erscheinung und Problematik einer Adelsfamilie’. In Geschichtsschreibung und geistiges Leben im Mittelalter. Festschrift für Heinz Löwe zum 65. Geburtstag, eds. K. Hauck and H. Mordek (Cologne), 71–101. Schrijver, P. 2017, ‘Frisian between the Roman and the Early Medieval Periods: Language contact, Celts and Romans’. In Frisians and their North Sea Neighbours. From the Fifth Century to the Viking Age, eds. J. Hines and N. IJssennagger (Woodbridge), 43–52. Seebold, E. 2001, ‘Die Friesen im Zeugnis antiker und spätantiker Autoren’. In Handbuch des Friesischen / Handbook of Frisian Studies, eds. H. H. Munske, N. Århammer, V. F. Faltings, J. F. Hoekstra, O. Vries, A. G. H. Walker and O. Witts (Tübingen), 479–87. — 2012, ‘Ptolemäus und die Sachsen. Mit Überlegungen zum Status von Namen im Materialteil der Geographie’, Beiträge zur Namenforschung 47, 191–206. Shuler, E. 2010, ‘The Saxons within Carolingian Christendom: Post-Conquest identity in the translationes of Vitus, Pusinna and Liborius’, Journal of Medieval History 36, 39–54. Springer, M. 2003, ‘Location in space and time’. In The Continental Saxons from the Migration Period to the Tenth Century: An Ethnographic Perspective, eds. D. Green and F. Siegmund, Studies in Historical Archaeoethnology (Woodbridge), 11–36. — 2004, Die Sachsen (Stuttgart).

246

Z01 Hines Book B.indb 246

07/05/2021 13:30

Frisians and Saxons Steuer, H. 2003, ‘The beginnings of urban economies among the Saxons’. In The Continental Saxons from the Migration Period to the Tenth Century: An Ethnographic Perspective, eds. D. Green and F. Siegmund, Studies in Historical Archaeoethnology (Woodbridge), 159–92. Taayke, E. 1996, ‘Die einheimische Keramik der nördlichen Niederlande 600 v. Chr. bis 300 n. Chr. Teil III: Mittel-Groningen’, Berichten van de Rijksdienst voor het Oudheidkundig Bodemonderzoek 42, 9–85. — 2000, ‘Onder Franken en Saksen. Friesland in de laat-Romeinse tijd’, De Vrije Fries 80, 9–28. Udolph, J. 1999, ‘Sachsenproblem und Ortsnamenforschung’. In Die Altsachsen im Spiegel der nationalen und internationalen Sachsenforschung, ed. H. Häßler (Oldenburg), 427–48. Van Egmond, W. 2005, ‘Radbod van de Friezen, een aristocraat in de periferie’, Millennium 19, 24–43. Van Es, W. 1990, ‘Drenthe’s plaats in de Romeinse tijd (en de vroege middeleeuwen)’. Nieuwe Drenthse Volksalmanak 107, 181–92. Versloot, A. and E. Adamczyk 2017, ‘Geography and dialects in Old Saxon: river basin communication networks and the distributional patterns of Ingwaeonic features in Old Saxon’. In Frisians and their North Sea Neighbours. From the Fifth Century to the Viking Age, eds. J. Hines and N. IJssennagger (Woodbridge), 125–48. Vos, P. and E. Knol 2015, ‘Holocene landscape reconstruction of the Wadden Sea area between Marsdiep and Weser’, Geologie en Mijnbouw 94 (2), 157–83. Wenskus, R. 1961, Stammesbildung und Verfassung: das Werden der frühmittelalterlichen Gentes (Vienna). — 1976, Sächsischer Stammesadel und fränkischer Reichsadel (Göttingen). Wood, I. 1990, ‘The Channel from the 4th to the 7th centuries AD’. In Maritime Celts, Frisians and Saxons, ed. by S. McGrail, Council for British Archaeology Research Reports 71 (London), 93–9. — 2000, ‘Before or after mission. social relations across the Middle and Lower Rhine in the seventh and eighth centuries’. In The Long Eighth Century, eds. I. L. Hansen and C. Wickham (Leiden), 149–66. — 2001, The Missionary Life: Saints and the Evangelisation of Europe, 400–1050 (Harlow). — 2003, ‘Beyond satraps and ostriches: political and social structures of the Saxons in the early Carolingian period’. In The Continental Saxons from the Migration Period to the Tenth Century: An Ethnographic Perspective, eds. D. Green and F. Siegmund, Studies in Historical Archaeoethnology (Woodbridge), 271–97. — 2013, ‘The pagan and the other: varying presentations in the Early Middle Ages’, Networks and Neighbours 1, 1–22. — 2015, ‘Who are the Philistines? Bede’s reading of Old Testament peoples’. In The Resources of the Past in Early Medieval Europe, eds. C. Gantner, R. McKitterick and S. Meeder (Cambridge), 172–87. Wormald, P. 2003, ‘The Leges Barbarorum: law and ethnicity in the post-Roman West’. In Regna and Gentes. The Relationship between Late Antique and Early Medieval Peoples and Kingdoms in the Transformation of the Roman World, eds. H.-W. Goetz, J. Jarnut and W. Pohl, The Transformation of the Roman World 1 (Leiden–Boston), 21–53.

247

Z01 Hines Book B.indb 247

07/05/2021 13:30

Z01 Hines Book B.indb 248

07/05/2021 13:30

• 9 • STRUCTURED BY THE SEA: RETHINKING MARITIME CONNECTIVITY OF THE EARLY-MEDIEVAL FRISIANS Nelleke IJssennagger-van der Pluijm

M

Framing the maritime Frisians

aritime identity and cross-sea connectivity are amongst the most interesting aspects of the Early-medieval Frisians, yet also some of the most difficult to pinpoint. Historical Frisians are undoubtedly considered maritime people, both by their contemporaries and by present-day scholars, and are distinctively portrayed as connected to the North Sea once known as Mare Frisicum. This is often illustrated by the image Pliny the Elder paints of a people living on elevated platforms in an area that is neither fully land nor sea as ‘resembling sailors in ships when the water covers the surrounding land, but shipwrecked people when the tide has retired’ in Roman times (Naturalis Historia, XVI, 1), or the poetic depiction of a Frisian seafarer’s wife in the Old English Maxims I as an example of the Frisian maritime identity and cross-sea links in Early-medieval context (Bremmer 1981, 74–5; Shippey 1972, 154; Whitbread 1946). Although evocative and important references, without their textual, historical, societal and physical context, these remain illustrations, which in themselves do not prove that early Frisian society was distinctively maritime. To classify Early-medieval Frisian society as one shaped by maritime connectivity, and Frisians as a people with a particular maritime identity, we need to question the more structural impact of the maritime on people and their perception. This article aims to shed some light on the elusive evidence we have for maritime connectivity and to rethink how it shaped society and mindset amongst Early-medieval Frisians. It does so by discussing how we can meaningfully understand maritime connectivity in the historical archaeoethnological frame of this volume on the one hand. On the other, it questions the available evidence, the absence of evidence, and whether we need to look further beyond the descriptive and the tangible. Although focussing on the Early Medieval Period, largely spanning from post-Roman times into the eleventh century, that is not used strictly to demarcate time limits for this article. Rather, through looking at traces of structural maritime connections in a wider frame of time and space, and comparing them with other cultures, this article presents some lines of thought about our understanding of how much the Early-medieval Frisians were structured by the sea. It makes the argument that, apart from visible finds, material and 249

Z01 Hines Book B.indb 249

07/05/2021 13:30

Frisians of the Early Middle Ages representations of anything Frisian which might be considered as an external picture, it should be sought in mentality, worldview and in the absence of a static footprint.

Sources and approaches The concept of connectivity is much more often used than it is defined, and defining it can be challenging (IJssennagger 2017, 21–2). In this paper, connectivity is thought of as a way of looking at maritime connections and of considering the extent to which there were structural connections both with and across the sea over time, that had an impact on a variety of aspects of the Frisian social landscape. This is not restricted to connections with people and areas across the sea but includes the structural relation Frisians had with the sea itself, and with aspects they imaged to be related to or situated across the sea. Rather than the sum of all connections, maritime connectivity is the consistent role of the (cross-)maritime in the physical and mental world of the Early-medieval Frisians. Looking at this connectivity from a historical archaeoethnological point of view means that we take a specific theoretical point of departure. As the initiators of the Historical Archaeoethnological seminar series explained, the three elements in this coined phrase all have multiple meanings and together present an interdisciplinary cultural history (Ausenda 1995, 303). The phrase primarily refers to the fact that we are looking at people in a historical archaeological period, rather than a prehistoric one, indicating the use of written texts as sources and the historical disciplines for assessing them (Ausenda 1995, 303). This presents an interesting issue in relation to Early-medieval Frisians, as we must acknowledge that the vernacular tradition of writing is not known in recorded form until the twelfth century, apart from a handful of succinct and sometimes puzzling runic inscriptions. The first recorded Old Frisian material from, for the most part, the thirteenth century onwards very likely encompasses an oral tradition recorded within it, creating sources with a great time-depth and plurality of layers (Bremmer 2014; Looijenga 2003, 325–8; Looijenga, this vol.; Nijdam, this vol.). These sources can be used in combination with contemporary sources from outside the Frisian area, as we most often hear about Early-medieval Frisians once they appear in sources from across the sea, sometimes reflecting their physical presence there or others visiting Frisia. This interesting and challenging situation of source material availability is one of the reasons Early-medieval Frisians remain so enigmatic. Using the various sources together to shed light on the coastal dwellers means that we automatically apply a multi-layered ethnological perspective; by looking back at a people in a different time through Frisian material filtered by the later medieval traditions as well as through contemporary references to the Frisians from outside, such as from AngloSaxon, Frankish, Roman and Scandinavian perspectives. This, therefore, is the broad context in which we should understand any information about maritime connectivity we may encounter. Another way in which we assess the topic is through the archaeological record reflecting a past people (Ausenda 1995, 303–4). Although there will be no place to fully assess the archaeological record in this paper, it is important to highlight and scrutinize some material culture that we associate with the maritime in Frisia and that we often compare it to finds elsewhere to form a picture. The primary evidence is that of the maritime landscape and ships, and both individual ship-finds and comprehensive 250

Z01 Hines Book B.indb 250

07/05/2021 13:30

Maritime Connectivity overviews over ship-finds from the Roman to High Medieval Periods are available for the Netherlands and Germany (Brouwers et al. 2015; Siegmüller and Peek 2015; Kegler and Thiemann 2014; Van de Moortel 2011). Whilst using these finds as a point of departure is very helpful, most of these ships may have been used in creeks or rivers rather than at sea (cf. Van de Moortel 2011, 68), so we must consider how much of the tidal landscape and the inland creek-systems connecting to the sea can be considered part of the maritime landscape. Other finds need to be researched further to provide detail on where they came from and when, and potentially what role they played. The latter can be particularly challenging and most remains tell us little of the societal context in which they were used. Here, material culture that was exchanged and shows cross-sea influences or that sheds light on daily life in the coastal zone is a helpful additional resource. For the societal context, the medieval Old Frisian legal codes are helpful sources that can be used in comparative analysis, and with a lack of contemporary vernacular Frisian sources, the late eighth-century Latin Lex Frisionum (ed. Von Richthofen 1863; Siems 1980) is often used as a contemporary insight into Frisian society. Utilizing these various historical and archaeological sources together indeed provides an interdisciplinary cultural historical approach to our topic. What is perhaps most challenging in this approach is connecting literary references on the one hand and material culture on the other to a people and a location that were not static over a longer period of time. The arguments in the critical debate about the danger of connecting a material culture to a people in general do not need repeating here. But similarly, caution is required when interpreting references to Frisians. The term can have varying meanings in different times, places, text types and contexts, and we should be careful not to apply a present-day understanding of Frisians to the historical situations (Hines and IJssennagger 2017). References on the one hand and finds on the other may not represent or refer to the same people at the same time, which is why we must constantly contextualize and often need to generalize. Considering that we are looking at material culture found in an area historically known as Frisia, and material culture elsewhere that may strongly relate to it, without calling it ‘Frisian material culture’ is key. Equally, we must remind ourselves that when written sources reflect on Frisians, this may mean many things, from people dwelling in Frisia to maritime traders in a more general sense (Lebecq 1990), without always assuming they refer to the people who we believe to have dwelt in the area known to us as Frisia (see Nieuwhof 2013 and 2015 for recent discussion). Ultimately, the ‘Frisians’ remain a construct.

Measuring the maritime Before zooming in on Frisians, it is helpful to consider what elements can constitute or contribute to a maritime profile for a society in diachronic context, other than its geographical location, which is a predetermining factor. To assess this, we can briefly consider and generalize about other groups that we identify as having strong maritime connotations to understand what elements collectively create this image. From the Vikings in Europe to ancient peoples in Oceania, the use of ships and importance of seafaring for their economies is an obvious start. But from it follows that representation of these societies in past and present equally highlights the use of ship symbolism. This includes depictions, miniatures, ship settings and ships in core stories. Furthermore, 251

Z01 Hines Book B.indb 251

07/05/2021 13:30

Frisians of the Early Middle Ages gods or heroes are connected to the sea and seafaring, and their stories are captured in written or oral traditions as well as in vocabulary. These traditions also refer to maritime connections in the form of overseas trade, the maritime and coastal geography, stories of travels and myths of origins. Anthropological studies indeed suggest the importance of the notion of gods, the afterlife, as well as peoples’ origins, being situated across the sea as a much-found phenomenon, in which the sea serves a dual role as connector and divider. This creates a structural relationship with the overseas and therefore with the sea, making cross-sea travellers and the items they bring elements of interest for many different cultures (Helms 1988). Although in no way pretending to provide a full analysis, the point to make here is that the maritime elements in these societies have an importance extending beyond the elementary level of subsistence and are part of the fabric of the group. What is striking about these elements is that they are aspects people themselves use or create as a reflection on the world around them, which are intrinsically embedded in their worldview and societal structure. Whilst these internal elements equally may have been stressed by others in portraying these people as distinctively maritime, the people are not classifiable as maritime-oriented by comparison with others only, nor through the lens of others alone. Their cultural identity is connected to the maritime because their world, worldview and way of living is structured by the relationship with the seascape, originating in a geographical situation. Although we may assume a strong relation with the sea because of the geography in Frisia, in contrast to the above examples, Frisians as we portray them mostly are considered maritime in comparison with others and through external references. Whereas the understanding of the Frisian maritime landscape is continuously improving, we currently lack a good understanding of the role of the maritime within society and culture, which brings us to assess what elements we have for the internal and external picture of the Frisian maritime elements in the Early-medieval context.

Between the tides: the maritime geography of Frisia Historical Frisia, defined in Lex Frisionum as between the Rivers Zwin and Weser around AD 800, was a stretched coastal area bordering the Wadden Sea and North Sea along what is now Germany and the Netherlands (Hines and IJssennagger 2017, 2–3). Arguably, parts of Flanders and North Frisia also formed part of the Frisian area at certain points in time. From the Late Roman Period into the Early Middle Ages, the landscapes of the Low Countries and adjacent areas changed. This applies to the geographical and physical landscape as much as to the socio-cultural, political, economic and religious landscapes, which all interlink (Van Lanen et al. 2015a, 1). Over the centuries, rivers changed their course, shores shifted, there were floods, and woodlands incrementally spread, all having an impact on navigability (Brouwers et al. 2015, 7). Between AD 100 and 800, two points in time that recent studies tend to measure, the landscape changed in a particularly drastic way along the coast (Van Lanen et al. 2015a, 1). For the Early Middle Ages, as for today, the southern North Sea coastal area that encompassed Frisia can be divided into geographical zones – the main zones being the western dune coast, the central tidal terp landscape and the North Frisian islands (Knol and IJssennagger 2017, 5–7). The western coast was permeated by the big, navigable rivers, whilst the central area was tidal and facing various islands. Inland, both areas 252

Z01 Hines Book B.indb 252

07/05/2021 13:30

Maritime Connectivity were bordered by fenny mires, making the inhabitable area throughout this period restricted to the dunes, river banks and higher rims in the landscape – natural or manmade. As the mires were difficult to cross, most navigation would have gone by sea, rivers and creeks. The mires in the Dutch and bordering German Frisian area started to be reclaimed in the late ninth century and throughout the tenth, changing the landscape and increasing the habitable area (Knol and IJssennagger 2017, 15). According to Van Lanen et al. (2015a, 14), the western and northern coastal areas, which were the most dynamic, were almost inaccessible by land routes. In Central Frisia, areas like Westergo in fact formed peninsulas, and further east and north the maritime landscape was even more centred on islands and tidal islands. The islands of Texel, Wieringen and the North Frisian Geestinseln all formed notable parts of this landscape as well. The Vlie as a central waterway through Frisia, connecting the sea to inland rivers, must have been extremely important. All of this must have had a big impact on patterns of habitation and navigation (Van Lanen et al. 2015a, 16). Although displaying serious regional variety, the people in these coastal areas were therefore forced to look outward to the sea and to the areas beyond it, and to each other. Despite increased understanding of the landscape and how its dynamics affected habitation and navigation, the organization and geography of seafaring in Frisia still remains elusive to a substantial extent, particularly in Dutch Frisia. We do not have many physical indications of mooring places, harbours, quays or shipyards, nor do we have the names of places where ships moored. This may be related to the strong dynamic of the landscape and therefore the necessary adaptability of people, leaving fewer traces, and the use of natural features rather than man-made ones. However, research into the archaeological traces of communication, such as transfer locations between cross-sea and regional exchange, in the Elbe-Weser area has revealed traces of landing-sites, beach markets and craft-production sites, highlighting how socially significant control of this infrastructure was throughout the first millennium (Aufderhaar 2017, 149–52). The terps as distinctive man-made landscape features to enable living in a tidal zone are naturally the most visible and informative elements, that most likely served as landing places. A small number of well-known and find-rich sites like Wijnaldum, where there is evidence of both production and exchange, may perhaps be seen in that light. In addition, based on a combination of circumstantial evidence we do have a general idea of where strategic stop-over locations and areas that played a structural role in cross-sea connections were.

Cores of connectivity After the collapse of the Roman Empire, the river network was reorganized, coinciding with the dip in habitation. The subsequent phase of the Migration Period was characterized by high mobility and the moving and merging of various peoples around the North Sea (Nieuwhof 2013). With trade relations intensified, the Merovingian Period saw a boom in inter-regional exchange across the North Sea, as well as with the hinterland. Into the later Merovingian and early Carolingian Periods, inter-regional trade kept increasing, with Dorestad as its flagship between the late seventh and mid ninth centuries (Willemsen and Kik eds. 2010; De Lange and Mol, this vol.). In this period, West and Central Frisians in addition seem to have stayed in connection with neighbouring peoples across the sea, as textual references and particularly the record 253

Z01 Hines Book B.indb 253

07/05/2021 13:30

Frisians of the Early Middle Ages of material culture indicate. These were people with whom they may have shared a Migration-period history, or had close relationships through family ties and cultural exchange, reflected in precious items of jewellery and dress accessories in particular (Nicolay 2014; 2005; and this vol.). Looking at the change in possible routes between AD 100 and 800 in the Netherlands, the considerable increase in the Central Frisian area, now displaying the densest cluster of routes, is striking (Van Lanen et al. 2015b, 154). The routes to the north went over the Vlie, creating a very important and potentially controllable sea-route, although we lack details of any systems of control here. Overseas trade networks were dependent upon locations where sailing routes met inland networks, and the Frisian coastal lands were placed most strategically for that (Kosian et al. 2015, 99). It is therefore not the farthest coastal edges where land meets the sea, but the whole tidal and creek zone where inland waters meet the sea that is important for maritime connectivity. The maritime landscape therefore consists of the seascape as well as extending inland into the landscape of creeks towards landing places, and exchange and building of connections largely take shape between the tides. We think very much of Frisians in the Early Middle Ages as middlemen in trade (Lebecq 1983; 1990; de Lange and Mol, this vol.). As Loveluck and Tys (2006) have stressed, this was an uncontrolled peasant trade, particularly from a centralized Frankish perspective, which made it adaptable and dynamic. The free exchange between the coastal zones resulted in a more common or possibly shared maritime group identity and an abundance of exchanged material in coastal zones, making it hard to pinpoint what is Frisian and what not. The closer affinity with other maritime groups rather than the inland elite may have created different links than we can discern in the textual sources, reaching beyond trade (Loveluck and Tys 2006, 161). Here, taking a non-­national and non-land-based or centralized approach of study may help us understanding this wider maritime element and its implications further in the future, e­ specially if we think about which liminal areas the textual evidence does not shed much light on (­IJssennagger-Van der Pluijm 2018). It is important to remember that where sources are silent on these zones, this does not equal a lack of important developments. After the decline of Dorestad and far into the eleventh century, trading networks dispersed to other places across Frisia and beyond. Apart from the geography of trade, centred in hubs such as now Tiel and Deventer in the Netherlands and Hedeby in Denmark, other areas may show high levels of maritime connectivity, whether through direct or indirect links. For example, for the seventh- and eighth-century North Sea area, finds of sceattas may be used to identify links and cores of connectivity around the North Sea. A recent study of the distribution of these coins (Theuws 2018), which questions our tradition of identifying the place where a type is most found as the place of origin rather than as the place of destination, highlights how important it is to keep rethinking mechanisms behind these patterns, and what they may represent. For the Viking Age, the combination of evidence from written sources and metal-detected finds, including important coin-finds such as the work of Coupland (cf. 2006; 2016) shows, provide a better indication of where these areas are (IJssennagger 2017). These are not necessarily connected to hubs as one might expect, but concentrate in coastal areas beyond them. Moreover, the material includes items that are not classified as traded ware or not particularly connected to what is often called elite exchange, reinforcing the idea of a free coastal exchange system. It remains important to stress that 254

Z01 Hines Book B.indb 254

07/05/2021 13:30

Maritime Connectivity this cannot simply be connected to trade alone. The finds, therefore, show a different pattern that may point to a continuation of a maritime connectivity from before the Viking Age reaching a wide societal group (IJssennagger 2017, 244–56). Over time, it seems that some core areas for cross-sea connectivity shifted, and there certainly are various areas with which Frisia had links through time directly and indirectly, primarily (eastern) England and (southern) Scandinavia, as well as the Frankish hinterland. This is partly due to the geographical circumstance to which the coastal dwellers adapted, but equally is affected by the socio-political changes; shifts in the political situation, the change of religion and the changes in society beyond Frisia in the North Sea world (cf. Nicolay 2014; IJssennagger 2017; Pestell 2017).

The materiality of the maritime and the symbolism of ships Not for want of effort, no Frisian counterpart to the Sutton Hoo, Oseberg and Gokstad ship burials has ever been found. Yet over the last decades, ship-related finds from the Romano-Celtic to High-medieval traditions have come to light along the southern North Sea coast and adjacent waterways (cf. Brouwers et al. 2015; Van de Moortel 2011). The Early-medieval ship-timber amongst these is often found re-used, in a secondary context, and usually only in fragments. Therefore, it is not always easily recognized, unless the details of construction like nails, nail-holes or shape give its previous life as part of a ship away. In a 2015 paper (Brouwers et al.), maritime archaeologists listed the Early-medieval ship-finds from the Netherlands, both riverine and maritime. They included finds from cemeteries, and pieces re-used in other ways or from former river beds, and came to the conclusion that the Utrecht area has the highest concentration of Early-medieval ship-finds. The chronological and geographical distribution of the finds conforms with the pattern of known cores of connectivity and the establishment and decline of trading hubs. The oldest, pre-seventh-century finds are from the Rhine estuary and the terp region, including the Wijnaldum area. For the seventh century, the finds are primarily from the Utrecht-Dorestad area and around AD 1000 finds are known from the new trade centres of Tiel and Deventer (Brouwers et al. 2015, 14). As highlighted by Van de Moortel (2011, 101), there is a shared tradition of boat-building in the entire southern North Sea coast, which may have developed from a combination of Romano-Celtic logboat traditions and planking traditions possibly brought by Germanic seafarers in the Migration Period. Up to the Carolingian Period, most finds were of logboat type, indicating small-scale navigation, although extended logboats would have been seaworthy. Some clinker-built vessels in the Nordic tradition, whether built locally or further away, may have been used for coastal and maritime navigation (Van de Moortel 2011, 101). For the late ninth century, the Anglo-Saxon Chronicle mentions that in commissioning his fleet, Alfred the Great wanted ships of a non-Danish and non-Frisian design (Anglo-Saxon Chronicle s.a. 896), implying that there was a distinction in tradition between these two groups (Whitbread 1946). However, we must also keep the option open that they could signal a distinction between ships for war and for trade that were attached to ethnic labels, with Danish signalling warfare, and Frisians referring to trade. Before c. 850, much of the timber used for ship-finds in the Netherlands came from the German Rhineland, whereas between c. 950 and 1050 it came mostly from 255

Z01 Hines Book B.indb 255

07/05/2021 13:30

Frisians of the Early Middle Ages the Ardennes, a source also represented in an Early-medieval ship found in Sheffield, Yorkshire (Brouwers et al. 2015, 14–15). One ship from Tiel was partly built with wood from the south-west of England (Brouwers et al. 2015, 20). Some of the ninth- and tenth-century finds, for instance, remains of one ship found in Dorestad, are thought to have been in the northern clinker-built tradition, whether built locally or elsewhere. It is worth emphasizing that in various recent excavations, especially along the central Frisian coast, clinker-nails have been found, often on burial sites, but not aligned like a ship (Brouwers et al. 2015, 22). With a relatively high number of the ship-timber finds being in re-used context, perhaps re-use or deposition of clinker-nails is not surprising. However, we must question if the deposition of clinker-nails in burial grounds could have an additional, symbolic meaning, similar to boat-burials. A seventh-century grave in the shape of a boat made from fragments of a clinker-built vessel was found in Solleveld, in the present-day province of South-Holland (Waasdorp and Eimermann 2008). Whilst researchers hesitate to classify it as an actual boat-burial, there appears to be a consensus that it is deliberately using ship imagery. In 2015, the impression of a ship was discovered in the eighth-century gravefield of Dunum, Ostfriesland (Siegmüller and Peek 2015). Although now gone, a body, most likely of a woman, had been interred in part of a boat that still measured 2.60 m in length. What did remain were finds of glass beads, metal and fabric, suggesting a burial with grave goods (Siegmüller and Peek 2015, 48). It appears, therefore, that the re-use of boats in burials does occur amongst other forms of re-use of ship material. The idea of boat-burials reflecting a journey to the afterlife, which is then thought of as across the sea, is well known (cf. Dobat 2015, 177). We can question if the finds mentioned here represent a local tradition of ship symbolism, and if so whether they are the best indicators of maritime connectivity for Frisians in general, or rather present a maritime connection for an individual. Tangible representations of ships and the maritime beyond ship-finds from Frisia are restricted to a handful of miniature ships or models that may have served different purposes from toy to container (cf. Van de Moortel 2011, 75) and depictions on coins, such as the Dorestad coinage (cf. Van de Moortel 2011, 93). Compared with other regions and periods that we consider maritime, this is very little and may reflect a very different pictorial tradition. To find depictions of the maritime, we need to look at oral and written traditions, both in sources looking at Frisians and as a representation of maritime imagery in Frisian sources.

Navigating maritime imagery References to Frisians and Frisia in maritime contexts can be scarce and sometimes constitute no more than naming a Frisian, but by combining and contextualizing them, Lebecq (1983) in his still relevant overview paints a picture of ‘the Frisian’ as an image and reality through the period. Those references indicate that, outside of Frisia, the Frisians and the sea are often paired, not dissimilar to the way Vikings and the sea get paired slightly later or at the same time (IJssennagger 2017, 169–71). Moreover, they show that Frisia at large is on the geographical radar for the Early-medieval people and is mentioned as destination and stop-over when seafaring (IJssennagger 2017, 159–65). Unfortunately, it often does not become clear what part of Frisia is meant with certain references, nor who is meant by Frisians. Sometimes the nature of the area as described can provide an 256

Z01 Hines Book B.indb 256

07/05/2021 13:30

Maritime Connectivity indication; the image the Icelandic Egils saga Skalla-Grímssonar paints of the Frisian landscape and how to navigate it is striking for the wet tidal landscape, whether this is a contemporary image or a later one (IJssennagger 2017, 160–1; Egils saga, ch. 70). We can add to this picture the maritime imagery in the Frisian material, with all its challenges and through its later Christian Carolingian veil (Nijdam 2009 and this volume). Looking through the various Frisian Law codes, the sea is often paired with danger. It is connected to incoming threats like Vikings and floods, against which Frisians need to defend their land with various weapons (IJssennagger 2017, 122–7). The Frisian areas, of which there are seven, are called the Frisian sea-lands, and if one of these areas is attacked, people from the others come to help. In Carolingian times, Frisians are even exempt from further military duty as their main task is to defend the coast and with that, the empire. Apart from the sea and the dangers it brings itself, the areas across the sea can have negative connotations as well. There is reference to people being taken on board ships to be taken away across the sea, and people could indeed be captured and taken across the sea as slaves. In the course of the Early Medieval Period, Frisia transitioned from being the southernmost non-Christian Germanic area to being the northern rim of the Christian and Frankish world. This change is reflected in the Law Codes as a transition between north and south, past and present, stating that all Frisians once belonged to the North. Consequently, the geographical focus shifts from the maritime North Sea area to the land-focus of the Continent. In the texts, the ties with the maritime North Sea world after the Viking Age are clearly cut, aiming to make the Frisian world a more continental one, and by doing so they stress the importance of the maritime Germanic world in Early-medieval times (Nijdam 2009; IJssennagger 2017, 129–32). Due to the fact that the Old Frisian corpus is primarily legal material viewed through a Christian Carolingian veil, the opportunities are not highlighted in the same way as the threats, yet they can still be seen between the lines, providing the impression that Frisians seem to have been able to adapt to Viking activity and act as mercenaries within this context. For example, many have highlighted how the reference to a Frisian being taken captive by Vikings and performing attacks with them, who upon his return is found innocent if he can swear he was forced, indicates others joined voluntarily and were guilty (Jesch 2004, 257; Vries 2007 17–18). Outside the Frisian texts, their venturesome role is particularly highlighted by the fact that Frisians appear in trading towns across Europe in the Early Medieval Period, where they sail into the light of the written sources such as the Frisian merchant colony in the 770s in York described in the Vita Liudgeri (Diekamp 1881, 17) or can be seen in the archaeological record such as in Kaupang (Skre 2010, 137–41).

Looking through the lens of Lex Frisionum With the vernacular Frisian sources being of later date, Lex Frisionum (ed. von Richthofen 1863; Siems 1980) recorded in the late eighth century, is often used as a contemporary insight into Frisian society, albeit in Latin and through Carolingian intervention, capturing a moment of transition. Moreover, the version of the text that is handed down to us appears to be in draft form, an inventory of various Frisian regulations to which Carolingian additions were made, and only survives in a 1557 printing. It is important to bear in mind that this text is not trying to paint a picture of society, but that the 257

Z01 Hines Book B.indb 257

07/05/2021 13:30

Frisians of the Early Middle Ages collection is one of Frisian customary, criminal law. Yet through the listed crimes and compensations we do get glimpses of important elements for late eighth-century life in the Frisian areas. At first glance maritime elements are conspicuously absent, but with rethinking, our perception changes. The two instances where one might expect maritime references which are lacking, are in the clusters of regulations concerned with fines for people who work with their hands and those concerned with blockages of routes. The person who hits a harp-player or a goldsmith on the hand has to pay four times as much wergild as for the same crime committed against any other man of the same status. This regulation is followed by the statement that the same applies to a woman who makes fresum, Frisian cloth (Lex Frisionum, Tit. III.23–24). Together these imply that as their hands were so valuable for their particular skills, which therefore must be a specialism, and thus for making their living, they need to be more ‘insured’ against damage than others. One might have expected to see shipbuilding or handcrafts in this list. However, that would imply it was a particularly specialist skill; and perhaps shipbuilding was more common and relied on communal efforts, and particularly on the skills of carpenters and blacksmiths that are not mentioned here either. Moreover, we cannot dismiss the possibility that fresum or Frisian cloth may have been used for sails as well as for clothing. Perhaps more interesting is the absence of the maritime in the list of obstruction of routes. Blockage of a road so a person cannot pass is listed as a crime, the subsequent regulation mentions the price one has to pay for throwing another man of a horse, indicating horse riding, whereas the third rule, about obstructing a river, reads: De Flumine obtruso. Si quis in flumine viam publicam occluserit, XII solidos componat.1 (Lex Frisionum, Tit. VI, Saxmund). Given the presumed essential role of ships and seafaring, it would not have surprised us to see one about blocking access to the sea or throwing someone off a ship, and if there had been such a regulation it would no doubt be quoted as an example of the maritime importance by all disciplines studying the Frisians. However, as established above, the rivers in Frisia are to a large extent essential parts of the maritime landscape and access to rivers means access to the sea and to sailing routes. Moreover, the regulation mentions public passage in rivers, indicating that the waterways to an extent are providing public access and that sailing routes which must include those to sea are as such regulated and protected. Rivers and water occur also where someone is pushed into the water and goes under, thrown into a river or deep water where he cannot stand and needs to swim, which is compensated with 12 solidi, whereas if one saves someone from the water the victim is compensated with 4 solidi. The fact that maritime elements at large are absent does not mean they were not important. There may be ample reasons why no maritime character is found in these regulations, the main one being the fact that Lex Frisionum, like the rest of the Leges Barbarorum and other legal codices, and their Frankish scribes, were primarily focused on the land in general. It has been pointed out (Ian Wood, pers. comm.) that in comparison with the other Leges Barbarorum, the Lex Frisionum in fact is relatively ‘wet’. It seems therefore that, although sparse, some of the water-rich character of the landscape and society still seeps through.

1 ‘Of the obstruction of a river. He who in a river obstructs public passage has to compensate with 12 solidi.’ (trans. author).

258

Z01 Hines Book B.indb 258

07/05/2021 13:30

Maritime Connectivity The sea is mentioned once, in the last and most quoted regulation of the Lex Frisionum. The regulation on the honour of the (heathen) temple stands out amongst the regulations in both content and in style. It dictates that: Qui fanum effregerit, et ibi aliquid de sacris tulerit, ducitur ad mare, et in sabulo, quod accessus maris operire solte, finduntur aures eius, et castratur, et immolatur Diis quorum templa violavit (Lex Frisionum Tit. XI).2 The gripping image this rule paints is seen as a pre-Christian remnant that may only have survived in the collection because it was still in draft. As has been stressed by various researchers (cf. Siems 1980), the rule stands out both in its pre-Christian content and in form. Compared with the other regulations, this one is in more narrative form. Using the sea and the tides as a means of punishment, is mirrored in a story in Vita Vulframni where punishment by being tied to a pole on the beach, waiting for the tide to roll in and drown the criminals, is described. These instances seem to reflect that the coast was an important location for such a symbolic act of punishment, of sacrifice related to a god (cf. Hines, this vol.). Where land meets the sea is related to the transition between life and the afterlife, and to ends and beginnings.

Origins of maritime mentality? Anthropological studies highlight the idea of the liminal coastal zone being one of ends and beginnings, as well as connections with the afterlife, for many cultures. This also includes myths of origins and the idea of ancestors coming from across the sea (cf. Helms 1988; 1998). We do not formally know a Frisian myth of origin, but glimpses in the orally transmitted legal collection and comparison with related traditions give us the idea that Frisians may have seen their origins in cross-sea context (IJssennagger 2017, 136–8). The Overkeuren law text (Hunsinger Recht, ed. Buma and Ebel 1969, 72) states: Tha alle Fresa skipad weren, tha leweden hia, hoc hira sa erest thene londgong nome, thet hia ene pictunnabernde end tha otherum thermithe kethe, thet hia londgung nimen hede.3 The line implies the Frisians came in several ships over open water and went ashore. It has been suggested that this is a fossilized remnant of a fuller story (Vries 2007, 50) and may be reflected in later medieval stories of Frisians coming from an island and arriving in Frisia (Bremmer 2004, 129–30). As Bremmer (2014, 32–4) notes, the appeal to the past and use of graphic details strongly signifies orality. In a way similar to the regulation on the desecration of the heathen temple in Lex Frisionum, this rule stands out in both content and form and is far more narrative than the other regulations. Therefore, we can suggest this is an older, orally transmitted migration myth, which has resemblances with Anglo-Saxon and Scandinavian myths of origin (IJssennagger 2017, 136–9). For pre-modern societies, the reflection of changes and innovations, threats and opportunities coming from the sea, has been theorized as power being of external origin. This, in turn, is for instance reflected in the presence of foreign rulers. For 2 ‘Whoever shall have broken into a shrine and have taken something from what is sacred there, he should be led to the sea, and in the sand which the tide of the sea is going to cover his ears will be cut and he shall be castrated and sacrificed to the gods whose temples he has violated.’ (trans. Hines, this vol.). 3 ‘When all Frisians were shipped in, then they promised that he who went ashore first, would light a barrel of pitch to indicate to the others that they had gone ashore.’ (trans. author).

259

Z01 Hines Book B.indb 259

07/05/2021 13:30

Frisians of the Early Middle Ages Frisia, this may be true for the Christian missionaries, but possibly also for the new laws introduced by the Christian Franks. The latter is represented in the Sage van Karel ende Redbad (ed. Vries 2007, 74–7) where the twelve law-speaking Frisians (asegas) who do not wish to share their laws are sent to sea, drifting until a thirteenth person appears who moors the ship and teaches law. The thirteenth person has been identified as Christ, as well as Odin, suggesting the divine and external origin of the laws, and the Frisians themselves (Vries 2007, 50; Schwartz 1973). Going back to the idea of ship burials and the idea of reflecting a journey to the afterlife across the sea, the reverse may be possible as well. In that line of thought, ship burials and related symbolism representing origins and newcomers such as kings who came from across the sea, have recently been explored by Dobat (2015). This idea fits with a wider tradition of stranger-kings, who are portrayed as the start of a lineage with foreign powers, including mythical kings and gods. As a tradition, this ties in with the idea of founder-graves that need marking out. Rethinking the ship as a symbol for arriving from across the sea, rather than leaving the shores, is what is required in these cases. Once we start looking at the finds in this way, it makes us wonder if ship-related elements in burials could signal something similar, as the need for a functioning ship or the impression thereof is no longer required. Do the burials in Frisia with ship-related material after all then provide a sense of long-term and strong maritime relations, whether real or perceived, for an entire group of people, and can they in turn have been used to strengthen the maritime connectivity and create identity?

Conclusion: maritime identity through connectivity Adding up the various elements mentioned here, there is evidence over time for the maritime character of Frisians, moving and shifting along the shores of the North Sea, but it remains challenging to fully grasp the picture. Some elements are very specific, such as the particular occurrences of the sea in Frisian legal documents, whilst others, such as the material evidence for ships and seafaring, remain rather general and tie in with a wider southern North Sea picture. For a large part, the challenge of the image is down to the challenges of the sources, with the lack of contemporary Frisian written material at its heart. The Frankish narratives including Frisia and the various Frisian legal collections that we see through a Carolingian veil are land-focused by their nature. It appears that an originally clearer maritime mentality and structural relationship with the sea both for subsistence and for the social and religious aspects of worldview in the earlier Medieval Period became overshadowed by the land-focused and centralized frame of the Carolingians that incorporated Frisia, Frisians and their traditions. Yet it is also in these partly external textual traditions that Frisians are connected to the sea, both by comparison with others and in the remnants of their internal stories. For instance, it is found in their specific duties of guarding the coast, which is prompted by their position on the edge of the Continent and their ability to navigate the coastal stretch. But perhaps more interesting, it can be found in the idea of origins, ancestors and the otherworld being related to the sea, seafaring and tidal zones. Moreover, it is important to remind ourselves that the radar of the textual sources is not for the most part directed at the sea, and we often only hear of Frisians when they, the material they exchange or the name they make for themselves reaches other shores. 260

Z01 Hines Book B.indb 260

07/05/2021 13:30

Maritime Connectivity Meanwhile the archaeological record does show us that navigation over water, both across the sea and on inland waters that together form the maritime landscape, was a continuous tradition, and this is similarly the case in neighbouring coastal areas. Material culture in Frisia does not stand out amongst the other Germanic coastal regions in terms of providing a maritime image, yet it is clear that its specifics are prompted by the local geography. The tradition of navigation in a dynamic landscape was tied in with exchange of people and their objects, and can particularly be seen in trading hubs as well as shifting cores of connectivity across the landscape, depending on the exact political, geographical and economic situation each time. The mobility and the adaptability that was needed in the dynamic coastal zone where land, sea and creeks all came together and constantly moved, may have created a way of life that leaves less of a material and textual footprint, with orality playing an important role, than in the case of more land-based groups. It could be argued that the absence of, for instance, centralized courts where texts are produced, roads, harbours, religious sites and thing-places – which we do know from, for example, Anglo-Saxon, Merovingian and Carolingian contexts – is in fact the best indicator of a truly maritime character of the society. Rethinking what is invisible to us and asking why that is the case, shows us that we are trying to find static clues for something that is dynamic and by its nature obscure, so we must look beyond. We can add to this the realization that we often only read of Frisians through the lenses of their neighbours when Frisians appear on their shores or when they travel to Frisia, indicating a strong mobility which becomes visible only when within view of external centres. Altogether, it can therefore be argued that maritime connectivity is at Early-medieval Frisian society’s core. We have to understand that living in a maritime environment and by means of maritime navigation was a reality for Frisians throughout the Middle Ages, and that this informed their mentality beyond the tangible. To fully grasp the maritime nature of these Early-medieval Frisians, we need to consider both this reality of maritime connectivity in the form of shared and exchanged (material) culture, routes and ships, and the perceived maritime connectivity, reflected in a sense of maritime identity and affinity with other maritime groups, origins and otherworldly power, rather than assessing it in our own frame of reference that remains land-focused. Once turned to the sea, the Frisians are almost outside the view of the scribes on land who created the textual sources we have at our disposal and outside the reach of archaeological investigation, making our historical and archaeological research methods inapt. Until we learn to understand the mindset connected to this maritime social landscape in a more anthropological way, we cannot begin to understand the Frisian coastal dwellers. Using the historical archaeoethnological light therefore helps us to see maritime connectivity as a way of life for the Early-medieval Frisians that connected past, present and future.

Discussion IJSSENNAGGER-VAN DER PLUIJM  In my paper I focus on how you can address maritime connectivity in a historical archaeoethnological way, and what we actually know about it if it really comes to the evidence. As well as listing some types of evidence (without attempting to give a full overview), perhaps this is just as much of a theoretical or conceptual topic, and the paper is part of a thoughtexperiment. Following from that I would be keen to discuss and get your input 261

Z01 Hines Book B.indb 261

07/05/2021 13:30

Frisians of the Early Middle Ages particularly on whether we can and whether we do distinguish enough between riverine, maritime and coastal navigation? How can we better understand where ships go – do we pay enough attention to the fact that they are meant to move? If indeed Frisians gathered at things as we have seen and heard, should there be relatively big landing places for ships there and can we recognize them? Do we know more details about who owned the ships, either individuals or as communal vessels? Who controlled the waterways and how? And can we distinguish between regional traditions of seafaring and changes over time? VERSLOOT  You are very familiar with the situation in the Viking Age. Do you have any reason to think that the situation – so the way of organizing ship ownership and exploitation – was fundamentally different one or two centuries earlier in this North Sea area? Different perhaps from the clearer picture of the Viking Age? IJSSENNAGGER-VAN DER PLUIJM  No, I don’t have any indication of that per se. The clearer picture I know is indeed from the Viking Age, but also from outside Frisia, and our image of other periods may shape our picture of the Early Medieval Period to some extent. So I am trying to think back to the pre-Viking period as well and to bear in mind differences in scales, circumstance and regional variation. The examples that we have from the Viking Age are larger ships and crews, and seagoing vessels. Also, there must be differences between navigation within Frisia and along the coast, with riverine and coastal vessels, and with seafaring and seagoing vessels. There must have been systems and ways of organizing these practices in the past, and we need to think about those variations in the practices as well rather than simply think ship means seafaring. KNOL  I think there are some clues. We have a growing quantity of ship-related finds, although still very limited. In this area of the northern Netherlands, where wood was scarce, the wood would have been burnt if possible. But there are finds of clinker-nails, as you list, which seem to be from clinker-built seagoing vessels. We also have more canoe-like finds; like a half-canoe found in Jemgum, just across the border (Kegler and Thiemann 2014). It is well radiocarbon dated to the seventh century and is half of a complete river boat. These two types could well fit in with a distinguished difference between riverine and more coastal navigation. In general, I assume ships coasted as much as possible, and rarely went out into the open sea – only if necessary. They would have needed to do so to cross to England, somewhere around Katwijk. One more ship I would like to add is the ship burial at Hogebeintum, which is a bit problematic. According to the excavation records, a skeleton was buried in a ship, but unfortunately no finds remain of it, only a picture. The picture is difficult to interpret as they chalked the alleged ship for it, making it difficult to identify even for consulted specialists. It has kept me wondering about the possibility. There have been ship-finds at Dunum and Solleveld (Siegmüller and Peek 2015; Waasdorp and Eimermann 2008) which you mention in your paper. At Hogebeintum the excavators would have been familiar with coffins as they excavated many, and with ships as they were part of everyday life, so I think this ship burial was genuine, even if the picture of it remains problematic. You also raised a question about gathering for the thing. There are very few gathering sites known, but the Upstalsboom is famous and that location has good river access via a small river nearby. 262

Z01 Hines Book B.indb 262

07/05/2021 13:30

Maritime Connectivity IJSSENNAGGER-VAN DER PLUIJM  These are very good and helpful points. I absolutely agree there are clues, and I mention some like the clinker finds in my paper, which is not intended as a full overview. Certainly, more evidence can be added and without doubt there are different types of vessels. There are papers listing the finds and they indeed clearly distinguish between them (Brouwers et al. 2015; Van de Moortel 2011). Although recognising the differences between seagoing and riverine vessels, in general the discussions of this topic tend not to go into the extent of what the different practices actually mean. We cluster them as shipfinds, but I think they represent different phenomena to some extent, which I am trying to understand. So perhaps we can focus on those questions in some of our discussions. KNOL  There is a third category to add. These are very small boats for very close to home use. They were used in the tidal marsh area, in small gullies. These were used in one’s own environment, for fishing and fowling. MAJCHCZACK  I would like to comment on the search for the harbour and thing sites, especially the big landing places. During the six years of our North Sea harbour project, we tried to find harbour sites along the coast, the tidal inlets and the rivers. It is very difficult to find archaeological evidence, nothing turns up. Within the coastal landscape, the settlements are located on the terps and on the elevated geest edge, with so many little creeks and rivers which form perfect natural harbours. What should we look for? The small coastal sites do not need elaborate harbour infrastructure as large trading sites with lots of maritime traffic do. For most of our sites, a sheltered waterway and a pole to tie up a ship will do. There are many well suited natural harbour situations close to the settlement sites, but little harbour infrastructure to find. Then there are the beaches, which can be used as they are. Instead of searching for harbour infrastructure, we then started to look for settlements with maritime elements in their material culture. The question is, how can we find the thing sites without the harbours? IJSSENNAGGER-VAN DER PLUIJM  That is exactly the question. MAJCHCZACK  On the question of who owned the ships, I think this would depend on the type of ship. The very small ones would have been quite common. The large troopships belonged to the kings or local leaders, while the smaller ships surely belonged to the traders. The Hessens terp in Lower Saxony is a good excavated example (Siegmüller 2010). A local terp resident started trading and had a quite unique slip construction of timber for bringing a small ship out of the sea. IJSSENNAGGER-VAN DER PLUIJM  That is indeed how we think about the ship ownership. But can we actually tell, perhaps from written sources, if this was the case? Or can we tell how this was seen and how the ships were used? MOL  For the later Middle Ages we have some clues about ships operated by monasteries, through the conversi, the lay-brothers. That is as late as 1250–1300, so we need to question if we can use that retrospectively. Nearly every monastery then was running a ship, to obtain wood and corn from places like Hamburg and Bremen. We see trading expeditions around four times a year. There is some evidence of trading with Flanders too; and from Mariengaarde to King’s Lynn on one occasion, possibly due to a storm (Mol 2001). They probably sailed along the coast, but could cross the sea as well. We don’t know what the ships, called skuta, themselves were really like; perhaps similar to a kogg. But this is all later evidence. 263

Z01 Hines Book B.indb 263

07/05/2021 13:30

Frisians of the Early Middle Ages NIEUWHOF  I think it is indeed good to summarize the evidence of the ships. This was of course a highly maritime landscape, so there must have been many boats, even if the evidence is sparse. The earliest type is the dug-out canoe; in the Roman Period they had expanded dug-outs, which could go along the coast. There is evidence of that from Jemgum mentioned earlier and Leeuwarden (Dijkstra and Nicolay 2008). These stay in use far into the Middle Ages. The clinker-built ships probably come in from northern Germany or Jutland, like the Nydam ship. This type could cross the North Sea. The third type is the flat-bottomed ship. We now have well-dated evidence from Britsum, dated between 685 and 780 (Nieuwhof and Van Holk 2018). So that is a good type for local, regional traffic in creeks. WOOD  There are also a few parallels to what Hans was saying. We have quite a lot of evidence for seventh-century monasteries controlling ships, for example at Quentovic and Kentish nunneries. Interestingly, these involve abbesses who controlled cross-Channel shipping; exactly why nobody really knows. Another example is Jarrow, which sits next to the largest natural harbour of northern Europe. The old photos of Jarrow show the layout in the mud of the tidal harbour. NICOLAY  I am interested in how open this northern landscape was, and how it could be controlled, in terms of ships coming in and out. There is more Merovingian pottery in Westergo than elsewhere; I think importation is controlled, from that area. How can you prevent people coming into your open landscape? There are sea-barriers in Danish coastal waters, but no evidence for this in the salt-marsh area of the northern Netherlands, although it has to be said that none of the creeks have been excavated in this area. How and who can control the maritime landscape is an interesting question. IJSSENNAGGER-VAN DER PLUIJM  I absolutely agree and think we can also ask this in relation to Lex Frisionum specifically (MGH LL 3, III; Siems 1980). The regulations do talk about controlling the roads and blockages to roads. There also is a little bit referring to rivers, but nothing at all about going to sea. There are no references to where the rivers enter the sea. This absence interests me. It made me question who controls but also who potentially owns these waterways, if any? All references to ownership are about land ownership, but the water in general was equally important. As we have not researched these creeks or looked in more detail at the contact points between land and sea, that would be interesting to do in the future. Where land and sea and rivers meet is where lot of activity of importance could have happened, or where people saw significance. The liminality of the tidal area is important. I find it very striking that in Lex Frisionum the one who desecrates a temple is brought to the tidal zone of the coast and executed specifically in the tide. That is a very symbolic act. The sacrifice is specifically performed in the marginal area of the tidal coastal landscape. There must be so much more to it and may also reflect mentalities. FLIERMAN  I think you’re right to stress that the written evidence is very fragmentary, but I also feel that it might nevertheless add up to something. I’m thinking of the many hagiographical and historiographical accounts that show Frisians turning up along major waterways, over and over again, and far inland as well: we frequently encounter them along the Weser and Elbe, and occasionally even the Danube. The mechanism underlying such activity seems to be that they 264

Z01 Hines Book B.indb 264

07/05/2021 13:30

Maritime Connectivity enter one river, then return along that same river towards the coast, and then go down another river. This ties in with your point that the riverine and coastal networks connect very closely. A specific instance underlining the extent of such networks, and Frisian participation in them, is the so-called Fossa Carolina, a canal that was supposed to connect the Rhine and the Danube (which was never finished, by the way: the project was abandoned in an advanced stage due to persistent flooding). Anyway, recent excavations show that this project was initiated near the end of 792 (Kirchner et al. 2018). Its purpose seems to have been economic as well as military: the Fossa Carolina was initiated just after the 791 Carolingian campaign against the Avars in modern Hungary. What is striking to me is that Frankish sources explicitly refer to Frisians aiding the Carolingians on this campaign: they apparently provided and operated ships with which a section of the army navigated the Danube towards the Avar heartland. Building on this, we might also think of Frisian involvement in the Fossa Carolina a year later: they would certainly have been among its beneficiaries and prospective users and could also have brought in expertise to aid its construction. IJSSENNAGGER-VAN DER PLUIJM  Those are very good and detailed examples, and it is of course true that the written sources have more to say, especially taken together. I do not mean to suggest there isn’t any evidence. But I suppose I try to deconstruct the apparent image of the seafaring Frisians and their maritime world first, before building it back up. We take it as self-evident, but I think it does need questioning and comparing. There is a lot to be said when we bring different sorts of sources together. With the evidence we do have I am trying to get a sense of what this maritime image actually is. Sometimes we too easily take one reference to a Frisian ship as meaning Frisians are seafaring people, but there is so much else we need to think of. Particularly we need to think about what maritime and seafaring mean, how we can see and assess it and what elements we should look for. For me, the sheer mobility may be a factor resulting in invisibility in some more static sources, whereas it makes Frisians clearly appear in sources elsewhere. Moreover, we perhaps too easily connect Frisian and seafaring without here questioning what Frisian means in the individual cases. Your example of the way in which Frisians sail along the various rivers and their role in the Fossa Carolina are very interesting in that respect. FLIERMAN  I agree, we need to reflect further on the context of these references to Frisian maritime activity. IJSSENNAGGER-VAN DER PLUIJM  There is a much more complex picture we need to look at, involving meaning, rather than simplistic associations. There is of course also the imagery. KNOL  Two small remarks. You are wondering about the lack of maritime elements in Lex Frisionum. In the later Middle Ages there was very limited sea law. There is only the Visby Law, that was used by all seafarers up to the sixteenth century, and I can imagine that it is a later invention. For a long time the sea was a very free, more or less lawless system. IJSSENNAGGER-VAN DER PLUIJM  That is at least how it may seem. At the same time agreements about control of the sea may have existed without making it into writing later and therefore without us being able to trace them. Or perhaps the water indeed was a free or contested zone. The legal material focused on land, and 265

Z01 Hines Book B.indb 265

07/05/2021 13:30

Frisians of the Early Middle Ages that could be either because that is what the laws focus on – see the other Leges Barbarorum – because the sea was a free for all or because that simply was the extent of the Carolingian rule. We are oblivious to whether there was agreement between various groups of Frisians and other coastal dwellers about who ruled what waterway, as we simply do not have their writing. KNOL  Turning back to the canoes, in the Neolithic you have mainly canoes. At that time a great deal of flint from Helgoland, already an island, was being used on the mainland. This means that they were able to cross over to Helgoland that early in date, and that the boats must have been seaworthy. I would not be surprised if the Jemgum find was seaworthy as well. IJSSENNAGGER-VAN DER PLUIJM  One of the other things we can address is the question of scale. All the things we’ve been talking about in this conference, from the linguistic evidence to the brooches, they represent so many dynamics, exchange and different connections. This must have been a vast thing and systematic. KNOL  I agree we underestimate the dynamics. You couldn’t use grain without a millstone, and Eifel millstones are found all over the place, even in the smallest settlements. They are not easy things to transport, but apparently it was possible. It is not just high-cost garnets and luxuries, also just ordinary material that was brought in. HINES  I was struck by your reference to an affinity between the truly maritime Frisians and other maritime groups of adjacent areas rather than with the inland populations. I’m not sure quite what you had in mind, but this perspective might help us to cope with the distinction we have identified between the external perception of the Frisians, as merchants (and male!), and as sailors/seafarers. Internally, meanwhile, we see a Frisian material culture in which the ship does not seem to have any great symbolic status, and the infrastructure associated with seafaring remains invisible. This is very different from Scandinavia, with the shipgraves such as Oseberg, Gokstad, etc. Frisia could be said to have had a maritime culture which is regular but invisible. This appears to me to similar to what Dagfinn Skre has argued for in relation to the outer coastal zone of Norway before the Viking Period: the distinctiveness and significance of this zone had largely been overlooked because there is an inner-fjord zone with its own conspicuous material culture that attracted attention away from the coastal zone (Skre 2018). This seems also to be the case in the Celtic West of Britain. The populations here are never thought of as a maritime people, and yet circumstantial evidence shows that seafaring was important to them: we have trading sites for Mediterranean imports; abundant evidence of overseas travel from missionary stories; the control of the Isle of Man from Gwynedd, etc. We are not talking here in terms of some shared maritime group identity, but there are interestingly common, parallel forms of behaviour – whereby in particular the seafaring people seem to keep to themselves, and their culture is very pragmatic. IJSSENNAGGER-VAN DER PLUIJM  It is about the external versus the internal image, as well as about internal differences within the Frisian area. You have very eloquently hit the nail on the head. It is a very interesting point you raise about that the maritime identity can be common but not shared, which is good to bear in mind. The question in my mind is, do we need to look into this further, or do we 266

Z01 Hines Book B.indb 266

07/05/2021 13:30

Maritime Connectivity just accept that the footprint is not actually a footprint at all but will be seen at best in different elements or ways. What I want to try and understand is how this works in terms of mentality and how certain cultural elements come about. HINES  I find that very interesting. I don’t know the answer, though. A related feature of cross-sea connnexions which we haven’t discussed, however, could be the fact that we tend to look for the shortest maritime route for connexions as implied by distribution patterns that seem to link opposite sides of an area of sea. Those direct crossings could be useful to look at: an example from the later fifth century is the link between Sogn og Fjordane in Norway and Humberside/East Anglia in England in respect of wrist-clasps (Hines 1984; Hines 1993, 87–9). This paints a picture that is quite the opposite of coastal navigation, which may have become more habitual in the seventh to ninth centuries compared with earlier. Wulfstan’s travelogue in the Old English translation of Orosius very strongly emphasizes that he followed a coasting route from Hedeby to Truso, keeping the mainland on the starboard side (Englert and Trakadas 2009; Hines, forthcoming). If there was a significant shift, maybe that had something to do with confidence. Possibly that is something that Scandinavians of the Viking Period overcame, retrieved, or had never lost. VERSLOOT  On the mentality issue and the contacts from the coast and inland, I would like to point out that linguistic patterns tend to spread along the coast, but spread inland in a more limited way. It is more than just contact. There is an interesting contrast then between contacts with people with whom you want to share your language and contacts with people with whom you choose not to do so. There are parallels in the Frisian area from later times. A maritime city with a very different dialect from the hinterland is Hindelopen, but you cannot imagine that they never met or talked to each other. This is perhaps more about language as an identity marker. Language sometimes was very archaic. Schiermonnikoog, one of the Frisian Wadden islands, has a very distinct dialect with a clear maritime profile, as it also archaic in some ways, highly innovative in others. If there were such a thing as a maritime identity along the coastline it could help to explain linguistic features and distributions, even when there were contacts. HINES  Pidgin languages have arisen very widely in maritime contexts in historical times. Sailors are very experienced in dealing with situations in which simple practical interlanguages have to be created. They can become creoles. VERSLOOT  I was rather thinking more of language as an identity marker, not necessarily pidginization. I don’t think pidginization applies to the period we’re talking about; it’s more a matter of difference in accent but then you still are on a level of communicating. KNOL  One question you didn’t ask is what proportion of the inhabitants in the different zones of the landscape were sometimes seafarers and how many never left the land but had to stay behind to do the agricultural work and create surplus. The former was probably a minority, perhaps a quarter or one-eighth, as many more people were needed for agriculture. IJSSENNAGGER-VAN DER PLUIJM  That is a very good point. I don’t think we can know, but it probably depends on time and place. It also depends on reasons for travelling, which would have been varied. It comes back to the point that ships do not necessarily make ‘a seafaring people’. Ships were meant to move, so if you want 267

Z01 Hines Book B.indb 267

07/05/2021 13:30

Frisians of the Early Middle Ages to play devil’s advocate you could say that all that finding a ship in Frisia tells you is that someone came to that location on a ship, at least once. VERSLOOT  It is a possibility. Another interpretation is that a smart merchant doesn’t need a local surplus. You can buy and trade goods within the network. On the islands, in later periods, most of the seafaring male population were away most of the trading season. Women remained to do the farming and that was not for surplus. An extremely maritime society can be imagined, which is equally a possibility. KNOL  If you look at modern Friesland, and look at certain spots like Schiermonnikoog, the seafaring and non-field rich areas are limited. VERSLOOT  In the pre-dyked period I think there were opportunities for a larger proportion of the population to be active in this way. There was probably also a considerable part only involved in agriculture though. MOL  In the case of Hindelopen, the comparison may be wrong. The coastal towns of the seventeenth and eighteenth centuries were all tied into Amsterdam, in a specialized merchant economy, which only arises in the Late Middle Ages. The islands were used to provide crews for the ships. KNOL  But also in Amsterdam – everybody was looking to Amsterdam. IJSSENNAGGER-VAN DER PLUIJM  In relation to this we can consider cultural elements and social life. The maritime is a way of life and it is about mobility and dynamics, which is structural in all aspects of life. This also reflects in mentalities and stories of origins and relations. The hint in the Frisian law codes to all Frisians coming to Frisia on three ships as a fossilized story, is that a remnant of a myth of origin? NICOLAY  I like the idea of going beyond the origin myth and trying to think about the whole cultural and also ideological world.

