Issuu on Google+


PUBLICATIONS OF THE INSTITUTE OF ARCHAEOLOGY, UNIVERSITY COLLEGE LONDON Director of the Institute: Stephen Shennan Publications Series Editor: Peter J. Ucko The Institute of Archaeology of University College London is one of the oldest, largest and most prestigious archaeology research facilities in the world. Its extensive publications programme includes the best theory, research, pedagogy and reference materials in archaeology and cognate disciplines, through publishing exemplary work of scholars worldwide. Through its publications, the Institute brings together key areas of theoretical and substantive knowledge, improves archaeological practice and brings archaeological findings to the general public, researchers and practitioners. It also publishes staff research projects, site and survey reports, and conference proceedings. The publications programme, formerly developed in-house or in conjunction with UCL Press, is now produced in partnership with Left Coast Press, Inc. The Institute can be accessed online at ENCOUNTERS WITH ANCIENT EGYPT Subseries, Peter J. Ucko, (ed.) Jean-Marcel Humbert and Clifford Price (eds.), Imhotep Today (2003) David Jeffreys (ed.), Views of Ancient Egypt since Napoleon Bonaparte: Imperialism, Colonialism, and Modern Appropriations (2003) Sally MacDonald and Michael Rice (eds.), Consuming Ancient Egypt (2003) Roger Matthews and Cornelia Roemer (eds.), Ancient Perspectives on Egypt (2003) David O’Connor and Andrew Reid (eds.), Ancient Egypt in Africa (2003) John Tait (ed.), ‘Never had the like occurred’: Egypt’s View of its Past (2003) David O’Connor and Stephen Quirke (eds.), Mysterious Lands (2003) Peter Ucko and Timothy Champion (eds.), The Wisdom of Egypt: Changing Visions Through the Ages (2003) Andrew Gardner (ed.), Agency Uncovered: Archaeological Perspectives (2004) Okasha El-Daly, Egyptology, The Missing Millennium: Ancient Egypt in Medieval Arabic Writing (2005) Ruth Mace, Clare J. Holden, and Stephen Shennan (eds.), Evolution of Cultural Diversity: A Phylogenetic Approach (2005) Arkadiusz Marciniak, Placing Animals in the Neolithic: Social Zooarchaeology of Prehistoric Farming (2005) Robert Layton, Stephen Shennan, and Peter Stone (eds.), A Future for Archaeology (2006) Joost Fontein, The Silence of Great Zimbabwe: Contested Landscapes and the Power of Heritage (2006) Gabriele Puschnigg, Ceramics of the Merv Oasis: Recycling the City (2006) James Graham-Campbell and Gareth Williams (eds.), Silver Economy in the Viking Age (2007) Barbara Bender, Sue Hamilton, and Chris Tilley, Stone Worlds: Narrative and Reflexivity in Landscape Archaeology (2007) Andrew Gardner, An Archaeology of Identity: Soldiers and Society in Late Roman Britain (2007) Sue Hamilton, Ruth Whitehouse, and Katherine I. Wright (eds.) Archaeology and Women (2007)


James Graham-Campbell Gareth Williams Editors

Walnut Creek, California

LEFT COAST PRESS, INC. 1630 North Main Street, #400 Walnut Creek, CA 94596 Copyright © 2007 by Left Coast Press, Inc. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior permission of the publisher. ISBN 978-1-59874-222-0 hardcover Library of Congress Cataloging-in-Publication Data Silver economy in the Viking age / James Graham-Campbell, Gareth Williams, editors. p. cm. Includes bibliographical references and index. ISBN-13: 978-1-59874-222-0 (hardcover : alk. paper) ISBN-10: 1-59874-222-1 (hardcover : alk. paper) 1. Coins, Scandinavian. 2. Coins, Medieval. 3. Silver coins—Europe—History. 4. Coinage—Europe—History. 5. Money—Europe—History. 6. Numismatics—Europe—History. 7. Viking antiquities. 8. Coin hoards—Europe. 9. Europe—Antiquities. 10. Europe—Economic conditions—To 1492. I. Graham-Campbell, James. II. Williams, Gareth. CJ3094.S57 2007 737.49363—dc22 2006035671 Printed in the United States of America Typeset in Times by Newgen Imaging Systems (P) Ltd., Chennai, India. Additional Production: Penna Design, Abbotsford, British Columbia The paper used in this publication meets the minimum requirements of American National Standard for Information Sciences—Permanence of Paper for Printed Library Materials, ANSI/NISO Z39.48–1992. Cover Image: Part of the silver hard from Cuerdale, Lancashire, deposited c 905. Reproduced by permission of the Trustees of the British Museum. 07 08 09 10 11

5 4 3 2 1

Contents List of Illustrations vii Notes on Contributors xi Preface xiii List of Abbreviations xv 1



































List of Illustrations 1.1 2.1

2.2 2.3 2.4 2.5



3.1 3.2 3.3 3.4 3.5 3.6 3.7 3.8

3.9 3.10 3.11

Distribution map of thrymsas and sceattas from England (i–iv) Selected coins, scale c 3/2 (© Gabriel Hildebrand, Royal Coin Cabinet, Stockholm). (i) Wodan/Monster sceatta, Frisia, from c 715; (ii) Charles the Great silver denier, Dorestad, from 771–93/94; (iii) KG 5, from the period of Louis the Pious, probably the 820s; (iv) KG 3, from the period of Louis the Pious, probably the 820s Diagram of Combination-Groups (KG) 1–6, percentages (after Malmer 1966, plate 42) Combination-Groups (KG) 3–6, scale c 1/1 (© B Malmer) Weights of Carolingian obols and of KG 3–6 (© B Malmer) Finds of KG 1–6, with geographical division into Areas I–VI: Area II = Schleswig-Holstein; Area IIIa = Jutland, including Ribe (after Malmer 1966, plate 55) Scale c 1/1: (1) KG 8, Area II (‘Hedeby’), from c 950; (2) KG 9b, Area II (‘Hedeby’), from c 975; (3) KG 10a, Area III (Denmark), from c 975/80 and (4) KG 11, Area III (Denmark), from c 975/80 (© B Malmer) KG 3 reverses showing fish symbol (© B Malmer): 1–3, with ship; 4, with enlargements of fishes with ships; and 5–7, with the Dorestad axe Hedeby (Haithabu): the topographical situation (after Elsner 1992) The chronology of minting at Hedeby (after Hatz 1984, fig 131) The distribution of hoards in Schleswig-Holstein The distribution of hoards in Schleswig-Holstein (790–980) The distribution of hoards in Schleswig-Holstein (980–1050) The distribution of Hedeby coins in Schleswig-Holstein The Steinfeld hoard, with Hedeby coins (i)–(v) Six coins from Hedeby: (i) sceatta, settlement-find; (ii) Abbassid dirham, with a runic graffito, settlement-find; (iii) Byzantine gold solidus, grave-find; (iv) Northumbrian styca, settlement-find and (v) Norwegian penny struck by Harald Hardråde, grave-find Coins of different provenances (in %) of the Hedeby finds (settlement and graves) Main coin groups of different provenances (in %) of the Hedeby finds (settlement and graves) Chronological distribution of the main coin groups at Hedeby (settlement and graves)


14 16 17 19



25 30 32 34 35 36 37 38

39 41 41 42

3.12 3.13 4.1 5.1 5.2

5.3 5.4 5.5 5.6 6.1. 6.2 6.3 6.4 6.5 6.6 6.7 6.8 6.9 6.10 6.11 6.12 6.13 6.14 6.15 6.16 7.1


Distribution of Viking-Age finds in the districts of Angeln and Schwansen, eighth and ninth century (after Willroth 1992, map 69) 44 Distribution of Viking-Age finds in the districts of Angeln and Schwansen, tenth century (after Willroth 1992, map 70) 45 A selection of Cuerdale coins to demonstrate the progressive extent of the pecking with age (© Trustees of the British Museum). 50 Distribution map of finds of gold coins in Britain, 700–1200 72 Histogram of gold coins in the British Isles, 700–1200 (Source: Appendices A and B; fractional coins counted as fractions, base forgeries omitted) 79 British finds of gold coins: A2–A14 83 British finds of gold coins: A15–A25 84 English gold coins with meaningful inscriptions: B1–B8, with comparative material 86 Gold ingots and hack-gold from England: C1–C10 88 Charles the Bald, deniers, GDR-type, mint of Rouen. © Yohann Deslandes. 103 Evolution at the mint of Rouen. 105 Evolution at the mint of Bayeux. 105 Evolution at the mint of Curtisasonien. 106 Degenerated GDR denier, mint of Bayeux. © Peter Woodhead. 106 Degenerated GDR denier, mint of Rouen. © Peter Woodhead. 107 Degenerated GDR denier, mint of Rouen. © Yohann Deslandes. 107 Degenerated GDR denier, mint of Rouen. 107 Degenerated GDR denier, mint of Rouen. © Peter Woodhead. 108 Degenerated Carolingian obol, possibly Clermont or Rouen. © Yohann Deslandes. 108 Temple-type imitations from the Coudres hoard. 110 Temple-type imitations from the Evreux hoard. 110 William Longsword, deniers, Rouen. © (c) The National Museum of Denmark; (d) Yohann Deslandes; (e) F. Dugue. 112 William Longsword deniers, Rouen. 113 Modern forgery of William Longsword denier. © Peter Woodhead. 113 Rouen, Saint-Ouen monastery, deniers. © (a) Yohann Deslandes; (b) The National Museum of Denmark. 114 The distribution of arm/neck-rings in hoards of different complexity: (a) hoards from Öland, tenth–eleventh centuries; (b) hoards from Denmark, ninth–twelfth centuries 128 The distribution of bars and ingots in hoards of different complexity: (a) hoards from Öland, tenth–eleventh centuries; (b) hoards from Denmark, c 970–1016 129


8.1 8.2

8.3 8.4 8.5 8.6 8.7 8.8

9.1 9.2 9.3 9.4 9.5

The distribution of hack-silver hoards of different complexity: (a) hoards from Öland, tenth–eleventh centuries; (b) hoards from Denmark, c 970–1016 ‘Permian’ rings found on Öland (© Birgitta Hårdh) Distribution of complete ‘Permian’ rings: (1–2) Perm/Vjatka area; (3–4) Moscow area; (5) Finland; (6) Estonia; (7) Gotland; (8) Öland and (9) Jutland Weight grouping of the ‘Permian’ rings in Russia, Sweden and Denmark Polyhedral weights found at Uppåkra, Skåne, Sweden (© B Almgren, LUHM) Some terminal knobs with graffiti (scale 1:1): upper row, Russia; lower row, Gotland (© Birgitta Hårdh) A small striated ring (after Stenberger 1958, Textabb 18) Distribution of weights of rings weighing c 200 g from Perm and Vjatka Part of a ring from the Perm gouvernement (State Hermitage Museum, St Petersburg): note the incised star figure (© Birgitta Hårdh) Distribution map of the early Viking-Age hoards from Ireland (Classes 1–5) Pie-chart showing the relative proportions of hoards of Classes 1–5 amongst the Viking-Age finds from Ireland Viking-Age silver hoard of Class 4, from Cloghermore cave, Co Kerry (University College Cork Audio-Visual Services) Viking-Age silver hoard of Class 5 from Kilmacomma, Co Waterford (© National Museum of Ireland) Viking-Age silver hoard of Class 5 from ‘Co Antrim’ (© Ulster Museum)

129 136

137 139 140 141 142 144

145 153 154 157 158 158

Notes on Contributors Marion Archibald is now retired, but was for over 30 years responsible for the curation of the early medieval coins in the British Museum. She has published extensively on AngloSaxon as well as Norman and late medieval coins, currency and numismatic history. Mark Blackburn is Keeper of Coins and Medals at the Fitzwilliam Museum and Reader in Numismatics and Monetary History at the University of Cambridge. He is currently also President of the British Numismatic Society. Märit Gaimster is currently working for Pre-Construct Archaeology as their Finds Manager. She is an established scholar in the field of early medieval archaeology in Britain and Scandinavia, with a focus on the use of metal and metalwork as different forms of media. James Graham-Campbell is Professor Emeritus of Medieval Archaeology, after recently retiring from the Institute of Archaeology, having worked at University College London since 1973. He has specialised in the study of Viking-Age silver hoards and brooches, and early medieval Insular metalwork. Birgitta Hårdh is Professor of Medieval Archaeology at the University of Lund, with her main research interests being in the Iron Age and Viking Age. She has published many important works on the Viking-Age silver of Scandinavia and the Baltic region. Susan Kruse has studied Viking-Age ingots and weights, with an interest in Viking-Age economic systems. She now lives in the Highlands of Scotland, teaching and working on archaeological, environmental and local history projects. Brita Malmer is Professor Emerita of Numismatics at the University of Stockholm, having been Director of the Royal Coin Cabinet (1971–79), and editor of the Catalogue of Coins from the Viking Age found in Sweden (1971–93). D M Metcalf is Professor Emeritus of Numismatics at the University of Oxford, where he is also an emeritus Fellow of Wolfson College, having been Keeper of the Heberden Coin Room in the Ashmolean Museum. Jens Christian Moesgaard has been Assistant Keeper at the Royal Collection of Coins and Medals, Copenhagen, since 1997, having worked previously at museums in both France and England. He has published several articles on coins and coin circulation in France, England and Scandinavia from the Viking Age to the Early Modern Period. John Sheehan is a Senior Lecturer in the Department of Archaeology, University College Cork. His research interests focus on the Viking Age in Ireland, particularly on VikingAge silver, but also Ireland’s early medieval ecclesiastical archaeology. Ralf Wiechmann is archaeologist and numismatist at the Museum für Hamburgische Geschichte, in Hamburg, where he is responsible for both the medieval department and the coin cabinet. His main area of research has been the monetary history of the Viking Age. Gareth Williams has been Curator of Early Medieval Coinage at the British Museum since 1996. His research interests are focussed on the history and archaeology of early medieval Britain and Scandinavia, and particularly on interdisciplinary approaches to the study of coinage in its broader context. xi

Preface During April to August 2000, a temporary exhibition was mounted at the British Museum by Dr Gareth Williams (Department of Coins and Medals), entitled ‘Paid in Burnt Silver: Wealth and Power in the Viking Age’, with the support of the Royal Norwegian Embassy, through their ‘Visions of Norway’ programme. The two of us considered that this important event provided an appropriate occasion for the British Museum and the Institute of Archaeology, UCL, to organise a joint symposium on the subject of the economic uses of Viking silver, to be co-sponsored by the Department of Coins and Medals (BM) and the Complex Societies Research Group (IoA). Some 40 international specialists, comprising archaeologists, historians and numismatists, were invited to participate, with 12 papers planned for delivery on the theme of ‘Silver Economy in the Viking Age’. In the event, Svein Gullbekk, who was to have spoken on ‘Monetary Development and State Formation in 11th-century Scandinavia’ was unfortunately unable to be present, and a paper on ‘Coinage, Kingship and State Formation in the Late Viking Age’, by Gareth Williams, was substituted in its place. As the organisers, we would like to record our appreciation of those colleagues who shared, together with James Graham-Campbell, the responsibility for chairing the six sessions: Barry Ager, Marion Archibald and Leslie Webster, all of the British Museum. The Symposium was held at the Institute of Archaeology, on 26–27 May 2000, and we are grateful to then Director, Professor Peter Ucko, for having enabled this most stimulating event to have been held there, with the support of the Institute’s Research Committee. We would also like to express our gratitude to the British Academy for their generous support for the symposium through a Conference Grant. At the same time, it is a pleasure to thank both the then Director of the British Museum, Dr Robert Anderson, and the then Keeper of Coins and Medals, Dr Andrew Burnett, for having invited participants to a private view of the exhibition, on our first evening, and last, but very far from least, Dr Per Sörbom and the Royal Swedish Embassy for having sponsored the reception in the Gallery that helped make this a particularly memorable occasion. Ten of the 12 papers delivered during the symposium are published here in edited form, although we have not sought to impose absolute uniformity of style on our speakers, who in any case represent seven different nationalities, so as to retain something of a sense of the occasion, which included much lively but good-natured discussion. As both organisers and editors, we allowed ourselves, then and now, the last word(s) and so there are, indeed, 12 papers in this volume. After discussion with the editors, it was agreed that the important paper delivered by James Barrett, on ‘Wealth, Power and Trade in Scandinavian Scotland’, would be held over for expansion and publication by him elsewhere. Mark Blackburn’s paper on ‘The Monetary Economy in the Danelaw under Scandinavian Control, 880–927’ overlapped heavily with two other papers already promised for publication elsewhere, so with the agreement of the editors he has chosen instead to expand here on one aspect of his symposium paper, which he has developed into a new and broader paper on ‘Gold in England during the “Age of Silver” (Eighth–Eleventh Centuries)’. Similarly, the paper by Gareth Williams on ‘Coinage, Kingship and State Formation in the Late Viking Age’ does xiii



not appear in the form that it was presented at the symposium, but the theme of that paper is one of those explored in his editorial overview paper in this volume, which also includes a more detailed exploration of the themes presented in the ‘Paid in Burnt Silver’ exhibition. The paper by Marion Archibald, on ‘The Evidence of Pecking on Coins from the Cuerdale Hoard’, is published here only in a summary version, being work in progress for her contribution to the catalogue of The Cuerdale Hoard and related Viking-Age Gold and Silver from Britain and Ireland, in the British Museum, in preparation by James Graham-Campbell. As the co-editors of these proceedings, we are most grateful to the Publications Committee of the Institute of Archaeology for having readily agreed to take this volume under their wing, recommending it in due course for publication by UCL Press, and to both the Viking Society for Northern Research and the Dorothea Coke Memorial Fund, University of Cambridge, for grants to help defray its printing costs. Finally, we would both like to express our gratitude to all the contributors for their patience during the not inconsiderable delay of the appearance of their papers in print (especially after impatient demands on our part for submission in one or two cases!); in other words, we appreciate their understanding that we have both found ourselves, in our different ways, undergoing unexpected and demanding challenges during the past two or three years. 30 April 2005

James Graham-Campbell Emeritus Professor of Medieval Archaeology University College London Gareth Williams Curator of Early Medieval Coinage British Museum

This book in its current form (as described by us above) was intended for publication in late 2005. Unexpected difficulties during the production phase and changes in the ownership of UCL Press have led to a delay of a year in its publication. To avoid the possibility of any additional delay, the editorial decision was taken not to allow contributors the opportunity of further revising their texts, so that any failure to address issues raised in very recent publications is thus our own responsibility and not that of individual contributors. We would like to express again our gratitude to them for their patience and forbearance, and to Mitch Allen and colleagues at Left Coast Press Inc. for having seen the volume through its final stages into print. 8 November 2006

JG-C and GW

List of Abbreviations Antiq J ASC ASSAH AU


Antiquaries Journal Swanton, M J (ed and trans) (1990), An Anglo-Saxon Chronicle, Exeter Anglo-Saxon Studies in Archaeology and History Mac Airt, S and Mac Niocaill, G (eds and trans) (1983), The Annals of Ulster (to AD 1131), Part 1. Text and Translation, Dublin British Archaeological Reports Bulletin de la Commission départementale des Antiquités de la Seine-Inférieure/Seine-Maritime Bulletin du Cercle d’études numismatiques British Museum, London Grueber, H A and Keary, C F (1887–93), A Catalogue of English Coins in the British Museum. Anglo-Saxon Series, 2 vols, London Bibliothèque nationale de France, Paris British Numismatic Journal Bulletin de la Société française de numismatique Council for British Archaeology ‘Coin Hoards’ (published in NC) Corpus Nummorum Saeculorum IX–XI Blunt, C E, Stewart, B H I H and Lyon, C S S (1989), Coinage in Tenth-Century England, Oxford Duplessy, J (1985), Les trésors monétaires médiévaux et modernes découverts en France, vol I (751–1223), Paris Whitelock, D (ed and trans) (1955; 1968), English Historical Documents c.500–1042, London Economic History Review Fitzwilliam Museum, Cambridge ‘Gratia Dei Rex’ coinage of Charles the Bald Isle of Man Natural History and Antiquarian Society Proceedings Journal of the Royal Society of Antiquaries of Ireland Musée départemental des Antiquités de la Seine-Maritime, Rouen Morrison, K F and Grunthal, H (1967), Carolingian Coinage, New York Medieval Archaeology North Munster Antiquarian Journal





Numismatic Chronicle Spink’s Numismatic Circular Northern History National Museum of Ireland, Dublin Nordisk Numismatisk Unions Medlemsblad Nordisk Numismatisk Årsskrift Poey d’Avant, F (1858–62), Monnaies féodales de France, 3 vols, Paris Publications of the National Museum [of Denmark] Proceedings of the Royal Irish Academy Proceedings of the Society of Antiquaries of Scotland Revue belge de numismatique Revue numismatique Royal Numismatic Society Sylloge of Coins of the British Isles Transactions of the Royal Historical Society Victoria and Albert Museum, London


Regions Around the North Sea With a Monetised Economy in the Pre-Viking and Viking Ages D M Metcalf

The circular letter calling for papers for our symposium spoke of the transition from the gift/status economies of the pre-Viking Age to the monetised ‘nation states’ which were emerging by the end of the Viking Age. This view of the English Viking Age being, in my opinion, seriously out of touch with current research, I felt some doubts whether I could make any useful contribution to the occasion, but I wrote to ask Dr Gareth Williams whether a paper that might be thought controversial would be acceptable. He replied with exemplary serenity, ‘… by all means be as controversial as you like … the whole point of a symposium like this is for people to share their ideas with others, whether or not everyone agrees in the end. Certainly it often makes for a more interesting debate when people argue (constructively!) with each other’s opinions.’ Encouraged by this wise response, I should like to try to set out as clearly and accurately as possible the framework of the argument, as I see it, which says that in the pre-Viking-Age parts of England, and also other regions across the North Sea, had already developed an intensive money economy comparable with that which they enjoyed in the eleventh century. Between the two there was a monetary downturn in southern England, and similarly on the Continent. Pirenne could see the monetary upturn which followed, and mistook it for the beginnings of monetisation in north-western Europe. He was unable to see the preceding downturn, and therefore failed to do justice to the character of the economy in the early middle ages. Tenth-century numismatics, which tends to be difficult, has been intensively and very ably studied by a group of scholars of whom Christopher Blunt was the doyen. Tenth-century monetary history, which is another kettle of fish altogether, requires to be set into a wide context if the perspectives are to be accurately drawn. Between the eighth century and the eleventh, much of the economic infrastructure in southern England fell on hard times; wics and proto-towns were replaced by a new urban network. There is a considerable but scattered literature on monetary history, and the general historian may well be grateful for an overview. What he or she is most likely to need is a sense of how the arguments mesh together. It is well enough known that in eighth-century England the degree of monetisation varied regionally, being much greater in the south-east and the east, and thinning out westwards to such an extent that the south-western peninsula, the whole of Wales and the north-west, let alone the Pennine region and Cumbria, were relatively coinless. The east-west 1



gradient was evidently generated by trade across the English Channel and around the coasts of the North Sea (Metcalf 1989, 267–274). It is to some extent a perennial pattern in English monetary history throughout the early middle ages and indeed into the high middle ages. It does not correlate closely with regional prosperity, as we may judge from the Domesday Survey (Darby 1979, 200). Rich farming landscapes, for example in Herefordshire or Shropshire or in Somerset, have yielded extremely few stray finds of coins compared with Kent or East Anglia, a discrepancy by a factor of 10 or 20, whether it is viewed per capita or per acre. Similarly, the gradient does not correlate closely with political power. Mercia became a major player in the progressive consolidation of smaller tribal groupings, as the Tribal Hidage reveals (Hart 1977, 43–61), largely without the benefit of a money economy, and it was able eventually to gain the upper hand over the kingdom of the East Angles (Dumville 1989, 123–140) in spite of the greater monetary wealth of that kingdom. It is a similar story south of the Thames: the early primacy of east Kent (which surely reflects its locational advantages) is challenged by Wessex, in spite of the much later and weaker monetisation of the West Saxon kingdom. What emerges as a conclusion clearly enough from all this is that monetisation in the pre-Viking Age was significantly linked with long-distance or inter-regional trade. Wics were the nodal points of the trade-routes: the wics of the Wantsum Channel; Ipswich; Lundenwic, Eoforwic, Hamwic. And the English wics are part of a larger picture, of trade around the North Sea coastlands, with Dorestad as the dominant centre, and again with some regions strongly monetised and others, such as north-eastern France and Belgium, much less so (Op den Velde, De Boone and Pol 1984, 117–145). The relative intensity of monetary exchanges in the hinterland of each wic seems to have varied, for reasons that are not clear. One might have supposed that the prosperity of the wic arose from the trade passing through it. But it seems that monetary exchanges may to some extent, and to a variable extent, have been concentrated in the wics, rather than evenly spread through the hinterland area which supported the urban economy: for example, farmers brought their cattle to market and, although they may have taken their money home again, it was in the marketplace that there was the greatest risk of accidentally losing a coin. Thus, Ribe, in Jutland, was a wic full of money (we can all, I hope, agree on that), yet finds of sceattas anywhere east of Ribe are exceedingly few, and not for want of searching. The concentration of money in wics could also have been partly because that was where wealth was generated – not just the value-added wealth of traders who lived in the towns, but that produced by craft industries – comb-making at Ribe (Bendixen 1981, 200), glass-making at Hamwic (Andrews 1997, 216–218) and so on. Specialised production tends to require cash payments rather than barter. When all is said and done, however, the wics are coastal: they are essentially ports prospering through maritime trade, and unless they were eking out a livelihood by taking in each others’ washing, that implies that a wic was functionally linked to its hinterland. There is a problem of method, in that our impressions of the degree of monetisation tend to be based on the relative numbers of stray losses from this region and that. Although the chances of coins being accidentally lost are most likely to have been in direct proportion to the number of occasions on which they changed hands (rather than, for example, the size of the accumulated currency in a region), there are so many other unknown variables that the conclusion may be deemed unsafe, unless the margins of difference are very substantial. Different styles of husbandry, different settlement patterns, coins of different purchasing power – each may have had a significant effect on the



statistics. Again, ratios of stray finds may not equate with ratios of stray losses for the purposes of the exercise, because of differences in the intensity of metal-detector searches or of archaeological excavations, or even of techniques of excavation. The uncertainties can to some extent be discounted, but it is well to remember that comparisons between regions based simply on global numbers of stray finds should be treated with caution. The east-west gradient of which we spoke is so pronounced and so widely evidenced that its significance is unlikely to be compromised by methodological uncertainties. The paucity of detector finds of eighth-century coins in westerly regions, for example, can be confirmed if one takes into account the relatively large numbers of Roman and other coins found by detectorists in those same regions. It is worth underlining that this paucity in the west is relative. It does not mean that coinage was virtually unknown there. At Bidfordon-Avon in Warwickshire, for example, a detectorist who has searched the area year in year out for 14 years, finding one or two sceattas a year, had at the last count pushed the total to 29 sceattas from this one village. Having established the broad, regional perspectives, one can turn to the local topography, to marshal another argument, based on the contexts of the find-spots. There are now over 650 localities on record from which in total more than 2,360 single-finds of thrymsas and sceattas have been recorded in England (see the map, Figure 1.1) – far more than for eleventh-century coins (Metcalf 1998, 249–271). As regards the alleged gift/status economy of the pre-Viking Age, need I say more, really? Of the 650 localities, some include several separate sites within the boundaries of the same parish. The point to be seized is that the great majority of these 650+ localities (necessarily) are just villages, with nothing special about them. There is no visible correlation with royal manors, or with minster churches. There is no concentration of finds from royal centres such as Winchester or Tamworth (both well excavated). Coins have not been found in proximity to Offa’s Dyke. Sceattas were being lost predominantly in the course of ordinary village life, and, one must suppose, by ordinary people – everyday farming folk. The acid test, general historians might be inclined to suppose, of whether coins were in widespread use for everyday purposes would be if they were ordinarily handled by women. A capitulary of Charles the Bald issued at Quierzy in 861, while specifying penalties for refusing good coin, goes on to exempt women from this provision, on the grounds that it is well known that they are accustomed to bargain before agreeing a price and making a purchase (Grierson 1990, 64). Legal documents rarely record such mundane details, which have the charm of immediacy. But one may well ask which is the better evidence, the anecdotal from which it may or may not be safe to generalise, or the more comprehensive review of a range of archaeological evidence. If the latter is open, as a whole, to some residual doubt because of the unknown variables referred to above, there are, fortunately, some aspects of the evidence of stray finds which are much more secure. When coins from different mint-places, but of the same date and the same face value, mingled freely in circulation, they will arguably have been lost at random in respect of their mint-place, and the composition of the finds from a site should provide excellent evidence of the proportions, in the currency of that place, of the respective kinds and, by extension, of the relative scale of inter-regional transfers. In other words we are on much safer ground if we rely on within-sample variation. Thus, for example, the sharp contrast between the eighth-century finds from York (including its southern suburbium of Fishergate) and from Whitby (Rigold and Metcalf 1984, 245–268) respectively raises intriguing problems of interpretation. Another example: the relative



Figure 1.1 Distribution map of thrymsas and sceattas from England.



scarcity of the locally minted sceattas of Series H anywhere outside Hamwic itself has prompted the suggestion that the wic enjoyed a positive balance of payments through its trading (and craft) activities, and that this was reflected in unusually small monetary outflows (Metcalf 1988, 17–20). The other leg of the argument from the mingling of coins of different mint-places depends on estimates of the quantities of coins minted, as judged from the numbers of dies employed. This argument is stronger when applied to active mints, which were using up reverse dies at such a rate that one need not worry too much about the possibility of serious under-use of dies. Taken by itself, the argument from quantities of coins minted might be depreciated by saying that most of the coins could have served only or mainly as a store of value, and that they should not be assumed to have supported an exchange economy. That is something which one cannot altogether discount, except by saying that if coins were in such widespread use, most of the people who were using them were necessarily ordinary people who would not have had large-scale surplus wealth. But when the two arguments are combined, and set into the context of regional currencies, one obtains very specific and powerful evidence of the scale of inter-regional monetary transfers. These are the twin pillars on which the case rests: not the seven pillars of wisdom, nor the five pillars of religion, but the twin pillars of monetisation – quantity in combination with velocity. On the conventional estimate that 100 dies could be used to strike something like a million coins, it is clear beyond reasonable doubt that millions of coins were involved in inter-regional transfers within England, and that the scale of the currency accumulated within Hamwic, for example, was of the order of some two or three million coins. We now know, through the patient researches of Derek Chick, that the coinage of King Offa was struck from an estimated 1,000 to 1,100 dies, with little room for statistical error, pointing us towards an output figure of the order of 10 million coins. For the preceding sceatta coinages, evidence of the same high quality is not yet available, except for Northumbria, but it has been conjectured that they were produced, admittedly over a period of nearly 100 years, from a grand total of roughly 8,000 dies, therefore perhaps 80 million coins in all. Of course these totals are dependent on the assumption of 10,000 as the average output of a die, as is the corresponding estimate for Offa, and they may accordingly be somewhat wide of the mark, but not so wide as to threaten the general conclusion that the pre-Viking-Age economy was strongly monetised in some regions, and relatively much less so in others, with maritime trade in the English Channel and on the North Sea coasts as the driving mechanism. Of all the sceattas found in England, a good 14 per cent are of the ubiquitous ‘porcupine’ type, minted in the Rhine mouths area, largely at Dorestad (Metcalf 1993–94, 174). Dorestad, with its miles of quays (Van Es and Verwers 1980), was the Rotterdam of the day, a major commercial port serving as intermediary between the trade of the Rhine valley, and in the other direction the trade passing through the wics of the North Sea coastlands, all around from Northumbria to Jutland. A seventh of all the money in England had originated in the Rhine mouths area. We can safely say that millions of porcupines accumulated in England. And then there are the plentiful Frisian runics, Series D, which also reached England in quantity, and the Wodan/monster sceattas of Series X, minted in Ribe. In light of our discussions at the symposium (see Malmer, Chapter 2, in this volume), it is necessary to add that the arguments for the attribution to Ribe (Metcalf 1984, 159–164) have been disputed, but have recently been restated without retraction



(Metcalf 1996, 403–409). Together the foreign sceattas found in England represent a massive balance-of-payments surplus, generated (how else?) by exports. They did not reach England through gift-exchange, or as status-symbols (Hodges 1982; Hodges and Cherry 1983, 131–183). The scale of the exchanges is such that most of those exports were almost certainly farm products of one sort or another. It would seem that regions such as East Anglia were producing for an export market to an appreciable extent. And on the Continent too, there seem to have been sharp contrasts between the heavily monetised and the lightly monetised regions – the Rhine mouths area, for example, in contrast with the coastlands further west in Belgium and Picardy (Delmaire, Gricourt and Leclercq 1990, 179–185). Detectorists’ single-finds accumulate steadily in the Netherlands: the KPK maintains a register of them, and coin dealers regularly sell provenanced sceattas. I have concentrated on the logical framework of the argument, with the minimum of details by way of example, and those drawn mainly from the first half of the eighth century, because that is where the case stands or falls. It offers a view of the pre-Viking Age which is sharply at variance with the general ideas of historians of an earlier generation. If this argument had been available to Pirenne, he could not have formulated his famous thesis in the way he did. No blame attaches to him: in the country of the blind, the one-eyed man is king. But there is no occasion to be one-eyed about the evidence available in 2000. Pirenne could not know what we know now, nor could he have shared the ideas that govern our thinking, for several reasons. First, the number of eighth-century coins above-ground has increased many times over, primarily by the activities of metaldetectorists, allowing us to construct die-catalogues large enough to serve as a basis for statistically reliable die-estimation. When there were only half-a-dozen specimens of a variety in existence, the absence of die-duplicates among them was neither here nor there, and there was no immediate reason to suspect the large scale of the issues. Second, there has been a 30-year-long debate among numismatists about the appropriate methods and practical problems of statistical analysis (Carcassonne and Hackens 1981), with a growing awareness of the historical implications of the numismatic facts, or better one could say, the facts about monetary circulation and monetary history that are being created through the recording of provenanced single-finds. Pirenne could allow himself to imagine that eighth-century coins had originally been so few that they made hardly any impact on an economy in north-western Europe characterised by non-monetary exchanges, local selfsufficiency, a restricted long-distance trade in luxury goods which reinforced the status of the ruling elite and so on. We know differently. Even though numismatists will no doubt continue to wrangle about how large is large (Buttrey 1993, 335–351; Buttrey 1994, 341–352; De Callataÿ 1995, 289–311), the sheer scale of the currency, in combination with the relative scale of transfers between regions – some regions, but not all – imply that there was a monetised economy based on trade. The fact that our surviving documentary sources are virtually silent about it in no way casts doubt; it merely illustrates why an interdisciplinary approach is fruitful in early medieval studies. The contexts of the findspots show that, in heavily monetised and in lightly monetised regions alike, this economy reached out to countless ordinary villages. In Denmark we may be talking about sceattas from high-status sites, but not in England. Where Pirenne detected the first beginnings of a widespread money economy in the years following the dislocation of the Viking raids, an upturn as he saw it somehow triggered by that dislocation, we can say categorically that that was not the beginning, but rather the recovery after a downturn caused by the Viking raids, from an earlier level of mint-output and monetary circulation that was not to be reached again much before the eleventh century. Pirenne simply did not reckon with that



earlier phase of monetisation. How deep and how dark the tenth-century recession was in southern England (and in other regions too) is a difficult subject for investigation, because our data-base of single-finds of coins is so much smaller and so problematic compared with either the eighth century or the eleventh century. That is in spite of die-estimation which indicates a high level of mint-output (Metcalf 1986, 133–157). But the story is certainly one of great wealth in certain regions – portable wealth which was in itself an inducement to Viking raiders – followed by dislocation and monetary recession, followed by recovery. The numismatic evidence which has been used to create the perspective necessarily sums up a diversity of local detail, and reveals the dominant theme. There may have been other local themes, which are submerged in the general statistics. If evidence of them shows up here or there, one should not be surprised, nor should one allow it to distort the over-all view. The tenth century – the Danelaw century – witnessed some curious reversals of polarity, of which none is more striking than the dramatic rise to prominence of the mint of Chester and eventually, at about the date of Eadgar’s reform of the coinage, its equally dramatic decline. Obviously this is intimately connected with trade in the Irish Sea coastlands and with the fortunes of the Norse kingdom of Dublin. There are numerous hoards of English coins found in Ireland and Man (Metcalf 1992, 89–106) and in the highland region of Scotland (Metcalf 1995, 16–25), which reflect a new situation. Until then, there had been little enough coinage in those regions. Interestingly, these hoards are by no means as heavily dominated by the mint of Chester as one might have expected. Money from eastern England was also quite prominent in the outflows. The bonds of society were very different in the highland zone, and so were the constraints on gaining a livelihood, and the ordinary levels of material well-being. There is more of a case for speaking of a status silver economy enjoyed by warrior leaders. We should be careful not to get our framework of argument back to front: we cannot by studying coin finds, detect the transactions, unilateral or otherwise, that coins were not used for. Gift-exchange, and so forth, no doubt existed alongside any monetary economy, and was relatively more prominent than in England, but numismatics cannot help us to explore it. In the highland zone, the hoards were not accompanied, as they were in England, by a thick carpet of stray losses, and one may justifiably infer that the Highlands and Islands were not a monetised region, in spite of the hoards. It is not until the time of King David I that we see the English monetary patterns spreading northwards to encompass the central lowlands of Scotland. Jean-Paul Devroey has sketched the characteristics of the ninth-century political economy, as revealed by polyptiques (Devroey 1985, 475–488). He points to traditional and very conservative agricultural strategies as the background to ideals of peaceful stability, regular production, foresight in the storage of surpluses, and political paternalism. The estate economy had few outlets: revenues derived from agricultural surpluses were often not applied to any economic purpose, but were used up in almsgiving, in the building of churches or in the purchase of precious objects such as books or vestments. Alongside this rather static survival-economy there are signs, which (Doevroey says) we do not understand at all clearly, of a money economy which, in tension with demographic trends, held out the possibility of progressive social change and a spiral of economic development. The signs, one may comment, are writ increasingly large. To speak about the two economies being ‘alongside’ each other merely affirms the obvious, namely that they could coexist, and leaves open the question of their relationship. Thus, York may



have had the character of a trading emporium which was socially very different from the broad acres over which the Northumbrian kings ruled, an island where money was in daily use, in a sea of subsistence and barter. Such a view is challenged by the stray finds of coins from the countryside which offer the crucial obstacle to traditional historical perceptions. Their evidence gains in strength by being submitted to systematic comparative analysis. In the tenth century, however, the numbers of stray finds from southern England decline quite sharply, to such an extent that the patterns of inter-regional monetary circulation in the southern shires are difficult to determine, simply because we have not accumulated enough data. If we compare the half-century from the accession of Athelstan to Eadgar’s coinage reform (925–c 973) with the immediately following sixty years to the death of Cnut (c 973–1035), the numbers of single-finds recorded in the British Numismatic Society’s annual Coin Register over the past five years is 22 before 973 against 88 after 973, a fourfold difference (cf. 293 thrymsas and sceattas!). The numbers are large enough to be statistically significant; we have to try to believe them. Of the 22, no fewer than 14 are finds from East Anglia, and a few of those, from the Thetford by-pass, may in fact be from an undisclosed small hoard. There is a paradox here, in that although there are so few single-finds from 925–c 973, Athelstan had a comprehensive network of some 30 mints scattered over Wessex and Mercia (but only two in East Anglia). The single-finds are the crux of the evidence for a monetary downturn or retreat in southern England in the tenth century. Perhaps the straightforward interpretation is the sensible one, namely that maritime trade from the Channel ports and the North Sea ports was severely dislocated by the unacceptable levels of risk, or that England’s trading partners were in economic disarray, and that the fall-off in trade fed back into the hinterland in the English countryside. At approximately the same date as that of Eadgar’s reform, c 973, the polarities swung back again, the Irish Sea trade ceased to bring money into Chester, monetary outflows to Scandinavia began to grow strongly, the southern English mints likewise resumed their activity, stray losses quickly become plentiful throughout southern England, and their analysis shows that money was moving freely across shire boundaries, and finally it seems that inflows of silver from across the Channel resumed (although foreign coins were not permitted to circulate in England, and had to be recycled) (Metcalf 1998, 85–89). In a word, the perennial patterns of English monetary history, based on trade with the continent, were restored, in a triangular situation that now also involved Scandinavia. Eadgar’s reform was more than just a renewal of the coinage. Its political dimension, in the early days, is illustrated by the importance of Winchester in implementing the reform (Metcalf 1998, 105), and the tardiness of London in participating (Petersson 1990, 295). If one were to compare the distribution map of stray finds of coins in England south of the River Trent, from the accession of Eadgar in the south in 959 to his coinage reform, with a similar map for the sexennium c 973–79, the contrast would focus attention on changes in monetary circulation which we do much better to understand in political than in social or economic terms. Again, a study of the ranking order of the reform mints and their respective shares of the national currency, allied with a survey of the mint-places of the stray finds from the territory of the Five Boroughs, quickly disposes of the old idea that the Danish settlement was a major stimulus to monetisation. If we turn back now, briefly, to the period from c 860 onwards – the period of the great heathen army, the creation of the Danelaw, the fight-back by Wessex and the recovery of East Anglia, Lincolnshire and eventually York – the monetary historian’s task of creating



an accurate and broad-based regional perspective is made more difficult by the politically fragmented character of the coinages. The mapping of stray finds and the analysis of within-sample variation are as usual the best way, indeed the only way, to describe the patterns of monetary circulation in southern and eastern England. But stray finds are few, and the survival-rate per die can be exceptionally low. If one focuses just on the St Edmund Memorial coinage, or just on the Cross and Lozenge coinage, for example (Blackburn and Keynes 1998, 125–150), the panorama is not wide enough to reveal the regional contrasts which should control our interpretation. Thus, the first question for the monetary historian to ask, probably, is whether the east-west gradient grew steeper in this phase. The Scandinavian settlers of the Danelaw adopted English ways with money quickly enough in those regions which had previously been monetised. And of course, it is not as if they came from coinless regions: Islamic dirhams were in use throughout Scandinavia in great quantities. At York in the closing years of the ninth century, the Danish rulers reintroduced a good silver coinage to replace the discredited stycas. The Cuerdale types, struck from some 700–750 reverse dies (Lyon and Stewart 1961, 106), clearly had an international dimension. If there was a significant change in English monetary history as a result of the Danish settlements, it would seem to be that the great Anglo-Scandinavian river-ports of England’s North Sea coastlands, Stamford, Lincoln and especially York, prospered at the expense of London and the Channel ports, a shift in the balance which was to persist up to the time of William I’s harrying of the North. In conclusion, one may sketch a context for the new quality in monetary affairs, which is apparent in England and in Germany from the late tenth century onwards, by pointing to new sources of silver in the Harz mountains and elsewhere, the benefits of which seem to have been quite quickly dispersed over wide areas (Metcalf and Northover 1986, 35–63). Second, royal power was gradually becoming relatively stronger vis-a-vis political power at a regional level. Europe was becoming richer, as in our own day, and the enlargement of the cultural community by the christianisation of Scandinavia became an option. Viking as a way of life became marginalised because this new tide of wealth made it look increasingly a poor option. There were better and less destructive ways to make one’s fortune in the prospering towns of the eleventh century. Just as the euro is merely the most visible aspect today of a political strategy for economic and monetary union, so coinage in the later Viking Age should be seen in a more complex political and regional context than Pirenne envisaged. Perhaps southern England in the tenth century had been susceptible to a monetary downturn specifically because the earlier prosperity of the country had been integrally dependent on foreign trade.

References Andrews, P (ed) (1997), Excavations at Hamwic, Vol 2. Excavations at Six Dials (CBA Research Report no 109), ‘The glass’, 216–218 Bendixen, K (1981), ‘Sceattas and other coin finds’, in M Bencard (ed), Ribe Excavations, 1970–1976, Vol 1, Ribe Museum, Esbjerg Blackburn, M A S and Keynes, S D (1998), ‘A corpus of the Cross-and-Lozenge and related coinages of Alfred, Ceolwulf II, and Archbishop Æthelred’, in M A S Blackburn and D N Dumville (eds), Kings, Currencies, and Alliances. History and Coinage of Southern England in the Ninth Century, The Boydell Press, Woodbridge, 125–150 Buttrey, T V (1993), ‘Calculating ancient coin production: facts and fantasies’, NC 153, 335–351



Buttrey, T V (1994), ‘Calculating ancient coin production, II: why it cannot be done’, NC 154, 341–352 Carcassonne, C and Hackens, T (eds) (1981), Statistics and Numismatics (=Pact 5), Strasbourg Darby, H C (1979), Domesday England, 2nd edn, Cambridge University Press, Cambridge De Callataÿ, F (1995), ‘Calculating ancient coin production: seeking a balance’, NC 155, 289–311 Delmaire, R, Gricourt, D and Leclercq, P (1990), ‘Chronique numismatique (IX)’, Revue du Nord 72, no 286, 179–185 Devroey, J-P (1985), ‘Réflexions sur l’économie des premiers temps carolingiens (768–877): grands domaines et action politique entre Seine et Rhin’, Francia 13, 475–488 Dumville, D N (1989), ‘Essex, Middle Anglia, and the expansion of Mercia in the south-east midlands’, in S Bassett (ed), The Origins of Anglo-Saxon Kingdoms, Leicester University Press, Leicester, 123–140 Grierson, P (1990), ‘The “Gratia Dei Rex” coinage of Charles the Bald’, in M T Gibson and J L Nelson (eds), Charles the Bald, Court and Kingdom, 2nd edn, Variorum, 52–64 Hart, C (1977), ‘The kingdom of Mercia’, in A Dornier (ed), Mercian Studies, Leicester University Press, Leicester, 43–61 Hodges, R (1982), Dark Age Economics: the Origins of Towns and Trade, A.D. 600–1000, London Hodges, R and Cherry, J F (1983), ‘Cost-control and coinage: an archaeological approach to economic change in Anglo-Saxon England’, Research in Economic Anthropology 5, 131–183 Lyon, C S S and Stewart, B H I H (1961), ‘The Northumbrian Viking coins in the Cuerdale hoard’, in R H M Dolley (ed), Anglo-Saxon Coins. Studies Presented to F.M. Stenton, Methuen, London, 96–121 Metcalf, D M (1984), ‘A note on sceattas as a measure of international trade, and on the earliest Danish coinage’, in D Hill and D M Metcalf (eds), Sceattas in England and on the Continent. The Seventh Oxford Symposium on Coinage and Monetary History (BAR, British Series 128), Oxford, 159–164 Metcalf, D M (1986), ‘The monetary history of England in the tenth century viewed in the perspective of the eleventh century’, in M A S Blackburn (ed), Anglo-Saxon Monetary History. Essays in Memory of Michael Dolley, Leicester University Press, Leicester, 133–157 Metcalf, D M (1987), ‘Introduction’, in D M Metcalf (ed), Coinage in Ninth-Century Northumbria. The Tenth Oxford Symposium on Coinage and Monetary History (BAR, British Series 180), Oxford, 1–10 Metcalf, D M (1988), ‘The coins’, in P Andrews (ed), Southampton Finds, vol 1. The Coins and Pottery from Hamwic, Southampton City Museums, 17–59 Metcalf, D M (1989), ‘The availability and uses of gold coinage in England, c.580–c.670: Kentish primacy reconsidered’, in U Ehrensvärd, N-U Fornander, K Jonsson, U Nordlind, F Olrog and I Wiséhn (eds), Festskrift til Lars O. Lagerqvist, Stockholm, 267–274 Metcalf, D M (1992), ‘The monetary economy of the Irish Sea province’, in J Graham-Campbell (ed), Viking Treasure from the North-West: the Cuerdale Hoard in its Context, National Museums and Galleries on Merseyside, Occasional Papers, Liverpool Museum, no. 5, Liverpool, 89–106 Metcalf, D M (1993–94), Thrymsas and Sceattas in the Ashmolean Museum, Oxford (RNS Special Publication no. 27), 3 vols, London Metcalf, D M (1994), ‘The beginnings of coinage in the North Sea coastlands: a Pirenne-like hypothesis’, in B Ambrosiani and H Clarke (eds), Developments around the Baltic and the North Sea in the Viking Age (The Twelfth Viking Congress), Birka Studies, no. 3, Riksantikvarieämbetet and Statens Historiska Museer, Stockholm, 196–214 Metcalf, D M (1995), ‘The monetary significance of Scottish Viking-age coin hoards, with a short commentary’, in J Graham-Campbell, The Viking-Age Gold and Silver of Scotland (AD 850–1100), National Museums of Scotland, Edinburgh, 16–25 Metcalf, D M (1996), ‘Viking-Age numismatics, 2: Coinage in the Northern Lands in Merovingian and Carolingian times’, NC 156, 399–428



Metcalf, D M (1998), An Atlas of Anglo-Saxon and Norman Coin Finds (RNS Special Publication no. 32), London Metcalf, D M and Northover, J P (1986), ‘Interpreting the alloy of the later Anglo-Saxon coinage’, BNJ 56, 35–63 Op den Velde, W, De Boone, W J and Pol, A (1984), ‘A survey of sceatta finds from the Low Countries’, in D Hill and D M Metcalf (eds), Sceattas in England and on the Continent (BAR, British Series 128), Oxford, 117–145 Op den Velde, W and Klaassen, C J F (2000), The Sceattas and Merovingian Deniers from Domburg Petersson, H B A (1990), ‘Coins and weights. Late Anglo-Saxon pennies and mints c.973–1066’, in K Jonsson (ed), Studies in Late Anglo-Saxon Coinage. In memory of Bror Emil Hildebrand (=Numismatiska Meddelanden 35), Stockholm, 207–433 Rigold, S E and Metcalf, D M (1984), ‘A revised check-list of English finds of sceattas’, in D Hill and D M Metcalf (eds), Sceattas in England and on the Continent (BAR, British Series 128), Oxford, 245–268 (a further revision is in preparation) Van Es, W A and Verwers, W J H (1980), Excavations at Dorestad, Vol 1. The Harbour: Hoogstraat I, The Hague/Amersfoort (and subsequent volumes)


South Scandinavian Coinage in the Ninth Century Brita Malmer

The development of silver economy in Viking-Age Scandinavia includes, of course, the development of native coinage. Before the 990s native coinage only existed in the southwestern part of Scandinavia. This native coinage is silent – the coins carry a variety of pictures but no legends. Silent coins require extra care in interpretation, the first difficulty being to establish a reliable hypothesis about the place at which, and the date when, a silent coinage was issued. The second aim is to produce hypotheses about perspectives on ‘monetary development and state formation’ – a specification which, according to the invitation sent out in 1999, to the ‘Silver Economy’ symposium, was a particularly welcome theme. This short chapter concentrates on the first problem, but it also contains an introductory discussion on the second.

METHODOLOGICAL INTRODUCTION Methodologically, the field of numismatics has much better conditions than most subjects in the humanities. This is derived from the presence of a considerable element of science, given that weight, fineness, dies and secondary treatment (such as pecks) are all characteristics which it is possible to count and measure, and thus to work up statistically. Consequently, the field of numismatics is not dependent on hypotheses to such an extent as most other subjects in humanities. However, anonymous or ‘silent’ coins imply special problems. Given that silent coins provide no information as to where they were minted, there is only one way to know this for certain and that is to discover the mint itself by archaeological excavation, something which happens extremely rarely. For the ninth-century Scandinavian coinage, no mint has yet been found by excavation. This means that it is only possible to point to the most likely area of minting in Scandinavia, and not to a specific town, as the source for any particular coin type. Should one, nevertheless, wish to link a particular coin type to a specific town, then it is necessary to emphasise that this is just a hypothesis – and nothing more.

MINTING IN THE EIGHTH CENTURY According to Michael Metcalf, Scandinavian coinage did not start in the ninth century, but early in the eighth, at the town of Ribe. Most sceattas are silent, and Metcalf moves one 13


14 (i)







b (iv)




Figure 2.1 (i–iv) Selected coins, scale c 3/2 (© Gabriel Hildebrand, Royal Coin Cabinet, Stockholm). (i) Wodan/Monster sceatta, Frisia, from c 715; (ii) Charles the Great, silver denier, Dorestad, from 771–93/94; (iii) KG 5, from the period of Louis the Pious, probably the 820s; (iv) KG 3, from the period of Louis the Pious, probably the 820s.

of the most numerous groups of the silent sceattas, the Wodan/Monster type, from a rather indistinct region of minting, namely Frisia, to a specific Danish town (Metcalf 1984; Figure 2.1(i)). No archaeological remains of any such mint have yet been found anywhere. Nevertheless, in his magnificent work, Thrymsas and Sceattas in the Ashmolean Museum, Metcalf puts them under the distinct heading ‘Danish Sceattas’. According to Metcalf, mass production of Wodan/Monster sceattas commenced at Ribe at the beginning of the eighth century. This earliest Scandinavian coinage was not at all hesitant, experimental or incomplete; on the contrary, from the very beginning, it was supposedly a regulated coinage, combined with the obligation to change all foreign coins to Danish sceattas. Such would indicate a rather sophisticated kind of coinage, like that we know to have developed in Denmark and Norway more than 300 years later, after a fairly complicated initial stage. Metcalf has to support his interpretation of the evidence by means of a number of suppositions that it is impossible to prove. One example is his assertion that the 1976 Föhr hoard, containing many Porcupines but no Wodan/Monsters (Hatz 2001), is an exception, rather than representative of finds from the ‘old’ Danish territory. Metcalf’s theory of a monetised economy at Ribe in the early eighth century was stimulated by the rich archaeological finds of Wodan/Monster sceattas in its original marketplace. These finds are extremely interesting from the point-of-view of urbanisation. They show a limited monetary economy in the market of a northern European town as early as the first half of the eighth century, but with regards to their interpretation in the history of Scandinavian coinage, there is a more natural, more simple – if still hypothetical – explanation. At the beginning of the eighth century trade flourished in Frisia. The excavations at Ribe show a considerable import of Frisian/Frankish utility goods, particularly pottery and glass. The finds of Wodan/Monster sceattas at Ribe should



instead be regarded as part of this influx of Frisian/Frankish imports. The foreign merchants brought their own coins with them to facilitate minor purchases during the market season (see also Williams, Chapter 11, in this volume). What is really needed to solve the problem of the location of the mint of the silent Wodan/Monster sceattas is, of course, going to be a detailed monograph with a careful description of every coin available, including both coins from hoards and stray finds. Unfortunately, such a monograph does not yet exist.

SURVEY OF SCANDINAVIAN COINAGE IN THE NINTH CENTURY In his presidential address, printed in The Numismatic Chronicle (1996), Professor Metcalf presented a number of further ideas about Scandinavian coinage, now concerning the ninth century. According to the present writer, the only way to discuss difficult numismatic problems in a constructive way, in order to get useful results, is (as just mentioned) to gather together all the available coins, and to note all the relevant characteristics of every single one. Since 1966, when Nordiska mynt före år 1000 (Malmer 1966) was published, the number of Scandinavian coins from the ninth century has increased from only 70 to almost twice as many (139). As a reply to Metcalf’s proposals and, hopefully to facilitate future research, my forthcoming study of the Scandinavian Pointed Helmet and Short Cross imitations (Malmer, forthcoming) will include an extensive introduction, namely a catalogue of all the Scandinavian ninth-century coins known today, with enlarged photographs of every specimen, and detailed descriptions.

Combination-Groups Die-links are rare in the early Scandinavian coinage and, moreover, dies are often difficult to distinguish. Therefore, in Nordiska mynt före år 1000, the coins were divided up into so-called Combination-Groups according to the combinations of obverse-pictures and reverse-pictures, as a substitute for links and chains. The word ‘Combination-Group’ was abbreviated to KG (after the initial Swedish spelling with K). Altogether 12 CombinationGroups were established, of which KG 7–12 belong mainly to the tenth century. In this chapter only KG 1–6 are relevant (Figure 2.2). KG 1 is the official Carolingian coinage of Dorestad before the reform of 793/94. KG 2, which is not quite homogeneous, consists principally of early imitations of Krinkberg type, minted in the 790s. The Scandinavian coinage starts with KG 3. There is a very distinct gap between KG 5 and 6 (Figure 2.2, bottom right). KG 3–5 are, on the other hand, more or less mixed together. The gap between KG 2 and 3 is not only chronological, but also technical. In KG 3 the coins are larger, thinner and lighter than in KG 2, and the Carolingian letters are mixed with Scandinavian symbols. The mean weight of the Carolingian pre-reform coinage is usually given as 1.3 g. For Dorestad alone, the figure is as low as 1.12 g (Malmer 1966, 80, Tab 1). Twenty undamaged coins of KG 2 have exactly the same mean weight as official Dorestad deniers, and there can be no doubt that KG 2 was minted according to the pre-reform weight-standard. This is especially clear when one compares the mean weight of the subsequent groups, which is as low as 0.77 g. In 1966, KG 2 was described as a Northern transitional group between KG 1 and KG 3, but still minted inside the Carolingian empire. A hoard from Hoogstrat, Dorestad, found in 1972, seems to confirm



Figure 2.2 Diagram of Combination-Groups (KG) 1–6, percentages (after Malmer 1966, plate 42).

this (van Gelder 1980). In any case, the southern Scandinavian coinage of the ninth century began with KG 3, not with KG 2. It is necessary to underline this because Metcalf writes of KG ‘2–6’ jointly, in spite of the obvious differences between KG 2 on the one hand, and KG 3–6 on the other (Metcalf 1996, 410).

Design KG 3–6 are the four true Scandinavian Combination-Groups of the ninth century. KG 3 consists today of 57 coins, all of them with a Carolus obverse (Figure 2.1 (iv-a); Figure 2.3, 1). A clear majority of them have a Dorestad reverse, as on the prototype (Figure 2.1 (iv-b); Figure 2.3,2), but no fewer than 23 coins have irregular reverses (Figure 2.3, 3–7). These include ships of different types, a house with slanting Scandinavian buttresses, a man and even a ‘monster’, similar to the reverse animal of the Wodan/Monster sceattas (Figure 2.1 (i-b)). KG 4 and 5 (Figure 2.3, 8–16) are rather inter-connected. Out of 53 coins, 38 have the regular combination of Wodan/Monster, with decorated border (Figure 2.3, 8–9), or with plain border (Figure 2.3, 10–11), which is a somewhat later characteristic. Two coins have a deer instead of a monster as the reverse design (Figure 2.3, 12). No fewer than 13 coins have no obverse design at all. Reverses are combined with reverses, and the pictures



Figure 2.3 Combination-Groups (KG) 3–6, scale c 85% (© B Malmer).

involved are two types of ship (the naturalistic type with a fish, and the simplified type with bows), as well as deer and cockerels (Figure 2.3, 13–16). KG 6 is the last and smallest Combination-Group of the ninth century (Figure 2.3, 17–20): there are 22 coins, all with the regular Wodan/Monster combination. Only plain, undecorated borders have been found in KG 6. The coils of the snake seen in KG 5 are here transformed into a circle. KG 5 or 6: seven corroded or fragmentary ninth-century Wodan/Monster-imitations.



Pictures or designs present a mixture of Carolingian, Frisian, English and, not least, Scandinavian elements, generally of a very high artistic quality. First to be mentioned among Carolingian models is, of course, Charlemagne’s Dorestad coinage from before 793/94 (Figure 2.1 (ii)). The picture of a ship, produced by the mints of Dorestad and Quentovic, and dated to the period 812–19, is obviously not a model for the Scandinavian ships, but merely a source of inspiration. The same is true for the Scandinavian house, which seems to have been inspired by Louis the Pious’ CHRISTIANA RELIGIO denier, which on one side has a cross, and on the other the picture of a temple or a church. The CHRISTIANA RELIGIO denier was struck throughout the whole Carolingian Empire and treated as the imperial type par excellence. The Scandinavian die-cutters seem not to have been interested in crosses, but the ship of Dorestad and Quentovic, as well as the imperial temple, inspired them to some of their very best pictures, full of details (Figures 2.3, 3, 5 and 13). Other models, or possible models, are Northumbrian animals, not quite unlike our deer, and cockerels, from English sceattas. The picture of a man (Figure 2.3, 6) has a very complicated origin, mostly Scandinavian, but a standing or walking figure occurs both on English and continental coins (Pedersen 2000). Finally, there are the sceattas of Wodan/ Monster-type (Figure 2.1 (i)). It is instructive to compare the original sceatta from the first half of the eighth century and the imitations from 100 years later (Figure 2.1 (i) and 2.1 (iii)). On the later coins, Wodan has acquired a broad border with masks and serpents, and the little monster has grown considerably and acquired a checked pattern on its body, and a serpent below it.

Chronology The prototypes of the designs of the Scandinavian ninth-century coinage cover a period of more than 100 years, and give a terminus post quem to the period of Louis the Pious. There are over 70 finds, but a majority of them cannot be dated more exactly, as in the case of many grave-finds. For example, KG 3 is represented in 30 finds, but only six of them are possible to date, and so on. Three finds with KG 3 and one find with KG 4–5 have a terminus post quem of 822, based on CHRISTIANA RELIGIO deniers of Louis the Pious’ Class 3 which, according to Grierson, date from 822–40. KG 3 appears only in finds from the ninth century, whereas KG 4–5 appear in finds from the ninth and the first half of the tenth century. Especially useful for the dating of KG 4–5 is the 1967 Russian Kislaya hoard, with nearly 700 Islamic and Khazarian coins, and a terminus post quem of 837, which fits extremely well with the CHRISTIANA RELIGIO deniers (Noonan 1981, 97, find no. 34). In recent years, the question of a ninth-century mint at Ribe has been connected to the idea of there having been a mint at Ribe in the early eighth century. The Wodan/Monster motif of KG 5–6 is believed to have been inspired not by Frisian sceattas but by native eighth-century Danish sceattas, in agreement with Metcalf. Consequently, there has been a trend to move back the ninth-century minting of Wodan/Monster imitations to the very beginning of the century, so as to increase the probability of there having been an eighthcentury mint at Ribe producing Wodan/Monster sceattas, followed by minting of KG 5 in the early ninth century at the same town. KG 6 belongs to the ninth century, in spite of the fact that it does not appear in any find from before the middle of the tenth century. The main reason for this early date is that



KG 6 has about the same weight as KG 3–5 (mean weight 0.77 g). When, after a break, the southern Scandinavian coinage started again with the following group, KG 7, then the mean weight had dropped by about one-third, down to 0.55 g. Another reason is that KG 7 is already present in the Cuerdale hoard (Archibald 1985). The terminus post quem for Cuerdale is c 905. The distance in time between KG 6 and KG 7, with their different weights and designs, cannot be very small. It seems that the proposal made in 1966, that KG 6 was minted around the middle of the ninth century, remains quite likely, even today (Malmer 1966, 219, 247, cf. 210 et seq).

Weight For comparison, the weights of 65 whole obols of Louis the Pious, following Morrison’s 1967 catalogue (up to and including coin no. 441), have been gathered (Figure 2.4). The mean weight is 0.74 g. Because of corrosion, cutting, and so on, only 39 Scandinavian KG 3–6 coins, out of a total of 139, are suitable for a comparative weight study (Malmer 2002b, 271–273). Their mean is 0.77 g, nearly the same as that of the Carolingian obols. Both Carolingian obols and Scandinavian coins peak in the same interval, but the difference between the lightest and the heaviest coin is naturally larger for the new, northern coinage. A rather obvious conclusion is that the KG 3–6 coinage of the ninth century was intentionally minted to the weight standard of the Carolingian half deniers or

Figure 2.4 Weights of Carolingian obols and of KG 3–6 (© B Malmer).



obols, a denomination which seems to have been introduced by Louis the Pious (Grierson and Blackburn 1986, 213). Scandinavian ninth-century coins were the same diameter as the contemporary Carolingian denier, c 19.5 mm, but only about half their weight (Malmer 1966, 143, Tab 18).

Dies, links and output Because of corrosion and cutting, there are many dies which cannot be identified. Nevertheless, it is possible to state that the Scandinavian ninth-century coinage presents an unusually large number of dies in relation to the number of coins for which dies can be identified. Die-links are rather unusual and concentrated in KG 4–5. This confirms what the Combination-Groups have already told us: that KG 3 and especially KG 6 represent a mint in rather good order, while the minting conditions of parts of KG 4 and 5 were in a certain state of confusion. There are different formulas to calculate the number of minted coins, but most methods suffer from a lack of statistical validity (cf. Malmer 2002a, 126). We do not know whether or not a Scandinavian die of KG 3 or 4 would have had the same average productivity as, for instance, a Danish die of the eleventh century. In this connection, it is enough to state that the Scandinavian coinage of the ninth century seems to have been of considerable proportions.

Geographical distribution Geographical distribution has often been used as a guide to the mint location of anonymous coins. The trouble with this is that different kinds of finds are not immediately comparable. When searching for the most probable location for a mint, it is impossible to give finds from graves, where all coins have been adapted for use in necklaces, the same weight of information as coins which have been lost unintentionally, during their circulation as coins. KG 3, the largest group, appears in 30 localities as just mentioned. When grave-finds are excluded there is an obvious concentration to south-west Scandinavia, and especially Hedeby. Coin finds from this area are discussed by Ralf Wiechmann elsewhere in this volume (Chapter 3). Grave-finds on the other hand are concentrated in Birka and central Sweden (8 graves: 15 coins) and are less common in Hedeby and Jutland (4 graves: 5 coins). Coins from graves at Hedeby and Jutland are in a rather fragmentary condition, but no traces of holes or rivets are to be seen. The coins from northern graves are much better preserved, and all except one have holes, often with the suspension loop still preserved. KG 4–5 appear in 35 localities. The coins are spread between Cuerdale in the west to Kislaya in the east. Only two hoards belong to the ninth century, all the others being later. Site and stray finds are, of course, more useful than late hoards to locate the most probable region of minting of anonymous coins. There is an obvious concentration of site and stray finds in and around Ribe. On the other hand, up until now, not a single coin of KG 4–5 has been found in or around Hedeby. Maps of the Combination-Groups produced in the 1960s, long before the excavations at Ribe, show no concentration of coins at all there (Figure 2.5). There was then no reason to discuss the possibility of there having been a second mint in south-western Scandinavia in the ninth century, but now the picture has changed. Two fragmentary coins, found at



Figure 2.5 Finds of KG 1–6, with geographical division into Areas I–VI: Area II = SchleswigHolstein; Area IIIa = Jutland, including Ribe (after Malmer 1966, plate 55).

Ribelund and at Rosenallé, are undoubtedly KG 5. The remaining seven pieces under discussion are all, except one, from the Post-Office excavation and very badly corroded. At least two of them are certainly KG 5. This means that KG 5 is now comparatively well represented at Ribe. The grave-finds of KG 4–5 are concentrated at Birka, with 12 out of the total of 14 known graves. All are women’s graves, and all the coins have been adapted to ornaments. Obviously, the women at Birka were delighted to wear coins from abroad around their necks, especially those with the Wodan/Monster motif, probably as a kind of amulet.



KG 6, the last and smallest Combination-Group of the ninth century, with only 23 coins, appears in 13 separate finds. No fewer than 11 are hoards from the mid-tenth century or even later. The two remaining finds are from Jutland: one from a grave, and the other from a site to the south-west of Ribe. Maybe KG 6 is represented inside Ribe as well; we do not know for certain because of the bad condition of most of the coins from the Post-Office excavation. KG 6 is not so far represented around Hedeby, and it is not represented at Birka either, which is indeed a remarkable circumstance well worth a paper of its own.

Scandinavian coinage in the ninth century: summary There are four Combination-Groups, KG 3–6, which were minted at the same weight as Carolingian obols or half-deniers, but with the same diameter as the Carolingian denier. The date was the period of Louis the Pious, probably the 820s, but KG 6 is a little later. The volume of the coinage seems to have been considerable. The coinage started with imitations of Carolus/Dorestad deniers and ended with imitations of Wodan/Monster sceattas. In between, there existed a mixture of other types: man, house, deer, cockerels and, most importantly, ships of different types. There is a clear gap between KG 5 and KG 6, with KG 6 appearing as a peculiar group.

WERE THERE ONE OR TWO MINTS IN SOUTH-WESTERN SCANDINAVIA IN THE NINTH CENTURY? The geographical distribution of the site and stray finds clearly indicates south-western Scandinavia as a region of minting in the ninth century. Within the south-western region of minting, the geographical distribution seems to indicate the existence of two different mints, one in Schleswig-Holstein (Figure 2.5, Area II) and one in southern Jutland (Figure 2.5, Area IIIa). Moreover, the border-line is not between KG 5 and 6, which would cause no problem at all, but between the interlinked groups KG 3, 4 and 5. Is there a reliable method by which we can single out a number of KG 5 Wodan/Monster imitations and refer them to a second mint? Let us call the two possible mints, Mint A (‘Hedeby’) and Mint B (‘Ribe’). Mint A (Area II) was strongly influenced by the coinage of the northern Carolingian Empire and started its coinage with Carolus/Dorestad imitations, that is KG 3. But Christian influence, and domestic, perhaps heathen ambitions, together with advanced domestic handicraft, caused the use of other motifs as well, such as deer, cockerels and Wodan/Monster, that is part of KG 3, KG 4 and part of KG 5. Then production at Mint A came to an end. More or less at the same time, Mint B (Area IIIa), started by striking only Wodan/Monster-imitations of KG 5, very like those which were struck at Mint A. After a couple of years it was time also for Mint B to close. Around the middle of the ninth century, a new variant of the old Wodan/Monster type, that is KG 6, was minted. Whether this was at Mint A or at Mint B it is impossible to decide, but we do know that Mint A started to strike a new type of Carolus-Dorestad imitation at the end of the ninth century, with a considerably reduced weight, namely KG 7. We also know that Mint A continued to work throughout the tenth century producing KG 7, 8 and 9, probably not continuously, but with longer or shorter breaks (Figure 2.6,1 and 2). Mint B,



Figure 2.6 Scale c 1/1: (1) KG 8, Area II (‘Hedeby’), from c 950; (2) KG 9b, Area II (‘Hedeby’), from c 975; (3) KG 10a, Area III (Denmark), from c 975/80 and (4) KG 11, Area III (Denmark), from c 975/80 (© B Malmer).

on the other hand, seems to disappear from the history of coinage, until the beginning of the eleventh century, when certain Quatrefoil imitations of Canute appear. However, there are certainly scholars who would find it natural to place KG 10, 11 and 12 – the Danish Combination-Groups of the last quarter of the tenth century (Figure 2.6, 3 and 4) – before the Quatrefoil imitations of Mint B. The above proposal should be regarded not even as a hypothesis, but as just a preliminary sketch. The problem is to find out which KG 5 coins could possibly have been minted at Mint B, and which could not. Another critical question is: are KG 3, 4 and 5 contemporary or not? When two contemporary coinages exclude each other geographically, it means that each ruler had enough power to exclude the other currency from his territory. It also means that both territories were monetised, at least to a certain degree. How should we combine this with political conditions in south-western Scandinavia under Godfred’s sons, also the time of Ansgar’s mission? The alternative explanation is, of course, that Mints A and B produced two contemporary, small local currencies that existed side by side without competition. There is a parallel in the tenth century, just mentioned above. At that time, only tenth-century Carolus-Dorestad imitations, the so-called Hedeby ‘half-bracteates’ (Figure 2.6, 1–2), appear in finds from Schleswig-Holstein. But in Denmark, to the north of Schleswig-Holstein, a considerable quantity of another type of coin has been found, the small ‘half-bracteates’ with crosses, which were contemporary Danish coins (Figure 2.6, 3–4). Those two coinages are mutually very different as to design and size, thus easy to distinguish, and the number of finds is considerable. The political background is clear. The Danish coinage with crosses belongs to the period of Harald Gormsson (Bluetooth), ‘that Harald who won for himself all Denmark and Norway and Christianized the Danes’ (Harald’s rune-stone at Jelling, quoted from Lund 1997, 160). But what about the same geographical area 100 years earlier? The Combination-Group in question, KG 5, presents hardly any internal differences as to design, and has the same size and weight as KG 3 and 4. Coin finds are very few. Outside Ribe, the finds are almost exclusively from



the tenth and eleventh centuries (von Heijne 2004, Region 10, Finds 1–22). Correspondingly clear political powers, as in the tenth century, or corresponding political boundaries are not easy to find during the rather turbulent decades c 820–50. Archaeologists have little to contribute to this debate. Compared to coins, other objects of silver are much more difficult to date (Malmer 2002c, 124, 126). This leads to chronological discrepancies. The conditions of ‘silver economy’ in Schleswig-Holstein, and especially the Hedeby area, have been studied in detail by Heiko Steuer and Ralf Wiechmann. In Steuer’s Period 1, the late eighth century to c 880 (Steuer et al 2002, 136–40, cf. Steuer 1987, 490, and Steuer 1997, 344), barter (Tauschwirtschaft) was predominant, but finds of scales and weights show that precious metal was weighed. During this period there was ‘a first attempt’ at a domestic ‘Münzgeldwirtschaft’ (that is KG 3–6). Steuer’s second period, c 880–950/90, starts with a kind of revolution, the introduction of foldable, standardised scales and weights (1997, 344). Now ‘Gewichtsgeldwirtschaft’ is predominant, but in the Hedeby area Steuer admits a ‘local numismatic room’ (KG 7–9), in agreement with Wiechmann (1996, 190–193). The change from barter to payment by weighed silver c 880 corresponds rather well with the Scandinavian Combination-Groups. There is certainly a rather long period of time between KG 6 and KG 7. KG 6 belongs to Steuer’s period 1 and KG 7 to his period 2. Steuer’s periods 1–2 are also in agreement with the fact that Scandinavian coins in finds from the ninth century are often fragmentary, but not cut. Scandinavian coins in finds from the tenth century are often cut into small pieces, in KG 4–5 as well as in KG 6. But this important agreement between archaeology and numismatics does not help to solve the problem of two more or less contemporaneous Combination-Groups, KG 3–4 and KG 5, representing two different, possibly mutual competing coinages. Steuer’s archaeological grouping is chronologically far too coarse. The same is true for Scandinavian archaeologists, for example Birgitta Hårdh in her study, Silver in the Viking Age. There are considerable geographical differences, where Storebælt indicates a border between a western and an eastern system, but no special economic contrast is found between Hedeby and Ribe (1996, 31, 170). More detailed chronological divisions are seldom possible. For example, in southern Sweden the first period of silver hoards lasts nearly 200 years, ending as late as c 970 (Hårdh 1976, 39). Conclusion: the hypothesis of there having been two mints in south-western Scandinavia in the ninth century creates a number of consequences, which need to be discussed thoroughly by numismatists, archaeologists and historians.

GENERAL IMPORTANCE OF THE COINAGE, ILLUSTRATED BY A SMALL DETAIL A little fish can be seen under some of the ships (Figure 2.7, 1–3). There are 10 such designs, of which eight are fairly discernible (Figure 2.7, 4). Some of the fish are rather naturalistic, with small fins. The Dorestad pre-reform coinage has an axe as a feature of the mint. This axe was quite accurately copied at one of the supposed Scandinavian mints, Mint A (Figure 2.1 (ii-b–iv-b)). A very interesting contribution to the question of intentional features was found at the site of Gudme on Funen (Figure 2.7, 5; von Heijne 2004; Find 7.23; Malmer 2000). On the reverse there is a small fish, which has been squeezed in under the axe. Earlier Dr Wiechmann noticed the back of a fish on a coin found at Hedeby (Figure 2.7, 6; Wiechmann, unpublished list). With the help of these two



Figure 2.7 KG 3 reverses showing fish symbol (© B Malmer): 1–3, with ship; 4, enlargements of fishes with ships; and 5–7, with the Dorestad axe.

coins, it was possible for me to identify a third fish, hidden behind the usual fragmentary suspension loop on a coin from a grave at Birka (Figure 2.7, 7). Not only the main motifs of the coins (such as a deer or a ship), but also the small details (such as a mask or a serpent) are likely to be meant as symbols; they are intended to express a certain message. In this case, it seems natural to ask whether the little fish was perhaps meant as a feature, a symbol for the probable first Scandinavian town with a mint, just as the axe was a symbol for Dorestad. If so, was it heathen or Christian, and what kind of organisation or political power, does this feature represent? On the other hand, this was the time of the Carolingian Christian mission in southern Denmark, a very active mission with a clear political aim: to incorporate southern Denmark to the Carolingian empire. Archbishop Ebo visited Denmark in 823, Ansgar and Authbert in 826. Motifs such as



deer/serpent and ship/fish are familiar Christian symbols (Malmer 2002d, and 2004). The deer (Christ) kills a serpent (the Devil), while the ship (the Christian church) is combined with a fish (the Christian faith). Consequently, concerning ‘silver economy’ in southern Denmark in the ninth century, there is certainly a strong economic influence from the Carolingian Empire, but also influences from a powerful Carolingian Christian mission (for further discussion of links between Christianisation and the introduction of coinage, see Williams, Chapter 11, in this volume). Silent coins do not, by themselves, solve problems concerning the nature of the silver economy, nor do they give any definite answer concerning the particular theme welcomed in the call for papers: ‘monetary development and state formation’. But they often throw light upon other, more detailed, not unimportant phenomena, which are impossible to discern only by means of the mass of unminted silver.