References Primary sources Altfrid, Vita Liudgeri. Ed. W. Diekamp 1881, Die Vitae Sancti Liudgeri, Geschichtsquellen des Bistums Münster 4 (Münster). Anglo-Saxon Chronicle. Ed. and trans. Michael J. Swanton 1996, The Anglo-Saxon Chronicle (London). Egils saga Skalla-Grímssonar. Ed. Sigurður Nordal 1933, Íslenzk Fornrit II (Reykjavík). Hunsinger Recht. Eds. Wybren Jan Buma and Wilhelm Ebel 1969, Das Hunsingoer Recht, Altfriesische Rechtsquellen. Texte und Übersetzungen Band 4 (Göttingen). Lex Frisionum. Eds. Karl August Eckhardt and Albrecht Eckhardt 1982, MGH Fontes iuris 12 (Hannover). Pliny, Naturalis Historia. Ed. and trans. H. Rackham 1945, Natural History, Vol. IV: Books 12–16 (London–Cambridge, MA). Secondary sources Aufderhaar, I. 2017, ‘Between Sievern and Gudendorf. Enclosed sites in the north-western Elbe-Weser triangle and their significance in respect of society, communication and migration during the Roman Iron Age and Migration Period’. In Frisians and their North

268

Z01 Hines Book B.indb 268

07/05/2021 13:30

Maritime Connectivity Sea Neighbours. From the Fifth Century to the Viking Age, eds. J. Hines and N. IJssennagger (Woodbridge), 149–72. Ausenda, G. 1995, ‘Editor’s postscript’. In After Empire. Towards an Ethnology of Europe’s Barbarians, ed. G. Ausenda, Studies in Historical Archaeoethnology (Woodbridge), 302–4. Bremmer, R. H. Jr. 1981, ‘Frisians in Anglo-Saxon England: a historical and toponymical investigation’, Fryske Nammen, 45–94. — 2004, Hir is eskriven. Lezen en schrijven in de Friese landen rond 1300 (Hilversum–Leeuwarden). — 2014, ‘The orality of Old Frisian law texts’. In Directions for Old Frisian Philology, eds. R. H. Bremmer Jr, S. Laker and O. Vries, Amsterdammer Beiträge zur älteren Germanistik 73 (Leiden–Amsterdam), 1–48. Brouwers, W., E. Jansma and M. Manders 2015, ‘Middeleeuwse scheepsresten in Nederland: de Vroege Middeleeuwen (500–1050)’, ArcheoBrief 19 (3), 6–24. Coupland, S. 2006, ‘Between the devil and the deep blue sea: hoards in ninth-century Frisia’. In Coinage and History in the North Sea World c. 500–1250. Essays in Honour of Marion Archibald, eds. B. Cook and G. Williams, The Northern World 19 (Leiden–Boston), 241–66. — 2016, ‘Recent finds of imitation gold solidi in the Netherlands’, Numismatic Chronicle 176, 261–9. Dijkstra, J. and J. A. W. Nicolay ed. 2008, Een terp op de schop. Archeologisch onderzoek op het Oldehoofsterkerkhof te Leeuwarden, ADC Monografie 3 (Amersfoort). Dobat, A. S. 2015, ‘Viking stranger-kings: the foreign as a source of power in Viking Age Scandinavia, or, why there was a peacock in the Gokstad ship burial?’, Early Medieval Europe 23 (2), 161–201. Englert, A. and A. Trakadas eds. 2009, Wulfstan’s Voyage: The Baltic Sea Region in the Early Viking Age as Seen from Shipboard, Maritime Culture of the North 2 (Roskilde). Heidinga, H. A. ed. 1975, De ‘Noordzeecultuur’. Een Onderzoek naar de Culturele Relaties van de Landen Rond de Noordzee in de Vroege Middeleeuwen (Amsterdam). Helms, M. W. 1988, Ulysses’ Sail (Princeton). — 1998, Access to Origins. Affines, Ancestors and Aristocrates (Austin). Hines, J. 1984, The Scandinavian Character of Anglian England in the Pre-Viking Period, B.A.R 124 (Oxford). — 1993, Clasps-Hektespenner-Agraffen. Anglo-Scandinavian Clasps of Classes A-C of the 3rd to 6th Centuries A.D. Typology, Diffusion and Function (Stockholm). — forthcoming, ‘Wulfstan in Truso: Old English text, Baltic archaeology, and World History’. In Global Perspectives on Early Medieval England, eds. K. L. Jolly and B. Brooks (Woodbridge). Hines, J. and N. IJssennagger 2017, ‘Frisians – who, when, where, why?’. In Frisians and their North Sea Neighbours. From the Fifth Century to the Viking Age, eds. J. Hines and N. IJssennagger (Woodbridge), 1–4. IJssennagger, N. 2017, ‘Central Because Liminal. Frisia in a Viking Age North Sea World’ (Ph.D. thesis, University of Groningen). IJssennagger-Van der Pluijm, N. 2018, ‘Centering the ‘liminal’ Middle Ages: reading between the historical and archaeological lines’. In Rural Riches & Royal Rags? Studies on Medieval and Modern Archaeology Presented to Frans Theuws, eds. M. Kars, R. van Oosten, M. A. Roxburgh and A. Verhoeven (Zwolle), 179–82.

269

Z01 Hines Book B.indb 269

07/05/2021 13:30

Frisians of the Early Middle Ages Jesch, J. 2004, ‘Vikings on the European continent in the Late Viking Age’. In Scandinavia and Europe 800–1350: Contact, Conflict, Coexistence, ed. J. Adams (Turnhout), 255–68. Kegler, J. F. and B. Thiemann 2014, ‘Der Einbaum aus Jemgum im Landkreis Leer oder das älteste Wasserfahrzeug Ostfrieslands’, Archäologie in Niedersachsen 17, 121–24. Kirchner, A., C. Zielhofer, L. Werther et al. 2018, ‘A multidisciplinary approach in wetland geoarchaeology: survey of the missing southern canal connection of the Fossa Carolina (SW Germany)’, Quaternary International 473 (part A), 3–20. Knol, E. and N. IJssennagger 2017, ‘Paleography and people. Historical Frisians in an archaeological light’. In Frisians and their North Sea Neighbours. From the Fifth Century to the Viking Age, eds. J. Hines and N. IJssennagger (Woodbridge), 5–24. Kosian, M. C., R. J. van Lanen and H. J. T. Weerts 2015, ‘Dorestad’s rise and fall: how the local landscape influenced the growth, prosperity and disappearance of an early-medieval emporium’. In Golden Middle Ages in Europe. New Research into Early-Medieval Communities and Identities, eds. A. Willemsen and H. Kik (Turnhout), 99–104. Kramer, E., P. Pentz, I. Stoumann and A. Greg, eds. 2000, Kings of the North Sea, AD 250–850 (Leeuwarden–Newcastle-upon-Tyne). Lebecq, S. 1983, Marchands et navigateurs frisons du Haut Moyen Âge. 2 vols. (Lille). — 1990, ‘On the use of the word “Frisian” in the 6th–10th centuries written sources: some interpretations’. In Maritime Celts, Frisians and Saxons, ed. S. McGrail, Council for British Archaeology Research Reports 71 (London), 85–90. Looijenga, T. 2003, Texts and Contexts of the Oldest Runic Inscriptions, The Northern World. North Europe and the Baltic c. 400–1700 AD: Peoples, Economies and Cultures Vol. 4 (Leiden-Boston). Loveluck, C. and D. Tys 2006, ‘Coastal societies, exchange and identity along the Channel and southern North Sea shores of Europe, AD 600–1000’, Journal of Maritime Archaeology, 140–69. Mol, H. 2001, ‘Mariëngaarde, het Friese kloosterwezen en de Friese maatschappij’. In Vitae abbatum Orti Sancte Marie. Vijf Abtenlevens van het Klooster Mariëngaarde in Friesland, eds. H. T. M. Lambooij and J. A. Mol (Hilversum–Leeuwarden), 59–130. Nicolay, J. A. W. 2005, ‘Nieuwe bewoners van het terpengebied en hun rol bij de opkomst van Fries Koningschap. De betekenis van gouden bracteaten en bracteaatachtige hangers uit Friesland (vijfde-zevende eeuw na Chr.)’, De Vrije Fries 85, 37–103. — 2014, The Splendour of Power: Early Medieval Kingship and the Use of Gold and Silver in the Southern North Sea Area (5th to 7th century AD), Groningen Archaeological Studies 28 (Groningen). Nieuwhof, A. 2013, ‘Anglo-Saxon immigration or continuity? Ezinge and the coastal area of the Northern Netherlands in the Migration Period’, Journal of Archaeology in the Low Countries 4, 53–83. — 2015, Eight Human Skulls in a Dung Heap and More. Ritual Practice in the Terp Region of the Northern Netherlands 600 BC–AD 300, Groningen Archaeological Studies 29 (Groningen). Nieuwhof, A. and A. van Holk 2018. ‘Een vondst van groot belang: de boot van Britsum (Fr.)’, Paleo-aktueel 29, 51–9. Nijdam, H. 2009, ‘Klinkende Munten en Botsplinters in de Oudfriese Rechtsteksten. Continuiteit, Discontinuiteit, Intertekstualiteit’, De Vrije Fries 89, 45–66. Pestell, T. 2017, ‘The kingdom of East Anglia, Frisia and continental connections, c. 600– 900’. In Frisians and their North Sea Neighbours. From the Fifth Century to the Viking Age,

270

Z01 Hines Book B.indb 270

07/05/2021 13:30

Maritime Connectivity eds. J. Hines and N. IJssennagger (Woodbridge), 193–222. Schwartz, S. P. 1973, Poetry and Law in Germanic Myth (Berkeley, CA). Shippey, T. 1972, Old English Verse (London). Siegmüller, A. 2010, ‘Die Ausgrabungen auf der frühmittelalterlichen Wurt Hessens in Wilhelmshaven. Siedlungs- und Wirtschaftsweise in der Marsch’, Studien zur Landschaftsund Siedlungsgeschichte im südlichen Nordseegebiet 1 (Rahden). Siegmüller, A. and C. Peek 2015, ‘Das Bootgrab aus dem frühmittelalterlichen Gräberfeld von Dunum, Ldkr. Wittmund’, Nachrichten Marschenrat zur Förderung der Forschung im Küstengebied der Nordsee 52, 46–8. Siems, H. 1980, Studien zur Lex Frisionum (Ebelsbach). Skre, D. 2010, ‘From Dorestad to Kaupaung’. In Dorestad in an International Framework. New Research on Centres of Trade and Coinage in Carolingian Times, eds. A. Willemsen and H. Kik (Turnhout), 137–41. — 2018, ‘Sea Kings on the Norðvegr’, In Avaldsnes – A Sea-Kings’ Manor in First-Millennium Western Scandinavia, ed. D. Skre, Ergänzungsbände zum Reallexikon der Germanischen Altertumskunde 104 (Berlin), 781–99. Theuws, F. C. J. W. 2018, ‘Reversed directions. Rethinking sceattas in the Netherlands and England’, Zeitschrift für Archäologie des Mittelalters 46, 27–84. Van de Moortel, A. 2011, ‘Medieval boat and ship-finds of Germany, the Low Countries, and northeast France. Archaeological evidence for shipbuilding traditions, shipbuilding resources, trade and communication’, Siedlungs- und Küstenforschung im südlichen Nordseegebiet 34, 67–105. Van Lanen, R. J., M. C. Kosian, B. J. Groenewoudt and E. Jansma 2015a, ‘Finding a way: modelling landscape prerequisites for Roman and Early-medieval routes in the Netherlands’, Geoarchaeology: An International Journal 30 (1), 1–23. Van Lanen, R. J., M. C. Kosian, B. J. Groenewoudt, T. Spek and E. Jansma, 2015b, ‘Best travel options: Modelling Roman and Early-medieval routes in the Netherlands using a multi-proxy approach’, Journal of Archaeological Science: Reports 3, 144–59. Vries, O. 2007, Asega, is het dingtijd? (Leeuwarden–Utrecht). Waasdorp, J. A., and E. Eimermann 2008, Solleveld. Een opgraving naar een Merovingisch grafveld aan de rand van Den Haag, Haagse Oudheidkundige Publicaties 10 (Den Haag). Whitbread, L. 1946, ‘The “Frisian sailor” passage in the Old English gnomic verses’, Review of English Studies 22, 215–19. Willemsen, A. and H. Kik eds. 2010, Dorestad in an International Framework. New Research on Centres of Trade and Coinage in Carolingian Times (Turnhout).

271

Z01 Hines Book B.indb 271

07/05/2021 13:30

Z01 Hines Book B.indb 272

07/05/2021 13:30

• 10 • ART, SYMBOLISM AND THE EXPRESSION OF GROUP IDENTITIES IN EARLY-MEDIEVAL FRISIA J. A. W. Nicolay

E

arly-medieval Frisia, like the axis of a wheel, was situated in between the Frankish, Scandinavian and Anglo-Saxon worlds. ‘Frisian’ works of art reflect this intermediate position, as a fascinating mixture of cultural influences. In this paper, different ways of expressing group identities within the area defined in Lex Frisionum are investigated, in relation to the shape and decoration of metal objects that circulated among its coastal population. These objects include copper-alloy brooches that were exchanged within family groups, prestige goods that were distributed as gifts among members of the regional and local elites, and items with a (pre-)Christian symbolism that could circulate both within family and among elite groups. Chronologically, this paper starts with the migration of ‘Anglo-Saxons’ in the late fourth and fifth centuries, leading to the replacement and partial re-shuffling of the original, Roman-period population of the Dutch coastal area. The sixth and seventh centuries saw the rise of regional and supra-regional kingdoms in this area, until Frisia had become part of the Frankish realm during the Saxon Wars of Charlemagne (772– 804). In this four-hundred-year time span, the transition can be seen from ‘Frisian’ communities to a Frankish population, from independent kingdoms to a Frankish district, and from non-Christian belief systems to Christianity – all affecting the way ‘Frisians’ felt related to one or more co-resident, political and religio-ideological groups.

Artistic expressions in Early-medieval Frisia European missionaries who travelled the world from the seventeenth century onwards, to spread the Christian message, came into contact with a tribal world, in which cult houses, statues and other forms of art played a crucial role in the worship of gods and ancestors. As part of their missionary work, indigenous tribal art was destroyed, some of it taken home as collectors’ items (e.g. Corbey 2000). In the coastal areas of Early-medieval Frisia, the first missionaries will have entered a world that was similar in many respects – and, according to historical sources, they even worked in a similar way, by destroying or removing symbols of non-Christian belief (Milis 2005, 156–8). In the footsteps of Pippin of Herstal, Charles Martel and Charlemagne, Willibrord, Boniface and others operated among coastal communities which knew a rich and varied 273

Z01 Hines Book B.indb 273

07/05/2021 13:30

Frisians of the Early Middle Ages

Fig. 10.1  Front and back of the leather scabbard of a seax from Usquert (Groningen), richly decorated with incised motifs. Probably dating to the seventh century. Scale 1:2. Groninger Museum (Photograph: Marten de Leeuw).

collection of material culture, decorated with a complex set of artistic expressions (see Schön 1999; Brandenburgh 2010; Nicolay 2014). Most artistic expressions in Early-medieval Frisia are to be expected on organic materials, as carved, painted or woven motifs (Fig. 10.1). Even when the preservation conditions are relatively good, as usually is the case for terp sites in the former saltmarsh area of the northern Netherlands, items of or decorated with wood, textile and leather are extremely rare. The reasons for this are obvious: most discarded wood in this treeless landscape ended up as fuel, and both textile and leather require a very specific acidity for their preservation. In the western Netherlands, where many settlements are situated on dunes, organic materials are even more scarce, only to be preserved in wells or other structures that reach below groundwater level. Both pottery and metalwork, on the other hand, are well represented in both the clay and sand regions. Although some pots are decorated and pottery in general will have been used to express co-resident identities (see Nieuwhof, this vol.), this paper will focus on metal finds for two reasons. First, objects of bronze, silver and gold, especially during the sixth century, show a large variety in shape and decoration, allowing the study of cultural, political and religio-ideological relations and related expressions of group identity for individual types of artefacts. Secondly, the systematic inventorization of Early-medieval brooches from the province of Friesland by Jurjen Bos (†) has led to a genuinely invaluable dataset that could be used for this paper (largely unpublished: see Bos 2006; 2008); in addition, many other metal finds from Friesland and other Dutch provinces are now available for scientific research thanks to Portable Antiquities of the Netherlands (see Kars and Heeren 2018). Although the analysis of metal composition can add valuable information to the study of metal finds (e.g. Roxburgh et al. 2014; 2016), this paper has a clear focus on the shape and decoration of metal objects. 274

Z01 Hines Book B.indb 274

07/05/2021 13:30

Symbolism and Group Identities

Symbolism and the expression of group identities A decorative motif or a collection of such motifs could be added to a functional item to embellish its appearance. Such motifs, however, were not randomly selected; instead, specific categories of items during a specific period usually show very similar or closely related decorations – chosen from a pool of motifs that must have been familiar to the people producing and using or wearing any particular item. Especially when symbolic messages were expressed in any specific period, it must have been important that such messages could be ‘read’ and as uniform (sets of) motifs were understood by the observers within local or regional communities. In different contexts, however, these messages may have had different meanings (e.g. Hodder 1982). In Early-medieval Europe the network of relations to which a person belonged, and the way these relations changed during the life path of this person, probably determined what ornaments were worn by males and females. In Beowulf, for example, every individual is introduced in relation to other members of society, and the various stages of Beowulf ’s life are visualized by the receiving of gifts from his father, his king and other kings (Bazelmans 1999). The uniformity of less prestigious items, like copper-alloy brooches, is further proof that certain rules about ‘accepted’ shapes and decorations must have existed. At the same time variation did occur and could be used both as a passive and an active tool (e.g. Conkey 1990) for expressing social relations and group membership at the following levels: 1. The co-resident level Within Early-medieval Frisia the population was not spread evenly along the North Sea coast. In Lex Frisionum, a geographical division is made between a western part (the western Netherlands), a central part (present-day Friesland) and an eastern part (present-day Groningen and Ostfriesland); a further division can be made into around twenty clusters of settlement sites (De Koning 2018, 146–147, figs. 1–2). These ‘nuclear regions’ are defined by Heidinga (1987, 154) as areas with a certain administrative status and cultural identity. When the older definition of so-called ‘adaptation groups’ by Waterbolk (1979, 4–6) is followed, these are groups of local communities that live at close distance, have close contact (e.g. to exchange marriage partners) and share a similar social, economic, political, military and religious tradition. These co-resident groups, each consisting of a few hundred people, also shared cultural traits, like similar house, pottery and burial forms. They can be seen as the basic units of larger ‘cultural groups’ (Waterbolk 1979, 2–3). 2. The socio-political level In tribal communities socio-political relations were established and consolidated by the exchange of gifts, between a king or other leader (the patron) and his personal followers (clients) (Bazelmans 1999). These patron–client relations determined the topdown distribution of objects within specific socio-political entities, allowing the elite to project ‘their’ identity to the group of followers they controlled. According to Geary (2002, 173), such political identities were most relevant for groups of followers who ‘agreed to accept a common leader and, in time, a common identity’. Since prehistoric times, so-called prestige goods of exotic origin or made of gold and silver were used for networking strategies (e.g. Roymans 1990, 41–2). As a result, the distribution of such 275

Z01 Hines Book B.indb 275

07/05/2021 13:30

Frisians of the Early Middle Ages goods, and the related expression of group identity, directly reflected the outreach of a leader’s network of followers and the size of the area he controlled. 3. The religio-ideological level An ideological border divided Early-medieval Europe into a pagan, Scandinavian north, and a Christian, Frankish south. In the same way as Christianity united the tribal groups that became part of the Frankish kingdom, the Nordic belief unified the Germanic people of southern Scandinavia and its periphery. Material culture and decorative motifs played a crucial role in marking the contrast between Christian and non-Christian, this way representing the markers of two important religious-ideological identities (Hedeager 1992; 1993). The first was characterized by animal motifs, man-between-animals motifs and runes, often referring to Odin/Wodan, the second by images of a cross, the Lamb of God, a saint and the Chi-Rho, clearly referring to Christ. This contrast, however, was not black and white. Frankish ornaments from the fifth to the seventh century reflect a period of transition: non-Christian symbols continued to be used, gradually incorporated into a Christian iconography from the late sixth century onwards (Wamers 2009; Kars 2018). Instead of a strict border between Christian and non-Christian, within northern France, Belgium and the Netherlands a wide transitional zone existed (Theuws and Hiddink 1996, 66–7, fig. 49). Group identities at these three levels were not expressed in isolation, but partly or fully overlapped with one another. It is to be expected that co-resident identities were most important at the local and regional level, and both socio-political and religious-ideological identities at the regional and supra-regional level. To investigate the way in which such identities were visualized in Early-medieval Frisia, three categories of metalwork are studied: copper-alloy brooches, gold and silver ornaments, and objects with a Christian or Nordic symbolism. For all three categories, objects from the following periods are compared: the late fourth and fifth centuries, the sixth and first half of the seventh centuries, and the mid-seventh to tenth centuries.

Copper-alloy brooches and the expression of co-resident identities Copper-alloy brooches were among the most common items that circulated in Early-medieval Frisia. For their manufacture scrap could be re-used, otherwise bronze, tin and, depending on the period of production, lead or zinc had to be imported as raw materials. Although detailed information about the number of production sites is lacking, brooches probably were manufactured widely. Imported brooches were also popular among the coastal population, resulting in an interesting mixture of local/ regional and foreign metalwork. Most brooch types were worn by women, as the grave finds from Oosterbeintum (Friesland) and other coastal cemeteries indicate (Tab. 10.1). An exception is the fifth-century, heavily shaped supporting-arm brooches, which were mainly worn by men (Böhme 1974, 42). At Oosterbeintum the only brooch of this type indeed came from a male grave (Fig. 10.2: grave 460). Brooches and other ornaments probably were not personal possessions, ‘owned’ by their users. Instead, they represented the identity of individual families and were transferred to temporary owners during important life-cycle rituals or family events, like a marriage (Kars 2013). This explains why brooches have rarely been found in graves 276

Z01 Hines Book B.indb 276

07/05/2021 13:30

Symbolism and Group Identities of older women; these persons’ ornaments were successfully returned to the ‘family treasure’, for example, by being given to a new bride. For one child, one male and nine females, buried with a (set of) brooches at Oosterbeintum, the age at death is known (Tab. 10.1): besides a child (c. 9 years old) and two females (>40 years old), a man and four women were young adults – like a fifth woman, buried at Wijnaldum (18–30/35 years old). During a specific life-cycle ritual, women especially may have received these brooches from their mother/father or another family member as a temporary gift. Only when they died as a young adult, or when they had no children or were married outside the area of their co-resident group, did the transmission fail – resulting in the burial (or melting) of an ornament. Inter-generational object transmission also explains why one of the graves at Hogebeintum contained an old and a newer type of cruciform brooch (Tab. 10.1: grave 129), and why some brooches are heavily worn or repaired. Late fourth to fifth centuries: Saxon-style and Anglian-style brooches Within the ‘Saxon’ part of northern Germany the development of various types of Armbrustfibeln (crossbow brooches) and closely related Stützarmfibeln (supporting-arm brooches), both with a rectangular or trapezoidal foot, can be seen during the second half of the fourth and early decades of the fifth century (Böhme 1974, 7–14). Their decoration with grooves, sloping notches and point-circle motifs is copied from contemporary, Late-Roman belt-fittings. A similar decoration is visible on fifth-century, more heavily shaped supporting-arm brooches (Böhme 1974, 51–2), which nowadays are seen as hybrid, ‘German-Roman’ products that may have been produced both in northern Germany and in northern Gaul (Heeren 2017, 171–2). Saxon-style crossbow and supporting-arm brooches from the coastal area of the northern Netherlands (for the second brooch form, see Fig. 10.2: grave 460; Fig. 10.3: nos. 1–5) and southern England reflect the migration of ‘Saxons’, probably not before the late fourth century. New settlers used these ornaments to signal their area of origin, as imported and locally imitated brooches demonstrate (for England, see Böhme 1986; Hirst and Clark 2009, 487–8). The lack of finds from the western Netherlands, north of the Rhine, is intriguing in this respect and may relate to a different habitation history: while large parts of the northern Netherlands were deserted during the second half of the third and fourth centuries (Nieuwhof 2011; 2016), habitation in the western Netherlands probably continued into the first quarter of the fourth century, followed by a break in habitation that may have lasted until the late fifth century (Dijkstra and De Koning 2017, 58–61). An association with migrating ‘Saxons’ has also been assumed for fifth-century equal-armed (gleichärmige) or butterfly brooches (Böhme 1974, 14–19; Bruns 2003), showing a chip-carved decoration that is directly copied from Late-Roman belt-fittings again. When silver specimens (see below) are included, the different types of equalarmed brooches that have been found outside northern Germany show a very distinct distribution: in the coastal area of the northern Netherlands, including its hinterland, and across southern England. Within England, Insular forms (e.g. Types Mucking and Berinsfield: Bruns 2003, 22–3) are proof that Saxon-style brooches were imitated locally in the areas where newcomers had settled. The material culture of the immigrants also included ‘Anglian’ brooches of the cruciform type, showing a development from simple to more complex types from the late fourth to the mid-sixth century (Reichstein 1975; Brouwer 2004; Martin 2015; Fig. 10.3: 277

Z01 Hines Book B.indb 277

07/05/2021 13:30

Frisians of the Early Middle Ages

Fig. 10.2  (Above and facing) Copper-alloy brooches deposited as grave goods at the cemetery of Oosterbeintum (Friesland). Scale 1:2 (after Knol et al. 1996).

nos. 6–13). The shape of these brooches goes back to the so-called Nydam brooches (Bemmann 1993; Rau 2010, 146–80), with their typical cross-shaped head, consisting of three knobs. Cruciform brooches are found all along the southern North Sea coast, 278

Z01 Hines Book B.indb 278

07/05/2021 13:30

Symbolism and Group Identities

as well as in southern Scandinavia (Reichstein 1975, map 1). The earliest continental forms, including brooches of Reichstein’s Type Witmarssum (compare Brouwer 2004, ‘Frisian’ type FA1) and Types Groß Siems and Midlum (compare Brouwer 2004, ‘Frisian’ types FA2 and FB; Fig. 10.3: nos. 6–9), show a more limited distribution, across northern Germany, the northern Netherlands and southern England (Reichstein 1975, Abb. 9–11) – confirming their assumed association with migrating ‘Anglo-Saxons’ (e.g. Bos and Brouwer 2005). Sixth to first half of the seventh century: regional-style brooches Around AD 500, a remarkable ‘regionalization’ of brooch types in the southern North Sea area can be seen, allowing the identification of regional identity groups. The older cruciform brooches represent a northern, Scandinavian and a southern, Continental group (Reichstein 1975, Karten 2–3). During the late fifth and first half of the sixth centuries within the group of cruciform brooches various Norwegian, English and also ‘Frisian’ types can be traced (Reichstein 1975, 42, fig. 11), including Type Achlum (compare Brouwer 2004, ‘Frisian’ type FC; Fig. 10.3: no. 10). The youngest cruciform brooches from Frisia can be dated to the mid-sixth century (Brouwer 2004, type FE; Fig. 10.3: nos. 11–13). They no longer terminate in a stylized animal head, but in a human head between two animals – a motif also known from English specimens (Palm and Pind 1992, fig. 5: no. 35). A similar development towards regional types is reflected in various brooch forms that belong to the group of ‘small long brooches’ (Leeds 1945; De Leeuw 2002; Penn and Brugmann 2007, 24–5). The earliest specimens, like contemporary cruciform 279

Z01 Hines Book B.indb 279

07/05/2021 13:30

Frisians of the Early Middle Ages Table 10.1  Copper-alloy brooches and a gold-filigree and silver-cloisonné brooch from grave contexts in Early-medieval Frisia. Cemetery site

Grave

Burial type, sex (age)

Brooch(es)

Publication

Oosterbeintum (Fr)

460

inhumation, male (20–30 y.)

supporting-arm brooch (heavy type)

Knol et al. 1996, 403

Hogebeintum (Fr)

129

inhumation

cruciform brooch, Type Hogebeintum cruciform brooch, Type Midlum

Knol 1986, 127–8

Hogebeintum (Fr)

47

inhumation

cruciform brooch, Type Midlum 2 x small long brooch

Knol 1993, fig. 61

Oosterbeintum (Fr)

398

inhumation, female (35–45 y.)

2 x cruciform brooch, Type Midlum small long brooch

Knol et al. 1996, 395–7

Oosterbeintum (Fr)

A

inhumation

cruciform brooch, Type Midlum small long brooch

Knol et al. 1996, 375

Oosterbeintum (Fr)

360

inhumation, female (40–50 y.)

cruciform brooch, Type Midlum small long brooch small long brooch

Knol et al. 1996, 392–3

Oosterbeintum (Fr)

160

inhumation, female (adult)

cruciform brooch, unknown (single knob)

Knol et al. 1996, 381

Oosterbeintum (Fr)

60

inhumation, female (35–45 y.)

cruciform brooch, Type Midlum cruciform brooch, Type Midlum? (fragment) annular brooch (fragment)

Knol et al. 1996, 377

Hogebeintum (Fr)

130

inhumation

Armbrustfibel, Type Ozingell small long brooch, Type Domburg

Botman 1994, 54 (no. 1)

Oosterbeintum (Fr)

428

inhumation, female (25–35 y.)

small long brooch, Type Domburg 2 x annular brooch

Knol 1993, fig. 60; Knol et al. 1996, 399–401

Driesumerterp (Fr)

-

inhumation

small long brooch, Type Domburg annular brooch

Knol 1986, 65

WijnaldumTjitsma (Fr)

-

single inhumation, female (18–19 y.)

2 x annular brooch

Cuijpers et al. 1999, 305–8

Oosterbeintum (Fr)

374B

inhumation, female (30–45 y.)

2 x annular brooch

Knol et al. 1996, 393–5

Cornjum (Fr)

-

inhumation

2 x annular brooch

Knol 1986, 55

Oosterbeintum (Fr)

295

inhumation, female (20–22 y.)

annular brooch

Knol et al. 1996, 389

280

Z01 Hines Book B.indb 280

07/05/2021 13:30

Symbolism and Group Identities Blija-Sijtsmaterp (Fr)

-

inhumation

annular brooch

Knol 1986, 53

Aalzum (Fr)

-

inhumation

disc-on-bow brooch (disc missing), garnet-cloisonné

Nicolay 2014, fig. 4.2

Hogebeintum (Fr)

112

cremation, Hessens-Schortens urn

annular brooch: sheet-gold front plate, silver edge/pin

Knol 1993, fig. 70; Nicolay 2014, 63, figs. 4.2–4

Oosterbeintum (Fr)

342

inhumation (c. 9 y.)

small equal-armed brooch

Knol et al. 1996, 391

Oosterbeintum (Fr)

393

inhumation, female (>45 y.)

small equal-armed brooch

Knol et al. 1996, 395

Oosterbeintum (Fr)

501

inhumation, female (25–35 y.)

small equal-armed brooch

Knol et al. 1996, 405

Den HaagSolleveld (SH)

1

cremation, Anglo-Saxon urn

bow brooch, Frankish type (fragment)

Dijkstra 2011, fig. 6.17

Rijnsburg-De Horn (SH)

215

unknown

silver disc brooch, garnetcloisonné/cross motif

Dijkstra 2011, fig. 6.5

Rijnsburg-De Horn (SH)

213

inhumation

small long brooch, Type Domburg annular brooch

Dijkstra 2011, figs. 6.5, 8.4

Den HaagSolleveld (SH)

7

cremation, Frankish urn

small long brooch, Type Domburg annular brooch

Dijkstra 2011, fig. 8.4

Den HaagSolleveld (SH)

479

boat grave

small equal-armed brooch

Waasdorp and Eimermann 2008, fig. 5.27

Rijksburg-De Horn (SH)

216

unknown

disc brooch, cross motif

Dijkstra 2011, fig. 6.5

Note: The graves, in Friesland (Fr) and South-Holland (SH), are presented in their (most probable) chronological order. A radiocarbon date is available only for the grave with the gold and silver brooch (c. AD 425–600): see Knol 1993, fig. 13: no. 3.

brooches, have a cross-shaped headplate. They are represented by different Saxon-style and Anglian-style types on both sides of the river Elbe (Böhme 1986, fig. 72), probably dating from the second quarter of the fifth to the early sixth century. Based on the shape of the footplate, among these early types two main forms can be distinguished (Leeds 1945, figs. 1 and 3; De Leeuw 2002, types VIb–c and VId): one with a shallow foot and a spatula-shaped terminal, another with a lozenge foot, terminating in a small knob. Outside northern Germany, both Saxon- and Anglian-style small long brooches have mainly been found in southern England (Böhme 1986, fig. 72); from the northern Netherlands only Anglian-style brooches with a spatula-shaped terminal are known, all from Friesland (Fig. 10.3: no. 14). In the second half of the fifth century, the older forms are replaced by a larger variety of small long brooches, showing more regional, German, Norwegian, English, Kentish and ‘Frisian’ characteristics (De Leeuw 2002, fig. 3.17; for the chronology of English 281

Z01 Hines Book B.indb 281

07/05/2021 13:30

Frisians of the Early Middle Ages

282

Z01 Hines Book B.indb 282

07/05/2021 13:30

Symbolism and Group Identities

Fig. 10.3  (Above and facing) Copper-alloy and pewter (nos. 51–2) brooches from settlement sites in Friesland. Both excavation finds from Wijnaldum-Tjitsma and metaldetector finds are included. Grey: enamel. Scale 1:2 (drawings by J. de Koning: nos. 32–7 after Knol 1993, fig. 60; nos. 38–53 after Bos 2004; 2006).

forms, see Penn and Brugmann 2007, figs. 5.21–22). Although the combination of characteristics within these regional groups has not yet been studied in detail, the sixth-century ‘Frisian’ specimens have a rectangular and sometimes a square or trapezium-shaped headplate (Fig. 10.3: nos. 15–18; Fig. 10.4: nos. 1–2). The footplate and terminal show different combinations of forms: a drop-shaped foot with a round or kidney-shaped knob, a lobed foot with a round knob, and in fewer cases a shallow or lozenge foot with a spatula-shaped or round knob. The closest parallels can be found in Kent, where brooches with a rectangular head are also found; these brooches, however, have a lozenge and sometimes a lobed foot and different forms of terminals (Leeds 1945, figs. 23, 25). One of the most distinctive types of small long brooches is relatively late, dating from the mid-sixth to the late-seventh century (Nicolay and Van Eerden 2021). These so-called Domburg brooches have a kidney-shaped headplate and a rhombic to dropshaped footplate with a kidney-shaped terminal (Botman 1994; see also Koch 1999a; Fig. 10.3: nos. 19–21). The earliest brooches depict three sets of opposing bird’s heads: two sets of curved heads, forming the kidney-shaped headplate and terminal, and a set of ‘hanging heads’, shown as bean-shaped cavities on the shoulders of the footplate. Interestingly, Domburg brooches feature a gradual development in decoration motifs: from brooches with Style I birds’ heads on the headplate and more stylized heads on the footplate and terminal (type A), and brooches with stylized heads both on the headplate and terminal (types B) or on the headplate only (type C), to specimens without any recognizable birds’ heads (type D) (Fig. 10.5). In some cases a stylized human head 283

Z01 Hines Book B.indb 283

07/05/2021 13:30

Frisians of the Early Middle Ages

Fig. 10.4  Copper-alloy brooches from a settlement site (nos. 1–2: Oegstgeest) and two cemeteries (nos. 3–5: Rijnsburg-De Horn; nos. 6–8: The Hague-Solleveld) in SouthHolland. Scale 2:3 (after Dijkstra 2011, figs. 6.5, 6.17, 8.4–5; Waasdorp and Eimermann 2008, fig. 5.27).

is visible on the central bow or on the footplate (type B2; see also Nicolay and Van Eerden 2021). The different forms of Domburg brooches have been found all along the Dutch coastal area and in the adjacent river area (Koch 1999a, fig. 5; Heeren and Van der Feijst 2017, fig. 8.36); despite the large number of ‘sub-types’, no clusters of specific forms in smaller regions within Early-medieval Frisia can be identified. Another variety of small long brooches is represented by a small number of specimens from Friesland (unpublished) and one find from the Dutch central river area (Heeren and Van der Feijst 2017, type 82i1, pl. 77), probably dating to the sixth century. They have the usual, rectangular headplate, however, with flat ‘knobs’ along the upper rim and side rims (Fig. 10.3: nos. 22–23). Although only fragmented brooches are known, the footplate probably had a lobed shape. On the central bow a disc is attached, similar to the Frisian-style gold and silver brooches discussed below. With one exception (Neresheim, Germany; Koch 1999b, fig. 4), direct parallels from surrounding countries are unknown, making this brooch another typically ‘Frisian’ ornament. 284

Z01 Hines Book B.indb 284

07/05/2021 13:30

Symbolism and Group Identities

Fig. 10.5  The stylistic development of Domburg brooches, showing a gradual ‘debasement’ of animal motifs. Not to scale (drawing by the author; typology largely after Botman 1994).

In the cemetery of Oosterbeintum a Domburg brooch was found in combination with an annular brooch (Tab. 1; Fig. 10.2: grave 428). At the cemeteries of Rijnsburg-De Horn and The Hague-Solleveld (South-Holland) a similar combination of brooch types can be seen (Fig. 10.4: nos. 3, 6). Annular brooches were worn from the mid-sixth to the late seventh century, although a wider date range cannot be excluded (e.g. Knol 1993, 67–8: c. AD 450–650/700). One of the annular brooches from Friesland (Fig. 10.3: no. 32) is showing clearly recognizable animal heads. Some of the other brooches have triangular remnants of such heads (Fig. 10.3: no. 33); most brooches, however, are only decorated with simple ribs, roundels and dots, or have no decoration at all (Fig. 10.3: nos. 34–37). The currently known finds from southern England (e.g. Hills et al. 1984) and the Dutch coastal area (Knol 1993, fig. 60) show remarkable similarities. Most of the ‘Frisian’ specimens, however, are more robust and decorated in different ways than those from England, which is in agreement with their relatively late date (see Palm and Pind 1992, fig. 3a). Besides these regional-style brooches showing a cultural orientation towards northern Germany and southern England, throughout the sixth century imported, Frankish-style bow brooches were also popular – including square-headed and especially radiate-headed brooches (Fig. 10.3: nos. 24–27; Fig. 10.4: no. 7; see Fig. 10.6). More numerous are figural brooches, with a dominance of bird brooches (Fig. 10.3: nos. 28–31). Their variety in form and decoration, however, may indicate that some bird brooches are local imitations. Interestingly, no single Frankish-style brooch was buried in combination with a cruciform, small long or annular brooch at the cemetery of Oosterbeintum (Tab. 1). While such ‘foreign’ brooches apparently were unproblematic when worn in daily life, they seem to have been inappropriate for the expression of (group) identity in the burial ritual. An exception is the Frankish-style bow brooch 285

Z01 Hines Book B.indb 285

07/05/2021 13:30

Frisians of the Early Middle Ages from The Hague-Solleveld (Tab. 1; Fig. 10.4: no. 7) and the Frankish-style garnet disc brooch from Rijnsburg-De Horn (Tab. 10.1; Fig. 10.4: no. 4), two cemeteries close to the Frankish realm. Mid-seventh to tenth centuries: towards supra-regional-style brooches During the seventh century, small equal-armed brooches appear as a new brooch form. The oldest form, with kidney-shaped terminals, is closely related to small long brooches of type Domburg (Bos 2006: type 1.1 and related type 1.2; Fig. 10.3: no. 38). Slightly younger are some types with a heavy bow, shaped like a wrapped toffee and usually decorated with ribs or grooves (Fig. 10.3: no. 39). Based on a context-dating from Wijnaldum-Tjitsma (Friesland), they must have been in use before the year 650 (Bos 2006, 460). At the cemeteries of Oosterbeintum no specimen was found in association with Domburg or annular brooches (Tab. 10.1), which points to a first appearance around the mid-seventh century. In the late seventh or early eighth century, the heavy specimens are replaced by a larger variety of mainly slighter brooch types (Fig. 10.3: nos. 40–43), including the well-known specimens of the ‘bow-tie type’ (Bos 2006: type 1.8.1.2; Fig. 10.3: no. 41). These brooches are relatively small again, and are usually decorated with simple groove patterns on both terminals. Small equal-armed brooches have been found across large parts of north-western Europe. Within their distribution area a distinction on the basis of their shape and decoration can be made between two regional groups: one in northern France/Belgium (Van Bellingen 1988; Thörle 2001) and a second in the Dutch coastal area. It is remarkable, however, that this distinction is also visible among finds from the western Netherlands (mainly the Domburg trading site; Capelle 1976) and the northern Netherlands (mainly Friesland; Bos 2006), the same area where similar types of the Domburg brooches were worn. Disc brooches were first introduced as round, engraved specimens during the sixth or seventh century (Fig. 10.3: nos. 44; Fig. 10.4: no. 5); after the mid-eighth century, these are replaced by a larger variety of shapes, ranging from round, flower-shaped, rectangular and moon-shaped to cruciform brooches (Bos 2008; see also Haseloff 1990, 91–107; Frick 1993; Fig. 10.3: nos. 45–53). Many forms are decorated with enamel, either as a small central ‘stone’ or as the fill of a cross, animal or saint’s motif. Contrary to small equal-armed brooches, different forms of disc brooches show an overlapping distribution across north-western Europe (Haseloff 1990, Karten 1–4; Frick 1993, Karten 1–17). Not a single regional type can be identified among this extensive and diverse group of ornaments. Both small equal-armed and disc brooches were worn into the tenth century. In the eleventh century, the variety of forms was probably limited to disc brooches with an image of the Lamb of God or Christ (Bos 2006, 773, types 2.7.2.2 and 2.7.4.3; Fig. 10.3: nos. 51–52) and especially pseudo-coin brooches, showing imitations of older solidi of Louis the Pious (Bos 2006, 755, type 2.6; Fig. 10.3: no. 50, and related forms). These brooches, of which some may have been worn into the twelfth century, were no longer made only of copper alloy but also of pewter, a lead-tin alloy. Figure 10.6 shows the chronological development of brooch types that were popular among the inhabitants of Early-medieval Frisia. After a population decline and in some parts of the northern Netherlands even depopulation during the fourth century, the arrival of newcomers is reflected in the appearance of ‘Saxon’ and ‘Anglian’ brooch 286

Z01 Hines Book B.indb 286

07/05/2021 13:30

Symbolism and Group Identities

Fig. 10.6  Chronology of copper-alloy brooches that have been found in the area of Earlymedieval Frisia, or are relevant for their dating. Datings after Rau (2010, fig. 35) or other publications mentioned in the text (drawing by the author).

types: supporting-arm brooches, equal-armed brooches and cruciform brooches. Throughout most of the fifth century, these ornaments referred to the newcomers’ homelands; at the same time a gradual development towards regional types can be seen, in the sixth and first half of the seventh century typical for different parts of the southern North Sea world. Within Early-medieval Frisia, the most distinctive types are Domburg brooches and disc-on-bow brooches, both found along the Dutch coast and also in the adjacent river area. From the mid-seventh century, new brooch types were introduced, as varieties of two forms that dominated the following centuries. During the eighth and ninth centuries, regional types can still be discerned among small equalarmed brooches; disc brooches of many different types, on the other hand, have been found across the wider area of north-western Europe. Despite the large variety of brooch forms and the sixth-century process of ‘regionalization’, co-resident identities were not expressed at the level of individual nuclear regions. Even Domburg brooches show no clustering of specific types at a truly regional level. According to Dijkstra (2011, 355; see also Dijkstra and De Koning 2017, 65–6), these ornaments were not meant to express differences between local or regional co-resident groups in an active way; instead, in a more passive way, they had an integrating function for the inhabitants of the Dutch coastal area as a whole. The distribution of Domburg brooches probably relates to the living areas of women who married both within and between ‘Frisian’ nuclear regions. Only few Domburg brooches have been found outside the Dutch coastal area (e.g. Koch 1999a), representing women who married outside this ‘Frisian’ network. After incorporation into the Frankish kingdom, the inhabitants of Frisia no longer expressed a uniform identity. Instead, a subdivision into 287

Z01 Hines Book B.indb 287

07/05/2021 13:30

Frisians of the Early Middle Ages smaller identity groups can be seen on the one hand (small equal-armed brooches), and into a larger, supra-regional identity group on the other (disc brooches).

Precious-metal ornaments and the expression of political identities For the production of gold and silver ornaments highly specialized craftsmen and access to precious metals, in minted or unminted form, were needed. While copper-alloy brooches probably circulated within or between families and were temporarily worn by women of specific age groups, ornaments of gold and silver relate to a different form of personal relations, between a leader and members of his retinue or Gefolgschaft (Bazelmans 1999; Nicolay 2014). In successive phases, the ornaments of gold and silver show cultural relations with the German Elbe-Weser region, Scandinavia, Anglo-Saxon England and the Frankish area (Nicolay 2017a). This way, ornaments of gold and silver not only reflect the top-down flow of prestigious items within elite networks, but also the ‘taste’ and cultural orientation of the ruling elite. It was this elite who decided what precious metals were used for the manufacture of a specific gift, how this gift should be shaped and how it was decorated. This way, coins or other raw materials from a leader’s treasure were transferred by gold or silver smiths into the ornaments that were desired. Late fourth to fifth centuries: Saxon-style brooches During the late fourth and fifth centuries, Late-Roman silver plate reached the North Sea area as tribute paid by the Roman government to Germanic leaders to assure their obedience. Fragmented plate could be used as a kind of currency, or as a raw material for the manufacture of regional-style silver ornaments (e.g. Rau 2013). Among the Saxon-style brooches that are typical for this period, the silver equal-armed or butterfly brooches with their Late-Roman-style, chip-carved decoration were most popular. The largest and most prestigious specimens are of the Dösemoor and Nesse types, made of gilded silver (Böhme 1974, 17–18; Bruns 2003, 16–19) and dating to c. AD 425–500 (Rau 2010, fig. 35). Silver equal-armed brooches, like their copper-alloy counterparts, have mainly been found in the northern part of the Elbe-Weser region (Nicolay 2014, fig. 9.4). Less pronounced find clusters, in the northern Netherlands and southeast England, relate to migrating ‘Anglo-Saxons’ again. Within the Dutch coastal area silver Saxon-style brooches are represented by a relatively small number of finds, all from Friesland: a supporting-arm brooch, three equal-armed brooches (Types Wehden and Dösemoor; Fig. 10.8: nos. 1–2) and two bronze saucer brooches with a silver front-plate (Types Jouswier and Lippspringe). Interestingly, no single Anglian-style cruciform brooch was executed in silver. Apparently, the leading elite chose to express a ‘Saxon’ instead of an ‘Anglian’ identity via the brooches they distributed as gifts. Sixth to first half of the seventh century: from Scandinavian-style to regional-style ornaments In the late fifth and first half of the sixth century, within the North Sea area a shift in the use of precious metal (from silver to gold) and in cultural focus (from the Late Roman Empire to Scandinavia) can be seen. The gold for making high-status ornaments was imported mainly via southern Scandinavia, as solidi of Eastern Roman origin. These coins were transformed into gold arm- and neck-rings, and above all gold bracteates of Scandinavian types (Fig. 10.7). 288

Z01 Hines Book B.indb 288

07/05/2021 13:30

Symbolism and Group Identities

Fig. 10.7  Scandinavian-type gold bracteates from the province of Friesland, found at Hitzum (no. 1: type A, settlement find), De Valom (no. 2: type B, secondary context/ replaced terp soil) and Achlum (no. 3: type D, hoard find). Scale 1:1 (Hitzum/Achlum: collection and photograph: Fries Museum, Leeuwarden; De Valom: private collection and photograph Streekmuseum Volkssterrenwacht, Burgum).

Within Scandinavia bracteates were in use from c. AD 440 to 530; most specimens from the southern North Sea area were executed in chip-carved relief and are relatively late (produced after c. 475; Axboe 1999; 2007, 146–8). Burial finds from the south-east of England indicate that bracteates were still worn and possibly also produced outside Scandinavia up to c. 570 (Axboe 2007, fig. 59). The neck and arm rings are more difficult to date, but most probably were contemporary with the Scandinavian-type bracteates. The currently known bracteates from the southern North Sea area show a remarkable distribution: they cluster in several small areas, separated by extensive empty zones (Nicolay 2014, 238–43, figs. 9.7–9). One of the clusters is evident in the area around Sievern (Elbe-Weser region), others in the northern part of Westergo (Friesland) and eastern Kent (England). Despite their close cultural links with southern Scandinavia, most bracteates from these areas show regional characteristics and were probably manufactured in regional workshops (e.g. Behr 2010, 65–71). Three of the bracteates from Westergo, for example, show human hands, a unique ‘Frisian’ feature that is unknown from any Scandinavian find (Fig. 10.7: no. 3). The same is true for the gold neck-rings of the so-called Mulsum type, only found in the Elbe-Weser region (Brieske 2001, 149–51). The sixth century saw a reversion from gold to silver ornaments and a shift in cultural focus back to the south, where the Frankish kingdom had gradually succeeded the Western Roman Empire. All along the southern North Sea coast the use of silver-gilt, Frankish-type brooches and silver or silver-inlaid belt-fittings is reflected in settlement and cemetery finds. Although such Frankish ornaments were also worn by people living in the Dutch coastal area, a substantial proportion must have been melted down to be transformed into regional-style brooches that in their shape and decoration were inspired by Scandinavian prototypes. In Friesland, silver-gilt brooches of the Achlum type were produced around the mid-sixth century (Nicolay 2014, 87–8; Fig. 10.8: nos. 3–6, Fig. 10.9: no. 2). Their overall shape, the small disc on the bow, the birds on the footplate and the decoration with niello were modelled on square-headed brooches of the ‘Jutlandic’, Scandinavian type (Haseloff 1981, 18–51). As a regional, ‘Frisian’ element, the kidney-shaped terminal of the footplate, also typical of Domburg brooches, is added. Across north-western Europe 289

Z01 Hines Book B.indb 289

07/05/2021 13:30

Frisians of the Early Middle Ages

Fig. 10.8  Saxon-style silver brooches (nos. 1–2) and regional-style silver-gilt brooches (nos. 3–6), all found at settlement sites in Friesland. Scale 2:3 (drawings by J. de Koning).

many copies and later imitations of ‘Jutlandic’ brooches have been found (Nielsen 2009). Thanks to a recent metal-detector find, such a prototype for the brooches of Type Achlum is also known from Friesland (Nicolay 2017b; Fig. 10.9: no. 1). Another metal-detector find is the bronze copy of an Achlum-type brooch, found on Wieringen, in the most northern part of North-Holland (Nicolay and Van Eerder, in prep.). The late sixth and early decades of the seventh century saw a further development of regional-style ornaments, not only in Friesland but in the wider coastal area of the northern Netherlands. Once more these ornaments are executed in gold, which had reached the research area for the largest part as Frankish solidi and tremisses now. The ‘Frisian’ jewellery is characterized by garnet- and filigree-decorated disc-on-bow brooches and several types of pendants, including some with elaborate, filigree-decorated borders (Nicolay 2014, 257–61). The most compelling examples are the disc-on-bow brooches from Wijnaldum, Wieuwerd and Hogebeintum (Fig. 10.9: nos. 4–6), belonging to the Hogebeintum type. They represent a further development of the Achlum-type brooch, showing the adoption of Frankish (garnet inlay) and Scandinavian (Style II animals) Fig. 10.9  (Facing) Selection of high-status ornaments of different types: silver-gilt brooches from the Holwerd area (Friesland, no. 1) and Achlum (Friesland, no. 2); gilt copper-alloy brooch from Hallum (Friesland, no. 3); gold (and silver) brooches from Wijnaldum, Wieuwerd and Hogebeintum (Friesland, nos. 4–6); gold-decorated, copperalloy belt-buckle from Rijnsburg (South-Holland, no. 7); gilt copper-alloy terminal of a drinking horn from Katwijk (South-Holland, no. 8); gold pseudo-coin brooches from Witmarsum and Noordbergum (Friesland, nos. 9–10) and gold finger rings from Tytsjerk and Oostermeer (Friesland, nos. 11–12). Scale 1:2 (nos. 1–8) and 1:1 (nos. 9–12) (Holwerd: private collection, photograph F. de Vries; Rijnsburg/Katwijk: collection and photographs Rijksmuseum van Oudheden, Leiden (inventory nos. h 1913/11.88 and h 1921/1.21); Hallum: collection Noordelijk Archeologisch Depot, Nuis, photograph ARCbv; Noordbergum: collection and photograph Streekmuseum Volkssterrenwacht, Burgum; Tytsjerk/ Oostermeer: private collections, photographs De Detector Amateur; others: collection and photographs Fries Museum, Leeuwarden).

290

Z01 Hines Book B.indb 290

07/05/2021 13:30

Z01 Hines Book B.indb 291

07/05/2021 13:30

Frisians of the Early Middle Ages decoration techniques. At the same time they reflect close cultural ties with AngloSaxon England, especially with Kent (Mazzo Karras 1985). In contrast to the northern Netherlands, no gold jewellery or other ornaments point to the development of a regional style in the western Netherlands. This area shows a cultural focus on two of its neighbouring areas now: some pendants with filigree-decorated borders, found in the province of North-Holland, are typical of the northern Netherlands (Nicolay 2014, 115–16), while a shield-on-tongue buckle from Rijnsburg (Fig. 10.9: no. 7), the terminal mount of a drinking horn from Katwijk (Fig. 10.9: no. 8) and a hanging-bowl from Oegstgeest, found in South-Holland, show a cultural link to East Anglia and Kent (Dijkstra 2011, 232–3, 243–4; Van der Meer, in prep.). Mid-seventh to tenth centuries: towards Frankish-style ornaments The following seventh century is characterized by gold and silver jewellery that is showing truly exotic, Byzantine influences (Geake 1997). The jewellery is characterized by finger rings (silver), linked pins (gold/silver), and most notably necklaces with gold and silver pendants and beads, as well as amethyst beads. Most of these seventh-century gold and silver ornaments are known from Anglo-Saxon England, but continental finds show that elements of this Anglo-Saxon fashion occurred in the wider North Sea world (Nicolay 2014, 261–3). Because of the more-or-less uniform character of such ornaments, it is no longer possible to trace individual elite networks. These networks fully disappear after their incorporation into the Frankish kingdom. Already during the Saxon Wars of Charlemagne, a gradual imbedding of ‘Frisian territory’ into the Frankish political structure can be seen. Among both the old and new pro-Frankish elite families competition for power is reflected in the eighth- and ninth-century burial ritual, resulting in a horizon of remarkably rich weapon graves in the Dutch coastal area and further east, in Ostfriesland (Knol and Bardet 1999). After gold coins had been replaced by silver denarii and sceattas in the late seventh century, Louis the Pious minted solidi for a few years again after he became king (Grierson 1951). These coins were imitated on a large scale, especially in the area of Early-medieval Frisia (Grierson 1951; Boersma 1977). Both original solidi and local imitations were worn as brooches, after a rim of gold beaded wires was added (Nicolay 2008; Fig. 10.9: nos. 9–10). Most probably, both (imitation) coins and coin as well as pseudo-coin brooches were distributed as gifts among local and regional groups of followers, this way symbolizing the new status of a pro-Frankish elite. A similar function can be assumed for silver small equal-armed brooches, partly decorated with niello (Bos 2006, e.g. types 1.6.1.1 and 1.6.2.1–2), and the gold finger rings mentioned below. Because of a stabilization in power structure after the ninth century, competition among members of the ‘Frisian-Frankish’ elite shifted to a different ritual arena (church building and donations) and hardly any high-status ornaments are found in the Dutch coastal area anymore. One of the few exceptions is the gold and silver hoard from Marsum (Groningen), buried in the late ninth century (Knol 2004). It contains a silver small equal-armed brooch, silver belt-fittings, imitation solidi and Frankish denarii, as symbols of the Frankish cultural world. The leading families among the ‘Anglo-Saxon’ settlers in the northern Netherlands continued to produce silver brooches that referred to their homeland, with a dominance of Late-Roman-style, ‘Saxon’ symbols. After the fall of the Western Roman Empire, this 292

Z01 Hines Book B.indb 292

07/05/2021 13:30

Symbolism and Group Identities elite turned to southern Scandinavia for inspiration: their status symbols are imitations of ‘northern’ prestige goods, including gold bracteates. In the mid-sixth century, the imitation of silver square-headed brooches, again of a Scandinavian origin, led to the development of regional-style brooches of Type Achlum, which were succeeded by those of Type Hogebeintum. The distribution of such identity markers as gifts in an active way reveals the size of regional and larger ‘Frisian’ elite networks during the sixth and early seventh centuries (Nicolay 2014, 353–9, figs. 13.2–3). At their greatest extent, these networks covered the northern Netherlands and the northern part of North-Holland, and the western Netherlands, respectively; northern Westergo, and possibly Wijnaldum, was the centre of the ‘northern kingdom’, and the Rhine mouth area the centre of the ‘western kingdom’. The step-wise conquest of both kingdoms by Charles Martel and Charlemagne, in 719 (western kingdom, up to the river Vlie), 734 (northern kingdom, up to the river Lauwers) and 804 (northern kingdom, eastern part), is strong proof of their status as independent elite networks (Nicolay 2014, 362, fig. 2.3). The fact that the northern kingdom was conquered in two steps further indicates that the sixth-century regional kingdoms may have kept their independence throughout the seventh century, perhaps as part of larger confederations, headed by an overlord. Already before the ‘Frisian’ regional and supra-regional kingdoms had been conquered by the Franks, regional-style ornaments were replaced by uniform, Byzantine-style ornaments. The actual incorporation of Frisian territory, between 719 and 804, led to competition for power both at the local and regional level. Investment in weapon graves and the distribution of (imitation) solidi, as well as gold and silver brooches, gave pro-Frankish families the opportunity to gain power, (partly) at the expense of the older, sixth- and seventh-century elite. In the ninth century, both the old and new elites had adopted Frankish status symbols, which showed their relationship to a new overlord: the Frankish king.

Animal decoration and the expression of religio-ideological identities Both copper-alloy brooches and more prestigious, gold and silver ornaments are richly adorned with decorative and more symbolic motifs. As symbols of family groups and elite networks such motifs referred to co-resident and political identities. On a larger geographical scale, a third level of group identities can be identified by the use of pre-Christian and Christian decoration motifs on copper-alloy and more prestigious ornaments. Although the iconographic study of Early-medieval artefacts has received limited attention in Dutch archaeology, an attempt will be made to trace the religious-ideological orientation of the ‘Frisian’ coastal population and its ruling elites. Late fourth to fifth centuries: a Roman-Germanic ideology Inspired by Late-Roman belt-fittings, the Saxon-style copper-alloy and silver equalarmed brooches show geometrical, floral and animal motifs. Within the Elbe-Weser region a regional distinction can be made between brooches without ‘bordering animals’ and those with such animals – respectively west and east of the River Oste (Genrich 1951, fig. 30). The animals on brooches from the eastern zone are imitations of similar animals on Late-Roman belt-fittings, like images of lions, sea creatures and griffins (Haseloff 1973). These Late-Roman symbols were copied with the purpose of 293

Z01 Hines Book B.indb 293

07/05/2021 13:30

Frisians of the Early Middle Ages demonstrating a regional identity that was built on cultural relations with the prestigious Roman Empire. Interestingly, a similar development can be seen in Scandinavia, with the development of the fifth-century Nydam Style (Haseloff 1981, 8–17). While the Saxon-style brooches closely followed Late-Roman prototypes and fit into a wider ‘Roman-Germanic ideology’, within Scandinavia the development towards more complex animal and also man-between-animals motifs is reflecting a different, Nordic worldview. Sixth to first half of the seventh century: a Scandinavian ideology The Nydam Style was succeeded by Animal Style I around AD 460 (Rau 2010, 297–301). This style is typical for silver ‘Jutlandic’ square-headed brooches and gold bracteates. From the Dutch coastal area one imitation of a ‘Jutlandic’ brooch is known, dating to the late fifth or first half of the sixth century (Nicolay 2017b). This silver-gilt brooch, from the vicinity of Holwerd (Friesland; Fig. 10.9: no. 1), shows the standard decorative scheme: a human head (Fig. 10.10) between two quadrupeds on the headplate, a second human head (Fig. 10.10) on the bow, and birds’ heads along the shoulders of the footplate. The dominant theme is the man-between-animals, both on the headplate and on the bow/upper part of the footplate. From the province of Friesland a total of twelve gold bracteates are known, dating to c. AD 475–550 (Nicolay 2017b, 506–7). One specimen, of type A (Fig. 10.7: no. 1), features the head of a bearded man, wearing a richly decorated diadem; on both sides of the head a runic inscription with unknown meaning is visible. Three further bracteates, of type C (Fig. 10.7: no. 2), show a similar human head, above what seems to be a horse. On one of the bracteates the triangle in front of the headdress terminates in a bird’s head with a curved beak. Eight more bracteates, of type D, feature the ribbon-like body of an animal with teardrop-shaped hips. The animals have two different types of head: horse- or wolf-like heads with an open mouth (Fig. 10.7: no. 3), or bird-like heads with

Fig. 10.10  Religious motifs on ‘Frisian’ ornaments: two human heads, probably of Wodan/ Odin, on the Holwerd brooch (see Fig. 10.9: no. 1), and the face of Christ on a copper-alloy disc from Rauwerd (Friesland, scale 2:1) (Holwerd: private collection, photograph F. de Vries; Rauwerd: collection and photograph Noordelijk Archeologisch Depot, Nuis).