References Archibald, M M (1985), ‘A Scandinavian coin of Carolingian type from the Cuerdale hoard’, Hikuin 11, 79–82 van Gelder, H E (1980), ‘Coins from Dorestad, Hoogstraat I’, in W A van Es and W J H Verwers, Excavations at Dorestad I: The Harbour, Hoogstraat I, (Nederlandse Oudheden 9), The Hague, 212–224 Grierson, P and Blackburn, M (1986), Medieval European Coinage, 1. The Early Middle Ages, Cambridge University Press, Cambridge Hårdh, B (1976), Wikingerzeitliche Depotfunde aus Südschweden. Probleme und Analysen (Acta Archaeologica Lundensia, Series in 8° minore, no. 6), Lund Hårdh, B (1996), Silver in the Viking Age. A Regional-Economic Study (Acta Archaeologica Lundensia, Series in 8°, no. 25), Lund Hatz, G (2001), Der Münzfund vom Goting-Kliff/Föhr (Numismatische Studien 14), Hamburg von Heijne, C (2004), Särpräglat. Vikingatida och tidigmedeltida myntfynd från Danmark, Skåne, Blekinge och Halland (ca 800–1130) (Stockholm Studies in Archaeology 31), Stockholm Lund, N (1997), ‘The Danish Empire and the end of the Viking Age’, in P Sawyer (ed), The Oxford Illustrated History of the Vikings, Oxford University Press, Oxford, 156–181 Malmer, B (1966), Nordiska mynt före år 1000 (Acta Archaeologica Lundensia, Series in 8°, no. 4), Lund Malmer, B (2000), ‘Skeppsmyntet från Okholm – danska 800-talsmynt med fisksymboler’, NNUM 2000, no. 5, 103–109 Malmer, B (2002a), ‘Münzprägung und frühe Stadtbildung in Nordeuropa’, in K Brandt, M Müller-Wille and C Radtke (eds), Haithabu und die frühe Stadtentwicklung im nördlichen Europa, Wachholtz Verlag, Neumünster, 117–132 Malmer, B (2002b), ‘The earliest Scandinavian coins and their metrology’, in B Paszkiewicz (ed), Moneta Mediævalis. Studia numizmatyczne i historyczne ofiarowane Profesorowi Stanislawowi Suchodolskiemu w 65. rocznice˛ urodzin, Polskiej Akademii Nauk, Warszawa, 269–277 Malmer, B (2002c), ‘Hedeby i blickpunkten’, NNUM 2002, nos 7–8, 124–126 Malmer, B (2002d), ‘Karolingisk mission och nordiska mynt’, NNUM 2002, nos 9–10, 182, 191 Malmer, B (2004), ‘Kristna symboler på danska mynt’, in N Lund (ed), Kristendommen i Danmark før 1050, Roskilde Museums Forlag, Roskilde, 75–85 Malmer, B (forthcoming), Serpents and Crosses. Scandinavian Coinages from the time of Louis the Pious, Knut the Great, Harthacnut and Anund Jakob (Commentationes de nummis saeculorum IX–XI in Suecia repertis, Nova Series 14), Stockholm



Metcalf, D M (1984), ‘A note on sceattas as a measure of international trade, and on the earliest Danish coinage’, in D Hill and D M Metcalf (eds), Sceattas in England and on the Continent (BAR, British Series 128), Oxford, 159–164 Metcalf, D M (1993–94), Thrymsas and Sceattas in the Ashmolean Museum, Oxford (RNS, Special Publication 276, Vols 1–3), London Metcalf, D M (1996), ‘Viking Age numismatics, 2. Coinage in the northern lands in Merovingian and Carolingian times’, NC 156, 399–428 Morrison, K F and Grunthal, H (1967), Carolingian Coinage (Numismatic Notes and Monographs, no. 158), New York Noonan, T S (1981), ‘Ninth-century dirham hoards from European Russia: a preliminary analysis’, in M A S Blackburn and D M Metcalf (eds), Viking-Age Coinage in the Northern Lands (BAR, International Series 122), Oxford, 47–117 Pedersen, A (2000), ‘A Striding Man from Tissø – a rare imitation of Charlemagne’s Dorestadcoinage’, NNÅ 1994–96, 22–40 Steuer, H (1987), ‘Gewichtsgeldwirtschaften im frühgeschichtlichen Europa. Feinwaagen und Gewichte als Quellen zur Währungsgeschichte’, in K Düwel, H Siems and D Timpe (eds), Untersuchungen zu Handel und Verkehr der vor- und frühgeschichtlichen Zeit in Mittel- und Nordeuropa. Teil IV. Der Handel der Karolinger- und Wikingerzeit (=Abhandlungen der Academie der Wissenschaften in Göttingen Philologisch-Historische Klasse, Dritte Folge, Nr 156), 405–527 Steuer, H (1997), Waagen und Gewichte aus dem mittelalterlichen Schleswig (Zeitschrift für Archäologie des Mittelalters, Beiheft 10), Köln Steuer, H, Stern, W B and Goldenberg, G (2002), ‘Der Wechsel von der Münzgeld- zur Geldwirtschaft in Haithabu um 900 und die Herkunft des Münzsilbers im 9. und 10. Jahrhundert’, in K Brandt, M Müller-Wille and C Radtke (eds), Haithabu und die frühe Stadtentwicklung im nördlichen Europa, Wachholtz Verlag, Neumünster, 133–167 Wiechmann R (1996), Edelmetalldepots der Wikingerzeit in Schleswig-Holstein (Offa-Bücher 77), Wachholtz Verlag, Neumünster


Hedeby and Its Hinterland: A Local Numismatic Region Ralf Wiechmann In Northern Europe, the eighth century is characterised by a remarkable economic development.1 Trade between Western Europe and Scandinavia increased by leaps and bounds and quickly expanded to the Slavic and Baltic coasts of the Baltic Sea. For a small settlement on the Schleswig isthmus, this meant a tremendous change and Hedeby developed as the point of trans-shipment so urgently required for the movement of goods both north and south, and east and west. Its topographical loction was especially wellsuited for this purpose (Figure 3.1). Traders coming from the east could reach the town directly, although situated in a protected position far inland, by means of the Schlei, a river extending 40 km inland. Towards the west, it was separated from the navigable rivers Treene and Eider by only a 16 km-wide land bridge. The north-south connection is marked by the so-called Army Road, which passes by Hedeby only a few kilometres to its west. There, on the narrowest point of the Jutland peninsula, where the loads had to be transported only a short distance over land under the protection of the ‘Danevirke’ (a Danish frontier rampart), Hedeby developed into the first place in Northern Europe where people lived together in a town-like settlement. It was at Hedeby that the transition from the customary rural village to the new urban form of life and settlement took place for the first time in Scandinavia. Soon the settlement also played a part nationally, and during the ninth and tenth centuries it advanced to become the major hub for foreign trade from Eastern to Western Europe and vice versa (Jankuhn et al 1984; Jankuhn 1986; Elsner 1992; Radtke 1999). Hedeby played a special role with regard to the development of minting in Northern Europe (Figure 3.2). The earliest firmly attributable coins in the Scandinavian region were obviously minted here (Malmer 1966; Hatz 1984a; Wiechmann 1999; Malmer 2002), although earlier minting has been suggested at the Danish town of Ribe (for a summary of the controversial discussion, see Hatz 2001a, 60 et seq). According to studies by Brita Malmer, a local coinage workshop must have existed at Hedeby already in the second quarter of the ninth century (see also Malmer, Chapter 2, in this volume). The heavy ‘denier’ first produced imitates West European coins, especially coins from the Frisian town of Dorestad. I shall not go any further here into the still open discussion as to whether earlier coins from the Combination-Groups 5 and 6 may not have been produced in Hedeby, but possibly in Ribe, as suggested by Michael Metcalf (Metcalf 1996a; 1996b; see Metcalf, Chapter 1, in this volume), for this is a problem to be discussed in detail by 29

Figure 3.1 Hedeby (Haithabu): the topographical situation (after Elsner 1992).



Figure 3.1 Continued.




Figure 3.2 The chronology of minting at Hedeby (after Hatz 1984, fig 131).

Malmer (forthcoming). I would like to mention, however, that neither the old excavations at Hedeby nor the more recent ones have brought to light any examples from these groups. After a hiatus at the end of the ninth century, it has been proved that intensive minting took place, especially in the tenth century, which ended in the 980s and accompanied the economic decline of the town. The coins of the ninth and tenth centuries are anonymous; thus, the lord for whom the coins were minted is unknown. Presumably, it was a royal mint, but the merchants settled in the town have also been considered as initiators (see Bendixen 1967, 17; Skaare 1976, 46; Hatz 1984a, 271 et seq). It was only in the eleventh century that minting resumed in Hedeby, within the kingdom of Denmark, although this is of no great importance here.



Figure 3.2 Continued.

It is difficult to recognise the ruling circumstances of this ‘early town’. Until about the year 934 it was under Danish authority, but from then on it was more or less strongly influenced by the German empire. Between 974 and 983, Otto II conquered Hedeby. From 983 onwards the Danish king was once again able to assert his power (Radtke 1999, 366 et seq). The complex balance of power of this town, as well as its heterogeneous population is apparent from the recorded material from the graves. The largest part of the population settled in Hedeby consisted of Saxons and Frisians, while the Scandinavian part of the population represented not more than 5 per cent (or 10 per cent at the most), taking into account the grave-sites furnished with many or few added grave-goods, as well as the projected number of those buried (Steuer 1984, 363).



At that time, the countryside in this area between two seas was characterised for the most part by widespread, partly impenetrable forests, extensive marshes and sparse heaths and moors. Nevertheless, this once ‘very large town at the extreme end of the ocean’ (so At-Tartuschi, an Arab merchant around the year 965; Elsner 1992, 16), ‘meeting point for the tradesmen from all over the word’ (so Archbishop Rimbert of Bremen; Radtke 1999, 376), was not in an isolated position. In the Chronicle of Æthelweard (written around 960) it says, ‘The country of the Angles lies between the Saxons and the people of Jutland; its capital is called “Schleswig” in the language of the Saxons, but “Hedeby” by the Danish’ (Elsner 1992, 13). Although new cultivation methods, such as the three-field system, which increased the amount of crops, gradually led to an increase in agricultural productivity (Radtke 1999, 377), it must be assumed that a town with a population of about 1,000 – as was Hedeby in its heyday – would have required an agricultural catchment area of approximately 600 km2 (Randsborg 1980, 83 et seq). This statement is closely connected to the questions of what the area surrounding Hedeby looked like, and of how it was structured.

HOARD-FINDS AND THEIR STRUCTURES Let us first look at the hoards of this period found in Schleswig-Holstein (Figure 3.3). Two areas in particular can be distinguished here: on the one hand, the then Slavic East

Figure 3.3 The distribution of hoards in Schleswig-Holstein.



Holstein, and on the other, the isthmus of Schleswig, including the district of Angeln. Further finds are on the North Frisian Islands. This picture changes when we separate the period up to 980 – the period during which minting took place in Hedeby – from the period 980–1050 (Figures 3.4 and 3.5). Although there is a definite focus around Hedeby and in Angeln during the first period, there are no more finds in this area after 980. It is particularly noticeable that there is no hack-silver in the Angeln region, which is especially remarkable, because hack-silver and foreign coins can be regarded as characteristic of finds in Scandinavia during this period. One exception here is a find complex near Schleswig, which has since disappeared and which can have only been buried around 914 at the earliest (Wiechmann 1996, no. 37). Furthermore, all the hoard finds of this region consist solely of coins. Only one find, from near the mouth of the river Schlei, is made up of Islamic dirhams (Weichmann 1996, no. 10). Apart from Hedeby coins, two further finds from the ninth century in Hedeby itself contained small quantities of coins from other mints (Weichmann 1996, nos 4 and 5). All other hoard finds in this region – six in total – contain only Hedeby coins. If we look at the total distribution of the finds containing Hedeby coins in SchleswigHolstein we discover something most interesting (Figure 3.6). Hedeby coins are found both in East Holstein and in Angeln, but the two regions also show fundamental differences. Hedeby coins are only found in small quantities in the finds in Slavic East Holstein. Moreover, the individual coins are almost always cut up, or have at least been

Figure 3.4 The distribution of hoards in Schleswig-Holstein (790–980).



Figure 3.5 The distribution of hoards in Schleswig-Holstein (980–1050).

examined with regard to their content of silver by means of pecking with a knife (Malmer 2000), as was typical for Scandinavian hoards at this time. The silver objects show characteristic signs of a money-weight economy, being cut into pieces and showing testmarks in the form of nicks and pecks. On the other hand, a region is apparent to the north of Hedeby, roughly equivalent to the district of Angeln, within which certain common features are evident in the structure of the hoards. All the coins found there are whole, showing no signs of having been examined (no pecking) and are only insignificantly bent, or not at all. This is well demonstrated by the hoard from Steinfeld, which consists of Hedeby coins from Combination-Group 7 (Figure 3.7). This hoard does contain fragmented pieces, but these have not been intentionally cut, rather they were broken by ploughing on the find-spot. This is evident because, when the first coins appeared in 1945, there were only very few with broken edges, whereas the second parcel from 1984 and the third parcel from 1985 are much more fragmented, especially at the edges. Those graves in the Angeln region in which coins have been found also contain only Hedeby coins, for example, Gelting, Nadelhöft, district Schleswig-Flensburg, unpublished; Sieverstedt, district Schleswig-Flensburg (Mechlenburg 1844, 40, no. 6); Süderbrarup, district Schleswig-Flensburg (Aner 1952, 65 et seq); Thumby Bienebek, district Rendsburg-Eckernförde (Müller-Wille 1976, 42). Thus, the numismatic material is definitely dominated by the local coins. Obviously, the increased production of coins during the first half, but especially during the second half, of the tenth century resulted in



Figure 3.6 The distribution of Hedeby coins in Schleswig-Holstein.

the replacement of hack-silver and also all foreign coins in the finds from around Hedeby. All of these coin finds must be seen in context of the local market trade in Hedeby, the point of trans-shipment. The population which had settled in the area surrounding the trading settlement probably ensured the provision of both food and the raw materials required for manual production (Jankuhn 1980; Randsborg 1980, 90, 149; Steuer 1987, 188 et seq, Abb 24; MĂźller-Wille 1988). The almost complete absence of hack-silver and foreign coins from this region could lead to the conclusion that a developed area of coin circulation existed, which was supplied with coins from Hedeby. There is nothing to show that the corresponding finds were accumulated through foreign trade.

FINDS IN HEDEBY AND THEIR STRUCTURES What importance, however, did minting have in Hedeby itself? Do the half-bracteates coined there also prevail in numismatic material? What role did the Islamic, Byzantine, English and German coins, which were also discovered there, play? What was their relationship to each other? How does their existence impact on coin minting in Hedeby? During the almost 100 years of exploration of the trading-centre of Hedeby, a total of 151 coins has so far been found. Of these, 118 specimens can be regarded as true



Figure 3.7 The Steinfeld hoard, with Hedeby coins.

settlement finds, while 19 have been found in graves. Also of interest in this connection are 10 lead-tin imitations of Islamic dirhams (Steuer et al 2002, 156). In addition, there are 14 coins from two small hoards, one of which, consisting of seven coins, was found above a Grubenhaus in the ‘South Settlement’ (Wiechmann 1996, no. 4), and the other was excavated in 1980 in the harbour area, containing seven coins and 600 blue beads (Wiechmann 1996, no. 5). To give an idea of the material, only a few highlights can be mentioned here, starting with a Frisian sceat of the Wodan-Monster type (Figure 3.8), which was struck c 730–65 (Komnick 1994, 107, no. 2; Metcalf 1993–94, vol 1, 275 et seq; 1984, 159). This type is the forerunner of the heavy deniers of the Wodan-Monster type, which were probably produced in Hedeby (or Ribe?) in the ninth century (see Malmer, Chapter 2, in this volume). As an example of the inflow of Islamic coins, an Abbassid dirham can be illustrated Figure 3.8 (ii). This coin dates from the years 781/82 and is one of the oldest coins to have been found in Hedeby. It is of interest because it has a grafitto in the form of a rune, although this is not particularly unusual because approximately 3 per cent of the Islamic dirhams from Sweden display runic graffiti or small pictures (Hammarberg and Rispling 1985, 63), especially of Thor’s hammers or crosses. In Schleswig-Holstein, however, only 2.3 per cent of the coins display any graffiti (Wiechmann 1996, 109), with seemingly a religious background. A very unusual piece is a Byzantine gold solidus (Figure 3.8 (iii)), struck in the name of Theophilius in 830/31–40 (Grierson 1973, 427, class IIIe). In general, coins of








Figure 3.8 (i)–(v) Six coins from Hedeby: (i) sceatta, settlement-find; (ii) Abbassid dirham, with a runic graffito, settlement-find; (iii) Byzantine gold solidus, grave-find; (iv) Northumbrian styca, settlement-find and (v) Norwegian penny struck by Harald Hardrüde, grave-find.



Byzantine origin are very rare in Scandinavia, and only a very few gold pieces reached the north. Even in older times it must have been a precious object, as we can see to be the fact, that the coin was transformed into a brooch. An extremely rare piece for this region is a Northumbrian styca (Figure 3.8 (iv)), struck in the name of Æthelred II, between 841 and 844, in York. The distribution of these coins shows clearly that only a few pieces left Britain. Apart from this coin from Hedeby, only a few more stycas have been found, including two in a Norwegian grave, one in Dorestad and one in Schuby, a settlement not far from Hedeby (Wiechmann 1998, 454). Single-finds are known from Zutphen (Willemsen 2004, 134–136) and Mainz (Stoess 1994, 178, no. 26), obviously transported via the river Rhine. Two specimens, struck in the name of Æthelred II, have been excavated from graves 29 and 176 at Birka, Sweden (Arbman 1943, 12, Abb 11,7 and 75; 1940, Taf 141,1–2; Blackburn and Jonsson 1981, 150). In addition to this small group of stycas known from outside Britain, there are two unpublished examples found at Menzlin in north-eastern Germany (for detailed identification, see Wiechmann forthcoming). Another coin uncommon in the South Baltic region is a Norwegian penny, struck in the name of Harald Hardråde (Figure 3.8 (v)), belonging to the Triquetra type produced between 1047 and 1066 (Skaare 1976, 200, no. 56). In Schleswig-Holstein there are only a few Norwegian pennies on record, one of which originates from a large hoard found in Lübeck, consisting of more than 2,800 coins (Wiechmann 1996, no. 17, B2419). Three Norwegian pennies have been excavated in Schleswig, all as single-finds; one of these was struck in the name of Harald Hardråde (1047–66), and the other two date to the end of the eleventh and the beginning of the twelfth century (Hatz 2001b, 184, nos 25 and 26). Another penny of Harald Hardråde has been found in the central hillfort of Starigard-Oldenburg, in eastern Holstein (cf. Hatz 1984b, 175, no. 32u, Taf 52,32; Skaare 1976). As we have seen, there are coins of very different provenances in the numismatic material of Hedeby. The various groups of coins are, however, represented in extremely different percentages. Generally, the numismatic material is heavily dominated by the local Hedeby coin (Figure 3.9). This group of coins represents 47 per cent of the total, or approximately half of the existing coins and, as such, it most clearly dominates the entire material. This is followed by coins from the Islamic world (20.5 per cent), Scandinavia – excluding Hedeby’s tenth/eleventh-century issues (7.3 per cent) and the German deniers (6 per cent). Further groups of coins are only represented in minor quantities. A simplified, comprehensive presentation of the coins of this period from the larger regions of (i) Western Europe, (ii) Eastern Europe, (iii) Hedeby and (iv) others being issued in Scandinavia (excluding Hedeby’s tenth/eleventh-century issues) makes it very clear that the overwhelming number of Hedeby coins were minted locally (Figure 3.10). The other West European and East European coins tend to be represented in equal amounts. The category ‘Scandinavia (excluding Hedeby’s tenth/eleventh-century issues)’, which mainly consists of the national Danish coinage dating from the eleventh century, still amounts to almost 6 per cent of the entire number of coins. If these large groups are marked off on a time axis, a distribution in chronological perspective becomes apparent (Figure 3.11). The Y-axis shows the number of existing pieces; the X-axis shows the chronological distribution of each kind of coin, arranged by decade. The decisive factor is always the terminus post quem of the individual coin. First,



Figure 3.9 Coins of different provenances (in %) of the Hedeby finds (settlement and graves).

Figure 3.10 Main coin groups of different provenances (in %) of the Hedeby finds (settlement and graves).

it is possible to observe that the numismatic material reaches beyond the period of time known from historical sources. However, it can be well paralleled by other archaeological finds, which makes it plausible that settlement began at Hedeby as early as the eighth century. Altogether, the distribution of the few coins from the eighth century shows only a thin density. Coin finds only increase after the year 800, which fits together well with the beginning of systematic building activity at Hedeby, estimated as having been about the year 811 (for the settlement, see Steuer 1974, 33 et seq; 1984, 342 et seq; for the



Figure 3.11 Chronological distribution of the main coin groups at Hedeby (settlement and graves).

dendrochronological datings, see Eckstein and Schietzel 1977, Abb 12; for the find of a Mannheim-type sword, see Mßller-Wille 1982, 121; for the reticella glass, see Haevernick 1979, 158 and Evison 1982; and for the early imported ceramics, see Janssen 1987, 30, 44, Taf 11,1–7). As is to be expected, the ninth century is dominated by coins from Eastern Europe, but during the first quarter of the ninth century a definite increase in local Hedeby coins can be observed. This picture changes completely in the tenth century, with the start of minting activities marking a decisive change. The mass of Eastern European, and especially Islamic coins, is dated to shortly before the main period of minting at Hedeby. During the period from about 900 to 960/70, the numismatic material is almost completely dominated by the local coins minted in the town. It is only with the minting of Denmark’s national coinage in the eleventh century that local minting becomes recognisable once again (for the end of the settlement activity at Hedeby, see Meyer 1998, 103 et seq). If we sum up our observations with regard to the hoard finds on the one hand and the analysis of the settlement finds on the other, these two views complement each other. The effects of a deliberate coinage policy can be recognised in the Hedeby area. The new type of coin was certainly issued for economic reasons, which were of advantage to the coin lord or lords in Hedeby. The establishment of a uniform coinage system in and around Hedeby appears to be the result of an increase in coinage production from the beginning of the tenth century, followed by mass production during the second half of the tenth century (Malmer 1966, 247), which replaced hack-silver and also all foreign coins in the finds from and around Hedeby. On the other hand, Hedeby



coins were used indiscriminately on the North Frisian Islands and in Slavic East Holstein. The coins obviously had a different function in Angeln, but also in the rural area called Schwansen, which lay to the south of the river Schlei and was also protected by the Danevirke, where finds from the Viking Age are particularly concentrated and which must be regarded as the hinterland of the large trading centre. Apparently, a regional circulation of coins developed in this area, which was supplied with coins from Hedeby. These finds are very clearly connected with the local market trade in Hedeby, the place of trans-shipment. The distribution of Viking-Age finds in Angeln and Schwansen from the ninth to the tenth century shows not only an increase in the number of sites, but also makes it clear that the distribution was more widespread (Figures 3.12 and 3.13). This is especially true for southern and northern Angeln, but less so for Schwansen (Willroth 1992, 455), situated south of the river Schlei. On the whole, the tenth century must certainly be considered to have been a phase of regional development (Kuhlmann 1958, 56 et seq; Jankuhn 1961, 151 et seq; Müller-Wille 1973, 117). The distribution of finds pointing towards settlement in the Angeln region corresponds surprisingly with the distribution of Hedeby coins in this period. It is doubtful whether the rural settlements grouped around the large trading centre were orientated solely towards agriculture. The results from the archaeologically investigated Viking-Age settlements of Schuby, in the district of Schleswig-Flensburg (Kühn 1986), Kosel, in the district of Rendsburg-Eckernförde (Meier and Reichstein 1984; Meier 1994), and Winning, in the district of Schleswig-Flensburg (Meier 1998), which turned up only a few longhouses and many small pit-houses, suggest manual rather than agricultural production (Meier 1994, 193; 1998, 123 et seq). Seemingly, weaving took first place there. In general, the archaeological objects from Grubenhaus-settlements have been interpreted as connected with handicrafts and participation in trade and exchange. This has been demonstrated several times not only in the Hedeby region, but also in other places, for example at the edge of the sandy uplands in the Ribe region (Jensen 1991, 65 et seq). The finds from Kosel, consisting of a weight, a small balance and a few pieces of hack-silver, which unfortunately cannot be dated exactly, indicate that payment was also made on the basis of weight, but this is not a direct contradiction. Likewise in Sigtuna, located in the Lake Mälar region of central Sweden, periods of a coin-weight economy vary with periods of a silver-weight economy (Gustin 1997, 174 et seq). In any case, the hoard finds tell another story. Even though it is only at Schuby that coins from Hedeby have yet been found in any of the settlements mentioned above, payment for these goods seems to have been made, in part at least, in coins. The result was that some of the rural population became financially very sound. It is unknown whether or not these people were also involved in production for overseas trade. In the end, this led to the expansion of ‘the circle for products for foreign trade beyond the narrow field of trading places’ (Jankuhn 1980, 149), to the surrounding region, which was not agriculturally structured. This is verified by a number of imported goods from the settlements of Schuby in the district of Schleswig-Flensburg, Kosel in the district of Rendsburg-Eckernförde (Müller-Wille 1988; Meier 1994, 194) and Winning in the district of Schleswig-Flensburg (Meier 1998, 121 et seq), which have already been mentioned. The question that arises in this connection is how far the strictly standardised coinage system in Hedeby, especially that of the tenth century, was orientated less towards foreign trade rather than more towards this domestic trade. Both the structure of the hoards, as well as the analysis of the settlement and grave finds, lead to this conclusion.



Figure 3.12 Distribution of Viking-Age finds in the districts of Angeln and Schwansen, eighth and ninth century (after Willroth 1992, map 69).

Taken in a wider perspective, the conclusions with regard to the circulation of coins in and around Hedeby are also of interest for other important trading centres in Scandinavia. I refer here to the finds of sceattas in and around Ribe, in Denmark, and to the finds of coins of Olof Skรถtkonung in and around Sigtuna, in Sweden (Kilger 1995). It



Figure 3.13 Distribution of Viking-Age finds in the districts of Angeln and Schwansen, tenth century (after Willroth 1992, map 70).

may be possible to recognise certain mechanisms and processes here, which are characteristic for the structure of early urban establishments.

Note 1. The original version of this paper, as delivered in May 2000, was a preliminary report on the author’s work in progress, concerning the numismatic evidence from the settlement area and



graves at Hedeby. This has since been completed and is shortly to be published in full (see Wiechmann, forthcoming); the text of the present article has consequently been updated and slightly modified.

References Aner, E (1952), ‘Die wikingerzeitlichen Kammergräber am Thorsberger Moor’, Offa 11, 60–78 Arbman, H (1943), Birka I. Die Gräber, K. Vitterhets Historie och Antikvitets Akademien, Stockholm Bendixen, K (1967), Denmark’s Money, Nationalmuseet, Copenhagen Blackburn, M A S and Jonsson, K (1981), ‘The Anglo-Saxon and Anglo-Norman element of North European coin finds’, in M A S Blackburn and M Metcalf (eds), Viking-Age Coinage in the Northern Lands (BAR, International Series 122), Oxford, 147–255 Eckstein, D and Schietzel, K (1977), ‘Zur dendrochronologischen Gliederung und Datierung der baubefunde von Haithabu. Berichte über die Ausgrabungen’, Haithabu 11, 141–164 Elsner, H (1992), Wikinger Museum Haithabu: Schaufenster einer frühen Stadt, Karl Wachholtz, Neumünster Evison, V I (1982), ‘Bichrome glass vessels of the seventh and eighth centuries’, Studien zur Sachsenforschung 3, 7–21 Grierson, P (1973), Catalogue of the Byzantine Coins in the Dumbarton Oaks Collection and in the Whittemore Collection. Vol 3, Leo III to Nicephorus III 717–1081. Part 1, Leo III to Michael III (717–867). Part 2, Basil I to Nicephorus III (867–1081), Washington, DC Gustin, I (1997), ‘Islam, merchants, or king? Who was behind the manufacture of Viking Age weights’, in H Andersson, P Carelli and L Ersgård (eds), Visions of the Past. Trends and Traditions in Swedish Medieval Archaeology (Lund Studies in Medieval Archaeology 19; Riksantikvarieämbetet Arkeologiska Undersökningar Skrifter, Nr 24), Central Board of National Antiquities, Stockholm, 163–177 Haevernick, T (1979), ‘Karolingisches Glas aus St. Dionysius in Esslingen’, Forschungen und Berichte zur Archäologie des Mittelalters 6, 157–171 Hammarberg, I and Rispling, G (1985), ‘Graffiter på vikingatida mynt’, Hikuin 11, 63–78 Hatz, G (1984a), ‘Zur Münzprägung in Haithabu’, in H Jankuhn, K Schietzel and H Reichstein (eds), Archäologische und naturwissenschaftliche Untersuchungen an ländlichen und frühstädtischen Siedlungen in deutschen Küstengebieten vom 5. Jahrhundert v. Chr. bis zum 11. Jahrhundert n. Chr. Bd 2, Handelsplätze des frühen und hohen Mittelalters, Weinheim, 260–273 Hatz, G (1984b), ‘Numismatische Bearbeitung’, in I Gabriel, Starigard/Oldenburg. Hauptburg der Slawen in Wagrien. Stratigraphie und Chronologie (Archäologische Ausgrabungen 1973–1982 (Offa-Bücher 52), Neumünster, 169–185 Hatz, G (2001a), ‘Der Münzfund vom Goting Kliff/Föhr’, Numsimatische Studien 14, Hamburg, 7–93. Hatz, G (2001b), with a supplement by R Wiechmann, ‘Die Münzfunde aus Schleswiger Grabungen’, Ausgrabungen in Schleswig. Berichte und Studien 15. Das archäologische Fundmaterial II, 169–206 Jankuhn, H (1961), ‘Die Entstehung der mittelalterlichen Agrarlandschaft in Angeln’, Geographische Annalen 43, 151–164 Jankuhn, H (1980), ‘Beobachtungen und Überlegungen zur “Infrastruktur” des wikingerzeitlichen Seehandels’, Offa 37, 146–153 Jankuhn, H (1986), Haithabu. Ein Handelsplatz der Wikingerzeit, Karl Wachholtz, Neumünster Jankuhn, H, Schietzel, K and Reichstein, H (eds) (1984), Archäologische und naturwissenschaftliche Untersuchungen an ländlichen und frühstädtischen Siedlungen in deutschen Küstengebieten vom 5. Jahrhundert v. Chr. bis zum 11. Jahrhundert n. Chr. Bd 2, Handelsplätze des frühen und hohen Mittelalters, Weinheim.



Janssen, W (1987), Die Importkeramik von Haithabu (Die Ausgrabungen in Haithabu 9), Karl Wachholtz, Neumünster Jensen, S (1991), Ribe zur Wikingerzeit, Den antikvariske Samling, Ribe Kilger, C (1995), ‘Myntfynd och den vikingatida silverhandeln’, in K Jonsson, U Nordlind and I Wiséhn (eds), Svenska Numismatiska Föreningen: Myntingen i Sverige, 995–1995, Stockholm, 27–42 Komnick, H (1994), Die Fundmünzen der römischen Zeit in Deutschland. VIII, SchleswigHolstein und Hamburg, Berlin Kuhlmann, H J (1958), Besiedlung und Kirchspielorganisation der Landschaft Angeln im Mittelalter (Quellen und Forschungen zur Geschichte Schleswig-Holsteins 36), Neumünster Kühn, H J (1986), Eine Siedlung des frühen und hohen Mittelalters bei Schuby (Kr. SchleswigFlensburg) (Bericht der Römisch-Germanischen Kommission 67), 479–489 Malmer, B (1966), Nordiska Mynt före år 1000 (Acta Arch Lundensia, Series 8, no. 4), Lund Malmer, B (2000), ‘Om registrering av cirkulationsspår på mynt från vikingatiden’, NNUM 2, 23–27 Malmer, B (2002), ‘Münzprägung und frühe Stadtbildung in Nordeuropa’, in K Brandt, M Müller-Wille and C Radtke (eds), Haithabu und die frühe Stadtentwicklung im nördlichen Europa (Schriften des Archäologischen Landesmuseums 8), Schleswig, 117–132 Malmer, B (forthcoming), Serpents and Crosses. Scandinavian Coinage from the time of Louis the Pious, Knut the Great, Harthacnut and Anund Jacob (Commentationes de nummis saeculorum IX–XI in Suecia repertis, Nova Series 14), Stockholm Mechlenburg (1844), ‘Fünf lithographische Tafeln mit Erläuterungen von Herrn Apotheker Mechlenburg in Flensburg’, Kieler Berichte 9, III, Zweiter Anhang, 33–40 Meier, D (1994), Die wikingerzeitliche Siedlung von Kosel (Kosel-West), Kreis RendsburgEckernförde. Siedlungsarchäologische Untersuchungen in Angeln und Schwansen 3 (OffaBücher 76), Neumünster Meier, D (1998), ‘Winning – Eine wikingerzeitliche Siedlung am Ufer der inneren Schlei’, in A Wesse (ed), Studien zur Archäologie des Ostseeraumes. Von der Eisenzeit zum Mittelalter, Neumünster, 117–126 Meier, D and Reichstein, J (1984), ‘Eine wikingerzeitliche Siedlung westlich von Kosel, Kreis Rendsburg-Eckernförde (LA 117). Grabungen des Landesamtes für Vor- und Frühgeschichte von Schleswig-Holstein (1975–76)’, Offa 41, 113–168 Metcalf, D M (1984), ‘A note on sceattas as a measure of international trade, and on the earliest Danish coinage’, in D Hill and D M Metcalf (eds), Sceattas in England and on the Continent. The Seventh Oxford Symposium on Coinage and Monetary History (BAR, British Series 128), Oxford, 159–164 Metcalf, D M (1993–94), Thrymsas and Sceattas in the Ashmolean Museum, Oxford (RNS Special Publication 27, 3 Vols), London Metcalf, D M (1996a), ‘Kan Hedeby mønter også præget Ribe og Birka?’, NNUM 5, 74–77 Metcalf, D M (1996b), ‘Viking-age numismatics, 2. Coinage in the northern lands in Merovingian and Carolingian times’, NC 156, 399–428 Meyer, J (1998), Keramik der Zeit um 1100 aus dem Hafenviertel von Schleswig. Archäologie in Schleswig/Arkæologi i Slesvig 5/1996. Symposium Hjemsted 1.11.1996–2.11.1996, Kiel, 89–116 Müller-Wille, M (1973), ‘Das jüngereisenzeitliche Gräberfeld’, in F Voss, M Müller-Wille and E-W Raabe (eds), ‘Das Höftland von Langballigau an der Felensburger Förde’, Offa 30, 82–123 Müller-Wille, M (1976), Das wikingerzeitliche Gräberfeld von Thumby-Bienebek (Kr. Rendsburg-Eckernförde) 1 (Offa-Bücher 36), Neumünster Müller-Wille, M (1982), ‘Zwei karolingische Schwerter aus Mittelnorwegen’, Studien zur Sachsenforschung 3, 101–154 Müller-Wille, M (1988), ‘Hedeby und sein Umland. Archäologische Zeugnisse des Handels’, in B Hårdh (ed), Trade and Exchange in Prehistory [Festschrift B. Stjernquist] (Acta Arch Lundensia 16), Lund, 271–278



Müller-Wille, M (1992), ‘Archäologischer Import im Umkreis von Haithabu’, in H Menke (ed), Die Niederlande und der europäische Nordosten. Ein Jahrtausend weiträumiger Beziehungen (700–1700) (Veröff. Inst. Landesforsch CAU Kiel 1), Neumünster, 29–50 Müller-Wille, M (1995), ‘Archäologische Untersuchungen ländlicher Siedlungen der Wikingerzeit im Umland des frühstädtischen Handelsplatzes Hedeby/Haithabu’, Acta Praehist et Arch 26/27 (1994–95), 39–56 Müller-Wille, M and Willroth, K-H (1983), ‘Zur eisenzeitlichen und frühmittelalterlichen Besiedlung von Angeln und Schwansen’, Offa 40, 275–320 Radtke, C (1999), ‘Haiðaby’, in Reallexikon der Germanischen Altertumskunde 13, Berlin/New York, 362–381 Randsborg, K (1980), The Viking Age in Denmark, London Skaare, K (1976), Coins and coinage in Viking-Age Norway, Oslo/Bergen/Tromsö Steuer, H (1974), Die Südsiedlung von Haithabu. Studien zur frühmittelalterlichen Keramik im Nordseeküstenbereich und in Schleswig-Holstein (Ausgrabungen in Haithabu 6), Neumünster Steuer, H (1984), ‘Soziale Gliederung der Bevölkerung von Haithabu nach archäologischen Quellen’, in H Jankuhn, K Schietzel and H Reichstein (eds), Archäologische und naturwissenschaftliche Untersuchungen an ländlichen und frühstädtischen Siedlungen in deutschen Küstengebieten vom 5. Jahrhundert v. Chr. bis zum 11. Jahrhundert n. Chr. Bd 2, Handelsplätze des frühen und hohen Mittelalters, Weinheim, 339–366 Steuer, H (1987), ‘Der Handel der Wikingerzeit zwischen Nord- und Westeuropa aufgrund archäologischer Zeugnisse’, in K Düwel, H Jankuhn, H Siems and D Timpe (eds), Untersuchungen zu Handel und Verkehr der vor- und frühgeschichtlichen Zeit in Mittel- und Nordeuropa. Teil IV, Der Handel der Karolinger- und Wikingerzeit. Bericht über die Kolloquien der Kommission für die Altertumskunde Mittel- u. Nordeuropas in den Jahren 1980 bis 1983 (Abhandl Akad Wiss Göttingen, Phil-Hist Kl Dritte F, no. 156), Göttingen, 113–197 Steuer, H, Stern, W and Goldenberg, G (2002), ‘Der Wechsel von der Münzgeld- zur Gewichtsgeldwirtschaft in Haithabu um 900 und die Herkunft des Münzsilbers im 9. und 10. Jahrhundert’, in K Brandt, M Müller-Wille and C Radtke (eds), Haithabu und die frühe Stadtentwicklung im nördlichen Europa (Schriften des Archäologischen Landesmuseums 8), Schleswig, 133–167 Stoess, Chr (1994), ‘Die Münzen’, in E Wamers (ed), Die mittelalterlichen Lesefunde aus der Löhrstrasse (Baustelle Hilton II) in Mainz (Mainzer Archäologische Schriften 1), 177–190 Wiechmann, R (1996), Edelmetalldepots der Wikingerzeit in Schleswig-Holstein. Vom ‘Ringbrecher’ zur Münzwirtschaft (Offa-Bücher 77), Neumünster Wiechmann, R (1998), ‘Souvenirs aus England? Zwei northumbrische “Stycas” gefunden in Schleswig-Holstein’, in A Wesse (ed), Studien zur Archäologie des Ostseeraumes. Von der Eisenzeit zum Mittelalter. Festschrift für Michael Müller-Wille, Neumünster, 453–460 Wiechmann, R (1999), ‘Haiðaby – Münzprägung’, in Reallexikon der Germanischen Altertumskunde 13, Berlin/New York, 382–384 Wiechmann, R (forthcoming), ‘Haithabu und sein Hinterland – ein lokaler numismatischer Raum? Münzen und Münzfunde aus Haithabu (bis zum Jahr 2002)’, Berichte über die Ausgrabungen in Haithabu Willemsen, A (2004), Wikinger am Rhein, 800–1000, Theiss, Bonn/Stuttgart Willroth, K-H (1992), Untersuchungen zur Besiedlungsgeschichte der Landschaften Angeln und Schwansen von der älteren Bronzezeit bis zum frühen Mittelalter. Eine Studie zur Chronologie, Chorologie und Siedlungskunde. Siedlungsarchäologische Untersuchungen in Angeln und Schwansen 1 (Offa-Bücher 72), Neumünster