294

Z01 Hines Book B.indb 294

07/05/2021 13:30

Symbolism and Group Identities a pointed, curved beak. On two of the bracteates from the first group human feet and human hands are attached to the animal’s hips; two bracteates from the other group show a single human ear. After the mid-sixth century, Scandinavian-style man-and-animal motifs keep occurring, now on regional-style, ‘Frisian’ jewellery. The silver-gilt brooches of Type Achlum have a similar shape to the ‘Jutlandic’ prototypes, with the exception of the kidney-shaped terminal (Fig. 10.9: no. 2). The headplate is showing the man-between-animals again; the S-shaped hair on both sides of the head at the same time represents highly stylized animals. On the disc, no second human head is depicted; the bird-shaped animals on the upper part of the footplate are flanking a plain roundel. A further debased, gilded bronze copy of a ‘Jutlandic’ brooch was excavated at Hallum (Friesland; Fig. 10.9: no. 3). Although the bird-like animals on the footplate are still recognizable, the human head and animals on the headplate are reduced to simple geometrical shapes. This debasement of animal motifs can also be seen on Domburg brooches: the oldest specimen shows two typical Style I birds’ heads, which on the later brooches are reduced to bean-shaped or point-circle motifs (Fig. 10.5; see also Nicolay and Van Eerden 2021). The disc-on-bow brooches of Type Hogebeintum, produced in the late sixth and early decades of the seventh century, represent a final phase of imitating Scandinavian-style brooches and their symbolic language, now executed in Animal Style II. Inspired by Frankish jewellery, the front of the Wijnaldum brooch is fully covered with garnet inlays (Fig. 10.9: no. 4). Although the individual decorative motifs are difficult to recognize, the headplate shows the familiar human head between two quadrupeds. Two snake-like animals, with the mouth open, are seen on the upper part of the footplate. When the curved legs of both animals indeed form the eyes, moustache and beard of a second male face, the man-between-animals motif is depicted twice again. Two closely related brooches, found at Wieuwerd and Hogebeintum (Friesland; Fig.9: nos. 5–6), show more debased motifs: a highly stylized animal with an intertwined body and curved beak (Wieuwerd) and simple interlaced patterns (Hogebeintum), both in filigree. An interlaced animal design similar to that on the Wieuwerd brooch also adorns contemporary pendants, including those with bird-like animal heads that are identical in shape to the terminal knob of Achlum-type brooches (Nicolay 2014, figs. 4.10, 4.24). More naturalistic is the depiction of four horned animal heads on the triangular sides of a sword pyramid from Ezinge (Groningen; Nicolay 2014, fig. 4.37). In the western Netherlands, the buckle from Rijnsburg and the terminal of a drinking horn from Katwijk show closely related motifs in Animal Style II (Fig. 10.9: nos. 7–8). The symbolic meaning of the Scandinavian-type square-headed brooches, including those from Achlum and Wijnaldum, was studied by Olsen (2006). A central theme on both brooches is the human head flanked by animals. The interpretation of this theme is based on Old Norse literature of pre-Christian mythology: the male person is seen as Odin, head of the Norse pantheon and king of the Æsir gods, who is flanked by his helping spirits, embodied as horses, wolves or birds. These animals had the capacity not only to prophesy, but also to enter the world of the dead – so that Odin himself could enter the Otherworld. This association of the male heads on the Scandinavian-type ornaments with Odin (or his Continental and Insular counterpart Wodan) is supported by other research (e.g. Hedeager 2011), also for the southern North Sea area (e.g. Gaimster 2011; Nicolay 2017b; Nicolay and Van Eerden 2021). 295

Z01 Hines Book B.indb 295

07/05/2021 13:30

Frisians of the Early Middle Ages The reference to Scandinavian mythological scenes, which probably visualized elements of (unknown) ‘Frisian’ origin myths, will have been most relevant among members of the North Sea elite, as distributers of gold and silver ornaments that feature such scenes. It is uncertain to what extent the symbolic language was fully understood by people of lower social status. The Domburg brooches are an attempt to copy more prestigious, silver brooches, but the Odin-message had become less obvious when birds’ heads gradually degenerated into stylistic motifs. On the gilded, Scandinavian-style brooch from Hallum, simple geometrical forms are all that remain of the original motifs. Despite such a process of degeneration, the basic shape of Domburg brooches remained unchanged and for the observer may have referred to Odin/Wodan and his helping spirits – especially when it is assumed that mythological and origin stories were told in the halls both of the (supra-)regional and local elites. Mid-seventh to tenth centuries: towards a Christian ideology Elements of the ‘Nordic’ belief system were also known further south, in the Frankish world. Copies and imitations of ‘Jutlandic’ brooches, showing the man-between-animals motif, have been found across northern Gaul (Roth 1996, fig. 502). Moreover, some of the younger, late sixth- and seventh-century Frankish objects show a fascinating mixture of Christian and Nordic motifs. An example is the seventh-century gold disc from Limons, in central France (Wamers 2013): the disc is decorated with obvious Christian symbols on the one hand (Christogram, Christ’s head, Alpha/Omega and Christus Rex), and complex Nordic symbols on the other (snake-like creatures and stylized human/animal heads). This combination of religious-ideological motifs is typical for this period of changing belief systems, in which objects with a Nordic-Christian symbolism ‘have the capacity to refer to both old and new worldviews’ (Kars 2018, 142; see also Wamers 2009). In Early-medieval Frisia, the transition to a Christian symbolic language went hand in hand with Frankish occupation. In the late seventh century, disc-on-bow and Domburg brooches were no longer produced, and were replaced by small equal-armed and disc brooches. Small equal-armed brooches, decorated with one to three roundels, start featuring Christian motifs (Fig. 10.4: nos. 42–43), probably not before the late seventh or early eighth century (Thörle 2001, Taf. 60): each roundel shows an equal-armed cross, sometimes, on silver specimens, inlaid with niello. When four dots are added, these may symbolize the four Evangelists (Bos 2008, 729). The cross is a more common feature on disc brooches. After the first crosses appear on early, sixth- and seventh-century specimens (Fig. 10.3: no. 44; Fig. 10.4: nos. 4–5), this is the most common motif on mid-eighth- to tenth-century disc brooches. The cross is depicted in different forms, partly filled with white or coloured enamel (Fig. 10.3: nos. 47–48). Other Christian motifs are depictions of a saint in coloured enamel, in some cases holding a raised cross (Fig. 10.3: no. 49). Finally, some brooches feature the Lamb of God (Fig. 10.3: no. 51), or the face of Christ (Fig. 10.3: no. 52). The large number of disc brooches, their very specific symbolic language and their uniform composition suggest a centralized production in Christian workshops, like abbeys (Roxburgh et al. 2016, 14–17), and a distribution as badges of the new, Christian belief (Kars and Heeren 2018, 25). Cross-shaped and crucifix pendants start appearing in the ninth century. A copper-alloy, cross-shaped pendant, inlaid with enamel, was buried in a ninth-century 296

Z01 Hines Book B.indb 296

07/05/2021 13:30

Symbolism and Group Identities grave at Beers (Friesland), together with a silver denarius with a Christian cross, minted under Louis the Pious (Knol 1993, 216, fig. 71). One of the earliest depictions of the face of Christ can be seen on a probably ninth-century copper-alloy disc, found underneath the floor of the present-day Laurentius church at Rauwerd (Friesland; Fig. 10.10). More prestigious are gold finger rings, also dating to the ninth century, which feature the Lamb of God or the four Evangelists around this lamb (Postma 2015; Fig. 10.9: nos. 11–12). Interestingly, alongside these Christian symbols imported or locally imitated, Nordic metalwork is found, which can be related to Viking attacks and (temporary) Scandinavian settlers (e.g. Besteman 2004). The growing body of ‘Viking finds’ from the Dutch coastal area is showing a renewed appearance of animal-style motifs, probably referring to Scandinavian mythological stories especially in the ninth and tenth century again – as can be seen on brooches, cloth-pins and strap mounts of horse gear (IJssennagger 2017, 176–242). In contrast to the earliest expressions of Christianity in more southern regions, no find from Early-medieval Frisia shows a mixture of Christian and Nordic symbols. Moreover, no clear animal-style motifs are shown on metalwork after the mid-seventh century anymore, in contrast, for example, to small equal-armed brooches from more southern regions (Thörle 2001, e.g. types IX B 1 and X B 1). It can be questioned, however, to what extent this abrupt transition to a Christian symbolism meant that the inhabitants of the Frisian province smoothly embraced Christianity as part of a new religious-ideological identity. According to Milis (2005), several phases in the spread of Christianity across the Low Countries can be discerned: after pagan elements of collective and individual behaviour had been forbidden and gradually made place for Christian behaviour (including the wearing of Christian symbols), a phase of ‘in-depth conversion’ was aimed at the individual’s mind and heart (cf. also Hines, this vol.). This final phase probably was not reached before the eleventh or twelfth century, when churches were systematically built in the northern Netherlands (De Langen and Mol 2017: first half 11th century) and pagan worship was generally considered heresy (Mostert 1993, 132–3: early 12th century). During the period of the ‘Anglo-Saxon’ migration, the new inhabitants of Early-medieval Frisia expressed a ‘Roman-Germanic ideology’ in the decoration of Saxon-style metalwork. The late fifth century saw the adoption and imitation of Scandinavian-type ornaments, including their Nordic mythological scenes. These ornaments were not simply copied, however, but their iconography was applied to regional-style brooches (Types Achlum and Hogebeintum) and even transformed into new motifs (type D-bracteates with human hands). There can be little doubt that related ideas about Odin/Wodan and the other Æsir gods were understood by the sixth- and seventh-century inhabitants of the Dutch coastal area, probably as part of a newly created, ‘Scandinavian-Frisian ideology’. Despite the general acceptance of Christian symbols since the eighth century, traditional values were not easily replaced: the worship of gods like Wodan and Donar still had to be abandoned by pronouncing the ‘Utrecht Baptismal Vow’ in the late eighth century, the first Christian personal names did not appear before the tenth century, and some of the Old Norse gods still survive in our naming of the days of the week – including Woensdag (Wednesday), or ‘Odin/Wodan’s day’ (Mostert 1993). 297

Z01 Hines Book B.indb 297

07/05/2021 13:30

Frisians of the Early Middle Ages

Conclusion According to Lex Frisionum, Early-medieval Frisia was an extensive coastal zone, bordered by the rivers Sincfal and Weser. To what extent the people living in this area expressed a feeling of belonging together throughout the fifth to tenth centuries is investigated by looking at ‘Frisian’ works of art, ranging from simple brooches to more exclusive pieces of gold and silver jewellery. Copper-alloy brooches, especially during the sixth and seventh centuries, circulated within and between the ‘Frisian’ nuclear regions, in a passive way reflecting a co-resident identity among the inhabitants of Early-medieval Frisia. In an active way, the size of contemporary elite networks was visualized to outsiders by the distribution of regional-style gold and silver jewellery among a leader’s following. Each elite network, in a deliberately chosen, active way again, was part of a larger religious-ideological world, first related to the Old Norse belief system in the Scandinavian north, later to Christianity in the Frankish south. Frisia, as it is referred to in Lex Frisionum, clearly had its cultural, socio-political and ideological roots in the Migration and Merovingian periods – not as a single entity, but as a collection of families, elite networks and ideological communities that felt related on the one hand and were diverse on the other.

Acknowledgements The writing of this article would not have been possible without the systematic inventory of Early-medieval brooches by my former colleague Jurjen Bos (†), the first inventory of this kind in the Netherlands. Moreover, Jan de Koning (Hollandia, Zaandijk) was so kind as to share his drawings of these brooches with me, as the basis for some of the figures in this paper. Earlier discussions about the views presented in this paper, with Annet Nieuwhof (University of Groningen), Rob van Eerden (province North-Holland), Gilles de Langen (province Friesland/University of Groningen/Fryske Akademy), Menno Dijkstra (University of Amsterdam), Mirjam Kars, Stijn Heeren (both Free University, Amsterdam), Andreas Rau (Centre for Baltic and Scandinavian Archaeology, Schleswig) and others, were indispensable during the writing process. The data from Portable Antiquities of the Netherlands can be consulted on www.portable-antiquities.nl

Discussion VERSLOOT  You showed distribution maps of various types of copper-alloy brooches, including cruciform and small long brooches, from the fifth to seventh centuries. Do you have these maps for all the different artefact types that can be dated to this period? NICOLAY  I often have to use older maps. For most brooch types, for example, there are no up-to-date maps available which include all the most recent metal-detector finds. I think such distribution maps are very important, especially when clusters of specific object-types are seen as a reflection of (supra-)regional identity groups and social networks.

298

Z01 Hines Book B.indb 298

07/05/2021 13:30

Symbolism and Group Identities VERSLOOT  I was interested to hear your opinion about Anglian and Saxon brooch types. How do they appear in your data? Is it only a difference in geographical variation, or is it chronological too? NICOLAY  Both types are contemporary. They have their specific geographical distribution in the assumed homeland of the ‘Angles’ and ‘Saxons’, respectively in Schleswig-Holstein and the Elbe-Weser region. HINES  My immediate reaction to these maps is to observe that the difference that you see between a unified material culture at one level and clear regional differences at another level would be reproduced several times over with comparable maps on the other side of the North Sea. A big change in England before the seventh century is the abandonment of older regional female costume styles, which can be labelled ‘Anglian’/‘Saxon’/‘Kentish-Isle of Wight’, etc. A much more uniform style appears, although partly because of the actual size of England and residual polarities there are still differences: e.g. the very south coast and Northumbria remain different, but the continuum in between seems to be unbroken. At the same time, however, the symbolism of elite archaeology remains quite distinct: this has been debated greatly in relation to the Staffordshire Hoard. We know what the kingdom structure was by this time. The situation is highly analogous, it would appear to me. NICOLAY  Absolutely. Interestingly, however, the great square-headed brooches cut across the assumed boundaries of several sixth-century kingdoms in the east. HINES  Although there are quite a number of individual groups of these brooches that have quite narrowly defined, discrete local territories. That can be a single valley system. NICOLAY  The Domburg brooches are interesting in this respect. They are only found in the Dutch coastal area, across the area of several kingdoms that can be reconstructed on the basis of gold ornaments, especially. The only exceptions are women who probably moved out through marriage. At a level that crosses socialpolitical boundaries, I see the Dutch coastal area as a very distinct area. In this case we can talk about unity and group identity, historically related to the area called Frisia. HINES  An interesting point to stress in addition to this is the fact that the society as a whole is using its women and the way that they are dressed to declare the group’s presence. IJSSENNAGGER-VAN DER PLUIJM  That continues right into the Viking Age, as does the areas where it happens. I would like to pick your brain on finds from AD 700–1000. You mention that these finds show a more Carolingian identity, related to Christianity. It is a tricky point to make but of course this is based on iconography. Two of the finds you showed were gold finger rings, and they are very different objects than the brooches you showed otherwise. In my view perhaps the rings represent a different layer of ideas or networks and different use. How do you see them, and in particular how do you see them in relation to the brooches? NICOLAY  I see the gold finger rings, which are decorated with clearly Christian symbols, like the four Evangelists, as worn by the elite, and related to the incorporation of ‘Frisian’ territory into the Carolingian realm. IJSSENNAGGER-VAN DER PLUIJM  How do you perceive this elite, then? Do you see it as a network? Or is it a ruling elite, or perhaps a religious elite? 299

Z01 Hines Book B.indb 299

07/05/2021 13:30

Frisians of the Early Middle Ages NICOLAY  I find this difficult to say. The different parts of Early-medieval Frisia from the early eighth century on were gradually incorporated into the Carolingian realm. The question is what happened to the old elite. Did they fade away to be replaced? What is very interesting is that in the eighth and ninth centuries you see a renewed use of gold and silver, as highly symbolic, Frankish-type tools in a phase of social competition, to consolidate old and/or to create new positions. NIJDAM  This seems to be a bit early – influence of Christianity from about 700. NICOLAY  From about the middle of the eighth century you see a transition in brooch forms, from the earlier bow brooches, including Domburg brooches, to disc brooches which depict saints and especially cross motifs. One of the brooches with a saint motif, found during the excavations at Wijnaldum, was discovered in a context dating to the second half of the eighth century, showing that Christian motifs were in use at least from that time onwards. The exact dating of other key finds, including crucifixes, is not before the ninth century. IJSSENNAGGER-VAN DER PLUIJM  It could be later, as the gold finger rings I mentioned are from the ninth century. NICOLAY  Indeed, these probably cannot be dated before the ninth century. When we look at the earlier introduction of disc brooches with Christian motifs, there still seems to be a gap of about half a century between the disappearance of the regional-style brooches with animal motifs that seem to refer to a northern ideological world (Domburg brooches, in use until the late seventh century) and the appearance of a clearly Christian symbolism. NIJDAM  The way you presented AD 500–700, with its Scandinavian influence, is that these people were looking to Scandinavia. I had understood that this population brought this material with it from the north. NICOLAY  Some time ago I published a paper talking about Jutish immigrants bringing in gold bracteates (Nicolay 2005). That is not my view now. The gold bracteates I presented in this paper show clear links with Denmark and especially Jutland, but – like most of the bracteates from England – were probably not brought here by immigrants. Small details, like the unique human hands on some of the bracteates from Friesland, are proof that these were produced locally. NIJDAM  I was still navigating on the older views! NICOLAY  The transition point for me was looking at the Kentish finds, where very few bracteates are imported and most are locally produced. Scandinavia in the late fifth century had become a new ‘power block’ to look at when newly established elites in the southern North Sea area were searching for power symbols after the fading of Roman power in the West. NIJDAM  The step I want to take is to question what this actually meant. Did they adhere to Scandinavian mythology? We have nothing about Germanic gods in Frisia, except for our days of the week (e.g. ‘Woden’s day’). Is it really Odin/ Woden who is depicted on some bracteates with a head motif and on some of the brooches, and did they ‘believe’ in Woden in some way? I would love it to be true, but I have to be sceptical. NICOLAY  I am positive about the evidence that is available now, but this is hard to prove. The historical evidence is thin. The Utrecht source mentions Saxnot and Woden. Fosite can be linked in too. I think we should dare to read the symbolism of these objects as real, not merely reproduced from Scandinavian prototypes. 300

Z01 Hines Book B.indb 300

07/05/2021 13:30

Symbolism and Group Identities Importantly, man-and-animal symbols are not simply copied and presented on imitated ornaments, but depicted in a structured way on the headplate and both the shoulders and terminal of the footplate of regional-style, ‘Frisian’ Domburg brooches. This implies that such symbols, which can be linked to Odin/Woden, were known, understood and incorporated in the regional repertoire of identity markers. NIJDAM  The person between two animals is a very common motif, for example on the Sutton Hoo purselid. It can be Daniel in the lions’ den, a motif that can easily be re-used. NICOLAY  Absolutely. But recent discoveries offer a wider perspective, which for me supports the idea of Woden, who in a period with a clear cultural focus on Scandinavia was presented as an Odin-like deity. There is an example of a Christian buckle-tongue with one eye in the depicted face of Christ removed (Elsfleth, Germany; Mückenberger 2013), which seems to be Odinic. Price and Mortimer looked at the Sutton Hoo helmet, which has wafered gold foil only behind the garnet stones of one eyebrow; when the helmet was worn inside the hall, by the fire, the king would present himself as the one-eyed Woden (Price and Mortimer 2014). NIJDAM  Parts of Hauck’s work have been strongly criticized. Jan de Vries describes how the figure of Woden evolved from the Migration Period to the Viking Period, when he became the supreme god and ‘allfather’ (De Vries 1957, 27–106). It’s a matter of which Woden are we talking about. NICOLAY  That does indeed make it very complicated. In this case it’s not the name of the god that matters but rather the visual language that can be read on brooches and bracteates, and the relationship between the depicted male head and his flanking animals. I find it difficult to dismiss Hauck’s ideas, although his publications are barely readable. HINES  We mustn’t get bogged down with the specific problem of Woden; the essence of the problem with Hauck in my view is that his work was so idiosyncratic. In a way he created modern cults of belief; even an academic Odinic cult of a sort. There are pro- and anti-Hauck schools of thought. My own position has moved over the years from relative hostility to most of his ideas to a greater willingness to emphasize that certain observations he made simply cannot be dismissed and are indeed important. The key point I would make about the character of a god like Woden is that we must realize that it will have been an evolving figure, and one that changed a great deal, whereas Hauck appeared to assume the god existed in some complete perfect form early on and is only decaying in the later representations of him. Thus even the motif of being one-eyed could have originated separately and have been merged into an idea of Odin/Woden rather than needing to be Odinic from the start. NICOLAY  In our region we have small additions to the D-bracteates, e.g. single human features. The depiction of a snake-like animal with human elements is interpreted by Olsen (2000, 118) as an image of Odin/Wodan, transforming from human to animal, this way being able to enter the Otherworld. I think this is a very interesting idea, that I tend to agree with. HINES  I have always been inclined to see the iconography as moving towards a more social expression, become secular and almost heraldic. It marks social status and 301

Z01 Hines Book B.indb 301

07/05/2021 13:30

Frisians of the Early Middle Ages connections rather than a story. The idea of the inclusion of human elements on D-bracteates does not work well empirically with the English examples. NICOLAY  Something different: John, do you agree with the chronological scheme I presented today? HINES  Yes, I think it’s essentially entirely reliable. If anything, we are definitely seeing things drifting earlier, so that ‘circa 600’ as a turnover point might move back towards the mid-sixth century. But that also assumes we can transfer what we might be confident about in England to the situation in the Netherlands and around the North Sea. Even regionally in England things are not fully coordinated. FLIERMAN A comment on Christian symbolism turning up ‘too early’ in supposedly pre-Christian areas (I think c. 700 was mentioned somewhere in this discussion). From a historical perspective, such an early chronology would not be that problematic. The written material shows that the boundaries between the Christian and the pagan world were blurred in this period: there were many nominally Christian areas – Francia, Bavaria, Thuringia – in which ‘pagan’ practices continued to be encountered (though we should take into account that for hardliners like Willibrord and Boniface, any practice that fell outside their own rigid Christian norms could be labelled ‘pagan’). The religious situation in pagan areas like Frisia and Saxony was similarly complex, especially in border regions. A recurring feature of hagiographical accounts about missionaries in eighthcentury Frisia and Saxony is the family that is already Christian or leaning towards Christianity, and in this capacity comes to the missionary’s aid. In short, a confused chronology – if this is indeed what the material record amounts to – wouldn’t be all that problematic in light of the textual evidence. NICOLAY  The datings definitely show the mid-eighth to early ninth century as the starting point for Christian elements. Further south, into central Gaul, brooches and other ornaments show Christian symbols alongside what I regard as Nordic pagan motifs from the late seventh century onwards. That sort of mixture of preChristian and Christian motifs on the same object is interestingly absent in our region. FLIERMAN On an abstract level, I understand that these objects can create identity and can be used as part of a, so to speak, ‘social strategy’. But how does this work in practice? Were people wearing these objects? And if so, were they doing so all the time or only on a special occasion? NICOLAY  It’s important to make a distinction between ornaments of copper-alloy and those of more precious metals. The former were probably distributed in and between families, and worn daily; the latter are top-down gifts in elite networks that may have been worn less frequently and mainly on special occasions. The latter were meant to visualize symbolic boundaries between elite networks, while the former were exchanged (with wives) much more freely across these boundaries. FLIERMAN Just to make sure I understand you correctly: ‘passive’ identification in your view means people signalling an identity through material culture without them consciously setting out to do so, and ‘active’ identification refers to people, mainly elites, using material objects strategically to claim an identity (social, ethnic?) and establish networks of allegiance? NICOLAY  It’s not clear if wearing a Domburg brooch expresses a ‘Frisian’ identity, or, more probably, the identity of a coastal population for which the wearing of 302

Z01 Hines Book B.indb 302

07/05/2021 13:30

Symbolism and Group Identities such a brooch simply expressed co-resident relations. The Wijnaldum brooch, on the other hand, expressed and represented an elite network in a much more active way. The construction and awareness of identities at different levels in society are things we need to think about. VERSLOOT  A very specific question on the distribution maps of Anglo-Saxon ornaments, relating to c. AD 600/550–650. The inland, westward depth of their distribution, as you presented it, isn’t extensive. NICOLAY  I have plotted the gold and silver objects which I think were produced in regional styles along the southern North Sea coasts, trying to visualize certain elite networks in southern England and also in the Dutch coastal areas. VERSLOOT  If this served as a proxy for the spread of speakers of Germanic languages in southern England, it implies that their spread westward was very limited. HINES  But there are very many more Anglo-Saxon finds which can be mapped that go much further inland. There is a very clear boundary line between east and west, but it is much further west in England than the coastal area covered by Johan’s research. NICOLAY  It is interesting now that we have a range of overlapping maps for the late fourth to ninth/tenth centuries which need to be interpreted comparatively, to reconstruct the extent and geographical development of co-resident and elite groups in the southern North Sea area. LOOIJENGA  In your paper you wrote about brooches and the idea that these were handed down within families and expressed relationship. These were worn by women, and end up in the grave, perhaps with the last woman of a line. Did you ever look at the brooches from Alemannia and Bavaria which have inscriptions on them, from the late sixth/early seventh century? Most of these are names, and terms of endearment. This might fit with the interpretation you have. NICOLAY  No, I haven’t looked at those. The idea of the family networks is by no means my own. It’s related to grave finds, and often to women of specific age groups. Younger and very old women tend not to have these brooches. LOOIJENGA  This is more of a Tracht, and younger girls wore different costumes. KNOL  I have a remark on the paper. You tried to compare the distribution of copper-alloy brooches with distributions in graves. It’s important to note that we have a limited number of cemeteries. We rely mainly on Hogebeintum and Oosterbeintum. Little else is available. We have to be careful with this. NICOLAY  I totally agree. But the few graves with sex/age information fit very well with the pattern of age-relationship, also found in Frankish areas. NIJDAM  The women in this age-group being prominent must have something to do with female fertility. NICOLAY  There can be different explanations, but women seem to play a crucial role in expressing group identity in the Early-medieval North Sea area. NIJDAM  I am developing an idea out of Annet Nieuwhof ’s idea of family archives (this vol., Ch. 3), and now it pops up again. Can I relate this to the legal material, I wonder? Where does family become important? In paying wergild, various members had to pay different parts. Can we somehow connect these notions and make the family archive a bit more tangible? 303

Z01 Hines Book B.indb 303

07/05/2021 13:30

Frisians of the Early Middle Ages NICOLAY  For my dissertation I worked on Roman militaria in the central Dutch river area. There we see many finds of the soldiers’ military equipment in rural settlements. I think that locally recruited people took their armour home after their twenty-five-years’ service, and this was kept as memorabilia in the family household. Archaeologically therefore you can see this as like an archive. NIEUWHOF  I call it an archive, because I think it must be connected to storytelling. NIJDAM  Then I come back to my silver spoons. I have found a will from the eighteenth century where a grandmother had four spoons with her initials engraved in them decades earlier and in the will the granddaughter bequeaths two of those spoons to two family members. Her granddaughter was named after her grandmother so it is no surprise that she owned two of the original four spoons. NICOLAY  Memorabilia from the past are still important today. They relate to stories about fathers/mothers and grandfathers/mothers, and to the identity of families or other social groups. KNOL  We have this collectively too. We are currently in a building stuffed with Frisian family memorabilia. WOOD  One simple point is a story in Paul the Deacon, where a cup is brought out of the royal treasury and the story that goes with it is told. We know that objects were associated with stories. KNOL  What is done by the elite with very valuable things is often also done by those lower down with simpler and humbler things. They have the same meaning and feelings associated. NICOLAY  This fits nicely with the gold and silver objects I have studied. There is hardly any item that is not heavily worn, so they must have been in circulation for some generations before deposition. Here the importance of the associated (origin) stories was for the royal family and their elite network, not just the family. KNOL  You also have it with religious objects. The church possessions are often very old and people are proud of that. The story in the Bible (Exodus) of the bronze serpent is associated with later worship of the snake, when then the snakes have to be destroyed. The object switched from memory to worship, and then it was important to destroy it. At a certain point heirlooms can be destroyed because there is a desire to get rid of a certain tradition or story. NIJDAM  A final point on the silver spoons. Being given access to these heirlooms is an important matter. There is a case of a boy who was not named after his grandfather, against family tradition. The boy was officially renamed after a year, and at that point he was given the silver spoon. So the objects serve to mark admission into a tradition. NICOLAY  It is relevant that historical and material evidence are conjoined there. Archaeological objects are not only datable pieces. Much more fascinating are the history and related story that goes with the object.

References Axboe, M. 1999, ‘The chronology of the Scandinavian gold bracteates’. In The Pace of Change. Studies in Early-Medieval Chronology, eds. J. Hines, K. Høilund Nielsen and F. Siegmund, Cardiff Studies in Archaeology (Oxford), 126–47. — 2007, Brakteatstudier, Nordiske Fortidsminder B-25 (Copenhagen).

304

Z01 Hines Book B.indb 304

07/05/2021 13:30

Symbolism and Group Identities Bazelmans, J. 1999, By Weapons Made Worthy. Lords, Retainers and their Relationship in Beowulf, Amsterdam Archaeological Studies 5 (Amsterdam). Behr, C. 2010, ‘New bracteate finds from early Anglo-Saxon England’, Medieval Archaeology 54, 34–88. Bemmann, J. 1993, ‘Die Nydamfibeln. Eine Fibelform der Stufe C3?’, Germania 71, 139–82. Besteman, J. C. 2004, ‘Two Viking Hoards from the former island of Wieringen (the Netherlands): Viking relations with Frisia in archaeological perspective’. In Land, Sea, Home, eds. J. Hines, A. Lane and M. Redknapp (Cardiff), 95–108. Boersma, J. W. 1977, ‘Een imitatie-solidus van het type munus divinum uit de omgeving van Ten Post (Gr.) en soortgelijke in Noord-Nederland gevonden exemplaren’, Groningse Volksalmanak 1976–1977, 161–82. Böhme, H. W. 1974, Germanische Grabfunde des 4. bis 5. Jahrhunderts, zwischen unterer Elbe und Loire. Studien zur Chronologie und Bevölkerungsgeschichte (München). — 1986, ‘Das Ende der Römerherrschaft in Britannien und die angelsächsische Besiedlung Englands im 5. Jahrhundert’, Jahrbuch des Römisch-Germanischen Zentralmuseums Mainz 33, 469–574. Bos, J. M. 2006, ‘Medieval brooches from the Dutch province of Friesland (Frisia): a regional perspective on the Wijnaldum brooches. Part I: small equal-armed brooches’, Palaeohistoria 47/48, 455–77. — 2008, ‘Medieval brooches from the Dutch province of Friesland (Frisia): a regional perspective on the Wijnaldum brooches. Part II disc brooches’, Palaeohistoria 49/50, 709–93. Bos, J. M. and E. W. Brouwer 2005, ‘De kruisvormige fibulae van Friesland. Een nieuwe benadering in het onderzoek naar de herkomst van de eerste migranten vanaf AD 380/400 en de culturele oriëntatie van het Friese terpengebied tot ca. AD 550’, De Vrije Fries 85, 9–36. Botman, A. 1994, ‘De Domburgfibula, een Fries type’ (MA thesis, Vrije Universiteit Amsterdam). Brandenburgh, C. R. 2010, ‘Early medieval textile remains from settlements in the Netherlands. An evaluation of textile production’, Journal of Archaeology in the Low Countries 2–1, 41–79. Brieske, V. 2001, Schmuck und Trachtbestandteile des Gräberfeldes von Liebenau, Kr. Nienburg/Weser. Vergleichende Studien zur Gesellschaft der frühmittelalterlichen Sachsen im Spannungsfeld zwischen Nord und Süd, Studien zur Sachsenforschung 5/6 (Oldenburg). Brouwer, E. W. 2004, ‘De kruisvormige fibulae van Friesland als uitgangspunt voor de chronologie en ontwikkeling van bewoning van het Friese kustgebied gedurende de periode AD 380/400 to 550’ (MA thesis, University of Groningen). Bruns, D. 2003, Germanic Equal Arm Brooches of the Migration Period. A Study of Style, Chronology and Distribution, Including a Full Catalogue of Finds and Contexts, BAR International Series 1113 (Oxford). Capelle, T. 1976, Die frühgeschichtlichen Metallfunde von Domburg auf Walcheren, Nederlandse Oudheden 5 (Amersfoort). Conkey, M. W. 1990, The Use of Styles in Archaeology (Cambridge). Corbey, R. 2000, Tribal Art Traffic. A Chronicle of Taste, Trade and Desire in Colonial and Post-Colonial times (Amsterdam). Cuijpers, A. G. F. M., C. M. Haverkort, J. M. Pasveer and W. Prummel 1999, ‘The human burials’. In The Excavations at Wijnaldum. Reports on Frisia in Roman and Medieval

305

Z01 Hines Book B.indb 305

07/05/2021 13:30

Frisians of the Early Middle Ages Times. Volume I, eds. J. C. Besteman, J. M. Bos, D. A. Gerrets, H. A. Heidinga and J. de Koning (Rotterdam–Brookfield), 305–21. De Koning, J. 2018, ‘Trans Flehum. Wijnaldum, Den Burg, Texel, Westergo: het Vlie als verbinder en grens’. In Fragmenten uit de Rijke Wereld van de Archeologie, eds. A. Nieuwhof, E. Knol and J. Schokker, Jaarverslagen van de Vereniging voor Terpenonderzoek 99 (Groningen), 146–56. De Langen, G. J. and J. A. Mol 2017, ‘Church foundation and parish formation in Frisia in the tenth and eleventh centuries: a planned development?’, The Medieval Low Countries 4, 1–55. De Leeuw, L. F. 2002, ‘Small long brooches. Chronologie, verspreiding en de iconografische betekenis van small long brooches in het perspectief van de Migratieperiode’ (MA thesis, University of Amsterdam). De Vries, J. 1957, Altgermanische Religionsgeschichte (Berlin). Dijkstra, M. 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’ (Ph.D. thesis, University of Amsterdam). Dijkstra, M. and J. de Koning 2017, ‘“All quiet on the western front?” The western Netherlands and the “North Sea Culture” in the Migration period.’ In Frisians and their North Sea Neighbours. From the Fifth Century to the Viking Age, eds. J. Hines and N. IJssennagger (Woodbridge), 53–74. Frick, H.-J. 1993, ‘Karolingisch-ottonische Scheibenfibeln des nordlichen Formenkreis’, Offa 49/50, 243–463. Gaimster, M. 2011, ‘Image and power in the Early Saxon Period’. In The Oxford Handbook of Anglo-Saxon Archaeology, eds. H. Hamerow, D. A. Hinton and S. Crawford (Oxford), 865–91. Geake, H. 1997, The Use of Grave-Goods in Conversion Period England, c. 600–c.850, BAR British Series 261 (Oxford). Geary, P. J. 2002, The Myth of Nations. The Medieval Origins of Europe (Princeton–Oxford). Genrich, A. 1951, ‘Über Schmuckgegenstände der Völkerwanderungszeit im nordöstlichen Niedersachsen’, Neues Archiv Niedersachsen 23, 251–81. Grierson, P. 1951, ‘The gold solidus of Louis the Pious and its imitations’, Jaarboek voor Munt- en Penningkunde 38, 1–41. Haseloff, G. 1973, ‘Zum Ursprung der germanischen Tierornamentiek – die spätrömische Wurzel’, Frühmittelalterliche Studien 7, 406–42. — 1981, Die germanische Tierornamentiek der Völkerwanderungszeit, Vorgeschichtliche Forschungen 17 (Berlin–New York). — 1990, Email im frühen Mittelalter. Frühchristliche Kunst von der Spätantike bis zu den Karolingern (Marburg). Hedeager, L. 1992, ‘Kingdoms, ethnicity and material culture: Denmark in a European perspective’. In The Age of Sutton Hoo. The Seventh Century in North-Western Europe, ed. M. O. H. Carver (Woodbridge), 279–300. — 1993, ‘The creation of Germanic identity. A European origin-myth’. In Frontières d’empire. Nature et signification des frontiers Romaines, eds. P. Brun, S. van der Leeuw and C.R. Whittaker (Nemours), 121–31. — 2011, Iron Age Myth and Materiality. An Archaeology of Scandinavia, AD 400–1000. (London–New York). Heeren, S. 2017, ‘From Germania Inferior to Germania Secunda and beyond. A case study

306

Z01 Hines Book B.indb 306

07/05/2021 13:30

Symbolism and Group Identities of migration, transformation and decline’. In Social Dynamics in the Northwest Frontiers of the Late Roman Empire. Beyond Decline and Transformation, eds. N. Roymans, S. Heeren and W. de Clercq, Amsterdam Archaeological Studies 26 (Amsterdam), 149–78. Heeren, S. and L. van der Feijst 2017, Prehistorische, Romeinse en middeleeuwse fibulae uit de Lage Landen. Beschrijving, analyse en interpretatie van een archeologische vondstcategorie (Amersfoort). Heidinga, H. A. 1987, Medieval Settlement and Economy North of the Lower Rhine. Archaeology and History of Kootwijk and the Veluwe (the Netherlands) (Assen–Maastricht). Hills, C., K. Penn and R. Rickett 1984, The Anglo-Saxon Cemetery at Spong Hill, North Elmham. Part III: Catalogue of Inhumations, East Anglian Archaeology 21, Norfolk Museums Service (Norwich). Hirst, S. and D. Clark 2009, Excavations at Mucking. Volume 3, The Anglo-Saxon Cemeteries (London). Hodder, I. 1982, Symbols in Action. Ethnoarchaeological Studies of Material Culture (Cambridge). IJssennagger, N. 2017, ‘Central because Liminal. Frisia in a Viking Age North Sea World’ (Ph.D. thesis, University of Groningen). Kars, M. 2013, ‘The early-medieval burial evidence and concepts of possession: questioning individual identities’, Neue Studien zur Sachsenforschung 4, 95–105. — 2018, ‘Animal symbolism, silver smiths, Merovingian graves and early Christianity: (how) are they connected?’. In Rural Riches & Royal Rags? Studies on Medieval and Modern Archaeology Presented to Frans Theuws, eds. M. Kars, R. van Oosten, M. A. Roxburgh and A. Verhoeven (Zwolle), 136–43. Kars, M. and S. Heeren 2018, ‘Archaeological small finds recording in the Netherlands: The framework and some preliminary results of the project Portable Antiquities of the Netherlands (PAN)’, Medieval Settlement Research 33, 18–27. Knol, E. 1986, ‘Het dodenbestel in de Noord-Nederlandse kuststreken tot de elfde eeuw. Een inventarisatie’ (unpublished report) (Groningen). — 1993, ‘De Noordnederlandse Kustlanden in de Vroege Middeleeuwen’ (Ph.D. thesis, Vrije Universiteit Amsterdam). — 2004, ‘Gold und Silber aus Marsum – karolingische Schatzfunde in den Niederlanden’. In Die Macht des Silbers. Karolingische Schätze im Norden, eds. E. Wamers and M. Brandt (Regensburg), 119–24. Knol, E. and X. Bardet 1999, ‘Carolingian weapons from the cemetery of Godlinze, the Netherlands’. In In Discussion with the Past, Archaeological Studies Presented to W. A. van Es, H. Sarfatij, W. J. H. Verwers and P. J. Woltering (Zwolle), 213–25. Knol, E., W. Prummel, H. T. Uyttenschaut, M. L. P. Hoogland, W. A. Casparie, G. J. de Landen, E. Kramer and J. Schelvis 1996, ‘The early medieval cemetery of Oosterbeintum (Friesland)’, Paleogistoria 37/39, 245–416. Koch, A. 1998, Bügelfibeln der Merowingerzeit im westlichen Frankenreich, Römisch-Germanisches Zentralmuseum Monographien 41 (Mainz). — 1999a, ‘Friesisch-sächsische Beziehungen zur Merowingerzeit zum Fund einder Bügelfibel vom Typ Domburg auf dem sächsischen Gräberfeld vom Liebenau, Ldkr. Nienberg (Weser)’, Nachrichten aus Niedersachsens Urgeschichte 68, 67–87. — 1999b, ‘Nordeuropäisches Fundmaterial in Gräbern Süddeutschlands rechts des Rheins’. In Völker an Nord- und Ostsee und die Franken, eds. U. von Freeden, U. Koch and A. Wieczorek, Kolloquien zur Vor- und Frühgeschichte 1 (Bonn), 175–94.

307

Z01 Hines Book B.indb 307

07/05/2021 13:30

Frisians of the Early Middle Ages Leeds, E. T. 1945, ‘The distribution of the Angles and Saxons archaeologically considered’, Archaeologia, or Miscellaneous Tracts Relating to Antiquity 91, 1–106. Martin, T. F. 2015, The Cruciform Brooch and Anglo-Saxon England (Woodbridge). Mazzo Karras, R. 1985, ‘Seventh-century jewellery from Frisia. A re-examination’, AngloSaxon Studies in Archaeology and History 4, 159–77. Miedema, M. 1983, ‘Vijfentwintig eeuwen bewoning in het terpenland ten noordwesten van Groningen’ (Ph.D. thesis, Vrije Universiteit Amsterdam). Milis, L. 2005, Religion, Culture, and Mentalities in the Medieval Low Countries: Selected Essays (Turnhout). Mostert, M. 1993, ‘De kerstening van Holland (zevende tot twaalfde eeuw). Een bijdrage aan de middeleeuwse religieuze geschiedenis’, Holland 15, 125–55. Mückenberger, K. 2013, ‘Eine frühe Wodan-/Odin-Darstellung an der Huntemündung’, Neue Studien zur Sachsenforschung 4, 159–68. Nicolay, J. A. W. 2005, ‘Nieuwe bewoners van het terpengebied en hun rol bij de opkomst van Fries koningschap. De betekenis van gouden bracteaten en bracteaatachtige hangers uit Friesland (vijfde – zevende eeuw na Chr.)’, De Vrije Fries 85, 37–103. — 2008, ‘Een gouden pseudo-munthanger uit het Groningse terpengebied. De 9e-eeuwse elite in beeld’, Paleo-aktueel 19, 161–7. — 2014, The Splendour of Power: Early Medieval Kingship and the Use of Gold and Silver in the Southern North Sea Area (5th to 7th century AD), Groningen Archaeological Studies 28 (Groningen). — 2017a, ‘Power and identity in the southern North Sea area: The Migration and Merovingian periods’. In Frisians and their North Sea Neighbours. From the Fifth Century to the Viking Age, eds. J. Hines and N. IJssennagger (Woodbridge), 75–92. — 2017b, ‘Odin in Friesland. Scandinavian influences in the southern North Sea area during the Migration and Early Merovingian periods’. In Interaktion ohne Grenzen. Beispiele archäologischer Forschungen am Beginn des 21. Jahrhunderts, eds. B. V. Eriksen, A. Abegg-Wigg, R. Bleile and U. Ickerodt (Schleswig), 499–514. Nicolay, J. A. W. and R. A. van Eerden 2021. ‘Wodan’s mythical birds. Symbolic language on a Domburg brooch from Heiloo, North-Holland (the Netherlands)’, Archäologisches Korrespondenzblatt 51, 110–34. Nielsen, K. H. 2009, ‘The real thing or just wannabes? Scandinavian-style brooches in the fifth and sixth centuries’. In Foreigners in Early Medieval Europe: Thirteen International Studies on Early Medieval Mobility, ed. D. Quast, Monographien der Römisch-Germanischen Zentralmuseums 78 (Darmstadt), 51–111. Nieuwhof, A. 2011, ‘Discontinuity in the northern-Netherlands coastal area at the end of the Roman period’. In Transformations in North-Western Europe (AD 300–1000), eds. T. A. S. M. Panhuysen and B. Ludowici, Neue Studien zur Sachsenforschung 3 (Hannover– Stuttgart), 55–66. — 2016, ‘De lege vierde eeuw’. In Van Wierhuizen tot Achlum. Honderd Jaar Archeologisch Onderzoek in Terpen en Wierden, ed. A Nieuwhof, Jaarverslagen van de Vereniging voor Terpenonderzoek 98 (Groningen), 83–98. Olsen, V. S. 2000, ‘Wodan in Wijnaldum: de dubbelzinnige, iconologische boodschappen van de beugelschijffibula van Wijnaldum-Tjitsma (Fr.)’, Paleo-aktueel, 116–21. —2006, ‘The development of (proto)-disc-on-bow brooches in England, Frisia and Scandinavia’, Palaeohistoria 47/48, 479–528. Palm, M. and J. Pind 1992, ‘Anglian English women’s graves in the fifth to seventh centuries

308

Z01 Hines Book B.indb 308

07/05/2021 13:30

Symbolism and Group Identities A.D. – a chronological analysis’. In Chronological Studies of Anglo-Saxon England, Lombard Italy and Vendel Period Sweden, ed. L. Jørgensen (Copenhagen), 50–93. Penn, K. and B. Brugmann 2007, Aspects of Anglo-Saxon Inhumation Burial: Morning Thorpe, Spong Hill, Bergh Apton and Westgarth Gardens. East Anglian Archaeology 119 (Gressenhall, Norfolk). Postma, J. 2015, ‘Vroeg-christelijke ringen in Friesland en hun mogelijke betekenis op het kersteningsproces. Een hypothese’, Detector Magazine 140, 16–18. Price, N. and P. Mortimer 2014, ‘An eye for Odin? Divine role-playing in the age of Sutton Hoo’, European Journal of Archaeology 17 (3), 517–38. Rau, A. 2010, Nydam mose 1. Die personengebundenen Gegenstände. Grabungen 1989–1999 (Aarhus). — 2013, ‘Some observations on Migration period “Hacksilber” hoards with Roman components’, Neue Studien zur Sachsenforschung 4, 189–203. Reichstein, J. 1975, Die kreuzförmige Fibel. Zur Chronologie der späten römischen Kaiserzeit und der Völkerwanderungszeit in Skandinavien, auf dem Kontinent und in England, OffaBücher 34 (Neumünster). Roth, H. 1996, ‘Kunst der Merowingerzeit’. In Die Franken, Wegbereiter Europas. Vor 1500 Jahren: König Chlodwig un seine Erben (Mainz), 629–39. Roxburgh, M., H. Huisman and B. van Os 2014, ‘All change? The end of a metalworking tradition in early medieval Frisia’, De Vrije Fries 94, 19–30. — 2016, ‘The cross & the crucible: The production of Carolingian disc brooches as objects of religious exchange?’, Medieval and Modern Matters 5, 117–32. Roymans, N. 1990, Tribal Societies in Northern Gaul. An Anthropological Perspective (Amsterdam). Schön, M. D. 1999, Feddersen Wierde, Fallward, Flögeln. Archäologie im Museum Burg Bederkesa, Landkreis Cuxhaven (Bremerhaven). Theuws, F. C. W. J. and H. Hiddink 1996, ‘Der Kontakt zu Rom’. In Die Franken, Wegbereiter Europas. Vor 1500 Jahren: König Chlodwig un seine Erben (Mainz), 66–80. Thörle, S. 2001, Gleicharmige Bügelfibel des frühen Mittelalters, Universitätsforschungen zur prähistorischen Archäologie 81 (Bonn). Van Bellingen, S. 1988, ‘Gelijkarmige fibulae uit de Merovingische en Karolingische periode in België en Noord-Frankrijk’ (MA thesis, Vrije Universiteit Brussel). Van der Meer, L. B. forthcoming, ‘The (hanging-)bowl of Oegstgeest’. In Oegstgeest. A Riverine Settlement in the Early-Medieval World System, ed. C. Bakels, Merovingian Archaeology in the Low Countries 6 (Leiden). Waasdorp, J. A. and E. Eimermann 2008, Solleveld. Een opgraving naar een Merovingisch grafveld aan de rand van Den Haag, Haagse Oudheidkundige Publicaties 10 (Den Haag). Wamers, E., 2009: ‘Behind animals, plants and interlace: Salin’s Style II on Christian objects’. In Anglo-Saxon/Irish Relations before the Vikings, ed. J. Graham-Campbell and M. Ryan (Oxford), 151–204. — 2013, ‘Goldscheibe von Limons’. In Credo. Christianisierung Europas im Mittelalter. Band II: Katalog, eds. C. Stiegemann, M. Kroker and W. Walter (Paderborn), 140–1. Waterbolk, H. T. 1979, ‘Siedlungskontinuität im Küstengebiet der Nordsee zwischen Rhein und Elbe’, Probleme der Küstenforschung im südlichen Nordseegebiet 13, 1–21.

309

Z01 Hines Book B.indb 309

07/05/2021 13:30

Z01 Hines Book B.indb 310

07/05/2021 13:30

• 11 • RELIGION AND CONVERSION AMONGST THE FRISIANS John Hines

I

t is an unwelcome fact to have to face, that studies of pre-Christian religion amongst the Germanic-speaking peoples have recently shown little sign of progressing in a very coherent manner, and indeed in the course of the past quarter-century or so have been moving away from rather than towards any consensus even over the proper critical methodology to be brought to this topic. Specialization in the disciplines relevant to some particular source of evidence can lead to powerful convictions in favour of certain views, perhaps nowhere more clearly exemplified than in Anatoly Liberman’s etymologically embedded reconstruction of a prehistoric Germanic religion dominated by the terror of chthonic spirit leaders and hordes (work of many years collected and revised in Liberman 2016). Different only in mode is the imposition of perceived analogies or even just casual assumptions upon material phenomena such as the burial of the dead with grave goods, or potential cases of graphic or numerical symbolism (e.g. Andrén 2014; review Hines 2015a; further, infra). It is quite possible for suggestions made through such diverse approaches to this aspect of the distant past to be historically correct – and perhaps they are even more likely to be partially correct – but they will still be essentially flawed if they are conclusions drawn for the wrong reasons. I stand resolutely by the contention that it is utterly fundamental to the study of religion that the phenomenon itself be adequately characterized at a general level to enable us to identify with defensible credibility where religion may be inherent in what are essentially very different material, linguistic or textual phenomena (Hines 1997; 2015b). Studies of ancient and historical religion are otherwise no more than a series of variously lucky or wild guesses which discredit both the object of study and the scholarly attempt to engage with it. This critique is based upon a view that posits that religion is essentially immaterial, residing at its very heart in a common human sense of numinism: a field of cognition which in a truly Kantian sense ‘transcends’ the material and social substrates of the culture within which this consciousness exists and to which it must necessarily also relate; perhaps not (to develop the Kantian analogy further) as an a priori category of understanding but certainly as an autonomous element of the culture’s ideology. The institutionalized and major religions that became established over large areas of the medieval and modern world have been able to attempt to define, explore and evaluate their beliefs in massive volumes of written scripture and commentary. That situation, however, also creates not only abundant space for radically divergent positions 311

Z01 Hines Book B.indb 311

07/05/2021 13:30

Frisians of the Early Middle Ages amongst equally dedicated scholars within each religious tradition, but also tends to produce a field of theological scholarship that is remote in the extreme from what is thought and comprehended by the great majority of express adherents of the religions. It is observably a common and regular fact that the latter largely maintain their fidelity in the form of no more than a partial knowledge and acceptance of some tenets of theological doctrine alongside routinized observance of particular practices – which may indeed be matters of abstinence from or the shunning of otherwise feasible and familiar behaviour rather than of doing certain things. There can thus truly be a plethora of thoroughly valid but completely different answers to such basic questions as ‘What is Christianity?’; ‘What is Islam?’; ‘What is Judaism?’. Et ceterae. Studying a population such as the group we know of as the Early-medieval Frisians through the process of the adoption of one of the sophisticated medieval and modern religions – conversion to Christianity in this specific case – can pose especial problems because of fundamental differences in the articulation of religion either side of the historical change, between oral tradition before and documented history afterwards. At the same time the production of documentary records relating to the history of conversion can provide unique and valuable data. It is a truism that, when seeking information on the whole range of religious culture before, during and after the Christianization of a population, one must be intensely aware of how dependent one is upon the perceptions recorded as nominally historical records from the passionately partisan quarter of the successful evangelists: a source convinced of its own divine authority and sacred importance, and concurrently of the truly diabolical nature not only of alternatives to itself but even of unorthodox, ‘heretical’, variants of its accepted doctrines. The Church also had a strong sense of religious history, rooted in traditions within the Jewish Old Testament going back some two thousand years, and so inevitably had a habit of perceiving, interpreting and representing other religious expressions it encountered in the terms those traditions provided it with. What is not a truism, though, is that early Christian sources on the conversion of the Frisians must therefore be discounted in their entirety as historical evidence for anything other than what the presuppositions of Early-medieval Christian orthodoxy were. Nor would we be justified in assuming or postulating that pre-Christian Germanic religious traditions can only have been matters of habitualized ritual combined with superstitious mythological traditions. It is as indefensible to construct, a priori, sharply contrastive pre- and post-Conversion characterizations of religion as it is to use familiar aspects of modern religions to generate identifications and interpretations of religious behaviour in the pre-Christian age, effectively forcing the past into an interpretative assimilation to the present. In the case of the Early-medieval Frisians, for explicit historical evidence of both the pre-Christian religious traditions of the area and people, and for accounts of the progress of conversion, we are heavily reliant on a series of missionary saints’ Lives: those of Willibrord, Wulfram, Boniface, Willehad, Lebuin and Liudger; although his role was different, that of Gregory of Utrecht is also important (see Primary Sources). These portray the period from the late seventh to the early ninth centuries but were composed between the mid-eighth and mid-ninth centuries. In the case of practically every alleged event and fact directly concerned with religion within them, we constantly have to ask ourselves whether we can give any credence at all what is claimed; and likewise whether or how we might validly deconstruct and interpret those sources in order to produce a plausible and meaningful historical analysis. There are, of course, other sources besides 312

Z01 Hines Book B.indb 312

07/05/2021 13:30

Religion and Conversion these: documentary sources such as Boniface’s letters (ed. Tangl 1916), or the Penitential ascribed to Willibrord (Meens 1993; 2000; Wood 2001, 79), and the extraordinary Titulus XI of the Additio sapientum to Lex Frisionum, ‘De honore templorum’ (see further below); and, as this chapter will illustrate, place-names and linguistic evidence provide further valuable pieces of information. Archaeological remains can also be compared with and integrated with these sources within an overall picture. Altogether, it seems both practical and tempting to consider the Early-medieval transition of Frisia from ‘heathen/pagan’ to Christian (the latter label could just as well be put in inverted commas) primarily in a geographically broad, international and cross-cultural perspective. Conversion, and resistance to conversion, had by the eighth century undoubtedly become fundamental to large-scale political relations and ambitions in western Europe. While both later Merovingian and earlier Carolingian rulers of Francia visibly took an active interest in strengthening and extending the grasp of the Church in their northern Continental territory, and Frankish clerics are prominent in early Church foundations within the Anglo-Saxon kingdoms of the early to middle seventh century (Wood 1994, 178–9), both south of the Rhine and beyond it into Frisia (and ultimately Saxony too), as is well known, the effective missionary effort amongst the Frisians was very much in the hands of Anglo-Saxons: especially Willibrord, but likewise Wynfrith/Boniface, Lebuin and Willehad (Levison 1946, 45–69; Wood 2001, 10–12, 57–115 passim). A close cultural relationship across the North Sea between Frisia and Anglo-Saxon England dating back to the sixth century if not earlier, together with particularly close (and not accidental) similarities in language, as well as strong contemporary economic ties, will have combined to make that an entirely practical matter (Hines 2017). The history of the spread of Christianity in Britain and Ireland from the fifth century onwards and the conversion of the Anglo-Saxons from the late sixth century meant that missionary models and ideals were exceptionally strong within the English Church – perhaps significantly more so than within the Gallo-Frankish Church of the Merovingian Period. Thus in the late seventh and eighth centuries, Anglo-Saxon Christians were in many ways best suited to the task of converting the Frisians. It is also apparent, throughout the eighth century, that evangelization in Frisia was rarely thought of or attempted in isolation from the wider efforts to impose Christianity upon the Saxons (e.g. McKitterick 1983, 60–3), or to establish more general orthodoxy further south in the Germanic lands east of the Rhine: in Hesse, Thüringia and Bavaria (Levison 1946, 72–93; Wood 1994, 305–21). Without rejecting the valuable insights that a comparative approach can yield, however, there is good reason also to emphasize the peculiarities and specificity of the Frisian situation. Not least of those distinct factors must surely be the proposition that the northern Netherlands had effectively been depopulated for a period up to the early fifth century AD, eventually to be repopulated with a Germanic-speaking population where previously a Celtic language had been in use, and with strong cultural ties both east and west around the North Sea littoral from southern Scandinavia through northern Germany to the east coast of Anglo-Saxon England (Knol and IJssennagger 2017; Schrijver 2017). Whatever we may be able to detect, then, of religious practice and organization in Germanic Frisia from the fifth century down to the gradual introduction of Christianity in the later seventh and eighth centuries should have been introduced less than three centuries previously. We can suppose it to reflect, in its adaptations, the particular needs of the reconstituted society of Frisia. To a significant degree, the same state of 313

Z01 Hines Book B.indb 313

07/05/2021 13:30

Frisians of the Early Middle Ages affairs must also have existed in fifth- and sixth-century England, although conversion started in earnest amongst the Anglo-Saxons a century earlier than in Frisia, and the presence of sub-Roman Christian survival even in the south and east of Britain, as well as the growth and flourishing of Churches amongst the Britons, Irish, Scots and Picts, also represent equally significantly different circumstances (Blair 2005, 8–78; Higham 1997).

Frisian ‘paganism’ On more than one occasion I have addressed the problem of the pre-Christian religious traditions of Anglo-Saxon England in what can be characterized as distinctly sceptical terms (Hines 1997; 2015b). To date, I see no cause to recant positions previously argued for, nor any new need to vindicate them. As implied in the introduction to this chapter (above), one aspect of my scepticism which may reasonably be stressed even more explicitly is that what may appear to be extreme caution in accepting inferences of the religious character of various phenomena is in fact based fundamentally on a commitment to considering what religion really is or was in principle. This means an emphasis on the essentially ideological nature of religion as a set of beliefs and understandings, and furthermore on the spiritual character of religious ideas and beliefs as that which distinguishes religious ideologies from their secular counterparts. Spirituality in this context itself is a broad category, capable of comprising many different forms of animism, numinism and theism, as well as potentially (although not necessarily) the idea of a human soul that is distinct from the human body and may live on after the physical death of the individual. That position is one which remains satisfied with one of the simpler nineteenth-century characterizations of ‘religion’ (Tylor 1871, 383: ‘it seems best … simply to claim, as a minimum definition of Religion, the belief in Spiritual Beings’), and may indeed be criticized as too simple as a universal definition when considered in diverse cultural perspectives (as Tylor, indeed, understood perfectly well). It nevertheless appears to fit and to work entirely satisfactorily with everything we can discern of traditional pre-Christian Germanic religion, and indeed of religions throughout Europe of the first millennium AD. An important corollary of this position then, is to insist upon a separation of religion and ritual, congruent with stressing the distinction between religious and secular aspects of ideology: religion, understood spiritually, is undoubtedly likely to involve rituals but it is highly unlikely to subsist in ritual alone, while many rituals are likely to be essentially or entirely non-religious in character. A final point to be made, however, is that it is fully recognized that this desire for a strict definition of religion is undoubtedly that of the analyst of cultural history, seeking to highlight what are argued to be significant nuances within the behaviour and practices of a distant past: there can be no suggestion that it is a position which exactly reproduces or even approximates in any significant way to cognitive frameworks operative within that past. An overlapping and confusion of religious and secular interests can be expected to have predominated in those contexts. In the case of the Frisians, caution nevertheless does allow us to conclude that a range of sources combine to provide a consistent and credible image of the infrastructure of the ‘pagan’ religion that the missionaries sought to extirpate and to replace with Christianity. In this case, indeed, the range of both similarities and differences from the evidence from across the North Sea in England also permits interpretation in terms of 314

Z01 Hines Book B.indb 314

07/05/2021 13:30

Z01 Hines Book B.indb 315

07/05/2021 13:30

Fig. 11.1  Map of the Frisian region showing the principal sites referred to. Place-names apparently referring to pre-Christian religious sites in italics. Note the orientation of the North point.