The Evidence of Pecking on Coins from the Cuerdale Hoard: Summary Version Marion M Archibald

My contribution to the symposium was a preliminary account of an on-going study of the pecking on the coins associated with the artefacts in the Viking silver treasure from Cuerdale, Lancashire, concealed in the first decade of the tenth century and found on 15 May 1840. This study is being undertaken for inclusion in the volume on the hoard being prepared for the Trustees of the British Museum by Professor James GrahamCampbell, so only this brief summary of the paper is included here. I have already published a general survey of ‘Pecking and bending: the evidence of British finds’, including Cuerdale, in K Jonsson and B Malmer (eds), Sigtuna Papers. Proceedings of the Sigtuna Symposium on Viking-Age Coinage, 1–4 June 1989 (Stockholm/London, 1990), 11–24. The quality of precious metal can be tested in a number of ways. Pecking is a specific form of testing in which a knife-point is inserted shallowly into the surface and the degree of resistance experienced from the metal alerts the tester to its fineness; a small diagnostic sprue of metal remains proud of the original surface (Figure 4.1). This technique is associated with the Vikings and is best known numismatically from its intensive use on coins in Scandinavian hoards of the later tenth and eleventh centuries. In Cuerdale the numismatic element represents about a quarter of the bullion value of the contents and comprised somewhere between 7,250 and 7,500 coins. Unlike AngloSaxon hoards of the period, which for the most part comprise coins of just the type currently being issued, this Viking treasure includes an exceptional number of issues. Most of the types of the later ninth-century coinage in England, including the earliest issues of the Viking invaders, are represented and there are also important Carolingian and Kufic groups as well as a few coins from other series. The hoard thus presents the opportunity, unique in England, to study the pattern of pecking on a number of successive types and on contemporary issues across a wide geographical spectrum. The original high total would seem to offer an unusually secure basis for statistical comparisons but the hoard was not preserved intact. Painstaking work, particularly by the late C E Blunt, has restored some of the more distinctive strays to the corpus, but the provenance of many commoner coins almost certainly from the hoard cannot be proved. The dispersal of the material would have been less serious if the hoard had been more evenly divided among its constituent series, but some types remain represented by an unsatisfactorily low numbers of coins. To date, this study has been based on about 49



1,500 coins but as further examples, particularly of the scarcer groups, are still being added to the material specific figures are not quoted here. As in the case of the later multi-type hoards from Scandinavia, there is a broad correlation between the date of the coins and the extent of pecking – the earlier the coins, the greater the number of pecks. On the Cuerdale coins, they rarely exceed 10, are more usually between one and five, and many are not pecked at all. The earliest Anglo-Saxon group in the hoard, coins of the Cross and Lozenge type of Alfred and Ceolwulf II of Mercia of the mid-870s, is the most heavily pecked and the following types are progressively less so until the latest coins of Edward the Elder of the first decade of the tenth century bear only the occasional peck (Figure 4.1). The English coins were thus reaching the Vikings, who pecked them successively, as each new type replaced the previous one, and not all together at some later date. It is not, however, simply a matter of reading off the number of pecks against a timescale; more complex chronological, geographical and organisational factors are also involved. For example, where regional groupings can be distinguished among the AngloSaxon issues (as in the case of the main Two-Line type of Alfred), the official English coins struck in areas closer to the Danelaw border are more heavily pecked than those struck elsewhere in the West Saxon kingdom. Geographical position may not be the only determining factor here, for imbalances in the relative levels of output at various West Saxon mints over the lifetime of the type may also be relevant. Even some of the earliest

Figure 4.1 A selection of Cuerdale coins to demonstrate the progressive extent of the pecking with age (Š Trustees of the British Museum). Top row: Alfred London Monogram (fairly heavily pecked, bent); Alfred Two-Line type (less pecked, also bent) and Edward the Elder (no pecks, flat). Bottom row: Athelstan (Guthrum) of East Anglia (fairly heavily pecked, less bent than the London Monogram); St Edmund Memorial (less pecked) and Cnut/Cunnetti (no pecks, flat).



coins have no pecks and it is clear that not all coins, but only a representative selection, were pecked each time they were checked. The positioning of some pecks shows that coins were occasionally pecked more than once at the same time in order to double-check the feel of the silver. Coins of aberrant style often display a greater number of pecks than their more normal contemporaries suggesting that anything unusual or suspicious in appearance was specifically and repeatedly singled out for special attention. Older types may occasionally be more heavily pecked, not because they reached Viking hands earlier, but because they were among a larger parcel of later coins from which they were picked out for testing because they were visibly different. It is often difficult to determine the relative influence of these various factors in a given group of material. None the less, all of this suggests that the pecking process involved a considerable level of sophistication and was likely, for the most part, to have been carried out by experts rather than by every Viking for himself. By far the largest part of the hoard is made up of the English Viking coinages, those of East Anglia, the Midlands and York. These display the same age-related pattern of pecking found among the Anglo-Saxon coins. In East Anglia the earliest issues in the name of Athelstan (Guthrum), struck in the 880s, are heavily pecked whereas later coins invoking the martyred St Edmund, produced from c 895, are much less so (Figure 4.1). Among the numerous coins of the latter series it is possible to distinguish the earlier issues, and these have a greater number of pecks than those struck shortly before the deposition of the hoard. This suggests that these coins were also reaching the place where pecking was being carried out successively as they were produced, although the caveat just mentioned should be borne in mind. The degree of pecking on the successive copies of West Saxon types produced in the Viking Midlands is similarly greatest on those struck first, and the copies are systematically more heavily pecked than their West Saxon prototypes. The large parcel of coins of Viking York in the hoard display the same broad daterelated pattern of pecking noted elsewhere. The earliest coins struck in the name of Siefred, from c 895, have more pecks than the coins in the name of Cnut, the latest series of which (the Cnut/Cunnetti type) show only the occasional peck (Figure 4.1). This probably indicates that the York coins were also reaching the pecking area successively, but the possibility must remain open that a few earlier survivors may have been included in a later consignment with the result mentioned above. The coins of the Mirabilia/Dns Ds O Rex group are significantly more pecked than even the early Siefreds, but their place later in the York series is established by the die-linked chain connecting them with other types. Here the explanation seems inescapable that these coins had raised suspicion, in fact unjustifiably, simply because of their unusual appearance and lack of a regal or mint name. The Carolingian coins in the hoard are divided into two groups: the larger originating in Aquitaine and containing mainly local issues, and the smaller from the Netherlands, most of which had arrived there via the Rhine corridor from the German part of the Empire and from Italy. The bulk of the first group are immobilised coins in the name of ‘Emperor Charles’ struck from the 880s onwards into the reign of Charles the Simple, from 897, with a small number of coins of Odo, 888–97. The pecking pattern of the Charles and Odo coins is broadly similar and suggests that most coins of both series had come into Viking possession at the same time, supporting the view that they were loot acquired in the raid on Aquitaine in 898. The coins of Odo have a greater number of pecks



than contemporary issues from England in the hoard, no doubt because they had changed Viking hands more often before reaching their common destination. The Netherlands group includes coins struck in Germany for Louis the Child, after 899, and in Italy for Lambert, 892–98, and for Berengar I, in the period 898–900. Although the numbers of coins involved here are lower, the pattern is clear. They are less pecked than those in the Aquitaine group, suggesting that they had reached the Vikings more recently and had not seen so many changes in ownership. They too are more heavily pecked than the contemporary English coins in the hoard, both West Saxon and English Viking, all of which by this period show only the occasional peck. There are three native Scandinavian coins in the hoard, two of them fragmentary survivors of the earliest Danish coinage struck from c 825 (see Malmer, Chapter 2, in this volume) and one a later coin of c 900. The degree of pecking appears to be no greater than that found on the English coins in the hoard. The more plentiful Kufic dirhems display the usual variety of other test marks acquired on their journey westwards from the Kaliphate, but the classic pecks present, numbering from six down to possibly one on the most recent coin, are again on a par with the English issues. None of the pecking on these issues which had crossed the North Sea therefore requires to have taken place in Scandinavia. This interpretation is supported by finds from Scandinavia, for example, West European silver pennies had entered Norway earlier, but there is no evidence of classic pecking there until the latter part of the tenth century. This raises the question of where the pecking of the Cuerdale coins was carried out. This is a complex subject, which will be dealt with at greater length in the hoard volume, but may be outlined here. The earlier Lunettes pennies produced in Mercia and Wessex before c 875 and absent from Cuerdale are not found pecked, even in undoubted Viking contexts such as those at Repton. The baseness of the coins complicates arguments here, but there are also no classic pecks present on the contents of the Croydon, Surrey, hoard of c 872. Identified as a Viking assemblage, it included ingots, hack-silver and fine silver coins as well as the base Lunettes. The earliest evidence for pecking on coins comes from Cuerdale itself during the issue of the first substantive type of the restored fine-silver coinage in the later 870s. Although there are exceptions, it is a general rule that people do not systematically test their own coinage and issuing authorities take steps to prevent the practice. On this point isolated losses are crucial: only the very occasional coin struck during the period covered by the Cuerdale hoard is found pecked. This would suggest that systematic pecking was not practised in Viking-held areas where the invaders were issuing their own coins. The pecked coins in the Ashdon hoard from the western borders of Essex, deposited c 895, do not disprove this since the political situation in the central and eastern Danelaw at the time is uncertain, and the velocity of coin-movement back and forth across ethnic and political boundaries could account for their presence. Ashdon was also deposited during the period between the death of Athelstan (Guthrum), in 890, and the beginning of the St Edmund coinage (absent from the hoard), in c 895, when no coins were apparently being struck in East Anglia. The area of pecking seems to be centred on Viking Dublin. Pecking appears to have been a Viking rather than an Irish invention as there are no pecks on the earlier ninthcentury English and Carolingian coins from the Delgany, Co Wicklow, hoard of c 830. There is virtually no evidence of isolated finds of the period from north-west England (the source of the Cuerdale hoard), but no coins were then struck in that area, so pecking



may have been adopted there by Viking groups with Dublin connections and the practice taken by them further afield in England. We have seen above that Scandinavia is not the source of the practice as pecking occurs there only later. The Cuerdale coins are thus more likely to have been pecked in Ireland and in non-coin-producing areas of England with Dublin Viking connections. This conclusion would fit well with the evidence of the artefacts in the hoard which Professor Graham-Campbell has established are mostly of Irish origin.

BIBLIOGRAPHY Archibald, M M (1990), ‘Pecking and bending: the evidence of British finds’, in K Jonsson and B Malmer (eds), Sigtuna Papers. Proceedings of the Sigtuna Symposium on Viking-Age Coinage, 1–4 June 1989 (CNS IX–XI, New Series 6), Stockholm and London, 11–24 Archibald, M M (1992), ‘Dating Cuerdale: the evidence of the coins’, in J Graham-Campbell (ed), Viking Treasure from the North West: the Cuerdale Hoard in its Context (National Museums and Galleries on Merseyside Occasional Papers, Liverpool Museum no. 5), Liverpool, 15–20 Blunt, C E and Archibald, M M (forthcoming), ‘The Cuerdale coins’, in J Graham-Campbell (ed), The Cuerdale Hoard and related Viking-Age Gold and Silver, from Britain and Ireland, in the British Museum, British Museum, London Graham-Campbell, J (1995), ‘Nicking and pecking: an excursus’, in J Graham-Campbell, The Viking-Age Gold and Silver of Scotland (AD 850–1100), National Museums of Scotland, Edinburgh, 33 Graham-Campbell, J (2002), ‘The dual economy of the Danelaw: the Howard Linecar Memorial Lecture 2001’, BNJ 71 (2001), 49–59 Kilger, C (2004), ‘Vad säger egentligen pecks och böjningar? Tankar kring metodiska och teoretiska frågor angående sekundära individuella data och silvernateringen under vikingatiden’, NNUM 1–2, 3–12 Kilger, C (2006), ‘Silver handling traditions during the Viking Age: some observations and thoughts on the phenomenon of pecking and bending’, in B Cook and G Williams (eds), Coinage and History in the North Sea World, c 500–1200 AD, Brill, Leiden, 449–466 Malmer, B (1981), ‘Methodological problems in editing and evaluating the Swedish Viking-age coin hoards’, in M A S Blackburn and D M Metcalf (eds), Viking-Age Coinage in the Northern Lands (BAR, British Series 122), Oxford, 391–403


Gold in England During the ‘Age of Silver’ (Eighth–Eleventh Centuries) Mark Blackburn

And 2000 mancuses of gold are to be taken and minted into mancuses; … and they are to [be distributed] throughout the bishoprics … And I grant to the Archbishop 200 mancuses of gold … ; and to each of my diocesan bishops 120 mancuses of gold; and to each of my ealdormen 120 mancuses of gold; and to each appointed seneschal and each appointed keeper of the wardrobe, and each appointed butler, 80 mancuses of gold. And to each of my mass-priests whom I have put in charge of my relics 50 mancuses of gold and five pounds in pence; and to each of the other priests five pounds. And to every appointed steward 30 mancuses of gold, and [the same] to every man in priest’s [orders] who has been employed since I came to the throne, and [the same] to each of those who are in my household, in whatever office he is employed, unless he be little connected with the royal dwellings [?] … (Will of King Eadred; Whitelock 1979, no. 107)

King Eadred (951–55) clearly believed that gold was an appropriate reward to leave to his noblemen, bishops and personal servants, and through its wider distribution it might be a means of redeeming his soul. By any reckoning his executors would have needed at least 5,000 mancuses1 of gold to pay the legacies in his will – 5,000 mancuses would have weighed about 20 kg and been worth more than £1.5 million in today’s money.2 This was in addition to other legacies, mainly for the benefit of people generally, of at least 2,300 pounds of silver, equivalent to some £5 million in modern money, and of various land estates to the Church and to Eadred’s mother. Royal distribution of gold seems to have been customary, for, in the only other surviving will of an Anglo-Saxon king, King Alfred (871–99) bequeathed some 1,700 mancuses of gold at the rate of 100 mancuses each to his bishops and ealdormen, while making other legacies of some 1,800 pounds of silver and of many land estates (Whitelock 1979, no. 96). Touchingly, he says that he is not certain whether he has so much money, but he thinks he has, and if indeed he has more then it is to be shared among those to whom he had bequeathed money. This was, then, intended to be a reasonable estimate of the amount Alfred possessed when the will was drawn up, probably in the 880s (Keynes and Lapidge 1983, 173). Asser tells us that Alfred’s father Æthelwulf (839–58), after leaving the kingdom and his own inheritance between his sons and other family, had divided his money between the needs of his soul, his sons and his nobles (Whitelock 1979, no. 7, ch. 16), just as Alfred did. This chapter surveys the evidence for the availability and use of gold during the eighth to eleventh centuries, when the coinage in circulation was predominantly in silver and 55



when other metalwork in gold is notably rare. Its focus is England, but in order to provide a wider context, the finds of gold coins and artefacts from the rest of the British Isles, Scandinavia and Western Europe are taken into account. In particular, it compares the occurrence of gold artefacts in Scandinavian areas of the British Isles with those in Scandinavia itself. The appendices provide inventories of finds of gold coins from the British Isles 700–1200, of gold coins struck in England and of finds of gold ingots and hack-gold, a category of material that was almost unknown from England until the mid1990s. The chapter concentrates on gold as a means of exchange, considering its decorative or social functions mainly as alternative evidence for the availability gold in society. The rather plentiful, if at times ambiguous, written sources for the use of gold in England at first sight seem at odds with the paucity of finds, until one recognises that the pattern of loss of special purpose money and high-value artefacts was quite different from other types in daily use. In assessing the role of gold, therefore, the evidence needs to be weighed carefully, and while I have attempted to do this for early medieval Britain, there is a need for further comparative research spanning a range of periods and places.

THE WRITTEN SOURCES We have seen how Anglo-Saxon kings in their wills used gold to reward the elite in their household and society. Nobles and bishops, in return, bequeathed gold to the king as part of their heriot or death-due, as in the case of Theodred, bishop of London (d 951 x 953), who bequeathed to King Eadred 200 mancuses of gold, two silver cups and the customary horses and armour, as well as two estates (Whitelock 1979, no. 106). Ealdorman Æthelmær in 982 left King Æthelred II (978–1016) ‘four armlets of 300 mancuses’ as well as horses and armour (Whitelock 1930, no. X). Under King Cnut (1016–35) heriots were codified at rates that included 200 mancuses of gold for ealdormen and normally 50 for king’s thegns (II Cnut 71; Whitelock 1979, no. 49), although these had already applied under Æthelred II (Brooks 1978, 85–90). Brooks argues that gold arm-rings were part of a nobleman’s battle gear, to identify his status on the battlefield, and hence they became part of the heriot. One of the most lavish gifts made by a nobleman must be that from Earl Godwine to King Harthacnut (1035–37, 1040–42), to recompense for his involvement in the murder of the ætheling Alfred (Florence of Worcester, sub anno 1040; Whitelock 1979, no. 10). This consisted of a well equipped and ornate galley with 80 hand-picked soldiers, each of whom wore two gold arm-rings weighing 16 ounces (about 110 mancuses) and had a helmet, sword and shield with gilt decoration. If the account is reliable, the gold arm-rings alone would have been equivalent to some 9,000 mancuses and weighed perhaps 38 kg, substantially more than the amount of gold Eadred bequeathed in his will. Nor was gold the exclusive preserve of royalty. A number of other wills include legacies of gold, and not only left to people of high status. Wynflæd in the mid-tenth century directed that Wulfflæd and Æthelgifu were to be supplied with nun’s vestments and the bequest was to be supplemented with gold so that each of them should have at least 60 pennyworth, that is, two mancuses (Whitelock 1930, no. III), and in the late tenth or early eleventh century Wulfwaru left to her four servants ‘a band of twenty mancuses of gold’ (Whitelock 1930, no. XXI).



The Danes were also keen on gold. The Codex Aureus, now in Stockholm, has an inscription describing how, in the ninth century, Ealdorman Alfred and his wife had paid a ransom to recover it from the heathen army ‘with our pure money, that was with pure gold’, and they had then presented the book to Christ Church, Canterbury (Whitelock 1979, no. 98). In 872, after the Danes had occupied London, Wærferth, bishop of Worcester, sold an estate for ‘20 mancuses of tested gold’ to pay the tribute being raised (Whitelock 1979, no. 94). Before the Battle of Maldon in 991 the Danish army had asked for tribute in gold (Whitelock 1979, no. 10), and the tribute paid in 994 of 22,000 pounds was ‘in gold and silver’ (II Æthelred 7.2; Whitelock 1979, no. 42). To help raise this Archbishop Sigeric sold land to the bishop of Dorchester, receiving ‘90 pounds of refined silver and 200 mancuses of the purest gold’, and on another occasion the king sold land in Oxfordshire for ‘a payment of one pound of silver in purest gold, for paying the tribute’ (Sawyer 1968, nos 882 and 943; Keynes 1991, 101–102). There are many references in charters and other documents to gold being used for the purchase of land from the late eighth century onwards; prior to that the payment was usually unspecified. As with most legacies the amount is normally expressed in terms of mancuses, though it can take the form of solidi, shekels (sicli), pounds or objects such as rings or vessels. Peter Sawyer (pers comm) has pointed out that in a number of cases it is specified that the payment was in ‘pure gold’ or ‘refined gold’. These can be found intermittently between the late eighth and early eleventh century, but there is a marked increase from the 840s to 880s and 920s to 970s, which may be significant. Both are periods during which the Anglo-Saxon silver coinage was debased, although references to ‘pure silver’ are rarely found in charters. But the mid-ninth century also saw the arrival of debased imitations of Louis the Pious solidi (discussed pp 67–73), which could have prompted the more frequent stipulations that payment be measured in fine gold.

THE MANCUS The origin and use of the word ‘mancus’ (Latin mancosus, mancusus) has been much discussed, most recently and thoroughly by McCormick (2001, 323–342) and Ilisch (2004). The view that it represented a debased Byzantine solidus, initially espoused by Grierson (1954), was later abandoned by him and others in favour of an earlier interpretation identifying it with the standard Arabic gold coin, the dinar (Akerman 1842; Allan 1914; Grierson 1979, Addenda et corrigenda, 2–4; Grierson and Blackburn 1986, 327). The term was first used in Carolingian Italy in the later eighth century, occurring in a number of documents from 778 onwards. A series of documents from the abbey of Farfa, near Rome, suggest that in the early 780s, for the prescription of fines in penalty clauses, mancuses replaced payments that had previously been valued in ‘gold shillings of Lucca’ (ie Lombardic and early Carolingian gold tremisses), at just the period when these tremisses ceased to be produced: in a monetary reform of 781 the tremisses were replaced by a new coinage of Carolingian silver deniers (McCormick 2001, 326–330). From Italy the term passed into use in England, where it first occurs in a charter of 799 (Sawyer 1968, no. 23), though it was probably used earlier since a papal letter of 798 refers to a pledge given by King Offa in 786 to send 365 mancuses a year to the pope (Whitelock 1979, no. 205). Here too it may initially have been used to represent an Arabic dinar or its equivalent weight of gold (4.25 g). Subsequently its use was extended to cover



other types of gold coin, a unit of weight for gold bullion or ornaments and a unit of value equivalent to 30 silver pennies (Lyon 1969, 207–209). The multiple meanings of the term is ably demonstrated by the first line of the passage from King Eadred’s will quoted at the beginning of this chapter, specifying that ‘2,000 mancuses of gold’ [by weight] were to be ‘minted into mancuses’ [coins]. Whether as a unit of weight it remained constant throughout the Anglo-Saxon period, or it fluctuated according to the weight of the silver penny and the accepted gold:silver ratio is debatable. The evidence of the surviving English gold coins, considered below, would suggest the latter, but it is likely that two different standards were used for the unit of weight and for the coins. The mancus as a unit of weight may well have been fixed, subject only to the occasional reform of weights and measures, while the weight of the coin known as a mancus had to fluctuate if its value equivalent to 30 silver pennies was to be maintained. The analogy with silver units is that the pound, divided into 240 pennyweights, had a constant value as a unit of weight, while the weight at which the penny as a coin was struck varied considerably. It follows that the 2,000 mancuses of gold (by weight) left by King Eadred could have been minted into rather more, or less, than 2,000 mancuses as coins. The value of a mancus of gold would often have been different from the value of a minted mancus or a mancus as a unit of account, both equal to 30 silver pennies. The penny, of course, circulated at a premium over the intrinsic value of its metal, perhaps by as much as 10–15 per cent, and the mancus as a coin would likewise have benefited from this overvalue. Various permutations have been discussed and calculations made of the theoretical weight standards and ratios, particularly by Chadwick (1905), Lyon (1969) and Nightingale (1984), but it would be fair to say that evidence is not strong enough to validate any single model. What is clear is that, despite the common origin of the term mancus, we can expect that over a period of time the value and weight of the English mancus would have diverged from that on the Continent, as did the value of the penny/denier. Since the term mancus had such fundamentally different meanings, it can be difficult to determine which one was intended in any particular document. The phrase ‘a gold ring containing 75 mancuses’ in a charter of 822 (Sawyer 1968, no. 186; Whitelock 1979, no. 83) looks like a weight, while in other cases describing objects composed of different materials a value may be intended, for example, ‘the æstel [a reading pointer or bookmark] worth fifty mancuses’ that King Alfred planned to distribute to each bishopric with copies of his translation of Gregory’s Pastoral Care (Keynes and Lapidge 1983, 126, 205); ‘the sword which King Edmund gave me, which was worth 120 mancuses of gold and had four pounds of silver on the sheath’ (Will of Ælfgar; Whitelock 1930, no. II); ‘her old filigree brooch which is worth six mancuses’ (Will of Wynflæd; Whitelock 1930, no. III). However, in most cases the payment or gift is merely described as so many ‘mancuses of gold’ in a context that suggests that gold actually changed hands, but whether in the form of weighed bullion, ornaments or struck coin we cannot tell. The term ‘mancus’ is usually reserved for gold, while silver is weighed in ‘pounds’, but occasionally ‘mancus’ is used as a unit of account for silver, as in the will of Bishop Ælfwold of Crediton (997–1016) in which he leaves several legacies in ‘mancuses of gold’, but to ‘Ælfflæd the offestre five mancuses of pennies’ (Whitelock 1979, no. 122). In other cases too, sums of money expressed in terms of mancuses of gold, may well have been satisfied in silver. In theory the surviving artefacts and coins could show the extent to which gold bullion and coinage were available. Hinton (1975; 1978) surveyed the extant Anglo-Saxon gold and silver ornamental metalwork, though the material has grown significantly in recent



years, and the intention of this chapter is to lay that alongside the finds of coins and Scandinavian metalwork to see if a consistent pattern emerges. Still needed, however, is a systematic survey of the plentiful documentary evidence for the monetary use of gold in Anglo-Saxon England, and the writer hopes that others better equipped than him may be inspired to undertake such a study in the future.

THE SCARCITY OF GOLD COINAGE IN WESTERN AND NORTHERN EUROPE AFTER THE SEVENTH CENTURY Gold coinage dominated monetary circulation in Western Europe between the fifth and seventh centuries (Grierson and Blackburn 1986). Gold coins based on the Roman solidus and its third, the tremissis, were standard currency, and little in the way of silver or bronze coinage was minted outside Italy. From the late sixth or early seventh century, the gold issues began to be debased by the addition of silver to the alloy, and this progressed, albeit at different rates, in every part of Western Europe. By the 670s in Francia and England the gold content had fallen to less than 25 per cent, and reforms were carried out replacing these base gold coinages with new silver ones. In due course Spain and Italy followed suit, and for the next four or five centuries the currencies of Western Europe were essentially in silver, gold appearing only exceptionally. Those exceptional issues will be discussed further below. A broadly similar pattern is found in Northern and Central Europe, where although coinage was not produced locally, imported coins and other precious metal were used for social, ritual and commercial functions. Thus in Scandinavia Roman and Byzantine gold solidi and locally produced bracteates were used and hoarded in the fifth and sixth centuries, while the next wave of imported coinage starting in the late eighth or early ninth century was of silver dirhems from the Islamic East. Fine gold coins continued to be struck in Byzantium and parts of the Arab world during Europe’s silver phase from the eighth to twelfth centuries, but they have left little trace in the find evidence in Europe. More than 40 years ago Duplessy (1956) reviewed the finds of Arabic coins in Western Europe and showed that the gold coins fell into two groups. In the earlier one, from the eighth and ninth centuries, Arabic gold coins were represented in two hoards from Italy, two from France and one sketchy find from England. This last amounted to two Islamic coins found on the beach at Eastbourne before 1845, one of which was identified as an Umayyad dinar of the first half of the eighth century (Appendix A1). Since 1956 only one further find of this period has been recorded from England, an Abbasid dinar of AH 147 (AD 764/65) found at Wickhampton, Norfolk, in 1988 (A2). A new listing of Arabic coins found in the West down to AD 900 (McCormick 2001, appendix 3) records two additional single finds from France, a substantial hoard of 400 dinars apparently found in Venice in 1592, a hoard from Sardinia, one from Croatia and a single-find from Serbia; while Ilisch has added a single-find of an imitative forged dinar (in gilt-copper alloy) from Lippe, Germany (Ilisch 2004, 96, II,3). In Duplessy’s corpus there was a complete absence of finds of Arabic gold from Europe from the tenth and eleventh centuries. From his second period (twelfth–thirteenth centuries), he cited seven finds from western France, two from the Low Countries, one from Germany and one English find of two Almoravid dinars of the twelfth century found at London (A27–28). Since 1956 only one more coin of this latter period has been



discovered in England, a twelfth-century Almoravid dinar from Oxford (A26). However, three new English finds bridge the gap between the ninth and twelfth centuries: Fatimid quarter-dinars of the later tenth and mid-eleventh centuries from the Isle of Wight and St Leonards (A21 and A24) and a late eleventh-century Almoravid dinar from near York (A25). The pattern for Arabic gold coins from England is thus now more of a thin scatter across the whole Anglo-Saxon and Norman period, though they remain extremely rare. The record of continental finds after 900 has not been systematically up-dated since Duplessy. In Scandinavia the number of Arabic gold coins that has been found is tiny in comparison with the copious finds of silver dirhems. The remarkable Hoen hoard from southern Norway, deposited in the later ninth century, contained some 2.5 kg of gold in the form of neck-rings and other ornaments, including 20 coins that had all been converted into pendants. Nine of these coins are Abbasid dinars, and the remainder are a mixture of late Roman, Byzantine, Carolingian and Anglo-Saxon coins. It has generally been thought that these coins were assembled in Francia in the mid-ninth century and exported to Norway as a group, but a recent reassessment based on a study of their loop attachments has suggested that they arrived piecemeal, and were mounted singly or in small groups (Blackburn, forthcoming). The dinars may well have been drawn from local Scandinavian sources, rather than directly from Western Europe, and the handful of other Arabic gold coins from Scandinavian finds supports this view. One of these is from the later eighth century – an imitative dinar, perhaps of West European origin – while five others belong to the period 871–932, after which there is a gap until the twelfth century from which nine Almoravid dinars have occurred in Sweden, probably all found on Gotland (Welin 1964). All of these coins could have arrived in Scandinavia via Western Europe, but a new find from Bornholm – the first Arabic gold coin found on Danish soil – provides key evidence as to how some Arabic gold coins followed the routes of the Arabic silver, across Russia and the Baltic. This find, from Vejrmøllegård, Bornholm, discovered in 2001, is a crude imitation which Moesgaard and Rispling (2002) attribute to the Khazars, who from the region of the Lower Don and Volga controlled the trade between Russia and the central caliphate. They date the coin to the late ninth century, comparing it with Khazar imitations of silver dirhems found in Russian and Scandinavian hoards. It appears, then, that in Scandinavia during the Viking Age there may have been a very small stock of Arabic gold coins that was used for special purposes and not hoarded with silver, some of which arrived via Western Europe and some via Russia and the Baltic. As for Byzantine gold coins none has been found in England after the seventh century, yet Cook (1999a) has demonstrated from a detailed survey of documentary sources that in the later twelfth and thirteenth centuries ‘bezants’ (ie Byzantine hyperpyra) were widely available and used in a range of transactions. This example of find evidence failing to reflect a documented gold currency is not an isolated one. Also in the thirteenth century, there are many references in Italian accounts and in French sources to the presence of Byzantine hyperpyra, at a time when West European participation in the affairs of Greece and the Latin Orient were at their height, yet none of these coins has been found in any West European country (Baker 2002, I, 141–145). In England during the first half of the fourteenth century, before the introduction of a regular English gold coinage, documents show that foreign gold coins, particularly the florin of Florence, were used extensively, but again these have left little trace among the English finds (Cook 1999b, 255–260). To take another, more modern example, in seventeenth- and eighteenth-century Ireland foreign



gold coinage made up a substantial proportion of the gold currency, but although many weights for weighing Portuguese and other foreign gold coins have been found, not a single gold coin – foreign or British – is known to have been discovered in a coin hoard or as a single find (Robert Heslip, pers comm). As Cook points out (1999a, 257), it is salutary to note the potential limitations of find evidence in assessing the role of high value money. Conversely, when there is even a modest distribution of high-value gold coinage, as with the Arabic coins in Western Europe, this may represent a substantial volume of such coinage in circulation. It is with this principle in mind that we turn to consider the surviving gold coins produced in England.

ENGLISH GOLD COINAGE, 675–1250 From the later seventh to mid-thirteenth centuries English coinage was essentially monometallic in silver, as also was French. However, in both countries some exceptional coins struck in gold survive from this period. There are only eight English gold coins with meaningful inscriptions from this 600-year period (Appendix B); a second category of anonymous imitative coins will be considered below. Of the eight coins, four date from the late eighth or early ninth century, one from the mid-ninth century, another is from the early tenth century, and two are from the eleventh century. Since 1909 only one new specimen has been found – the Cenwulf mancus (B4), in 2001 – and this despite the large quantity of metal-detector finds made over the past three decades, indicating that such gold coins will always be extremely rare. What is difficult to judge is whether these coins are survivors of a few very exceptional issues, or whether gold coins were struck more frequently but rarely lost because of their high value. The earliest and best known of these coins is the dinar of King Offa (757–96), which is a copy of a gold dinar of the Abbasid caliph Al-Mansur dated AH 157 (AD 773–74), with the addition of the words OFFA REX in Roman letters in the centre of the coin (Appendix B1). It first surfaced in Rome before 1841 and is presumably a local Italian find. This extraordinary piece connects remarkably well with the documentary evidence, dating as it does from just the period when the use of the term ‘mancus’ for the Abbasid dinar first appears in documents in Italy and England (see, p 57). The Italian provenance for this English coin underlines the monetary connection existing between these two countries and their common interest in the dinar. Since its first discovery, the coin has been associated with Offa’s promise in 786 to pay an annual render to the Holy See of 365 mancuses (Longpérier 1841), and it has even been argued that the coin was produced specifically for that purpose (Carlyon-Britton 1908, 65–66), although this is an unnecessary supposition. If dinars had recently become recognised as an international currency appropriate for certain transactions (McCormick 2001, 379–384), but were in short supply, it is natural that the English should want to strike some of their own, carrying over from the silver coinage the practice of including Offa’s name and title on all his coins. It is likely that coins of this type were included in Offa’s payments to Rome, but they may also have enjoyed some wider use and circulation. As Archibald (1990, 12) has pointed out, this coin is somewhat battered and repeatedly test-marked, showing that it has seen considerable circulation. A few other imitations of Abbasid dinars have been identified as of West European origin – Anglo-Saxon or Carolingian (Allan 1914; Welin 1964). None has western inscriptions, but one has three crosses inserted into the design



(Figure 5.5, B1A; Lowick 1973; Webster and Backhouse 1991, no. 148b: 3.89 g, c 94 per cent gold). At one time this group was thought to be quite extensive, but Lowick has reasonably argued for a much more limited series in view of the small number of dinar finds from Western Europe. More recently, Ilisch (2004) has identified other groups of imitations, mainly with the year AH 157, that he thinks were probably struck in the Carolingian Alpine region. To England he assigns just the Offa dinar and Lowick’s piece with three crosses in the field. However, the existence of even a few such derivatives reinforces the view that dinars were used fairly regularly for certain transactions in Western Europe, particularly in Italy, during the later eighth century. Subsequently, the English gold coins took a variety of different forms and the meaning of the term mancus appears to have expanded to cover at least some of these. The second inscribed coin of this period (B2) was known in the early seventeenth century when it was in the collection of Sir Robert Cotton and illustrated by Speed (1611), but it was then forgotten until 1961 when it turned up in a London saleroom. The coin was recognised by Christopher Blunt, who bought it and presented it to the British Museum. It is a curious piece, with designs copied from a gold aureus of Augustus of the mint of Lugdunum (Lyons) and with no king’s name, but the name of the moneyer Pændræd on both sides. Pændræd worked, probably in London, as a moneyer during Offa’s light coinage, and this gold coin belongs to the same period, that is, 780s or early 790s. There was clearly a lively interest in Roman coins among the London moneyers or die-cutters at this time, since a number of silver pennies of this issue have quite faithful copies of Roman busts (Blunt 1961). The choice of a rare gold aureus as the prototype for Pændræd’s mancus emphasises the sophisticated approach to coin design at London in the 780s. Another anonymous gold coin with only a moneyer’s name (A3/B3), found near Manchester in 1849, had been regarded as a late sixth- or seventh-century issue (Sutherland 1948, no. 76), until it was reattributed by Pagan (1965). Its obverse inscription appears to carry the name Ciolheard, who was a London moneyer in the heavy coinage of Offa and issues of Cenwulf/Coenwulf (796–821) and Ceolwulf I (821–23) of Mercia. Its designs combine the obverse of a fourth-century Roman gold solidus with a cross-on-steps reverse typical of seventh- and eighth-century Byzantine and Beneventan solidi. This choice of types, poor lettering and a blundered reverse legend (in which Pagan has read traces of the London mint-name), is a little surprising, for after Offa’s monetary reform of c 792 there was greater uniformity in the designs and a high standard of literacy. It suggests that, although the product of an official moneyer, this coin has not been subject to the same degree of official regulation as his silver pennies. Indeed, both the Pendræd and the Ciolheard gold coins do not claim to be royal coins – their validity is backed by the moneyer in a personal capacity. This implies that in this early period when Arabic dinars were starting to be used in Western Europe for particular transactions, the moneyers provided a means of converting other forms of gold into coins that could be used as substitutes for dinars. The Offa dinar may well be of similar status, despite carrying Offa’s name, for the moneyer in an experimental phase may have thought it more correct to put the king’s name on the coin than his own. These three earliest pieces, then, seem to belong to an early transitional phase. A dramatic new find, a wonderful mancus of Cenwulf of Mercia discovered by the Riverl Ivel at Biggleswade (Beds) in 2001 (A4/B4), provides a fourth coin of this period, and the earliest gold coin to be struck from dies closely related to the official penny



coinage. The design and inscription of the obverse are virtually the same as on the silver pennies of the Portrait/Cross-and-Wedges issue at Canterbury (Blunt, Lyon and Stewart 1963, 11 and plate II) except for the omission of the inner circle on the gold piece. The die has been cut by the same hand using exactly the same methods of composing the bust and forming the letters, so that the smallest details can be paralleled among Portrait/Cross-and-Wedges coins of Cuthred of Kent, issued c 805–07, and of Cenwulf, issued 807–c 810 (Grierson and Blackburn 1986, 288, for the dating). The mancus can thus be dated to c 805–10, and it appears to belong relatively early in the stylistic sequence, possibly during Cuthred’s reign or at the beginning of Cenwulf’s Canterbury coinage in 807. The die-cutting is very competent, but no special treatment was given to this die over the normal dies for pennies – the portrait and epigraphy are no more elaborate – and the coin’s exceptional beauty derives from the care taken in striking it and its fine preservation. It is interesting, then, that a standard penny die was not used, and it was thought necessary to prepare a new die with a subtle but deliberate difference – the omission of the inner circle. It is likely that this was intended either to prevent gilded pennies being passed off as mancuses, or to facilitate control of the mancus dies so that they could not be used to strike pennies and vice versa. The reverse is very different from any other Anglo-Saxon coin, in both its geometric daisy pattern and especially its legend (+DE VICO LVNDONIAE), for while the use of the genitive and the inclusion of the mint-name is characteristic of this Canterbury diecutter, the term vicus is unparalleled. No other coin refers to the vicus or wic (ie trading settlement) of London. The term does occur in a handful of early documentary sources, the latest being a charter of 811 referring to Lundaniae vicus (Sawyer 1968, no. 168), and it is preserved in the street name Aldwych (‘the old wic’). The significance of the archaeological and scant documentary evidence was only recognised in the mid-1980s, when it was suggested that the old walled Roman city of London was scarcely occupied before the late ninth century and that the earlier settlement was located to the west of the city on the land around the Strand, the Aldwych and Covent Garden (Biddle 1984). Since then the name ‘Londonwic’ has been adopted by many historians and archaeologists for this early trading settlement, which suffered repeated Viking raids in the mid-ninth century, and was relocated by King Alfred in 886 within the refurbished and refortified Roman city (Vince 1990). The use of DE VICO LVNDONIAE in Cenwulf’s mancus inscription is highly significant, for such official endorsement on the royal coinage emphasises the importance of the wic settlement. It also implies that in the early ninth century the ‘mint’ of London (in effect probably a series of private workshops of moneyers, rather than a central building) was located in this area rather than in the city, and if so the same may have been true of other royal officials and functions. The absence of a moneyer’s name is very unusual, but if only one moneyer were licensed to strike mancuses, perhaps under an arrangement whereby the profits went straight to the Crown, there would be no need for the moneyer to be named to hold him accountable for coins of low weight or fineness. The simple geometric daisy pattern on the reverse of the mancus is a quite novel motif in the coinage of this period and would undoubtedly have stood out. It can be found in other Anglo-Saxon contexts, for example engraved at the centre of one of the silver bowls, of east Mediterranean workmanship, from the Sutton Hoo ship burial (Evans 1986, plate 45), or in manuscript decoration such as the roundels above King David in the Vespasian Psalter, folio 30v, c 725 (Webster and Backhouse 1991, no. 153). A new variety of gold shilling of the mid-seventh century also uses a form of daisy pattern (Blackburn 2006), though there is unlikely to be a link between that and the mancus.



The care with which Cenwulf’s mancus coinage has been planned, being synchronised with the silver pennies but carefully distinguished from them, and issued at London, the main Mercian mint, while drawing upon the resources of the designer and die-cutter of the Canterbury issue, puts it in a quite different class from the previous gold mancuses. It would appear to be the product of a carefully considered policy decision implemented by the royal administration, rather than a coinage arranged by moneyers in response to an ad hoc request. If so what was the motive? Was it a special issue produced solely to make one particular royal payment or donation of political significance? Or was it intended to have wider application, an attempt to standardise the English mancus and introduce a second denomination into the monetary system, albeit one only used for certain limited purposes? There are several possible motives for such a change. Cenwulf may simply have wanted to regulate the weight and fineness and ensure high standards of production. More probably, though, he would have wished to take the right to strike mancuses away from the moneyers and impose a royal monopoly, so that he might take the profits of minting gold for the Crown. He may also have wanted to exclude foreign gold coins from circulation and insist that they be exchanged at the mint for English coins, as was the policy for silver. How far the plans went, and how far they were achieved would have been two different matters, but this coinage does have all the appearances of effecting a significant monetary reform. If so it represents a clear precedent for the introduction a few years later of gold solidi by the Carolingian king Louis the Pious (814–40). That issue of c 816–19 is analogous to Cenwulf’s mancus in many respects, with an obverse conforming to that of Louis’ silver coinage but a distinctive reverse. It was probably issued on a larger scale than Cenwulf’s mancus as some 20 specimens from several pairs of dies survive and its influence was considerable. Yet it was a short-lived experiment to produce a substantive gold coinage that was not sustained. Other Carolingian gold coins appear to be more ad hoc, including some struck from regular dies for silver deniers, as in England at a later date. The Cenwulf mancus is an appropriate precursor to the fifth gold coin, the solidus of Archbishop Wigmund of York (c 837–54?) (B5), which has been known since the eighteenth century. Like the Cenwulf coin, it is of fine quality, carefully designed and without the name of a moneyer, but its design and reverse legend follow those of the solidi of Louis the Pious, adapting Louis’ profile bust to a front-facing image of the tonsured archbishop. This extraordinary piece was struck at a time when the York mint was only producing base-metal stycas, and for this reason the possibility has been considered that the coin may have been struck at a mint in the south (Grierson and Blackburn 1986, 330). However, this is unlikely since, as we shall see, the coin does not use the weight standard of the southern English mancus, but that of the Carolingian solidus. It is surprising that the only surviving gold coin of Northumbria should be in the name of an archbishop rather than a king, for one might reasonably ask what need the Church would have had for special gold solidi since it would generally have been a recipient of gifts and payments rather than a lavish donor. However, Rollasson (Tweddle, Moulden and Logan 1999, 131–134) has argued that in the ninth century the archbishops were the dominant political and economic power in York, and while numismatists looking at the styca coinage would find it difficult to agree that the kings played a subsidiary role, the view that the archbishop of York had a wider secular function could explain the existence of this gold solidus. The remaining three inscribed Anglo-Saxon gold coins – one of Edward the Elder (899–924), one of Æthelred II (978–1016) and one of Edward the Confessor (1042–66)



(B6–8) – mark another change, for they were all produced from regular penny dies. Indeed, for two of them silver pennies from the same obverse die survive, showing that the gold was part of the same production process. Still one would expect special care to have been taken in the striking in view of their high value, but this was not the case with the Æthelred coin (B7), for the reverse die is poorly aligned on the flan. If this was a special presentation piece the recipient might justifiably have been disappointed. The mints at which they were struck – an unidentified southern mint for Edward the Elder, Lewes for Æthelred II and Warwick for Edward the Confessor – include at least two that were not major administrative or economic centres, nor even towns associated with royal estates.

THE STATUS AND ROLE OF ENGLISH GOLD COINS These eight gold coins are, therefore, rather diverse in their nature, and they have usually been regarded as occasional ad hoc strikings for special purposes, rather than rare survivors of a more regular production in gold (Blunt and Dolley 1968, 157). Five bear the king’s name, one that of an archbishop and two are in a sense anonymous, though they have the name of an official moneyer. All eight coins can be regarded as having been produced in an official mint, in so far as that term is relevant during the Anglo-Saxon period, for we believe that moneyers probably minted coins in their own private workshops rather than a central mint building. The eight coins are die-struck, like the regular silver coinage, which would suggest that they were intended to be money, rather than ornaments or jewellery for which casting would be a more appropriate method of production. Indeed, only two of the surviving coins (B5–6) have been pierced for conversion into jewellery and one other (B2) shows signs on the edge of having been mounted, but possibly in modern times. If they were produced primarily for royal donations one might expect more time and care to have been lavished on their design and technical standard of production, like Roman multiple solidi. The three latest coins (B6–8) used regular penny dies, and two others (B4–5), although from dies specially made, reflect the standard of die-cutting typical of their periods; the remaining three – in fact the earliest three – are quite different from the contemporary silver coinages, but still their designs are not particularly intricate. Moreover, as Hugh Pagan has pointed out to me, most of the extant mancuses date from the middle of the ruler’s reign – only B3 and B5 are so imprecisely dated that they might belong to the period of a reign change – so that a testamentary legacy such as Eadred’s cannot have been the main reason for striking mancuses. While the first five were from major mints (London and York), in that period there was no alternative network of minor mints. When coin production had become less centralised in the tenth and eleventh centuries it is significant that provincial mints such as Lewes and Warwick should have been producing these exceptional gold coins. It would seem that if someone needed gold mancuses in this period they could take the metal to their local mint and ask the moneyer to strike them using his normal dies. If I am right in my theory that Cenwulf tried to bring the production of mancuses under direct royal control, this practice did not last; indeed, it would have been incompatible with the expansion of local minting implemented by Alfred and Edward the Elder. The result was a compromise: moneyers could strike mancuses locally, but they should use their official dies. That Eadred in his will had to leave specific instructions for just that implies that this



was somewhat unusual and that 2,000 mancuses would not otherwise have been readily available, as silver pennies clearly were. On the other hand, gold coins cannot have been so unusual that they would have been of little practical use to the many recipients, for the king would scarcely have distributed such a large sum of money in gold coins if this had been so. It would have been easier and more effective for Eadred to distribute 60,000 silver pennies, but the gold evidently had symbolic importance as a royal gift. The weights of the eight coins – 4.28, 3.74, 4.12, 4.33, 4.39, 4.80, 3.34 and 3.51 g – look quite erratic when compared with those of seven undamaged solidi of Louis the Pious: 4.32, 4.32, 4.34, 4.35, 4.39, 4.41 and 4.45 g (Grierson 1951, 23–25). Offa’s dinar (B1, 4.28 g) matches well the dinar weight standard (4.25 g), and Wigmund’s solidus (B5, 4.39 g) falls in the centre of the range of Louis’ solidi which it emulates. Cenwulf’s gold coin (B4, 4.33 g) could also be a solidus, though at the very light end of the range. All of the other coins fall well outside these two standards, and they have been interpreted as possible mancuses valued at 30 silver pennies, linked to the current weight of the silver penny, assuming the gold: silver ratio to be about 1:10 (Lyon 1969, 207–208; 1976, 183, 189). Thus B2 (3.74 g) is about three times the weight of Offa’s light coinage (c 1.25 g), B3 (4.12 g) three times that of Offa’s heavy coinage (1.40 g), B6 (4.80 g) three times Edward the Elder’s penny (c 1.60 g), B7 (3.34 g) barely three times Æthelred II’s Helmet penny (several standards 1.10–1.55 g; a die-linked penny weighs 1.27 g, and another dielinked to that 1.50 g) and B8 (3.51 g) about three times that of Edward the Confessor’s Expanding Cross type, Light issue (c 1.15 g). Cenwulf’s coin (B4, 4.33 g) could also be treated as a mancus, weighing about three times a penny of the Portrait/Cross-andwedges issue (c 1.42 g). The correspondence with the weight of the relevant penny is reasonably close in all cases except in that of Æthelred II’s Helmet coin, which is at the very bottom of the plausible range. However, all of these calculations assume that the gold:silver ratio was 1:10, when in fact it may have fluctuated according to supply and demand in Europe generally. Typical ranges in the Middle Ages are between 1:9 and 1:12, but they could vary outside that (Spufford 1988, 51). Ratios of 1:9 and 1:10 can be deduced from some Anglo-Saxon documents, while 1:12 seems to have applied in ninth-century Francia based on a paragraph in the Edict of Pîtres of 864 (Stewart Lyon, pers comm, developing on Lyon 1969, 214–216). To what extent the ratio was held stable in particular kingdoms or allowed to fluctuate with market forces we cannot tell. The element of overvalue in the coinage would have provided some cushion against the fluctuations of the market. If the ratio in England in the early eleventh century were 1:12, the Æthelred coin would fall in the centre of the expected range of a mancus, but that may be coincidental. The only coin among the eight that cannot reasonably be interpreted as a mancus is that of Archbishop Wigmund, and the explanation could lie in the fact that Northumbria had a different monetary system from the southern kingdoms. In the mid-ninth century there were no silver pennies for its mancus to be linked to, so the adoption of a Carolingian standard may have been reasonable. Lyon (1969, 221–222) has argued that Louis the Pious’s gold solidus was in fact a new Carolingian ‘mancus’ worth 30 Carolingian deniers, using the 1 :12 gold : silver ratio. None of these calculations takes account of the fineness of the gold, since only three of the coins have been analysed: B1 93 per cent, B2 96 per cent and B5 87 per cent. Yet herein lies another means of adjusting the relative values, and the compositions of the remaining five coins should be tested in order to understand the picture more clearly.



Whatever the precise calculations were by which the Anglo-Saxon moneyers arrived at the weight for the gold coins, it is clear that the weight varied from period to period with some relationship to the current weight standard of the silver penny. To that extent, at least, one can argue that the gold coins were intended to fulfil an economic function, and were not merely prestigious gifts. What would the recipients of Eadred’s legacy have done with their gold coins? Some may have melted them down to make other things, as with the gold-adorned wooden cup bequeathed by Wynflæd to Eadwold ‘in order that he may enlarge his armlet with the gold’ (Will of Wynflœd; Whitelock 1930, no. III). Others could have been mounted as a brooch for display, as with the Wigmund solidus, for we know that nummular brooches in various metals, particularly lead, copper-alloy, silver and silver-gilt, sometimes incorporating original coins, were popular in the Anglo-Saxon period. However, as already noted, display was not the primary function of these gold coins. It is most likely that they would have been kept for those special transactions that required them – purchase of land, payment of heriot, prestige gifts, tribute payment, etc – and in times of need they may well have fetched a premium over their normal value. The fact that two of the six pieces with known provenances (ie 33 per cent) were found on the continent (in Italy and Switzerland) suggests a possible role in alms-giving to Rome or at least in international transactions. However, direct comparisons with the tiny proportion of English silver pennies found in continental Western Europe cannot be made, as these would have been systematically reminted, whereas Anglo-Saxon gold coins could not have been since there were no regular local gold coinages into which they could be converted. Thus, although only eight of these literate gold coins survive from a period of almost 600 years, c 675–1255, there are various features that suggest they were a part of the monetary system, and were produced on a sufficient scale to be useful for a limited number of purposes. The earliest three (B1–3) seem to represent a transitional phase in the late eighth/early ninth century (soon after appearance of the Arabic dinar in Western European sources), when moneyers seem to have been responsible for producing dinar substitutes which could take a variety of forms. The next two pieces (B4–5), from the first half of the ninth century, are official issues produced under central authority without reference to moneyers, and, in the case of the Cenwulf coin, with designs that synchronise with the silver coinage. By the early tenth century the initiative seems to have been back with the moneyers, and they were permitted to strike gold mancuses when requested using their regular penny dies. One must stress that this interpretation is based on the very meagre evidence that is available, and if more gold coins were to be found the picture could change considerably. The English mancuses were, as we have seen, supplemented by Arabic dinars, and more significantly during the ninth century by Carolingian solidi and their imitations. Moreover, as Philip Grierson pointed out in 1951, some of these imitations appear to have been produced locally in England.

LOUIS THE PIOUS SOLIDI AND THEIR IMITATIONS Early in his reign Louis the Pious (814–40) instituted a limited issue of gold solidi carrying his profile bust on one side and a cross within a wreath on the other. The inscription of the reverse, MVNVS DIVINUM (‘divine gift’), is taken to refer to the



imperial wreath with which Louis was crowned in October 816, and this is a likely date for the issue (Grierson and Blackburn 1986, 329). Although there are fewer than 20 specimens of this solidus surviving today, it inspired a rash of barbarous copies or imitations of which about 100 exist. They have occurred in three hoards from the Netherlands, each deposited in the last quarter of the ninth century, and in the Hoen hoard from Norway. One of the Dutch hoards, that from Delfjizl, contained 17 die-duplicates, persuasive evidence that solidi were still being imitated in the late ninth century (Grierson 1951, 29–30). There are also single-finds from the Frankish empire and the British Isles. Of the 20 single-finds listed by Haertle (1997) from Francia, only two were found in France. These were separate finds but both from the Vendée, north-west Aquitaine, a département that otherwise boasts just one ninth-century coin hoard, and it is difficult to explain why these imitations should have occurred there and not elsewhere in France, even if Viking raiders or coastal trade are likely to be part of the explanation. One imitation was found at the emporium of Hedeby in southern Jutland, and the remaining 17 coins were found in the Netherlands, mostly in the provinces of Friesland and Groningen. With this pattern of finds there can be little doubt that the majority of the imitations were produced and used in Frisia. Frisia was different from other parts of the Carolingian Empire. It was a border region between the Frankish kingdoms and the Scandinavian North. The area was controlled by Danish kings or sub-kings, who had been granted rights by the emperor as a way of buying peace and securing their services in the protection of the northern border. The region does not have such strong Scandinavian cultural traits as the English Danelaw, but there are signs of Scandinavian presence in the local finds of metalwork and coins. The clearest and most spectacular examples of this are two recent and quite remarkable silver hoards found in the same field at Westerklief (Wieringen, Noord-Holland). The first, discovered in 1995, was deposited c 850 and contains 1.5 kg of metalwork – 16 silver ingots, six bandshaped arm-rings, an arm-ring and neck-ring of plaited silver rods and a fragment of a silver strap mount – with 78 Carolingian coins and three dirhems (Besteman 1997; Coupland 1997). The second, from 1999/2001, is later in date (c 885), and quite different in composition. It contains 34 Carolingian deniers, one obol and no less than 96 dirhems or pseudo-dirhem flans (many of them fragmentary), together with two complete ingots, 27 pieces of cut ingots or hack-silver and a silver coin-brooch based on a solidus of Louis the Pious (Arent Pol, pers comm). Apart from these two hoards there is a scatter of other finds showing Scandinavian influence (Coupland 2006), though nothing like the wealth of Scandinavian material found in the British Isles. If the majority of the imitative gold solidi were produced in this region, who was responsible for their issue? Were they, as Moesgaard (forthcoming) has suggested, a coinage of the Scandinavians in Frisia, or as Coupland argues, the local Frisians reacting to a local shortage of silver? Seven of the 20 stray finds from Frankish territory have come to light since Grierson (1951) published his survey and corpus of the solidi. In the same period, the number of recorded finds of solidi from Britain has more than doubled, rising from six to 14 (Appendix A6–19), in part due to further archival research. Grierson pointed out that two of the coins found in England were struck from the same pair of dies: one from Lewes in Sussex (A7) and one from Cambridge (A6), while a third specimen from these dies has since been shown to have been found at Therfield, near Royston (A8), 15 miles south-west of Cambridge. A fourth coin from different dies, but in a very similar style, is in the British Museum collection without find provenance, and it too could have been an English find



(Dolley and Morrison 1966, no. 80; Webster and Backhouse 1991, no. 147b). This is the group that Grierson suggested may have been struck in England. Another interesting group comprises three die-linked coins with find-spots facing each other across the North Sea, for two specimens occurred in the Hoen hoard from Norway and the third is a single-find from near Elgin, in Morayshire, far up in north-east Scotland (A12); a fourth specimen is unprovenanced. They are particularly literate and considered to belong early in the imitative sequence. That the three most northerly finds from Europe should share one obverse and two reverse dies is unlikely to be a coincidence. Moreover, all three had been converted for use as jewellery, the Hoen coins being looped and the Elgin specimen pierced twice for mounting as a brooch or in a necklace. In each case the mount is oriented to display the portrait rather than the cross side of the coin. We know that there was contact across the North Sea and around the North Atlantic (Blackburn 2005a), but to appreciate how the Elgin and Hoen coins might be connected, we need to understand how the Hoen hoard was assembled. As indicated above (p 60), it now seems likely that the 20 gold and silver-gilt coins came from a variety of different sources. The two imitative solidi had probably remained together and been mounted as a pair, but whether this was in Frisia or Scandinavia we cannot tell (Blackburn, forthcoming). It would be too much of a coincidence for the Elgin coin originally to have been part of the same necklace, and in any event its method of mounting is different. However, it could have come from the same Frisian goldsmith’s workshop, either via Scandinavia or more directly in trade with Frisia. In any event, the Hoen link suggests a Frisian rather than an English origin for the Elgin coin. Two other imitations (A13–14), bought in Paris in 1841, were said by two contemporary sources to have been ‘found in Scotland’ (Pagan 1988, 72), a provenance which is plausible in view of the Elgin find. Although we know no more than this, it is quite likely that Scandinavians were involved in their transportation or loss in Scotland. The same may also be true of the solidus found in Maughold churchyard in the Isle of Man (A15), although its loss could pre-date the Viking settlement of the island, which it has been suggested may not have occurred until the mid-tenth century (Bornholdt Collins 2003). If a Viking deposit, it would have been an heirloom already a century old. It has a different reverse design from the usual Munus Divinum one probably inspired by a fourthcentury Roman coin. Although from a highly proficient, perhaps official, obverse die, its blundered and unusual reverse suggests that it is not an official Frankish issue. Returning to the English finds, two were found on the south coast, one at Porchester (A9) and another in excavations at the nearby port of Hamwic (Southampton) (A10). A find from Stamford Bridge, near York (A11), belongs to the most literate group (Grierson’s class I) and, like the Elgin and Hoen coins, is likely to be early in the imitative series. Its loss may well pre-date the Scandinavian conquest of York in 866 and subsequent settlement of Yorkshire. The finds from Britain have been building up slowly, but commentators have been reluctant to see in these evidence for any monetary role in England for the solidi. Based on the evidence just outlined, Pagan has stated (1988): It is not easy to determine what function solidi of this nature were intended to fulfil in ninth-century Britain. The occurrence of solidi at Southampton and Porchester suggests that they may have performed a role in cross-Channel commerce, filling the gap left by the absence of any officially struck gold coins. It seems less likely that they played any great part in internal commerce, for the solidus was a heavier coin


MARK BLACKBURN than the mancus, the recognised unit of Anglo-Saxon gold currency, and it is likely that the solidi that found their way inland were valued for their gold weight rather than as coins of fixed international value. More find evidence is necessary before the imitative series can be properly understood.

The last decade or so has been rich in that respect, not merely for the number of finds but for the fascinating range of new material. Three of these are finds of plated forgeries of solidi – a category that was barely known previously. One of these is a very corroded copper core found in excavations at Exeter (A16), and unfortunately it is in such a poor state that one cannot say much about the original design, or whether it had been mounted or treated in any other way. Another is a detector find from near Gainsborough, Lincs (A17), which Archibald regarded as a currency forgery, since there was no sign of jewellery fittings (Archibald and Dolley 1984, 250). This piece is significant, for the style of the drapery is clearly influenced by Anglo-Saxon portrait coins, such as those of Ecgberht or Æthelwulf of Wessex or early coins of Berhtwulf of Mercia – issues of the late 830s to mid-840s that are unlikely to have remained in circulation much after 850. This demonstrates not only that it was made in England, but also that it broadly belongs to the second quarter of the ninth century, contemporary with the solidus of Archbishop Wigmund (B5) and before the Viking settlement of Lincolnshire. The third gilt forgery was found at Hingringham, Norfolk, and is a deliberately cut half coin (A18). The clean chisel-cut is certainly ancient, and very accurately done, suggesting that it had perhaps been intended to circulate as a half-solidus, rather than merely chopped in two to show it was a fake. The existence of these three plated forgeries suggests that solidi were present in England in sufficient quantity to make it worthwhile copying them. But why were they made? If it was as items of jewellery we could expect to find that they had been pierced for suspension, been mounted in a frame or had a pin and catch attached in some way, yet the specimens show no evidence of this. Nor are they likely to have been made for use as grave-goods, as some Merovingian plated coins were, since the practice of putting objects in graves had died out among the English by the early eighth century. It is true that some of the Scandinavian settlers in the later ninth century had coins put in their graves (Biddle et al 1986, 25–32), but the Gainsborough find pre-dates the Scandinavian settlement and Exeter lies outside the area of Scandinavian occupation. The most probable interpretation is that these were counterfeits produced to deceive the unwary into accepting them in payments. Another highly significant find is of a cut-quarter solidus from near Louth, Lincs, found in 1996 (A5). In this case it is not one of the imitations, but the much rarer original issue of Louis the Pious, displaying fine style, good literacy and a high gold content (87 per cent). The coin would have been struck c 816, and the fact that it shows no appreciable sign of wear or damage suggests that it was lost reasonably soon after that – at least during the first half of the ninth century, and again before the Scandinavian settlement of Lincolnshire. The care with which it was cut, neatly with a chisel into a quarter weighing 1.15 g (accurate to within 5 per cent, assuming a solidus of c 4.40 g), suggests that it was used for the payment of a quarter solidus as a sum of money, rather than an arbitrary weight of gold. Of course, we do not know whether it was cut in France or England; in either case it is quite exceptional. The division of silver pennies into halves and quarters to make lesser denominations became standard practice in England after Edgar’s coinage reform of c 973. Before that it was very unusual and, as Kristin Bornholdt’s



research has shown, mainly confined to the Isle of Man in the mid-tenth century. France never seems to have divided coins for currency. This quarter solidus, with perhaps the Hindringham plated half-solidus, is one of the earliest known examples of a coin being cut for currency purposes, and it is possible that the practice originated with such highvalue gold coins, to be applied to the normal silver currency only much later. Still more remarkable is a piece of lead sheet impressed with a pair of dies for striking imitative solidi found in 1997 at Torksey, Lincs (A19). It came from a site that has been prolific in metal-detector finds of the ninth century – coins, metalwork, hack-silver, ingots, weights, etc — much of which appears to be associated with the winter camp established and used by the Danish army in 872/73 (Blackburn 2002). This lead striking has been acquired by Scunthorpe Museum, and it can be described as follows: Obverse: DNHIVID / OVX[ ] (first D reversed; for D N HLVDOVVICVS IMP AVG), diademed bust right. Reverse: NDIDV[ ]VM (for MVNVS DIVINVM), cross in wreath, with tie breaking legend at bottom. Dimensions: 4.6 ⳯ 4.2 cm max. Thickness: 0.2 cm. Wt: 22.64 g, damaged. No coins struck from these dies are known, nor is the style sufficiently close for them to be attributed to one of Grierson’s groups of solidi. They are relatively literate and accomplished compared with most of the imitations, and to that extent would be comparable with the coins from Elgin or Stamford Bridge (Grierson class I). The lead had suffered recent and severe damage in the ground, probably from a plough, badly distorting it and tearing away parts of the right- and left-hand side. The only original edges to survive are the bottom (viewed from the obverse), the lower sections of the left and right, and possibly part of the top. It is difficult to determine the original shape or weight of the object. If symmetrical, it may have been octagonal, but not precisely so as the oblique angles formed by the base and sides are not equal. The impressions of the dies do not align, and they appear to have been applied separately, turning the lead sheet over between strikings. Archibald has considered the function of lead strikings in the Anglo-Saxon and Norman periods, arguing that many of them were used as a form of customs receipt, but allowing that some may have been weights (Archibald 1991; Williams 1999). For this item, struck with dies for an imitative coinage, the customs interpretation seems improbable. A weight would be more plausible, given the number of Viking weights found on the Torksey site, and a relatively heavy weight such as this might appropriately be stamped with dies of a high-value coin. Yet its uneven shape and crude production, even allowing for the recent damage, makes this interpretation less likely than regarding it as a trial piece to test or record the dies. Whatever the function of the lead sheet, its significance lies in the evidence it provides for the production of imitative solidi in England. Lead strikings do not generally stray far from their place of production (Archibald 1991, 331), and while this is an item that could have been brought to Torksey, it might equally well have been produced there. There is evidence of metal working on the Torksey site, in the form of many pieces of roughly broken-up copper-alloy and silver objects presumably destined for the crucible. We do not know how many of the finds and how much of the activity were associated with the occupation of the Viking camp during 872/73, for while the majority of finds fit well with such a date, there are some artefacts (eg Anglo-Scandinavian strap-ends) that suggest



Figure 5.1 Distribution map of finds of gold coins in Britain, 700–1200.

some activity there continued into the tenth century and beyond. It would certainly be unusual for a mobile Viking army to be striking coins, for coin production normally involves an administrative infrastructure to issue them and regulate their use. But different considerations may have applied to imitative coins such as these solidi, and one could envisage goldsmiths being asked to convert gold into solidi rather than say gold ingots. The dies may, indeed, have been intended for the production of plated forgeries. However, on balance it seems more likely that the dies and the lead striking were made elsewhere in England, perhaps by Anglo-Saxons – officially or unofficially – and brought to Torksey as loot or through trade. In any event, the piece is important evidence supporting the view that some of the imitative solidi were made in England. Do the new finds allow us to understand the role of the imitative solidi any better? The dating of the Gainsborough forgery provides welcome evidence that solidi were used in England during the second quarter of the ninth century and were not primarily introduced by the Vikings. This early dating is also supported by the solidus of Archbishop Wigmund (c 810–40) and by the finding of a cut-quarter of an original Louis the Pious solidus near Louth. If, as seems likely, the production of imitations began soon after the official solidi were issued in c 816, the Delfjizl hoard shows that they continued to be struck until the late ninth century, if not beyond – a long period of production. From this late period, there is one final ‘Carolingian’ gold coin, found at Congham, Norfolk (A20). It has a quasi-official status, being struck from a regular, though rusty, reverse die for a silver denier of Chartres of c 870–85, but an obverse that is loosely based on a fourth-century Roman solidus. Weighing only 1.89 g with c 37 per cent gold, it might



be classed as a gold denier, valued at only four or five silver deniers. Yet it may not be a Frankish production, for I have suggested that it could have been struck in the Danelaw c 880, like the imitative Temple type pennies that use official Quentovic reverse dies (Blackburn 2005b). In terms of circulation, it is clear that the use of imitative solidi was not restricted to cross-Channel commerce. The distribution (Figure 5.1) is now quite widespread, and looks fairly typical of stray finds for the ninth century (cf. Metcalf 1998). Pagan’s observation (above, p 69) that the solidi would have been incompatible in circulation with mancuses is compounded in the case of the imitations, for they have a much wider range of weights, typically varying from 3.9 to 4.5 g, and lower fineness (Grierson and Blackburn 1986, nos 752–756, five specimens ranging between 76 per cent and 35 per cent gold). With such a wide range in their intrinsic value, it is difficult to see how the imitative solidi could have been accepted in commercial transactions by tale. They would surely have been weighed and tested by streaking them on a touchstone. It is not surprising, then, that a number of charters, particularly in the mid-ninth century, should specify that payment was made in ‘pure gold’.

ANGLO-SAXON GOLD JEWELLERY AND ARTEFACTS Despite the many documentary references to payments in gold and to gold armlets, jewellery and other artefacts, surviving Anglo-Saxon gold ornaments from the eighth century onwards are rare. Finger-rings are the most substantial group (Hinton 1978, listing some 60 rings of the eighth to eleventh centuries). A fashion for elaborately decorated rings of high status developed in the ninth century, the grandest of which carry royal inscriptions (Webster and Backhouse 1991, nos 243–244), while others have Trewhiddle-style decoration (Webster and Backhouse 1991, nos 201–203) or simple inscriptions (Wilson 1964, nos 27, 30, 85 and 145). Apart from gold finger-rings, other objects made of solid gold are exceptional. Among the few pieces known are the Alfred Jewel and related socketed gold terminals from Minster Lovell (Oxon), Bowleaze Cove (Dorset) and Warminster (Wilts) that may well have been the handles for sumptuous manuscript pointers or æstels, such as those referred to in the preface to Alfred’s translation of Gregory’s Pastoral Care (Keynes and Lapidge 1983, 203–206; Webster and Backhouse 1991, nos 258–260; National Art Collections Fund Review 1999, no. 4740, Warminster). The tiny gold and niello fitting from Bidford on Avon (Warwks) (Treasure Annual Report 1998–99, no. 70), although superficially similar, must have had some other function. Equally puzzling is the animal-headed gold fitting from Sutton on the Forest (Yorks) (Treasure Annual Report 2000, no. 82). A clearly ecclesiastical item is the inscribed gold plaque from Brandon (Suffolk) depicting John the Baptist (Webster and Backhouse 1991, no. 66a). A gold filigree and glass button found recently at Luton (Beds) has been dated to the second half of the eighth or early ninth century (Treasure Annual Report 2002, no. 63). The hoard from Trewhiddle (Cornwall), deposited c 868, originally included a filigree gold pendant and small hexagonal gold ingot (C10), both now lost (Wilson and Blunt 1961, 85–86). The pair of filigree gold roundels from Lilla Howe (Yorks), and a similar piece from Swinderby (Lincs), could be of Scandinavian rather than Anglo-Saxon origin (Webster and Backhouse 1991, no. 249). Apart from the Luton button, all these items are probably of ninth-century date.



From the tenth and eleventh centuries, there are a few very elaborate gold filigree and gem- or enamel-inlaid secular brooches, such as that from Dowgate Hill, London, and the lost brooch from the St Mary-at-Hill, London hoard (deposited c 1074), and two new fragmentary finds, one an Ottonian piece from Fauld (Staffs) (Treasure Annual Report 2000, no. 37) and the other from Holberrow Green (Worcs) (Treasure Annual Report 2001, no. 42). A large eleventh-century strap-end from Norfolk (Norwich Castle Museum 1986.254) is of gold-alloy (c 41 per cent; Tim Pestell, pers comm). Gold could be used economically to adorn silver items, such as the King’s School, Canterbury brooch (Wilson 1964, no. 10), or the small portable sundial from Canterbury (Backhouse et al 1984, no. 77). The V&A ivory crucifix, with its cross covered in gold sheet, of mixed Anglo-Saxon and German origin (Backhouse et al 1984, no. 118) hints at what other altar crosses might have been like, such as those presented by Eadgar and Cnut to Ely and Winchester respectively (Dodwell 1982, 201). Yet even scraps of metal from composite objects such as these are very rare (cf. the little filigree plate from the Winchester excavations; Backhouse et al 1984, no. 78). In all, the number of extant Anglo-Saxon gold objects, other than finger-rings, from the eighth to eleventh centuries is fewer than 20, and most of these are small, so that in total the weight of gold can barely exceed 1 kg. Compared with any other category of ornament from the late eighth century onwards, finger-rings are indeed prolific. Considering this and the frequent references to arm-rings in documentary sources, Hinton (1978, 139–141, 146) posed the question whether these were simply personal ornaments or whether they may have been a way of storing bullion and even fulfilled some role as a gold currency. Looking at the weights of 19 rings, he found that there was some loose clustering which might be interpreted as representing 1, 2, 3 and 5 mancus units for finger-rings, and possibly 12–15 and 37–44 mancuses for two arm-rings. The correlations were only very approximate, because, he suggested, the mancus weight fluctuated considerably with the current weight of the silver penny. However, as discussed above (p 58), it is more likely that the mancus as a unit of weight was stable and was not linked to the weight of the mancus as a coin or that of the pennies, so that one would expect weight-adjusted items of bullion to follow a more coherent pattern. The concept of ornaments fulfilling a dual role as both personal adornments and units of bullion has been shown to be valid in the case of Scandinavian arm- and neck-rings (Graham-Campbell 1999), but for the Anglo-Saxon finger-rings it is not proven. The scarcity of surviving gold ornaments from this period provides a dramatic contrast with the wealth of later sixth- and seventh-century gold jewellery. The availability of gold for the production of ornamental metalwork appears to have declined in the same period that gold coinage in Western Europe was being debased and superseded by silver (Hawkes, Merrick and Metcalf 1966; Webster and Backhouse 1991, 47, 220–221). A simple comparison of the number of extant gold artefacts from different periods is, however, a misleading measure of the extent to which they were produced, for one has to take into account the fact that, after the end of furnished burials in England (c 700), high-status metalwork has a much lower survival rate. It is impossible to quantify the effect of this, as we do not know how many stray finds of gold artefacts of the sixth and seventh centuries came from disturbed burials. Even so there must have been a significant shift away from solid gold artefacts to those of silver, silver-gilt and particularly gilt-copper-alloy, which are prolific. The taste for gold-looking objects certainly survived. Dodwell’s survey of the literary and documentary references to Anglo-Saxon metalwork



in gold and silver presents a sumptuous picture of the trappings of the Church and the Anglo-Saxon elite (Dodwell 1982, 188–215). Gold is mentioned repeatedly in descriptions of reliquaries, altar crosses, church plate, book covers, jewellery, arm-rings, swords and scabbards, ships prows, furniture fittings and so forth. Yet many of these must have been gilded objects or fittings, like the small spouted jug in the British Museum (Backhouse et al 1984, no. 72). If solid gold had been widely used over such a range of artefacts, one would not expect the highly specialised socketed gold mounts to represent such a significant proportion of the few gold objects that survive.