Frisians of the Early Middle Ages chronological development: in other words, what we can detect for Frisia represents how the sort of self-consciously pagan reaction and resistance to Christianity that can be argued for in seventh-century England (Hines 1997, esp. 387) could have been carried forward had it survived into the eighth century. Even allowing for inter-textual influence, the historical sources are, for instance, quite consistent in stating that the religion was focussed upon a series of shrines serving a polytheistic pantheon (fana deorum, as it is put in Altfrid’s Vita Liudgeri [I, 16]). Worship and cults focussed upon idols (varia[e] cultura[e] idolorum: Vita Liudgeri, loc. cit.) are also repeatedly referred to. After the death of Radbod, when Willibrord could return to Utrecht with Charles Martel’s support, the first of the miracles associated with him is the rather grim tale of the punishment by God of the (sword-wielding) custos of an idol on the island of Walcheren which Willibrord destroyed (Vita Willibrordi, ch. 14; Talbot 1954, 12–13). That location at the mouth of the Scheldt estuary is at best marginal to Frisia proper (see Fig. 11.1; Hines and IJssennagger 2017, 2–3), but the sources do not suggest that the Rhine boundary between Francia and Frisia constituted a sharp divide between Christendom and heathen Europe: far from it. Fositeland (Helgoland) had temples to the eponymous god Fosite (cf. Old Norse Forseti, recorded as a son of Baldr by Snorri Sturluson and thus one of the Æsir [Gylfaginning, ch. 32; Skáldskaparmál, ch. 5 and verse 432]), to whom the entire island appears to have been sacred, according to Alcuin in Vita Willibrordi (Vita Willibrordi, chs. 10–11; Talbot 1954, 10–11). The key specifics of the taboos associated with this place, however, are that the grazing animals there were treated as sacred, and a freshwater spring had to be approached with awe and reverence, in silence. Willibrord slaughtered some of the beasts for food, and baptised three people in the spring, with no terrible retribution from the god. The island was also under the protection of the king, Radbod the Frisian, who followed a traditional judicial practice of determining liability for the sacrilege and whether or not execution should follow by the casting of lots. The outcome of the procedure was that one, but only one, of Willibrord’s party was put to death and thus martyrio coronatus est, ‘has been crowned as a martyr’. Towards the end of the eighth century, long after the disappearance of any powerful king of the Frisians to protect it, Liudger supposedly returned to Helgoland to re-enact and complete Willibrord’s attempt to destroy the shrine (Altfrid, Vita Liudgeri I, 22; Wood 2001, 110–11). So far as I am aware, no buildings identified as cult-houses or shrines have yet been identified at any Frisian site. Such buildings remain extremely elusive in England too, with Yeavering (Northumberland) Building D2 still the sole really convincing example (Semple 2007). In this respect, Scandinavian archaeology has seen much more substantial empirical advances over the past twenty-five to thirty years, with the exploration of the part-sacral landscape at Gudme (< guð-heim: gods-home) on Fyn, and the excavations of the cult sites at Tissø (< Týs-sær: Týr’s lake) on Sjælland and at Uppåkra in Skåne … all within the Early-medieval Danish kingdom (most recently, Jørgensen 2014). Further sites of votive sacrifice have been identified in Sweden (see Price 2014, 178–84), while, continuing even beyond the Viking Period into the 11th century, the oscillating interpretations of the site of Hofstaðir near Mývatn in Iceland have now settled on acceptance of the evidence for regular, cultic rituals here too (Lucas 2009). Where, even recently, it could be stated that evidence for ‘heathen Germanic’ shrines or temples was practically non-existent (with reference, especially, to Olsen 1966), we now have a clear idea that such sites could exist, and even of their potential longevity and 316

Z01 Hines Book B.indb 316

07/05/2021 13:30

Religion and Conversion how chronological changes might be evident, and thus can form an informed opinion on what such a site might look like. As in seventh-century England – possibly earlier too – the case for there having been designated religious sites marked by something other than purely natural or topographical features is supported in Frisia by a small number of specific place-name elements (Quak 2002; van Loon 2017, 62–4). The noun harg (Old Norse hǫrgr; Old English hearg or hearh) appears in Harich in Friesland (spelled Harch in a source dated to 1132), and in Hargen, from a plural in Haragum (AD 822 x 825) near Schoorl north of Alkmaar in North-Holland (Fig. 11.1). The noun wīh (Old Norse vé; Old English wēoh) appears in Wehe (c. AD 1000 Uuie) near Groningen. Both elements also occur in names further south in the Netherlands, along with the term alh (Old English ealh), proposed for Aalburg, Noord-Brabant. All of these terms could be translated ‘shrine’, fanum, although hearg/hearh in early Anglo-Saxon England does appear to be linked to larger, more central cult sites (Hines 1997, 384–7; cf. further Semple 2007). There are no convincing examples from the Frisian territory of the northern Netherlands of place-names containing the name of a god or goddess implicitly venerated there: even the more southerly cases of Dunreslo in Limburg and Woensel in Eindhoven may have been generated as archaizing references to pre-Roman and Roman-period pre-Christian cult centres (van Loon 2017, 43–56; Semple 2007 implies that this practice may have occurred more widely in Early Anglo-Saxon England too). Although recorded only in very late medieval Frisian historical prose, the legend of the creation of the Frisian Laws by twelve lawspeakers (āsegan) who were cast adrift and were eventually divinely guided by Christ does, however, include the detail that they finally come to land and are given the answer they need and seek by Eswei, the first element of which name appears to be an umlaut-inflected genitive singular form of ōs (Old Norse Áss): a kin-group term for one branch of the traditional Germanic gods (text: Bremmer 2009, 174–6; see ibid., 12–14 for the source in MS U; Krogmann 1967; 1971, 167–70).1 In Anglo-Saxon England, we find a number of recurrent topographical features treated as hallowed religious sites – above all the leah, ‘grove’ (cf. Dutch lo: van Loon 1 The grammatical history of this noun in the Germanic languages is rather more complicated than most studies referring to the term reflect. The term is evidenced across the three principal branches of the Germanic language group (East, West and North) but particularly as the first element of personal names: only in the case of the Old Norse (ON) noun with nom. sg. áss, nom. pl. æsir, do we have a full range of grammatical forms in freely generated text. The declension in ON is characteristic of u-stem masculine nouns, with i-mutation in the nominative plural. Consequently a Common Germanic (CGmc) root *ansuz is conventionally reconstructed, particularly as the original name of fourth rune in the fuþark rune-row. In standard Old English (OE) historical grammars, however (Campbell 1959, §§121, 622; Hogg 1992, §2.111), OE *ōs as a noun is assigned to the same ‘athematic’ declension as a small number of other high-frequency nouns including mann, ‘man’, and fōt, ‘foot’, which would imply a CGmc nominative singular *anss. The fact that the reconstructed CGmc genitive singular in *-iz would create i-mutation in this case – which has been completely replaced, analogously, by the inflexional ending -es (from CGmc -as) in all other Old English nouns of this inflexion – and so produce an early Old Frisian (OFris) ēs, lends support to this classification of the noun in OE and OFris at least. The redistribution of the effectively anomalous athematic nouns to more regular declensions is characteristic of all recorded branches of Germanic, so the ON evidence does not count against this suggestion. ēsa in the phrase ēsa gescot in the charm Wiþ færstice in the OE collection known as Lācnunga is to be parsed as the genitive plural, implying the extension of the mutated vowel of the root syllable from the nominative and accusative to all cases of the plural. That is perhaps most likely with a low-frequency, relatively unfamiliar noun.

317

Z01 Hines Book B.indb 317

07/05/2021 13:30

Frisians of the Early Middle Ages 2017, 3–108), but also feld, ‘field’, denu, ‘valley’, mere, ‘mere’, beorg and hoh, ‘hill’ (Hines 1997, 384–6). The site of Heiloo in North-Holland combines the toponymic ‘generic’ lo with a ‘specific’ meaning ‘holy’ (van Loon 2017, 61–2). This is the site of an early founded mother-church and a sacred spring, both traditionally associated with Willibrord. An extraordinary recent find at the Zuiderloo site here is half of an anthropomorphic oak ‘idol’ figure, deposited in the packing surrounding the wooden frame of a well, the construction of which can be dated to the late 620s (pers. comm. Menno Dijkstra; Dijkstra and Verhoeven 2012; illustrated in Lange 2017, 293, Cat. No. 165). The well is associated with a substantial longhouse. More contextual information, and undoubtedly much more discussion of this tantalizing find, must for now be awaited and anticipated with patience. It can come as little surprise that overall, though, it is primarily the seascape of islands and coasts that is endowed with this role in Frisian religious conceptions. The Eswei, where a freshwater spring had allegedly miraculously appeared, is found by the shore. Helgoland (the name recorded from the thirteenth century, and probably meaning Holy Island, although a topographic alternative has been suggested: Siebs 1909) stands out distinctly in the North Sea. Vita Vulframni (esp. ch. 8) includes a report on the ritualistic use of the tides and the sea, albeit in the context of a particularly grim and thus relatively implausible account of the Frisians’ extensive practice of human sacrifice: Mos pessimus, diabolica fraude inventus, praedicto incredulorum duci inerat, ut corpora hominum damnatorum in suorum solemniis deorum – et non deorum, sed daemoniorum execrabilia – sępissime diversis litaret modis, quosdam videlicet gladiatorum animadversionibus interimens, alios patulibus appendens, alios laqueis acerbissime vitam extorquens; praeterea et alios marinorum sive aquarum fluctibus instinctu diabolico submergebat. The most wicked practice, devised by the deceit of the Devil, resided in the aforesaid leader of the heathens, in that he would very often sacrifice as part of the rites of their gods – the appalling rites not of gods, but of demons – the bodies of those condemned to death: condemning some to the attentions of men with swords; hanging others on public gallows; most cruelly squeezing the life out of others using garrottes; while some, with devilish instinct, he would also drown in the tides of the sea or the creeks.2

The question of whether or not the pre-conversion burial rites of a Germanic cultural group such as the Frisians had anything to do with the religious beliefs of the burying population is as inadequately discussed in this context as it is anywhere in northern and western Europe. Both cremation and inhumation were practised by the Germanic-speaking Frisians, effectively side-by-side and in a relatively unconstrained manner, at least until the eighth century, when cremation was gradually superseded by inhumation as the norm, apparently under the influence of ecclesiastical standards (Knol 1993a). At the same time, though, there was a striking upsurge in the frequency of weapon burial with high-ranking males in inhumation graves, which is interpreted 2

The translations offered are the author’s. They are in many cases cross-checked and influenced by those of Talbot (1954), although, quite characteristically of his time, in seeking to achieve a fluent representation of the sense of the original, Talbot is inclined to paraphrase the original rather more liberally than I prefer to.

318

Z01 Hines Book B.indb 318

07/05/2021 13:30

Religion and Conversion in purely secular terms as a direct reflex of the demonstrative imposition of Carolingian Frankish rule and its administration. Particularly interesting burial phenomena in Frisia are the number of cases of burial in oak-log coffins, from as early as the sixth century (Knol 1993b, 176). This practice has recently been found to be paralleled in a Christian eighth-century cemetery at Great Ryburgh (Norfolk), but it is suspected also to have been used with sixth-century inhumations at Mucking (Essex) and Snape and Eriswell (Suffolk) too (Hirst and Clark 2009, 409–10; Caruth and Hines forthcoming, ch. 3). Another distinct aspect of early Frisian cemeteries is the relative frequency of animal inhumations, especially horse burials, followed by dogs (Prummel 1993; Knol, this vol.). This, however, has to be viewed in terms of a ritual that may be of purely secular and even practical ideological character and for which there is no need to create a facet of animal totemism in the pre-Christian Frisian religious traditions. In evaluating the evidence for attributing a distinct religious character to, and therefore religious explanations for, burial practice in England before the widespread conversion of Anglo-Saxon England to Christianity (Hines 2015, 69), I concluded, ‘What the Church would like to see in terms of burial practice did not have to replace a pagan burial practice in England because, I would argue, there was no such thing. Pre-Christian burial practices were not only variable, but also inessential to their religious context.’ I take the same view of near contemporary and closely related Frisia.

Radbod and the politics of religion The figure of Radbod, king of Frisia c. AD 680–719, looms large as both a key character and a profoundly fascinating figure in the missionary stories. It is essential to stress from the outset that this figure is to be understood entirely as a literary construct of precisely those sources. That in itself may nevertheless imply much about the underlying political and historical realities reflected in the documentary evidence, and not least because the Radbod thus portrayed is anything but a caricature, demonic pagan king. An instructive contrast is to be drawn, from Alcuin’s Vita Willibrordi, with the representation of Ongendus, king of the ‘fierce tribes’ of the Danes. He is described uncompromisingly as homo omni fera crudelior et omni lapide durior, ‘a human being more savage than any wild beast and harder than any stone’ … although even he received Willibrord with care and respect. Willibrord found him and the Danes to be obduratam moribus et idolatriae deditam et nullam melioris vitae spem habentem, ‘utterly stubborn in their practices, committed to idolatry, and having no hope of a better life’,3 and soon departed; Willibrod was reportedly, however, able to select and take with him thirty Danish boys to be brought up as Christians (Alcuin, Vita Willibrordi, ch. 9; Talbot 1954, 9–10). Later Frisian sources giving versions of the history of the primary missionary period in the territory do in fact move in the direction of the same form of stereotyping by making Radbod a Danish rather than a Frisian king (Schmidt 1971, 522–3). The contrasts simply encourage all the more careful evaluation of the earlier stories involving Radbod.

3

Sic, Levison ed. It is not clear why the adjectives here are in the feminine singular, although it could imply allusion to a feminine noun such as gens or natio; and some versions of the text have them in the masculine.

319

Z01 Hines Book B.indb 319

07/05/2021 13:30

Frisians of the Early Middle Ages Radbod’s role was allowed to eclipse that of his father Aldgisl, who had received, hosted and protected the turbulent seventh-century Bishop Wilfrid (Eddius Stephanus, Vita Wilfridi, 26–8; Bede, Historia Ecclesiastica, V, 19 [cf. also ibid. III, 13]). Aldgisl and Radbod are in fact the only two known rulers of an extensive, and effectively unified, Frisian kingdom, which must therefore have been formed in or shortly after the middle of the seventh century. It is clearly implied that Aldgisl considered conversion for himself and thus the Christianization of his kingdom, but he evidently held off from any definite decision in that direction. With his better recorded relations and interactions with the final Merovingian and first Carolingian rulers, Radbod inherited not only that position but also the policy challenge it related to. The famous story of his change of mind on the very threshold of baptism told in Vita Vulframni (ch. 9) implies that he came very close indeed to accepting the faith: cum ad percipiendum baptisma inbueretur … ad fontem processerat, ‘when he should be bathed in order to receive baptism … he had made his way to the font’. Ultimately, however, he declined to do so as a result of deliberate reflection. Alcuin describes Radbod as receiving Willibrord in his court humilitatis gratia benigne, ‘kindly, with the grace of humility’ … but nullis tamen vitae fomentis saxeum eius cor emollire potuit, ‘he was unable to soften his stony heart with any salves of Life’ (Alcuin, Vita Willibrordi, ch. 9; Talbot 1954, 9). The wording is by no means incompatible with the interpretation of Vita Vulframni just given. It was also Radbod who, as the king over Fositeland, sent Willibrord and his companions back to Pippin II as ruler of the Franks despite their attempted desecration of the sacred sites and practices of the special island (Alcuin, Vita Willibrordi, ch. 11; Talbot 1954, 11). In doing so, Radbod is portrayed as displaying restraint and diplomacy, as well as respecting the traditional juridical procedures, involving the casting of lots, of his own society. With the potentially unreliable benefit of hindsight, the decision of the Frisian kings not to adopt Christianity can seem surprising. A kingdom such as theirs could convert and flourish, as above all the example of Mercia across the North Sea would have shown them. In the late seventh century Mercia had expanded its power to the wīc of London with which the Frisians’ own Dorestad was in strong and regular contact (see ­Willibald, Vita Bonifatii, ch. IV; Talbot 1954, 35). Of course there were counter-examples that were equally conspicuous. Mercia’s expansion was at the expense of other similarly recently converted kingdoms, overshadowing its principal rivals Wessex and Northumbria, and dominating the smaller Kent, Essex and Lindsey. Being a Christian kingdom was manifestly no guarantee of security. England also differed from Frisia, as could have been recognized, in being a former Roman territory effectively reclaimed for the Church, whereas only on the very borders of Frisia, for instance in Utrecht, were there the earlier Roman sites at which churches may also have been located (Levison 1946, 48; de Groot 1996, 17–20; Parsons 1996, 30). Geographically, the Frisia of Aldgisl and Radbod was contiguous on one side with the still extensive non-Christian areas of the Continent and Scandinavia occupied by Saxons, Danes and Slavs. Moreover the later Merovingian kings and Church had not promoted mission and conversion with great fervour or force (see Discussion). All the same, both Aldgisl and Radbod should have been fully conscious of the potential legitimization and support for an independent Frisian kingdom that a Frisian Church with its principal see at Utrecht could offer. Wulfram’s story implies that Radbod made a last-minute decision in favour of the traditional practices and structures of his people, culture and society instead of 320

Z01 Hines Book B.indb 320

07/05/2021 13:30

Religion and Conversion religious innovation. The objection he is presented as raising, that his ancestors were condemned to Hell and his own assured afterlife would be in another place (however blissful), has been referred to in terms of an ‘ancestor cult’ (see Geary 1994, 35–6). This idea is now strongly promoted by self-styled contemporary ‘Pagans’ and so must be treated with caution; the ancestor cult is in fact a rather vague concept in terms of comparative religious studies, and to invoke it as a key characterization of Frisian paganism is almost certainly misleading. In fact the narrative of the last-minute raising of the question, and Radbod’s laconic response to Wulfram’s orthodox answer, is presented rather whimsically: … pedem a fonte retraxit, dicens, non se carere posse consortio praedecessorum suorum principum Fresionum et cum parvo pauperum numero residere in illo caelesti regno; quin potius non facile novis dictis adsensum praebere, sed potius permansurum se in his, quae multo tempore cum omni Fresionum gente servaverat. … he drew his foot back from the font, saying that he was not able to do without the company of his ancestors who had been princes of the Frisians and to dwell with a small number of poor individuals in that heavenly kingdom; on the contrary it was not easy to give assent to these new teachings, and he was instead going to continue with those that he had observed along with the whole Frisian people for a long time.

As an example of a king responding to earnest ecclesiastical doctrine and rules, this is strikingly similar both in focus and in tone to Oswiu’s pragmatic reliance on the practical power attributed by orthodoxy to St Peter in holding the keys of Heaven to resolve the dispute at the Synod of Whitby in AD 664 (Bede, Historia Ecclesiastica, III, 25). The image the narrative conveys is far less one of the naivety and stubbornness of a foolish pagan king than one of personal courage and dynastic loyalty. Meens (2015) presents a case for interpreting the story in terms emphasizing ecclesiastical orthodoxy; Caciola (2016, 11–13), conversely locates the story within a widespread survey of afterlife traditions. With the Old Frisian term for ‘kin’, kenn, visibly at the etymological core of the position of kening (Old Frisian ‘king’), Radbod’s ancestors stood for his people as a whole, and he held with them. It is certainly valid to postulate that Radbod foresaw major popular division and strife as a consequence of a decision to convert, and opted against that. Admittedly, the specific relevant case we can cite from historical evidence is that of Wrssing, the grandfather of Liudger, a leading nobleman who fell out with Radbod but who converted to Christianity in exile amongst the Franks (Vita Liudgeri, chs. 1–4; Wood 2001, 108–9, 113–14). From AD 714 we can argue also that Radbod’s hand was forced by Charles Martel’s assumption of power after the death of Pippin II, putting aside Grimoald, Radbod’s Frankish son-in-law and Pippin’s son (Wood 1994, 267–72). At that point, we are told that Radbod’s re-occupation of Frisia Citerior involved the destruction of recently established churches there and the refurbishment of pagan shrines and centres (Willibald, Vita Bonifatii, ch. IV). As it was on Helgoland, in this context, historically and territorially, the pagan religious infrastructure clearly became directly linked to and part of the political institution of the Frisian kingdom. Following Radbod’s death in 719, however, Frisia was effectively annexed to the nascent Carolingian Empire (Wood 1994, 293–303). The fact that the pre-Christian traditional religion was no longer closely tied to a serious local and rival monarchical power would plausibly explain why evangelization does not then appear to have been pursued 321

Z01 Hines Book B.indb 321

07/05/2021 13:30

Frisians of the Early Middle Ages with immediate and intense vigour in Frisia. To begin with, certainly, for a three-year period, Willibrord and Wynfrith/Boniface do appear to have made determined efforts (Vita Bonifatii, ch. 5; see Wood 2001, 86–8 and refs.), and the preaching of the gospel plainly did not simply come to an end after the early 720s. Utrecht, however, seems to have functioned primarily as the middle of a sphere of influence for Willibrord which extended from the coast in the west along the Rhine to the east rather than as the southern gateway into his Frisian parochia. Hence the story of his actions on Walcheren, and indeed his particularly close association with the monastery at Echternach (Wood 2001, 81–3, 89–91). Boniface then spent the best part of three decades building up the organization of the Church and Christian observance in the Hesse and Thüringian diocese, before, in the early 750s, as a very old man he returned to what we are told had been his youthful ambition to be, in effect, the apostle to the Frisians. It is a familiar view, that in his great old age he actively sought the martyrdom he suffered at Dokkum, be that in a Christianization of the heroic ideal (Mayr-Harting 1991, 226–8) or in emulation of the Roman martyr whose name he took in place of Wynfrith on consecration as bishop – or, of course, both of these. There is some evidence to suggest that Dokkum was an established assembly or central place in the northern coastlands (Nicolay 2014, 41, 81–2); by holding a confirmation ceremony there, Boniface was ostentatiously staking a claim to the primacy of the Christian order and the Church in the formal life of the territory, and perhaps quite consciously acting provocatively. Willibald’s Vita Bonifatii claims that he was taken unawares by the armed pagan band (hostiumque ingens … bibrantibus armis, astata ac scutata … multitudo: ‘a huge multitude of enemies, equipped with spears and shields, brandishing their weapons’) which rushed his camp. This turn of events does not, however, sound anything like a chance crossing of paths with a vagrant mob of bandits or outlaws (cf. Wood 2001, 60). The relics and books that the Christian party had with them were seized, reportedly mutilated, and then scattered and hidden away. Some were subsequently retrieved, and venerated all the more for the dramatic events they were witness to. Military vengeance was then swiftly organized and taken by the Christian neighbours of the area where this massacre had taken place: in other words, the sequence of events led to an armed showdown between the pagans and the Christians, as could well have been predicted and may well be inferred to have been expected and intended. Boniface may personally have eschewed violence, and sacrificed both himself and his companions as martyrs. He cannot be absolved of the realistic suspicion that he both foresaw and desired the full consequences of his actions. Following Boniface’s death, his loyal follower Gregory (who was evidently spared from martyrdom at Dokkum) was given the responsibility of managing the further conversion of the Frisians from Utrecht (Vita Gregorii, ch. 10). In his time Lebuin (Old English Leofwine) came from England, in fact explicitly to pursue mission amongst the Saxons rather than the Frisians, although he established his oratory to begin with at Wilp near Deventer on the River IJssel. The final three decades of the eighth century, and even the first of the ninth, are dominated in terms of religious policy in north-western Europe by Charlemagne’s wars of suppression and forced conversion of the Saxons – which carried an uncompromisingly militant Christianity to the border of Denmark and will thus unquestionably have contributed to the changing attitudes in Scandinavia towards the surrounding world that would issue in the violent raiding culture and expansionism of the Viking Period (Myhre 1993). The effective organization of 322

Z01 Hines Book B.indb 322

07/05/2021 13:30

Religion and Conversion pagan resistance amongst the Saxons must have provided some inspiration and support for Frisian rebels and resistors. As Ian Wood notes, Frisian pagans in Liudger’s Vita Gregorii are portrayed first and foremost as political rebels rather than as demon-worshippers or heretics (Wood 2001, 105). Sometime apparently in the second half of the 770s another Northumbrian missionary appeared in Frisia, Willehad (Wood 2001, 90–1). He went first to Dokkum, where the Frisian Liudger, who had been appointed by the bishop of Cologne to evangelize the Oostergo region, had raised a church (Vita Willehadi, ch. 2). Willehad then proceeded to the Humsterland district in the province of Groningen. Here the hagiographical motif of lots being cast to decide whether he should be put to death for sacrilege is repeated (Vita Willehadi, ch. 3); he then proceeded into Drenthe, before the Carolingian authorities sent him to Wigmodia in north-eastern Frisia (Vita Willehadi, chs. 4–5). Both Liudger and Willehad abandoned their posts in face of the threat posed by Widukind’s rebellion against Charlemagne in Westphalia in 782, and both headed initially for Italy: Liudger to Monte Cassino and Willehad to Rome (Vita Liudgeri, I, 22; Wood 2001, 110; Vita Willehadi, ch. 7). Willehad subsequently found refuge in Echternach (Wood 2001, 81–3, 89–91). Willehad returned to Wigmodia in the mid-780s but in 787 was appointed bishop of the new Saxon see of Bremen (Vita Willehadi, ch. 8); at this point Liudger returned to the Groningen province east of the River Lauwers, from where he completed the Christian cleansing of Helgoland (Vita Liudgeri, I, 22). Liudger too was subsequently moved into Westphalia, to found the monastery of Werden and eventually to become bishop of Münster (Vita Liudgeri, I, 23). A miracle attributed to the canonized St Liudger in Frisia some time in the late eighth century can be considered a concatenation of standard hagiographical and indeed wider literary topoi but may still cast considerable light on the complex cultural situation in conversion-period Frisia. Bernlef is introduced to the story as a blind Frisian poet, accustomed to sing songs celebrating antiquorum actus regumque certamina, ‘the deeds of the ancients and the battles of kings’ (Vita Liudgeri, I, 25–6). Liudger cures him by converting him and getting him to repent of his former life; he is then taught the Psalms rather than such traditional, vernacular heroic verse that had previously been his métier. We have no surviving ancient Frisian heroic poetry, although the Old English Finnsburh Fragment set in Frisia may serve as a glimpse of what this could have included (Klaeber ed. 2008, 273–90). Interestingly, not least in comparison with the eighth-century Cædmon miracle recorded by Bede (Historia Ecclesiastica, IV, 24), the tale is not presented as an origin legend for a vernacular Christian poetry: nothing is said about Bernlef paraphrasing the psalmos (which it is assumed means the Psalms rather than ‘hymns’ more generally) in Frisian. Bernlef ’s earlier, traditional material, however, was implicitly thoroughly secular; he was a pagan poet in the sense that he was yet to convert, but his poetry was not pagan in any attested way. The story hints, however, how far, at the end of the eighth century, the missionaries were faced with the challenges of inserting Christianity into the culture and mores of a secular order and its social interests and power-structures as much as those of combating militant paganism.

Religion in Early-medieval Frisia So far, so predictable, one might suggest. From the late seventh century to the early ninth, the history of religion in Frisia can be understood in primarily political terms, 323

Z01 Hines Book B.indb 323

07/05/2021 13:30

Frisians of the Early Middle Ages as a facet of if not even as a proxy for the processes of competition for social power, and the attempt to constitute a well-ordered, acquiescent and productive state. The period in which the traditional Germanic religious elements inherited and adapted by the Early-medieval Frisians were opposed to and in conflict with the Roman Christian Church – mediated in practice through its Frankish and Anglo-Saxon provinces – was of course one in which matters of religious allegiance were of especial historical significance, as well as being one in which religious features are therefore particularly likely to be salient in the available sources of evidence. That does not, of course, mean that what we see of religion in such circumstances can be taken as entirely representative of the religious traditions concerned in all and any circumstances. Although the story, in the most general terms, is one of the progress of Christianity and the victory of the Church at the expense of Frisian ‘paganism’, the disappearance of a strong Frisian kingship quite early in the eighth century appears in practical terms to have diminished the intensity of the drive towards comprehensive conversion as would be reflected in complete, enforced conformity of behaviour. In consequence, a surprising degree of space was left for a heterodox religious spectrum to survive. We may take the continuation of cremation burial as a token – a funerary practice the Church explicitly opposed as anti-Christian. The practice is less common in the eighth century than previously, but still running at a visible frequency, and cremation burials are known that are datable into the ninth century (Knol 1993, 167; Lippok 2020). The Viking Period of the ninth and tenth centuries is widely understood to have slowed, if not halted and in some cases even reversed, the process of Christianization in Frisia. As a result, a ‘second Christianization’ was needed in the eleventh century; which, as was typical over much of Europe at the time, saw the proliferation of local parish churches in a blanket system of pastoral provision and control (Schmidt 1987). It may be argued, at the same time, that the case of the Early-medieval Frisians is a particularly good one for demonstrating that this sort of ‘external’ history of religion, even though it illustrates unambiguously how substantially the religious dimension of social life can be affected by contextual crises and dramatic, intrusive, essentially political, interference, this is far from all there is to consider when seeking to understand religion as a major component of the culture of a historical people. We can return to Anatoly Liberman’s recent book for what I find a reductive proposition about traditional Germanic religion, which seems to characterize it as utterly primitive and effectively mindless: Throughout his career as a historian of religion, De Vries remained true to the idea that Óðinn was, first and foremost, a god of intellectual (spiritual) power. However, no primitive deity had such a lofty function. (Liberman 2016, 36; my emphasis)

The story of Radbod’s last-minute reconsideration in Vita Vulfhramni is not presented in such a way as to construct the pagan king as a deeply philosophical individual. His chain of thought and reasoning comes across essentially as an emotional predisposition which found utterance in a sequence of a simple question, decision and justification. Nonetheless, in sources which scarcely had it as their remit to build up the character of Radbod as a neo-pagan paragon of prudence, care and reason, this is not the only episode that portrays him as a cautious and thoughtful judge of situations and options. He thus presents a figure of a rational, and even a just pagan (for which there was 324

Z01 Hines Book B.indb 324

07/05/2021 13:30

Religion and Conversion an exemplary model in traditions current from the seventh century concerning the second-century Roman emperor Trajan: Trumbower 2001, 141–53). In contrasting him with Ongendus king of the Danes, as above, we may only be comparing two stock figures of literary composition, but at the very least the difference between them points to a contemporary acknowledgment of the fact that there was a range of forms and realizations of paganism to be dealt with. That pagans could and should be reasoned with at a relatively sophisticated level is made perfectly clear in a letter of Bishop Daniel of Exeter to Boniface, dated AD 723–4, just after Boniface’s period of mission in Frisia with Willibrord and at the outset of his work in Hesse (Tangl ed. 1916, no. 23). The essence of Daniel’s recommendations is that the pagans should be brought to reflect upon and understand the different concepts of eternity and temporality, with the True God existing beyond and before the genealogical structure within which their own gods and goddesses were attributed special histories: Cum vero initium habere deis utpote alios ab aliis generatos coacti didicerint, utrum initium habere hunc mundum an sine initio semper exstitisse arbitrentur. Si initium habuit, quis hunc creavit? Cum procul dubio ante constitutionem saeculi nullatenus genitis diis inveniunt subsistendi vel habitendi locum; mundum enim non hanc visibilem tantum cęlum et terram, sed cuncta etiam extanta locorum spatia, quę ipsi quoque pagani suis imaginare cogitationibus possunt, dico. When, indeed, they shall have confessed that their gods had a beginning, since they were begotten as individuals by others, they should consider whether the world had a beginning or has always been in existence with no beginning. If it did have a beginning, who created it? For there is no doubt that, before the cosmos was created, in no way could they find a place in which these created gods could subsist or dwell. By ‘the world’ I speak not of this visible heaven and earth but the whole extensive range of spaces which even these pagans are able to conceptualize in their thoughts.

Boniface, conversely, is better remembered for his violent confrontations, at Gaesmere and at Dokkum, than for anything resembling the calm and patient strategy advocated by Daniel (who was also quite happy to recommend exploiting the temptation of the economic and political advantages enjoyed in Christian Europe). The institutionalization of the Church, with centres of worship and learning and a full-time clericy of monks and nuns, priests and deacons, obviously provided conditions in which sophisticated contemplation and the elaboration of doctrine to reinforce the faith could flourish. The same is true of Judaism and Islam and of major religious traditions in Asia by that time. In this respect, the traditional religions of the agrarian Germanic, Slavic and Baltic populations of Europe would have been at the disadvantage of not having the same facilities for intellectual development, although we should not merely assume that such capacity was totally lacking in those traditions. Throughout that range we can also see how the close approach and intrusion of Christianity widely stimulated adaptive but reactionary borrowings from and imitations of Church institutions. The degree of institutionalization we can detect in Frisian paganism by the early eighth century would have created some space for some degree of reflection of this kind and for the definition of a body of lore. Such claims may appear essentially hypothetical rather than readily substantiated. The question of whether an established pagan infrastructure existed in any significant 325

Z01 Hines Book B.indb 325

07/05/2021 13:30

Frisians of the Early Middle Ages form outside of a short period of direct royal support that can be inferred for the period 714–719 is a pertinent one. However, one relevant aspect of traditional Frisian religious practice which we can draw particular attention to is the recurrent reference to the casting of lots as a judicial process. As noted, this appears in the judgment of Willibrord and his companions after their desecration of the sacred site on Fositeland, and in the case of Willehad in Humsterland several generations later. It is transferred into a carefully defined Christian process in Lex Frisionum Titulus XIIII (De homine in turba occiso: ‘If a man is killed in the course of a riot’). The Christian version (which is specified for the region between the Lauwers and the Vlie) is explicit that God, through the procedure, reveals who is guilty and who innocent: Deus exorandus, si illi septem, qui de homicidio commisso iuraverunt, verum iurassent, evidenti signo ostendat, ‘God being prayed that He will show by a clear sign if these seven men who have sworn oaths about the homicide committed had declared the truth.’ We cannot suppose that the antecedent pre-Christian procedures were viewed and accepted in the prior context purely as an arbitrary method of selecting a scapegoat: they must, therefore, presuppose a sense of justice, and of a divine power as law-giver and judge, whose procedures were to be treated and observed with great respect and obedience. In that light, the legend of the Christian origin of the Frisian Law as given to the twelve law-speakers may also be regarded as rather more than merely a pious imposition upon a fantastic tale. Truly remarkable and challenging in the context of a desire to produce a constructive interpretation of Lex Frisionum and its complex religious background are the references to the fanum – which must imply a non-Christian shrine or holy place – and the threat of it being broken into or broken up (the verb used is effringere). The first reference is in Titulus V of Lex Frisionum, De hominibus qui sine compositione occidi possunt, ‘Of people who can be killed without liability for compensation’, where the list of outlaws includes qui fanum effregit, ‘one who has broken into a shrine’, alongside those killed in duels or battle, adulterers, and thieves and arsonists caught in the act, besides neonates killed by infanticide. Much fuller is the final clause (Titulus XI) of the Additio sapientum, De honore templorum, ‘On the honour due to temples’: Qui fanum effregerit, et ibi aliquid de sacris tulerit, ducitur ad mare, et in sabulo, quod accessus maris operire solet, finduntur aures eius, et castratur, et immolator Diis quorum templa violavit. Whoever shall have broken into a shrine and have taken something from what is sacred there, he should be led to the sea, and in the sand which the tide of the sea is going to cover his ears will be cut and he shall be castrated and sacrificed to the gods whose temples he has violated.

Some commentators have been happy to accept these clauses as pure pagan survivals: elements of earlier law that were preserved rather than being edited out of the otherwise undoubtedly Christian Lex Frisionum. It is perhaps easier, although possibly over-simple, to regard them alternatively as deliberately placed reminders, plausibly even fictively produced as such, of a different though not remote pagan past. The parallelism in wording (qui fanum effregit/qui fanum effregerit) suggests that the additional Titulus XI is a deliberate expansion of the terse reference in Titulus V, but it could still either quote a law that had once held in openly pagan contexts or imaginatively create such a law. We should surely assume these provisions to appear in the law-code by design rather 326

Z01 Hines Book B.indb 326

07/05/2021 13:30

Religion and Conversion than by accident. In that case, they might represent a truly extraordinary tolerance of pagan worship and holy sites in conditions of continuing deep heterodoxy; it is also interesting, then, that the legal provisions acknowledge that what we must assume to be militant Christians are likely to attack and destroy shrines, but not vice versa. Alternatively, and perhaps most credibly, the preservation of these as in reality obsolete clauses acknowledged the roots of the Law of the Frisians in a different religious tradition to that of Christian Europe and the Church. At the same time, it acknowledged yet again the institutional and constitutional basis of that prior religious order. Consequently, while the parallelism of what appear to be unexpected laws protecting the survival of pre-Christian religious practices, particularly concerning sacrificial livestock, in Pactus Legis Salicae is of course striking and important, we may resist the temptation or encouragements to seek to combine those with the evidence of Lex Frisionum under a common explanation. Altogether, then, we can detect that there was much in the early Frisians’ handling of religion that was pragmatic. Considerations of political expediency alone, however, fall far short of a satisfactory account of the range, changes and forms of behaviour that embodied the religion of this people. It can also be tempting from a modern historical standpoint to think of Frisia itself as a backwater – whatever the impact of the Frisians in North Sea navigation – and so of the settled population there as relatively backward in matters of religion; thus ending up neither pagan nor very convincingly Christian, left only with the absence of regular Christian customs and accusations of the suspicion of the widespread practice of sorcery. From around the middle of the eighth century, first the Saxons and then the Vikings, subsequently also Slavonic eastern Europe and the Balts, drew the main force of either invasive or defensive Christian response to neighbouring unconverted populations, and, in relative terms, Frisia was neglected by the Church. Our sources thus are limited, but enough to reveal that there was there in fact a constructive and genuinely interesting example of how West Germanic religious traditions could develop.

Discussion FLIERMAN  With regard to the special clause in Lex Frisionum, there are also Christians in the Saxon context who admit that non-Christians had the best laws (Translatio S. Alexandri, ch. 2). The notion that destroying temples should be punished harshly is not that odd. HINES  Nevertheless the missionary saints’ lives are very conspicuously focussed upon the destruction of temples and idols, which are reported with approval. The vocabulary was undoubtedly available for Lex Frisionum to have sought to protect churches rather than any fanum or fana. The term does signify a non-Christian place of worship. The point that religion per se is to be respected is an important one, but they could have been very general if they had wanted to, whereas in the context this appears to refer very specifically to the immediate, traditional religious infrastructure. WOOD  When talking about the Merovingians and their policy or practices, it depends on the period. They were very keen on mission up to about 640, but with the decline in royal power there is a general (but not complete) loss of interest in mission, until the Anglo-Saxon missionaries arrive. 327

Z01 Hines Book B.indb 327

07/05/2021 13:30

Frisians of the Early Middle Ages HINES  Was that particularly in southern Germany? WOOD  No; Dagobert champions mission directly across the Rhine in Flanders etc. You’re right for the late seventh century, but not the early seventh century. Aldgisl and Radbod do not convert: I suspect there’s a generational issue here. There are other cases where rulers do not convert but allow their sons to. It’s a careful ruler’s way of working with their own generation and allowing the next one to develop in a different direction. There’s the whole issue of deconstructing the hagiography. Another thing that went through my mind is whether Lex Frisionum had Lex Salica in mind, where again you have the wise men, and again one clause on paganism (Pactus legis Salicae, II.16). Is that being used as a model? HINES  I’d be happy with that, although do not see it as the complete and only explanation we need. As a model, however, I can see how it could be significant. If we also use the situation in Iceland several centuries later for comparison, where the formal decision to adopt Christianity at the general assembly (Alþingi) was linked to limited but explicit tolerance of continuing pre-Christian practices (Íslendingabók, ch. 7), we would have both a precedent and a much later example. In seventh-century Anglo-Saxon England there is legislation against pagan worship in the Kentish laws (Bede, HE, III.8), but it would fit with the pragmatic instructions to work within the existing framework of sites and practices. IJSSENNAGGER-VAN DER PLUIJM  In the case of the story of Liudger and Bernlef, and the transition from heroic vernacular verse to psalms – around AD 800 there is a desire to present a ‘before-and-after’ situation; after this watershed, the culture is more Continental. Is there anything in the facts of how Radbod is portrayed as a pagan in successive sources that works with this view? HINES  I think firstly that we have to be very cautious in saying that Bernlef is presented as ‘pagan’. He isn’t presented as a religious poet, singing of the gods; he was talking about kings and battles. What represents paganism in his story is the fact that he is portrayed as blind. He doesn’t seem very religious at all. He was taught the psalms, but the story does not parallel that of Cædmon in seventhcentury Northumbria who then versified the Biblical and Christian texts in Old English (Frantzen and Hines eds. 2007). It may be implicit, but it isn’t stated. IJSSENNAGGER-VAN DER PLUIJM  Is a cultural contrast being emphasized? HINES  Yes, Bernlef represents the old culture which is going through a process of conversion. It may also be part of a wider cultural shift attributed to around the year 800. This is also the point at which the successive Saxon Capitularies insist that Saxonia is now Christian, whether people like it or not (Hines 2003). The border of Christendom has been pushed to the Elbe, if not to the Dannevirke. In respect of the sources recording the story of Bernlef I have seen different dates attributed, but I don’t know which is correct. WOOD  With regard to Liudger’s activity, all you can say is it must pre-date 793, when he moves to Westphalia. NIJDAM  You say Bernlef is not said to have recounted pagan songs. Is he typical in that respect? Do other poets sing songs of the gods, in Anglo-Saxon England? HINES  We don’t have a body of mythological poetry in Old English as we do in Old Norse, apart from small fragments, for instance amongst the ‘charms’, where there are a few references to Woden/the Ese (= ON Æsir) etc. It is an important question whether those are genuine survivals from much earlier or buried strata, 328

Z01 Hines Book B.indb 328

07/05/2021 13:30

Religion and Conversion or no more than new derivations from what has become – in religious terms – ‘merely’ folklore. We do seem to see quite a lot of evidence for coinages of the latter character in place-names for major topographical features. But outside of Scandinavia, there really is next to no survival of pre-Christian religious source material of this kind. That cannot mean no such material ever existed. NIJDAM  The other side of the coin is that, in respect of recounting the deeds of the old kings, we can look at the story of Radbod’s baptism, and his concern for his ancestors expressed in that. That would mean that a consciousness of the ancestors was tied to the old religion. HINES  I could not agree more. I have tried to keep a clear view of the separation between the secular and the religious, but I recognize that that is a modern analytical approach rather than being at all likely to have been how things were perceived then. It also comes back to this father-and-son issue: if Radbod was Aldgisl’s successor or son, we could postulate that Aldgisl thought Radbod would convert, and that would leave Radbod in an even more havering or problematically balanced and complicated position. The sources certainly make it look as if Radbod was seriously considering conversion for twenty years or more. Perhaps the political developments involving Charles Martel as a rival eventually pushed him in the other direction. VERSLOOT  The motif of blindness has Classical examples; the parallel in Iceland is indeed intriguing. When it comes to the temples though, that won’t work. Paganism is allowed but must be private. You also say you don’t want to see paganism as a sort of superstition. But the descriptions, even in Icelandic literature (where the sources are much later), imply that religion did not seem to affect people very much; it was a matter of keeping the gods calm; Fate overrides everything and the gods are in effect the oldest ancestors, so that heroic songs about ancestors can indeed be relatively religious in a way. In those terms, preChristian belief involves a very different way of looking at religion from the monotheistic point-of-view. It did not require any specific behaviour, and lacked moral implications. HINES  My response to those points is not really to disagree with any of them, but rather to take the view that such characterizations of religion plot out parameters of a diverse field of experience rather than compelling us to decide: it is or was like this or like that. On the point of the temples, though: as I noted, the site of Hofstaðir now suggests that it continued as a cult site well into the eleventh century. It is indoors, but within a colossal longhouse that has some form of regional pre-eminence. I would also turn to the skaldic poetry of the tenth to early eleventh centuries: if you look at Egil’s Sonatorrek, that does not represent blind fatalism. There are also the poets of Óláfr Tryggavson facing conversion, and their poetic diction represents rather more of a thoughtful religion (Clunies Ross 2005, 114–40). I think therefore that there is scope for these dimensions not just to be allowed for but respected. At the same time, though, it is the immediately immanent presence of Christianity that makes these religions visible to us, just as the presence of an anthropologist affects everything he or she may observe in terms of the behaviour and response of a population or culture being studied. Christianity, of course, can also be blind superstition. But I do very much like Daniel’s letter to Boniface, asking him to engage with the pagans intellectually, 329

Z01 Hines Book B.indb 329

07/05/2021 13:30

Frisians of the Early Middle Ages while at the same time what he says undeniably endorses your point that the pagan conception of religion was contained entirely within a temporal framework. There is a middle ground there; or, better, a house with many rooms (John 14.2). KNOL  One comment on the cremations. It is well established that the conversion passed from west to east in Frisia; Holland first; then Groningen; and Ostfriesland last. The cremation urns from eastern sites are later than those in Westergo and Oostergo, while those in Holland are earlier still. While the evidence is limited, it does seem to suggest that this west–east process is reflected in the disappearance of the cremation rite. The question of whether or not burials show Christianity is a different and difficult issue. Inhumation with grave goods was allowed by Christianity to start with. But cremation is definitely non-Christian, and we can see the change in that respect across this area. At Osterbeintum and Hogebeintum in particular, graves with grave goods are not found from after around AD 725. In Groningen the cemeteries do still have grave goods and in Dunum they are even younger, but Dunum is not within the diocese of Liudger: it belongs to another conversion history. I presume that under some of the oldest churches we may have had Christian burials, but those will be nearly impossible to find. If we went to dig as in the Martinikerk in Groningen we would find very little: just a couple of postholes, and very few burials. The layers are very much disturbed by later activity. One more remark on Radbod, who gave his daughter in marriage to a Christian king’s son. There are many references to female saints who are given by Christian kings to pagans, and who refuse to marry and so are martyred. In the transitional period it appears that Christian kings would give their daughters in this way to make alliances. There is the story of St Wilgefortis, daughter of the king of Portugal, who grew a beard to put off her intended Muslim husband-to-be, the Saracen king of Sicily (Lacey 1982). There is in fact a medical condition involving anorexia nervosa that could explain this as a realistic physiological response. MOL  This may be old-fashioned history, trying primarily to establish the facts. In the Pippin-Radbod-Willibrord triangle, what happened to the latter in the geopolitical context? He started at Radbod’s court in Utrecht in 690; then Pippin of Herstal seizes Utrecht in the mid-690s. What happened to Radbod and Willibrord? We know Willibrord went to Rome. He is appointed archbishop of the Frisians, and established in Utrecht, by 697/698; some suggest it was already a cathedral see but there was not yet any diocese of Utrecht: that came only in the eighth century. Was he really active there? I doubt it. Pippin created a base for him. Willibrord then gets a huge donation, and exactly when he should have been in Frisia, he is converting the Toxandria area … not in Utrecht at all. Only after 719 did it become possible for a conversion of Frisia Citerior to take place. I see a sequence here, and the question is whether or not Pippin gave Radbod room within which he could exert authority: in fief as it were. We have to see Willibrord as a real person moving around and acting under the lordship of the Austrasian majordomo. HINES  I very much agree, although your hypothesis concerning the relationship between Pippin and Radbod over Frisia Citerior and what Pippin’s objectives or strategy there were go beyond what I had thought of. I strongly agree that Willibrord’s sphere of influence runs primarily east to west, in the river area, rather than south to north. Interestingly, that makes the three years in which Willibrord 330

Z01 Hines Book B.indb 330

07/05/2021 13:30

Religion and Conversion and Boniface were reportedly operating together after 719 a major change in activity. In respect of the establishment of Utrecht and its association with the bishop of Cologne, I would add that the status of Utrecht as a former Roman centre, with supposedly an old Roman church there, is significant, even if it was not really established as an effective see until as late as 719. In my view it could well have been earmarked as a centre of the church, for intended missionary activity. I am clear, too, that the activities of those who were responsible for consolidating and developing the Church had to be focussed on the fact that the Church’s position in lands south of the Rhine was also still weak, even if politically those areas were part of the Merovingian realm. There was still a lot of work required there. MOL  Yes, Utrecht may have been earmarked. From a political point of view, in terms of Pippin’s strategy, he must have been ready to accept it could not be developed yet. The fact that Willibrord was archbishop of the Frisians rather than of Utrecht may be significant. Only later could Charles Martel give him considerable lands; twenty-five years or so further on. Pippin was not able to do that, or to create a firm infrastructure. What he does, in those circumstances, is to provide Willibrord with Echternach, Susteren, and an enormous property so that he can function as an archbishop. HINES  I’m very happy with that reconstruction. WOOD  I agree with Hans over the non-geographical nature of Willibrord’s power in Utrecht. I have severe doubts over the power of the Pippinids. One of the problems is that we understand them in terms of historical sources written after 751. There are even problems with the reconstruction of the family. I would possibly be even more sceptical about Pippin II’s power and his ability to control anything in the Frisian region. The Carolingian retrospective viewpoint affects a lot. NIJDAM  What you see suggests that what Pippin tries to do, in creating a sphere of influence in the river area, is to come to terms with the powerful man who is there. This fits with how Willibrord gains his financial basis in the Austrasian heartland. NIEUWHOF  I tend to look at religion from a cognitive perspective, as a set of beliefs and practices associated with the supernatural. This is not the same as a belief in the gods; it comprises both historical perspectives and more individual ones, relating to families, households and villages. All of those entities have their own and different levels of dealing with religion. What a person does is as interesting as whether he or she believes in Odin or not. In that respect, archaeology is a better starting point for the study of the phenomenon. We should also consider the amulets found in graves, and special deposits in houses, as aspects of religion. The evidence for temples is very sketchy. We know something about Helgoland. Were others found in Frisian lands within the former Roman Empire? There are references to tidal sites for rituals, as you noted (Vita Vulframni, ch. 8; cf. also Vita Liudgeri I, chs 6–7, for the use of drowning in an accepted procedure for infanticide). HINES  To take the latter question first: views have changed greatly over time. Twenty years or so ago, I would have been one of the many who would quite dogmatically have said that we have no evidence for Germanic ‘pagan’ temples of this period. The terms used in the sources can, of course, be misinterpreted, and give an impression that does not relate to the factual situation on the ground, and in the 331

Z01 Hines Book B.indb 331

07/05/2021 13:30

Frisians of the Early Middle Ages hagiographical sources Old Testament images of the gentiles will also affect the representation of paganism. There is, however, plenty of archaeological evidence for idols in the Germanic world, even if not precisely from the places and dates we’d want to see them in terms of understanding the pre-Christian religious traditions of the Anglo-Saxons, Old Saxons and their Frisian contemporaries (Van der Sanden and Capelle 2002). On the other hand there is the consistent reference in the literary sources to the treasure associated with sacred sites, and it being sent down to Rome, which seems to imply the existence of holy places, groves, trees etc. The references to temples seem to me to go beyond purely topographical features, although we do not know what any such cult buildings will have looked like. As I noted, we do have some archaeological examples now in Scandinavia: at Uppåkra, Tissø and more. These sites have buildings, sanctuary spaces, and ritual activities; likewise Yeavering in early seventh-century Northumberland. And there is the place-name evidence too. I would judge that this material is cumulatively more than sufficient in the way of hints that such sites existed, but the details are very elusive. I absolutely take the point about individual or socially highly localized practices of religion, and the valid criticism that I am focussing selectively on group religion: yet more rooms that I have to allow for in the house of this ‘religion’. NIJDAM  I agree with Annet, and I think you are overharsh on historians such as Rudi Künzel (1992). Returning to Christianity as a sort of deep experiment, however, and the reference to Bishop Daniel’s letter, and further to Radbod’s dream. There are Old Icelandic sources on Valhǫll and the idea of an afterlife: is this a later development? It matches Radbod’s dream of a beautiful golden palace. A pagan development of an idea of an afterlife could have been a response to what Christians were telling them. HINES  It could indeed, and a common interpretation of the image of Valhǫll is that it was an adaptive reaction to what Christianity was offering (Schiødt 2007). So far as I have been able to discover, afterlife beliefs are not universal but are very common (see Davies 2017, 137–42, on contemporary Judaism and Islam). What also appears to be the case is that societies of a certain level of development tend to develop the conception of an individual afterlife for their highest elite strata to begin with, and the idea of access to that afterlife gradually spreads outwards and downwards in society. By the Roman Period, mystical initiates such as the Mithraists adopted the beliefs. It was quite a dramatic religious reform introduced by Christianity (in keeping with other branches of Judaism at the time) to offer a universal afterlife. MAJCHCZACK  I should like to throw some light on the situation of conversion in North Frisia. We really don’t have any written sources, but we can agree that it happened much later than in the Frisian core area. The archaeological sources show that the tradition of cremation continued until the mid-ninth century. After that there is very little archaeological evidence to shed light on the progress of Christianity. There is no early evidence of church buildings before the late twelfth century. We remain very much in the dark about the goings on in North Frisia and how they relate to the southern regions.

332

Z01 Hines Book B.indb 332

07/05/2021 13:30

Religion and Conversion HINES   It may be useful to add to that the history of the church in Ribe can now be traced back to the mid-ninth century, with a Christian churchyard from the time of Ansgar onwards (Søvsø 2020, 180–204). MAJCHCZACK  A connection with Ansgar’s mission in Ribe is quite possible, analogous to the strong trade ties between the North Frisian islands and the market site at Ribe from the eighth century onwards. Looking at the material culture, I believe there could be faint traces of Christianity represented by brooches even from the late ninth century on the islands. There are a few finds of enamelled disc brooches with the cross motif, for example from Goting/Föhr. NICOLAY  This symbolic material starts to appear in the late eighth century. But it’s hard to know what it actually meant to people. KNOL  A thought about the temples and the treasures. You say the stories are consistent. That would mean real gold and silver was coming to the Church. What about the seventh-century hoards that we have? Following the line of thinking of Heidinga (1990, 14–19), I have suggested that these could be votive offerings as well (Knol 1993, 235–6). There are several in the archaeological records. They were treasure hoards that were not meant to be retrieved; but they would not immediately have been forgotten, and some tradition concerning them would have been passed down. They could then have been disinterred when conversion was taking place, and those finds could in turn have been represented in the hagiographical sources as the ‘temples’. It is unprovable, but this could explain why we have so few treasure hoards of the eighth century. VERSLOOT  Or were they buried to keep the treasure out of the hands of the Christians? NICOLAY  The chronology isn’t right; the hoards go back to around the year 600. KNOL  But if you retrieve a lot of these treasures then of course you will have a gap now. FLIERMAN  I was also going to emphasize the link between fanum and treasure: the first thing Charlemagne does is to destroy the Irminsul and take away the treasure. I agree with Knol in this respect; the link is a strong one. DE LANGEN  Did the Church neglect the Frisian area? I don’t think so; as soon as Radbod is gone the Church moves in, and churches are founded. These early churches were rebuilt in the late tenth century; during which period many more (daughter) churches arose as well. HINES  In suggesting that, I was thinking particularly of the contrast between the trajectory that was followed in the areas of Frisia and Saxonia, and then in Scandinavia. It is a relative matter, I concede, but I would suggest less of a drive in the case of Frisia. I am also particularly struck by the contrast suggested by the foundation of the truly great religious houses in Saxonia. I agree I should pay more attention to geographical conditions, but wonder if that would entirely work in relation to Holland and West Frisia. I accept I may have overstated the point, but there remains a point there to make, I believe.

333

Z01 Hines Book B.indb 333

07/05/2021 13:30

Frisians of the Early Middle Ages

References Primary sources Alcuin, Vita Willibrordi. Ed. W. Levison 1920, MGH Scriptores, SS Rer. Merov. 7 (Hannover). — Vita Willibrordi. Ed. A. Poncelet 1910, Acta Sanctorum 3 (Brussels), 435–57. — Vita Willibrordi. Trans. C. H. Talbot 1954, The Anglo-Saxon Missionaries in Germany (New York), 1–22. Altfrid, Vita Liudgeri. Ed. W. Diekamp 1881, Die Vitae Sancti Liudgeri, Geschichtsquellen des Bistums Münster 4 (Münster). Bede, Historia Ecclesiastica Gentis Anglorum. Eds. and trans. B. Colgrave and R. A. B. Mynors 1969, Bede’s Ecclesiastical History of the English People (Oxford). Boniface, Epistolae. Ed. M. Tangl 1916, S. Bonifatii et Lulli Epistolae. MGH Epistolae, Epp. 3 VI (Berlin). — Epistolae. Selection trans. C. H. Talbot 1954, The Anglo-Saxon Missionaries in Germany (New York), 65–149. Eddius Stephanus, Vita Wilfridi. Ed. and trans. B. Colgrave 1927, The Life of Bishop Wilfrid by Eddius Stephanus (Cambridge). Fan tha koningen Karle ande Redbad. Ed. R. Bremmer 2009, An Introduction to Old Frisian (Leiden), XVII, 174–6. Íslendingabók. Ed. Jakob Benediktsson 1958. Íslenzk fornrit 1,1 (Reykjavík). — Trans. Sian Grønlie 2006. Íslendingabók—Kristni saga (London). Lex Frisionum. Eds. Karl August Eckhardt and Albrecht Eckhardt 1982, MGH Leges, Fontes iuris 12 (Hannover). Liudger, Vita Gregorii. Ed. O. Holder-Egger 1887, MGH Scriptores, SS 15 (Hannover). Pactus Legis Salicae. Ed. K. A. Eckhardt 1962, MGH Leges IV,2 (Hannover). Snorri Sturluson, Gylfaginning and Skáldskaparmál. Ed. A. Faulkes 1988–98, Snorri Sturluson: Edda, 3 vols (London). Sonattorek. Ed. and trans. E. O. Gabriel Turville-Petre 1976 (Clarendon Press, Oxford), 24–41. The Fight at Finnsburg. Ed. Fr. Klaeber 2008, Klaeber’s Beowulf, eds. R. D. Fulk, R. E. Bjork and J. D. Niles, 4th ed. (Toronto), 273–90. Translatio s. Alexandri. Ed. B. Krusch 1933, ‘Die Übertragung des H. Alexander von Rom nach Wildeshausen durch den Enkel Widukinds. Das älteste niedersächsische Geschichtsdenkmal’. In Nachrichten von der Akademie der Wissenschaften in Göttingen. Philol.-hist. Klasse 4 (Göttingen), 405–36. Vita altera Bonifatii. Ed. W. Levison, Vitae sancti Bonifatii, MGH Scriptores, SS rer. Germ. 57 (Hannover). Vita Lebuini antiqua. Ed. O. Hofmeister 1934, MGH Scriptores, SS 30, 2 (Leipzig). — Trans. C. H. Talbot 1954, The Anglo-Saxon Missionaries in Germany (New York), 227–34. Vita Liudgeri Secunda. Ed. W. Diekamp 1881, Die Vitae Sancti Liudgeri, Geschichtsquellen des Bistums Münster 4 (Münster). Vita Vulframni. Ed. W. Levison 1910, MGH Scriptores, SS Rer. Merov. 5 (Hannover). Vita Willehadi. Ed. A. Poncelet 1910, Acta Sanctorum 3 (Brussels), 842–6. Willibald, Vita Bonifatii. Ed. W. Levison 1905, Vitae sancti Bonifatii, MGH Scriptores, SS rer. Germ. 57 (Hannover).

334

Z01 Hines Book B.indb 334

07/05/2021 13:30

Religion and Conversion — Vita Bonifatii. Trans. C. H. Talbot 1954, The Anglo-Saxon Missionaries in Germany (New York), 23–62. Secondary sources Andrén, Anders 2014, Tracing Old Norse Cosmography: The World Tree, Middle Earth, and the Sun in Archaeological Perspectives (Lund). Blair, John 2005, The Church in Anglo-Saxon Society (Oxford). Bremmer, Rolf 2009, An Introduction to Old Frisian: History, Grammar, Reader, Glossary (Amsterdam). Campbell, Alastair 1959, Old English Grammar (Oxford). Caciola, Nancy Mandeville 2016, Afterlives: The Return of the Dead in the Middle Ages (Ithaca NY). Caruth, Jo and John Hines forthcoming, The Anglo-Saxon Burial Grounds at RAF Lakenheath, Eriswell, Suffolk: Excavations 1997–2007. East Anglian Archaeology. Clunies Ross, Margaret 2005, A History of Old Norse Poetry and Poetics (Cambridge). Davies, Douglas 2017, Death, Ritual and Belief: The Rhetoric of Funerary Rites, 3rd ed. (London) Dijkstra, Menno F. P. and A. Verhoeven 2012, ‘Heiloo — Zandzoom/Zuiderloo’, De archeologische kroniek van Noord-Holland 2012, 57–62. Frantzen, Allen J. and John Hines (eds) 2007, Cædmon’s Hymn and Material Culture in the World of Bede (Morgantown WV). Geary, Patrick 1994, Living with the Dead in the Middle Ages (Ithaca NY). De Groot, Huib L. 1996, ‘Utrecht and Dorestad: fifteen miles apart, a world of difference’. In Utrecht, Britain and the Continent: Archaeology, Art and Architecture, ed. Elisabeth de Bièvre, British Archaeological Association Conference Transactions 18 (Leeds), 12–21. Heidinga, H. A. 1990, ‘From Kootwijk to Rhenen: in search of the elite of the central Netherlands in the Early Middle Ages’. In Jos C. Bestemann, Jurjen M. Bos and H. A. Heidinga (eds), Medieval Archaeology in the Netherlands: Studies presented to H. H.van Regteren Altena (Assen-Maastricht), 9–40. Higham, Nick 1997, The Convert Kings: Power and Religious Affiliation in Early Anglo-Saxon England (Manchester). Hines, John 1997, ‘Religion – the limits of knowledge’. In The Anglo-Saxons from the Migration Period to the Eighth Century: An Ethnographic Perspective, ed. J. Hines, Studies in Historical Archaeoethnology (Woodbridge), 375–410. — 2003, ‘The conversion of the Old Saxons’. In The Continental Saxons from the Migration Period to the Tenth Century: An Ethnographic Perspective, eds. Dennis H. Green and Frank Siegmund (Woodbridge), 299–328. — 2015a, ‘Book review: Tracing Old Norse Cosmography: The World Tree, Middle Earth, and the Sun in Archaeological Perspectives’, European Journal of Archaeology 18 (2), 369–74. — 2015b, ‘Burial and religion in pre-Christian Anglo-Saxon England’. In Dying Gods: Religious Beliefs in Northern and Eastern Europe in the Time of Christianisation, ed. C. Ruhmann and V. Brieske, Neue Studien zur Sachsenforschung 5 (Hannover), 63–71. — 2017, ‘The Anglo-Frisian question’. In Frisians and their North Sea Neighbours. From the Fifth Century to the Viking Age, eds. J. Hines and N. IJssennagger (Woodbridge), 25–42. Hines, John and Nelleke IJssennagger 2017, ‘The Frisians – who, when, where, why?’. In Frisians and their North Sea Neighbours. From the Fifth Century to the Viking Age, eds. J.

335

Z01 Hines Book B.indb 335

07/05/2021 13:30

Frisians of the Early Middle Ages Hines and N. IJssennagger (Woodbridge), 1–4. Hirst, Sue and Dido Clark 2009, Excavations at Mucking, Vol. 3: The Anglo-Saxon Cemeteries, 2 vols. (London). Hogg, Richard M. 1992, A Grammar of Old English. Part 1: Phonology (Oxford). Jørgensen, Lars 2014, ‘Norse religion and ritual sites in the Scandinavia in the 6th–11th century’. In Northern Worlds – Landscapes, Interaction and Dynamics, ed. H. C. Gulløv (Copenhagen), 129–50. Knol, Egge 1993a, ‘De kerstening van Noord-Nederland in het grafritueel weerspiegeld?’. In Het Tweede Leven van Onze Doden, eds. E. Drent, W. A. M. Hessing and E. Knol (Ammersfoort), 61–74. — 1993b, ‘De Noordnederlandse Kustlanden in de Vroege Middeleeuwen’ (Ph.D. thesis, Vrije Universiteit Amsterdam). Knol, Egge and Nelleke IJssennagger 2017, ‘Paleography and people. Historical Frisians in an archaeological light’. In Frisians and their North Sea Neighbours. From the Fifth Century to the Viking Age, eds. J. Hines and N. IJssennagger (Woodbridge), 5–24. Krogmann, Willy 1967, ‘Die friesische Sage von der Findung des Rechts’, Zeitschrift der Savigny-Stiftung für Rechtsgeschichte, Germanische Abteilung 84, 72–127. — 1971, ‘Altfriesische Literatur’. In Kurzer Grundriß der germanischen Philologie bis 1500. Band 2: Literaturgeschichte, ed. L. E. Schmitt (Berlin), 164–85. Künzel, R., ‘Paganisme, syncrétisme et culture religieuse populaire au Haut Moyen Age: Réflexions de méthode’, Annales: Histoire, Sciences Sociales 47 (1992) 1055–69. Lacey, J. H. 1982, ‘Anorexia nervosa and a bearded female saint’, British Medical Journal 285, 1816–17. Lange, S. 2017, Uit het juiste hout gesneden: Houten gebruiksvoorwerpen uit archeologische context tot 1300 n.Chr., Nederlandse Archeologische Rapporten 54 (Amersfoort). Levison, Wilhelm 1946, England and the Continent in the Eighth Century (Oxford). Liberman, Anatoly 2016, In Prayer and Laughter: Essays on Medieval Scandinavian and Germanic Mythology, Literature, and Culture (Moscow). Lippok, Femke 2020, ‘The pyre and the grave: early medieval cremation burials in the Netherlands, the German Rhineland and Belgium’, World Archaeology 52, 147–62. Lucas, Gavin 2009, Hofstaðir: Excavations of a Viking Age Feasting Hall in North-Eastern Iceland (Reykjavík). Mayr-Harting, Henry, 1991, The Coming of Christianity to Anglo-Saxon England (London). McKitterick, Rosamund 1983, The Frankish Kingdoms under the Carolingians (Harlow). Meens, Rob 1993, ‘Willibrords boeteboek?’, Tijdschrift voor Geschiedenis 106, 163–78. — 2000, ‘Christentum und Heidentum aus der Sticht Willibrords? Überlegugen zum Paenitentiale Oxoniense II’. In L’évangélisation des régions entre Meus et Moselle et la fondation de l’abbaye d’Echternach (Ve–IXe siècle), ed. M. Polfer (Luxembourg), 415–28. — 2015, ‘With one foot on the font: the failed baptism of the Frisian king Radbod and the eighth-century discussion about the fate of unbaptized forefathers’. In Early Medieval Ireland and Europe: Chronology, Contacts, Scholarship, eds. P. Moran and I. Warntjes (Turnhout), 577–96. Myhre, Bjørn 1993, ‘The beginning of the Viking Age – some current archaeological problems’. In Viking Revaluations, eds. A. Faulkes and R. Perkins (London), 182–204. Nicolay, Johan A. W. 2014, The Splendour of Power: Early Medieval Kingship and the Use of Gold and Silver in the Southern North Sea Area (5th to 7th century AD), Groningen Archaeological Studies 28 (Groningen).

336

Z01 Hines Book B.indb 336

07/05/2021 13:30

Religion and Conversion Olsen, Olaf 1966, Hørg, Hov og Kirke: Historiske og Arkæologiske Vikingetidsstudier (Copenhagen). Parsons, David 1996, ‘England and the Low Countries at the time of St Willibrord’. In ­Utrecht, Britain and the Continent: Archaeology, Art and Architecture, ed. Elisabeth de Bièvre, British Archaeological Association Conference Transactions 18 (Leeds), 30–48. Price, Neil 2014, ‘Nine paces from Hel: time and motion in Old Norse ritual performance’, World Archaeology 46, 178–91. Prummel, Wietske 1993, ‘Paarden en honden uit vroeg-middeleeuwse grafvelden’. In Het Tweede Leven van Onze Doden, eds. E. Drent, W. A. M. Hessing and E. Knol (Ammersfoort), 53–60. Quak, Arend 2002, ‘Germaanse sacrale plaatsnamen in de Nederlanden’, Naamkunde 34, 57–71. Schiødt, Jens Peter 2007, ‘Óðinn, warriors, and death’. In Judy Quinn, Kate Heslop and Tarrin Wills (eds), Learning and Understanding in the Old Norse World (Turnhout), 137–51. Schmidt, Heinrich 1971, ‘Friesische Freiheitsüberlieferungen im hohen Mittelalter’. In Festschrift für Hermann Hempel zum 70. Geburtstag, Vol. 3 (Göttingen), 518–45. — 1987, ‘Kirchenbau und „zweite Christianisierung“ im friesisch-sächsischen Küstengebiet während des hohen Mittelalters’, Niedersächsisches Jahrbuch für Landesgeschichte 59, 63–93. Schrijver, Peter 2017, ‘Frisian between the Roman and the Early Medieval Periods: Language Contact, Celts and Romans. In Frisians and their North Sea Neighbours. From the Fifth Century to the Viking Age, eds. J. Hines and N. IJssennagger (Woodbridge), 43–52. Semple, Sarah 2007, ‘Defining the OE hearg: a preliminary archaeology and topographic examination of hearg place-names and their hinterlands’, Early Medieval Europe 15, 364–85. Siebs, Theodor 1909, Helgoland und seine Sprache (Cuxhaven–Helgoland). Søvsø, Morten 2020, Ribe 700–1050: From Emporium to Civitas in Southern Scandinavia (Aarhus). Talbot, Charles H. 1954, The Anglo-Saxon Missionaries in Germany (New York). Trumbower, Jeffrey A. 2001, Rescue for the Dead: the Posthumous Salvation of Non-Christians in Early Christianity (Oxford). Tylor, Edward B. 1871, Primitive Culture: Researches into the Development of Mythology, Philosophy, Religion, Art and Custom (London). Van der Sanden, W. and T. Cappelle 2002, Götter Götzen Holzmenschen (Oldenburg). Van Loon, Jozef 2017, Lo, Donk, Horst: Taalkunde als Sleutel tot de Vroege Middeleeuwen (Gent). Wood, Ian 1994, The Merovingian Kingdoms 450–751 (Harlow). — 2001, The Missionary Life: Saints and the Evangelisation of Europe 400–1050 (Harlow).

337

Z01 Hines Book B.indb 337

07/05/2021 13:30

Z01 Hines Book B.indb 338

07/05/2021 13:30

• 12 • TRACES OF A NORTH SEA GERMANIC IDIOM IN THE FIFTH–SEVENTH CENTURIES AD Arjen P. Versloot

T

he concept of ‘Anglo-Frisian’ as a common branch on the Germanic language tree was coined in the nineteenth century and appears prominently in the title of Theodor Siebs’s (1889) dissertation Zur Geschichte der Englisch-friesischen Sprache (see also Nielsen 1985; Stiles 1995; Hines 2017). Nevertheless, people were already aware of the similarity between Frisian and English much earlier, perhaps even as early as the time of the Anglo-Saxon mission in Frisia in the seventh and eighth centuries. In the late sixteenth century, the language of parts of North-Holland is described as Half Vries, half Engels by woorden ghebroken [‘Half Frisian, half English with broken words’] (Valcooch 1599, fol. A7r), Franciscus Junius, who was an early Old English philologist, visited Friesland in 1646–8 to learn Frisian and study Old Frisian. The Tegenwoordige Staat der Vereenigde Nederlanden [‘The current state of the United Netherlands’] (Schouten et al. 1785), describing Friesland, contains a West Frisian text with a parallel English translation to exhibit the similarities between the languages. Examples of English-Frisian parallels are easy to find, well known and seemingly convincing. Some older place-names and dialectal words in traditional western Dutch dialects attest to the earlier existence of some form of Anglo-Frisian far beyond the present-day western border of the province of Fryslân (Tab. 12.1). Table 12.1  Examples of similarities between North Sea Germanic language varieties. German

Dutch

English

Frisian (West)

Western Dutch dialects

Gans

gans

goose

goes

Goes (place-name)

brachte

bracht

brought

brocht

brocht

Brücke

brug

bridge

brêge, brich

breg, brigge

ab

af

of

ôf

of

This paper addresses the following issues. The first part is concerned with the concept of Anglo-Frisian and how it has been critiqued, and subsequently presents a new understanding of the concept. This is made possible by various new results from historical linguistics, including a new interpretation of the early phonological history of English (Versloot forthcoming a; forthcoming b); a close scrutiny of the North Sea Germanic traces in western Dutch (De Vaan 2017); fuller understanding of the early 339

Z01 Hines Book B.indb 339

07/05/2021 13:30

Frisians of the Early Middle Ages history of Frisian (Versloot 2014a; Versloot and de Vaan, in prep.); and better access to the dialectal nuances of Old Saxon (Tiefenbach 2010; Versloot and Adamczyk 2017). The second part aims at establishing a link between the linguistic and the archaeological evidence. Special emphasis is put on the spatial organization of the linguistic traces and the actual speakers of these idioms by an analysis of the geographical patterns in the distribution of runic inscriptions, place-names commonly linked to the migration of populations in the Early Middle Ages, and some archaeological artefacts associated with the early Frisians and Anglo-Saxons.

The Anglo-Frisian hypothesis The deconstruction of the Anglo-Frisian hypothesis The ‘old-school’ mode of reasoning about North Sea Germanic (including Anglo-Frisian commonalities) finds its climax in Schwarz’s (1951) Goten, Nordgermanen, Angelsachsen, where various Stämme/Völker (‘tribes/peoples’) with apparently clearly distinguishable ethnic and linguistic profiles move around, and ad hoc and in a fairly eclectic way exchange linguistic traits and features. Schwarz is very explicit about the origins of Anglofriesisch: ‘Es wird mit Recht geschlossen, daß die gemeinsamen Züge des Fries. und Ae. bereits auf dem Festlande ausgebildet waren.’ [It will be correctly concluded that the common traits of Frisian and Anglo-Saxon had been established already on the Continent.]. The Anglo-Frisian hypothesis of a common idiom that was taken to England by the Germanic invaders was severely criticized in Kuhn’s (1955) seminal article, which was in fact an expanded review of Schwarz’s book. He introduced a contrasting scheme, arguing for a fairly undifferentiated North-West Germanic language prior to the movements of the Migration Period. As in many other instances, ideas that later become mainstream can often be found long before that time. Chadwick’s (1889) account of events is entirely in line with Kuhn’s approach. Chadwick wrote (1889, 264): In saying that none of the existing distinctions between Scandinavian, Anglo-Frisian and West Germanic go back to a period much before the beginning of the fifth century I do not mean that before that time dialectical differences were entirely wanting. Considering the size of the area over which these languages are (and were even at that time) distributed, this would be improbable. I mean rather that the divisions and dialects, which existed before that time, were probably different from those which appear later. The political events of the fourth and fifth centuries involved a general dislocation of the Germanic world; new groups were formed and old differences would naturally often be obliterated.

The relationship among and differentiation of the Germanic languages has been carefully analysed in work by Nielsen (1985; 1998; 2000; 2010), revealing a nuanced picture of various early changes, gradually deploying into the differentiation between North Germanic, North Sea West Germanic and Continental West Germanic, before, during and after the landnam of England (adventus Saxonum). A view of this delicate and complicated process, without sharp boundaries, but largely taking place in a language continuum, where different changes spread over smaller or larger areas, was also developed in Stiles (2013). Most of the ‘classical’ North Sea Germanic features fall in the 340

Z01 Hines Book B.indb 340

07/05/2021 13:30

North Sea Germanic Idiom period during and after the landnam. Little to nothing is left of an Anglo-Frisian idiom on the Continent prior to the year 400. Instead of a pre-landnam Anglo-Frisian on the Continent, Kuhn and others suggest the development of a common language in the period during and after c. 400, where Frisian and English, with the inclusion of the Dutch coastal region (see Tab. 12.1), largely developed in a parallel vein for the next two centuries. But this version of Anglo-Frisian is nowadays mostly discarded as well. To quote Stiles (1995, 211): [T]he evidence does not support the notion of an ‘original Anglo-Frisian unity’ or sub-protolanguage. This is because it is not possible to construct the exclusive common relative chronology that is necessary in order to be able to establish a node on a family tree.

What we end up with is a set of features that spread at different times and different pace over varying regions around the North Sea. The reconstructed timing of the events is often so different that nothing more than a vague common tendency is left. The lack of similarity in the chronologies of phonological changes between English and Frisian, as mentioned by Stiles, is illustrated in Table 12.2. Table 12.2  Selected ‘Anglo-Frisian’ sound changes and their relative chronological ordering in English and Frisian according to the current conventional understanding. Sound changes in Old English

Old English dialects

Sound changes in Old Frisian

1. Raising of /æ:/ > /e:/

Mc, Nh

2. Rounding */a(:)/ > /ɔ(:)/

[WS], Mc, Ke, Nh

3. Monophthongization of */ai/

Mc, WS, Ke, Nh

4. Fronting */a/

Mc, WS, Ke, Nh

Fronting */a/ - restricted

5. Breaking: /i(:), e(:), æ(:)/ > io, eo, ea

WS, Mc, Ke, Nh

Monophthongization of */ai/ - restricted

6. Combinative ‘breaking’ /æ/ > /a/

Mc, Nh

12. Monophthongization of */au/ 

7. Restoration of /a/

WS, Mc, Ke, Nh

8. Restoration of /a:/ (partly)

WS

9. Palatal diphthongization: /e(:), æ(:)/ > ie, ea

WS, Nh

10. i-Mutation

WS, Mc, Ke, Nh

i-Mutation

11. Smoothing >

Mc, Nh

13. Labial mutation and breaking

Rounding */a(:)/ > /ɔ(:)/

Raising of /æ:/ > /e:/ Abbreviations -

Ke = Kentish, Mc = Mercian, Nh = Northumbrian, WS = West Saxon -

Examples OE/OFris: 1. rēd/rēd ‘advice’; 2. mōnað/mōnath ‘month’; 3. clāþ/clāth ‘cloth’; 4. þæt/thet ‘that’; 5. weorc,seax/werk,sax ‘work,knife’; 6. barna/berna ‘children’ (gen.pl.); 7. dagas/*dagar ‘days’; 8. māgas/mēgan ‘kinsmen’; 9. scēap/skēp ‘sheep’; 10. dēman, doeman/dēma ‘to judge’; 11. (cf. 5.) werc,sæx/werk,sax ‘work,knife’; 12. rēad/rād ‘red’; 13. þicce/thiukke ‘thick’.

341

Z01 Hines Book B.indb 341

07/05/2021 13:30

Frisians of the Early Middle Ages Table 12.2 illustrates that there is only a limited overlap in changes between English and Frisian and that most of the changes that are common are allocated to different places chronologically. The unavoidable conclusion is that, despite the similar outcomes, these changes were not ‘common’ in terms of timing and probably then also in terms of causation. Kortlandt (1999; 2008) suggests a common basis, with many developments actually shared in chronology and time by English and Frisian. Some of his ideas are reflected in the suggestions I will make but differ from it in other points. A full discussion goes beyond this chapter but some of my arguments can be found in Versloot (2017a, 318). The evidence reconsidered The archaeological evidence The cumulative archaeological and genetic evidence (for a recent summary, see, e.g., Colleran 2016) points in the direction of a re-settlement of the southern North Sea littoral (Frisia) and a landnam by speakers of Germanic languages from a common stock of people with origins in the Elbe-Weser Triangle, the southern Jutlandic peninsula (including Angeln), with some further admixture of people from further north up into Norway (Nicolay 2005; Higham and Ryan 2015, 70–125). This potentially opens up a way back to the ‘old-school’ thinking about a common Anglo-Frisian Continental ‘tribe’, that took its language to its new homelands. However, we adhere to the linguistic reconstructions of the last decades in the spirit of Chadwick, Kuhn, Nielsen and Stiles and conclude on the basis of their research that these people spoke a fairly undifferentiated North-West Germanic at the beginning of the fifth century. The new dialectal configuration of post-migration Germanic may incidentally show linguistic traces which existed already earlier in restricted parts of the Germanic continuum but cannot be pinpointed in a one-to-one relationship to specific homelands.1 The phonological history of English I recently suggested an extensive reform of the earliest phonological history of English (Versloot, forthcoming a; forthcoming b). This proposal is based on an in-depth analysis of the oldest Early Old English records: Cædmon’s Hymn, Bede’s Death Song (Sweet and Whitelock 1967, 181–3), the Franks Casket (Findell 2014, 44–54), minor sources that are traditionally linked to Northumbria, and the Épinal and Erfurt Glossaries, comprising around 1,100 parallel tokens (Pheifer 1974). All of these sources are from the late seventh or early eighth centuries; only the Erfurt version of the Glossaries is usually dated somewhat later in the eighth century. Together with my analysis of the development of Proto-Germanic *ai (Versloot 2017a) they suggest a major change to the chronologies shown in Table 12.2. The pivotal issues are the place of the Old English (OE) breaking in the relative chronology and the scope of fronting. OE breaking primarily concerns the change of short Proto-Germanic (PGmc) *a and e into ea and eo respectively, as in OE weard, weorc ‘guard’, ‘work’. Old Frisian (OFris) did not have this type of breaking and its 1

In a similar vein, some traits of American English can be found in dialects in Britain, but there is no direct relation between geographical variation in American English and specific British English dialects (see also Trudgill 2010).

342

Z01 Hines Book B.indb 342

07/05/2021 13:30

North Sea Germanic Idiom cognates of these OE words are therefore ward and werk, preserving the PGmc vowel qualities. Anglo-Frisian fronting appears in words such as OE þæt and dæg ‘that’, ‘day’, OFris thet, dei; cf. Dutch dat and dag, Old Norse þat and dagr. The current theory, which can be found in, for instance, Campbell (1959) and Hogg (2011), works with a general implementation of fronting of PGmc *a and a subsequent nearly exceptionless breaking of vowels before *rC, *lC and *x.2 The application of these two rules would predict, however, many more words with fronted or broken vowels than are in fact attested in historical OE texts. To make up for the disparity, a set of additional sound laws has been introduced (‘retraction’ and ‘smoothing’; nos. 6, 7 and 11 in Table 12.2) to account for the actually attested forms, making the chronology of OE according to the current theory extremely complicated and packed with contradictory and sometimes phonologically questionable sound changes. There are clear dialectal differences in this respect. The dialect of Wessex, commonly referred to as West Saxon, has many more broken vowels than the dialects more to the north, commonly labelled Mercian and Northumbrian. But even then, many actually attested forms, even in West Saxon, have to be ‘explained’ by positing mixtures of spelling traditions, sloppy scribal practices such as ‘scribal error(s)’ and ‘scribal confusion’, or otherwise archaic spellings at a time when no body of writing in the vernacular existed that people could take these spellings from.3 The following sequences leading from PGmc *a to e in Old English and Old Frisian in a similar phonological context, may serve as an example of the extra steps considered by the current theory for Old English. -

Old English: -

PGmc *harwian- ‘to abuse, treat with contempt’ > *hærwian (fronting) > *hearwian (breaking) > herwan (Anglian i-mutation)

Old Frisian: PGmc arƀijō- (form attested on the third- or fourth-century Tune   stone, Norway) ‘heir’ > erva (i-mutation by earlier -ij-) -

The intermediate steps posited for OE *hærw- and *hearw- do not appear in any form of early writing. Moreover, the alleged effect of i-mutation in OE, namely *ea > e, is phonologically illogical; the normal manifestation of i-mutation is directly from [a] > [ɛ]. The ‘Frisian’ scenario is much easier and more straightforward: *a was fronted by i-mutation to e and the same could be assumed for pre-OE. In the alternative scenario, suggested by myself and based on the relative order of events as reconstructed in Frisian, there is an only partial fronting of PGmc *a in OE, 2 3

‘C’ = any consonant; ‘x’ represents Dutch and German in reconstructed forms. A typical example is the frequent occurrence of ‘old’ without breaking in sections of the West Saxon Parker Chronicle from the ninth century, whereas the spelling with breaking predominates in entries describing events from the tenth century. The current theory ignores this transition in spelling as an indication of a change in pronunciation around the year 900 because breaking is supposed to have taken place three or four centuries earlier. Campbell (1977, §143) describes the variation in the chronicle only in terms of changes in ‘spelling’. Hogg (1992, §5.15) assumes that there were two contemporaneous dialects side by side, one with and one without breaking and that the older sections were written by speakers of the dialect without breaking. This equally assumes that breaking is of much earlier date than even the oldest sections of the chronicle.

343

Z01 Hines Book B.indb 343

07/05/2021 13:30

Frisians of the Early Middle Ages under nearly identical conditions as in Frisian, and a much later date for the breaking of PGmc *a and *e. An allocation of OE breaking to the late seventh century, deploying further in the eighth and ninth centuries both geographically and in respect of phonological environments, eliminates changes 6, 7, 8 and 11 from the chronology of OE in Table 12.2. The relative order of the remaining sound laws can subsequently be adjusted, and their relative chronology brought into line with the Frisian chronology. The linguistic shape of the Dutch coastal region De Vaan (2017) offers the most comprehensive discussion of the potentially North Sea Germanic features in the coastal dialects of Dutch. His conclusions are sobering for the Anglo-Frisian hypothesis in that many of the features interpreted as residues of an earlier North Sea Germanic idiom in the coastal region of the Netherlands, Belgium and northern France can just as well or better be interpreted as local developments, unrelated to what happened in Frisian and English. This careful scrutiny of the evidence secures, at the same time, a number of features or relic words (often only in place-names), that can be positively identified as relics of a North Sea Germanic idiom. For the sake of this research, the latter group is of interest. De Vaan enumerates these features and identifies them clearly (2017, 515–22). They are our only access to this variety of North Sea Germanic.4 The shape and extent of the North Sea Germanic idiom The ‘slimmed’ OE chronology turns out to fit very well with the developments as reconstructed for Frisian. Many of these developments are attested in relic words and placenames in the coastal Dutch zone as well; note that almost none of them is an integral part of the mainstream phonology of traditional coastal Dutch dialects. Many of these developments can also be found in Old Saxon, in particular the central and eastern varieties spoken in the river basins of the Weser and Elbe (Versloot and Adamczyk 2017). The evidence from those four varieties of Germanic (English, Frisian, coastal Dutch and Old Saxon) can be combined in a ‘North Sea Germanic tick-list’ containing the most prominent North Sea Germanic features. Table 12.3 presents this list and the geographical scope of the various developments by mentioning which varieties participated in which sound changes. Where necessary and reconstructable, sub-dialectal patterns have been included. Table 12.3 illustrates two important features. First of all, there is a largely shared chronology, both relative and absolute, for English and Frisian, including most of the Dutch and Flemish coast all the way to Calais, and the North Sea-oriented varieties of Old Saxon. This strong linkage remains intact up to i-mutation, which is a general, widespread phenomenon in North Sea Germanic, as opposed to the restricted ‘primary i-mutation’ of short PGmc *a in the Continental West Germanic varieties (Salmons 2018, 130–1). i-mutation is dated around the year 600 for English, perhaps with a slight delay in Frisian, and has a much more restricted application in Old Saxon. 4

I have recently identified a 48-line poem from 1643 as a likely specimen of Frisian from North Holland (Versloot 2018). Next to a one-line proverb from the Waterland region near Amsterdam, it seems to be the most extensive witness of Anglo-Frisian locally developed west of the Vlie. As far as attested in this brief text, it shows the Frisian features mentioned in Table 12.3.

344

Z01 Hines Book B.indb 344

07/05/2021 13:30

North Sea Germanic Idiom Table 12.3  An overview of the dating and geographical spread of prominent North Sea Germanic linguistic developments. North Sea Germanic changes

Date

Examples from modern English/ modern West Frisian < Proto(West) Germanic

Distribution

1. Rounding of /a/ and /ā/

fifth century

month/moanne < *mānōþ-

OE-AK(S) OF

FZH

EF,[WF]

2. Fronting of ā < PGmc ē1

fifth century

sheep/skiep < *skāpa-

OE

OF

FZH

EF

3. Fronting of PGmc /a/

fifth century

tail/teil < *tegl- < *tagla-

OE

OF

FZH

[EF]

4. Monophthongization PGmc /ai/ > ā

early sixth century

toe/tean < *taixwō-

OE

[OF]

[H]/ [ZF]?

[EF, WF]

5. Monophthongization PGmc /au/ > ā

early sixth century

beam/beam ‘tree’ < *bauma-

(OE?)

OF

FZH

[EF,WF]

6. Loss of nasals before spirants

400–600

us/ús < *uns

OE

OF

FZH

EF,WF

7. Palatalization of initial /g/ and /k/ before front vowels

sixth century

yield,church/ jilde,tsjerke < *geldan-,*kyrikō-

OE

OF

[FZH]

EF

8. i-mutation (‘secondary’) c. 600

green/grien < *grōni-

OE

OF

FZH

9. Rising diphthongs < PGmc eu

early seventh century

steer/stjoere < *steurijan-

[OE]

OF

H

10. OFris. Breaking

seventh century right/rjocht < *rexta-

[OE-K]

OF

H

11. OEng. Breaking

seventh– ninth centuries

heart/hert < *hertan-

12. Shortening of ai > æ

late seventh century

ladder/ljedder < *hlaidrō-

OF

ZH

EF,WF

13. Raising of ǣ (PGmc ē1) > ē

> 700

meadow/miede < *māþwō-

OF-S

FZH

EF

14. Delabialization of mutated vowels

eighth–tenth centuries

thin/tin < *þunwi- OE

OF

FZH

OE

OE-AK

Abbreviations -

OE = Old English: S = Saxon, A = Anglian, K = Kentish OF = Old Frisian (all dialects): S = ‘South Frisian’ – all varieties, except for Insular North Frisian Traces in Dutch coastal varieties: F = Flanders, Z = Zeeland, H = Holland EF = Eastphalian Old Saxon; WF = Westphalian Old Saxon […] = only under specific phonological conditions or with very little attested evidence.

345

Z01 Hines Book B.indb 345

07/05/2021 13:30

Frisians of the Early Middle Ages Developments 4, 6, 7(?), 8, 9 in Table 12.3 – and also, in a specific North Germanic form, 10 and 11 – are also found in North Germanic (Haugen 2004, 53–70). This shared chronology is not categorical: already before 600 we can detect differentiation between English and Frisian in the monophthongization of PGmc *ai and possibly also of *au; Old Saxon hardly participates in the fronting events, but evidence from the northern part of Saxonia is lacking. In a few instances, there is also dialectal differentiation within Old English and Old Frisian. The commonalities in the ‘inner circle’ of North Sea Germanic do not end abruptly after 600 but continue into the seventh and eighth centuries. This is very much in line with the concept of waves rather than a discrete node on the language tree. A hybrid model, combining both shared, areal phenomena as well as simultaneous gradual differentiation is sometimes called the linkage model (see e.g. Schulte 2018, 20). The coastal language of Zeeland and Flanders was not Frisian in a strict sense, perhaps already in the case of PGmc *ai, but certainly from the seventh century onwards with the lack of Frisian breaking. The whole Dutch coastal zone did, however, participate fully in the wider ‘Anglo-Frisian’ developments. Especially the varieties of Old Saxon spoken in the river basins of Weser and Elbe, which were apparently oriented more towards the North Sea than the speech of Saxons in the south-western regions near the Rhine, are involved in various of the early changes but hardly in any of the later ones mentioned in this overview.5 A link with North Germanic was particularly active from the sixth to the eighth centuries. The North Sea region with strong linguistic links is archaeologically characterized by small-scale regionalization. The gradual break-up of the North Sea Germanic dialectal continuum since the seventh century goes together with the formation of larger political and cultural units, which can be understood as a period with growing inter-regional contrasts and shrinking intra-regional contrasts (for the archaeological evidence, see Nicolay 2017, passim and esp. 88): the building of larger territories in England, and Frisia and Saxonia becoming a part of the Frankish Empire, with Scandinavia remaining ‘pagan’ territory beyond those Christian kingdoms. Is there an Anglo-Frisian node in the West Germanic family tree? Not in a strict sense, just as little as there was a West Germanic node (see the schema of differentiation of Germanic in Nielsen 2000, 287–93).6 But there was a period in the history of Germanic, roughly between AD 450 and 650, when varieties that later evolved into Frisian and English shared most of their innovations (see, in a similar vein, Deckers 2017, 178–9). The innovative power of this region remained active into the ninth century. Many of these innovations were at the time also found in the Dutch and Flemish coastal regions and extensive parts of present-day northern Germany, but largely disappeared from them after 800, leaving Frisian as an increasingly shrinking North Sea Germanic 5 Versloot and Adamczyk (2017) provide examples of a few more Old Saxon developments from the seventh and eighth centuries, not discussed in Table 12.3, which Old Saxon shared with Old Frisian and partly with Old English. Old Frisian and Old Saxon also share some morphological innovations not found in English, such as the productivity of the ending ‑u as a marker of the instrumental, the replacement of the nom.sg. ending of the ō-stems by the acc.sg. and the productivity of the i-stems, although much more limited in Frisian than in Saxon. 6 Even though English and Frisian are typologically definitely West Germanic in many aspects, they share many traits with Scandinavian that Continental West Germanic does not: traits that developed later than the typically West Germanic ones. This implies that there was not one, decisive split; see Euler (2013, 41–58) for an overview of features, with different emphases in his conclusions.

346

Z01 Hines Book B.indb 346

07/05/2021 13:30

North Sea Germanic Idiom relic language area on the Continent (Århammar 1990). As a consequence, Frisian and English are the heirs of an ‘Anglo-Frisian’ sub-group of West Germanic which encompassed a much wider geographical area than only England and Friesland at the time of its emergence in the Early Middle Ages. On the origin of North Frisian One of the major questions of Frisian historical linguistics has concerned the relationship between the so-called Insular North Frisian dialects, spoken on the islands of Sylt, Helgoland, Föhr and Amrum, and the rest of Frisian. The issue boils down to the following question: does the Insular North Frisian dialect group represent a branch on the Frisian node, the result of migration from ‘South Frisia’, or does it represent an – albeit closely related – autochthonous branch of a language spoken by a pre-Migration-period population that did not move to Britain or Frisia? Although situated in the zone with tidal flats along the southern North Sea, Nordfriesland is historically not a continuation of the southern regions: it was part of the Danish kingdom from the Early Middle Ages and it is separated from the rest of Frisia by the region of Ditmarsch without any historical reference to ‘Frisianness’ or traces of the Frisian language there. This spatial interruption is also visible in the map in Figure 12.1, showing an intermediate zone with very few place-names in *haim or *ing, and it implies that the dialects of the North Frisian islands are either an archaic relic zone or the result of historical colonization from the south. There is quite some research on this topic and excellent discussions and summaries can be found in Hofmann (1956) and Århammar (1968; 1990; 1995; 2001). There are striking differences between the North Frisian dialects of the coast and the Halligen on the one hand and the islands listed above on the other, in historical phonology and lexicon (see the references). Moreover, the speakers of the island dialects do not identify their speech as ‘Frisian’, contrary to the speakers on the coast and Halligen.7 The latter group is historically attested for the first time in the twelfth century and historical geographical research points to a settlement from the south in the eleventh century (Bantelmann et al. 1995, 62); the ‘Frisianness’ of the coastal and Halligen dialects is not disputed in any way. An essential methodological issue is the question in what respects the two groups of varieties (Insular North Frisian and all other dialects, here referred to as ‘South’ Frisian, including the language of the North Frisian coast and Halligen) should be similar or differ in order to consider either of the two hypotheses as being refuted or proven. This poses the subsequent question: what makes Frisian linguistically unique, compared with all other attested Germanic languages? Such linguistic features will preferably represent linguist innovations, as archaisms do not prove a common stage of development and can be found in any Germanic variety.8 Common 7

8

Although an incentive to open this discussion, this argument bears no weight in practice: the speakers’ self-designations of the other Frisian island varieties are not ‘Frisian’ either, but simply the adjectival form of the name of the island or even of ‘islanders (speech)’. Simultaneously, dialects such as ‘Westfries’ in North Holland (Netherlands) or ‘Ostfriesisch’ in Lower Saxony (Germany) owe their names to the regions and are varieties of Dutch and Low German respectively, although the regions were historically Frisian speaking. It should be kept in mind that the application of terms such as ‘language’, ‘dialect’ or ‘variety’ are largely interchangeable in historical linguistics. Germanic is a ‘dialect’ that developed from Proto-Indo European, whereas it is often considered to include various ‘languages’, such as English, Dutch, Frisian etc.,

347

Z01 Hines Book B.indb 347

07/05/2021 13:30

Frisians of the Early Middle Ages innovations that are fully and only implemented in one (group of) varieties are hard to find: most linguistic innovations spread over larger areas and it is the combination of a series of innovations and archaisms that characterize a language (group). Another complication is that the varieties under consideration are closely related under both hypotheses anyway and have been in constant contact with and influenced by other, also related, neighbouring varieties. Last but not least, the dates associated with a potential colonization scenario point to a time in history, the eighth century, when the Germanic languages were not highly differentiated anyway. This hints at another aspect to take into consideration. The idea of an autochthonous population on the islands, speaking some form of North Sea Germanic, was brought forward based on the assumption that many of the typical North Sea Germanic features had already developed on the Continent before the Migration Period (see Schwarz 1951, discussed above). If that were correct, the island dialects could be truly relics of an otherwise vanished (emigrated) language of the southern Jutlandic peninsula. However, the current interpretation works from the reconstruction that most of these features only developed between the fifth and eighth centuries. If the island dialects developed largely parallel but not identical to Frisian and English during that period, why, for instance, did Ditmarsch or other parts of Jutland not participate, whereas the North Frisian Islands did? This consideration weakens the ‘autochthonous’ hypothesis substantially. Every interpretation starts from the well-founded observation that the Insular North Frisian dialects’ closest genetic relatives are the ‘South’ Frisian dialects.9 This narrows the question to be answered down to: is there a considerable body of innovative features that are fully and only implemented in ‘South’ Frisian in comparison with any third Germanic language and which are also found in Insular North Frisian? The answer is ‘yes’. Key witnesses can be found in phonology, but also morphology and lexicon (Århammar 1995, 73–7; 2001, 533): -

1. The differentiation in the monophthongization of PGmc *ai (c. 500) as demonstrated by the following examples: PGmc *klaiþ- ‘cloth, dress’ > OE clāþ > ModE cloth; OFris clāth(ar) (pl.) > ModWFris klean, ModINFris (Föhr) kluaser ‘clothes’; ModDu. kleed; PGmc *staina- ‘stone’> OE stān ‘stone’ > ModE stone; OFris stēn > ModWFris stien, ModINFris (Föhr) stian; ModDu. steen.

9

each with its own ‘dialects’. So, ‘dialect’ primarily refers to a genetic and diachronic relationship whereas ‘language’ refers to a group of linguistically closely related dialects. All are ‘varieties’ of language. The use of these terms here should not be confused with their sociolinguistic application, where ‘language’ is mostly associated with a codified, normalized and standardized language variety with high prestige in the society while the term ‘dialect’ is mostly applied to spoken vernaculars. This is not to be confused with the linguistic distance among the present-day Frisian varieties. Most of these are not mutually intelligible and the distances between them are greater than among the three Scandinavian languages Danish, Norwegian and Swedish, or between Dutch and Afrikaans (Swarte et al. 2013, 297–8). Both among and even within the two North Frisian subgroups, mutual intelligibility is not obvious (Århammar 1975). The strong contrast between the modern Frisian variants (e.g. under 1., immediately below) attests to the disparate development of the varieties of Frisian varieties since the Early Middle Ages.

348

Z01 Hines Book B.indb 348

07/05/2021 13:30

North Sea Germanic Idiom The lexical distribution of forms pointing to an earlier *ā or *ē is nearly identical across ‘South’ Frisian and Insular North Frisian (Versloot 2017a, 304–5). -

2. The breaking of PGmc *e before *xt to iu, as in OFris riucht < *rexta- and instances of breaking by following *u/w as in OFris niugen < *nigun < *niun, ModWFris rjocht, njoggen, ModINFris (Föhr) rocht, njüügen ‘right’, ‘nine’. These breaking phenomena are also found in some varieties of Old English (e.g. incidental reoht, niogen alongside dominant riht, nigon) as well as in Old Norse, but the phonological conditioning and the actual lexical distribution of breaking phenomena is unique across all Frisian varieties. 3. Generalization of the fronting of PGmc *a in the feminine ō-stems, e.g. PGmc *talō-, OE talu, ModDu. taal, OFris tele > ModWFris tel(tsje), ModINFris (Sylt) tial ‘story’. 4. The ending -ar in the nominative plural of the masculine a-stems, OFris bāmar ‘trees’, ModINFris (Föhr) buumer, cf. OE bēamas, Old Saxon boumas. Frisian is the only West Germanic language with this ending (Versloot 2014b; 2017b). The ending is common in Old Norse, but in a different distribution than in Frisian. Its morphological distribution is unique for Frisian, where it appears also in the s-stems, e.g. Föhr ai – aier ‘egg(s)’, but where it is absent in Old Norse: egg (sg. and pl.). 5. The semantic shift of PGmc *texxō- ‘group, community’, OE tiohh, ModHG Zeche, to the fieldname OFris *tiuche ‘land parcel’, found on Föhr as Tjüüch, in the place-name Tjuchem (prov. Groningen) and the historical fieldname ‘t Joch (near Leiden in Holland) (Hofmann 1972). -

The aggregated evidence leads to a common reconstructable ancestor for all Frisian varieties, including Insular North Frisian, that can be dated to the period somewhere between 650 and 850. This dating is based on runic inscriptions and names in early medieval Latin texts (Gysseling 1962; Versloot 2014a). The timeframe matches the archaeological evidence concerning new or strongly extended settlements in the North Frisian Islands (see Majchczack, this vol.).

The speakers of Anglo-Frisian: origins and distribution The demographic and societal context Assuming that the proposed reform of the Old English phonological history fits, and that Table 12.3 is an adequate interpretation of the attested linguistic facts, one may wonder whether there was any reconstructable historical reality that supports the idea that language changes were able to spread so easily and in many instances so completely across the North Sea. The reconstruction given in Table 12.3 suggests in fact that the speakers on either side of the North Sea made up one speech community, with only minor dialectal differences, for at least two hundred years (c. AD 450–650) (for archaeological evidence of such a common zone, see Nicolay 2005; 2014). This presupposes intensive language contacts. These contacts may have run primarily through close contacts at various levels and among specific groups in the society, such as trading communities, or elite and warrior classes in these early societies, including marriages (Bos and Brouwer 2005); see Deckers (2017) and Kauhanen (2017) for a modelwise 349

Z01 Hines Book B.indb 349

07/05/2021 13:30

Frisians of the Early Middle Ages underpinning of such processes. One may assume that in a pre-feudal belligerent landnam society, more people were involved in overseas contacts than at later dates, comparable with the differences we can observe in geographical and social mobility between Viking-age Iceland and late- and post-medieval Iceland. When considering the people speaking Anglo-Frisian, a relevant question is: where exactly did these people live and where did they come from? The Frisian marshlands were in the process of being repopulated from the late fourth century, a process that lasted until the seventh century for the easternmost region of Land Wursten (Aufderhaar 2017). The innovative, developing Proto-Frisian spread from west to east. The various stages of migrations are visible in the place-name landscape: in most of Frisia, compounds containing PGmc. *‑haim are dominant, but in the east (Wangerland, Butjadingen, Land Wursten) names in *‑ing- are more numerous (see also Fig. 12.1), although both exist side by side (Versloot 2011, 133–4). In a similar vein, the anglicization of England will have taken some time, but this took place in a country with a population present. The character of the process of anglicization (the ‘how’, the speed, the numbers) is still heavily disputed (Higham and Ryan 2015, 70–125). The estimations of percentages of Germanic immigrants vary considerably. A recent study estimates ‘the proportion of Saxon ancestry in Cent./S England as very likely to be under 50%, and most likely in the range of 10–40%’ (Leslie et al. 2015, 313). Note that Leslie et al. use the term ‘Saxon’ for people associated both with northern Germany and Denmark, not distinguishing between the potential regions of origin of Angles, Saxons and Jutes (cf. below). Some studies suggest a strong gender bias among the settlers: the proportion of male DNA may have touched the mentioned upper limit, the female DNA rather the lower limit (Vonderach 2015, 284). This interpretation supports the classical view as presented, for example, in Gildas’s chronicle (Higham and Ryan 2015, 57–62) that especially the first wave of Anglo-Saxon immigrants consisted primarily of bands of male warriors, subjugating Britain and its inhabitants by military force. More research, in particular based on contemporaneous genetic material, is needed. The warrior-like behaviour and the overall confined (not necessarily small) number of the invaders make it unlikely that the whole of what we later call England was directly Germanic-speaking from the outset of the Anglo-Saxon invasion. A period of conquest was probably followed by a much longer period of cultural and linguistic anglicization of the remaining population. Linguistic evidence of the first Anglo-Frisian settlers An indication of the concentration of populations speaking Anglo-Frisian varieties may come from various sources, both archaeological and linguistic. Place-names are often mentioned as an indication of earlier demographic movements, such as settlement and colonization. It is generally accepted that place-names using the PGmc suffixes *haim ‘house, village’ and *ing ‘belonging to’, often with a proper name as the first element, mark Germanic settlement from the period of the Great Migrations (Berger 1993, 13–15). Apart from those two suffixes, a few more suffixes are commonly associated with Germanic-speaking settlers of the fifth and sixth centuries: the names in leben/-lev ‘remains, legacy’, with high densities in Thüringen; place-names containing PGmc *stadi- ‘spot, locality’; and names in court and ville in France, with counterparts in hofen and wihl,weil(er) and more in German (Nielsen 2000, 303–12; König 2001, 58). 350

Z01 Hines Book B.indb 350

07/05/2021 13:30

North Sea Germanic Idiom The elements court, ville/wihl etc. are of Romance origin. Place-names containing any of these suffixes have been gathered from a digital gazetteer and regions with dominant patterns computed (for a technical account of the sources and selection filters applied, see Appendix). The result is shown in Figure 12.1. The special focus of Figure 12.1 is on the relation between the locations of runic finds from the fourth to sixth centuries and the place-names commonly associated with the Migration Period. Most of the latter are located south-west of the former Roman Limes, including Britain. The coloured areas in Figure 12.1 cover 20% of the territories of the UK, the Benelux countries, Germany and Austria, whereas 69% of all runic finds are found against the background of one of these name-types. People able to write (West Germanic) runic texts lived predominantly in those regions which place-name evidence would associate with new settlers.

Fig. 12.1  A range of name-types commonly held (predominantly) to represent settlements from the Early Middle Ages, together with the distribution of runic finds. The mapping identifies the dominant variant among the six name-types in a particular region, and picks out regions where the sum of names of these six types is higher than the median.

351

Z01 Hines Book B.indb 351

07/05/2021 13:30

Fig. 12.2  The relative proportions of the name-types in Figure 12.1 (grey: ‘overall’) and the distribution of runic finds of the third–eighth centuries correlated with the name-types. Areas and runic finds that are not associated with any of the six name-types (32 out of 103) are not included.

Fig. 12.3  The correlation between the Anglo-Frisian and Continental runic finds of the third–eighth centuries and the various name-types. Runic finds not associated with any of the six names analysed (32 out of 103) are not included.

Z01 Hines Book B.indb 352

07/05/2021 13:30

North Sea Germanic Idiom The comparison with the six different name-types also reveals another special correlation: among these name types, the *haim and *ing names show the closest association with the runic finds. Figure 12.2 shows that, while the *haim and *ing are the most frequent types in the map among the six types investigated, the runic finds are even more strongly associated with these two name-types than following a random distribution over the six types. However, names with the *haim and *ing suffixes are not evenly associated with the various sub-groups of inscriptions. There is a clear link between Anglo-Frisian runes and *haim names on the one hand and between Continental runes and *ing names on the other, as illustrated in Figure 12.3. This correlates with the uneven geographical spread of the *haim and *ing suffixes. This analysis confirms and underlines the close relationship between the settlers of the Migration Period and the production of runic inscriptions outside Scandinavia. Apart from Frisia, which was newly settled from the fifth century after the hiatus in the fourth century (Nicolay 2005; Nieuwhof 2016), almost all inscriptions are found on former Roman territory. Why the geographical association between runic finds and place-names is almost exclusively restricted to names with the *haim and *ing, and appears to a much lesser extent with other names that are usually considered to date to the same period, is not easy to answer. In addition, there is the question why the use of runes seems to be so heavily restricted to newly settled regions and much less to the

Fig. 12.4  Evidence of Germanic settlement and the distribution of early runic inscriptions of the fifth and sixth centuries (© author, using coordinates for Continental runes from Findell 2012).

353

Z01 Hines Book B.indb 353

07/05/2021 13:30

Frisians of the Early Middle Ages regions the Germanic peoples settled before. Note that the few Continental inscriptions north of the Limes are mostly found in or near to regions with a clear presence of one or more of the six name types investigated. Following these observations concerning the six name-types studied, the focus was narrowed to names with the suffixes *haim and *ing, whose spread was combined in the map in Figure 12.4, again as the background to the distribution of runic finds of the fifth and sixth centuries. This distribution map confirms the observations made on the basis

Fig. 12.5  (above) The overall areal proportion of names ending in *-haim or *-ing in the West Germanic speaking countries: UK (GB), the Netherlands (NL), Belgium (B), Luxembourg (L), Germany (D) and Austria (A); (below) the proportional correlation of these points with a runic find. The y-axis represents the number of map-pixels; the x-axis represents ranges of proportions for the specific names amongst all other names within the given area. Note that >= 0 means >0–0.09 and the ranges of the bars are exclusive.

354

Z01 Hines Book B.indb 354

07/05/2021 13:30

North Sea Germanic Idiom of Figure 12.1: there are high concentrations of names with the suffixes *haim and *ing in the regions south and west of the Roman Limes (in present-day Flanders, south-western Germany and Alsace), in Friesland and in eastern England. They confirm their interpretation as settler names from the fifth and sixth centuries. The over-representation of runic finds in regions with these name types is illustrated diagrammatically in Figure 12.5. The figure shows that runic finds are often located in regions with 10–50% *haim and *ing place-names, whereas such regions are altogether a rarity in the Germanic-speaking western European countries (lower bar chart).10 The mapped evidence in this section implies that we now have two independent sources about language use, that are in strong agreement with each other: people writing runic texts (mostly brief inscriptions) lived in regions where many villages had a Germanic Migration Period settlement name, in particular with the suffixes *haim and *ing. This again poses the question why we have inscriptions from just these places but not (to the same degree) from other regions. People elsewhere apparently did not feel the need or were not able to write runic inscriptions. This makes sense for England and Continental regions south of the Limes, where many places were most likely still inhabited by the descendants of the Roman-period populations that had been living there before the arrival of Germanic-speaking groups. This hypothesis may particularly apply to the names in *villa, which is a Latin suffix after all. In the case of the regions north of the Limes, it has been claimed that these witnessed high levels of mobility and demographic relocation, producing relatively low population densities in northern Germany for longer or shorter periods (Siegmund 2003, 77–83; Meier 2003, 49; Dörfler 2003, 143). It seems that the elites in particular left the region in the period of the fifth–seventh centuries (Siegmund 2003, 90). Major factors were probably migrations to Frisia, Britain and the former Roman Empire as well as climatic circumstances (the Late Antique Little Ice Age, LALIA: Büntgen et al. 2016). Note also Bede’s remark about Anglia (Angeln) still being deserted in the seventh century (Bede, HE, I, 15). Combining linguistic and archaeological evidence This section will concentrate on the Anglo-Frisian settlers in Frisia and Britain, identified by Anglo-Frisian runic inscriptions. The hypothesis of a specific regional spread of ‘Anglo-Saxon’ migrants in the fifth and sixth centuries can be compared with archaeological evidence. We have compared the spread of Migration Period place-names and runic inscriptions with the locations of various types of brooch dated to the fifth century. One such object-type is the cruciform brooch, an object of daily use in female clothing (Nicolay 2005, 74–5; Martin 2015). The earliest specimens were produced in the homelands and relatively long preserved as ‘souvenirs’ from home (Bos and Brouwer 10 A caveat is due. The map is based on modern place-names and it is well known from diachronic evaluations of the onomastic landscape that there have been some substantial changes. Blok (1968) shows that the southern part of Holland – entirely white in Figures 12.1 and 12.4 – may have had around 25% ‑haim names in the Early Middle Ages but many of these names simply disappeared as a result of changes in the landscape. Such drastic changes are in particular likely where dramatic changes in the landscape occurred. Holland is such a region but far greater stability can be assumed for many other regions. Another complication is the exchange of suffixes, such as in the place-name Dokkum (Fryslân), counted here as a name with the ending *‑haim, but historically attested as Dockinga, which would qualify it as a member of the *‑ing group.

355

Z01 Hines Book B.indb 355

07/05/2021 13:30

Frisians of the Early Middle Ages

Fig. 12.6  The distribution of early cruciform brooches, c. AD 380–450 (after Nicolay 2005, 74) and fifth-century Saxon-style brooches (after Nicolay 2017, 78), against the background of Migration-period place-names (© author).

2005, 20–2, 30). Another type of brooch, the ‘Saxon’ jewellery, is usually associated with the formation of power (Nicolay 2005, 76–8, 87; see also Higham and Ryan 2015, 85–6, 107). Both types are shown against the background of the Migration Period placenames in Figure 12.6. The distribution of the brooches in Frisia and Britain is closely linked to the ­Migration-Period place-names, as were the runic inscriptions. Stray finds of both types of artefact and *haim and *ing names are found in Flanders, as well as an Anglo-Frisian runic inscription from Amay (Looijenga 2003, 303; see Fig. 12.4). This inscription, from a grave near Liège, is considered to be Anglo-Frisian because it contains a specifically Anglo-Frisian runic character, but it is not necessarily Frisian in any narrower linguistic, ethnic or cultural sense. It may represent a local community that migrated there from the same area as the other ‘Anglo-Frisians’ and shared linguistic features, placename types and artefacts with the nascent Frisian and Anglo-Saxon populations. However, the *haim and *ing names are notoriously absent from the regions of origin of the settlers, where such brooches are also found. The villages from which the settlers of Frisia and Britain came apparently had different names, or were deserted – as Bede describes for Angeln – and their original names have not been preserved due to a hiatus in habitation (Siegmund 2003, 80–1). This mismatch in place-names was extensively studied by Udolph, who concluded that the Schleswig-Holstein region in 356

Z01 Hines Book B.indb 356

07/05/2021 13:30

North Sea Germanic Idiom particular could not be considered a region of origin of the Anglo-Saxons, given the lack of onomastic links between that region and England (Udolph 2006, 338). However, he did not consider the possibility of a break in habitation. After the Migration Period, Angeln was taken over by the Danes, who brought now-Germanized Scandinavian place-names to the region, such as: Tastrop, Gelting, Böklund, Schuby. The low population level in this region correlates with the absence of runic inscriptions from these regions from the sixth (Fig. 12.4) and seventh centuries. We have extended the scope for Britain by including two additional layers of information. In the first place, we have included furnished Anglo-Saxon cemeteries from the fifth–seventh centuries. We also included a set of place-names with a Celtic suffix, namely place-names in don or dun, such as in Swindon.11 These names are relatively frequent in England but not in the western Celtic regions Wales and Cornwall. The set of names in don or dun therefore seems to continue an earlier layer of British placenames. The result is shown in Figure 12.7. The map and the underlying map calculations show that the concentration of names in haim and ing on the one hand and names in don or dun on the other are geographically complementary. The colours in the map are potential markers of the geographical spread of ethnicities in England in that period. The map again shows the distribution of Anglo-Saxon runic inscriptions of the fifth to seventh centuries. The spatial correlation between the linguistic/onomastic evidence and the archaeological evidence of cemeteries is clearly very low (Dodgson 1966). There is positive agreement in Kent, East Anglia and Sussex, and negative agreement (no Anglo-Saxon place-names, no Anglo-Saxon graves and no runic inscriptions) in Hertfordshire. But there are also large mismatches, for example, in Oxfordshire, Warwickshire and near surroundings. At this point, it is not possible to give an explanation. It would be good to have a further differentiation in time of the cemeteries. It may be interpreted in a way that the archaeological information shows us the progress of acculturalization of the British population to the Anglo-Saxons’ burial practices. This acculturalization or assimilation of Britons to Anglo-Saxon society was essential for the former to be able to function on an equal footing with the newcomers (Higham and Ryan 2015, 109–10). The contrast with the spread of the runic finds would imply that material and customary culture were adopted earlier than the language, which is in line with processes of culture and language shifts in many other places in the world.

Excursus I: ‘Angles’, ‘Jutes’ and ‘Saxons’ The earlier inclination to treat the populations discussed as cultural, political and linguistic unities, in the way it is applied by, inter alios, Schwarz (1951) as an example of an earlier paradigm of scholarship (see also Higham and Ryan 2015, 71), has been abandoned over the last decades. The author of the Anglo-Saxon Chronicle from the late 11 Gelling and Cole (2000) interpret Celtic *dūno- as ‘fortified place’ while dūn in English place-names occurs ‘as a generic in village names in regions where there are clusters of level-topped hills suitable for settlement sites’. The lack of fortifications at English places with dūn names persuades Gelling and Cole that dūn and *dūno cannot be associated. However, Celtic fortified oppida were situated on hill tops so that a semantic shift from ‘fortified site on a hill top’ to ‘settlement on a hill top’ is hardly implausible (cf. the semantic spread of Germanic *tūna- to German Zaun ‘fence’, Dutch tuin ‘garden’, Icelandic tún ‘meadow’ and English ‘town’).

357

Z01 Hines Book B.indb 357

07/05/2021 13:30

Frisians of the Early Middle Ages

Fig. 12.7  Possible evidence of Anglo-Saxon and British populations in England during the fifth–seventh centuries. The distribution of place-names is based on a modern gazetteer (see Appendix). Using place-name databases with onomastic etymologies might improve, or even alter, the picture. (© author; distribution of cemeteries after Higham and Ryan 2015, 80).

358

Z01 Hines Book B.indb 358

07/05/2021 13:30

North Sea Germanic Idiom ninth century is said to have ‘had no reliable sources for the fifth century’ (Higham and Ryan 2015, 72). Higham and Ryan are reasonably representative of current mainstream opinion in archaeology. Bede (HE, I, 15) is the classical source for a tripartition of England between Angles, Saxons and Jutes, an account that is currently interpreted as having been ‘framed according to the geopolitics of the present’ (i.e., Bede’s time), with ‘little historical validity’ (Higham and Ryan 2015, 75). Bede’s eighth-century description of the three peoples, settling in Britain in the fifth century, is nowadays considered ‘too late to be a useful primary source’ (Higham and Ryan 2015, 72). Bede himself, however, is recognized as having ‘believed that the major kingships of the seventh century had their origins in the settlements of different Germanic tribes in the fifth century’ (Higham and Ryan 2015, 137). The concept of clearly distinguishable peoples is replaced by the concept of occasional coalitions of ethnically diverse people, united under a leader for the purpose of a military campaign (e.g. Todd 2004, 29, 202–3). Archaeology traditionally distinguishes between Saxon and Anglian artefacts, but Higham and Ryan consistently put those designations in quotation marks, to stress their relativity (2015, 78–82). Although they acknowledge a strong correlation between ‘Anglian’ areas and cremation, and ‘Saxon’ areas and inhumation, they stress that those patterns are becoming more and more blurred with new data being available. This is supported by their maps (ibid., 80), showing different types of ‘Anglian’ and ‘Saxon’ artefacts with some geographical inclination but a large spatial overlap as well. And despite the affinity between ‘Saxon’ artefacts and practices and those found in the Elbe-Weser triangle, against overlap between ‘Anglian’ items and artefacts found north and east of the Elbe, they state that ‘these styles were already mixing on the Continent by the late fourth century, and the degree of regional conformity in England is less than was once suggested. Such is the diversity of artefacts that to date no Anglo-Saxon cemetery has been definitively tied to a specific place of origin in Germany as if the result of migration en bloc to Britain.’ (ibid., 82). Given all these misgivings, the pattern displayed in Figure 12.8 is astonishing. It demonstrates that nearly all the runic inscriptions from the fifth–seventh centuries, except for Watchfield but including the two that are mostly considered to be of Frisian origin (Hamwic and Folkestone), have been found in or very near the territories identified by Bede as Anglian and Jutish. This spatial correlation provides independent evidence – unknown to Bede (!) – for the geographical tripartition along the lines described by Bede in the eighth century. The implications of this correlation work in two directions. On the one hand, we may have to reconsider the refutation of Bede’s tripartition for the fifth and sixth centuries. Bede’s sketch may have more ‘historical validity’ than currently assigned to it and its rejection seems to be an example of over-application of ‘critical theory’ in the interpretation of historical sources and the relativization of their historical accuracy. A clear ethnic, ‘tribal’ or national (whichever term one prefers) identity for a settler community does not presuppose the uniform origin of all its members, but rather a clear destination territory in combination with a substantial founder community. Examples are found in countries such as New Zealand, Australia, Canada and the United States. We do not need full regional conformity nor en bloc migration from one place in order to end up with fairly distinct cultural provinces or nations in the new settlements. The correlation shown in Figure 12.8 also implies that the Saxons did not use/know the runic script, whereas the Angles and Jutes did. Given the dominant regions of 359

Z01 Hines Book B.indb 359

07/05/2021 13:30

Frisians of the Early Middle Ages

Fig. 12.8  The find spots of Anglo-Frisian runic inscriptions of the fifth–seventh centuries against the background of a traditional ethno-political map of England in the sixth century. The base map ‘England in 584’ has been taken from A School Atlas of English, ed. Samuel Rawson Gardiner 1907 (London).

360

Z01 Hines Book B.indb 360

07/05/2021 13:30

North Sea Germanic Idiom origin of these groups, south-west and north-east of the Elbe respectively, it suggests that the runic script was not much used in regions south-west of the Elbe before the Migration Period. It also implies that there was a substantial Anglian and/or Jutish contribution in the settlement of the Frisian marshlands, where runic inscriptions have been found. Figure 12.6 shows indeed various ‘Anglian’ cruciform brooches in the western Frisian terp region. Scandinavian cultural ‘input’ is particularly noticeable from the late fifth century and sixth centuries (Nicolay 2005, 78), which overlaps exactly with the increased production of runic texts in the Frisian region (see further, Excursus II, below). The Saxons and (western) Franks apparently did not know the runic alphabet, or at least did not use it extensively on inscriptions. The majority of the Continental runes are found in the Alemannic region, and the core of this population is generally associated with the Elbe-Germanic region (Quak 2016, 316–17). This means that the origins of the West Germanic peoples making extensive use of the runic alphabet between the fourth and eighth centuries can all be associated with a region of origin roughly north and north-east of the Elbe, contiguous with Scandinavia. This geographical bias is supported by a diachronic bias: whereas runic finds are relatively common in Scandinavia from before 400, the bulk of the West Germanic inscriptions are from the fifth and especially the sixth century. Still, all this may be not that simple: first, it is questionable to what extent a concept such as ‘Elbe-Germanic’ is still supported by modern archaeologys secondly, the Alemannic people are already mentioned near the Rhine and Danube in the late third century, which is three hundred years before the bulk of the ‘South Germanic’ inscriptions in the sixth century.