SCANDINAVIAN GOLD ORNAMENTS, INGOTS AND HACK-GOLD Gold is also very scarce in the Scandinavian world, in comparison to the vast quantity of silver from Viking-Age finds. Given the existence of the Scandinavian bullion or metalweight economy, archaeologists quite rightly have drawn a distinction between finds of whole ornaments and ones where they have been cut up for use as money (Hårdh 1996, 18–27; Sheehan 2000b). To the cut-ornaments (or hack-gold) we should add gold ingots, which served a similar purpose, yet both of these are notably rare in Scandinavia and areas of Scandinavian settlement. In Scandinavia itself, Hårdh (1996, 132–134, 160–161) has observed that gold, when it occurs, normally consists of complete ornaments deposited separately from the silver hoards. There are very few mixed hoards, and little in the way of hack-gold or gold ingots has been found. If one makes a comparison with silver finds, a much larger proportion of gold takes the form of whole ornaments, though there are regional differences. Hårdh suggests that gold had more of a prestige role than an economic one in Scandinavia, and some gold deposits left in or near water may have been part of a religious ritual. In England only a few whole gold ornaments of Scandinavian character have been found, notably the twisted-rod gold arm-ring from Wendover Dean (Bucks) (Smith 1923, 117, plate iii, no. 7), the filigree roundels just mentioned (p 74), six small gold bracteates from the Halton Moor (Lancs) hoard deposited c 1030 (Blackburn and Pagan 1986, no. 212) and several gold finger-rings, such as two in the Soberton (Hants) hoard, deposited c 1068 (Smith 1923, 117, plate iii, nos 4–5), single-finds from Chester (St Werburgh Street) and York (Hungate), and another from Oxford (Graham-Campbell 1988). No ingot or hack-gold was known from England, besides the tiny Trewhiddle hoard ingot (C10), until the mid-1990s, since when several discoveries have been made in the Danelaw, thought to be Scandinavian losses of the later ninth or early tenth centuries (described in Appendix C). Three are whole ingots, apparently single-finds: a small transverse-hammered ingot from excavations in Norwich (C1), and two larger cast ingots from Fenstanton (Cambs) (C2) and West Dereham (Norfolk) (C3). From the ‘productive site’ at Torksey, which was probably the location of the Danish winter camp of 872/73, there is a mixed group of ingots, hack-gold and hack-silver that appears to be part of a hoard, the gold element of which comprises the cut end of an ingot (C5) and a section of gold rod (C6). From the same site, but probably part of a different find is a piece of hackgold in the form of a small piece of thin rod (C8), found in association with two small silver ingot terminals. Another piece of hack-gold from the site was a single-find, a piece



probably cut from a twisted-rod neck- or arm-ring of Scandinavian type (C7). From Springthorpe, near Gainsborough (Lincs), just 10 miles from Torksey, a piece of hackgold occurred as a single-find, again taking the form of a piece of cut gold ingot or rod (C4). Finally, and most recently, a section of very finely hammered gold band, probably cut from a finished or partially finished arm-ring, was a single-find from West Wratting (Cambs) (C9). These pieces of hack-gold and the hack-silver from Torksey, cut from ingots and ornaments, including some of Scandinavian origin, were evidently intended for use in a Scandinavian-style metal-weight economy. The two associated groups in which they were found show that they retained that function on the Torksey site – in GrahamCampbell’s terminology, these were ‘active hoards’ where the contents represented a means of payment (Graham-Campbell 1989). They contrast with the many roughly broken pieces of Anglo-Saxon ornaments, mainly of copper-alloy, but some of silver, that appear to have been destined for the melting pot – the evidence for metal-working on the site, mentioned above (Blackburn 2002). In Scotland, the pattern is similar (Graham-Campbell 1995, especially pp 54–55; 2005). There are several finds of whole gold ornaments, namely a neck-ring from Braidwood Fort (Midlothian), two gold arm-rings from the Broch of Burgar (Orkney) and single ones from Pierowall (Westray), Oxna (Shetland) and the Sound of Jura, six complete gold finger-rings and one fragment from the ‘Hebrides’ hoard, four from the Loch of Stenness (Orkney), as well as single rings from several other sites. There are no recorded finds of complete gold ingots from Viking-Age Scotland, but two cut pieces of ingot were present in the Hebrides finger-ring hoard, together with a fragment of gold rod. Another piece described as a tiny ingot fragment was found in excavations on the Broch of Birsay (Orkney). Hack-gold is generally very scarce in Scotland, with the odd piece occurring in the Iona hoard and the Hebrides hoard; some tiny fragments of gold have also been noted in excavations. From a pre-Viking phase, a small hammered gold ingot and a piece of rod were found in excavations at Whithorn Priory (Lothian and Galloway), a Northumbrian monastic site, where they were associated with a fine-metal-workshop dating from c 700 (Hill 1997, 397–399). Other scrap gold was found in a later metal-workshop at Whithorn dating from the eleventh–thirteenth centuries, and in both cases the finds were accompanied by crucibles, tools and metalworking debris. From Ireland there are quite a number of finds of complete ornaments, the most spectacular being the remarkable Hare Island (Co Westmeath) hoard of 1802, comprising 10 gold arm-rings weighing some 5 kg, twice the weight of the Hoen hoard (GrahamCampbell 1974; 1976, 50–51). Broad-band gold arm-rings have been found at Edenvale (Co Clare), Glengarriff (Co Cork), Vesnoy (Co Roscommon) and Wicklow (Co Wicklow), and other gold arm-rings have come from Dublin High Street, Rathkeale (Co Limerick), Virginia (Co Cavan), Rathedan (Co Carlow) and Co Westmeath (Graham-Campbell 1976; Sheehan 1998a; 1998b; 2000a). As in other regions, finger-rings are also reasonably plentiful, such as the two from the Dublin High Street excavations, which also yielded a piece of gold braid (National Museum of Ireland). Four or five gold ingots are known, one from an unpublished hoard from Knock (Co Mayo) found with two whole silver ingots and one fragment (Sheehan 1998a, 200), a transverse-hammered ingot from Askeaton (Co Limerick) (Cahill 1993, 9–11; Sheehan 1998b, 163), another from the Wood Quay excavations in Dublin (Cahill and Ó Floinn 1995, 80), an ingot from Co Cork now in the British Museum (Graham-Campbell, forthcoming) and a chisel-cut ingot terminal



recently reported from near Dublin. This last is one of the very few pieces of cut gold recorded from Ireland. Apart from this piece, there is a cut-fragment of an arm-ring without provenance in the National Museum of Ireland (W.70; Bøe 1940, 104), and James Graham-Campbell has drawn my attention to a fragment of a Hiberno-Viking arm-ring found near Clonmel (Co Tipperary) in 1842, but now lost (Cahill 1994, plate vii, no. 3, as Bronze Age). The Isle of Man has yielded one twisted rod gold arm-ring from the large Ballaquayle hoard, deposited c 970 (Graham-Campbell 1983, no. 5, 1, fig 8), and a plaited-rod fingerring from Greeba in 1981 (Cubbon 2003), previously thought to be a single-find, is now to be associated with a fragment of a second gold finger-ring, also metal-detected at Greeba, with cast decoration (Wilson 2004). Apart from these, there is a suggestion in one account that the partly dispersed Kirk Andreas (1874) hoard, deposited c 1045, may have included some gold as well as silver ornaments, together with its coins (Bornholdt Collins 2003). There are, then, broad similarities between the pattern of gold finds in the Scandinavian areas of the British Isles and those in Scandinavia itself. Everywhere gold is rare in comparison with silver. Yet the balance between whole ornaments, on the one hand, and ingots or hack-gold, on the other, varies from region to region. In Ireland, where there are the most gold finds, whole objects dominate and there is very little hack-gold and ingot. One might infer that gold was used there more for display than in an economic role. However, although the metalwork is of Scandinavian character, some of the deposits were probably put down by native Irish people, and the preference for whole gold ornaments will in part reflect their practice. Scotland also has a fair number of whole ornaments among its finds, but ingot fragments and hack-gold, although rare, represent a slightly higher proportion of the gold finds than in Ireland. Among the English finds, ingots and hack-gold represent a much higher proportion than anywhere else in the Scandinavian world, suggesting that gold played a greater economic role. However, the number of finds from England is still small, and most of the hack-gold and ingots come from Torksey and nearby Springthorpe. These finds may be associated with one event, the wintering of the Viking army in 872/73, and certainly reflect the remarkable character of the Torksey site, where trade and metalworking were clearly important.

CONCLUSIONS The written sources leave one in no doubt that gold was regularly encountered in late Anglo-Saxon England, at least in the church and among the ruling elite. It was used particularly for royal gifts and bequests, both from and to the king, and within his household, and this duly became formalised in the payments of heriot. The wearing of gold, most notably in the form of finger-rings, but also for men as arm-rings or ornaments on a sword and for women as jewellery, would have been a status-symbol, but how far down the social classes it percolated we cannot tell from the written sources alone. Quite apart from such display, gold had an economic role, recorded particularly for the purchase of land and the payment of tribute, although one has to consider how often gold – be it in bullion or coin – actually changed hands. In some cases we can be sure that it really was used, but in others it could have been merely a formal way of expressing the price. Equally, the sumptuous descriptions of golden vessels, ornaments and jewellery found in



many literary sources should perhaps be read with a sceptical eye, for some of it will surely have been in gilt-silver or gilt-bronze, the latter being plentiful among the finds. The surviving material in gold is no less tricky to interpret. There can be no doubt that far fewer Anglo-Saxon gold objects exist today from the eighth century onwards than from the sixth and seventh centuries, but it is difficult to determine what effect the ending of furnished burial practice had on survival rates. The rise in the proportion of high-quality gilded objects suggests there was a real move away from the use of solid gold. The fact that such specialist items as the socketed mounts or æstels feature so prominently among the small corpus of ninth-century gold objects, apart from finger-rings, implies that they were rather exceptional in being produced in gold. The decline in gold ornaments after the seventh century does then appear to be a genuine one, and their use was probably limited to people of higher social status. Turning to the coinage, we should be wary of comparing the number of surviving gold coins before c 675, when they formed the sole currency in England, with the number from later periods, when silver dominated and gold was effectively special-purpose money. The furnishing of some burials with grave goods before the early eighth century is one factor, but another is the rate of accidental loss from circulation, which may well have been higher in the seventh century than in later periods, as the role of gold coinage was rather different. It is indeed remarkable just how many seventh-century gold coins were accidentally lost, represented by single-finds particularly on ‘productive sites’ such as Riby, ‘South Lincs’, Bawsey, Coddenham, Hollingbourne, etc (Bonser 1997; Blackburn 2003), showing that these coins, despite their high value, were regularly being used in trade on these sites. Later, such transactions would generally have used silver or copper coins, and gold when available would have been reserved for high value payments and been less often carried in pockets or purses at markets and fairs, vulnerable to accidental loss. Thus, in the later medieval and early modern periods when gold coins were plentiful, they are scarce as single-finds even on sites that have produced hundreds of silver or copper coins of the period. Yet this change of role in the late seventh century was part of an economic trend, for the progressive debasement and then abandonment of gold coinages was a phenomenon that affected all parts of Western Europe during the seventh and eighth centuries, and must have been born out of a general and deepening shortage of gold. The gold once released from its monetary use might in theory have been available in larger quantity for ornamental purposes, but it seems that the economic forces that drained the supply of gold to the East during the period of progressive debasement in the seventh century had also depleted the stock of ornamental gold as well. The eighth century is probably the thinnest period of all for Anglo-Saxon gold metalwork and finds of gold coins. By contrast the ninth century presents something of a revival. Rather more gold coins have been found in the British Isles from the ninth century than for any period down to the fourteenth century (Figure 5.2). Although 17 gold coins from the ninth century may seem rather few, when one remembers that not a single bezant or florin from the thirteenth or early fourteenth centuries has yet been found in Britain, despite extensive documentary evidence for their use (above, p 60), it becomes much more significant. Ninth-century gold metalwork is probably also less scarce than that from the tenth or eleventh centuries. The same is true in respect of gold coin finds from France and from the Netherlands, and it may be no coincidence that the largest Scandinavian gold hoard – that from Hoen – was deposited during the second half of the ninth century. The tenth century is notably thin for English



Figure 5.2 Histogram of gold coins in the British Isles, 700–1200 (Source: Appendices A and B; fractional coins counted as fractions, base forgeries omitted).

finds of gold coins, with just a paltry quarter-dinar, despite the fact that there are more references to gold in wills and charters than are found from other centuries. In the eleventh century the finds are a little more productive, and the amount of surviving Anglo-Saxon metalwork also increases to some extent. Even if such small samples will not bear much statistical comparison, there does seem to be a broad agreement between the coin and artefact evidence. English gold coins of the period 675–1255 are immensely rare, yet the eight surviving specimens with coherent inscriptions (ie not including anonymous copies of Arabic dinars and Carolingian solidi, some of which may have been made in England) have characteristics that indicate they were produced in official mints and were intended to fulfil a role within the established monetary system. They were probably produced as and when required, but in sufficient quantity to perform a recognised function. Three phases of production have been tentatively discerned among the eight surviving coins. During an initial transitional phase in the later eighth century, moneyers seem to have produced mancuses in a personal capacity to their own designs, and often solely in their names without that of the king. The second phase, probably inaugurated by Cenwulf c 807, saw the crown taking control of the production of gold coinage, providing special distinctive designs, which in the case of Cenwulf’s mancus related to the design of the penny. If these were intended to replace the foreign coins already in use they did not succeed, since continental solidi and their imitations dominate the ninth-century finds. In the third phase, during the tenth and eleventh centuries, moneyers at any mint seem to have been able to strike gold mancuses on request using their regular penny dies. The precious record in King Eadred’s will of a direction to strike 2,000 mancuses may well be typical of the type of circumstances under which gold coins were produced. Yet it need not have been in such quantity or at the direction of a king; provided the gold could be found, anyone could probably have commissioned a moneyer to produce some mancuses. To what extent such coins may have been required to complete a particular transaction we cannot tell. Did they have a special function, that imported solidi or Arabic dinars could not fulfil? The



imitative Louis the Pious solidi have such variable weights and finenesses that it is difficult to believe they were not used as a convenient form of bullion, rather than coins of recognised value. But simply being in the form of coin could have given the gold a premium value, as we also see with the Scandinavian imitative silver coinages of the late tenth and early eleventh centuries (Malmer 1989; 1997). The role of gold in England during the Viking Age is a tantalising subject. Although the evidence is woefully incomplete, when one puts together everything we know from the coins, artefacts and written sources it does begin to build a coherent picture. On the one hand gold was scarce and of high value, giving it a prestige that made it appropriate for certain functions in Anglo-Saxon society. On the other, there was sufficient availability for it fulfil those roles, in particular as a standard means of payment for the purchase of land, and for there to be a recognised tradition of moneyers striking gold mancuses over a period of some 300 years. Given their special use, it is to be expected that gold coins would not have been hoarded with silver. We think of the English mancus as an exceptional coin, largely because of its rarity, but that perception would change if a hoard of them were to be found. One day, I hope, there will be such a find – most likely to be from the later ninth or the eleventh century, periods with the highest rates of hoard deposit and non-recovery – and then we will be able to flesh out the bones of this particular chapter of English monetary history.


Islamic coins A1.

Eastbourne, Sussex, pre-1847. Found on the beach. Two dinars, one said to have been cast in gold from a silver one (?), perhaps a dirhem that has been gilded; one identified as Umayyad, Hisham (724–43) by Prof Wilson. NC 1st series 9 (1846–47), 95; Duplessy 1956, 124, no 12. Neither has ever been illustrated.


Wickhampton (near Acle), Norfolk, 1988. Abbasid dinar, al-Mansur (754–75), dated AH 147 (AD 764/65). Unpublished, information Barbara Green. Norwich Castle Museum.


Anglo-Saxon coins A3.

Nr Manchester, 1849. Anonymous gold mancus, temp Offa (757–96) or Cenwulf (796–821), kings of Mercia, struck by the London moneyer Ciolheard, c 792–805. Wt 4.12 g. BM. See Appendix B3.


Nr Biggleswade, Beds, 2001. Cenwulf of Mercia (796–821), gold mancus, Portrait type, c 807, London mint. Wt 4.33 g. See Appendix B4.



Carolingian coins and derivatives Official issue of Louis the Pious A5.

Louth, Lincs, 1997. Cut quarter of a regular solidus of Louis the Pious. Wt 1.15 g; 87 per cent gold (SG analysis), assuming an alloy of gold and silver. Coin Register 1997, no. 101. FM; ex Spink sale, 14 July 1998, lot 270.

Imitations of Louis the Pious A6.

Cambridge (in the River Cam by Magdalene Bridge), before 1904. Imitative solidus of Louis. Wt 4.30 g; c 66 per cent gold (SG analysis). Grierson 1951, XVIc; Grierson and Blackburn 1986, no. 753. FM; ex Neville pre-1904. Same dies as Lewes and Therfield finds; probably struck in England.


Lewes, Sussex, 1884. Imitative solidus of Louis. Wt 4.33 g. Grierson 1951, XVId. Ashmolean Museum; ex Evans. Same dies as Cambridge and Therfield finds; probably struck in England.


Therfield, nr Royston, Herts, before 1860. Imitative solidus of Louis. Wt 4.59 g. Grierson 1951, XVIb; Dolley and Morrison 1966, no. 78; Pagan 1988, 71 (for the find provenance, citing Proc Soc Antiq 2nd series 1 (1859–61), 306); Webster and Backhouse 1991, no. 147c. BM; ex Beldam 1860. Same dies as Cambridge and Lewes finds; probably struck in England.


Porchester, Hants, 1832. Imitative solidus of Louis. Pagan 1988, 70; Sotheby, 25 March 1850, lot 247. Never illustrated.


Southampton excavations, 1971. Imitative solidus of Louis. Wt 4.56 g. Metcalf 1988, no. 148; Pagan 1988; Gunstone 1992, no. 2424. Southampton Museum. Same dies as O’Hagan sale 1908, lot 3 (Grierson 1951, VIIa).


Stamford Bridge, Yorks, early 1940s. Imitative solidus of Louis. Wt 4.08 g, pierced. Same obverse die as BM coin (Grierson 1951, Ie; Dolley and Morrison 1966, no. 81). Stewart 1986. Richard Falkiner collection; ex Christies sale, 17 November 1970, lot 46a; ex W Lee. Cast in FM.


Elgin, Morayshire, before 1864. Imitative solidus of Louis. Wt 4.34 g, pierced twice. Grierson 1951, Ia; Dolley and Morrison 1966, no. 79. BM; ex Frazer 1864. Die-links with two specimens in the Hoen hoard, Norway; from the same obverse and reverse as one coin and the same obverse as the other.


Scotland, before 1841. Imitative solidus of Louis. Wt 4.52 g. Fillon 1853, plate viii, 11; Gariel 1883–84, II, plate xiv,13; Grierson 1951, XIXa (and note in Grierson 1979, addenda); Pagan 1988, 72. BnF, Paris; ex Smith-Lesauëf; ex Gariel (Hoffmann sale, 8 November 1885, lot 683); ex J Lelewel.


Scotland, before 1841. Imitative solidus of Louis. Wt 4.04 g. Fillon 1853, plate viii, 8; Prou 1896, no. 1076; Grierson 1951, XXa; Pagan 1988, 72 (with discussion of Scottish provenance). BnF, Paris; ex Feuardent 1896; ex J Lelewel. Annulet eyes and line of annulets for drapery; Pagan calls these Anglo-Saxon features.


Maughold churchyard, Isle of Man, 1884. Variant of a Louis solidus, perhaps from an official obverse die, but with a standing figure and blundered legend on the reverse. Wt 4.36 g. Grierson 1951, 17, C; Bornholdt Collins 2003, 213–215. Manx Museum.



Plated forgeries of Louis the Pious A16.

Exeter (excavation find), Devon, 1971–80. Base metal core from a plated forgery of an imitation of Louis solidus, badly corroded and bust obscure, but elements of the original inscription legible. Wt 2.43 g (corroded core only). Not pierced and no evidence of having been mounted. Archibald and Dolley 1984, F1.


Nr Gainsborough, Lincs (between Gainsborough and Lincoln), early 1995. Giltbronze forgery of a solidus of Louis. Wt 4.06 g. Coin Register 1995, no. 127 (‘This a currency forgery as there is no sign of jewellery fittings’, M M Archibald). Crude imitation, with pellet in each rev quarter, drapery of ‘Anglo-Saxon’ style, most similar to early coins of Berhtwulf or Rochester coins of Æthelwulf, ie of the early-mid 840s. This would be contemporary with the solidus of Archbishop Wigmund (c 837–54). Probably struck in England.


Hindringham, nr Fakenham, Norfolk (SMR no. 25071), winter 1988/89. Plated forgery of imitative solidus of Louis, cut half. Wt 1.79 g; gold plated on copper-alloy core, using mercury gilding technique. To be published by Blackburn and Northover.

Lead trial-piece or weight for Louis the Pious solidus A19.

Torksey, Lincs, 1997 (metal-detector find from prolific site that appears to be the location of the Viking camp of 873/74). Lead plate impressed with dies for imitative Louis the Pious solidi; probably a trial striking of the dies rather than a finished object such as a weight. See text above for description. Blackburn 2002, 93–94. Scunthorpe Museum.

Derivative issue after Charles the Bald A20.

Congham, nr King’s Lynn, Norfolk, 11 March 1990. Anon gold denier temp Charles the Bald or later, c 875–c 925. Struck from a regular reverse die of Chartres of the period c 870–85, but this gold coin could be rather later. Possibly struck in the Danelaw. The obverse is derived from a Roman coin of Constantine I or II. Wt 1.89 g, pierced for suspension; 37 per cent gold (SG 12.68) assuming an alloy of gold and silver. Blackburn and Bonser 1990a; 1990b; Blackburn 2005b; Spink auction, 3 November 2001, lot 639 (unsold).


Islamic coin A21.

Carisbrooke Castle, Isle of Wight, pre-1999. Fatimid, quarter-dinar (tari), temp al-Mu’izz (953–75), mint Filastin?, AH 358 (AD 968/69). Wt 1.00 g. Cook 1999b, 269, no. 42; additional information from Cécile Bresc.


Anglo-Saxon coins A22.

Hellingly, East Sussex, c 1808. Æthelred II (978–1016), gold mancus, Helmet type, c 1003–09, Lewes mint. Wt 3.34 g. BM. See Appendix B7.


Figures 5.3 British finds of gold coins: A2-A14.




Figures 5.4 British finds of gold coins: A15-A25.




Worcester (St Clement’s Church, during demolition), c 1824. Edward the Confessor (1042–66), gold mancus, Expanding Cross type, c 1050–53, Warwick mint. Wt 3.51 g. BM. See Appendix B8.

Islamic coins A24.

St Leonards on Sea, Sussex, 1986. Found on the beach, having probably fallen off the cliff. Fatimid, 1/4 dinar (tari), Sicily, temp al-Mustansir (1036–94), blundered legends, c 1050–70. Wt 0.9 g. Blackburn and Bonser 1986, no. 124.


Nr York, 1752. Almoravid dinar, probably of Yusuf b Tashufin (1087–1106), mint and date unknown. Wt ‘60 grains’. Metcalf 1958, 95 (citing Society of Antiquaries Minute Book vii, 18, 7 May 1752, with drawing, reproduced Figure 5.4).


Islamic coins A26.

Oxford (St Aldates, ‘during the digging of a sewer in the street opposite Christ Church’), 1825. Almoravid dinar of ‘Ali b Yusuf (1106–42), Denia (Spain), AH 500 (AD 1106/07). Brown 1990; Blackburn and Bonser 1986, note to no. 124. Ashmolean Museum, Christ Church collection (no. 364).

A27–28. London (‘within site of St Paul’s cathedral’), pre-1879. Two Almoravid dinars of ‘Ali b Yusuf (1106–42), Almeria (Spain), AH 525 (AD 1130/31). American Journal of Numismatics 13 (1878/79), 92–93; Duplessy 1956, 133, no. 36. Never illustrated.


Offa of Mercia (757–96), gold dinar (mancus), copying an Abbasid dinar of Caliph al-Mansur (754–77), AH 157 (AD 773–74), but with ‘OFFA REX’ inserted into the inverted reverse. Wt 4.28 g; 93 per cent Au (SG 18.2), assuming a gold-silver alloy. BM; ex Carlyon-Britton (Sotheby sale, 17 November 1913, lot 269); bt Rollin and Feuardent (Paris) 1907; ex Duc de Blacas; acquired in Rome before 1841; presumed local find. Longpérier 1842; Akerman 1842; Carlyon-Britton 1908; Allan 1914; Blunt 1961, 50–51; Stewart 1978, A120; Archibald 1990, 12; Webster and Backhouse 1991, no. 148c.


Temp Offa of Mercia (757–96), gold mancus, struck by Offa’s London moneyer Pændræd, c 785–92. Inspired by a gold aureus of Augustus (27BC–AD14) of the mint of Lugdunum (Lyons). Obv +PAENdRAEd MVNITA RE (for Pændræd mynetere, ‘Pændræd the moneyer’) around diademed bust; rev PEND/REd, standing figure of Diana with spear and bow. Wt 3.74 g; 96 per cent Au (SG 18.8), assuming a gold-silver alloy. The reverse die is smaller in diameter than the obverse, suggesting that they may not have been made as a pair, a view that is perhaps supported by the different spellings of the name Pændræd. Stewart Lyon (pers comm) has wondered whether a silver penny with this reverse design may one day



Figures 5.5 English gold coins with meaningful inscriptions: B1-B8, with comparative material.



be found, yet the two designs derive from the same Roman prototype and the obverse was therefore intended to accompany this reverse. There are signs on the edge of it having been mounted, but whether in Anglo-Saxon or more modern times is unclear. BM, given by C E Blunt; ex Christie sale, 30 May 1961, lot 16; ex Sir Robert Cotton (1571–1631). Illus John Speed’s Historie of Great Britaine (1611), 315. Dolley 1964, 15, frontispiece A; Blunt and Dolley 1968; Stewart 1978, A122; Tite 1992; Van der Meer 1994, 224 (arguing that Cotton acquired it between 1606 and 1611). B3. Temp Offa of Mercia (757–96) or Cenwulf (796–821), gold mancus, struck by Ciolheard, a London moneyer c 792–805. Wt 4.12 g. Obv CIOLH/EARD, diademed bust after a late fourth-century solidus; rev cross potent-on-steps after seventh/ eighth-century solidus. BM; ex Rev A Mallinson (Spink sale 39, 6 December 1984, lot 73); ex Lord Grantley (Glendining, 27 January 1944, lot 602); ex W and T Bateman (Sotheby, 4 May 1893, lot 241); found near Manchester 1849. Pagan 1965; Sutherland 1948, no. 76. B4. Cenwulf of Mercia (796–821), gold mancus, London, c 805–10. Wt 4.33 g. Obv +COENVVLF/REX M (contraction mark over M), diademed bust right; rev +DE VICO LVNDONIAE, eight-lobed geometric daisy design. Found by the River Ivel at Biggleswade (Beds), 2001. Final disposition uncertain; ex Spink, 6 October 2004, lot 493. B5.

Wigmund, archbishop of York (c 837–54?), gold solidus inspired by those of Louis the Pious. Obv VIGMVND ARE/P, facing tonsured bust of the archbishop; rev MVNVS DIVINVM, cross in wreath. Wt 4.39 g, pierced twice for ornamental display; 87 per cent Au (SG 17.5), assuming a gold-silver alloy. Grueber and Keary 1887–93, I, 193, no. 718; Blunt 1960, iv-vii; Stewart 1978, A123. BM (BMC 718); ex Sotheby, 31 August 1848, lot 34; ex 8th Earl of Pembroke (d 1733).


Edward the Elder (899–924), gold mancus, struck from penny dies of HCT 1 type (Early II, Winchester style), c 910, moneyer Deorwald (CTCE, no. 144). Wt 4.80 g, pierced. Lausanne, Musée Cantonal; found at Lutry, near Lausanne 1909. Blunt 1948; Stewart 1978, A125; Blunt, Stewart and Lyon 1989, 64, no. 144, plate 1,30. From same obverse as penny in BM (Grueber and Keary 1887–93, II, 90, no. 31; illus below B6A).


Æthelred II (978–1016), gold penny mancus, struck from penny dies of Helmet type, c 1003–09, Lewes mint, moneyer Leofwine. Wt 3.34 g. BM; bt Rev A Carter; ex Mrs Holroyd; ex Mr Martin; found Hellingly, East Sussex, c 1808. Grueber and Keary 1887–93, II, 208, no. 1; Allen 1948, 266; King 1957, 518; Dolley 1964, frontispiece B; Stewart 1978, A126; Backhouse et al 1984, no. 205. From the same obverse die as a silver penny in the BM, wt 1.27 g (King 1957, no. 61; Backhouse et al 1984, no. 204; illus below B7A); the silver penny appears to have been the earlier striking, to judge from the condition of the obverse die (Lyon 1969, 209). That coin’s reverse die was apparently used, without the addition of a pellet in each of three quarters, to strike a coin in Stockholm, Hildebrand 1460, which weighs 1.50 g (King 1957, no. 60). These two coins thus span a decrease in weight standard early in the Helmet issue.


Edward the Confessor (1042–66), gold mancus, struck from penny dies of Expanding Cross type, light issue, c 1050–53, Warwick mint, moneyer Lufinc; specimen in silver not known. Wt 3.51 g. BM; gift Mrs R C Lockett 1951; ex T H Spurrier and descendants; found Worcester (St Clement’s Church, during demolition), c 1824. Allen 1948; Dolley 1964, frontispiece C; Stewart 1978, A127.



Figures 5.6 Gold ingots and hack-gold from England: C1-C10.



APPENDIX C. INVENTORY OF GOLD INGOTS AND HACK-GOLD FROM ENGLAND C1. Norwich (Library Car Park Excavations), Norfolk, 1999. Gold ingot, in the form of an oblong bar of sub-rectangular section, with transverse hammering on its upper face creating parallel ridges across the ingot. It tapers slightly to rounded ends. No nicks or pecks. Length 36 mm, width 6 mm, height c 3 mm. Wt 7.06 g, c 85 per cent gold (SG 16.89, which would be equivalent to 83 per cent gold, assuming a gold-silver alloy, but a non-quantitative X-ray analysis, because of its porous surface, revealed some copper content, in addition to gold and silver. This would have the effect of decreasing the SG, thus after adjustment c 85 per cent gold is a fair estimate.) Treasure Annual Report 2000, no. 66; Percival and Hutcheson, forthcoming. Norwich Castle Museum. The distinctive transverse hammering on this ingot is found on Viking-Age silver ingots. In England it is a scarce technique, occurring on only 16 of the c 360 known ingots (most data here cited from Kruse 1988). Twelve of these were in the Cuerdale hoard (dep c 905), two were in the Chester 1950 hoard (dep c 970), one was a single-find from Ditchingham, Norfolk (Blackburn and Rogerson 1993), and one is among the group of ingots and pieces of hack-gold and hack-silver from Torksey cited below (C5–6). If this Torksey group can be associated with the army’s wintering of 872/73, it would provide the earliest find context for a transverse-hammered ingot in the British Isles or Scandinavia. In Ireland, two gold transverse-hammered ingots have been found, one at Askeaton (Co Limerick) now in the NMI (1929:1332) and one from the excavations at Wood Quay in Dublin (Cahill and Ó Floinn 1995) and a silver transverse-hammered ingot was found at Feltrim Hill (Co Down) (NMI 1947:426; Cahill 1993, fig 3). In Scandinavia, transverse-hammered ingots are widely spread during the second half of the tenth and the eleveth centuries. They occur in southern Swedish hoards from the mid-tenth century, and extensively in hoards from Gotland after 1000. Several Latvian hoards of the eleventh century have long ingots or rods with similar hammering (Hårdh 1996, 144, fig 33). In Denmark, they occur mainly in hoards deposited after the mid-tenth century, and during the last quarter of the century they are the most common type of ingot. The earliest occurrence of the technique is therefore insular, but whether these ingots were produced in the British Isles or in Western Scandinavia is an open question. The Norwich ingot’s dating is somewhat uncertain for, although found in an excavation, it was in an unstratified context (Hutcheson 2000). The technique indicates that it is of Viking-Age date, and within the range c 850–1050. The most likely date is the later ninth or early tenth century when the Southern Danelaw had a dual economy – a bullion economy subsisting alongside an emerging regulated monetary economy – a transitional state that ended in the 920s (Blackburn 2001; Graham-Campbell 2001). But it could have arrived somewhat later, as an import from Denmark in the later tenth century, and served merely as a piece of a goldsmith’s stock of bullion to be used for metalworking. C2. Fenstanton, Cambs, 1999, metal-detector find by Tim Jackson, declared Treasure on 6 July 2000. Gold ingot, in the form of an oblong bar of D-section, with rounded ends, cast and unworked. No nicks or pecks. Width (max) 11 mm, height 7.5 mm, length 51.3 mm.



Wt 38.57 g. Analysis (X-ray spectrography, at BM): 77 per cent gold, 15 per cent silver, 7 per cent copper. Treasure Annual Report 1998–99, no. 82. FM. This simple cast ingot is typical of many silver ingots from Viking-Age hoards in the British Isles and Scandinavia, for example Croydon hoard, dep c 872; Cuerdale hoard, dep c 905 or Chester 1950 hoard, dep c 970. In gold it is paralleled by the ingot from West Dereham (C3), the terminal fragment from Torksey (C5), and ones reported from near Dublin and the ‘Hebridean’ hoard. Such simple ingots could in theory have been made in any period, but in practice where known their context is almost always of Viking-Age date. This specimen, as a metal-detector find, is without a datable context, for no other Viking-Age or Anglo-Saxon finds have been recorded from the same or adjoining fields. Yet it was found in a field adjoining the River Ouse, which would have led through the river system out to the Wash and the North Sea – a plausible location for a Viking loss. As with the Norwich find (C1), the most likely date of loss for an ingot found in the Southern Danelaw is c 865–925. The composition of the metal, with c 7 per cent copper, was a cause for concern initially, since early medieval gold alloys, at least those of the fifth–seventh centuries, do not normally contain significant amounts of copper. However, recent analyses of items in the Hoen hoard have demonstrated that Carolingian and Scandinavian gold artefacts of the ninth century do often have 5–10 per cent copper, with gold and silver, and the Norwich ingot also has a copper element in its composition. C3. West Dereham, Norfolk, October 2002. Metal-detector find by M Webb. Gold ingot, in the form of an oblong bar of D-section, with rounded ends, cast and unworked. No nicks or pecks evident from photographs. Width (max) 10 mm, height 6 mm, length 76 mm. Wt 58.2 g. Analysis (X-ray spectrography, at BM): c 79 per cent gold. Treasure Annual Report 2002, no. 36. Norwich Castle Museum. See comment under C2. C4. Springthorpe, nr Gainsborough, Lincs, 16 January 2000. Metal-detector find by S Trainer. Fragment of a sub-rectangular gold bar/ingot with surfaces worked to some extent by hammer; the fragment chisel cut at both ends from above and below. No nicks or pecks. Width 7.5 mm, height 6 mm, length 13 mm. Wt 6.7 g. Analysis (X-ray fluorescence, at BM): 93 per cent gold, 6 per cent silver, 1 per cent copper. Treasure Annual Report 2000, no. 69. Lincoln City and County Museum. C5. Torksey, Lincs, 1995 or earlier; assumed to be from the ‘productive’ site; sold to the trade with C6 and seven cut fragments of silver ingots or ornaments, presumably part of a hoard. Gold ingot terminal fragment, chisel cut from a bar of oval section, with rounded end. No signs of nicks or pecks. Width 8 mm, length 7 mm. Wt not recorded. Dispersed in trade, 1995. This cast ingot is similar to C2. The silver pieces in the group from which this and C6 came comprised one small complete transverse-hammered ingot, two hammered ingot terminals, two cast unworked ingot terminals, a section of broad-band penannular armring and a section of spiral- or ‘Permian’ ring. Apart from the transverse-hammered ingot and the gold, these same elements were present in the Croydon hoard, the deposition of which has been associated with the Viking army’s activities around London the year



before it came to Torksey (Brooks and Graham-Campbell 2000). Many of the finds of coins and other metalwork from the site point to their association with the wintering of the army at Torksey in 872/73 (Blackburn 2002). This group of ingots, hack-gold and hack-silver appears to represent the whole or part of a hoard that was probably deposited on that occasion. C6. Torksey, Lincs, 1995 or earlier; see C5. Gold rod fragment, of octagonal section having been hammered flat on all surfaces, chisel cut at both ends. No signs of nicks or pecks. Width and height 6 mm, length 21 mm. Wt not recorded. Dispersed in trade, 1995. The rod from which this was cut had probably been formed into a simple arm-ring. James Graham-Campbell (pers comm) draws a parallel with the large silver arm-ring from the Cuerdale hoard that had been converted into a buckle (illustrated in Bjørn and Shetelig 1940, fig 11, middle). C7. Torksey, Lincs, mid-1990s, from the ‘productive’ site. Gold rod fragment, circular section with some flattened surfaces, chisel cut at each end, with a twist and curve suggestive it has been cut from a twisted- or plaited-rod gold neck- or arm-ring. No nicks or pecks. Diam 4 mm, length 9 mm. Wt 1.30 g, c 44 per cent gold (SG analysis, assuming a gold-silver alloy, but higher if significant copper element present). FM. The distinctive twist and curve on this fragment and the signs of working on its surface suggest that it was cut from an item similar to the twisted-rod gold arm-ring from the Sound of Jura (Graham-Campbell 1995, S24, plate 72, c–d; note particularly the very similar working and twisting of the gold rod, which is of comparable diameter, 4–5 mm). The plaited-rod gold arm-ring from Oxna (Graham-Campbell 1995, S11, plate 72, a–b), by contrast, is made of thinned rods (1.5–2 mm) of more even circular section, but there is great variety in the size and construction of these rings, and the rods can taper enormously between their centre and ends, as on the Hoen neck- and arm-rings (GrahamCampbell 1999). It is difficult to judge from the shallow curve of this fragment whether it comes from an arm-ring or a neck-ring and whether it made of simple twisted rods or the more complex plaited construction. Graham-Campbell dates the Sound of Jura arm-ring to the tenth century, but the Hoen hoard shows that both twisted and plaited gold neckand arm-rings go back to the second half of the ninth century, and Graham-Campbell (1999) has argued that they were made in Southern Scandinavia (Denmark or Norway) rather than Russia, as had previously been thought. C8. Torksey, Lincs, 1995, from the ‘productive’ site; found in association with two small cut-terminals of silver ingots. Gold rod fragment, of pentagonal or hexagonal section, with a slight curve and chisel cut at both ends. Diam 2.5 mm, length 11 mm. Wt 0.66 g, 62 per cent gold (SG analysis, assuming a gold-silver alloy, but higher if significant copper element present). Blackburn 2002, plate 5, d. FM. This fragment is unlikely to have come from the same object as C6, as it has a higher fineness, is thinner, with a more distinct cross-section, and shows only slight twisting. Even though the rods in such arm- or neck-rings can taper considerably and vary in cross section, this piece appears to be of different workmanship and from a different batch of



gold. Whether it comes from some kind of twisted-rod arm-ring another artefact is not clear. Nonetheless, like C7 the piece is cleanly cut at both ends, apparently to form a small piece of bullion, and it may not be a coincidence that C8 is precisely half the weight of C7. Although from the same productive site, these two pieces were discovered by different finders on different occasions (though we do not know the precise locations), and they may well not have been associated losses. C8 was, however, found within a few metres of two silver ingot terminals, of small rectangular cross section weighing 1.36 g and 1.15 g. These items are on a different scale from the nine pieces discussed above (under C5), and I would judge them to be from a separate deposit. C9. West Wratting, Cambs, January 2003, metal-detector find by Mr L Eeles, declared Treasure, April 2003. Fragment of a gold ingot or arm-ring, chisel-cut from a broad band of rectangular section, decorated with close parallel hammer marks on both surfaces and the two sides. The band tapers and thins slightly in one direction. It has been cut at both ends with a sharp chisel from one side; a preliminary attempt to cut it at the wider end has left a parallel channel close to the severed surface. The band is fairly flat, except at its wider end where the force of the chisel blow has caused some bending, with stress cracks visible on the sides. No nicks or pecks, but some scuffing on one surface, which may have been more exposed to damage if the band was originally penannular. Length 25–26.5 mm, width 10 mm tapering to 9 mm, thickness 3.5 mm narrowing to 2.5 mm. Wt 11.37 g, c 72 per cent gold (SG analysis (15.71) at FM, assuming a gold-silver alloy, but higher if significant copper element present). Treasure Annual Report 2003, no. 81. FM. This piece of hack-gold has transverse-hammered decoration of similar technique to the Norwich ingot (C1) and others cited above. However, the present piece has been more thoroughly and more finely worked than most hammered ingots, to form a thinner and more uniform band of metal. Particular care has been taken to decorate the sides in the same manner as the main faces. The width and thickness of the band is similar to that of many penannular broad-band arm-rings, and the slight tapering and thinning of the piece suggests that it may have been cut from such an object. This form of arm-ring probably originated in Denmark in the mid-ninth century, but was copied and developed by the Scandinavians in Ireland during the later ninth and early tenth centuries (Brooks and Graham-Campbell 2000, 75–77; Sheehan 1998a, 194–197). Most of the Scandinavian and Irish arm-rings of this type have punched decoration covering the length of their outer surface, in some cases with parallel transverse grooves giving an effect somewhat analogous to transverse hammering. A few broad-band arm-rings are entirely plain, such as one from the ninth-century Sønder Kirkeby hoard from Denmark or a group in gold from Ireland that have been reattributed to the Viking period (Ó Floinn 1983, 7–8; Sheehan 2000a, 34–35). I am not aware of any complete broad-band arm-ring from the British Isles that has hammered decoration, as on this piece, though from the Skaill hoard there is a silver hammered band that has been bent double, the function of which is ambiguous (Graham-Campbell 1995, 40 and 102, no. 24,109). There are comparable pieces also in silver from Scandinavia, especially Mainland Sweden and Gotland, in hoards of the later tenth and eleventh centuries, and these are regarded as ‘broad-band hammered ingots’ (eg Stenberger 1947–58, II, plates 182,9 and 191,21). However, the distinction between an ingot and an arm-ring is not as sharp as one might imagine, since arm-rings were formed from ingots, often of an adjusted weight, and they retained the function of being a store of



wealth suitable for cutting up or melting down when needed as money. I am inclined to think that the object this piece was cut from was either a finished arm-ring or one in the course of preparation. A probable dating to the later ninth or early tenth century, suggested by its context in the Southern Danelaw, would not be inconsistent with the transverse hammering. C10. Trewhiddle, Cornwall, 1774, from a hoard of coins and ornamental metalwork, deposited c 868. Small gold ingot, hexagonal with tapering sides. Wt not recorded, now lost. Wilson and Blunt 1961, 86, plate xxvi, c; Blackburn and Pagan 1986, no. 59.

ACKNOWLEDGEMENTS I have benefited from the help and advice of a number of scholars who have read drafts of this paper: James Graham-Campbell and Leslie Webster particularly on finds of AngloSaxon and Scandinavian metalwork, Stewart Lyon on matters of metrology and weight systems, and Hugh Pagan, Lord Stewartby and Gareth Williams on numismatics generally. Gareth Williams also kindly supplied illustrations of several coins in the British Museum’s collection. May Sinclair provided information and photographs about the new Cenwulf mancus when it was first found. I am grateful to them all.

Notes 1. The term ‘mancus’ (discussed below, p 57) can have several different meanings, including a gold coin, an amount of gold of a particular weight or a value equivalent to 30 silver pence. 2. For this calculation I have assumed Eadred had 10 bishops, seven ealdormen, one seneschal, one keeper of the wardrobe, one butler, five mass priests and 10 other office holders in the royal household, which may well be a considerable underestimate. I have taken the mancus as equal to 30 silver pennies. For the translation into modern money, I have assumed a tenth-century penny to have a purchasing power of about £10, which is a conservative estimate based on the provision in VI Athelstan 6.1–2, where compensation for loss of a sheep is reckoned five pence, a pig 10 pence, a cow 20 pence, an ox 30 pence and a horse 120 pence (Whitelock 1979, no. 37). 3. Most items in the Appendices are illustrated on Figures 5.3–6

References Akerman, J Y (1842), ‘The gold “mancus” ’, NC 5, 122–124 Allan, J (1914), ‘Offa’s imitation of an Arab dinar’, NC 4th series 14, 77–89 Allen, D F (1948), ‘Edward the Confessor’s gold penny’, BNJ 25, 259–276 Archibald, M.M. (1990), ‘Pecking and bending: the evidence of British finds’, in K Jonsson and B Malmer (eds), Sigtuna Papers. Proceedings of the Sigtuna Symposium on Viking-Age Coinage, 1–4 June 1989, Kungl. Vitterhets Historie och Antikvitets Akademien and Spink, Stockholm, 11–24 Archibald, M M (1991), ‘Anglo-Saxon and Norman lead objects with official coin types’, in A G Vince (ed), Aspects of Saxon and Norman London 2: Finds and Environmental Evidence (London and Middlesex Archaeol Soc Special Paper 12), London, 326–346 Archibald, M M and Dolley, M (1984), ‘The coins’, in J P Allen (ed), Medieval and PostMedieval Finds from Exeter, 1971–1980 (Exeter Archaeological Reports 3), Exeter, 249–251



Backhouse, J, Turner, D H and Webster, L (eds) (1984), The Golden Age of Anglo-Saxon Art 966–1066, British Museum, London Baker, J R (2002), ‘Coinage, Monetary Policy and Monetary Economy in Greece, 1204–ca 1350’, unpublished PhD dissertation, University of Birmingham Besteman, J (1997), ‘Die Vondst van Westerklief, Geemente Wieringen: Een Zilverschat uit de Vikingperiode’, Oudheidkundige Medelingen uit het Rijksmuseum van Oudheded te Lejden 77, 199–226 Biddle, M (1984), ‘London on the Strand’, Popular Archaeology 6.1, 23–27 Biddle, M, Blunt, C E, Kjølbye-Biddle, B, Metcalf, M and Pagan, H (1986), ‘Coins of the AngloSaxon period from Repton, Derbyshire: II’, BNJ 56, 16–34 Bjørn, A and Shetelig, H (1940), Viking Antiquities in Great Britain and Ireland. Part IV: Viking Antiquities in England, with a Supplement of Viking Antiquities on the Continent of Western Europe, H Aschehong & Co, Oslo Blackburn, M (2001), ‘Expansion and control: aspects of Anglo-Scandinavian minting south of the Humber’, in J Graham-Campbell, R Hall, J Jesch and D N Parsons (eds), Vikings and the Danelaw. Select Papers from the Proceedings of the Thirteenth Viking Congress, Oxbow Books, Oxford, 125–142 Blackburn, M (2002), ‘Finds from the Anglo-Scandinavian site of Torksey, Lincolnshire’, in B Paszkiewicz (ed), Moneta Mediævalis. Studia numizmatyczne i historyczne ofiarowane Profesorowi Stanislawowi Suchodolskiemu w 65. rocznice˛ urodzin, Instytut Archeologii i Etnologii Polskiej Akademii Nauk, Warsaw, 89–101 Blackburn, M (2003), ‘ “Productive” sites and the pattern of coin loss in England, AD 600–1180’, in T Pestell and K Ulmschneider (eds), Markets in Early Medieval Europe. Trading and ‘Productive’ Sites, 650–850, Windgather Press, Macclesfield, 20–36 Blackburn, M (2005a), ‘Coinage and contacts in the North Atlantic during the 7th to mid-10th centuries’, in S V Arge and A Mortensen (eds), Viking and Norse North Atlantic: Select Papers from the Proceedings of the Fourteenth Viking Congress, 2001, Tórshavn, 112–120 Blackburn, M (2005b) ‘Presidential Address 2004. Currency under the Vikings. Part 1: Guthrum and the earliest Danelaw coinages’, BNJ 75, 18–43 Blackburn, M (2006), ‘Two new types of Anglo-Saxon gold shillings’, in B Cook and G Williams (eds), Money and History in the North Sea World, AD 500–1200, Brill, Leiden, 127–140 Blackburn, M (forthcoming), ‘The loops and graffiti as a guide to how and when the coins were acquired’, in S H Fuglesang and D M Wilson (eds), The Hoen Hoard. A Viking Gold Treasure of the ninth Century, Norwegian Institute in Rome, Rome Blackburn, M and Bonser, M (1986), ‘Single finds of Anglo-Saxon and Norman coins – 3’, BNJ 56, 64–101 Blackburn, M and Bonser, M (1990a), ‘A Carolingian gold coin struck from a die of Chartres and found at Congham, Norfolk’, NCirc, 304–306 Blackburn, M and Bonser, M (1990b), ‘Une monnaie d’or carolingienne inédite portant le nom de l’atelier de Chartres’, BSFN (April 1990), 796 Blackburn, M and Pagan, H E (1986), ‘A revised check-list of coin hoards from the British Isles, c. 500–1100’, in M A S Blackburn (ed), Anglo-Saxon Monetary History, Leicester University Press, Leicester, 291–313 Blackburn, M and Rogerson, A (1993), ‘Two Viking-age silver ingots from Ditchinham and Hindringham, Norfolk: the first ingot finds from East Anglia’, Med Arch 37, 222–224 Blunt, C E (1948), ‘A gold penny of Edward the Elder’, BNJ 25, 277–281 Blunt, C E (1960), ‘The President’s address: Ecclesiastical coinage in England (Part 1. To the Norman Conquest)’, NC 6th series 20, i–xvii Blunt, C E (1961), ‘The coinage of Offa’, in R H M Dolley (ed), Anglo-Saxon Coins, Methuen, London, 39–62 Blunt, C E and Dolley, M (1968), ‘A gold coin of the time of Offa’, NC 7th series 8, 151–160 Blunt, C E, Lyon, C S S and Stewart, B H I H (1963), ‘The coinage of southern England, 796–840’, BNJ 32, 1–74



Blunt, C E, Stewart, B H I H and Lyon, C S S (1989), Coinage in Tenth-Century England, British Academy Oxford Bøe, J (1940), Viking Antiquities in Great Britian and Ireland. Part III: Norse Antiquities in Ireland, H Aschehoug & Co, Oslo Bornholdt Collins, K A (2003), ‘Viking-Age Coin Finds from the Isle of Man: a Study of Coin Circulation, Production and Concepts of Wealth’, unpublished PhD dissertation, University of Cambridge Bonser, M J (1997), ‘Fifteen years of coin finds from productive sites’, Yorkshire Numismatist 3, 39–45 Brooks, N P (1978), ‘Arms, status and warfare in late-Saxon England’, in D Hill (ed), Ethelred the Unready (BAR, British Series 59), Oxford, 81–103 Brooks, N P and Graham-Campbell, J (2000), ‘Reflections on the Viking-Age silver hoard from Croydon, Surrey’, in N Brooks, Communities and Warfare 700–1400, Hambledon Press, London, 69–92 Brown, H (1990), ‘A dinar for St Frideswide?’, The Ashmolean 18 (Summer/Autumn 1990), 9–10 Cahill, M (1993), ‘Some unrecorded Bronze Age gold ornaments from Co. Limerick’, N Munster Antiq J 35 (1993–94), 5–23 Cahill, M (1994), ‘Mr Anthony’s bog oak case of gold antiquities’, PRIA 94C, 53–109 Cahill, M and Ó Floinn, R (1995), ‘Two silver kite brooches from near Limerick city’, N Munster Antiq J 36, 65–82 Carlyon-Britton, P W P (1908), ‘The gold mancus of Offa, king of Mercia’, BNJ 5, 55–72 Chadwick, H M (1905), Studies on Anglo-Saxon Institutions, Cambridge University Press ‘Coin Register’: an annual section in BNJ Cook, B J (1999a), ‘The bezant in Angevin England’, NC 159, 255–275 Cook, B J (1999b), ‘Foreign coins in medieval England’, in L Travaini (ed), Local Coins, Foreign Coins: Italy and Europe 11th–15th Centuries, Milan, 231–284 Coupland, S (1997), Review of J Besteman, Vikingen in Noord-Holland? (Haarlem, 1996), in NC 157, 273–274 Coupland, S (2006), ‘Between the devil and the deep blue sea: hoards in 9th-century Frisia’, in B Cook and G Williams (eds), Money and History in the North Sea World, AD 500–1200, Brill, Leiden, 241–266 Cubbon, A M (2003), ‘A Viking-age plaited gold finger-ring from Greeba, Isle of Man’, IMNHASP 11:2 (1999–2001), 249–257 Dodwell, C R (1982), Anglo-Saxon Art: a New Perspective, Manchester University Press Dolley, M (1964), Anglo-Saxon Pennies, British Museum, London Dolley, R H M and Morrison, K F (1966), The Carolingian Coins in the British Museum, British Museum, London Duplessy, J (1956), ‘La circulation des monnaies arabes en Europe occidentale du VIIIe au XIIIe siècle’, RN 5th series 18, 101–163 Evans, A C (1986), The Sutton Hoo Ship Burial, British Museum, London Fillon, B (1853), Lettres à M. Ch. Dugast-Matifeux sur quelques monnaies françaises inédites, Paris Gariel, E (1883–84), Les monnaies royales de France sous la race carolingienne, Strasbourg Graham-Campbell, J (1974), ‘A Viking-Age gold hoard of Ireland’, Antiq J 54, 269–272 Graham-Campbell, J (1976), ‘The Viking-Age silver hoards of Ireland’, in B Almqvist and D Greene (eds), Proceedings of the 7th Viking Congress, Dublin, 15–21 August 1973, Royal Irish Academy, Dublin, 39–74 Graham-Campbell, J (1980), Viking Artefacts: a Select Catalogue, British Museum, London Graham-Campbell, J (1983), ‘The Viking-age silver hoards from the Isle of Man’, in C Fell, P Foote, J Graham Campbell and R Thomson (eds), The Viking Age in the Isle of Man, Viking Society for Northern Research, London, 53–80



Graham-Campbell, J (1988), ‘The gold finger-ring from a burial in St Aldate’s Street, Oxford’, Oxoniensia 53, 263–266 Graham-Campbell, J (1989), ‘The coinless hoard’, in H Clarke and E Schia (eds), Coins and Archaeology (BAR, International series 556), Oxford, 53–61 Graham-Campbell, J (1995), The Viking-Age Gold and Silver of Scotland (AD 850–1100), National Museums of Scotland, Edinburgh Graham-Campbell, J A (1999), ‘Rings and things. Some observations on the Hon hoard’, in G Fellows-Jensen and N Lund (eds), Beretning fra attende tværfaglige vikingesymposium, Forlaget Hikuin, Højbjerg, 53–64 Graham-Campbell, J A (2001), ‘The Dual Economy of the Danelaw. The Howard Linecar Memorial Lecture 2001’, BNJ 71, 49–59 Graham-Campbell, J (2005), ‘The Viking-age gold and silver of the North Atlantic region’, in S V Arge and A Mortensen (eds), Viking and Norse North Atlantic: Select Papers from the Proceedings of the Fourteenth Viking Congress, 2001, The Faroese Academy of Sciences, Tórshavn, 125–140 Graham-Campbell, J (forthcoming), The Cuerdale Hoard and related Viking-Age Silver from Britain and Ireland in the British Museum, British Museum, London Grierson, P (1951), ‘The gold solidus of Louis the Pious and its imitations’, Jaarboek voor Munt- en Penningkunde 38, 1–41 (reprinted in Grierson 1979, article 22) Grierson, P (1954), ‘Carolingian Europe and the Arabs: the myth of the mancus’, Revue Belge de Philolgie et d’Histoire 32, 1059–1074 (reprinted with important revisions in Grierson 1979, article 3, and Addenda, 2–4) Grierson, P (1979), Dark Age Numismatics, Variorum Reprints, London Grierson, P and Blackburn, M (1986), Medieval European Coinage 1. The Early Middle Ages (5th–10th Centuries), Cambridge University Press Grueber, H A and Keary, C F (1887–93), A Catalogue of English Coins in the British Museum. Anglo-Saxon Series, 2 Vols, British Museum, London Gunstone, A J H (1992), South-Eastern Museums. Ancient British, Anglo-Saxon, and Later Coins to 1279 (SCBI 42), Oxford University Press Haertle, C M (1997), Karolingische Münzfunde aus dem 9 Jahrhundert, Böhlau Verlag, Cologne, Weimar and Vienna Hårdh, B (1996), Silver in the Viking Age. A Regional-Economic Study (Acta Archaeologica Lundensia, series in 8°, no. 25), Almquist & Wiksell International, Stockholm Hawkes, S C, Merrick, J M and Metcalf, D M (1966), ‘X-ray fluorescence analysis of some Dark Age coins and jewellery’, Archaeometry 9, 98–138 Hill, P (1997), Whithorn and St Ninian. The Excavation of a Monastic Town, 1984–91, The Whithorn Trust/Sutton Publishing, Stroud Hinton, D A (1974), A Catalogue of the Anglo-Saxon Ornamental Metalwork 700–1100 in the Department of Antiquities, Ashmolean Museum, Oxford University Press Hinton, D A (1975), ‘Late Anglo-Saxon metal-work: an assessment’, Anglo-Saxon England 4, 171–180 Hinton, D A (1978), ‘Late Saxon treasure and bullion’, in D Hill (ed), Ethelred the Unready (BAR, British series 59), Oxford, 135–158 Hutcheson, A R J (2000), ‘The French Borough’, Current Archaeology 170, 64–69 Ilisch, L (2004), ‘Die imitativen solidi mancusi. “Arabische” Goldmünzen der Karolingerzeit’, in R Cunz (ed), Fundamenta Historiae. Geschichte im Spiegel der Numismatik und ihrer Nachbarwissenschaften. Festschrift für Niklot Klüssdendorf zum 60. Geburtstag am 10. Februar 2004, Hannover, 91–106 Keynes, S (1991), ‘The historical context of the Battle of Maldon’, in D Scragg (ed), The Battle of Maldon AD 991, Basil Blackwell, Oxford, 81–113 Keynes, S and Lapidge, M (1983), Alfred the Great, Penguin Books, Harmondsworth King, H H (1957), ‘The coins of the Sussex mints. Part III: Lewes’, BNJ 28 (1955–57), 518–536.



Kruse, S (1988), ‘Viking-Age Ingots from England and Wales and their Economic Implications’, unpublished PhD thesis, University College London Longpérier, A de (1842), ‘Remarkable gold coin of Offa’, NC 4, 232–234 Lowick, N M (1973), ‘A new type of solidus mancus’, NC 7th series 13, 173–182 Lyon, C S S (1969), ‘Historical problems of Anglo-Saxon coinage – (3) denominations and weights’, BNJ 38, 204–222 Lyon, C S S (1976), ‘Some problems in interpreting Anglo-Saxon coinage’, Anglo-Saxon England 5, 173–224. Malmer, B (1989), The Sigtuna Coinage c.995–1005 (Commentationes de Nummis Saeculorum IX–XI in Suecia repertis Nova Series 4), Kungl. Vitterhets Historie och Antikvitets Akademien and Spink, Stockholm Malmer, B (1997), The Anglo-Scandinavian Coinage c.995–1020 (Commentationes de Nummis Saeculorum IX–XI in Suecia repen Nova Series 9), Kungl. Vitterhets Historie och Antikvitets Akademien and Spink, Stockholm McCormick, M (2001), Origins of the European Economy. Communications and Commerce, AD 300–900, Cambridge University Press Metcalf, D M (1958), ‘Eighteenth-century finds of medieval coins from the records of the Society of Antiquaries’, NC 6th series 18, 73–96 Metcalf, D M (1988), ‘The coins’, in P Andrews (ed), The Coins and Pottery from Hamwic (Southampton Finds, Vol 1), Southampton City Museums, Southampton, 17–59 Moesgaard, J C (forthcoming), ‘The Vikings on the Continent: the numismatic evidence’, in I Skipsted Klæsø (ed), The Vikings on the Continent Moesgaard, J C and Rispling, G (2002), ‘Guldefterligning af en arabisk mønt fundet på Bornholm’, NNUM, 91–94 Nightingale, P (1984), ‘The ora, the mark, and the mancus: weight-standards and the coinage in 11th-century England. Part 2’, NC 144, 234–248 Ó Floinn, R (1983), ‘A gold band found near Rathkeale, Co. Limerick’, North Munster Antiq J 25, 3–8 Pagan, H E (1965), ‘A third gold coin of Mercia’, BNJ 24, 8–10 Pagan, H E (1988), ‘The imitative Louis the Pious solidus from Southampton and finds of other related coins in the British Isles’, in P Andrews (ed), The Coins and Pottery from Hamwic (Southampton Finds, Vol 1), Southampton City Museums, Southampton, 71–72 Pegge, S (1772), An Assemblage of Coins Fabricated by Authority of the Archbishops of Canterbury, London Percival, J W and Hutcheson, A R J (forthcoming), Excavations within the French Borough (Between Theatre Street and Bethel Street), Norwich, 1998–99 (East Anglian Archaeology) Prou, M (1896), Catalogue des monnaies françaises à la Bibliothèque Nationale. Les monnaies carolingiennes, Rollin and Feuardent, Paris Sawyer, P H (1968), Anglo-Saxon Charters. An Annotated List and Bibliography, Royal Historical Society, London Sheehan, J (1998a), ‘Early Viking Age silver hoards from Ireland and their Scandinavian elements’, in H B Clarke, M Ní Mhaonaigh and R Ó Floinn (eds), Ireland and Scandinavia in the Early Viking Age, Four Courts Press, Dublin, 166–202 Sheehan, J (1998b), ‘Viking Age hoards from Munster: a regional tradition?’, in M A Monk and J Sheehan (eds), Early Medieval Munster: Archaeology, History and Society, Cork University Press, 147–163 Sheehan, J (2000a), ‘Viking Age silver and gold from County Clare’, in C Ó Murchadha (ed), County Clare Studies: Essays in Memory of Gerald O’Connell, Seán Ó Murchadha, Thomas Coffey and Pat Flynn, Ennis, 30–41 Sheehan, J (2000b), ‘Ireland’s early Viking-Age silver hoards: components, structure, and classification’, in S Stummann Hansen and K Randsborg (eds), Vikings in the West (Acta Archaeologica 71), Copenhagen, 49–63



Smith, R A (1923), A Guide to the Anglo-Saxon and Foreign Teutonic Antiquities in the Department of British and Medieval Antiquities, British Museum, London; reprinted as British Museum Guide to Anglo-Saxon Antiquities 1923, Ipswich, 1993. Spufford, P (1988), Money and its Use in Medieval Europe, Cambridge University Press Stenberger, M (1947–58), Die Schatzfunde Gotlands der Wikingerzeit, 2 Vols, Kungl. Vitterhets Historie och Antikvitets Akademien, Stockholm Stewart, I (1978), ‘Anglo-Saxon gold coins’, in R A G Carson and C M Kraay (eds), Scripta Nummaria Romana. Essays Presented to Humphrey Sutherland, Spink, London, 143–172 Stewart, I (1986), ‘A solidus from Yorkshire’, BNJ 56, 182–183 Sutherland, C H V (1948), Anglo-Saxon Gold Coinage in the Light of the Crondall Hoard, Oxford University Press Tite, C (1992), ‘Sir Robert Cotton and the gold mancus of Pendræd’, NC 152, 177–181 Tweddle, D, Moulden, J and Logan, E (eds) (1999), Anglian York: A Survey of the Evidence (Archaeology of York 7/2), CBA, York Van der Meer, G (1994), ‘An early seventeenth-century inventory of Cotton’s Anglo-Saxon coins’, Jaarboek voor Munt- en Penningkunde 81, 215–233 Vince, A (1990), Saxon London: an Archaeological Investigation, Seaby, London Webster, L and Backhouse, J (eds) (1991), The Making of England. Anglo-Saxon Art and Culture AD 600–900, British Museum, London Welin, U S L (1964), ‘Elias Brenners arabiska guldpenning och andra dinarer funna i Sverige’, NNÅ, 5–25 Whitelock, D (1930), Anglo-Saxon Wills, Cambridge University Press Whitelock, D (1979), English Historical Documents I. c.500–1042, 2nd edn, Eyre Methuen, London Williams, G (1999), ‘Anglo-Saxon and Viking coin weights’, BNJ 59, 18–36 Wilson, D M (1964), Anglo-Saxon Ornamental Metalwork, 700–1100, in the British Museum, British Museum, London Wilson, D M (2004), ‘A ring from Greeba’, IMNHASP 11:3 (2001–03), 437–439 Wilson, D M and Blunt, C E (1961), ‘The Trewhiddle hoard’, Archaeologia 98, 75–122


A Survey of Coin Production and Currency in Normandy, 864–945 Jens Christian Moesgaard

When the Viking chief Rollo received the lower Seine Valley from the Frankish king Charles the Simple in 911/12, he faced the problem of organising his new lands. Should he introduce the institutions he knew from back home in Scandinavia, or should he draw on local traditions? Coinage is one of the fields in which one can examine what he and his followers did. The purpose of this chapter is to provide a survey of the currency and coin production during the first decades after the Vikings gained control of the area that was later to be known as Normandy. It is necessary to treat the monetary situation just before the arrival of the Vikings in some detail so as to provide the background for the subsequent evolution. It is a matter of fact that many issues carried the same type for decades (the so-called types immobilisés) and that imitations were frequent, so one must examine the material – the coins themselves and the hoards – most thoroughly to sort out the internal chronology of the issues and the evolution in the structure of the currency. It should then be possible, by comparing the monetary situation before and after the arrival of the Vikings, to answer the question: what effect did the Viking take-over have on coinage and currency? Did it mean rupture or continuation? In this chapter, the word Viking will be used for the Scandinavians before their settlement in Francia, whereas Norman will be used for the same people after their installation.

COIN PRODUCTION, 864–77 The last nation-wide Carolingian issue was the GDR-type in the name of Charles the Bald (840–77), introduced by the Edict of Pîtres (dép Eure), in 864. It has a KAROLUSmonogram with the inscription GRATIA DEI REX (hence the name of the type) on the obverse, and a cross and the mint name on the reverse (Grierson and Blackburn 1986, 232–233). In those parts of the empire that were to become Normandy, the GDR-type was struck at the following mints: Rouen (dép Seine-Maritime) (MG 866–80), Le Talou (dép Seine-Maritime) (MG 881), Evreux (dép Eure) (MG 882–84), Lisieux (dép Calvados) (MG 885–86), Bayeux (dép Calvados) (MG 887–90), Deux-Jumeaux (dép Calvados) 99



(MG 891–92), Coutances (dép La Manche) (MG 893–94) and Curtisasonien (an unlocated place, probably in dép Orne: see discussions, Dumas 1971, 106–107; Grierson and Blackburn 1986, 635–637; Dumas and Pilet-Lemière 1989, 130, note 2) (MG 895–903). Judging from the number of surviving specimens and the composition of local hoards, Rouen and Curtisasonien were important, Bayeux and Lisieux of medium size, with the remainder being small mints (Table 6.1; Rey 1972, 168–170). Rouen, Evreux, Lisieux, Bayeux and Coutances were cities with episcopal sees (although no coins are known from the Norman sees of Avranches and Sées); Deux-Jumeaux was a monastery, and Le Talou an administrative unit, or pagus (the exact location of its mint is unknown), whereas the status of Curtisasonien is not known at all (the word curtis being part of the name). There is no positive evidence of an ecclesiastical involvement in the coinage, which seems to have been under firm royal control, as is demonstrated by its typological and metrological unity, even in the case of such monastic mints as Deux-Jumeaux. Table 6.1 Hoards containing GDR-coins, with CAROLUS monogram, from Norman mints Commune F-St-Brieuc F-St-Denis F-St-Michel F-Vire F-Imbleville F-Nourray F-Pt-St-Pierre F-Cosne F-Mercurey F-Autun F-Issy F-Compiègne? F-Vrigny F-Savigné F-Saumeray F-Chalo F-Ablaincourt F-Bligny GB-Laxfield F-Glisy F-Montrieux F-Féchain F-Monchy GB-Ashdon F-Montrieux GB-Cuerdale F-Luzy F-Juaye-Mondaye

Buried D.293 D.298 D.314 D.376 D.168(1) D.243 (2) D.110 D.210 D.23 D.172 D.104 D.378 D.341 D.338 D.79 (4) D.47 (5) D.154 D.231 (6) D.217 (7) D.229 (8) D.202 D.177

864–77 864–77 864–77 864–77 869–77 869–77 869–77 876–77 876–77 876–922(3) 876–922(3) 877–79 877–79 877–82 877–82 879–84 879–84 881–87 882–87 887–98 887–98 887–98 c 890–98 c 895 898–922 c 905 c 910/30 c 920 (9)

No. of coins


54Ⳮ 16Ⳮ 458 20Ⳮ 66 18Ⳮ 8Ⳮ 53Ⳮ c 1,700 c 1,500 35Ⳮ 288 5 234Ⳮ 57Ⳮ 240Ⳮ 368Ⳮ 98Ⳮ 9ⳭⳭ 568Ⳮ 672Ⳮ 441 475Ⳮ 71 300Ⳮ c 7,500 ? c 900

– 1 – – 18 – 1 – – – – 39 – 1 – 19 21 10 1 35 8 23 7 1 – – – –

Ta Ev – – – – – – – – – – – – – – – 2 1 – – – – ? – – – – – –

– – – – – – – – – – – 1 – – – – – – – – 1 ? – – – 1 – –





– – – X – – – – – – 5 1 – – 1 1 3 – – – – ? – – X – X –

– – 3 x – 2 – 1 14 – – 1 1 1 – 1 4 7 – 3 10 ? 1 – – – – 21

– – – – – – – – 1 – – – – – – – – – – – – ? – – – – – –

9 – – x – – – – – c 500 20 10 – 3 1 26 5 2 – 3 14 ? 13 2 x 10 x c 700




Table 6.1 Continued F-Rennes F-Evreux F-Fécamp F-Foissy

D.274 D.136(10) D.137 D.144

c 920/23(9) 943/45 980/85 1002/24

136 58 8,781Ⳮ 5

– 1Ⳮ – –

– – – –

– – – –

– – – –

– – 3 –

– – – –

1 – 1 1

Key to mints: Ba ⳱ Bayeux; Cu ⳱ Curtisasonien; DJ ⳱ Deux-Jumeaux; Ev ⳱ Evreux; Li ⳱ Lisieux; Ro ⳱ Rouen; Ta ⳱ Le Talou D ⳱ Duplessy 1985 Notes (1) Post-869, because of coins of Gembloux and Curange/Couvin. (2) CH 1997, no. 65 (NC 1997, 235–236). (3) Issy-l’Evêque and ‘near Autun’ may be parcels of a single hoard. The dating of this hoard (or hoards) is much debated, although it must be placed after 876 because it contains the imperial issue of Nevers. A more precise date depends on that assigned to the coins of Saint-Nazaire-d’Autun. Some authors attribute them to the GDR-issue of Charles the Bald, struck 864–75/77 (Grierson and Blackburn 1986, 542; Haertle 1997, nos 59 and 70), others to Carloman, 882–84 (Duplessy 1985, nos 172 and 23), and yet others to Charles the Simple, 898–922 (Morrison and Grunthal 1967, 298; Armstrong 1998, 164). (4) Morrison and Grunthal 1967, no. 84, unfortunately give a rather imprecise list of part of this hoard, which is much larger (see RN 1965, 271; 1967, 291–295). (5) Dolley and Morrison 1963, 79, no. 11; on the probable deposition date, see Grierson and Blackburn 1986, 235 (coin of Saint-Géry of Cambrai). (6) Dhénin 1985. (7) Blackburn 1990. (8) Hawkins 1843, 57–58, nos 29 and 31; Gariel 1883, 141–142; the reconstruction by Blunt has yet to be published (see Archibald 1992, 20). (9) Dates of deposit based on Lafaurie’s dating of the two-line type of Charles the Simple to the very end of his reign (Lafaurie 1965, 267, 283–284). (10) The coin(s) of Rouen is/are omitted by Duplessy 1985, but mentioned by Longpérier 1870. NB: I have followed Armstrong 1998 for the deposition dates of the hoards, with the exception of Issyl’Evêque/Autun and Imbleville. No hoard containing coins of Coutances or of Curtisasonien in the name of Eudes is on record; no hoards containing Norman coins are recorded either by Haeck 1996 (for Belgium) or by Boeles 1915 (for the Netherlands).

As will become evident below, the future Normandy did not form a single currency zone at that time. Rouen, Le Talou and Evreux belonged to a currency zone comprising present-day Upper Normandy, Picardy, Champagne and Ile-de-France, whereas Lisieux, Bayeux, Deux-Jumeaux, Coutances and Curtisasonien were part of the zone between Loire and Seine. Metcalf calls the latter zone ‘Neustria’, and the former ‘Austrasia’ (Metcalf 1990, 90), but this last designation is somewhat out-dated for the Carolingian period, and it is better to use the term ‘Francia’. The style of the coins seems to indicate that this border was also valid for the organisation of coin production. Indeed, the coins of Curtisasonien, Bayeux, Deux-Jumeaux and Lisieux have some stylistic resemblance to those from Maine and the Middle Loire Valley region, both for the GDR-type of Charles and for the subsequent issues in the names of Louis II and III (877–79–82) and of Eudes (887–98). The coins of Rouen, Evreux and Le Talou are of a different style, although this impression is based only on a rapid examination of photographs of a few published specimens,1 and so it needs to be validated by a thorough examination of a larger body of material. Nevertheless, it suggests that there may have been a central die-cutting workshop for the whole region between the Seine and the Loire.2



COIN PRODUCTION, 877–911/24 In a few mints the GDR-type was replaced in 875/76 by an issue with Charles’ new imperial title (Grierson and Blackburn 1986, 233–235). In other mints, it was replaced by coins in the names of the successors of Charles, after his death in 877. In addition, some other mints kept the GDR-type unchanged with the name of Charles the Bald until well after his death (a so-called type immobilisé). Under Charles the Simple (898–922), the GDR-type in the name of Charles was revived in some mints. None of the mints in the future Normandy struck coins in the name of the successors of Charles the Bald,3 except for Curtisasonien, which struck a very limited issue in the name of Eudes (887–98) (Prou 417; MG 1289).4 What happened during the decades following Charles’ death? Knowing the frequent Viking raids on the Seine valley, one might think that the disarray meant a cessation of coin production. Recent research has, however, suggested that the Carolingian king kept a firm grasp on social structures right up until the concession of the area to Rollo in 911/12 (Le Maho 1995; 1998; Bauduin 1999; Deniaux et al 2002, 392–394), so there is no reason to believe that the authorities should not have been able to maintain coin striking. Most likely, the issue in the name of Charles continued as a type immobilisé, or at least it was resumed under Charles the Simple. The best way to establish whether a coin type was immobilised is to look at its occurence in hoards deposited a long time after its official period of striking. Specimens with few signs of wear would then have a fair chance of being late immobilised issues. Indeed, hoard evidence does support the suggestion that coin production continued after 877 in the future Normandy. If we examine the situation in Lower Normandy first, the hoards demonstrate that coins struck in Bayeux and Curtisasonien still circulated massively during the early tenth century, and a few coins occur in later hoards (Table 6.1). Some of these coins were probably old Charles the Bald coins still in circulation, but the bulk should be regarded as later issues, at least in hoards from the last decade of the ninth century onwards. The wear can be used as a guide to distinguish old and recent coins in these hoards, but unfortunately this feature was not always dealt with in older find reports. As for the mint of Lisieux, Lafaurie thinks that it closed at the end of the ninth century as a result of Viking raids (Lafaurie 1965, 279). There are, however, Lisieux coins known from two early tenth-century hoards: Montrieux (dép Loir-et-Cher) (D.229) and Luzy (dép Nièvre) (D.202). Unfortunately, the hoard reports are not precise enough to record how many, and the coins could be old specimens still in circulation. If the Issy-l’Evêque (dép Saône-et-Loire) hoard (D.172), with five Lisieux coins, really dates from the reign of Charles the Simple (898–922), it would be a clue that this mint was indeed in operation during the first decade(s) of the tenth century. But the date of deposit of this hoard is most insecure, and it may well date from some decades earlier (see Table 6.1, note 3). Thus, we do not know, at present, when the mint of Lisieux closed. We have no evidence that the scarce GDR-issue was continued after 877 at the mint of Coutances. Turning to Upper Normandy, the evidence is less clear cut. Coins of Rouen are plentiful in hoards until about 900 (Table 6.1). In the tenth-century hoards, they are completely absent, except for an odd specimen in the Evreux (dép Eure) hoard (D.136; Longpérier 1870). Their absence from the Coudres (dép Eure) hoard, deposited about 920 (D.111 and 135; Moesgaard 1995, 99), on the border between the two currency zones (as defined above), is particularly significant and may hint that their production had ceased by then. Unfortunately, no early tenth-century hoard has yet been recorded north of the



Seine, Rouen’s home circulation zone, so that we cannot be certain as to whether or not their circulation really ceased then. On the present evidence, however, this is the obvious conclusion. It is difficult to distinguish specimens minted during the lifetime of Charles the Bald from the types immobilisés and the issues of Charles the Simple. In his fundamental paper on the Rennes hoard, J Lafaurie established a distinction on the basis of weight. The weight of the denier was fairly constant on a standard of 1.60–1.80 g until the early tenth century. At some stage in the reign of Charles the Simple (898–922), it declined to c 1.50 g, and then declined even further during the following decades (Lafaurie 1965, 268–270; see also Grierson and Blackburn 1986, 247; Moesgaard 1997). As for the mints of the future Normandy, the weight evolution can be following to a certain degree by examining the published weights of specimens from well-dated hoards. The results of a quick survey of the material are presented in Tables 6.2–6.4. Unfortunately, this material is rather sparse because weights were not always recorded in the published hoard reports so that the results cannot be relied on statistically. Nevertheless, broadly speaking, they fit the weight evolution described by Lafaurie. The only intriguing feature is the weight (1.60 g), observed by Charvet, of the coins of the Rouen mint in the Glisy hoard, from the 890s. This is very low compared to that of other mints (1.70–1.75 g). So, is the quoted weight an average of the15 specimens of Rouen mentioned by Charvet, or just the weight of a single specimen? If it is an average, then it points to an earlier decline in weight in Rouen than in Lower Normandy. At present, it is impossible to tell, but the publications of the Ablaincourt (dép Somme) and Féchain (dép Nord) hoards will hopefully provide solid new information. The weight variation within an issue was, at any given moment, so large that the weight of an individual coin cannot be used as the sole guide for its dating. To give an example, a whole series of coins of Rouen display the same faults: DI instead of D-I; retrograde S in the monogram; sometimes X instead of an initial cross on the obverse and P instead of R (PA, plate III,1–2, 4–5, 8, 10; Prou 382, 388; Dumas 1978, nos 17–18; MDA inv 94.5.1 (Figure 6.1a) and 96.4 (Figure 6.1b); Vinchon sale, 30 Nov–1 Dec 1993, no. 160). These shared features most likely point to a series struck within a short period. Nevertheless, the weights range from the official full-weight of 1.81 g (PA III,1) to a mere 1.43 g (MDA 94.5.1). The best method of determining the place of an individual specimen in the chronological sequence of the issue would be a full-scale die study of all the coins of a particular mint, because then the occurrence of particular specimens in well-dated hoards would give fixed points for an absolute chronology. However, no die study has yet been undertaken for any of the Norman mints. So, for the time being, the only way

Fig. 6.1. Charles the Bald, deniers, GDR-type, mint of Rouen. (a) MDA, Rouen, inv. 94.5.1 Photo: Yohann Deslandes; (b) MDA, Rouen, inv. 96.4. Photo: Yohann Deslandes.



to follow the chronological development of the style of the various mints is by examining specimens from successive hoards. Many hoards are, however, only known from the most summary descriptions, without illustrations, and others are not fully published as yet. I am, therefore, only able to present here a summary sketch of the development of the three major mints, based on the published material (Tables 6.2–6.4; Figures 6.2–6.4). In this context, one needs to note that Gariel (1883–84) mainly illustrated hitherto unpublished varieties, and thus his plates overstress odd specimens compared to normal ones. Table 6.2 Mint of Rouen, c 890–c 900: specimens in dated hoards Date



887–98 (F-Glisy) 887–98 (F-Féchain) c 890–98 (F-Monchy) c 895 (GB-Ashdon)

Gariel 1883, pl XIV,123–126 Dhénin 1985, 418 Gariel 1883, pl XVII,31 Blackburn 1990, no. 70

1.60 (Charvet 1870, 434) – – [1.16 fragment]

Table 6.3 Mint of Bayeux, c 890–c 980: specimens in dated hoards. Date



887–98 (F-Glisy) c 920 (F-Juaye-Mondaye) c 950/80 (F-Fécamp)*

Gariel 1883, pl X, 36–37 Doucet 1882, figs 2–6 Dumas 1971, nos 6047–6049

1.65–1.70 (Charvet 1870, 425) 1.50, 1.60, 1.65, 1.70(2) 1.28, 1.10, [0.98 clipped]

* hoard buried c 980/85, but coins presumably older.