Excursus II: thi koning Redbad fan Danemercum The late-medieval Frisian text sources include a remarkable piece, known as the ‘Legend of Charlemagne and Redbad’ (Vries 2007, 50, 74–7; Bremmer 2009, 174–6; for a recently improved reconstruction of the oldest version in Codex Unia, see http://tdb. fryske-akademy.eu/temp/karel-en-redbad/). The story begins with: Tha thi koning Karle and thi koning Redbad fan Danemercum in that land comen … When King Charles and King Redbad from Denmark came into the country …

and tells of the victory of Charles over Redbad (Radbod) and subsequently the origin of Frisian law. ‘Karle’ is generally conceived of as being Charlemagne, who is also otherwise mentioned as the benefactor of Frisian law and Frisian freedom in the medieval Frisian tradition (Noomen 2001). In reality, it was his grandfather Charles Martel who fought a battle with the Frisian king Radbod in 716 and eventually initiated the conquest of the western parts of Frisia after Radbod’s death. Radbod is often associated with the ‘North’ as in the addition ‘fan Danemercum’. The seventh statute of the ‘Seventeen Statutes’ (version U: Sytsema 2012) states: That jef him thi koning Karle that truch se Cristen worden, and henzich, and herich tha sutherna koninge […] hwand alle Fresan er nord herden an tha grimma hirna.

361

Z01 Hines Book B.indb 361

07/05/2021 13:30

Frisians of the Early Middle Ages This [privilege] gave King Charles them [= the Frisians], because they became Christians, and obedient, and dependent on the southern King [of the Franks] … because all Frisians earlier belonged to the north in the dark outskirts [of civilization].

The ‘North’ is repeatedly mentioned in medieval sources as the origin of the Vikings and their attacks on Frisia (IJssennagger 2012, 38–40) and another OFris passage in R1 (Buma and Ebel 1963, 36–7) explicitly associates Redbad with the violent attackers from the north: thet wi er north herdon Redbate tha unfrethemonne ‘because we earlier belonged to the violent, northern [king] Redbad’ (see Popkema 2013, 197). So, the legend seems to contain a double anachronism: ‘Karle’ is both Charles Martel and Charlemagne while the Frisian king Redbad is associated with the Viking raids, which post-dated the reign of Charlemagne. Noomen (2001, 15) stated that De sêge is dêrmei gjin selsstannige boarne foar de kennis fan dy wrald, ‘the legend cannot be taken as an independent source of knowledge about the world [of the Early Middle Ages]’, but it may echo real events from earlier times. The ruling classes in Frisia in the second part of the fifth and early sixth centuries had close ties to Scandinavia (Jutland and Norway) and felt and maintained them well into the seventh century (Nicolay 2005, 78–85; 2017, 87).12 This brings us chronologically close to the appearance of Radbod in history. The collocation Redbad fan Danemercum may, therefore, not (only) reflect a late-medieval confusion between seventh-/eighth-century pagan Frisians and ninth-/tenth-century pagan Vikings, but just as well echo the self-purported Scandinavian link or even origin of the sixth- and seventh-century Frisian rulers.

Discussion and conclusions We conclude that the North Sea region witnessed the formation of the Anglo-Frisian idiom by a series of shared innovative linguistic changes, which at that time mostly also included the coastal zone of the Low Countries and parts of the continental Saxon region. The heydays of these developments fall in the fifth–seventh centuries. The geographical spread of these changes was not monolithic and differed per feature: some spread wider than the North Sea Germanic area, including sometimes (parts of) North Germanic, sometimes also parts of the Franconian regions, others encompassed only parts of this ‘Anglo-Frisian’ Sprachbund. The question about a separate branch on the Germanic language tree or a matter of convergence is actually superfluous under a more subtle interpretation of the developments: after a period of strong linguistic convergence, divergent tendencies became more prominent from the seventh century, although common innovations could still develop in the following centuries (Hines 1995; Deckers 2017). The Anglo-Frisian speakers were dispersed over Frisia (in its widest sense down to Calais), the (north)eastern parts of the Continental Saxon region (compare the Eastphalian involvement in most of the early sound changes in Table 12.3) and in Britain, in particular the eastern parts, in regions close to the North Sea. The language area was – certainly to begin with – no solid geographical territory, although spatial 12 In Nicolay (2005) this is attributed to a separate wave of immigration, but Nicolay (pers. comm.) now considers it rather to be a cultural phenomenon.

362

Z01 Hines Book B.indb 362

07/05/2021 13:30

North Sea Germanic Idiom concentrations existed. Anglo-Frisian in England was concentrated in some areas but otherwise rather a sociolect or ethnolect, whose speakers were foremost connected by their political and military power and their common ethnicity, including their religion (compare the position of the Goths in Italy and Spain and the Vandals in North Africa). It is revealing to see that, at least in the fifth and sixth centuries, regions such as Mercia and Wessex, later considered to be constituting the OE dialect landscape and well represented on the map with Anglo-Saxon cemeteries, seem barely to have been inhabited by speakers of Anglo-Saxon idioms, but remained numerically dominated by Romano-British descendants during the first two centuries after the landnam (cf. Grimmer 2007). The Frisian region was largely resettled in the same period as Britain, but here the settlers found a nearly empty region. Both regions are strongly marked by the toponymic landscape and their inhabitants knew how to write runic texts. This contrasts strongly with the Elbe-Weser region from where those populations had come. There are very few *haim and *ing names there, and basically no runic inscriptions after the fifth century (Sievern, Fallward, Heide and Geltorf II are all from the fifth century). A clearer picture of the historical-geographical landscape of those days of present-day northern Germany may help us to better understand the dynamics of Anglo-Frisian on the Continent.

Appendix: Sources for the place-name mapping The maps with place-names are based on the Geonames gazetteer: https://www.geonames.org/ -

The data were analysed and mapped using ILWIS 3.31 Academic GIS-software: https://www.itc.nl/ilwis/download/ilwis33/ -

The following text filters were applied: -

TypeFilter -court/-hofen: *hofen,*hoven,*court *-haim: *heim,*um,*hem,*ham,*gem,*hjem *-ingi: *ing,*ingen,*inghen,*ingh,*ingi,*inge,*ange; *ens (only for Frisia) -leben, -lev: *leben,*lev *stadi: *sted,*stedt,*stett,*stede,*stead,*stätten,*städten *villa: *ville,*viller,*vijl*,*weil,*weiler,*wihl* -

In total, 13,430 points were selected. -

The map in Figure 12.1 was smoothed using a moving window of 13 x 13 pixels. Before smoothing the image, those pixels with a summed proportion of any of the six nametypes >12% were selected. This is the sum of the average proportion in the entire map area plus one standard deviation. The average proportion was computed for a moving window of 21 x 21 pixels. Every pixel corresponds to one square kilometre.

363

Z01 Hines Book B.indb 363

07/05/2021 13:30

Frisians of the Early Middle Ages

Discussion HINES  Is there any place in your model for Peter Schrijver’s ideas that Celtic substrates were present and had a significant influence on both Old Frisian and Old English? VERSLOOT  I can imagine Celtic influence in England, because of the population situation when speakers of Germanic arrived. But Frisia was largely depopulated. I don’t dispute the case that Frisia could have been Celtic-speaking in the Roman Period. But I find his arguments a bit far-fetched. I very much admire his creativity and his ability to open new windows, but on some points he is too bold. However, breaking in English could well be Celtic-influenced. Breaking is more prevalent in West Saxon than in Northumbrian Old English and Celtic influence was probably greater in the west.13 HINES  Without going too deeply into details of the chronology of sound changes, one of the immediate problems that appears is that the reason why breaking is put before i-mutation is because we have i-mutated broken diphthongs. How do you get around that? Do you propose that breaking is not a single process? Could there have been a West Saxon breaking that is early, and potentially Celtic-influenced? You have both primary and secondary i-mutation too. VERSLOOT  Yes, it does require very lengthy analysis. The earliest evidence I can use is non-West Saxon. This is a problem of the evidence we have. The West Saxon situation is a specific problem, but that would have been spoken very far to the west, deep into a linguistic contact zone. I see its features as a bit of a side-track, rather than central to Old English. In Northumbria and Mercia, the i-mutation product of a so-called broken vowel is a monophthong. HINES  The Old English runic inscriptions are overwhelmingly ‘Anglian’ in terms of geography. Those from Kent and the Isle of Wight are particularly challenging in terms of legibility and intelligibility. From the heart of the Saxon area there is only the Watchfield inscription from the sixth century, and in fact that whole assemblage looks strikingly Kentish in key respects. On the whole, the runic inscriptions give us really valuable evidence for Anglian dialect areas of East Anglia, Lincolnshire, and subsequently Northumbria, but very much less information outside of that zone. VERSLOOT  In Anglian dialects, it was assumed there was widespread breaking that was later undone, but I’d like to see evidence of that. There is none. There is rather much to suggest there was no early breaking. DE LANGEN  When you discussed North Frisian you said you saw influence from the south. Where is that? And is that area ‘monolithic’? VERSLOOT  Yes, from the features listed, it is a monolith. DE LANGEN  And it includes the Rhineland? VERSLOOT  It includes North-Holland. 13 The chronology of sound changes reconstructed for Gallo-Romance in the period AD 400–700 is revealing – including: 1. fronting of a/ā; 2. palatalization of velars before fronted vowels; 3. retraction of æ > a before w, n (among other things); 4. diphthongization of e when following a palatal consonant (≈ palatal diphthongization) (Kerkhof 2018, 120). The close resemblance of these to the developments in pre-Old English sheds new light on the areal distribution of so-called ‘North Sea Germanic developments’ and their potential sources. APV

364

Z01 Hines Book B.indb 364

07/05/2021 13:30

North Sea Germanic Idiom DE LANGEN  Where is the boundary? VERSLOOT  Somewhere south of the Oude Rijn (near Leiden). DE LANGEN  That is interesting! VERSLOOT  I may perhaps mention that up to now there has been no Frisian text identified from North-Holland, but I found one that was already known which can be reinterpreted as originating from there. It is about love; it has been in the newspapers. Salacious stories attract attention! That is the first piece we have (Versloot 2018). It’s also different from West Frisian, spoken in the province of Fryslân. But in terms of the early features, it shares all of those. In terms of linguistic features, the boundary between the Maas and the IJ is much more important than the Vlie. DE LANGEN  A little bit more about the Rhineland. How much difference is there: when do the language dialects part? VERSLOOT  Up to a given point (Tab. 12.3, point 9) we have relic features in Flanders, Zeeland and Holland. This appears to be early seventh-century to me. After that Zeeland and Flanders no longer follow, but it doesn’t mean they can’t share in innovations afterwards. The wave model allows for variability. But up to that point there seems to be a single linguistic zone, except for some differentiation a little earlier on (see the Old Saxon column in Table 12.3). Later on, it gets more problematic, when the Westphalian region diverges, but in the eighth/ninth centuries there are still common elements in Frisian and Eastphalian. The model is flexible. DE LANGEN  The boundaries inland are clear? VERSLOOT  We have very little data. DE LANGEN  I’m particularly interested in the central river area. VERSLOOT  There’s practically nothing there, except very early and difficult runic inscriptions (Bergakker and Borgharen: see Looijenga, this vol., Ch. 13), which do not contain Anglo-Frisian traces. DE LANGEN  How much data do you really have? When a new find comes, do you see new boundaries? VERSLOOT  Not really boundaries, but I’m sure new finds will improve the chronology. Much of it is the result of recent scrutiny of the available linguistic and onomastic material dating from c. 800 and later, including modern dialects. MOL  Do you distinguish South- and North-Holland? VERSLOOT  No, Holland is taken as one. The boundary between Zeeland and Holland is around the Maas. DE LANGEN  But you see Holland as one? VERSLOOT  Yes: it’s just that Frisian disappears from South-Holland much earlier than it does from North-Holland. When you take the single relics in place-names, though, Holland is nicely in line. FLIERMAN  Is this diagram the totality of changes or a selection? VERSLOOT  No, it’s a representative selection, which concentrates on vowels, which are most prominent in historical reconstructions of Germanic languages. There is more, but this is the core for the Anglo-Frisian varieties. LOOIJENGA  In runic inscriptions, we have famous instances of the Frisian combs with, variously, kabu and kobu. How do you explain that? 365

Z01 Hines Book B.indb 365

07/05/2021 13:30

Frisians of the Early Middle Ages VERSLOOT  The development of the a+nasal consonant to o is as a gradual/‘reluctant’ change. In Old English too, it is a tendency rather than a full change: for example, we have Modern English long but also hang; comb but lamb. It’s therefore unproblematic to have both spellings. LOOIJENGA  They’re from very close to each other, with only one river between them. VERSLOOT  I don’t think the contrast should be interpreted in terms of a dialectal difference; it rather reflects the fact that the phonological status of this development was very weak. LOOIJENGA  And the ending -u? How do you explain that? VERSLOOT  I see it as an approximation of a schwa-like sound (cf. Nedoma 2018). NIJDAM  When we move into Old Frisian, there are only a few phonological changes that are exclusive to the language, in particular eht-breaking (reht > riucht) and w-breaking (singwan > siunga), which are also found in Insular North Frisian. There aren’t many distinctive features. VERSLOOT  Indeed; but also the differentiation of ai into ē and ā is important (e.g. English toe, stone; Dutch teen, steen; Old Frisian tāne, stēn). Some developments differ from Old Saxon and Old English only in the details. It is the total mix of features, some shared with Old English, some with Old Saxon, some with Old Norse, that characterizes Frisian. NIJDAM  That is my question, based on Roger Lass (1994) etc.: you say riuht appears just after 600. I thought it would only be in the late eighth century that it would have appeared. But you now see a North Sea Germanic continuum in which Frisian is distinct from the seventh century onwards? VERSLOOT  The concept of ‘Frisian’ is imposed from hindsight. I think a Germanic speaker of c. 700 would have had difficulty identifying something as ‘Frisian’; someone from the northern part of Saxonia would speak a very similar language. In hindsight we can see the changes becoming part of a ‘language’. It has to do with the family-tree model. With Franconian, there could have been a perceptible difference by about 700. NIJDAM  With things like final devoicing and loss of h etc.? VERSLOOT  Final devoicing is typical of Continental Germanic and much later in coastal areas – even in the nineteeth/twentieth centuries in Frisian. NIJDAM  Can you give us examples of how Frankish and Frisian differed? VERSLOOT  In all these things (Tab. 12.3), Frankish does not participate. Franconian is a very archaic language variant. What we find in Lex Salica is basically protoWest Germanic in form. Nothing seems to have happened. That’s also remarkable. NICOLAY  I wonder what you can say about the difference west and east of the former Vlie (between present-day Friesland and North-Holland), in relation to the habitation hiatus in the fifth century, and the sources of repopulation from the south (Frankish area) rather than the east (Anglo-Saxon Germany)? Does this fit with your model? VERSLOOT  What perhaps it does is to confirm the idea that the changes took place in situ and had nothing to do with where people came from. They came in with a quite undifferentiated West Germanic language, and once settled they began to participate in a North Sea Germanic configuration. It doesn’t matter where they came from. This confirms Kuhn’s idea that the innovations took place during and 366

Z01 Hines Book B.indb 366

07/05/2021 13:30

North Sea Germanic Idiom after the Migration Period. It’s an interesting oxymoron that immigration seems to have involved markedly distinctive groups in the archaeology while the linguistic picture is very different: more of a blank slate to start with. DE LANGEN  At the same time, you also stressed that you have a need for a certain amount of time for a change. But up to 600 the languages stayed together and then diverged. Does that give us enough time? VERSLOOT Between c. 400 and 600 these languages share a couple of innovations. DE LANGEN  They have to accelerate at a certain stage. VERSLOOT  Basically, yes. DE LANGEN  You also stressed a boundary with the hinterland, which made me wonder where the new settlers came from to join in so quickly. VERSLOOT  I don’t really see the problem. DE LANGEN  We have a hinterland which speaks one language, and people join in from somewhere. NICOLAY  In the western Netherlands you mean? DE LANGEN  As I read this as an archaeologist, by the fifth century they had already joined up. And then they split up. But when did they join up? VERSLOOT  Between 400 and 450 we have the start of settlement in Frisia and England, and they were in contact with one another. DE LANGEN  …but not in Holland. VERSLOOT  OK. That means that they would have to catch up in a way, through a sort of course of integration. DE LANGEN  But you also need time. And then they split up again. My question is, how much time do you need? VERSLOOT  The split-up is not a problem. You need them to stay in contact. DE LANGEN  But when did they join up? VERSLOOT  400, 450? Up to 500? That’s what I get from the archaeological story. Joining up didn’t require them to sit down together. The Anglo-Saxons were already in England, and you need North Sea Germanic contact and networks. We hear a lot about those. I’m interested also in the sociolinguistic reality of it. It’s true that there are rapid changes between 450 and 700, which we see often in languages in a dynamic situation. It’s possible for that then to freeze subsequently. It’s not a single linear thing. We see the same thing in Frisian in the fifteenth century. Around 1400 the language is typologically similar to late Old English; around 1500 it’s like the modern language. HINES  The answer to the questions on your slide, is that we should wait and see. Archaeology is developing very rapidly, and ancient DNA (aDNA) data are now becoming available as a source of study on past demography. Advances in stable isotope analysis and interpretation are proceeding in an encouraging way too. We shall be able to study population mobility much better in the course of the next ten to fifteen years, and just for now we have to be patient. Nevertheless, the distributions of artefacts do show connected networks, and those networks absolutely presuppose contacts between people. I would draw attention to the fact that a concept that quite rapidly came to the fore in historical linguistics/language history about twenty years ago was the idea of ‘convergence’ between languages in contact as a major factor. Languages do not just split and diverge but can become more similar too. A more sociolinguistic viewpoint would 367

Z01 Hines Book B.indb 367

07/05/2021 13:30

Frisians of the Early Middle Ages emphasize that the practical circumstances have to be right for particular modes of development to take place: whether that is the movement of people or just contact is an interesting point. FLIERMAN  I deal mostly with Latin. Can we postulate the absence or presence of a written culture as a major factor? Can a literary tradition affect innovations? VERSLOOT  Written language is definitely a stabilizing factor nowadays. Thus rapid changes are more likely to take place in primarily oral language circumstances. We don’t really see that stabilizing influence until the modern era. FLIERMAN  So, you wouldn’t use it to compare the less innovative situation in Franconia with the north. VERSLOOT  Too little was written in the vernacular in Franconia then to make a difference. KNOL  In relation to changes in the fifteenth century, I was expecting something very different. In Groningen and Ostfriesland then, Frisian was completely replaced by Saxon. Less than a century was required for complete language replacement. You could ask the same questions of that situation, but only some can be answered. Subsistence and social organization did not change in Groningen then. Frisian Freedom remains in Groningen. But the language changed, except for place and personal names. This would be a very interesting point of comparison. VERSLOOT  I have given this some thought. I don’t have clear answers yet as to what was going on. In light of the Frisian Freedom, external political changes cannot be identified as the cause. KNOL  I wasn’t aware of a change in Frisian in Friesland then, although now I hear that the Frisian language changed there too. VERSLOOT  It was an internal change, but towards something much closer to Middle Dutch in type. There must have been widespread bilingualism, and this explains why the cities switched to Dutch. But I don’t think it’s exactly the same. Looking at the spread of linguistic innovations or fashions of pronunciation in the Early Middle Ages is different from the fifteenth century when it’s a matter of languageshift and strong contact situations. I hadn’t though thought about the relationship between the internal changes in Frisian and the concurrent language-shift in Groningen. The same phenomenon should underlie them. LOOIJENGA  I think the situation involves a lot more than Frisians and AngloSaxons alone. Britain was not an unpopulated place. There were lots of contacts with France/Gaul too. I don’t know what language they would have had. This makes the situation very complicated, and that must have sociolinguistic consequences too. It could also help explain why it was so innovative, and interesting fashions spread. VERSLOOT  Yes, you are right. It’s an interesting consideration; thank you (cf. fn. 13, above). IJSSENNAGGER-VAN DER PLUIJM  Does population size matter and can you take that into account? VERSLOOT  The smaller the group, the easier it is for changes to spread. There are interesting network-model analyses of this: depending on the number of participants and the form of the network, changes can spread quickly. IJSSENNAGGER-VAN DER PLUIJM  So, could it be that because the networks were originally small, in the early phase, innovations would spread more quickly? 368

Z01 Hines Book B.indb 368

07/05/2021 13:30

North Sea Germanic Idiom VERSLOOT Yes. DE LANGEN  It could also be that they switched language. VERSLOOT  At that time, switching language basically meant switching accent. NIEUWHOF  That answers my question: why was the Frankish language so stable, then? It is probably because it was a large population. VERSLOOT  Yes, otherwise we’re talking about a mobile situation with small numbers. The comparison with Saxon is important. Old Saxon is the most conservative form in the eighth/ninth centuries. In Franconian you do get the High German consonant-shift. Saxony remains very stable. NIEUWHOF  …while the fifth-century Frisian area had a new population forming, and so was very flexible.

References (including Discussion) Primary sources A School Atlas of English History. Ed. Samuel Rawson Gardiner 1907 (London). Bede, Ecclesiastical History, Volume I: Books 1–3. Trans. J. E. King 1930, Loeb Classical Library 246 (Cambridge, MA). R1 = Rüstringer Recht. Eds. Wybren Jan Buma and Wilhelm Ebel 1963, Das Rüstringer Recht. Altfriesische Rechtsquellen. Texte und Übersetzungen 1 (Göttingen). U = Codex Unia. Ed. Johanneke Sytsema 2012, ‘Diplomatic Edition Codex Unia’ (via http:// tdb.fryske-akademy.eu/tdb/index-unia-en.html#). Sweet, Henry and Dorothy Whitelock 1967, Sweet’s Anglo-Saxon Reader in Prose and Verse (Oxford). Valcooch, Dirck Adriaensz 1599, Chronijcke van Leeuwenhorn (Hoorn) (via https://dbnl. org/tekst/valc007chro01_01/valc007chro01s_01_0001.php , accessed 22 August 2018). Secondary sources Århammar, Nils R. 1968, ‘Die Herkunft des Inselnordfriesischen im Lichte der Wortgeographie’, Philologia Frisica anno 1966 (Grins), 49–76. — 1975, ‘Historisch-soziolinguistische Aspekte der nordfriesischen Mehrsprachigkeit’, Zeitschrift für Dialektologie und Linguistik (2), 129–45. — 1990, ‘Friesisch und Sächsisch. Zur Problematik ihrer gegenseitigen Abgrenzung im Früh- und Hochmittelalter’, In Aspects of Old Frisian Philology, eds. R. H. Bremmer, G. van der Meer and O. Vries, Amsterdamer Beiträge zur älteren Germanistik 31–32 (Groningen–Amsterdam–Atlanta), 1–25. — 1995, ‘Zur Vor- und Frühgeschichte der Nordfriesen und des Nordfriesischen’, In Friesische Studien II. Beiträge des Föhrer Symposiums zur Friesischen Philologie vom 7.–8. April 1994, eds. Volkert F. Faltings, Alastair G. H. Walker and Ommo Wilts (Odense), 63–96. — 2001, ‘Die Herkunft der Nordfriesen und des Nordfriesischen’, In Handbuch des Friesischen / Handbook of Frisian Studies, ed. Horst Haider Munske (Tübingen), 531–7. Aufderhaar, Iris 2017, ‘Between Sievern and Gudendorf. Enclosed sites in the north-western

369

Z01 Hines Book B.indb 369

07/05/2021 13:30

Frisians of the Early Middle Ages Elbe-Weser triangle and their significance in respect of society, communication and migration during the Roman Iron Age and Migration Period’. In Frisians and their North Sea Neighbours. From the Fifth Century to the Viking Age, eds. J. Hines and N. IJssennagger (Woodbridge), 149–72. Bantelmann, Albert, Albert Panten, Rolf Kuschert and Thomas Steensen 1995, Geschichte Nordfrieslands (Heide). Berger, Dieter 1993, Duden: Geographische Namen in Deutschland: Herkunft und Bedeutung der Namen, Duden-Taschenbücher 25 (Mannheim–Leipzig–Wien–Zürich). Blok, D. P. 1968, ‘Plaatsnamen in Westfriesland’, Philologia Frisica anno 1966., 11–19. Bos, J. M. and E. W. Brouwer 2005, ‘De kruisvormige fibulae van Friesland: een nieuwe benadering in het onderzoek naar de herkomst van de eerste migranten vanaf AD 380/400 en de culturele oriëntatie van het Friese terpengebied tot ca. AD 550’, De Vrije Fries 85, 9–36. Bremmer Jr, Rolf H. 2009, An Introduction to Old Frisian: History, Grammar, Reader, Glossary (Amsterdam). Büntgen, Ulf, Vladimir S. Myglan, Fredrik Charpentier Ljungqvist et al. 2016, ‘Cooling and societal change during the Late Antique Little Ice Age from 536 to around 660 AD’, Nature Geoscience 9 (3), 231–6. Campbell, Alistair 1959 (2nd edn 1977), Old English Grammar (Oxford). Chadwick, H. Munro 1889, Studies in Old English (London). Colleran, Rebecca A. B. 2016, ‘Keeping it in the family; disentangling contact and inheritance in closely related languages’ (Ph.D. thesis, University of Edinburgh) https://era. ed.ac.uk/handle/1842/25919 De Vaan, Michiel 2017, The Dawn of Dutch. Language Contact in the Western Low Countries before 1200 (Amsterdam). Deckers, Pieterjan 2017, ‘Cultural convergence in a maritime context’. In Frisians and their North Sea Neighbours. From the Fifth Century to the Viking Age, eds. J. Hines and N. IJssennagger (Woodbridge), 175–92. Dodgson, John McN. 1966, ‘The significance of the distribution of the English place-name in -ingas, -inga- in south-east England’, Medieval Archaeology, 10, 1–29. Dörfler, Walter 2003, ‘Rural economy of the Continental Saxons from the Migration Period to the tenth century’. In The Continental Saxons from the Migration Period to the Tenth Century: An Ethnographic Perspective, eds. Dennis Howard Green and Frank Siegmund, Studies in Historical Archaeoethnology (Woodbridge), 133–57. Euler, Wolfram 2013, Das Westgermanische: von der Herausbildung im 3. bis zur Aufgliederung im 7. Jahrhundert; Analyse und Rekonstruktion (Berlin). Findell, Martin 2012, Phonological Evidence from the Continental Runic Inscriptions (Berlin–Boston). — 2014, Runes (London). Gelling, Margaret and Ann Cole 2000, The Landscape of Place-Names (Stamford). Grimmer, Martin 2007, ‘Britons in early Wessex: The evidence of the Law Code of Ine’. In Britons in Anglo-Saxon England, ed. Nick J. Higham, Publications of the Manchester Centre for Anglo-Saxon Studies 7 (Woodbridge), 102–14. Gysseling, Maurits 1962, ‘Het oudste Fries’, It Beaken 24, 1–26. Haugen, Odd Einar 2004, ‘Frå urnordisk til norrønt språk’. In Handbok i norrøn filologi, additional file online (http://folk.uib.no/hnooh/handbok/ekstra/urnordisk.html) (Bergen).

370

Z01 Hines Book B.indb 370

07/05/2021 13:30

North Sea Germanic Idiom Higham, Nicholas J. and Martin J. Ryan 2015, The Anglo-Saxon World (New Haven–London). Hines, John 1995, ‘Focus and boundary in linguistic varieties in the north-west Germanic continuum’. In Friesische Studien II. Beiträge des Föhrer Symposiums zur Friesischen Philologie vom 7.–8. April 1994, eds. Volkert F. Faltings, Alastair G. H. Walker and Ommo Wilts (Odense), 35–62. — 2017, ‘The Anglo-Frisian question’. In Frisians and their North Sea Neighbours. From the Fifth Century to the Viking Age, eds. J. Hines and N. IJssennagger (Woodbridge), 25–42. Hofmann, Dietrich 1956, ‘Probleme der nordfriesischen Dialektforschung’, Zeitschrift für Mundartforschung 24 (2), 78–112. — 1972, ‘Fries “tiuche”, deutsch “zeche”, griech “dikè” und Verwandte’, Us Wurk 21–2, 55–80. Hogg, Richard M. 2011, A grammar of Old English Vol. 1, Phonology (Chichester). IJssennagger, Nelleke 2012, ‘Vriend, vazal en vijand? Contacten tussen Friezen en Scandinaviërs in de Vikingtijd: Teksten en objecten’, De Vrije Fries 92, 37–59. Kauhanen, Henri 2017, ‘Neutral change’, Journal of Linguistics 53 (2), 327–58. Kerkhof, Peter Alexander 2018, ‘Language, law and loanwords in early medieval Gaul’ (Ph.D. thesis, Leiden University). König, Werner 2001, Dtv-Atlas Deutsche Sprache (München). Kortlandt, Frederik 1999, ‘The origin of the Old English dialects revisited’, Amsterdamer Beiträge zur Älteren Germanistik 51 (Amsterdam), 45–51. — 2008, ‘Anglo-Frisian’, NOWELE. North-Western European Language Evolution 54/55, 265–78. Kuhn, Hans 1955, ‘Zur Gliederung der germanischen Sprachen’, Zeitschrift für deutsches Altertum und deutsche Literatur 86, 1–47. Lass, Roger 1994, Old English. A Historical Linguistic Companion (Cambridge). Leslie, Stephen, Bruce Winney, Garrett Hellenthal et al. 2015, ‘The fine-scale genetic structure of the British population’, Nature 519 (7543), 309–14. Looijenga, Tineke 2003, Texts and Contexts of the Oldest Runic Inscriptions (Leiden–Boston). Martin, Toby F. 2015, The Cruciform Brooch and Anglo-Saxon England (Woodbridge). Meier, Dirk 2003, ‘The North Sea coastal area: Settlement history from Roman to early medieval times’. In The Continental Saxons from the Migration Period to the Tenth Century: An Ethnographic Perspective, eds. Dennis Howard Green and Frank Siegmund, Studies in Historical Archaeoethnology (Woodbridge), 37–76. Nedoma, Robert 2018, ‘Voralfriesisch -u im Nominativ und Akkusativ Singular der maskulinen a-Stämme’. In Directions for Old Frisian Philology, eds. R. H. Bremmer Jr, S. Laker and O. Vries, Amsterdamer Beiträge zur älteren Germanistik 73 (Amsterdam), 343–68. Nicolay, Johan A. W. 2005, ‘Nieuwe bewoners van het terpengebied en hun rol bij de opkomst van Fries Koningschap. De betekenis van gouden bracteaten en bracteaatachtige hangers uit Friesland (vijfde-zevende eeuw na Chr.)’, De Vrije Fries 85, 37–103. — 2014, The Splendour of Power: Early Medieval Kingship and the Use of Gold and Silver in the Southern North Sea Area (5th to 7th century AD), Groningen Archaeological Studies 28 (Groningen). — 2017, ‘Power and identity in the southern North Sea area: the Migration and Merovingian Periods’. In Frisians and their North Sea Neighbours. From the Fifth Century to the Viking Age, eds. J. Hines and N. IJssennagger (Woodbridge), 75–92. Nielsen, Hans Frede 1985, Old English and the Continental Germanic Languages: A Survey of

371

Z01 Hines Book B.indb 371

07/05/2021 13:30

Frisians of the Early Middle Ages Morphological and Phonological Interrelations, Innsbrucker Beiträge zur Sprachwissenschaft 33, 2nd edition (Innsbruck). — 1998, The Continental Backgrounds of English and its Insular Development until 1154. (Odense–Portland OR) — 2000, The Early Runic Language of Scandinavia: Studies in Germanic Dialect Geography (Heidelberg). — 2010, ‘The early runic language of Scandinavia: proto-Norse or north-west Germanic?’ In Zentrale Probleme bei der Erforschung der älteren Runen: Akten einer internationalen Tagung an der Norwegischen Akademie der Wissenschaften, eds. John Ole Askedal, Harald Bjorvand, James E. Knirk and Otto Erlend Nordgreen (Frankfurt am Main.), 95–114. Nieuwhof, Annet 2016, ‘De lege vierde eeuw’. In Van Wierhuizen tot Achlum. Honderd Jaar Archeologisch Onderzoek in Terpen en Wierden, ed. A Nieuwhof, Jaarverslagen van de Vereniging voor Terpenonderzoek 98 (Groningen), 83–98. Noomen, Paul N. 2001, ‘Hachens en Wachens. Feit en fiksje yn midsieusk Fryslân’. In Speculum Frisicum: Stúdzjes oanbean oan Philippus H. Breuker, eds. Rolf H. Bremmer Jr, Lammert G. Jansma and Piet Visser (Leeuwarden–Leiden). Pheifer, Joseph J. M. D. 1974, Old English Glosses in the Épinal-Erfurt Glossary (Oxford). Popkema, Anne T. 2013, ‘On the origin of all evil: The Old Frisian collocation grimme herne and its modern use’. In Twenty-nine smiles for Alastair: Freundesgabe für Dr. Alastair G. H. Walker zu seinem Abschied von der Nordfriesischen Wörterbuchstelle der Christian-Albrechts-Universität zu Kiel am 4. Juli 2013, ed. J. Hoekstra (Kiel–Groningen), 195–207. Quak, Arend 2016, ‘Urgermanisches e1 im Kontinentalgermanischen’. In From Variation to Iconicity (Festschrift for Olga Fischer), eds. A. Bannink and W. Honselaar (Amsterdam), 311–30. Salmons, Joseph 2018, A History of German: What the Past Reveals about Today’s Language. 2nd edition (Oxford). Schouten, P., J. de Groot, G. Warnars, S. Luchtmans, J. Luchtmans, A. Blusse and P. Blusse 1785, Tegenwoordige Staat der Vereenigde Nederlanden; dertiende deel; vervattende eenen aanvang der beschryving van Friesland (Amsterdam). Schulte, Michael 2018, Urnordisch: eine Einführung, Wiener Studien zur Skandinavistik 26 (Wien). Schwarz, E. 1951, Goten, Nordgermanen, Angelsachsen. Studien zur Ausgliederung der germanischen Sprachen (Bern–München). Siebs, Theodor 1889, Zur Geschichte der englisch-friesischen Sprache (Halle an der Saale). Siegmund, Frank 2003, ‘Social relations among the Old Saxons’. In The Continental Saxons from the Migration Period to the Tenth Century: An Ethnographic Perspective, eds. Dennis Howard Green and Frank Siegmund, Studies in Historical Archaeoethnology (Woodbridge), 77–95. Stiles, Patrick 1995, ‘Remarks on the “Anglo-Frisian” thesis’. In Friesische Studien II. Beiträge des Föhrer Symposiums zur Friesischen Philologie vom 7.–8. April 1994, eds. Volkert F. Faltings, Alastair G. H. Walker and Ommo Wilts (Odense), 177–220. Stiles, Patrick V. 2013, ‘The pan-West Germanic isoglosses and the subrelationships of West Germanic to other branches’, NOWELE. North-Western European Language Evolution 66 (1), 5–38. Swarte, Femke, Nanna Haug Hilton and Charlotte Gooskens 2013, ‘Onderlinge verstaanbaarheid tussen Noord- en Westerlauwers Fries’, Us Wurk 62 (1–2), 21–46. Tiefenbach, Heinrich 2010, Altsächsisches Handwörterbuch / A Concise Old Saxon Dictionary

372

Z01 Hines Book B.indb 372

07/05/2021 13:30

North Sea Germanic Idiom (Berlin). Todd, Malcolm 2004, The Early Germans (Malden–Oxford). Trudgill, Peter 2010, Investigations in Sociohistorical Linguistics: Stories of Colonisation and Contact (New York–Cambridge). Udolph, Jürgen 2006, ‘England und der Kontinent: Ortsnamenparallelen’. In Language and Text: Current Perspectives on English and Germanic Historical Linguistics and Philology, eds. A. J. Johnston, F. von Mengden and S. Thim (Heidelberg), 317–43. Van Berkel, Gerald and Kees Samplonius 2018, Nederlandse plaatsnamen verklaard (Amstelveen). Versloot, Arjen Pieter 2011, ‘Frisian place-names and place-names in Friesland’, Onoma 46, 127–50. — 2014a, ‘The Runic Frisian vowel system: the earliest history of Frisian and Proto-Insular North Frisian’, Amsterdamer Beiträge zur älteren Germanistik 72, 35–62. —2014b, ‘Die –ar-Plurale im Altwestfriesischen mit einem Exkurs über die sächlichen Plurale im Westfriesischen’, Us Wurk 63, 93–114. — 2017a, ‘Proto-Germanic ai in North and West Germanic’, Folia Linguistica 51 (s38), 281–324. —2017b, ‘The Riustring Old Frisian -ar plurals: borrowed or inherited?’. In Frisian Through the Ages. Festschrift für Rolf Bremmer, eds. M. de Vaan and S. Laker, Amsterdamer Beiträge zur älteren Germanistik 77 (Leiden–Boston), 442–56. —2018, ‘Frysk út Noard-Hollân út de 17e iuw?’, Us Wurk 67, 1–37. — forthcoming a, ‘Reconstructing the historical phonology of Old English I’, English Language and Linguistics. — forthcoming b, ‘The historical phonology of Old English: a critical review’, Papers in Historical Phonology. Versloot, Arjen Pieter and Elżbieta Adamczyk 2017, ‘Geography and dialects in Old Saxon: river-basin communication networks and the distributional patterns of North Sea Germanic features in Old Saxon’. In Frisians and their North Sea Neighbours. From the Fifth Century to the Viking Age, eds. J. Hines and N. IJssennagger (Woodbridge), 125–48. Versloot, Arjen Pieter and Michiel de Vaan in preparation, A Grammar of Proto-Frisian; the Beginnings of the Frisian Language. Vonderach, Andreas 2015, Anthropologie Europas: Völker, Typen und Gene vom Neandertaler bis zur Gegenwart, 2nd edition (Graz). Vries, Oebele 2007, Asega, is het Dingtijd? De Hoogtepunten van de Oudfriese Tekstoverlevering (Leeuwarden).

373

Z01 Hines Book B.indb 373

07/05/2021 13:30

Z01 Hines Book B.indb 374

07/05/2021 13:30

• 13 • RUNIC LITERACY IN NORTH-WEST EUROPE, WITH A FOCUS ON FRISIA Tineke Looijenga

T

his chapter discusses runes and runic writing in the Early Middle Ages in general, and in Frisia in particular. Objects with runes on them have been found in the Frisian terp-area dating from around the sixth to the ninth century. In this chapter I aim to discuss how we should assess ‘Frisian’ runes and their relation with other runic traditions. Although the oldest known runic objects date from the second century, it is assumed that the runic alphabet was created sometime in the first century AD.1 But why, when and where runes were developed from an archaic Mediterranean model is still an unsolved question. Runic writing, the exclusive indigenous script of Germania, emerged in a period when Germanic-speaking people grew increasingly in touch with the Roman imperial world, a literate world. Many Germanic men served as mercenaries in the Roman army. They received Roman citizenship after twenty-five years’ service, and we may assume that some of them returned home after this service. Because this citizenship was hereditary, the sons of these auxiliaries were also Roman citizens, even before entering the army. Some of them made a career in the army, and learned to read and write. This became increasingly common in the second and third centuries AD, exactly the period in which we infer runic knowledge to have spread across a large part of northern Europe (Stoklund 2006, 358). Most scholars now believe that runes were based on a Mediterranean alphabet, most likely the roman/Latin alphabet.2 I believe they will have been devised in close connection with Roman culture and possibly on Roman territory. The obvious context would then be creation through one of these Germanic soldiers, with their literary training in the army and the implicit ability to create a script of their own (Derolez 1998, 26; Pollington 2016, 79). The survey produced by René Derolez (1998, 5–6) contains the observation that ‘there was a near-perfect agreement between the runes of the futhark and the phonemes of Germanic as reconstructed for the beginning of the period under consideration. Could such a phonemic fit be the result of a gradual adaptation or does it presuppose a conscious and systematic arrangement?’. And Antonsen (1996, 11)

1 2

The oldest known runic inscriptions date from the second century AD. As it can hardly be supposed these were the first to be made, the creation of the runes is assigned to ‘some time between the birth of Christ and the first half of the second century AD’ (Barnes 2012, 9). There are other candidates, such as the Etruscan alphabet or one of the Alpine scripts.

375

Z01 Hines Book B.indb 375

07/05/2021 13:30

Frisians of the Early Middle Ages claimed that ‘no alphabetic writing system can survive from one generation to the next without a spelling tradition’. Perhaps we should consider the possibility of one or more Germanic individuals being taught in rhetorica and grammatica in Rome. Germanic soldiers served, for instance, as bodyguards for the Imperial family, so this supposition is not so far-fetched as it might seem. They may have passed their knowledge on – and introduced it to their homeland – but somehow and for unknown reasons it was not the roman alphabet that came into use in Germania but a variant of some archaic alphabet such as existed in the Alps and northern Italy. We do not know what the story was, but runic evidence first appears far away from the Limes, in Scandinavia, with a concentration in Denmark. Interestingly, futhark inscriptions (the complete runic alphabet) seem to occur around the periphery: in Bosnia, in Hungary, in Thuringia, in Burgundy, near London, and on Gotland. A separate group are the fifth-century bracteates with a futhark or part of it. Does this mean that copies of the whole or partial futhark had some specific purpose in the spread of literacy? Runic inscriptions appear in Scandinavia from the second century onwards on weapons, weapon parts, jewellery, tools, utensils, combs, personal gear, bracteates and coinage. But we should note that also ‘from the second century onwards, Roman military items began to appear in southern Scandinavia including sword-blades with stamped factory marks; alongside these are a few examples of terra sigilata ceramics and a large quantity of Roman coins’, according to Pollington (2016, 70, quoting Imer 2010). Runes came to be known and used over a large area in Europe, at first in Scandinavia, from where they spread to the south-east of Europe,3 to Poland, Pannonia, Romania and Ukraine, along with groups such as the Goths in the fourth century. From the fifth century onwards runic use also spread to western and central Europe: from Jutland or Schleswig-Holstein along the coast to the Elbe-Weser triangle (Aufderhaar 2017), to Frisia and on to Anglo-Saxon England. From the sixth century onwards runic objects appear in Merovingian France and southern Germany. It is not quite clear along which routes knowledge of the runes will have travelled. For instance, runologists have wondered why runic practice came rather late (the sixth century) to the south of Germany, where the remarkable quantity of nearly a hundred runic objects has been found found, dated to within about 150 years. But since the recent runic finds in Merovingian France from the first half of the sixth century (the ring-swords of Saint-Dizier, Grenay and Fréthun: Fischer and Soulat 2010), it seems conceivable that runic knowledge may have come to southern Germany from the west, from Francia (see Fig. 13.1). The current idea (Düwel 1996) is that runes could have reached southern Germany only from the north, and only after the fall of the Thuringian kingdom (AD 531/2). But that kingdom was conquered by the Franks. The sixth and seventh centuries were the runic prime time in Europe; subsequently runes only remained in use in Scandinavia and in England. In the first centuries of runic writing, the use of runes seems to be reserved for a small exclusive group of people, one could call them an elite. To answer the question of who was literate in those days in Germanic-speaking society, we can only guess; there must have been some literates because those runes were written and read by people 3 The Frienstedt comb is dated to the third century, and was found near Erfurt. It has a reading kaba interpreted as WGmc for ‘comb’. It is a small portable object, and this type of comb is known from a large area, including the northern coasts. In fact, all the early runic objects are portable, so their findplace and origin may be quite different.

376

Z01 Hines Book B.indb 376

07/05/2021 13:30

Fig. 13.1  The ‘Frisian’ and ‘Frankish’ runic items, part of a Continental North Sea group. Drawing: RuneS.

Z01 Hines Book B.indb 377

07/05/2021 13:30

Frisians of the Early Middle Ages – but by whom, and by how many? Derolez (1998, 5) noted that ‘we have no more than a couple of hundred inscriptions spread over five centuries’. Katrin Lüthi (2006, 169–82) has analysed the ‘South Germanic’4 runic inscriptions in respect of the level of literacy among the people who wrote these runes and concluded that their writing skills are comparable to those of children learning to write, including the kind of mistakes and writing imitations. This is in fact the situation in all regions where runic writing was practised in the early centuries. If we suppose only a select few were able to read and write, this form of literacy may have served as a means for cultural and elite display. This is consistent with runic use in the first centuries in Denmark, and also in later centuries in southern Germany, Francia, Frisia and England. Here we find signs of ownership, marking personal items, and expressing personal relations and gift-and-exchange customs. We find runic objects used as charms and amulets everywhere. It may be useful to point out that there was almost certainly no role for script as an administrative instrument, since administration and state building among the Germanic nations was still a long way off. For what, then, did they develop a script? Just in imitation of Roman practices? That seems partly to be so: the practice of marking weapons and small personal objects with names, even if in some cases the inscription was not meant to be read by the public, resembles everyday Roman inscriptions on earthenware and personal belongings.5 Even if this supposition is true, it does not explain why Germanic people preferred runes above roman script – or was that just because of a feeling that their own language needed its own writing system?

Literacy Anyone involved in runic studies is confronted with the question: who could read or write runes in the Roman Imperial Period and the Early Middle Ages? How ordinary was this script, and what does it mean that we have so few attestations left? The question at this symposium regards especially people living in Frisia after the habitation gap of the fourth century AD. This is because to determine to what extent Frisians were literate depends on the period you study; in Roman times one may suppose that some Frisians could at least read and perhaps also write Latin, as is shown by inscriptions found near Hadrian’s Wall, presumably made by Frisian auxiliaries in the Roman army (Galestin 2007/2008 provides a comprehensive overview of Roman inscriptions mentioning Frisii and/or Frisiavones). Some classical authors wrote about Frisia, situated approximately between the estuaries of Meuse and Ems along the continental shore of the Oceanus Germanicus. Pliny the Elder, Tacitus, Cassius Dio, Ptolemy and Pomponius Mela provided limited information and say nothing about any literacy. Pliny actually visited the area. Epigraphic examples in Frisia in Latin or an indigenous tongue have not been found, except for the first-century wax tablet of Tolsum, in Latin. We have some names of Frisian kings, Malorix and Verritus, recorded by the Romans. Tacitus, for instance, mentions a certain Cruptorix (Annales IV, 73) and a goddess Baduhenna 4 5

Südgermanisch is a conventional term in runic studies for the Continental inscriptions, mostly in what was probably the ancestor of Old High German. The Roman practice of putting up memorial stones with elaborate texts meant to be read by the public might be mirrored by the large, Viking-age stones with runic inscriptions found in Norway, Denmark and Sweden.

378

Z01 Hines Book B.indb 378

07/05/2021 13:30

Runic Literacy in North-West Europe apud lucum quem Baduhennae vocant (‘at the grove which is named after Baduhenna’). As Schrijver (2017, 44) points out, these names do not appear to be Germanic names but Celtic. One of the names mentioned on the Tolsum tablet may also be Celtic: Caturix, slave of one Iulia Secunda (Galestin 2009/2010, 9–26). This means that we probably have to regard those Roman-period Frisians as a Celtic-speaking community. When we look at the newly inhabited Frisia after the Roman Period, and see what traces of literacy the post-Roman Frisians have left, we only have a few runic inscriptions. One of the first remarkable observations is that the Frisian runic texts have little in common with the kind of texts the Roman-period Frisians produced (or had produced for them), such as we find on altars near Hadrian’s Wall or elsewhere in Roman territory. However, the oldest runic texts we find on weapons and personal belongings have much in common with Roman graffiti: names mostly, and makers’ inscriptions. Monumental and public texts, so common for Roman society, are completely lacking in the early runic period.6 Only in later centuries, from about the seventh century onwards, do monumental runic texts on stone appear, almost exclusively in Scandinavia. (One might consider the runic inscriptions on the so-called megaliths as having inspired by Roman culture, but the kind of texts they bear is divergent. Take for instance Noleby: runo fahi raginaku(n)do ‘a rune I mark, from the advisory (gods)’, or otherwise Einang: [ek go]daga[s]tiz runo faihido translated as ‘I Godagast marked a rune’. These texts have nothing to do with the Roman kind of texts.) According to Bernard Mees (2015, 2), early epigraphic texts produced by ancient societies just adopting literacy tend to fall into standard types, such as makers’ inscriptions on objects like weapons and jewellery. Further, there are commemorative epigraphs and magico-religious texts. ‘The restriction of inscriptions to these three basic categories is common not just to Old Latin and archaic Greek tradition, but also to Etruscan, archaic Italic, and archaic Celtic or Lepontic epigraphy.’ These categories occur in the runic tradition as well, and also in the Old Frisian texts. So far, nothing out of the ordinary: we can only conclude that the Frisian tradition is no exception. Frisian runic texts such as Westeremden A: adugislu meth gisuh[i]ldu, two personal names, can be

Fig. 13.2  Westeremden A. A ‘weaving-slay’ (or ‘reed’), inscribed with the runes adujislu me(þ)gisuhldu. Groninger Museum (Photograph: Heinz Aebi).

6

Inscriptions with mixed Latin/Roman-vernacular contents, such as can be found in Gallo-Roman Gaul after the Roman conquest, are also lacking in the Germanic realm, in all probability because Germania was never conquered by the Roman Empire.

379

Z01 Hines Book B.indb 379

07/05/2021 13:30

Fig. 13.3  Runes inscribed on the plates of the Oostum comb. (a) aib kabu; (b) deda habuku. Groniger Museum (Photograph: Heinz Aebi).

Z01 Hines Book B.indb 380

07/05/2021 13:30

Fig. 13.4  A miniature yew-wood sword from Arum inscribed with the runes edæ boda. Photograph: Fries Museum, Leeuwarden. Collectie Het Koninklijk Fries Genootschap.

Fig. 13.5  The word kobu (‘comb’) inscribed in runes on the Toornwerd comb. Groninger Museum (Photograph: Heinz Aebi).

Z01 Hines Book B.indb 381

07/05/2021 13:30

Frisians of the Early Middle Ages

Fig. 13.6  Westeremden B. A stick of yew-wood, inscribed with runes. Groninger Museum (Photograph: John Stoel).

compared to many Scandinavian, South Germanic and Anglo-Saxon ‘commemorative’ inscriptions (Fig. 13.2). Also Oostum, aib kabu, deda habuku translated as ‘Habuku (modern speech would have Habbeke) made the comb for Aib’ (‘Aib made the comb for Habbeke’ is equally possible), is one of very many so-called makers’ inscriptions found all over the runic world (Fig. 13.3a–b). The Wijnaldum bone piece, with an inscription in ‘ornamental’ runes, may be read as inguz, the name of a Germanic god, and might have been a kind of amulet. That is the extent of potentially magico-religious inscriptions in Frisia. The Arum miniature wooden sword (Fig. 13.4) has eda boda, which may be taken as a means to summon people to come to a trial and take oath. An owner’s inscription may be found on the Ferwerd comb-case with the inscription em ura or me ura, to be taken as ‘I am from Ura’, or possibly ‘Ura has me’. We also have more personal names on coins: Hada, Welad, Skanomodu and Æniwulufu, plus appellatives like ‘comb’ on a comb – Oostum kabu and Toornwerd kobu (Fig. 13.5) – as well as Hamwic katae on a knucklebone, meaning exactly what it is: ‘koot’ in Dutch. Only the long Westeremden B inscription (Fig. 13.6) is absolutely out of the ordinary; the reason why it was supposed to be a forgery. After investigations into the weathering of the wood by the archaeologist and wood specialist Dr Casparie and the art historian Professor Van Asperen de Boer there is no reason to regard it as such (Looijenga 1991). Nevertheless, Gaby Waxenberger prefers to keep it out of her investigations, because ‘the inscription is at least partially in a poor state, particularly regarding the critical graphemes’ (2017, 97). The inscription does indeed show some extraordinary graphs, and seems to have been influenced by Scandinavian younger futhark runes, but some ‘mirror runes’ may be present, otherwise found in East Anglia (Spong Hill) and on some bracteates. Whatever its problems, the Westeremden B inscription needs more serious attention.

Runic writing in Frisia (and beyond) As has been argued widely (for instance, Taayke 2003; Knol 2009; Gerrets 2010; Lanting and Van der Plicht 2009/10; Nieuwhof 2016; Knol and IJssennagger 2017), Frisia was virtually depopulated in the fourth century and became inhabited again after a hiatus of about a century – apparently by people who migrated from the east, along the coast, or perhaps also from the Pleistocene sandy hinterland to the south. In any event, this new population was Germanic-speaking. The written evidence is from runic inscriptions carved in objects that became buried in dwelling mounds (terpen or wierden) built in the marshes against the tides. This is the only kind of literacy we have from the post-Roman pre-Christian Frisians, until the writing down of the Old Frisian lawbooks 382

Z01 Hines Book B.indb 382

07/05/2021 13:30

Runic Literacy in North-West Europe in the twelfth–fourteenth centuries. Other written sources about the Frisian population derive from missionaries and historians and are not discussed here (see Wood, and Hines, this vol.). The runic period in Frisia may have lasted some centuries – but dating the inscribed objects is difficult since they are all stray finds from the terpen and no context has been recorded. We only have observations from people who attended the commercial digging of fertile soil from the ancient dwelling mounds in the nineteenth and twentieth centuries. There are, however, also a few objects considered ‘Frisian’ not found in these mounds, but found elsewhere, in England, Belgium and Ostfriesland. Some ‘Frisian’ inscriptions show peculiarities that are supposed to be characteristic of the Frisian language and the runes of the Early Middle Ages. I will come to that subject further on. Runic literacy in Frisia is limited and enigmatic. Here we find the difference of langue and parole, terms distinguished by Ferdinand de Saussure. It is not so much langue, the principles of language, we have to take into consideration but parole, the concrete instances of the use of language.

The Frisian corpus Let us first review what objects we are talking about. There are nineteen in total: four gold coins, six combs, four objects made of yew wood, three objects made of whale bone or whale ivory, one knuckle bone and an antler point.7 Furthermore there are many silver coins (sceattas) and one gold bracteate, probably an import. The bracteate is an unlocated find from the Hitsum terp, but stylistically similar to a bracteate from Sievern (Nicolay 2014, 155–6). Boeles (1951, 325) recorded another bracteate found in the terp at Achlum with a longer runic inscription and unfortunately sold to an unknown person in 1944. Two other runic objects found in the Netherlands must be noted: a silver-gilt scabbard mount from the Rhine-Meuse estuary near Tiel and a copper-alloy belt buckle from a Merovingian burial site on the Meuse just north of Maastricht. The scabbard mount was part of a large assemblage of metal objects, maybe the stock of a smith or part of a (votive) deposit. The belt buckle is one of the few runic grave finds from the Netherlands and is an example of a widespread type of buckle. The name on the buckle, Bobo, could be Frankish or Alemannic. I shall deal with these objects separately (below), since they cannot be counted as ‘typically Frisian’ as far as their runes or language goes. ‘Typically Frisian’ refers to the use of certain rune-forms and diagnostic linguistic features. The runic objects found in the terp-area are anomalous when compared with objects from Denmark, Germany or England. There we sometimes find runes on precious objects such as gold and silver brooches, personal outfits, tools, weapons and bracteates. The Frisian runic corpus is rather eclectic as regards the range of objects. We have humble objects made from unremarkable materials such as bone, antler and wood, but also a remarkable whale-bone staff with a nearly eroded inscription: only the name 7

In earlier surveys a golden pendant from Wijnaldum is mentioned, with a runic inscription hiwi, but this object should be deleted from the list. On closer inspection the alleged scratches are not runes at all. Although not discussed in detail in this paper, for convenience sake, images of the inscribed Britsum yew-wood stick and Bernsterburen tau-staff of whale bone have been included at the end as Figures 13.14 and 13.15.

383

Z01 Hines Book B.indb 383

07/05/2021 13:30

Frisians of the Early Middle Ages

Fig. 13.7  Ring-sword from Saint-Dizier, inscribed with the runes alu on the pommel. Photograph: S. Culot, Inrap.

Tuda is certain and possibly the verbal form kius ‘chose, selected’. Further we have a whale-bone sword handle, with a likewise eroded inscription: here only the name Oka is certain. The ‘Frisian’ gold coins are striking, although it cannot be ruled out that they were in fact Anglo-Saxon products – we shall return to his tricky subject later on. But all objects were inscribed in a notably careful way. As noted, all of the terp finds are stray finds, without precise contexts. In Denmark, England, Francia and Germany, runic objects emerge from graves or ritual deposits such as votive bog finds. The great majority of the early Anglo-Saxon runic finds in the UK are grave finds. There are almost no Frisian runic objects from graves, since cemeteries are relatively scarce in the terp-area.8 It should be noted that most ‘Frisian’ runic objects were found in the dwelling mounds – terpen and wierden – which were most favourable for preserving the organic objects. They are in fact settlement finds. If we did not have those mounds, we would have a runic landscape comparable to that in France: only a number of metal pieces found by metal detectors, because of road works, or in the occasional dig. France has three runic ring-swords (from graves); one bow brooch (from a grave); and one disc brooch (from a grave). The ring sword from Saint-Dizier, found in 2002, has a legible inscription: alu, a well-known, but not fully understood word, found only in runic inscriptions (Fig. 13.7; Truc and Paresys 2010). The Netherlands would then have one gilt-silver scabbard mount from a hoard or a smith’s stock, one copper-alloy belt buckle from a grave, and four possibly Frisian gold coins. One of those is from an antique shop at Harlingen: in de buurt van Harlingen gevonden, kennelijk in een terp (‘found in the region of Harlingen, apparently in a terp’: Boeles 1951, 345). Another gold coin belonged to a royal collection in England (possibly originating from Hanover); the third is an unlocated find from a field in Ostfriesland, Germany. The gold tremissis from Folkestone is now lost, and a similar one (see below, Fig. 13.10) in the Hunterian Museum (Glasgow) was possibly bought on the Continent, and appears to be modelled 8

The comb from Hogebeintum is from a grave assemblage, and the comb from Amay (Wallonia; Fig. 13.8) is also from a grave find although far away from the terp-area. This grave was part of a a Merovingian row-grave cemetery close to the Meuse, south of Liège.

384

Z01 Hines Book B.indb 384

07/05/2021 13:30

Runic Literacy in North-West Europe upon a Merovingian tremissis from south-western France (Blackburn 1991, 143). In these circumstances, we would have described the runic landscape of the north-western part of the Continent from Ostfriesland to Burgundy as ‘Merovingian’. Nothing mirrors the uncertainty of assigning runic finds to an ethnic or cultural entity better than this situation. This conclusively demonstrates that one should be very careful in assigning specific runic knowledge to certain areas as opposed to other areas which seem to be ‘empty’.

What kind of inscriptions do we have? We have names and appellatives, of persons and objects, which is typical for (early) runic (and other archaic) inscriptions in general. An outstanding example of a long, elaborate runic text is found on the small wooden stick of Westeremden, which seems to express some kind of a call for good fortune. According to the Elfde en Twaalfde Jaarverslag van de Vereniging voor Terpenonderzoek over de verenigingsjaren 1 mei 1926 tot 31 april 1927 en 1 mei 1927 tot 31 april 1928, this object was found by the mound diggers J. de Vries and H. Boukema in 1917. A very accurate description of the findplace is given (ibid., 52), and also an indication of the date of deposition: no earlier than the fifth century, and no later than the eighth, because in the layers above the runic stick some Saxon pottery was found that could be dated to the eighth century. Interesting is the remark by Van Giffen, the excavator, that it was found in the vicinity of a small burial ground. The combs According to Düwel (2018, 325), there are fifty runic combs (an extraordinary tally for one category of inscribed object). A separate sub-group is the combs inscribed with a word meaning ‘comb’, found along the coastal areas of the North Sea, from Frisia to Ribe in Jutland, Denmark. These are five: Oostum and Toornwerd in Frisia (above, Figs. 13.3 and 13.5), Elisenhof in Schleswig-Holstein, Ribe in Jutland, and Frienstedt near Erfurt. More combs have a runic inscription although not the word ‘comb’: Kantens, Fer­werd, Hogebeintum (all Frisia), (Wallonia) (Fig. 13.8), Whitby (Yorkshire, England),

Fig. 13.8  The Amay comb, inscribed with runes on the left, alongside the break. Photograph: Musée Le Grand Curtius, Liège.

385

Z01 Hines Book B.indb 385

07/05/2021 13:30

Frisians of the Early Middle Ages Vimose (Fyn, Denmark), Lauchheim (Baden-Württemburg), Setre (Norway), Belgorod (Russia). Combing one’s long hair may have had a special meaning since long hair was a privilege among high-standing Frankish men (cf. Childeric’s signet ring with his long-haired face on it, and the famous gravestone from Niederdollendorf at the Landesmuseum in Bonn). Writing ‘comb’ on a comb could have meant something specific. The coins There are four gold coins with runes: the hada, welad (Looijenga 2013) and skanomodu solidi plus the æniwulufu tremissis, often ascribed to the Frisians, although sometimes an Anglo-Saxon source is considered (Figs. 13.9–13.10). These solidi were not meant as everyday cash. A general market economy with its central places and emporia was yet to develop. Before the end of the seventh century, the demand for coins to buy goods in regular transactions caused the introduction of a silver coinage, the sceattas. A gold coin such as the solidus had too high a value to be used in such transactions. Solidi were used as payment to soldiers, or to pay taxes, or they were used as precious gifts. A high-ranking warrior under Theoderic the Great earned three solidi per year. ‘There was a gradual devaluation of the post-Roman solidus. In the middle of the seventh century the solidi contain only 10% gold and 90% silver. Perhaps this was done as a solution to obtaining coins that could be used in the market.’ (Steuer 2003, 186). The four runic gold solidi mentioned above show different levels of gold fineness. The skanomodu coin has the highest measured gold content, at 90.5%. This coin may be dated ‘partly by type but more confidently by gold fineness, no later than the early seventh century’ (Hines 2017, 31, referring to Williams and Hook 2013). The hada and welad solidi can be dated to around 640, according to Hines. They contain c. 50–73% gold and are thus probably a little younger than the skanomodu coin. The fourth gold coin, the æniwulufu tremissis, has not yet been analysed to the same standards. These runic coins may have served other purposes than a medium for payment; they may have been gifts, the more precious because of the runes. The skanomodu solidus once had a loop for suspension. In all four cases the inscriptions show names, which, as Hines demonstrates, could be Frisian male names. But one may ask: is this so? Couldn’t these names be Merovingian? There is a tendency to add an ending -us to indigenous names, cf. LEVDARDVS (Bishop Liudhard), VANIMVNDVS (mint-master, listed by Hook and Williams 2013 and mentioned by Blackburn 1991, 145). The æniwulufu coin is a copy of a Merovingian coin; according to Blackburn (1991, 143), ‘The design and legends copy those of a Merovingian tremissis from south-western France, with the insertion into the obverse legend of runes … The Kentish provenance (Folkestone) for the coin would not preclude a Frisian attribution.’ I wonder why a Merovingian provenance has not been considered? Because of the runes there has been a desire to assign it to either the Anglo-Saxon or the Frisian runic corpus, understandably so since France was not, when Blackburn was writing that article, in the picture as an area where runes were used. As John Insley (1991, 173) observed in the same volume, in an appendix to Blackburn’s paper: ‘[T]he first element of the name is western Ingvaeonic */aun-i-/, a side-form of the etymologically obscure West Germanic name element */aun-a-/ contained in such names as Frankish Aunobertus, Aunogisil, Aunecharius, Aunemundus, Aunulfus’. As regards the ongoing discussion of the ‘typically Frisian -u endings’ in personal names, we should note that Insley declares that the final -u in the 386

Z01 Hines Book B.indb 386

07/05/2021 13:30

Fig. 13.9  The ‘Schweindorf ’ runic solidus. Photograph: Christina Kohnen, Ostfriesische Landschaft.

Fig. 13.10  The æniwulufu tremissis. Hunterian Museum, Glasgow. Photograph: author.

Z01 Hines Book B.indb 387

07/05/2021 13:30

Frisians of the Early Middle Ages supposedly ‘Frisian’ names can plausibly be explained as the Old Saxon instrumental case: ‘there is no reason why the language of the Frisian runic inscriptions should not have shared this feature [the retention of the instrumental -u in the Gmc a-declension] with Old Saxon’ (Insley 1991, 174). The Bergakker and Borgharen inscriptions The Bergakker silver-gilt scabbard mount was found with a metal detector in 1996 near Tiel, between the rivers Rhine and Waal (Fig. 13.11). It is dated to c. AD 425. It has a runic inscription on the reverse. The runes are of the older futhark but it also contains an anomalous rune formed of a double chevron which most probably denotes the vowel e or u. Uncertainty about its transliteration means there is no agreed transcription and/ or interpretation. I propose to read the inscription as haþẹþẹwas: ann: kẹsjam: logẹns, with the anomalous sign interpreted as e (Looijenga 2003, 319). Several others prefer the reading u (e.g., Bammesberger 1999). This, however, is not the place for discussing such matters in detail. According to its ornamentation, in provincial Roman style, the mount seems to belong to a group of swords from northern Gaul into the Lower Rhineland, the area known as Toxandria, a territory of the Salian Franks. The runic inscription might therefore be considered Frankish. The same identity may be assigned to the belt buckle found in 1999 in a Merovingian cemetery near Borgharen on the Meuse, just north of Maastricht (Fig. 13.12). This has been dated to the third quarter of the sixth century. Its runes are unproblematic and read bobo. It might have belonged to a Frankish miles settled in the area (Looijenga 2003, 323).

Fig. 13.11  The Bergakker scabbard mount. Photograph: Courtesy of the Museum Valkhof, Nijmegen.

Fig. 13.12  The Borgharen belt buckle, front, inscribed with the runes bobo. © Centre Céramique, Maastricht.

388

Z01 Hines Book B.indb 388

07/05/2021 13:31

Runic Literacy in North-West Europe

The range of runic literacy If we consider the distribution of runic objects from the period of the fourth–fifth centuries, we may get an idea of the dispersal of runic knowledge. It has been noted that the oldest known objects appear in Scandinavia, especially Denmark, and Schleswig-Holstein. They are in fact followed by a small group of objects along a route towards the south-east of Europe in the third century: the spearheads of Dahmsdorf, Rozwadów and Kowel (in Brandenburg, Poland and Volhynia, respectively). Then, from the fourth and fifth centuries there is the spindle-whorl of Leţcani, Romania (fourth century), the silver buckle of Szabadbattyán, Hungary (early fifth century), the silver buckle of Sukhodil/Shydlivtsy, Ukraine (early fifth century), the comb of Belgorod, Russia (fifth century), and the gold neck-ring of Pietroassa, Romania (early fifth century). Towards the south and south-west, meanwhile, there is the silver neck-ring of Aalen, Baden-Württemberg (fifth century), and to the west and south-west the bracteates from Sievern (fifth century)9 and the Fallward footstool (early fifth century), both in Lower Saxony, the Bergakker scabbard mount from the Rhine/Meuse estuary (early fifth century, Fig. 13.11), the Loveden Hill urn, Lincolnshire (fifth or sixth century), the Spong Hill urns,

Fig. 13.13  A core area in which runes occur to the south and south-west of Scandinavia in the fifth and sixth centuries. 9

The diffusion of bracteates occurred in the fifth century in the same regions, mainly in northern Europe and the countries around the North Sea, but also some in southern Germany.

389

Z01 Hines Book B.indb 389

07/05/2021 13:31

Frisians of the Early Middle Ages Norfolk (fifth century), and the Caistor-by-Norwich astragalus, Norfolk (early fifth century). The picture roughly presents two groups, one expanding south-east towards the ‘Gothic’ area of the Çernjachov Culture, and one extending west and south-west, the route of the Anglo-Saxon and Frankish expansion (Fig. 13.13). In the sixth century the diffusion reaches its maximum extent: Frisia, Francia, England, southern Germany, Switzerland, Bosnia and Pannonia. It seems that we may identify two main groups carrying runic knowledge with them: Goths on the one hand, and Angles, Saxons, Jutes, Frisians and Franks on the other. If no one other than these groups were responsible for this diffusion, that could account for the widely noted uniformity and continuity of runic writing, and the linguistic unity as well (Derolez 1998, 5–6), and also for the phonemic fit; the standardization of, inter alia, word-division markers, the standardization of the futhark, and the acrophonic principle. But if we take into account the poor quality of the texts themselves, the near absence of eloquency, and the very limited choice of words and expressions (formulae) we do not get an impression of great literacy. If these two groups were the bearers of runic knowledge through Europe, they may not be seen as the bearers of a literate culture, but as those who ‘sich vielmehr im Stadium der “begrenzten Literalität” befanden’ (‘found themselves rather in the state of “limited literacy”’: Lüthi 2004, 335).

Conclusion There is still no complete corpus edition of the Anglo-Saxon and Frisian runes. In order to compile such a useful study, I would plead for a broader context for studying the Frisian runic reality than is commonly in use. I would propose to include in analysis and discussion the whole Continental region of the North Sea coast and the Channel from the mouth of the Elbe to the mouth of the Somme, and also the coastal regions of Britain in East Anglia and Kent opposite. This wide coastal area is the stage where all kinds of interactions between the peoples of the Late Roman Period and the Early Middle Ages took place (cf. Wood 1990; Lebecq 1990; Soulat 2014). That means we ought to study the runic finds from the Continental area between Sievern/Fallward in Lower Saxony and Fréthun and Grenay in Pas de Calais, and across the North Sea the area from Southampton to the Humber, as a whole. And we should also include the ring-sword from Saint-Dizier and the Bergakker scabbard mount. I consider the latter to be of Frankish origin as far as its ornamentation goes, and as regards the runes I would argue these were made by a Frank or a Saxon, based upon the observance that the runic inscription of the Fallward footstool has some peculiarities in common with

Fig. 13.14  A stick of yew-wood from Britsum with double carved runes. Photograph: Fries Museum, Leeuwarden. Collectie Het Koninklijk Fries Genootschap.

390

Z01 Hines Book B.indb 390

07/05/2021 13:31

Fig. 13.15  (a) An inscribed whalebone tau-staff from Bernsterburen; (b) runes reading tuda on the Bernsterburen staff. Photographs: Fries Museum, Leeuwarden. Collectie Het Koninklijk Fries Genootschap.

Z01 Hines Book B.indb 391

07/05/2021 13:31

Frisians of the Early Middle Ages the Bergakker inscription – such as the writing of double consonants, not otherwise normal runic practice). I would propose to call this runic area the North Sea/Channel group. In general one may say that the question of who was who in the fifth century is quite confusing. Saxons, Frisians and Franks are frequently mixed up (Wood 1990; Seebold 2003). Merovingian objects occur in graves of south-west England, and Anglo-Saxon objects occur on the other side of the Channel in equally copious numbers (Soulat 2014, 94–5). The Bergakker piece was found in the area of the Lower Rhine and Meuse (Betuwe), which was re-settled around AD 425 by a migrant group (Heeren 2017, 158– 63). The area had apparently become depopulated a century earlier. That the migrants came from the north is argued by their house-building techniques and the growing of rye. Only along the Limes were there surviving settlements, all belonging to military fortifications and associated with foederati (Heeren 2017, 166).10 This is the context of the Bergakker find, and also of the runic buckle from Limburg. To place the Frisian runes in a broader context and to study the connectivity of the runic area of the entire North Sea–Channel region will provide more insight into runic literacy. It appears from the archaeological and historical evidence that the whole region was, in a dynamic sense, in close contact and interaction. In order to get a complete picture, one must therefore include all finds from both sides of the sea. Literacy and runic use in these areas may reveal greater cohesion in terms of reading and writing in the transformational period of the fifth–seventh centuries.

Discussion LOOIJENGA  I would like to investigate the possibility of seeing the oldest runic inscriptions as part of the decoration of the object – sometimes the combination of runes and ornamental lines make the impression of being part of a single design. Of course the runes offer a text too – in nearly all cases a name. NIEUWHOF  It had already struck me that there are some indications in the material which seem to me possible signs of gift-exchange: some wordings you mention. For example, the inscription on a comb from Oostum translates as: ‘Habuku made the comb for Aib’. Another, on a comb-case from Ferwerd, as: ‘I am from Ura’. Or the one from Freilaubersheim, that ends with ‘greetings’. Such inscriptions would be very understandable in the context of gift-exchange. We were talking about the family archive when we discussed my paper (Ch. 3). Well, these seem to me the kind of objects that end up in family archives. There are stories attached to them, about from where they come, who made them, and by whom they were given. That may be the explanation for part of the puzzling runic inscriptions, though of course not for all. What do you think of that? 10 ‘In short, the material culture of the fifth century contains Roman elements and traditional Germanic forms but, most importantly, new forms were created and used well into the Merovingian period. Instead of referring to the mix of Roman and Germanic elements in terms of assimilation as is previously the case, it might be better to see this material culture as an independent hybrid group and to acknowledge that most of it developed in the fifth century in the context of newly inhabited territories in the former Roman province. … In short, an essentially post-Roman society took shape in the northern part of the former Roman province and adjacent area north of the Lower Rhine, in close conjunction with and inseparable from the decline of Roman authority.’ (Heeren 2017, 166).

392

Z01 Hines Book B.indb 392

07/05/2021 13:31

Runic Literacy in North-West Europe LOOIJENGA  Objects with runes are very rare, and it seems that these objects did have a very special meaning; they are not just objects with some text on them. The idea that these special objects belong to a family archive may be quite right. I am inclined to believe that runes were designed just for that single purpose: to preserve the names of special people; to prevent them from being forgotten. You see the same mechanism in other archaic epigraphy around the Mediterranean; almost all ancient script is especially concerned with the writing of names. It seems to be the most important feature. NIJDAM  The Pforzen belt buckle has a runic inscription that seems to refer to an old heroic tale – what could that be? LOOIJENGA  That inscription reads: Aigil andi Ailrun ltahu gasokun, which can be translated as ‘Aigil and Ailrun fought …’ . The sequence ltahu makes no sense, so you probably have to amend the sequence with an extra graph: possibly an ‘a’ since alliteration would predict that: i.e., Aigil and Ailrun altahu gasokun. ‘altahu’ might be the name of a river, say modern Alzach, and indeed one runologist came up with Illzach (Nedoma 2004). The story has been explained by several runologists, but I think the interpretation the Innsbruck scholar Max Siller gave is most convincing (Siller 2011). Important here is that the name ægili occurs in the runes on the Franks Casket, and that the Old Norse Vǫlundr (Weland) story has been handed down to us with the names of Egill and his wife Ǫlrún. Egill or Aigil was a famous archer and Ǫlrún was one of the swan maidens who came flying to where Egill and his brothers were living. Part of it is a fairy tale. Aigil married one of the swans (Ǫlrún). He is shown defending a stronghold on the Franks Casket. The source of the tale has been found, according to Max Siller; he relates it to a famous classical town in north Italy, Aquileia, and he identifies the archer on the Franks Casket as Egill while the woman who is sitting behind him in a kind of castle with towers would be Ǫlrún. Siller claims that the background is a historical event that eventually turned into a heroic tale. The archer who seems to be defending the castle may be the Alemannic general Agilo, serving in the Roman-Byzantine army during the reign of the Emperor Julianus. The date was March 362, and the story was recorded by the historian Ammianus Marcellinus (Siller 2011). There is no mention of a river, and besides, there is no river known named Alzach or Illzach. But, according to Siller, there is a River Elz which in the Middle Ages was called Elzach. There is even a town called Elzach. The river flows along the Kaiserstuhl in the Black Forest. There was an important Roman castellum here, a capital of the civitas Breisgau, at a junction of several Roman roads (Siller 2018, 148). This would typically be an ideal place for telling stories and get these spread all over Europe. KNOL  I have several points. You say that there are no runic inscriptions from Frisian graves; we have one on a comb from a grave at Hogebeintum. In northern Germany, at Fallward, we have the stool with runic ksamella (scamella, Latin for footstool), also from a burial site. And there is the supposed bracteate from Achlum, sold in 1944: this was claimed to be genuine by Boeles but Morten Axboe has discovered that Boeles had the information from a person called de Jong, who became famous for selling a false Viking hoard to the Fries Museum (poster presentation at the 68. Sachsensymposion, Canterbury, 2018). It seems very likely that the runic bracteate was also false. The Spong Hill alu urn: this is a funerary 393

Z01 Hines Book B.indb 393

07/05/2021 13:31

Frisians of the Early Middle Ages urn, but only in its final use – the pot itself could have been a normal functional one to start with. It’s quite usual to have vessels with the word for ‘beer’ on them. LOOIJENGA  You are right about Hogebeintum – indeed a grave find. But the runic find from Fallward is not Anglo-Frisian I would say, it is too early for that (dated c. AD 425). The Spong Hill urn is found in Norfolk, not Frisia. And whether or not the word alu really only means ‘beer’ is a matter of dispute – there seems to be a connection between the word alu and death; we have several instances of that (Heizmann 2011). KNOL  As for the Betuwe (Bergakker) and Limburg inscriptions, are these in AngloFrisian runes? LOOIJENGA  No, Bergakker is also too early for Anglo-Frisian runes; it is dated to the beginning of the fifth century. This is in itself quite remarkable – to have a runic object of that date in the Rhine-Meuse estuary. The Borgharen buckle with runes is sixth century and belongs to the so-called Continental tradition (or maybe we should say Merovingian). KNOL  So they are very different. LOOIJENGA  The Bergakker find is from an area that was essentially Roman, with an admixture from an indigenous Celtic population, down to the fourth century. After that date the area became populated with all kinds of Germanic folk; it was a transitional period. People speaking different languages would have met and the prevailing Roman culture gave way to other cultures, those of people coming in from several directions. You’ll find people with Celtic, Germanic and Roman backgrounds, and there was some depopulation in certain areas. That means in fact that the language of the Bergakker inscription need not necessarily be Germanic. There is one anomalous rune in the inscription, which occurs four times, and must reflect a vowel, e or u. There is no consensus, though, and therefore no commonly accepted interpretation. The inscription though is clearly a sentence starting with the name of a Germanic man which begins with h, followed by a verb: he gave swords to his comitatus, or something like that. Sentences are rare among runic texts. KNOL  But you could have had people moving in from the north. The surprising thing is that you have a runic inscription there at all, because they are so rare. My final remark is about the ‘secret’ character of the inscriptions. It reminds me of long inscriptions on clocks and bells in church towers: they can be long and meaningful, but no one will ever read them. It’s also the case with elements of the sluices. What is stated on them is ‘I have made this’. Perhaps all that is important is that the words are there. WOOD  It’s the same in churches: there are lots of paintings with inscriptions that are quite impossible to read. LOOIJENGA  Egge (Knol) says that the meaning lies in the fact that there is a text, even if no one can read it. I also think that that is the really significant thing; it is all about the writing itself, the fact that something important is being inscribed to be kept for ever. WOOD  It’s a bit different with a prophet with a scroll. KNOL  Also similar is the immensity of the war memorials at places like Ypres. You cannot read all the names. They just have to be there. 394

Z01 Hines Book B.indb 394

07/05/2021 13:31

Runic Literacy in North-West Europe MAJCHCZAK  We still today put inscriptions on objects for gift-exchange; for instance on the inside of wedding rings. It’s very important those names are there. It’s a good analogy. In that case it is personal. KNOL  You can remove the ring. MAJCHCZAK  But it’s also a personal thing to remove it. VERSLOOT  I am puzzled by the idea that the runes come from the south, inspired by archaic alphabets. LOOIJENGA  At the very least, the inspiration must have come from a literate society . The only possible model for the runic alphabet is one of the archaic Mediterranean alphabets. The puzzling thing is the geographical distance, and the time gap: between the latest archaic Mediterranean inscriptions and the first runes there is a gap of more than a century. Geographically the nearest ancient script, Raetic, is dated between the sixth and the first centuries BC. So it is suggested that the source alphabet might have been the Roman alphabet; if so, you will have no time gap. But the shape of the runes looks very much like the archaic Etruscan and other Italic alphabets, and the practices are also very much the same. I mean they all are epigraphic, funerary, mostly short, mostly names and appellatives, makers’ formulae, alphabets. VERSLOOT  With regard to the shapes of the letters, there is discussion of possible origins in northern Italy. The earliest runic inscriptions are in the North, in Denmark; but perhaps this gap is now being filled up a bit. But these examples are later. There is nothing from the first century in the Rhine area, for instance. There is what appears to be a former Roman officer buried at Fallward, with his military paraphernalia, and an object with a runic inscription. If writing is only symbolic, why would he use runes? Latin would be better for display. I don’t want to claim widespread literacy, but the runic script was around, and used more than just for ‘fun’. LOOIJENGA  One can speculate but I do not have the answer. He may have showed off his Latin knowledge with the name of the object. And added to his Latin knowledge he showed off his knowledge of runes. Perhaps this was not extraordinary, but among the very few objects with Latin and runes this is quite a sensation! VERSLOOT  skamella is a Latin loanword, although it may be incorporated into Germanic with the -a ending. The morphology seems to be Germanic. Latin would be scamulum, neuter singular. LOOIJENGA  This Fallward inscription is really an eye-opener: it shows a mix of Roman and Germanic culture, nicely embedded in the mortal remains of a Roman officer of Germanic descent. He used runes, and he used a Latin word in this one inscription; he was very sophisticated! And the inscription is not random; it fits with a tradition: the name of the object and name of the man himself, or perhaps of his dog. HINES  My first point is about reluctance to believe in a widespread runic tradition. I would be more positive. It is widespread in so far as it covers a very large area of northern Europe. Also it has a very long history, at least a thousand years as a living script which develops. With development we can recognize that despite the difficulties of interpretation of the texts, there is consistency in the letter forms. We can debate the exact circumstances, but we have quite a good sense of what was 395

Z01 Hines Book B.indb 395

07/05/2021 13:31

Frisians of the Early Middle Ages happening overall. There are certain places where we seem to get sudden, quite intense changes, sometimes with strange forms; a particularly good example is the Ash Gilton sword pommel (Hines 2006). Another key point I would like to make, goes back to the Pforzen inscription. The text there is a single line of alliterative verse, or a well-formed sentence with a subject, main verb and possibly a direct object: you can take the text away from the object and it would still make sense, linguistically, in whatever medium you read it. The great majority of the really early inscriptions make very little sense separated from the objects: kabu/‘comb’ is comprehensible, but it belongs to the object. The only properly formed sentence in pre-Old English is the Loveden Hill urn. The distinctive change in England is that a new form of literacy comes in with conversion in the seventh century, so that eighth-century inscriptions are texts that make sense on their own, once the use of literacy has really begun to separate from the material-cultural context (Hines 2019). But in the case of the Frisian inscriptions, I do not see that development taking place in Frisia. If that is valid, somehow the introduction of ‘literacy’ in Frisia hasn’t taken place that early and does take place at a later date. It’s a more positive way of looking at the problem of the Frisian inscriptions. It’s also a good bit more positive than just calling the writing symbolic or decorative. KNOL  A lot of the material which is used for runic inscriptions in Frisia – wood, bone and antler – will not be preserved over a very large area: for instance, the whole Pleistocene area. So I suppose that very many such inscriptions have been lost. The number made must have been very much greater. LOOIJENGA  Yes, it must have been thousands. VERSLOOT  Inscriptions may be difficult to interpret, but there are two or three sentence-like examples from Frisia. Compared with South Germanic and AngloSaxon this is really quite rich, and also gives quite a positive impression if you compare it with all the question marks in Martin Findell’s book on the Continental inscriptions (Findell 2012). It is a problematic old corpus though. NIJDAM  Two observations. One is that the Anglo-Frisian alphabet is not as consistently attested in both England and Frisia as the term suggests. In Frisia only two or three Anglo-Frisian runes occur in inscriptions. Most of the new ‘AngloFrisian’ runes are attested in England. It also seems to me that the Frisian runic corpus has a comparatively large number of strange, non-standard forms. LOOIJENGA  Yes, in a way that is right. You are thinking of Westeremden, by all means an intriguing inscription which seems to contain runes from the younger futhark, mixed with Anglo-Saxon runes. NIJDAM  And Bernsterburen. LOOIJENGA  I don’t think Bernsterburen has anomalous runes. The inscriptions on the staff are very much eroded, and really only the name tuda is clear, and possibly a verb form kius, although the s-rune there might be a short-twig form from the younger futhark. This is puzzling indeed. IJSSENNAGGER-VAN DER PLUIJM  Tuda is a recorded name, and we noted in the discussion of Egge’s paper (this vol., Ch. 2) that there was a known bishop of Lindisfarne of that name. Should we distinguish between the carving of runes, and the inclusion of runes that are part of the design of an object? 396

Z01 Hines Book B.indb 396

07/05/2021 13:31

Runic Literacy in North-West Europe LOOIJENGA  If the main purpose was to write a name, a specific person’s own name, not just anybody’s – and at the same time you knew that nearly no one would be able to see or read the runes, the purpose is not the reading, but the preservation of that name (identity) for eternity. I think that maybe it was not ornamentation that mattered, but the use of a kind of secret, votive, language, with a confined purpose; especially names and appellatives (such as ‘comb’) were important. Think of the fairy tale of Rumpelstiltskin and the fact that nobody should know his name, otherwise he would lose power. The fact that the word ‘rune’ means ‘secret’ in several old Germanic languages, and in the Gothic Bible it was used to translate Greek mysterion, may point to the purpose of writing only for initiates.

References Antonsen, E. H. 1996, ‘Runes and Romans on the Rhine’. In Frisian Runes and Neighbouring Traditions. Proceedings from the First International Symposium on Frisian Runes at the Fries Museum, Leeuwarden, 26–29 January 1994, eds. T. Looijenga, A. Quak, Amsterdammer Beiträge zur älteren Germanistik 45 (Amsterdam–Atlanta), 5–14. Aufderhaar, I. 2017, ‘Between Sievern and Gudendorf. Enclosed sites in the north-western Elbe-Weser triangle and their significance in respect of society, communication and migration during the Roman Iron Age and Migration Period’. In Frisians and their North Sea Neighbours. From the Fifth Century to the Viking Age, eds. J. Hines and N. IJssennagger (Woodbridge), 149–72. Bammesberger, A. 1999, Pforzen und Bergakker. Neue Untersuchungen zu Runeninschriften (Göttingen). Barnes, M. 2012, Runes: A Handbook (Woodbridge). Blackburn, M. 1991, ‘A survey of Anglo-Saxon and Frisian coins with runic inscriptions’. In Old English Runes and their Continental Background, ed. A. Bammesberger (Heidelberg), 137–89. Boeles, P. C. J. A. 1951, Friesland tot de Elfde Eeuw, Zijn Vóór- en Vroege Geschiedenis (’s Gravenhage). Campbell, A. 1983, Old English Grammar (Oxford). Derolez, R. 1998, The Origin of the Runes: an Alternative Approach, Academiae Analecta (Brussel). Düwel, K. 1996, ‘Die Macht der Schrift im östlichen Frankerreich’. In Die Franken, Wegbereiter Europas. Vor 1500 Jahren: König Chlodwig und seine Erben. Katalog zur Ausstellung im Reiss-Museum Mannheim, ed. A. Wieczorek (Mainz), 540–52. — 2018, ‘Kämme als Toten- und Opfergaben’ In Auf Archäologischer Spurensuche, Gedenkschrift für Dr. Wolf-Dieter Tempel, 15.08.1937–07.03.2017, ed. S. Hesse, Archäologische Berichte des Landkreises Rotenburg (Wümme) Band 21 (Oldenburg), 325–46. Elmevik, L., ‘De urnordiska runinskrifternas alu’. In Runor och Namn, Hyllningsskrift till Lena Peterson den 27 januari 1999 (Uppsala), 21–8. Findell, M. 2012, Phonological Evidence from the Continental Germanic Runic Inscriptions, Ergänzungsbände zum Reallexikon der Germanischen Altertumskunde 79 (Berlin). Fischer, S. and J. Soulat 2010, ‘Les inscriptions runique de France: les pommeaux du type Bifrons-Gilton’. In La Gaule mérovingienne, le monde insulaire et l’Europe du Nord (Ve-VIIIe s.): tome XVI des Mémoires publiés par l’Association française d’archéologie mérovingienne, ed. C. Lorren (Condé-sur-Noireau).

397

Z01 Hines Book B.indb 397

07/05/2021 13:31

Frisians of the Early Middle Ages Galestin, M. C. 2007/2008, ‘Frisii and Frisiavones’, Palaeohistoria 49/50, 687–708. — 2009/2010, ‘Tolsum revisited: how the Frisian ox disappeared’, Palaeohistoria 51/52, 9–26. Gerrets, D. A. 2010, Op de grens van land en water. Dynamiek van landschap en samenleving in Frisia gedurende de Romeinse tijd en de volksverhuizingstijd, Groningen Archaeological Studies 13 (Groningen). Heeren, S. 2017, ‘From Germania Inferia to Germania Secunda and beyond. A case study of migration, transformation and decline’. In Social Dynamics in the Northwest Frontiers of the Late Roman Empire, ed. N. Roymans, S. Heeren and W. De Clercq (Amsterdam), 149–78. Heizmann, W. 2011, ‘Die Formelwörter der Goldbrakteaten’. In W. Heizmann and M. Axboe (eds), Die Goldbrakteaten der Völkerwanderungszeit — Auswertung and Neufunde (Berlin), 535–601. Hines, J. 2006, ‘The early runic inscriptions from Kent and the problem of legibility’. In Das fuþark und seine einzelsprachlichen Weiterentwicklungen, eds. A. Bammesberger and G. Waxenberger, Ergänzungsbände zum Reallexikon der Germanischen Altertumskunde 51 (Berlin), 188–208. — 2017, ‘The Anglo-Frisian question’. In Frisians and their North Sea Neighbours. From the Fifth Century to the Viking Age, eds. J. Hines and N. IJssennagger (Woodbridge), 25–42. — 2019, ‘Practical runic literacy in the Late Anglo-Saxon Period: inscriptions on lead sheet’. In Anglo-Saxon Micro-Texts, eds. U. Lenker and L. Kornexl (Berlin), 29–60. Imer, L. M. 2010, ‘Runes and Romans in the North’, Futhark, International Journal of Runic Studies 1, 41–64. Insley, J. 1991, ‘Appendix 2: The personal name *Āniwulf/*Æniwulf’. In Old English Runes and their Continental Background, ed. A. Bammesberger, Anglistische Forschungen Heft 217 (Heidelberg), 172–4. Knol, E. 2009, ‘Anglo-Saxon migration reflected in cemeteries in the northern Netherlands’. In Foreigners in Early Medieval Europe: Thirteen International Studies on Early Medieval Mobility, ed. D. Quast (Mainz), 113–30. Knol, E. and N. IJssennagger 2017, ‘Palaeogeography and people’. In Frisians and their North Sea Neighbours. From the Fifth Century to the Viking Age, eds. J. Hines and N. IJssennagger (Woodbridge), 5–24. Lanting, J. N. and J. van der Plicht 2010, ‘De 14C-chronologie van de Nederlandse pre- en protohistorie VI: Romeinse tijd en Merovingische periode, deel A: historische bronnen en chronologische thema’s’, Palaeohistoria 51/52, 27–168. Lebecq, S. 1990, ‘On the use of the word “Frisian” in the 6th–10th centuries written sources: some interpretations’. In Maritime Celts, Frisians and Saxons, ed. S. McGrail, Council for British Archaeology Research Reports 71 (London), 85–90. Looijenga, T. 1991, ‘Yew wood and runic inscriptions in the Frisian terp-area’. In Old English Runes and their Continental Background, ed. A. Bammesberger, Anglistische Forschungen Heft 217 (Heidelberg), 335–42. — 2003, Texts and Contexts of the Oldest Runic Inscriptions, The Northern World 4 (Leiden–Boston). — 2013, ‘Die goldenen Runensolidi aus Harlingen und Schweindorf ’. In Land der Entdeckungen. Die Archäologie des friesischen Küstenraums/Land van ontdekkingen. De archeologie van het Friese kustgebied, eds. J. F. Kegler, A. Nieuwhof, K. Nowak-Klimscha and H. Reimann (Aurich), 430–1.

398

Z01 Hines Book B.indb 398

07/05/2021 13:31

Runic Literacy in North-West Europe — forthcoming, ‘Frisian runes revisited and an update on the Bergakker runic item’. In Old English Runes: Interdisciplinary Perspectives on Approaches and Methodologies, eds. G. Waxenberger, K. Kazzazi and J. Hines (Berlin). Lüthi, K. 2004, ‘Von Þruþhild und Hariso: Alemannische und ältere skandinavische Runenkultur’. In Alemannien und der Norden. Internationales Symposium vom 18.-20. Oktober 2001 in Zürich, ed. by H.-P. Naumann (Berlin–New York), 318–39. — 2006, ‘South Germanic runic inscriptions as testimonies of early literacy’. In Runes and their Secrets, Studies in Runology, ed. M. Stoklund, M. L. Nielsen, B. Holmberg and G. Fellows-Jensen (Copenhagen), 169–82. Mees, B. 2015, ‘Work songs and whetstones: from Sutton Hoo to Straum’, Scandinavian Studies 87 (4), 514–30. Nedoma, R. 2004, ‘Noch einmal zur Runeninschrift auf der Gürtelschnalle von Pforzen’. In Alemannien und der Norden. Internationales Symposium vom 18. – 20 Oktober 2001 in Zürich, ed. H. P. Naumann, Reallexikon der Germanischen Altertumskunde, Ergänzungsband 43 (Berlin–New York), 340–70; Tafel 3–5. Nicolay, J. A. W. 2014, The Splendour of Power: Early Medieval Kingship and the Use of Gold and Silver in the Southern North Sea Area (5th to 7th Century AD), Groningen Archaeological Studies 28 (Groningen). Nieuwhof, A. 2016, ‘De lege vierde eeuw’. In Van Wierhuizen tot Achlum. Honderd Jaar Archeologisch Onderzoek in Terpen en Wierden, ed. A. Nieuwhof, Jaarverslagen van de Vereniging voor Terpenonderzoek 98 (Groningen), 83–98. Pollington, S. 2016, Runes, Literacy in the Germanic Iron Age (Hockwold cum Wilton). Schrijver, P. 2017, ‘Frisian between the Roman and the early medieval periods: language contact, Celts and Romans’. In Frisians and their North Sea Neighbours. From the Fifth Century to the Viking Age, eds. J. Hines and N. IJssennagger (Woodbridge), 43–52. Seebold, E. 2003, ‘Die Herkunft der Franken, Friesen und Sachsen’. In Essays on the Early Franks, eds. E. Taayke, J. H. Looijenga, O. H. Harsema and H. R. Reinders, Groningen Archaeological Studies 1 (Groningen), 24–34. Siller, M. 2011, ‘Storie del Mediterraneo antico nell’Europa del nord medievale: The Franks Casket (British Museum, ca. 700)’. In Ernst Robert Curtius e l’identità culturale dell’Europa. Atti … 2009, eds. I. Paccagnella and E. Gregori, QCFLPadovano 25 (Padova), 293–300. — 2018, ‘Aigil, Ægili, Egill, Aïlrun, Ǫlrún – des Rätsels Lösung? Ein Vorbericht’. In Synergien – 25 Jahre Germanistik und DAAD an der Philosoph Konstantin-Universität Nitra. Sammelband: Internationale Tagung 2017, eds. D. Krause and O. Wrede (Nitra), 143–50. Soulat, J. 2014, ‘La circulation transmanche des objets de type Anglo-Saxon et Mérovingien (Ve-VIIe siècles)’. In Les Anglais en Normandie, Congrès des Sociétés historiques et archéologiques de Normandie 16 (Caen), 83–95. Steuer, H. 2003, ‘The beginnings of urban economies among the Saxons’. In The Continental Saxons from the Migration Period to the Tenth Century: An Ethnographic Perspective, eds. D. Green and F. Siegmund, Studies in Historical Archaeoethnology (Woodbridge), 159–92. Stoklund, M. 2006, ‘Chronology and typology of the Danish runic inscriptions’. In Runes and their Secrets, Studies in Runology, ed. M. Stoklund, M. L. Nielsen, B. Holmberg and G. Fellows-Jensen (Copenhagen), 355–84. Taayke, E. 2003, ‘Wir nennen sie Franken und sie lebten nördlich des Rheins, 2.–5. Jh.’. In Essays on the Early Franks, eds. E. Taayke, J. H. Looijenga, O. H. Harsema and H. R.

399

Z01 Hines Book B.indb 399

07/05/2021 13:31

Frisians of the Early Middle Ages Reinders, Groningen Archaeological Studies 1 (Groningen), 1–23. Truc, M.-C. and C. Paresys 2010, ‘Des chefs francs à St Dizier? Découverte de trois tombes exceptionelle en 2002’, Societé des Lettres, des Sciences, des Arts, de l’Industrie, et de l’Agriculture de Saint-Dizier, tome IX, 2ème serie, 1–40. Waxenberger, G. 2017, ‘How English is the early Frisian runic corpus? The evidence of sounds and forms’. In Frisians and their North Sea Neighbours. From the Fifth Century to the Viking Age, eds. J. Hines and N. IJssennagger (Woodbridge), 93–124. Williams, G. and D. Hook 2013, ‘Analysis of gold content and its implications for the chronology of the Early Anglo-Saxon coinage’. In Sylloge of Coins of the British Isles 63. British Museum Anglo-Saxon Coins Part I: Early Anglo-Saxon and Continental Coins of the North Sea Area, c. 600–760, ed. A. Gannon (London), 55–70. Wood, I. 1990, ‘The Channel from the 4th to the 7th centuries AD’. In Maritime Celts, Frisians and Saxons, ed. by S. McGrail, Council for British Archaeology Research Reports 71 (London), 93–9.

400

Z01 Hines Book B.indb 400

07/05/2021 13:31

FINAL DISCUSSION VERSLOOT  With reference to the problem of interpreting what can only be scarce evidence, and ideas of the ‘deconstruction’ of historical narratives, the evidence we can put together seems to me largely to be adding details to a traditional scheme that has been around for some time. Has that scheme changed fundamentally, or has it just become more precise? We are still talking about migrations, essentially. I find this quite comforting. Johan (Nicolay) is filling out the concept of migration, but not saying it is not there. IJSSENNAGGER-VAN DER PLUIJM  I agree with that, but there’s still more to it. The focus has changed; the perspective is now more anthropological. Some of the papers do try to go beyond the traditional scope. It is not about trying to throw away an old image, or even just filling it out, but a different perspective that enables us to see new things. Many of the papers were about a wider world than Radbod’s. NIJDAM  I would make two points. It is important not to forget that Boeles’s idea of a fourth-century break, and the proposition that the same Frisians had not always inhabited the region, was revolutionary and unpopular in the early twentieth century. Around that time, the Frisian Movement also made use of Radbod as a figure; they would not want to see him deconstructed as a Frankish duke. Things changed in the post-World War II period. Jos Bazelmans wrote on this in the De Vrije Fries (2002), being surprised that this question of a long continuity of the Frisians back into the Roman Period is no longer perceived as important to Frisian identity. Moreover, new insights have been gained in recent times. Hans and Gilles’s location of Radbod in North-Holland/Texel (de Langen and Mol) poses the question of what his power was over the rest of Frisia. Meanwhile, the reaction to Ian (Wood)’s public lecture does show the continuing desire for information on origins. IJSSENNAGGER-VAN DER PLUIJM  The public may not always be happy with the answer. HINES  We have to respect the desire to ask the question; and responding with respect will mean giving an answer that is as accurate as we can, but not necessarily the answer that is wished for. Correspondingly, there is a desire in some quarters in Wales now to validate Welsh identity precisely by emphasizing that it has gone through all the constructive processes typical of European national identities throughout history rather than somehow being eternal. IJSSENNAGGER-VAN DER PLUIJM  It might be interesting to have a further meeting outside of what is now Friesland. There has been an internal debate in the Netherlands about how we study the northern area, with a Frisian view opposed to a Frankish one; it would be good to combine them. NIJDAM  To explore the forging of a Dutch identity? IJSSENNAGGER-VAN DER PLUIJM  No, not just a Dutch one. 401

Z01 Hines Book B.indb 401

07/05/2021 13:31

Frisians of the Early Middle Ages FLIERMAN  It also strikes me that modern views of Frisians put identity back on the table. Did we work around this rather than really problematizing it? Archaeologists, for instance, use terms of this kind as typological labels but avoid assuming identities; ditto, in the study of languages. To approach the public we need to re-address that question; are we willing to do so? NIJDAM  This surprises me in a way, because I take from the presentations and discussions how layered identity is: for instance, in the case of Johan (Nicolay)’s Domburg brooches. This shows that kin groups and families go up to a middle range of identity and then there is a top-level elite. Our sources show us that identity is multi-layered; we need to tell the public that. FLIERMAN  That is one of the answers; and works in relation to speech-acts and material behaviour. Will it appeal in terms of what people are looking for? NIJDAM  Is there a tension between what we do and what the public wants? I think people want anchors, for instance, in history. FLIERMAN  As experts we have been able to present a series of pictures but I don’t think these amount to a complete definition of what it means to call someone or anything ‘Frisian’. Can we provide that? LOOIJENGA  People ask ‘where do we come from?’, and what interests me is the two-hundred-year period before Radbod, starting from the habitation hiatus. What were these people? How did they stand out from their neighbours, the Saxons and Franks? What distinguished ‘Frisians’ then? I think we need to address this in terms of origins. IJSSENNAGGER-VAN DER PLUIJM  What we’ve learnt from here is how wide in time and space the contexts in which we could look for Frisians are: should we look at that in a different perspective, but also at differences amongst the Frisians? For me the essence of the Frisians is in fact that they are mobile and dynamic, and their identity will be created both by people coming into the group and some going out. NIEUWHOF  I can speak of my own first-hand experience of Frisian identity … reactions to my book on funerary rituals and excarnation (Nieuwhof 2015) saw it as an affront to Frisian identity. You never know what will hurt and what will not hurt; we are not writing for a public that does not like new scientific insights. We must strive to improve. MAJCHCZACK  The situation is indeed very multi-layered and complex; we have a narrative, but it is not a simple one. We have to try to get that story across. Bad media experiences are typical, unfortunately. You cannot help it. IJSSENNAGGER-VAN DER PLUIJM  Do we have an idea as to whether we should always use the term ‘Frisia’, rather than ‘southern North Sea coast’? Vikings have become a ‘Viking phenomenon’. Is there an alternative? HINES  To answer that question but also to respond to the point raised by Tineke (Looijenga) concerning the same issue back in the fifth century, because the points are pertinent to one another. In a rather in-between position chronologically we have Procopius, at the other end of Europe: did he have genuine information about ‘Frisians’ in the north-west, or had he got only fragments of contemporary and historical information that are utterly garbled and allow us to conclude nothing? I now have a much stronger sense from this symposium that the re-adoption of the Frisian identity was political and came out of circumstances in the midseventh century; that process gave a totally new meaning to what the inherited and 402

Z01 Hines Book B.indb 402

07/05/2021 13:31

Final Discussion existing terms meant. Both archaeologically and linguistically, I would continue to emphasize the close parallels either side of the North Sea. Things which are ‘Frankish’ and ‘Anglo-Saxon’ in the sixth century, we can distinguish. VERSLOOT  I have no discomfort with the relevance of the term ‘Frisian’. Linguistically too, Frisian is indeed something that is emergent: it overlaps with the area of Lex Frisionum, and with later areas called ‘Frisia’. I would not say that these people in c. 400 were ‘Frisians’, but to call this complex ‘Frisian’ is not problematic to me. Perhaps it does help enquiring people to put question marks on things. IJSSENNAGGER-VAN DER PLUIJM  I would carefully emphasize that it is not the same as modern ‘Frisian’. HINES  We have a duty and responsibility to facilitate knowledge transfer/impact, and not to restrict reflection to an elite which is privileged to meet here [laughter] and have these discussions; there is a point at which a degree of internal reflection is justified too. I’m not so surprised that we have papers which are going in one direction and discussions in another; the nature of cross-disciplinary discussion is always likely to move away from subtleties in individual areas towards simpler and more old-fashioned issues. I was struck, from the detail in the papers, and thinking again of Giorgio (Ausenda)’s paradigm and about questions that can be asked or which can’t be answered — there’s a point at which we look at this population like a crustacean; we see the shell but not what’s happening underneath it. We haven’t discussed kinship and family, as Giorgio would have insisted on. This is a field where cognition within the period we are looking at would be very different from modern ideas. It comes up in relation to navigation too: externally, we have the automatic stereotype of the Frisians as argonauts of the southern North Sea; but in their culture that is practically invisible. On religion: I emphasize the accessible group aspects; Annet (Nieuwhof) the aspects of individual belief, which I tend to push into a folklore category. There are issues here that are partly source-related, partly matters of methodology, but are also the challenges of cross-disciplinary dialogue. IJSSENNAGGER-VAN DER PLUIJM  There’s a difference of langue and parole here: this relates to Tineke (Looijenga)’s ideas on the runes. LOOIJENGA  In this way an important thing I have learnt or become aware of, is the idea that something has to trigger the mind in relation to the very short inscriptions; one word can trigger a whole story. NIJDAM  Let me raise a few more general points. They go back to Giorgio Ausenda and the anthropological model he used. Ian (Wood) talked about cross-fertilization between history and anthropology. It strikes me that Ausenda’s scheme is very much that of a traditional anthropological monograph. This anthropological schema had become outdated since the beginning of the twentieth century. In 1991 a monograph was published by Donald Brown on human universals. He showed that since the 1930s a new type of anthropology emerged, which was rooted in the nature versus nurture debate. The nurture approach won and then came to dominate anthropology. I think we are on the brink of moving back to giving nature its due place in this, so we can assume stem categories, for instance, what you just said about family and kin. A related point is about the ‘otherness’ of the Frisians. I have missed this so far: we approached this on religion. We have not explicitly stated what we think about how 403

Z01 Hines Book B.indb 403

07/05/2021 13:31

Frisians of the Early Middle Ages the Frisians perceived their ‘otherworld’ permeating their everyday life. We have been discussing things from a modern, non-magical perspective. I would like to reconsider Jan de Vries’s idea that ‘might’ (Proto-Germanic *mahti-) was in some way intangible and associated with magic (de Vries 1956/57); I see this as associated with the concept of honour. WOOD  Of course Ausenda laid down a model, but when he did so initially he had very little knowledge of the Early Middle Ages, and practically every paper written for him rebelled against his template. The model is, however, helpful in providing direction. What I have learnt from anthropology is that you cannot use ‘an’ anthropological model – every case is unique; what you can do is read the work and learn how to question a society as ‘other’. It shocks you into seeing the apparently ‘normal’ as not normal. IJSSENNAGGER-VAN DER PLUIJM  That’s very helpful. The otherness of religion takes us into the mentality question: how do we find a mind-set and see what things mean? HINES  Precious metals and elite art-styles did allow an elite to express itself and to interact with other elites; this can be linked to the law codes, which in a malecentred way define personal status (e.g., wergild) in clear monetary terms. Social recognition of a personal status/value could well involve a material practice of gift-giving which directly reflects that social identity. In terms of the recent tradition of approaches to Germanic religion, which (I suggested) are extremely cautious over moving beyond the immediate context of the sources, perhaps I have not put myself as much out on a limb as I had thought. I am confident from the connections pointed out that this line of argument could be taken a lot further. But will it end up returning to sacral kingship? Will we then just reinvent debates of the 1930s or 1950s? If so, will that mean that those were in fact more justified than has recently been thought? Or should we try to look at this in a different way? FLIERMAN  For me it has been a big temptation always to move towards simplification to try to find common ground: e.g., Widukind – what can we say about him? Interdisciplinarity can therefore be a redefinition of one’s own field – or is that just a historian’s view, as it being almost a betrayal to engage with the questions? HINES  I would personally say ‘no’. I am conscious of terminology which in one field we take for granted but which needs careful explanation in talking outside the field. There are, for instance, technicalities – this is very clear with sound changes, as years of training are needed to understand philological details and facts, and it’s hard to present those to a wider audience. But I don’t see it as compromising the integrity of a particular field, or as simplifying things regrettably. VERSLOOT  I would add that I have a positivistic attitude. I do believe things can be discerned. I would like ten times as much evidence, and more reliably dated sources, of course. Even if much can be debated, you genuinely can reconstruct an enormous amount about a ‘lost’ language. That cannot be done in history or archaeology – you can’t design the missing evidence. I feel therefore that I am on solid ground. I don’t have to force myself in any way. FLIERMAN  Perhaps I should say instead being forced to adapt by taking the evidence in a direction you would not otherwise have taken it in? 404

Z01 Hines Book B.indb 404

07/05/2021 13:31

Final Discussion HINES  Only in a good way, in my view. WOOD  I think historians have been far too certain that we know how to reconstruct. Traditionally most scholars just joined up the dots with direct lines, but now we are isolating the spots of evidence rather more. In a way, it’s a good thing we’re being forced to look at our sources a great deal more carefully. VERSLOOT  There’s something similar in linguistics. What I see is that conclusions drawn from haphazard observations, and the rejection of other evidence as scribal errors, are being re-evaluated: taking corpora as they are in their entirety produces new information. But this is essentially doing the same thing as has always been the case; just doing it properly. NIJDAM  You have already yourself pointed out that the advantage of studying a language is that you can depend on the fact that you are reconstructing systems that are valid. Historians cannot do that: they are not looking for something machine-like and on the whole rarely approach their material from a certain model. This is why I am keen on the concept of human universals because that gives me cogwheels comparable to the building blocks of linguistics.

References Bazelmans, J. 2002, ‘Is Frieslands oudheid tegenwoordig zonder belang?’, De Vrije Fries 82, 302–9. Brown, D. E. 1991, Human Universals (New York). De Vries, J. 1956/57, Altgermanische Religionsgeschichte, 2 vols. (Berlin). Nieuwhof, A. 2015, Eight Human Skulls in a Dung Heap and More: Ritual Practice in the Terp Region of the Northern Netherlands, 600 BC–AD 300, Groningen Archaeological Studies 29 (Groningen).

405

Z01 Hines Book B.indb 405

07/05/2021 13:31

Z01 Hines Book B.indb 406

07/05/2021 13:31

CONTRIBUTORS Dr Robert Flierman is a postdoctoral research fellow and lecturer at Utrecht University. He has published widely on the Continental Saxons and is currently part of the NWO Vici research project ‘Citizenship discourses in the Early Middle Ages, 400–1100’ (2017–2022). Prof. John Hines has taught at Cardiff University since 1983, firstly as lecturer and reader in English and subsequently as professor of Archaeology. He publishes extensively on integrated approaches to cultural evidence and its transmission from late Antiquity to modern times, especially across northern Europe. Dr Nelleke (Laure) IJssennagger-van der Pluijm is Director of the Fryske Akademy (Leeuwarden) since March 2021. Her interdisciplinary research focusses on Early-medieval and Viking Age North Sea and Atlantic coasts, as well as on cultural heritage in (coastal) minority areas in the longue durée. Dr Egge Knol is curator of Archaeology, History and Old Regional Arts & Crafts at the Groninger Museum. As an archaeologist and biologist, he has published widely on the (zoo-)archaeology of the northern Netherlands, especially regarding terps, burial customs and material culture. Prof. G. J. (Gilles) de Langen is provincial archaeologist of Fryslân and researcher in Medieval Archaeology at the Fryske Akademy (Leeuwarden). He is attached to Groningen University as extraordinarius professor for the archaeology of the Frisian coastal area. Dr Tineke Looijenga studied Scandinavian and Old Germanic languages and Archaeology at Groningen University. She wrote her dissertation on the oldest runic inscriptions. Dr Bente Sven Majchczack is an archaeologist affiliated with the University of Kiel. His archaeological and geophysical research on settlement in the North Frisian Wadden Sea area focuses on the first and second millennia AD. Prof. J. A. (Hans) Mol works at the Fryske Akademy (Leeuwarden) as researcher in medieval history. He is also attached to Leiden University as extraordinarius professor for the history of the Frisian lands in the Middle Ages. Dr Johan (Auke Willem) Nicolay is assistant professor at the Groningen Institute of Archaeology, University of Groningen. Central themes in his research are settlement archaeology in the coastal area of the northern Netherlands, the first-millenium history of present-day Noord-Holland, and the ‘reading’ of symbolic messages on early-medieval metalwork. Dr Annet Nieuwhof is an archaeologist specializing in the habitation history of the northern Netherlands coastal area. She is presently working as a research fellow and teaching at the Groningen Institute of Archaeology of the University of Groningen. 407

Z01 Hines Book B.indb 407

07/05/2021 13:31

Contributors Dr Han Nijdam is researcher in Old Frisian and Old Frisian law at the Fryske Akademy (Leeuwarden). His research on Old Frisian law ranges from its Early-medieval beginnings up to the influence of learned law in the fifteenth century, working with classical philological as well as historical anthropological methods. Prof. Arjen Versloot is professor of Germanic Linguistics (German, Scandinavian, Frisian) at the University of Amsterdam. His main field of research is Frisian with special emphasis on variation and change across all periods and dialects. Prof. Ian Wood is professor emeritus of Early Medieval History at the University of Leeds. He is the author of books on the Merovingians, the hagiography of early-medieval mission, the modern historiography of the early Middle Ages, and the transformation of the Roman West.

408

Z01 Hines Book B.indb 408

07/05/2021 13:31

INDEX A, River  91 Aalburg 317 Aalsum  29, 36, 57, figs. 2.10, 3.6 Åparken 179 Acca, Bishop of Hexham  119, 207 Achlum  278, 289–90, 293, 295, 297, 383, 393 Achtkarspelen 91 Additiones Sapientum, see Lex Frisionum Ado (Atto), see Wrssing Adorp 30 adventus Saxonum  226, 340 Æniwulufu, runic coin (Folkestone)  359, 382, 384, 386, fig. 13.10 Æsir, Ēse  295, 297, 316–17, 328 Æthelbehrt, King of Kent  145, 157 Africa (incl. Egypt, Eritrea, the Hadendowa, the Nuer, Sudan)  7–9, 32, 64, 363 Afrikaans, language  195, 348n Agilo 393 akkerterp (‘crop-field terp’)  107, 111; see also terp Alcuin  119, 203, 207, 209, 213–14, 218, 316, 319–20 Aldgisl  2, 79, 83, 94, 116–18, 121–3, 125, 155, 203, 205–6, 208–10, 214–15, 218, 227, 240, 320, 328–9 Alemanni, Alemannia (Lati Ala-)  1, 140, 145, 208, 212–14, 303, 361, 383, 393 Alkersum 176 Alkmaar  110, 317 Almere  93, 100–2 Alpaida 211 Alphen  120, 210, 215 Altfrid  117, 119, 208, 210, 213–15, 229, 231, 316 Amandus 203 amber  95, 177, 181, 184–5, 190 amethyst 292 Ammianus Marcellinus  225, 393 Amrum  3, 171, 173–4, 187, 347 Amsterdam  268, 344n Amtum 35 Angeln  342, 355–7 Angles  227, 299, 350, 357–61, 390 Anglo-Frisian, language (Anglofriesisch) 339–63, 365, Tab. 12.2 Anglo-Frisian, parallels  145, 183 Anglo-Frisian runes  5, figs. 12.3, 12.8 Anglo-Saxon Chronicle  255, 343, 357–8

Anglo-Saxon(s) (angelsaksisch)  1, 4, 9, 19–22, 35, 45, 47, 51, 55–67, 70, 79, 97–8, 116, 119, 141, 155–7, 161–2, 164, 184, 203, 205, 207–10, 213, 218–19, 227–8, 237, 250, 259, 261, 273, 278, 281, 288, 292, 297, 303, 313–14, 317, 319, 324, 327–8, 332, 339–40, 350, 355–7, 359, 363, 366–8, 376, 382, 384, 386, 390, 392, 396, 403, figs. 3.4, 3.6, 3.9, 12.7 animal motifs  276, 278–9, 285, 290, 293, 294–5, 297, 300 (Salin’s) Animal Style I  282, 294, 295 (Salin’s) Animal Style II  290, 295 Anjum 15 Annales Fuldenses, Annals of Fulda,  101, 212 Annales Mettenses Priores  209, 210, 211 Annales Regni Francorum, Frankish Annals  207, 228 Ansbert 207 Ansgar  187, 333 Anso of Lobbes  207, 212–13 antler  5, 14, 17, 19–20, 22, 25, 30, 34, 95, 383, 396 Antwerp  109, 118–19 Aquileia 393 Aquitaine, Aquitania  213, 217 archaeoethnology, ethnoarchaeology  1–2, 6–8, 10, 249, 261, 403–4 Archsum, Alt-Archsum  172, 174–5, 188, fig. 6.1 Ardennes  118, 120, 212, 256 arm-ring 288–9 Arnhem  229, 231 Arnulfing family  205 Arwerd 115n āsega (‘law-speaker’)  153, 159, 260, 317 assembly, see thing attacks, raiding and plundering  46, 101, 102, 121, 123, 137, 164, 165, 209, 214, 216, 225, 226, 297, 234, 297, 322, 362 Attigny 229 Attingahem  94, 121 Audulfus  155, 206, 217 Aurich  19, 152 Ausenda, Giorgio  1, 6–11, 403–4 Austrasia, Austrasians  116, 118, 213, 215, 330–1 Avars  229, 265 Baduhenna 378–9 Baflo 127

409

Z01 Hines Book B.indb 409

07/05/2021 13:31

Index Bainus, Abbot  207 Baiuvarii, Bavaria  8, 208, 227, 302, 303, 313 Baltic peoples  325 Baltic Sea and region  188, 236 ban (Lat. bannus)  141–3 baptism  157, 213, 218, 229, 316, 320, 321, 329 Barward 54 basketry, see craft production Batavians 236 Bayeux 236 beach-market 174 bead(s), bead necklace  19–20, 26, 70, 175, 181, 183, 184, 185, 190, 292 beadmaking  35, 181, 184 Bede  5, 91n, 119, 162, 203, 207–10, 218, 219, 228, 323, 356, 359 Bede’s Death Song 342 Beers 297 Beetgum  16, 56, 64–5, fig. 3.6 Belgium  1, 109, 276, 286, 344, 354, 383 belt and belt-fittings  19, 225, 277, 289, 292, 293 Benserwatt  22, fig. 2.7 Beowulf  147, 155, 206, 275 bereskinzes kempa 35 Berlikum 113 Bernlef  323, 328 Bernuthsfeld 19 Betuwe  232, 392, 394 Biessum  107, fig. 4.11 boat, see ship Boeles, P. C. J. A.  80–1, 383, 393, 401 bog-body/-ies 19 Bolsward  88, 109, 112 Boniface, Saint  5, 88, 94, 121, 203, 205, 208, 213–14, 273, 302, 312–13, 322, 325, 329, 331 Boorne, Battle at the (AD 734)  88, 95, 99, 121, 203, 214 Boorne, River  84, 90–1 Borgsum, Borgsumburg  172, 185–90, 192, figs. 6.1, 6.8 bow, see weaponry bowl(s)  21–2, 32 hanging-bowl (Oegstgeest)  292, 295 Brabant  93, 205 (see also Noord-Brabant, province) bracteate  288–9, 293, 294–5, 297, 300–2, 376, 382, 383, 389, 393 breaking (diphthongization)  342–4, 346, 349, 364, 366, Tabs. 12.2, 12.3 Breisgau 393 Bremen  46, 79, 83, 230–1, 263, 323 Breukelen  94, 121 Britain  2, 4, 61, 64, 83, 162, 225–6, 232, 236, 266, 313–14, 342n, 347, 350–1, 355–7, 359, 362–3, 368, 390 Britsum  264, 383n

Brittany 208 bronze-working, see craft-production brooch(es)  13, 15, 19–20, 22, 30, 32, 34, 35, 56, 59, 68, 70, 115, 183, 185, 189, 225, 266, 273, 274, 276, 277–288, 289–93, 295, 296, 297, 298–9, 300–1, 302, 303, 333, 355–6, 383, 402 Achlum type  278, 289–90, 293, 295, 297 ‘Anglian’ 286–7 Anglo-Saxon 19 annular  20, 55–6, 285 ‘bow-tie type’  286 brooch (Frankish-style)  286, 300, 384 butterfly 277 coin and pseudo-coin  30, 286, 292 copper-alloy  13, 19, 273, 275, 288, 293, 298 crossbow (Armbrustfibel) 277 cruciform  19, 276, 277–81, 285, 186, 277, 278–9, 281, 287, 298–9, 355–6, 361 sub-types 278 disc 30, 286, 288, 296, 300, 384 enamelled  183, 296, 333 disc-on-bow  115, 156, 287, 290, 293, 295, 296–7, 302 Domburg type  20, 68, 183, 189, 283–4, 285, 286, 287, 289, 295, 296, 299, 300, 302, 402 equal-armed (gleichärmige)  56, 64, 183, 185, 189, 277, 286, 287, 288, 292, 293 Borre-style 185 small equal-armed  286, 287, 288, 296, 297 figural 285 Frankish 289 ‘Frisian’ 284–5 gold  15, 187, 284 Hallum 296 moon-shaped 286 Nydam 278–9 penannular 64 rectangular (plate)  183, 286 ring 187 saucer 288 ‘Saxon’  286–7, 288, 293, 299, 356 Scandinavian  185, 187, 295 silver  15, 187, 275, 277, 284, 288, 296 small-long  19, 68, 281–2, 285, 286, 298 square-headed (Anglo-Saxon ‘great’ and Contintental ‘Jutlandic’ types)  289–290, 293, 294, 295, 296, 299 supporting-arm (Stützarmfibeln)  276, 277, 287, 288 tortoise 189 Bruges  79, 109 Bruk  182–4, fig. 6.7 Bruno 228 Bubo (Poppo), dux  121, 155, 203, 214 buckle  301, 383–4, 388, 389, 392, 393, 394 shield-on-tongue (Rijnsburg)  292, 295

410

Z01 Hines Book B.indb 410

07/05/2021 13:31

Index Burgundians, Burgundy  164, 217, 376, 385 Butjadingen 350 Byzantine Empire, Byzantium  141, 145, 163, 292–3, 393 Cædmon (Cædmon’s Hymn)  323, 328, 342 Calendar of Willibrord  209–10 Cammingehunderi  151 Campus Madius (Marchfield)  212, 217 capitaneus (OFris. havedling, ‘chiefain’)  148 Carolingian Empire/Period, Carolingians  2, 4–5, 13, 24, 29–30, 57, 80, 82, 88–9, 93, 95–6, 99, 101, 112, 115, 118, 124–5, 127–8, 137–43, 152, 157, 163, 165–6, 197, 207–12, 216–17, 227–33, 235–6, 238–9, 240–1, 255, 257, 260–1, 265–6, 299–300, 313, 319–20, 321, 323, 331 Cassander, George  139 Cassius Dio  378 castellum/-a  49, 117, 119, 120, 203, 209, 393 Castricum-Oosterbuurt 54 castrum  99, 203 Caturix 379 Celtic, languages  218, 224, 313, 357, 364, 379, 394 Lepontic epigraphy  379 Central Frisia  5, 81, 83, 86, 88, 90, 93, 95, 98–100, 102, 110, 112, 115, 124, 126–7, 138, 141, 145, 148, 151–2, 253–4, 256 central place(s)  47, 121, 162, 164, 227, 322, 386; see also production sites; trading sites central river area, the Netherlands  92–5, 98, 116–17, 119, 122–3, 284, 304, 365 Çernjachov Culture  390 Channel, English  99, 225–6, 234, 390, 392 Charlemagne  13, 88 152, 155, 209, 217, 227–30, 239–40, 273, 292–3, 322–3, 333, 361–2 Charles Martel  88, 116–17, 120–1, 128, 155, 205, 208, 211–15, 215–16, 227, 273, 293, 316, 321, 329, 331, 361–2 Charms (Old English verse)  328 chastity 149–50 Chauci  46–51, 68, 223, 225–6, 232, 234, fig. 3.2 Chelles 209 Chilperic, King  206, 212–13, 215 Chlochilaichus (Hygelac)  206 Chlothar I, King  206 Christian iconography  20, 273, 276, 286, 293, 296–7, 299–302 Christianity, Christianization  2, 4, 18, 20, 32, 88, 121, 124, 125–6, 155, 187, 192, 208–10, 214, 218–19, 227, 229, 255, 273, 296, 299, 302, 312–6, 320–6, 328–30, 332–3; see also Church ‘second Christianization’  324 Chuningus 208 Church, the  5, 20, 29, 32, 88–91, 95, 111, 124, 126, 127, 209–10, 213, 219, 230, 233, 235, 276, 299,

312–4, 319, 320, 322, 324, 325, 327, 328, 329, 331, 333 civitas 152 clay district (Friesland)  79, 86, 91, 103, 124, 274; see also terp region clog, see footwear cloth pin  297 clothing (cap, cape, cloak, tunic)  19 Codex Unia  361 coffin(s)  26, 319 coin hoards  182–3, 188, 195 coin(s), coinage  5–6, 15–16, 29–30, 36–7, 81, 87, 94, 95, 97–9, 101, 103, 109, 110, 115, 125, 140–2, 146–7, 150, 155, 175–6, 179, 183, 187–8, 190, 192, 193, 195, 205–6, 217, 227, 254, 256, 258, 286, 288, 290, 292, 293, 297, 329, 376, 383–8; see also denier; dirham; Dronrijp; penny; runic coins; sceatt; solidus; tremissis Audulfus  155, 206, 217 Cologne (Köln)  35, 36, 118–19, 120, 203, 212–13, 215, 223, 323, 331 comb(s)  5, 14, 15, 16, 19, 20, 33, 34, 59, 181, 189, 365–6, 376, 380–2, 383, 385–6, 389, 392, 393 connectivity, maritime connectivity  5, 82n, 249, 250, 253–5, 260–1, 392 Continent [of Europe], the  10, 20, 89, 92, 155, 196, 207, 213, 225, 257, 260, 279, 292, 295, 313, 320, 328, 340–8, 352–5, 359, 361, 362–3, 366, 378, 384–5, 390, 394 continuum, cultural  299 continuum, linguistic  233, 340 conversion, see Christianization coopering, see craft production Cornjum  32, 280 Corvey 230 cowrie shell  30, 34 craft production  20, 22, 24, 25, 35, 47–8, 62–3, 66, 69, 96, 103, 177–8, 179, 184, 185, 188, 190, 192, 197, 253, 288, 296; see also production sites; workshops Cruptorix 378 cult  30, 273, 301, 316, 321 cult site, sacred site  316–8, 320, 329, 332, 379 cultural groups, ethno-cultural groups  1, 275, 318 curtes, see manors and manorial sites Cuxhaven(-Altenwalde, -Sahlenburg)  188 Dagobert I or II, King  203, 205, 206, 208, 212, 328 Dagobert III, King  210 Daniel, Bishop  325, 329, 332 Dannevirke  178, 186–7, 193, 328 Danube, River  229, 264–5, 361 Deae Matres  219

411

Z01 Hines Book B.indb 411

07/05/2021 13:31

Index Delft  110, 239 denier(s) (denarius/-i)  97, 179, 183, 188, 292, 297 Denmark, Danes, Danish  6, 36, 46, 89, 123, 128, 156, 188–90, 191–3, 197, 227, 229, 254, 255, 264, 300, 316, 319–20, 322, 325, 347, 350, 357, 361–2, 376, 378, 383, 384, 385–6, 389, 395 Deventer  102, 194, 231, 235, 254–5, 322 dialect(s), dialectal  193–4, 195–6, 197, 340, 343, 344, 346–9, 363, 364–5 Diesen 120 diet 185 dike(s)  16, 87, 107, 124, 173–4 Dingen 54 dirham 30 distribution patterns and mechanisms  6, 9, 24, 30, 35, 36, 37, 48, 57, 97, 98, 103, 175n, 188, 190, 267, 275–6, 277, 278, 284, 286, 287, 288, 289, 293, 296, 298, 302, 303, 340, 351, 354, 356, 35, 358, 367, 389–90 district(s)  3, 82–3, 84, 86, 88, 89, 91, 92, 93, 94, 96, 97, 98, 100, 102, 103, 109, 110, 112, 115, 122, 123, 124, 125, 141, 151, 152, 158, 159, 164, 231–2, Ditmarsch 347–8 DNA  350, 367 Dodo 211 dog(s)  29, 30, 147; 319 Dokkum  88, 109, 112, 152, 203, 205, 214, 322–3, 325, 355n Domburg  80, 85, 98, 101–2, 118–19, 126, 205, 286 domesticus 211 Donar  297 (see also Dunreslo) Dongjum  113, fig. 4.18 Dordrecht 109 Dorestad (Wijk bij Duurstede)  4, 5, 25–6, 33–4, 36, 80–2, 93–102, 112, 116, 118–20, 122–3, 126–7, 188–9, 192–3, 203, 205–6, 208–10, 253–6, 320 Drenthe, province  22, 34, 45, 48, 50–5, 58–9, 63, 127, 194, 224, 226, 231, 234, 323 drinking horn, terminal mount  292, 295 Dronrijp  113; see also tremisses Duisburg 179 dune(s) and sand ridges (West and North Frisia)  13, 16, 84, 85–6, 87, 174, 183, 252–3, 274 Dunreslo  317 (see also Donar) Dunum  16, 26, 256, 262, 330 Dutch, language  112, 151, 195–6, 317, 339, 343–8, 357n, 366, 368, 382 dux, duces  (116–7, 121, 142, 155–6, 161, 163, 208–9, 211, 218–9, 230, 232 *druhtin (Gmc)  155, 161 dux Francorum 209 dwelling mound(s), see terp

East Anglia  287, 292, 357, 364, 382, 390 East Frisia, see Ostfriesland  16, 24–6, 30, 56, 81, 83, 86–103, 110–11, 115–16, 125, 126–7, 141, 151–2, 178, 188, 197, 229, 256, 275, 292, 330, 368, 383–5 Eastphalia, Eastphalians  228, 345, 362 Ebroin  155, 205–6, 208, 214–15 Ecgbert, King  209 Echternach  120, 123, 157, 205, 230, 322–3, 331 Eelde-Grote Veen  58, fig. 3.6 Eersel  210, 215 Egill Skalla-Grímsson  257 Egmond 100 Ehe 91 Eider, River and estuary  51, 172, 174, 177–8, 186, 188–9, 195 Eiderstedt  171–4, 177, 187–90, 195–6 Eifel  25, 34, 120, 266 Eilrat 229 Eindhoven, province  317 Einstrassensiedlung 81 Elbe, River  55, 223, 225, 227–30, 235, 264, 281, 318, 344, 346, 359, 361, 390 Elbe-Germanic 361 Elbe-Weser, region/Triangle  22, 56, 58–9, 64, 66–7, 174, 178, 188, 197, 225–6, 229, 231, 253, 288–9, 293, 299, 342, 359, 363, 376 Elisenhof  172, 174, 177–8, 187–90, 385, fig. 6.1 elite networking  58, 112–3, 116, 118, 120, 124, 125, 211–15, 225, 229, 231, 233, 273, 275–6, 288, 292–3, 296, 298, 299, 302–3, 304, 349, 403 Elz, River  393 Elzach 393 Emden  81–2, 91, 110, 112, 126, 178, 192 emporium/-a  79–80, 82, 85, 94, 98, 99, 101–2, 118, 127, 184, 187, 188, 189, 236, 386 Ems, River and estuary  13, 26, 30, 46, 50, 56n, 63, 91, 93, 107, 109, 126, 194, 196, 197, 223, 228, 230–1, 233, 378 Emsgo, Emsgouw  88, 91 fig. 4.4 Ems-Weser region  230 Engelbert  210, 215 England  6, 20, 22, 29, 36–7, 60–1, 65, 68, 79, 93, 97–8, 101, 118–19, 155–6, 189, 206, 209, 213, 218–19, 236, 255–6, 262, 267, 277, 279, 281, 285, 288–9, 292, 299, 300, 302–3, 313–14, 316–19, 320, 322, 328, 340, 346–7, 350, 355, 357–60, 363, 364, 367, 376, 378, 383–5, 390, 392, 396 English, language  5, 151, 339–49, 357n, 366 American English  342n Old English  147, 155, 162, 233, 241, 249, 267, 317, 322, 323, 328, 341–9, 364–7, 396 Old English dialects Anglian  343, 345, 359–61, 364 Kentish  341, 345 Mercian  341, 343

412

Z01 Hines Book B.indb 412

07/05/2021 13:31

Index Northumbrian  341, 343, 364 Saxon  345–6, 368–9 West Saxon  241, 341, 343, 364 Englum  25, 50–1, 57, fig. 3.6 Épinal Glossary  342 Epistolae Austrasicae 206 Erfurt  376n, 385 Erfurt Glossary  342 Eriswell 319 Ermin (Ermino), Abbot  207, 213 Essex  319, 320 ethnogenesis 237 ethnos, ethnonym  224, 225, 232, 237 Etruscan, epigraphy  375n, 379, 395 Eudo of Aquitaine  213 Eurinus 208 Ewa Frisonum  240 excarnation  52, 70, 402 exchange, see trade exile  215, 231, 321, Ezinge  15–16, 25, 33, 46–57, 61–2, 64–5, 68, 104, 295, fig. 3.3 Fallward  22, 363, 389–90, 393–5 family archives  49, 54, 59, 69–70, 276–7, 303–4, 392–3 family relations, see kinship; social and societal structure(s) fanum/-a (shrine)  316–17, 326, 327, 333; see also shrine farming, arable  3, 18, 25, 33, 79, 81, 85, 87, 96n, 102, 104–8, 115, 172, 177, 185, 191, 197 farming, livestock  3, 18, 33, 79, 80, 81, 82n, 87, 88, 96, 96n, 102, 104–8, 115, 122, 172, 177, 185, 197 Feddersen Wierde  55, 58, 65, 104 fehting (artificial pond)  18 Femburg 231 fenland(s)  173, 188; see also marsh; mire; peatbog; wetland Ferwerd Burmania  30, 35, 59, 64, 88, 382, 285, 392 filigree  290, 292, 295 finger-ring, ring  15, 30, 292, 297, 299–300, 395 Finn Folcwalding  154–5 Finnsburgh fragment  241, 323 Firdgum 106 Fivelingo  109, 113, 115n Flachsiedlung(en) (level-ground settlements)  174, 177, 189 Flanders (Flemish area)  45, 65, 68, 93, 126, 194, 252, 263, 328, 344–6, 355–6, 365 Flögeln(-Eekhöltjen)  58, 174 Föhr  70, 171–84, 185–7, 189, 190, 191, 194–5, 333, 347–9 Folcer 232

Folkestone, see Æniwulufu Fontenelle 156 footwear  18–9, 24, 35, 36 Fosite  156, 203, 219, 300, 316 Fositeland, see Helgoland Fossa Carolina  265 Foudgum  107–8, fig. 4.12 France  236, 276, 286, 296, 344, 350, 368, 376, 384–5, 386 Franconian, language  362, 366, 369 Franeker  88, 124, 152 Franks Casket  342, 393 Franks, Frankish, Francia  1, 4–5, 29–30, 32, 45, 47, 62, 80, 82, 84, 86, 87–8, 91, 93, 94–102, 116–25, 127–8, 137, 139, 141–3, 150, 155, 157–9, 161, 163, 165–6, 174, 176, 178, 183, 188, 192, 194, 195, 203–19, 225–41, 250, 254–5, 257–60, 265, 273, 276, 281, 285–8, 289–90, 292–3, 295, 296, 298, 300, 303, 313, 319–21, 324, 346, 361, 362, 366, 369, 376–7, 383, 386, 388, 390, 392, 401, 402–3 Fredegar  207, 217–18 Fredegar continuator  203, 207, 209, 210–14 fredus (peace)  142 Fresio 208 fresum (Frisian cloth)  96n, 258; see also pallia fresonica; textile Freyr 155 Friesland (Fryslân), province  1, 5, 13, 15–16, 18, 20, 22, 29, 36, 45–6, 48, 50, 51, 54–6, 62–3, 64, 68, 81, 84, 86, 91, 98, 115, 121, 127, 139, 140–1, 197, 224, 226, 268, 274–8, 281–90, 294–7, 300, 339, 347, 355, 366, 368, 401 frisa kiltar (Frisian guilds)  192 Frisia Citerior  5, 82, 91, 93–4, 116–23, 157, 203, 208–10, 214–15, 321, 330 Frisia Ulterior  5, 91, 93, 203, 208, 214 Frisian Freedom  10, 138, 140, 148, 150, 152, 361 Frisian, language  195–6, 347, 368, 383 Old Frisian  137–54, 233, 241, 250–1, 257, 317n, 321, 339–69, 379, 382 dialects Insular North Frisian  345, 347–9, 366 North Frisian  196, 347–9, 364 South Frisian  345 West Frisian  345, 365 Frisii, Frisiavones  2, 223, 224, 378 fronting  342–4, 364n Fryslân, see Friesland Fulda  18, 228, 231 furniture (bench, chair)  22 Fallward footstool  21–2, 389, 390, 393 futhark  375–6, 382, 388, 389, 396; see also runes; runic inscriptions Gaesmere 325

413

Z01 Hines Book B.indb 413

07/05/2021 13:31

Index Gardulf 123 garnet  32, 34, 266, 286, 290, 301 Garnwerd 14 Garrelsweer  109, 113 Gaul  223, 225–6, 232, 236, 277, 296, 302, 368, 379n, 388 geest  171, 172, 174, 176, 177, 178, 182, 183, 188, 194–5 Geest-settlements  56, 176, 178–185 Gefolgschaft (retinue)  59, 158, 219, 288, 298, 392 Genesis (Old Saxon poem)  241 gens Saxonum et Fresonum commixta  230–1 gentes, gens  217, 225, 227, 233, 235–6, 319n3 German, language  112, 147, 151, 217, 339, 343n, 350, 357n, 369, 378n Germania, Germani  223, 225, 234, 375–6, 379n Germanic religion, see pre-Christian religion Germanic, language North Germanic  340, 346, 362 North Sea Germanic  5, 195, 339–69 North-West Germanic  340, 342 Proto-Germanic  342, 404 West Germanic  194, 195, 196, 340–9, 351, 354, 361, 366, 386 Germany  1–3, 13, 16, 18, 22–6, 45–8, 50, 52, 55, 56, 59n, 62–3, 68, 74, 112, 141, 224, 229, 251, 252, 264, 277, 279, 281, 284–5, 301, 313, 328, 346, 347n, 350–1, 354–5, 359, 363, 366, 376–8, 383–4, 389–90, 393 Gerulf, Gerulfingen, family  86, 95, 123, 128 Gervold, Abbot  208 Gesta Abbatum Fontanellensium  207 Ghent 109 Gildas  225, 350 Gisela 209 glass vessel(s)  26, 29, 35, 174, 181, 184, 188, 190 glass  26, 35, 95, 181, 184, 190 glass-working, see beads/beadmaking, craftproduction, production sites Glockentummler (bell-shaped tumbler)  35 Godel, River  178–9, 182 Godfrey, King of the Danes  128 Gokstad, ship burial  255, 266 gold  15, 29–30, 32, 80, 87, 95, 97, 98, 99, 115, 141, 156, 176, 187, 225, 227, 275, 276, 288, 289, 290, 292, 293, 297, 298, 299–300, 304, 383–4, 386, 389 foil 301 wire  181, 292 goldsmithing  115, 288 Gothic, language  397 Goths  363, 376, 390 Goting  70, 173, 175, 177, 182–5, 187–8, 189, 190–1, 194–5, 197, 333, figs. 6.1, 6.7 Great Ryburgh  319 Greek, epigraphy  379

language 397 Gregory of Tours  206, 217, 236 Gregory of Utrecht  312, 322 Grimersum 178 Grimoald  119–20, 210–11, 214–15, 321 Groningen, province  6, 13–16, 17–18, 20, 22, 24, 25, 28, 29, 31–2, 33–5, 46–51, 55, 61, 63, 68, 81, 98, 103–4, 110–12, 126–8, 141, 224, 226, 234, 240, 274, 275, 292, 295, 323, 330, 349, 368 town  109, 194, 317 Groothusen 178 Groß Strömkendorf (Reric) 236 Grubenhäuser, see sunken hut(s) Gudme 316 Gwynedd 266 Haarlem 110 habitation hiatus and depopulation  16, 45–6, 47, 51, 58, 63, 67, 174, 175, 177, 190, 197, 205, 224, 253, 273, 277, 286, 356–7, 364, 366, 378, 382, 392, 402; see also migration hacksilver 187 Hada, runic coin  382, 384–6 Hadrian’s Wall  151, 378–9 hagiography, hagiographical  5, 119, 161, 205–7, 230, 234–5, 264, 302, 312, 327, 328, 332–3 Haithabu, see Hedeby Hallig Hooge  173 Halligen  19, 26, 36, 195, 347 Hallum(-Jousemaburen)  56n, 61, 64, 295–6, figs. 2.3, 3.7, 10.9 Hamburg 263 Hameland 232 hand pin  60–1 harbour  174, 176, 179, 182, 183, 188 harg (n. OFr.; ON hǫrgr; OE hearg; hearh), see shrine Hargen  317, fig. 11.1 Harich  317, fig. 11.1 Harlingen, runic coin, see Hada harrow 24–5 Hattersum  25, fig. 2.8 Hedeby (Haithabu)  35, 178, 185, 188, 193, 254, 267 Hedehusum 187 Heiloo(-Zuiderloo)  30n, 91, 318 Heinrich II, Emperor  179 heirloom(s) and inheritance  19, 49, 69–71, 123, 128, 144, 276–7, 303–4 Helgoland (Heligoland), also Fositeland  157, 163, 185, 191–2, 193–4, 196, 203, 205, 266, 316, 318, 320–1, 323, 326, 331, 347 Heliand (Old Saxon poem)  233 Herbrum 30 Hermegisl, King  206

414

Z01 Hines Book B.indb 414

07/05/2021 13:31

Index Herold, Johannes  139 Hesse  313, 322, 325 Hessens  18, 263 Hessi  228, 239 Hindelopen 267–8 Historia Ecclesiastica gentis Anglorum, Bede  162, 203, 207 historical archaeoethnology, see archaeoethnology Hitsum (Hitzum), bracteates  26, 383, fig. 10.7 hoard(s)  15, 30, 36, 70, 95, 99, 187, 188, 190, 195, 225, 292, 333, 384, 393; see also coin hoards, treasure Goting 188 Marum 30 Marsum 292 Morsum 187 Hofstaðir  316, 329 Hogebeintum  16, 19, 22, 26, 29, 56, 64–6, 104, 107, 115, 262, 277, 280, 290, 295, 297, 303, 330, 384–5, 393–4, figs. 2.6, 3.6, 4.9 Hoksent 120 Holland  68, 79, 85–6, 93, 110, 126–8, 224, 330, 333, 345, 349, 355n, 365, 367 Holland, Count/county of  123, 141, 152 Holstein 189 Holwerd  30, 88, 294, fig. 10.9 honour  144–6, 150, 154, 159, 163, 259, 326, 404 Hoogterp (Heechterp)  28–9, fig. 2.10 Hopperus, Johannes  139 horse(s)  29, 81, 150, 177, 238, 258, 294, 295, 319 hostage(s)  142n, 150, 229 Humsterland  323, 326, fig. 11.1 Hunsinger Recht  164 Hunsingo 109 Huntebrück 46 Hunze, River  55, 57 Hygelac, see Chlochilaichus Iceland, Icelandic  9, 69, 147, 152, 154, 157–8, 164, 237, 241, 257, 316, 328–9, 332, 350, 357n identity  1, 5, 6, 10, 19, 46, 47, 49, 52, 54, 62, 69, 83, 158, 171, 188–9, 190, 224, 225, 236, 237, 238, 239, 249, 252, 260, 266, 267, 273–6, 287–8, 293, 294, 297, 298–9, 293, 294, 298, 299, 301–4, 397, 401–4 idiom  341, 344, 362 idol(s), idolatry  30, 229, 316, 318, 319 327, 332 IJ, River  93, 100, 365 IJssel, River  102, 230, 233, 322 i-mutation  317n, 341–5, 364 Ingomar 208 Ingvaeonic, language  386 interpretatio Germanica  151 Ireland  2, 209, 313

Irmina van Oeren  120 Irminsul 333 iron (bog iron, iron ore)  26, 34, 95 ironworking, smithing  181, 184, 185 Islam, Islamic  30, 184, 216, 312, 325, 332 Isle of Man  266 Isle of Wight  299, 364 isogloss(es) 233 Issendorf 58 Italic, epigraphy  379, 395 Italy  7, 32, 164, 236, 323, 363, 376, 393, 395 Jarrow 264 Jemgum, Jemgumkloster  26, 46, 262, 264, 266 fig. 2.9 Jever  110, 194 jewellery and dress-accessories  15, 19, 30, 103, 187, 227, 253, 290, 292, 295, 297–300, 356, 376, 379; see also arm-ring; bead necklace; brooch; buckle; belt-fitting; finger-ring; hand pin; neck-collar, neck-ring, necklace; ornament; pendant, wrist-clasp Jorwerd 113 Judaism  312, 325, 332 Junius, Franciscus  339 Jutland, Jutes  58–9, 63, 65, 66, 174, 179, 189, 190, 206, 223, 226–7, 229–30, 264, 300, 348, 350, 357–61, 362, 376, 385, 390 Kaiserstuhl 393 Katwijk  16, 262, 292, 295, fig. 10.9 Valkenburg  49, 68, 100, 102 Kaupang  192, 201, 257 Kennemerland  84, 88, 91, 93, 100, 121, 123, 156, 232, fig. 4.5 Kent, Kentish  139, 145, 157, 264, 281, 283, 289, 292, 299–300, 320, 328, 357, 364, 386, 390 key  32, 189, 321 King’s Lynn  263 kingship  71, 116, 122, 123, 125, 137, 151, 154–6, 157, 159, 203–6, 214, 219, 227–8, 232, 241, 273, 324, 359 kinship, kin groups  8–9, 144, 148, 162–3, 317, 321, 329, 402, 403 Klappholtal 193 knife, knives and sheaths  19, 25, 181, 189, 192 Kongeå, river  189 Kromme Rijn, River  99, 205 kwelder, see salt marsh Lambert 211 Land Wursten  104, 350 landing-sites, landing places, mooring places  115, 253, 262–3 landnam  340–1, 342, 350, 363 land-ownership, see landscape organization

415

Z01 Hines Book B.indb 415

07/05/2021 13:31

Index landscape organization and land-ownership  3, 79, 83, 86–7, 89–91, 102–8, 111, 115, 116, 117, 118, 120, 122–3, 124, 127, 128, 158, 176–7, 185–7, 189, 190, 208, 215, 219, 227–8, 229, 230, 231–2, 233, 235, 240, 242, 292, 322, 324, 330–1, 350 Langobards  (Lombard) 8, 164, 212, 216, 227 Late Antique Little Ice Age (LALIA)  355 Latin, language and sources  85, 163, 217, 218, 230, 251, 257, 349, 355, 368, 375, 378, 379, 393, 395 Lauwers, River  50, 86, 88, 91, 93, 98, 121, 140–1, 145, 148, 152, 166, 230, 231, 293, 323, 326 Lauwerszee 57 Law Kentish 328 Old Frisian (High Med. OFris)  4, 35, 36, 137–143, 144–7, 150, 151, 153–4, 158–9, 162, 165, 237, 240, 242, 250, 251, 257, 268, 317, 361, 382–3, 404; see also Lex Frisionum; Overkeuren; Hunsinger Recht; Riustring Law; The Seventeen Statutes; The Synodical Law of Central Frisia; Ewa Frisonum leather  14, 19, 95, 274 Lebuin  162, 235, 237, 312–13, 322 Leer  91, 110 Leeuwarden  1, 10–11, 29, 68, 88, 109, 112–13, 264 Oldehove 15 Leges Alamannorum  140, 145 Leges Barbarorum (barbarian law codes)  3, 145, 149, 150, 163, 164, 238, 240, 258, 266 Lex Frisionum  2–5, 13, 83, 86, 93, 124, 137, 138, 139–43, 144–7, 148–50, 151–4, 156, 158–60, 163, 164, 165, 166, 195, 196–7, 208, 228, 229–30, 251, 252, 257–9, 264–5, 273, 275, 298, 313, 326–7, 403 Lex Ribvaria/Lex Ripuaria  206, 228, 240 Lex Salica  164, 166, 240, 327, 328, 366 Lex Saxonum  5, 139, 228 Lex Thuringorum  96n, 139, 161 Leiden  110, 349, 365 Leiderdorp  91, 102, 117 Lek, River  93–4, 99, 101 Lekkum 30 Liber Aureus  205 Liber Historiae Francorum  207, 209, 210, 211, 213 liber/-i (freemen)  148–9, 150, 162 Liège  118, 205, 211356, 384n Limburg, province  317, 392, 394 Limes  117, 120, 224, 351, 354–5, 376, 392 Limons 296 Lincolnshire  364, 389, Lindisfarne  33, 396 Lindsey 320 linkage model 346

List  176, 187 literacy  5, 375–83, 389–90, 396 litus/-i (half-free men/serfs)  81, 127, 148, 162 Liudger  5, 91, 117, 119, 121, 125, 205, 208, 210, 213–15, 229–31, 235, 312, 316, 321, 323, 328, 330, 331 Liudhard, Bishop  386 Lobbes  207, 212–13 Loenen 231 Loire, River  226 Lombard, see Langobards London  320, 376 longhouse(s)  174, 177, 179, 183, 184, 185, 190, 318, 329 loom, loomweights  18, 179–81, 184 Lothar I, Emperor  230–2, 235 Louis the German, King  232 Low German, language  166, 347n Lower Saxony (Niedersachsen)  48, 50–1, 54, 56, 58, 63, 65, 68, 171, 189, 194, 263, 347n, 389, 390 Luisel 120 Maas, see Meuse Maasland  84, 93, 101, 123–4 Maastricht  118, 383, 388 Madelinus, moneyer  99, 176 Magna Frisia  83, 116 majores palatii; majordomo, maior domus (mayors of the palace)  116, 119, 120, 205, 206, 208, 210, 211, 214, 215, 227, 330 Malorix  71, 224, 378 manors and manorial sites  85–6, 88, 124, 127, 147 Mare Frisicum  81n, 249 Mariengaarde 263 Marklô  152, 162 Marne, River  57n marriage, marriage exchange  8, 47, 49, 55, 57, 63, 68–9, 119–20, 206, 210–1, 214, 230, 233, 275–6, 287, 299, 302, 330, 349 Mars Thingsus  151 marsh  171, 173, 174, 176–8, 185, 185, 189, 194–5, 197, 236; see also salt marsh; fenland(s); mire; peat-bog; wetland Marum 30 Maxims (Old English poem)  241, 249 Medemblik  100–1, 109, 121, 123, 126 Melenknob 174 Mercia  218, 320 Merovingians  2, 34, 47, 57, 82, 86, 88, 95–6, 113, 155, 176, 203, 205–8, 212–14, 217, 237, 253, 261, 264, 298, 313, 320, 327, 331, 376, 383–5, 386, 388, 392, 394, metalworking, see craft-production Metz 205

416

Z01 Hines Book B.indb 416

07/05/2021 13:31

Index Meuse (Maas), River  13, 16, 83, 84, 93, 96n, 101, 118, 120, 122, 123, 152, 208, 228, 365, 378, 383, 384n, 388, 389, 392, 394 Middelburg  102, 110 Middelstum  113–14, fig. 4.17 Middelzee  46, 56–7, 113 Midlaren-De Bloemert  54–5, 65, fig. 3.6 Midlum (Föhr, Nordfriesland)  177–8, fig. 6.1 Midlum (Harlingen, Friesland)  15–16, 113 Midlum-Northum (Landkr. Cuxhaven, Lower Saxony) 55–6 migration(s), immigration and emigration  4, 22, 45, 51, 56–8, 66–7, 103, 137, 171, 174, 175, 190, 193, 195, 196, 197, 224, 232, 234, 236, 273, 277, 278, 288, 297, 300, 340, 347, 348, 350, 355–9, 363, 366–7, 369, 382, 392, 401–2; see also habitation hiatus millstone(s) (incl. Eifel; basaltic lava)  25, 34, 174, 178, 179, 183, 188, 190, 266 Minor Frankish Annals  212, 214 minting and mint(s), see coins, coinage miracle  213, 316, 323 mire(s)  4, 253; see also fenland; marsh; peatbog; wetland missionary activity  88, 91, 94, 95, 111, 112, 118, 120, 121, 124, 125, 159, 187, 191, 203–9, 213, 214, 218, 227, 229, 230, 266, 273, 302, 312–3, 314, 316, 319, 320–3, 325, 327–8, 330–1, 333, 339 Mithraism 332 mitten  19, 35 mobility  3, 69, 92, 113, 190, 196–7, 223, 240, 253–4, 261, 266, 268, 349, 367,402 money, see coins, coinage monophtongization  346, 348 Monte Cassino  323 morphology  233, 346n, 348–9 Morsum 187 Mucking 319 Muiden, see Almere Muiderberg 231 Münster  90–1, 125, 231, 323 murder, homicide  120, 128, 142, 144, 148, 149, 150, 211, 214, 326 natio, nationes  223–4, 236, 319n naval campaign(s), expeditions  214, 216–17, 225, 235 Nebel 187 neck-collar, neck-ring, necklace  32, 288, 289, 292, 389; see also jewellery Nederhorst ten Berg  94, 121 needle case  189 needles, nails  181, 189 Nehalennia 219 Neresheim 184

Nesse 91 Neustria, Neustrians  118, 120, 122, 205, 206, 208, 210–13, 215, 217 ‘New Frisians’  4, 45, 47, 56–8, 60, 61, 62, 103, 137, 218, 224–5, 236, 273, 277, 286–7, 292, 297, 363, 382 Nieblum  173, fig. 6.1 Niederdollendorf, gravestone  386 Niedersachsen, see Lower Saxony niello  289, 292, 296 Nivelles 205 nobiles (noblemen)  124, 125, 127, 148–9 Noord-Brabant, province  120, 317 Norddorf 174 Nordic, see Scandinavia, Scandinavian Nordliudi  228 Nordstrand 195 Norsemen, see Vikings North Frisia (Nordfriesland), North Frisian Islands  2, 3, 34, 35, 58–9, 66–7, 70, 166, 171–97, 252–3, 332–3, 347–9 North Germanic, see Germanic, language North Sea  1–6, 47, 59, 62–3, 66–7, 79, 81–3, 91–2, 94, 99, 101, 118, 121, 125, 171, 175, 177–9, 182, 186, 188–90, 192–3, 195–7, 206, 219, 223, 226, 249, 252–7, 260, 262, 264, 278, 287–9, 292, 295–6, 299, 300, 302–3, 313–14, 318, 320, 327, 341–2, 344, 346–7, 349, 362, 385, 389–90, 392, 402–3 North Sea Germanic, see Germanic, language North-Holland (Noord-Holland), province  30n, 45, 51, 54, 58, 63, 91, 140, 195, 215, 216, 228, 240, 290, 292–3, 298, 317, 318, 339, 364–5, 401 Northumbria  36, 299, 320, 323, 328, 342, 364 North-West Germanic, see Germanic, language Norway  266–7, 342, 343, 362, 378n, 386 nuclear regions  275, 287, 298 numismatics, see coins, coinage Nydam Style  294 Odin (Óðinn), Wodan/Woden  260, 276, 294–7, 300–1, 324, 331 Oegstgeest  16, 102, 284, 292 Óláfr Tryggvason  329 Old English, see English, language Old Frisian, see Frisian, language Old High German, see German, language Old Norse (Old Icelandic)  147 literature  295, 316, 393 Old Saxon  233, 241, 340, 344–9, 365–6, 369, 388 Oldeboorn  91, 109, 111–12, fig. 4.16 Oldenburg Klunderborg  35 Oldorf 16 Omgård 179 Ongendus  191, 319, 325

417

Z01 Hines Book B.indb 417

07/05/2021 13:31

Index Oosterbeek 231 Oosterbeintum  16, 19, 25, 26, 29, 47, 56, 147, 276–8, 280–1, 285–6, 303, fig. 3.6 Oosterbierum  113–14, fig. 4.18 Oostergo  24, 34, 48, 50, 54, 56, 63, 83, 86, 88–91, 96, 104, 109, 113, 115, 124, 152, 164, 203, 214, 330 Oostum  55–6, 59, 382, 385, 392, fig. 3.6 Oostwold 31–2 oral tradition(s), orality  217, 250, 252, 256, 259 origin myths  252, 259, 268, 296, 304, 323 ornament(s)  15, 16, 20, 30, 225, 226, 275, 277, 284, 288, 289, 290, 292, 293, 295, 296, 297, 299, 302; see also jewellery and dressaccessories; pendant high-status, precious metal  15, 225, 276, 288, 289, 290, 293, 299, 302; Orosius  226, 267 Oseberg, ship burial  255, 266 Oste, River  293 Ostfriesland (East Frisia)  16, 24–6, 30, 56, 81, 83, 86–9, 93–6, 98, 100, 103, 110–11, 115–16, 125, 126, 127, 141, 151–2, 197, 256, 275, 292, 330, 368, 383–5 Ostrogoths 164 Oswiu 321 otherness  228, 403 Otherworld  295, 301 Oude Rijn, River  84, 93, 100, 102, 115, 117, 120–1, 365 Overkeuren  259 overseas connections and relations  2, 5, 6, 29, 20, 47, 59, 62, 80–2, 82n, 92–3, 96, 101, 119, 125, 175, 178, 181, 183, 184, 187–8, 189, 190, 191, 192, 213, 218, 226, 232, 236, 249, 250, 253–4, 259, 260, 264, 266–7, 277–8, 289, 292–3, 296, 297, 300, 303, 313, 320, 339, 342–50, 355, 359–363, 367, 392, 403; see also trade and exchange Ovo 156 Pactus Legis Salicae, see Leges Barabarorum, Lex Salica Paddepoel 68 pagan, paganism, see pre-Christian religion pagus/-i, see district(s) palatalization 364n pallia fresonica (Frisian cloth)  18, 188; see also fresum; textile parochia (‘parish’)  322 Peada 218 peat bog(s), peat  45, 83, 84, 86, 90–1, 101, 117, 121, 123, 152, 229, 230, 239, 240; see also fenland; mire; wetland; marsh peat district  88, 92, 93, 103, 109, 195 Pellworm  173, 195, fig. 6.3

Penda, King  218 pendant  15, 20, 30–2, 61, 184, 290, 292, 295, 296, 331, 378, 382, 383n penny  36, 94, 95, 97–9, 101, 141; see also sceatt(a) peregrina (stranger)  231 personal law  239–40 Petrus, Sufridus  139 phonology, phonemics  339–49, 366, 375 Picts 314 pidgin language(s)  267 Pietersbierum  113–14, fig. 4.18 Pingjum 35 Pippin II, King (Pippin of Herstal)  116–23, 126, 208–12, 214–16, 227, 320–1, 330–1 Pippin III, King  99–100, 205, 214 Pippinid family  116–18, 205, 207, 210–12, 331 piracy  46, 223, 226 place-name(s)  61, 171, 313, 317, 329, 332, 339, 340, 344, 347, 349–59, 358, 363, 365 don, dun (Celt. *dūno- ‘fortified place’; OE dūn ‘village’) 357 *haim (PGmc ‘house, village’)  347, 350–7, 363 *ing (PGmc ‘belonging or pertaining to’)  347, 350–7, 363 *villa 355 Plectrudis  119–20, 210–13 Pliny the Elder  45–6, 249, 378 plough 25 Pogum  26–7, fig. 2.9 Pomponius Mela  378 Poppo, see Bubo portus  100, 110 pottery, pots  13, 15, 16, 17, 19, 22–4, 30, 33, 45, 47–63, 64, 65, 66–8, 70, 88, 95, 96, 175, 177, 179, 183, 184, 185, 188, 189, 190, 193, 195, 196, 199, 225, 274, 275, 376, 378, 385, 389 African Red Slip Ware  64 Anglo-Saxon style pottery  19–20, 35, 51, 53, 55–67, 175 Buckelurnen  65 ‘Chaucian’  49–51, 68; see also Wierum-style Driesum Ware, Driesum-style  51, 53, 55 egg-shaped  183, 185, 189 ‘Frisian’  48, 50–51 Frisian shell-tempered ware  183–4 globular 183 hand-made  22, 24, 47, 62, 179 Hessens-Schortens Ware, also weiche (frühe) Grauware  59, 60, 66, 67, 193 Jutish 189 Kugeltopf 22 Nordseeküstennahe Fundgruppe (formerly Chaucian–style)  48, 51, 56; see also Driesum Ware Oxfordshire Ware  61, 64

418

Z01 Hines Book B.indb 418

07/05/2021 13:31

Index Pingsdorf Ware  185 Rhenish ware  184–5 Rhine-Weser-Germanic (RWG) ware  48, 50 Samian Ware, see also terra sigillata 61 shell-tempered pottery  188 Stade-Harburger Gruppe  48 Trichterpokale  56n Tritsum Ware, also chaff or grass-tempered ware  60n, 65 völkerwanderungszeitliche verzierte Ware, see Anglo-Saxon pottery wheel-thrown  24, 34, 47, 87, 95, 96, 102n, 185, 188, 190 Wierum-style  50, 68 pottery production see craft production Praal 231 pre-Christian religion, pre-Christian  5, 6, 125, 143, 156, 214, 219, 223, 233, 237, 259, 273, 276, 293, 295, 296, 297, 300–2, 311, 313–27, 328–30, 331–2, 346, 362, 378–82, 404 princeps/-cipes (princes)  116, 209, 212, 229 Procopius  205–6, 218, 402 production sites  35, 97, 174, 177–8, 179, 184, 185, 253; see also craft-production Proto-Germanic, see Germanic, language psalmos, psalms  323, 328 Ptolemy  5, 50, 223, 225, 235, 378 Quentovic 264 quernstone(s), see millstone(s) Radbod (Redbad)  2, 36, 79, 83, 94, 98, 116–24, 125–8, 137, 154–7, 163, 164, 191–2, 196, 203, 206–18, 227, 228, 231, 240–1, 316, 319–21, 324, 328–30, 332, 361–2, 401 Radigis 206 Raganfred / Ragamfred  120, 211–12, 214 Rantgar 211 Rantrum  173, 174 Rasquert  17, 30 Rathmelsigi 209 Rauwerd  294, 297 rebellion  214, 229, 234, 322–3 Red Sea  30, 34 Redbad, see Radbod Reepsholt 138 regnum/-a  227, 231 Reitdiep, River  14, 57 relic(s) (literary, linguistic)  237, 344, 347, 348, 365 relic(s) (material)  119, 143, 154, 231, 322 reticella decoration (glass)  181 retraction 341–2 rex/reges (king)  116, 117, 123, 155, 156, 161–2, 196, 208–9, 218–9, 296, see also kingship Germanic terminology  155, 161–2, 218, 321

Rhine, River  13, 16, 25–6, 34, 48, 68, 71, 91, 98, 99, 101–2, 108, 117–22, 127, 152, 162, 205–6, 210, 213–14, 217–19, 223, 225–8, 232, 234–5, 255, 265, 277, 293, 313, 316, 322, 328, 331, 346, 361, 383, 388–9, 392, 394–5 Rhineland, Rhenish  24, 26, 84, 93, 99, 123, 128, 184–5, 188–9, 208, 214–15, 219, 255, 364–5, 388 Ribe  37, 126, 181, 183–5, 187–90, 191–2, 236, 333, 385 Rijnland  84, 88, 91–3, 95, 117, 123 Rijnmond 100 Rijnsburg(-De Horn)  16, 110, 281, 284–6, 290–2, 295 Rijswijk  49, 68 ringfort(s)  171, 172, 177, 185–7, 188, 190, 191, 192–3, 194 Ringwallburgen, see ringfort(s) ritual practice  30–2, 33–4, 49, 52, 54, 62, 185, 224, 229, 276, 277, 285, 292, 312, 314, 316, 319, 320, 331–2, 383, 384, 396 Riustring Old Frisian law  148 Roma, people  7–8 Romance, language  351, 364n Rome  71, 118–19, 209–10, 224, 231, 323, 330, 332, 378 Rorik 128 Ruimel 120 rune stone(s)  192, 378n, 379 runes  276, 352, 353–4, 361, 375–88, 390, 393–5, 403 runic coins  37, 383–4, 386–8 runic inscription(s)  4, 13, 32, 33, 59, 192, 195, 250, 340, 375, 349–60, 363, 364, 365, 375–97, 403 Aalen neck-ring  389 Achlum bracteates  383, 393 æniwulufu tremissis  386 Amay comb  356, 384–5, fig. 13.8 Arum miniature sword  382, fig. 13.4 Ash Gilton sword  396 Belgorod comb  386, 389 Bergakker scabbard mount  365, 388, 389, 390–2, 394, fig. 13.11 Bernsterburen whale bone staff  32, 33, 383–4, 391, 396, fig. 13.25 Borgharen buckle  365, 388, 392, 394, fig. 13.12 Britsum wooden stick  383n, fig. 13.14 Caistor-by-Norwich roe deer astragalus  390 Dahmsdorf spearhead  389 Einang stone  379 Elisenhof comb  385 Fallward stool  389, 390, 393 Ferwerd comb case  59, 382, 385, 392 Fréthun ring-sword  376, 390 Frienstedt comb  376n, 385 Geltorf II bracteate  363

419

Z01 Hines Book B.indb 419

07/05/2021 13:31

Index Grenay ring-sword  376, 390 hada solidus 386 Hamwic (Southampton) knucklebone  359, 382 Heide bracteate  363 Hitsum bracteates  383 Hogebeintum comb  385, 393 Kantens comb  385 Kowel spearhead  389 Lauchheim comb  386 Lețcani spindle-whorl  389 Loveden Hill pot  389, 396 Noleby stone  379 Oostum comb  59, 382, 385, 392, fig. 13.3 Pforzen buckle  393, 396 Pietroassa neck-ring  389 Ribe comb  385 Rozwadów spearhead  389 Saint-Dizier ring-sword  376, 384, 390, fig. 13.7 Setre comb  386 Sievern bracteates  389 skanomodu solidus 386 Spong Hill pot stamp  382, 398–90, 393–4 Sukhodil buckle  389 Szabadbattyán buckle  389 Toornwerd comb  381–2, 385, fig. 13.5 Vimose comb  386 Watchfield purse-mount  359, 364 welad solidus  382, 386, fig. 13.9 Westeremden A weaving slay  379, 385, 396, fig. 13.2 Westeremden B wooden stick  382, 396, fig. 13.6 Whitby comb  385 Wijnaldum bone piece  382, cf. 383n Rustringen 229 sacrifice  19, 156–7, 163, 166, 223, 229, 259, 264, 316, 318, 322, 326, Sædding 179 Sage van Karel ende Redbad  260, 361–2, St Wandrille, Abbey  207–8 salt marsh  13–14, 16, 20, 22, 25, 30, 34, 45, 81–2, 85, 86–7, 93, 95, 96, 104, 106, 109, 122, 128, 174, 177, 178, 224, 225, 230, 239, 263–4, 274, 350, 361, 382; see also marsh San Marino, Republic  1–2, 8–10 satrapa, satrapas plurimos 228 Saxmund  140–1, 145, 153–4, 258 Saxnot 300 Saxon Wars  229, 273, 292 Saxons, Continental  1, 4, 152, 157, 162, 164, 206, 217, 223–42, 277, 299, 313, 320, 322–3, 327, 332, 346, 361, 390, 392, 402 Scaltheim 101

Scandinavia, Scandinavians  1, 2, 4, 22, 80–2, 93, 95–6, 155–6, 158, 163–4, 183–9, 191–2, 194, 196, 197, 219, 226, 250, 255, 259, 266–7, 276, 279, 288–98, 300–2, 313, 316, 320, 322, 329, 332, 333, 346, 353, 361–2, 376, 379, 382, 389 Scandinavian, language  9, 197, 346n, 348n, 357 sceatt(a), sceattas  6, 29, 36, 37, 94, 97, 183, 188, 195, 205, 254, 383, 386 Anglo-Frisian, Frisian, Porcupine, transitional and Wodan-monster types  183 Continental rune type  37, 383–4, 386 hoards  188, 195; see also coin hoard, Goting Primary Phase, Series D and E (c. AD 680–715) 98 Schagen-Muggenburg 54 Scheldeland  96, 98, 101, 110, 123–4 Scheldt, River  13, 16, 84–5, 93, 101, 110, 118–19, 122, 205–6, 316 Schiermonnikoog 267–8 Schlei, fjord  186 Schleswig-Holstein  19, 51, 58, 63, 65, 68, 178, 226, 230, 299, 356, 376, 385, 389 Schweindorf, see Welad Scots 314 screona (weaving hut) see sunken hut; textile production seax, see weaponry seend (OFr: synodal jurisdiction)  89 Seine, River  207–8, 226 self-perception; self-identification  224–5, 226, 236–7, 238. serf, see litus/-i Sergius, Pope  118, 209 servus/-i (slaves)  146, 148, 149; see also slave(s) Seventeen Statutes, the (Old Frisian Law)  138, 140n Sexbierum  113–15, figs. 4.15, 4.18 ship(s) and seafaring  24, 25, 26, 27, 46–7, 68, 82, 83, 94, 100, 101, 102n, 108–9, 112–3, 118, 178, 184, 185, 186, 188, 190, 191, 214, 216, 223, 230, 250, 251, 255–6, 257, 258, 260–266 canoe(s), canoe-like, half-canoes  26–7, 46, 262, 264, 266 clinker-built vessel  26, 46–7, 255–6, 262, 264 logboat(s)  46, 255 miniature(s), model(s)  26–7, 33, 251, 256 ship-related finds  250–1, 255–6, 260 ship symbolism, imagery  251, 255–6, 260, 265 ships’ timbers  26, 184, 188, 255–6 ship- (and boat-) building  184–5, 258 nail(s), clinker(s), rivet(s)  26, 46, 184,185, 188, 255–6, 262 ship and boat burials  26, 255, 256, 260, 262, 266 ship settings  251 shoe(s), see footwear

420

Z01 Hines Book B.indb 420

07/05/2021 13:31

Index shrine  203, 211, 259, 316–7, 321, 326–7 Siccama, Sibrandus  139–40 Sidonius Apollinaris  223 Siebs, Theodore  318, 339 Sievern  197, 289, 363, 383, 389, 390 Sigibert of Gembloux  211, 214 Sigifrid, King  229 silver  6, 9, 15, 29, 32, 60, 61, 97, 141, 156, 179, 183, 187, 188, 190, 192, 225, 276, 277, 288, 289, 292, 293, 297, 298, 300, 303, 304, 383, 386; see also hacksilver; coins, coinage; jewellery silver-gilt  295, 383, 388 silversmithing 288 Sincfal, River (now the Zwin)  3, 13, 83, 88, 94, 121, 125, 140, 141n, 148, 166, 196, 227, 239, 252, 298 skaldic poetry  329 Skanomodu, runic coin  382, 386 slave(s)  95, 143,148–9, 150, 163, 257, 379 Slavs, Slavic, Slavonic  163, 187, 229–30, 235, 320, 325, 327 smoothing 341–2 Snape 319 Sneek 22 soapstone  178, 183 social structures and societies  7–11, 45–51, 54–7, 58, 59, 60–1, 62, 63, 71, 81, 86, 112–3, 116, 124–5, 127, 137, 142–3, 147, 148–9, 150, 151, 158, 162, 163, 164, 216, 219, 224, 225, 227–8, 231, 232, 234, 237, 240, 242, 252, 253, 275–6, 287, 288, 292, 293, 298, 300, 301–3, 313, 320–1, 323–4, 329–330, 349–50, 357, 367–8, 378 Soissons  207, 209 solidus/-i  15, 29–30, 141, 142, 147, 150, 258, 286, 288, 290, 292, 293, 386 Solleveld (Den Haag)  26, 256, 262, 281, 284–6 Sonatorrek (Old Norse poem)  329 South Germanic, inscriptions  361, 378, 382, 396 Southampton  37, 359, 382, 390 South-Holland (Zuid-Holland), province  26, 46, 49, 68, 152, 256, 281, 284–5, 292, 365 spade 24 Spain  147, 210, 363 spear, see weaponry spindle-whorl(s)  16–7, 389 Spong Hill  61, 65, 382, 389, 393–4 Staveren  88, 100–1, 109 Stephen of Ripon  203, 206–8 stirrup(s)  29, 35 strap mount (horse gear)  297 streepband (decoration)  51 Süderende 187 Suevi 217 Suidbert, see Swithbert

sunken-feature building (SFB) (Grubenhaus), see sunken hut sunken hut  52, 62, 70, 96n, 150, 174, 178–9, 180–1, 183, 184, 185, 188, 190, Sussex 357 Susteren  120, 210, 331 Sutton Hoo  156, 255, 301 Swithbert (Suidbert)  119, 209 sword, see weaponry Sylt  3, 58–9, 67, 171–7, 184–5, 187, 190, 193–5, 347, 349 Synodical Law of Central Frisia, the (OFr law) 138 Tacitus  5, 45–6, 71, 155–7, 223–4, 234, 378 Taxandria, Toxandria  120, 205, 210, 330, 388 temple(s)  121, 140, 143, 166, 214, 219, 259, 264, 313, 316, 326, 327, 329, 331–2, 333 terp,(pl) terpen  5, 13–16, 18–29, 30, 45, 47–8, 50, 51, 52, 56, 61, 70, 86–7, 103, 104–7, 110, 112, 113, 124, 127, 172, 174, 176–8, 189, 190, 194, 205, 252, 253, 255, 263, 274, 382–4 terp area, region, district  34, 35–6, 45–7, 49, 54, 58, 59, 64, 65, 79–80, 82, 108, 109, 126–8, 195, 224, 226, 240–1, 361, 375, 383–4 terra sigillata, see also pottery, Samian Ware, African Red Slip Ware  61, 64, 376 territorial law  239–40 tesserae  26 Texel  16, 58, 84, 93, 194, 253, 401 textiles and textile production  16–19, 25, 95–6, 177–8, 181, 184, 185, 188, 190, 274; see also screona; sunken hut; weaving batten; weaving hut; loom; loomweight; pallia fresonica; spindle-whorl linen  16, 95 wool  16–8, 96n, 177–8, 190 The Hague (Den Haag)-Solleveld, see Solleveld Theoderic the Great, King  386 Theoderic, Count  229 Theudebert I, King  206 Theudesinde, Thudsinda  119–20, 211 Theudoald  211, 213 thing (assembly)  137, 142, 145, 148, 151–2, 157–8, 162, 164, 217, 237, 238, 239, 241, 262–3, 322, 328; see also Upstalbam Thiofrid 157 Thuringia (Thüringen), Thuringians  8, 208, 302, 313, 322, 376 tidal marsh, see salt marsh tides and tidal zone  6, 86, 91, 100, 104, 113, 177, 178, 182,186, 239, 251, 252, 257–8, 260, 263–4, 318, 382 Tiel  102, 254–6, 383, 388 Tinnum, Tinnumburg  67, 173, 175–7, 184–8, 190, 193, fig. 6.1

421

Z01 Hines Book B.indb 421

07/05/2021 13:31

Index Tissø  316, 332 Tiwaz, Týr  151, 316 Tofting 174 toll  80, 93, 96, 99, 100–1, 119 Tolsum, writing tablet  378–9 Toxandria, see Taxandria trade and exchange  3–4, 6, 20, 29, 34, 36, 47, 49–50, 62, 68–9, 79–82, 87–8, 92, 93, 94–103, 107–9, 112–6, 121–3, 125, 126, 150, 175, 176, 177–8, 181, 183, 184, 186, 187–8, 189, 190, 191, 224, 230, 237, 251, 252, 253–5, 260–1, 263, 266–8, 273, 275, 288, 292, 293, 302, 349, 386, 404; see also overseas connections elite exchange  62, 95–6, 187, 254, 273, 275, 288, 292, 293, 298, 302 exchange of knowledge  46, 49, 62, 68–9, 297, 366 ‘Frisian trade’  94, 122, 188, 189 gift-exchange, gift-giving  9, 47, 49, 51, 55, 59, 273, 275, 288, 292, 293, 302, 378, 386, 392, 395,404 import and export  22, 25, 26, 35, 47–8, 69, 81, 88, 95, 96, 175, 177–8, 181, 183, 184, 185, 188, 189, 190, 233, 266, 275–7, 285, 288, 297, 300, 383 trading sites  4, 5, 13, 37, 79, 81, 82, 94, 97, 99–102, 103, 112–3, 115, 118, 119, 126, 176, 177, 179, 184, 186, 188, 190, 192–3, 196, 255, 257 trade terps (Du handelsterpen, Germ. Langwurten, Langwarf)  81, 82, 112, 126, 178 Trajan, Emperor  325 treasure, treasury  15, 70, 212, 219, 277, 288, 303, 332–3; see also hoard(s) Treaty of Verdun (AD 843)  230 Treene, River  178, 186, 188 tremissis/-es  147, 149, 155, 176, 193, 195, 290, 384–5 Dronrijp type  176, 193 tribute  209, 288 trough 22–3 Truso 267 Tuda  33, 384, 396, fig. 13.15 tumbler, see glass vessel, Glockentummler tweezers 19–20 Týr, see Tiwaz Tzum  32, 88 Ulrum 15 Unno 229 upchirica  127 Uppåkra  316, 332 Upstalbam assembly (Upstalboom)  152, 262 Usquert 274 Utersum 187

Utrecht  5, 89–91, 94, 98, 100–2, 109, 116–21, 123–5, 127, 137, 203–10, 213, 217, 228, 230–1, 255, 297, 300, 312, 316, 320, 322, 330–1 Valhǫll 332 Valkenburg, see Katwijk Vandals 363 Vanimundus 386 Vecht, River and region  93–4, 100, 117, 119–23, 205 Velsen  91, 94 Veluwe  26, 232 Venantius Fortunatus  206, 237 Verdun 230–1 Verritus  71, 224, 378 Vikings (Norsemen), Viking Period/Age  29–30, 33, 86, 89, 91, 95, 101–2, 111, 123, 126, 137–8, 158, 165–6, 191, 194, 219, 225, 227, 230, 251, 254–7, 262, 266–7, 297, 299, 301, 316, 322, 324, 327, 350, 362, 378n, 393, 402 Visigoths  8, 163–4 Vita Vulframni (Life of Wulfram of Sens)  163, 205, 207–8, 213, 218, 318, 320 Vlaardingen  46, 68, 102, 109 Vlie, River  79–88, 93–100, 103, 108–9, 113, 120–5, 127–8, 137–8, 140–1, 145, 151–2, 166, 216, 229, 253–4, 293, 326, 344n, 365–6 Vǫlundr (Weland)  393 Vorbasse  174, 179 Waal, River  93, 101–2, 117–18, 123, 388 Waalre 120 Wadden islands  86, 267 Wadden Sea  13, 16, 32, 91, 173–4, 189, 195, 252 Walcheren  118–19, 205, 316, 322 Wales, Welsh  357, 401 Waltbert, Count  231 Wandregisl 207 Wangerland 350 warfare and military power  7, 29, 46, 68, 99, 116, 117, 120, 121, 123, 124, 137, 142, 147, 151, 155, 156, 185, 190, 191, 193, 209–10, 212–15, 216–7, 223, 225–6, 227, 228–9, 234, 236, 238, 255, 257, 265, 275, 293, 303–4, 322, 328, 350, 359, 363, 375, 378, 392, 394–5 Warni 206 weapon grave(s)  238–9, 292, 293 weaponry, weapon(s)  26, 29, 33, 95, 225, 239, 257, 376, 379, 383 bow 28–9 seax  26, 29, 35; scabbard fig. 10.1 spear and spearhead (incl. angones)  26, 29, 35, 389 swords, blade grip and pommel, fittings, scabbard  29, 35, 295, 316, 376, 383–4, 389–90, 396

422

Z01 Hines Book B.indb 422

07/05/2021 13:31

Index ring-sword  376, 384, 390 throwing axe  29 weaving batten(s)  18, 379 weaving hut(s)  96n, 152, 181; see also sunken hut(s); screona weaving, see textile production Wehe 317 Welad, runic coin  382, 386, fig. 13.9 Wends  214, 217 Wenningstedt 176 Werden  18, 96, 115n, 230–2, 235, 240, 323 wergild  140, 141, 142–3, 144–9, 150, 163, 242, 258, 404 Weser, River  3, 13, 55, 63, 80, 83, 86, 88. 93, 94, 121–2, 125, 137–8, 141, 145, 151–2, 162, 166, 174, 196, 206, 225, 227–30, 252, 264, 298, 344, 346 Wessex  320, 343, 363 West Frisia (West Frisian)  79, 81, 83, 85–6, 87, 88–9, 91, 93, 94–5, 96n, 98–9, 108, 110, 115–16, 120–1, 122–4, 141, 145, 156–7, 188–9, 203, 333 West Germanic, see Germanic, language Westerbierum  113, fig. 4.18 Westergeest Weerdeburen  35 Westergo  24, 29, 34, 48, 54, 57–8, 83–4, 86, 88–91, 93, 95–6, 109, 113, 121, 123–4, 152, 164, 165, 203, 214, 227–8, 232, 253, 264, 289, 293, 330 Westerland 187 Westerschelde 83 Westerwanna 58 Westerwijtwerd 35 Westflinge  100 Westphalia, Westphalians  228, 231–3, 323, 328, 365 wetland  178, 179, 182–4, 208, 223; see also fenland; mire; peat-bog; marsh; salt marsh whale bone, ivory  32, 33, 383–4, 391 wheel 26 whetstone(s)  25, 177, 183, 184, 188, 190 Whitby 385 Synod of  321 wīc  4, 81, 320; see also trading; trading sites Widele, River  101 Widukind  228–9, 231, 232, 233, 323, 404 Wiedingharde  171, 173–5 wierden, see terp(s) Wieringen  16, 84, 86, 93, 100, 121, 123, 156, 253, 290

Wieuwerd  290, 295 Wigmodia 323 Wijnaldum(-Tjitsma)  15, 26, 35, 45, 47, 51, 57, 108, 113–15, 156, 253, 255, 277, 286, 290, 293, 295, 300, 303, 382–3, figs. 4.13, 4.18 Wijster 54 Wildeshausen 231 Wilfrid  155, 205–6, 208–9, 320 Wilgefortis, Saint  330 Willehad  230, 312–13, 323, 326 William of Orange  239 Willibald  151, 207, 209, 213–14, 322 Willibrord  91n, 94, 118–21, 123, 156–7, 191, 196, 203, 205, 207–10, 213–15, 218, 273, 302, 312–13, 316, 318, 319–22, 325–6, 330–1, Wilp 322 Wiltaburg  109 Winsum 109 Witla 101–2 Witsum  173, 176 –7, 178–90, 191–2, 197, fig. 6.1 Wittewierum 152 Wlemar  96n, 140–1, 145, 153, 161 Wodan/Wodan, see Odin Woensel 317 Woerden  94, 121 Woltersberg 194 Wons Hayumertille  35 wood, yew  28–9, 381, 382, 383, 390, figs. 2.10, 13.14 wood-turning, see craft-production workshop(s)  47, 289, 296 Wouters, Cornelis  139 wrist-clasps 267 Wrixum  173–5, fig. 6.1 Wrssing  117, 119–20, 210, 231, 321 Wulfram  5, 156, 207–8, 213, 218, 312, 320–1 Wurten, wurts, see terp(s); ‘commercial wurt’, ‘long wurts’ see ‘trade terps’ Yeavering  316, 332 yoke 25 York  155, 257 Zeeburg 194 Zeeland, province  140–1, 152, 194, 345–6, 365 Zierikzee 110 Zuilen 215 Zwin, River, see Sincfal

423

Z01 Hines Book B.indb 423

07/05/2021 13:31