Table 6.4 Mint of Curtisasonien, c 820–c 920: specimens in dated hoards. Date



887–98 (F-Glisy) c 890–98 (F-Monchy) 898–922 (F-Montrieux)

Gariel 1883, pl XI, 53 Gariel 1883, pl XVI, 16 –

c 895 (GB-Ashdon)

Blackburn 1990, nos 68–69

1.70–1.75 (Charvet 1870, 423) – 1.53–1.69 (RN 1838, 349, ‘29–32 grains’) [1.28, 0.99 fragments]

c 905 (GB-Cuerdale)

Dolley & Morrison 1966, no. 125

c 905/20 (F-Fécamp)* c 920 (F-Juaye-Mondaye)

Dumas 1971, no. 6050 Doucet 1882, figs 10–16

c 920/23 (F-Rennes)

Lafaurie 1965, no. 28

* hoard buried c 980/85, but coins presumably older.

1.73 (Dolley & Morrison 125), 1.62–1.68 (Hawkins 1843, ‘25–26 grains’) [1.275 clipped] 1.40(2), 1.45, 1.50, 1.60(2), 0.70(ob) [0.40 very corroded forgery]



Fig. 6.2. Evolution at the mint of Rouen. (a-d) hoard of Glisy, 887-98; (e) hoard of Monchy, c 89098. After Gariel.

Fig. 6.3. Evolution at the mint of Bayeux. (a-b) hoard of Glisy, 887-98; (c-g) hoard of JuayeMondaye, c. 920. After Gariel and Doucet.



Fig. 6.4. Evolution at the mint of Curtisasonien. (a) hoard of Glisy, 887-98; (b) hoard of Monchy, c. 890-98; (c-i) hoard of Juaye-Mondaye, c. 920. After Gariel and Doucet.

Specimens without hoard provenance are much more difficult to date. Nevertheless, specimens of low weight and odd style, with mis-spellings, are often likely to be late. This kind of argument is naturally less secure than one based on their occurrence in well-dated hoards, as there could be considerable weight (see above) and style variations within a single issue. The following specimens may be late: Degenerate GDR-denier, mint of Bayeux (Figure 6.5). Obv: ⳭCRATIA D-I PIX (starts at 9h) – monogram of CAROLVS. Rev: ⳭBAI◊CAS CIVITS – cross. 1.54 g; Peter Woodhead coll. Fig. 6.5. Degenerated GDR denier, mint of Bayeux, coll. Peter Woodhead. Photo: PW.



Degenerate GDR-denier, mint of Rouen (Figure 6.6). Obv: ⳭGRATIA DI REX (starts at 1h30) – monogram of CAROLVS. Rev: retrograde ⳭROTVMACVS CI – cross. 1.25 g; Peter Woodhead coll. Fig. 6.6. Degenerated GDR denier, mint of Rouen, coll. Peter Woodhead. Photo: PW.

Degenerate GDR-denier, mint of Rouen (Figure 6.7). Obv: X CRATIA RI REⳭ (legend starts at 3h) – retrograde CAROLVS monogram. Rev: Ⳮ ROTVMACVZ CIVII (legend starts at 11h30) – cross. 1.23 g; die axis: 0h30; peck: 0/1; MDA, Rouen, inv 94.5.2. Fig. 6.7. Degenerated GDR denier, mint of Rouen, MDA, Rouen, inv. 94.5.2. Photo: Yohann Deslandes.

Poey d’Avant used style to single out GDR-specimens from the mint of Rouen as types immobilisés, but unfortunately not of any other Norman mints (PA 93–109, plate III, 1–16). However, many of these coins, which are admittedly poor stylistically, are too heavy to be regarded as tenth-century issues. They may be late ninth-century issues (post877), but one should bear in mind that the official pre-877 issue was so large that misspellings and other faults are bound to have occurred. For example, the Chalo-Saint-Mars (dép Essonne) hoard contained a Rouen specimen with several mis-spellings GRAITA D REX (Gariel 1883, 83). This hoard was buried in about 880, so that the coin must be quite early. On these criteria, only PA 103–105, plate III,11–13, can be retained as late issues with a certain degree of confidence, although PA 99–102, plate III, 7–10, may also be considered to be late. As for style and mis-spellings, the specimen PA 103, plate III,11 (Figure 6.8), weighing 1.40 g, is the most degenerate of those illustrated by Poey d’Avant, and one would instinctively suppose it to be tenth century. However, another specimen with a very similar obverse, and almost the same mis-spellings, exists in the Merson collection: Fig. 6.8. Degenerated GDR denier, mint of Rouen. After Poey d’Avant.



Degenerate GDR-denier, mint of Rouen (Figure 6.9). Obv: xCRATIA CIRV PEX (starts at 10h) – monogram of CAROLVS with retrograde S. Rev: ⳭROTVIIACVS CIVII – cross. 1.63 g; die axis: 8h45; pecks: 0/3; R A Merson coll, possibly from the Laxfield (Suffolk) hoard. Fig. 6.9. Degenerated GDR denier, mint of Rouen, coll. R.A. Merson, possibly from the Laxfield hoard. Photo: PW.

This specimen is almost full-weight, which points to rather an early date; indeed, if it is from the Laxfield hoard, as circumstantial evidence seems to indicate (R A Merson, pers comm), it will have to date from before the deposition of that hoard in the 880s (Figure 6.1, note 5). This example shows once again how difficult it is to judge date from a single specimen. An unpublished obol, unfortunately without find-provenance, in the collection of the MDA, Rouen, may be related to this period, but its attribution is most uncertain:5 Degenerate Carolingian obol, possibly Clermont or Rouen (Figure 6.10). Obv: ⳭXCARLV(-I?) (S retrograde) – cross. Rev: retrograde ⳭCIAROIVM (starts at 6h) – monogram of CAROLVS. 0.64 g; die axis: 1 h; MDA, Rouen, inv 94.11.1. Fig. 6.10. Degenerated Carolingian obol, possibly Clermont or Rouen, MDA, Rouen, inv. 94.11.1. Photo: Yohann Deslandes.

The type and the legends clearly show that this coin is a blundered version of the CRF (CAROLVS REX FR) type with a cross on the obverse and the monogram and mint name on the reverse. This type was struck by Charlemagne from 793/94 to 812, revived by Charles the Bald from 840 to 864 and maintained in the Aquitaine mints even after 864. The most likely interpretation is that the reverse legend is a deformation of the name of Clermont, which normally appears as ⳭCLAROMIINT, albeit the style of the products of this mint is usually quite different.6 However, another possible reading is ⳭCIVROTVM for CIVITAS ROTVMAGVS, the ‘cité’ of Rouen, even though this order of the elements of the legend is very unusual. Only one specimen of the CRF-type, attributed to Charlemagne, is known for Rouen (MG 141; Völckers 1965, 142; found at Dorestad). On the other hand, obols are rare at this time, and the style is not right. These features would fit better at the end of the ninth and the beginning of the tenth century. Indeed, it would be tempting to see this as an issue of Charles the Simple (898–922) at Rouen, or as an irregular issue by the Viking invaders. The first of these possibilities is to be preferred, because the name on the obverse very clearly reads Charles. Of course, all these proposals are only hypotheses. We need a specimen from a well-dated hoard to get any further with the attribution of this coin.



For all the mints discussed above, it is impossible to say – on the basis of our present knowledge – whether the issue was continuous or occasional. Neither can we determine who was in charge of this coinage, but the firm grasp of this region by the Carolingian kings (see above) suggests that the king himself maintained control of the coinage.

UNSIGNED ISSUES 911/24–C 945 The lower Seine valley with its hinterland (approximately present-day Upper Normandy) was given to the Viking chief Rollo, c 911/12. Central Normandy was conceded in 924 and the Cotentin and Avranchin in 933 (Renaud 1989). The first signed issues by the new rulers were struck in the name of Rollo’s son and successor, William Longsword (c 927–42). Did the Vikings strike coins before these signed issues? Hoard evidence perhaps allows us to think so. The three GDR-specimens of Bayeux in the Fécamp (dép Seine-Maritime) hoard, deposited about 980/85, must be of a late issue, since they are not worn and of a low weight standard, of about 1.10–1.30 g. Were they struck during the brief rule of Louis IV in 944–45, as Dumas tentatively suggests, or under Hugues the Great, who was in charge of Bayeux 942–44/45, or under Norman rule before 942 or after 945 (Dumas 1971, 104–105 and plate VI, 6047–6049)? It is impossible to tell. The issue was probably rather limited; indeed, two of the coins were struck with the same obverse die. All three coins display some mis-spellings, but otherwise they are of good workmanship. On the other hand, the single GDR-specimen of Curtisasonien in the Fécamp hoard is worn and of the same style as the specimen in the Cuerdale (Lancs) hoard, buried c 905 (Table 6.4). Dumas thinks therefore that it was struck well before the burial of the hoard about 980/85, and that the mint was probably closed before the Normans took over in 924 (Dumas 1971, 106–107, plate VI, 6050). As with the period before 911, we have no firm evidence to establish whether or not the mint of Rouen was active after 911. The low weight of the specimen from the Woodhead collection quoted above would point to an early Norman date, but one cannot draw conclusions with any degree of confidence from the weight of a single specimen. Yet another coin type is a good candidate for being a Norman issue. This is the lowweight imitation of the Louis the Pious (814–40) temple type recorded in the hoards from Coudres (dép Eure), tentatively dated to 920/23 (D.111 and 135; Moesgaard 1995, 99), Evreux/Saint-Taurin (dép Eure), buried about 943/45 (D.136; Longpérier 1870; Moesgaard 2003) and Haute-Isle (dép Val-d’Oise), traditionally dated to 898/923 (D.161). The prototype was struck c 822–40 and has the inscription ⳭHLVDOVVICVS IMP on the obverse and ChiRhoISTIANA RELIGIO on the reverse; the imitations have more or less blundered versions of these legends. No other find provenances are known for these issues, and the concentration of finds in Eastern Normandy and its immediate border areas indicates a Norman origin (cf. Longpérier 1870). The dates would fit an issue struck soon after the arrival of the Vikings. Unless it should be regarded as a specific Upper Norman issue, or a very short-lived or sporadic one (which is unlikely, see below), its absence from the Juaye-Mondaye (dép Calvados) hoard (D.177; Doucet 1882), buried about 920, would be an argument for a date after 920/25 (thus making it a Norman issue). Unfortunately, the coins of Haute-Isle are known only from a very brief description, and their weights were not recorded (Gaudichard 1931). The coins of Coudres and of Evreux are



Fig. 6.11. Temple-type imitations from the Coudres hoard. After Gariel.

relatively well recorded, and they clearly belong to two different phases of this coinage: the former (Figure 6.11) weigh 1.01–1.41g,7 and the latter (Figure 6.12) 0.62–0.76 g,8 with the deformations of the legends being different. The many variations in legends show that several dies were used, and the issue must have been on a substantial scale. These features tell us that this coinage must have lasted for some while, from the 920s (or before?) to the 940s. However, we do not know if the striking was continuous. The Evreux/Saint-Taurin temple coins have been considered as obols, because of their low weight. Their diameter makes this suggestion unlikely, and they should rather be regarded as light deniers.

Fig. 6.12. Temple-type imitations from the Evreux hoard. After Longpérier.

It is impossible to tell who took the initiative with this coinage in the first instance. It has been suggested that the Evreux/Saint-Taurin temple coins were a royal issue in the name of Louis IV, struck during his brief control of Rouen, 942–45, but the obverse legends contain deformations of IMP ( ⳱emperor) rather than REX, as seen on the genuine Louis IV coins, so it is probably an independent Norman issue continuing the issue from the 920s and imitating the original Louis the Pious legend without any involvement from Louis IV. These late coins have a W on the obverse (as the real Louis IV coins, see below), which may be regarded as the signature of William Longsword, or – if they were struck after his death – as a way of making the coins acceptable to people used to William’s name (Longpérier 1870,



73–82; Dumas 1971, 73, note 1; 1979, 88–89, note 2). The only specimen with REX instead of IMP is from the Coudres board, and thus earlier than Louis IV (Prou 1896, no. 1055). The temple type had been very efficiently demonetised in West Francia in 864 (Coupland 1991a, 154), and no specimen would have been around in Normandy in the early tenth century. Why was this type chosen as prototype for this issue? We do not know, but one should note that this type is one of the best represented among the admittedly few Carolingian coins found in Scandinavia, and that it was still being used during the tenth century (Galster 1959; Skaare 1966; Hatz and Jonsson 1976, col 28–29; Skaare 1976, 43–47; Bendixen 1988; Coupland 1991b; Metcalf 1996 and later finds). It would go beyond the evidence to state that the idea was brought to Normandy from Scandinavia, but the coincidence is indeed puzzling. The temple type became very popular in Normandy and was used for the signed issues of Richard I (942/45–96), and as the starting point for the typological evolution embracing most Norman coins for the following two centuries (Dumas 1971; 1979).

SIGNED ISSUES C 930/40–45 The signed issue at Rouen in the name of William Longsword (c 927–42) was among the very first Frankish issues where the local issuing power put its own name so clearly on the coins. The regionalisation of coinage had already been going on for half a century, but at other mints the royal name was either kept as a type immobilisé or simply omitted, and any allusion to the local magnate behind the issue was limited to a single letter or so (Dumas 1973, 70). Longpérier suggested that the Rouen coin of William was an imitation of the coins of the pre-Viking East Anglian kings (RN 1849, 46; Longpérier 1870, 83), which carry a cross on both sides, sometimes with crescents in the angles: Æthelstan I, c 825–45 (North 1994, nos 445–448). Æthelweard, c 845–55 (North 1994, nos 452–453, especially no. 452 with crescents). Eadmund, 855–69 (North 1994, nos 457–458, especially no. 458 with crescents). This suggestion has generally been accepted by later numismatists (Dieudonné 1936, 303; Dumas 1971, 72, note 1; 1979, 87, note 5; Caron 1882, 15–16; Dhénin 1990, 18). Dumas adds another possible prototype: Wessex, Æthelwulf, 839–58 (North 1994, nos 612–613). One could add others: Wessex, Æthelwulf, 839–58 (North 1994, nos 603–607). Kent, Baldred, c 823–25 (North 1994, nos 213–215). These coins were, though, far more than half a century old when William started his coinage, and they were no longer in circulation in England (Archibald 1988, 273–274). No finds are recorded from France. They would thus probably not have been known either in France or in England, around 930. Only the odd specimen could have been available for copying, but this is not very likely, and I would prefer to see the Rouen type as an original, local type. Only very few specimens of this coin are known. I have made a rapid survey, which may not be complete:9



(1) Paris, BNF féod 940 (Figure 6.13a) (Duby 8; Longpérier 1843, RN, 53–54, 56, plate V,1 (not broken); PA 115, plate IV,1; Dumas 1971, plate I). Obv: ⳭVVILEL[M]VS; rev: [Ⳮ]ROTOMAGS; 0.99 g (broken). Unknown provenance. (2) Present location unknown, coll Rousseau in 1860 (Figure 6.13b) (Fillon 1860, 2, no. 7, plate I, 6; Le Roy and Moesgaard 1999, 143–144, no. 1). Same obv; rev: ⳭROTOMA CIV; 0.96 g (broken). Provenance: Soleure hoard, Switzerland (Dumas 1979, find no. 12). (3) Copenhagen, Royal Collection of Coins and Medals (Figure 6.13c). Same obv; rev: ⳭROTOMA CIVITAS; 1.25 g. Provenance: Terslev hoard, Denmark (Dumas 1979, find no. 14). (4) PLR coll, Le Havre (Figure 6.13d) (Le Roy and Moesgaard 1999). Same obv; rev: ⳭROTOMA CIVI; 1.03 g. Provenance: possibly the Rouen area. (5) Rouen, MDA, Rouen, inv 83.2 (Figure 6.13e) (Dhénin 1992). Obv: ⳭVVILELMV(S retrograde); rev: ⳭROTOMA CIVI; 1.23 g. Provenance: probably the Fécamp hoard. (6) Present location unknown (Dumas 1984, 503, after ms 720 of the Society of Antiquaries of London). Described by Dumas from a poor photocopy of a drawing: Same obv; rev: ROTOMA(G?)IS. The type seems to be slightly different from the other five specimens. Provenance: Berkhamstead, Herts, England, mid-eighteenth century.

Fig. 6.13. William Longsword, deniers, Rouen. (a) BNF, Paris, féod 940. After Poey d’Avant; (b) Soleure hoard. After Fillon; (c) Copenhagen, Royal Collection of Coins and Medals. Photo: The National Museum; (d) PLR coll., Le Havre. Photo: Yohann Deslandes; (e) Rouen, MDA. Photo: F. Dugue.

Ducarel published three varieties in 1757, all with various forms of the legend ROTOMAGIS/ROTOMA CIV on the reverse. One is of the usual type (Ducarel, plate 1.2, after the drawings of M de Boze) (Figure 6.14a); one has a trefoil instead of the crescents in the fourth quarter of the reverse cross (Ducarel, plate 1.1, drawing de Boze) (Figure 6.14b). The third is rather stylised without any symbols in the quarters of the cross (Ducarel, plate 8.99, after a drawing of Venuti) (Figure 6.14c). The first is straightforward and the third is probably just a simplified version of a real coin. The second is, however, problematic. No genuine specimens have been seen since the eighteenth century. On the other hand, modern forgeries are known (eg the specimen in the collection of Peter



Fig. 6.14. William Longsword deniers, Rouen. (a-c) after Ducarel.

Woodhead, weighing 1.34 g (Figure 6.15), from Glendining sale, 22 Oct 1970, no. 591. The forger John White was active in the mid-eighteenth century and, until a genuine specimen appears, the most likely interpretation of the Ducarel 1.1 is that it is a fake.10 Other forgeries of these types have since appeared, particularly those made by Benassis in the 1830s, allegedly from a hoard found at Pacy-sur-Eure (dép Eure) (D.249).11 The coins 1–5 are from different dies (allowing for imprecise drawing, nos 2–3 could be of the same obverse die). This indicates an issue of some importance. On the other hand, one should note that this coin type disappeared extremely quickly from circulation (see below), which suggests that the issue was rather limited (unless one supposes a renovatio monetae operated by Louis IV). Fig. 6.15. Modern forgery of William Longsword denier. Coll. Peter Woodhead. Photo: PW.

The second type in the name of William is known only from a single specimen, found at the Norman Abbey of Mont-Saint-Michel (dép Manche). It carries the title of Duke of Brittany, which probably reflects the acquisition of the Cotentin and Avranchin, formerly under Breton rule, by William in 933, or on a more general level the expansionist policy carried out by William towards Brittany (Dolley and Yvon 1971, plate XL, 5; Renaud 1989, 87). This coin is definitely of English type (circumscription cross) and style, which is not surprising given that the Normans kept contact with their Viking cousins in England until the beginning of the eleventh century. Sometime by the mid-tenth century, an issue in the name of Saint-Ouen was struck in Rouen (Figure 6.16), of which several varieties are known.12 It was probably struck at the Monastery of Saint-Ouen, which was the first religious institution to be re-established in Viking-controlled territory, as early as 918 (Le Maho 1996, 46).


JENS CHRISTIAN MOESGAARD Fig. 6.16. Rouen, Saint-Ouen monastery, deniers. (a) MDA, Rouen, inv. 93.4.1. Photo: Yohann Deslandes; (b) Copenhagen, Royal Collection of Coins and Medals. Photo: The National Museum.

When King Louis IV (936–54) took over Rouen after the assassination of William Longsword, in December 942, he initiated an apparently plentiful issue in his own name. This must have been interrupted at the latest in July 945, when the king was captured by Norman barons wanting to place William’s son on the throne. The issue thus lasted at most two and a half years (Longpérier 1870; Dumas 1979, plate XV, 2; Albuquerque, VSO 16; Salaün 2000). As the number of variants of legend and type shows, many dies were used to strike this issue, which quickly dominated the currency (see next section). Its purpose was clearly to underline the claim of the king to Normandy by spreading his name. But the coins carried a V before the name of the king, apparently to make them look like the legend of William’s coin in order to make them more acceptable. These two purposes were thus partly contradictory. After the return to power of Richard I in 945, a plentiful coinage commenced (Dumas 1971; 1979), but such is beyond the scope of this chapter. Except for a single mention of Bayeux in an eleventh-century document, we do not hear of any Norman mints other than Rouen for the whole of the ducal period.

CURRENCY 864–C 945 What role did the coins struck in Normandy, 864–c 945, play in the local currency? Were they plentiful enough to fullfill the need for coins, or did Normandy require to be supplied also from other regions? Were they meant to be the sole means of payment in the region, or to be part of a larger pool of currency covering several regions or the whole empire? To answer these questions, one needs to study the individual coin finds. Unfortunately, the recorded stray finds are too few to allow for a thorough analysis of this currency13 (see Moesgaard 1995). We have to rely on the few properly recorded hoards as the basis for our description, even though hoard compositions can easily be biased as a result of special circumstances surrounding their burial, and more well-recorded finds are needed to arrive at secure conclusions. Hoard evidence, as analysed by Metcalf, shows that the future Normandy did not form one single currency zone during the last third of the ninth century. The east belonged to a currency zone comprising present-day Upper Normandy, Picardy, Champagne and Ile-deFrance (‘Francia’). The centre and the west were parts of the zone between Loire and Seine



(‘Neustria’). Coins mingled in circulation within each zone, but rarely penetrated into the other zone, with ‘Francia’ being the more active and important of the two (Metcalf 1990, 90). Three hoards from the reign of Charles the Bald, or shortly after, are recorded from Norman territory: Imbleville (dép Seine-Maritime) (D.168); Pont-Saint-Pierre (dép Eure) (CH 1997, no. 65) and Vire (dép Calvados) (D.376). The first is fully recorded, whereas the latter two are only partly recorded. These finds were not used by Metcalf, but they appear to confirm his conclusions. The composition of the first two hoards, which are both from Upper Normandy, is typical for the ‘Francian’ currency zone, whereas the latter hoard, from Lower Normandy, clearly belongs to the ‘Neustrian’ zone. Local coins make up a large part of the currency, with other intra-regional coins forming the bulk, together with a few intruders (Table 6.5). There are no well-recorded Norman hoards known from the next 35–40 years; however, two hoards are recorded from about 920, which have different compositions. The Juaye-Mondaye hoard (dép Calvados) (D.177; Doucet 1882) is totally dominated by the local coins of Curtisasonien (c 700 from a total of c 900). There are 21 deniers of Bayeux, most of the rest being from ‘Neustrian’ mints (Angers, Blois, Chartres, Le Mans, Orléans, Châteaudun), with only a few from other regions (Paris, Saint-Denis, Rennes). Le Mans is the best represented mint after Curtisasonien, which is logical given that this mint is quite close to the find-spot. This fact is nevertheless noteworthy because Le Mans coins are also numerous in the Fécamp (dép Seine-Maritime) hoard, buried about 980/85 (Dumas 1971). It seems that the immobilised GDR-type was struck on a large scale at this mint for almost a century after the death of Charles the Bald, when production had already stopped at other mints. About 920, the clergy of Bayeux were fleeing, along with Breton clerics, presumably from Viking attacks (Chédeville and Guillotel 1984, 384), and Juaye-Mondaye is not far from Bayeux. J Le Maho suggests that there may well be a connection between these events and the burial of the hoard (pers comm), just as the deposition of the Rennes (dép Ille-et-Vilaine) hoard (D.274; Lafaurie 1965) has been related to Viking raiding (Chédeville and Guillotel 1984, 384). It has been suggested that the hoards from ‘near Autun’ and Issy-l’Evêque (dép Saône-et-Loire) (D.23 and 172), which may be parcels from the same hoard, represent the savings of a refugee from a Viking invasion who was fleeing insecure Neustria for inland

Table 6.5 Composition of Norman hoards, 864–77 Local




18 Rouen

1 Tours, 1 Gembloux, 1 Curange/Couvin


1 Rouen

26 Quentovic, 7 St Denis, 1 Soissons, 2 Attigny, 2 Reims, 4 Palais, 1 Paris, 2 Amiens 2 Palace, 2 St Denis


Bayeux, Lisieux, Curtisasonien

Angers, Orléans, Blois, Chartres, Le Mans

1 Nevers, 1 Mouzon, 1 Valenciennes Noyons, St Denis, Valenciennes, Amiens?



Burgundy, because the coins of Curtisasonien, Lisieux and Le Mans form the largest part of the hoard (Grierson and Blackburn 1986, 636; see also Metcalf 1990, 80). The dating of this hoard (these hoards?) is much debated (see Table 6.1, note 3), but it could be as late as c 920. If this is correct, it is puzzling how close its composition is to that of the Juaye-Mondaye hoard because such would indicate that the two sums derived from the same pool of currency. The Coudres (dép Eure) hoard (D.111 and 135, with bibliography, cf. Moesgaard 1995, 99) is quite different. The GDR-coins of the ‘Neustrian’ mints (Angers, Orléans, Blois, Chartres, Châteaudun, Le Mans, Tours, Vendôme, Chinon-Tours) still make up the bulk of the hoard (119 of the 144 coins described), but there are not any coins from either Curtisasonien or Bayeux, and only one from Le Mans. The best represented mints are Tours-Chinon, Blois, Chartres, Orléans and Angers, which cover the eastern and southern part of ‘Neustria’. Moreover, there are nine ‘Francian’ coins (Paris, Saint-Denis, Reims), as well as one from Bourges and one of Troyes, and 14 of the (presumably) local Upper Normandy imitation of Louis the pious’ Temple-coins (see p 109–111). Many of the coins are quite new, such as the two-line type from Paris and St-Denis, and the coins of Angers, Blois and Reims, with symbols in the quarters of the reverse cross, as well as the very numerous Tours-Chinon coins. This demonstrates a lively circulation with recent flows of coins mainly from the Loire Valley, but also from the Paris region. The interpretation of the Coudres hoard is nevertheless difficult because it is from the Evreux area, which was not under the firm control of the counts of Rouen (the official title of the first Norman rulers) until the middle of the reign of Richard I, 942/45–96 (Bauduin 1999). This hoard may therefore be representative of a Loire Valley/Paris region currency rather than of a Norman one. A hoard from the Lower Seine Valley or the Pays de Caux is needed before one is going to be able to judge the currency of early Normandy. Thus, at the beginning of the Norman period, the currency of the various parts of Normandy was very different. Central Normandy still had active mints, both locally (Curtisasonien and Bayeux) and in neighbouring areas (Le Mans), providing the bulk of the currency, whereas eastern Normandy (if the Coudres hoard can be taken as representative of all Upper Normandy) only had a limited local coinage. Coins were flying in from the south and east, but not from the west. The last recorded hoard is from the very end of the period studied here. It is from Evreux (dép Eure) (D.136; Longpérier 1870; Moesgaard 2003), not far from Coudres, and was buried c 943/45. It shows survivors from the currency at the time of the Coudres hoard, some 20 years before: at least 10 coins (out of the 58 recorded) are from ‘Neustrian’ mints (Beaugency, Blois, Châteaudun, Le Mans, Orléans, Vendôme), with at least one GDR-coin from Rouen and one from Bourges. The bulk of the hoard is nevertheless made up of the recent issue at Rouen in the name of Louis IV (38 specimens), and the low-weight, presumably Norman, imitations of Louis the Pious’ Temple coins (5 specimens). It is remarkable that the Louis IV issue had managed to become dominant so quickly; the issue must have been quite massive to achieve this goal. It is even more remarkable that the hoard does not contain even a single specimen of the recently struck coins in the name of William Longsword. Had Louis IV operated a successful renovatio monetae, so as both to obtain metal for his new issue and to get rid of the coins that symbolised the independence of Normandy? This is not very likely, as the older coins of other mints still circulated, unless such a renovatio was aimed solely at the William coins. Otherwise, one has to suppose that William’s coins only constituted a somewhat limited



part of the currency before 942, which would in turn suggest a rather limited output. Another possibility is that the coins of William never reached the Evreux region, which was only under loose Norman control (Bauduin 1999). One needs a hoard from William’s reign to obtain answers to such questions about the role of the first signed Norman issue. The locally dominated currency of the Evreux/Saint-Taurin hoard reminds one of the well-managed currency in the Fécamp (dép Seine-Maritime) hoard (D.137; Dumas 1971) and in later Norman hoards, where the most recent issue counts for at least two thirds of all the coins (Moesgaard 1998). Pecking is a typical Viking method of checking the quality of metal (cf. Archibald, Chapter 4, in this volume). To my knowledge, there is only one coin from France with this feature – and, unfortunately, it has no secure find-spot. It is the degenerate GDR-type with a KAROLVS-monogram, struck at Rouen, as described above; this appears to have a single peck on its reverse (6.7). It was acquired by the Musée départemental des Antiquités de la Seine-Maritime, in 1994, from the heirs of a collector from Rouen; however, this collector had not only bought coins locally, but also from dealers. The coin is not a particularly nice specimen, but even though the collector was interested in all kind of coins and not just coins from the Rouen mint, it would seem to be an unlikely one to have been ordered from afar, and thus a local Norman provenance is likely, even if it cannot be proved. If, indeed, this coin circulated and was pecked in Normandy, we have the first example of a coin that had demonstratively been used by a Viking in Normandy. Pecking has, to the best of my knowledge, not been noted on any other coin found in Normandy. One odd specimen might not have been noticed by earlier scholars, but if the phenomenon had been widespread, it would certainly by now have been noted by numismatists and archaeologists.

COINS, CURRENCY AND STATE FORMATION What role did coinage play in the formation of an independent Normandy under the Viking counts of Rouen/dukes of Normandy? It is clear that the coins and currency of the newly founded Normandy belong to a Frankish tradition. The Vikings did not introduce their own ideas about coin striking; indeed, at this period, it seems that the only coin production was that of the so-called Hedeby-issues, which were southern Jutish in their scope (cf. Malmer 1966, and Chapter 2, in this volume; Wiechmann 1996, and Chapter 3, in this volume). Nor did the Vikings introduce their bullion economy to their new territories (the use of whole and fragmented coins by weight rather than by tale, along with jewellery, ingots and hack-silver for its bullion value). At first, they appear to have been content to continue with the then current issues, which were types immobilisés of the old GDR-type, with KAROLVS monogram. On present evidence, we can only be certain that this took place at the Bayeux mint, but it is likely that other mints were also involved. Imitations of Louis the Pious’ temple type were probably struck in Upper Normandy, but still without any reference to the ruling power. This fitted well into the general pattern of the on-going regionalisation of coinage in the Frankish realm, which consisted of immobilisations and the copying of well-established royal coin types. Eventually, coins were used deliberately to affirm the new power, when William Longsword became one of the first local magnates in West Francia to put his own name, in



full, on a coin. When the king endeavoured to recover Normandy after the assassination of the duke, the first thing he did was to initiate an issue in his own name, and when Richard I took over power, he struck coins in his own name as well. The role of coins as a symbol of power and as a means of propaganda was thus clearly expressed. It should be stressed that it was also a sign of the integration of the incomers into their surrounding society. Indeed, recent research has shown that the Vikings had strong contacts with the local Franks even before their definitive installation in Normandy (Bauduin 1999; Deniaux et al 2002, 392–412), which would therefore have facilitated their speedy integration. What was to become Normandy belonged to different currency zones. Their unification under the new ruler does not seem to have brought about monetary unification immediately. This conclusion is, however, based on slight evidence, and new hoards and more stray finds are necessary to confirm it.

ACKNOWLEDGEMENTS I would like to thank Jacques Le Maho, Jacqueline Pilet-Lemière, Peter Woodhead, R A Merson, Jean-Pierre Garnier and Michel Dhénin for useful information and/or discussions.

Notes 1. FM (Grierson and Blackburn 1986); Charleville-Mézière, Musée de l’Ardenne (Doyen 1991); BM (Dolley and Morrison 1966); BnF (Prou 1896); Paris, Monnaie de Paris (Belaubre 1987); MDA (Dumas 1978). See also Vinchon sale, 30 Nov–1 Dec 1993; Crédit de la Bourse sale, 26–28 Apr 1993, no. 148 (now at the Musée d’Evreux); Marquigny 1983, 3 (Marquigny collection). 2. On the basis of material of other regions of the Carolingian Empire, some scholars have suggested the possibility of a central mint striking coins for a number of issuing places. The place names appearing on the coins would then be the issuing places rather than the mints. This seems unlikely, as it must have been too dangerous to transport so many coins over long distances. 3. The Lisieux coin Gariel 1884, L, 38, MG 1400 is probably pre-864. A Rouen coin in the name of Carloman (879–84) has been published (BSFN (2005), 78). 4. The Raoul coins of Curtisasonien (RBN 1870, 424; Prou 418; and Grierson and Blackburn 1986, no. 1508) are fakes (cf. Dumas 1980, 222). 5. I would like to thank J-P Garnier for valuable ideas in discussing the attribution of this coin. 6. At least the specimens from the Cuerdale hoard; see MG, plate XXXIV,1078, 1082–1083; Dolley and Morrison 197–98 (both from Cuerdale); Belaubre 178; Prou, plate XVII, 764, 768 (Cuerdale); Gariel 1884, plates XXII, 48-XIII, 51; Vinchon sale 30 Nov-1 Dec 1993, no. 134. 7. Longpérier 1847, 138–139, nos 315–321; Gariel 1884, 273–274, plates XLIII, 6, 8, 15–16, XLIV, 39; Prou 1896, nos 1036–1041, 1055. 8. Longpérier 1870, 75, nos 15–19; Dumas 1978, no. 34. 9. I have not had the occasion to check the Vergne, Jean dit Cazaux, Dubern sale in Bordeaux, 6–7 Dec 1985, nor have I made any attempt at checking sales catalogues systematically. While this book was in press, I became aware of NC 6–8 (1948), proceedings, 10. 10. Other forgeries of William coins are known: 2 specimens at Cambridge, FM, and one at the BM, London. The specimen sold by Frankfurter Münzhandlung E Button, sale 128, 3–4 Oct 1979, no. 1169 seems false, but the poor quality of the photo in the sale catalogue makes it



impossible to be sure. In particular, the very small dots of the inner rings on the obv and the rev are suspicious. 11. BnF, Paris, has four specimens of three different types from Benassis, and six specimens from other sources (kind information from M Dhénin). 12. Prou 1896, nos 394–395; Galster 1959, 74, no. 48; Dumas 1971, 100, note 2; Dumas 1979, plate XV,16; Albuquerque VSO 16; MDA, Rouen, inv 93.4.1 (⳱Crédit de la Bourse, 26–28 Apr 1993, no. 179); unpublished specimen from the excavations at Place Foch, Rouen; Dolley & Yvon 1971. 13. Metal-detecting is forbidden in France, so the detectorists’ finds are not recorded. Excavation finds are too few to permit statistical analysis.

References Archibald, M M (1988), ‘English medieval coins as dating evidence’, in J Casey and R Reece (eds), Coins and the Archaeologist, 2nd ed, 264–301, Seaby, London Archibald, M M (1992), ‘Dating Cuerdale: the Evidence of the Coins’, in J Graham-Campbell (ed), Viking Treasure from the North West: the Cuerdale Hoard in its Context, 15–20, Liverpool Museum Armstrong, S (1998), ‘Carolingian coin hoards and the impact of the Viking raids in the ninth century’, NC 158, 131–164 Bauduin, P (1999), ‘Aux origines du duché de Normandie’, Bulletin de l’O.U.E.N., no. 7 (Office universitaire d’études normandes, Université de Caen, Janvier 1999), 11–12 (cf. paper given at the conference, ‘Le Danemark et la France au Moyen Age à travers les images et les textes’, University of Copenhagen, November 1999). Belaubre, J (1987), Administration des Monnaies et Médailles, Les collections monétaires, II. Monnaies médiévales, 1. L’époque du denier, Monnaie de Paris Bendixen, K (1988), ‘Nyere danske fund af merovingiske, karolingiske og ældre danske mønter’, in P Berghaus, J Bracker, J S Jensen and L O Lagerkvist (eds), Commentationes Numismaticae 1988, Festgabe für Gert und Vera Hatz, 37–50, Tietjen, Hamburg Blackburn, M A S (1990), ‘The Ashdon (Essex) hoard and the currency of the Southern Danelaw in the late ninth century’, BNJ 59 (1989), 13–38 Boeles, P C J A (1915), ‘Les trouvailles de monnaies carolingiennes dans les Pays-Bas, spécialement celles des trois provinces septentrionales’, Jaarboek voor Munt- en Penningkunde, 1–98, Johannes Müller, Amsterdam Caron, E (1882), Monnaies féodales françaises, Rollin & Feuardent, Paris Charvet, J (1870), ‘Monnaies découvertes à Glisy’, 417–439 Chédeville, A and Guillotel, H (1984), La Bretagne des saints et des rois Ve-Xe siècle, éditions Ouest-France, Rennes Coupland, S (1991a), ‘The early coinage of Charles the Bald, 840–864’, NNÅ 151, 121–158 Coupland, S (1991b), ‘Carolingian coinage and Scandinavian silver’ (1985–86), 11–32 Delaporte, J and Lafaurie, J (1985), ‘Monnaies carolingiennes’, in P Périn and L-C Feffer (eds), La Neustrie, 326–329, Musée départemental des Antiquités de la Seine-Maritime, Rouen Deniaux, E, Lorren, C, Baudouin, P and Jarry, T (2002), La Normandie avant les Normands, éditions Ouest-France, Rennes Dhénin, M (1985), ‘[The hoard of Féchain]’, in P Périn and L-C Feffer (eds), La Neustrie, 416–419, Musée départemental des Antiquités de la Seine-Maritime, Rouen Dhénin, M (1990), ‘Les monnaies normandes’, Les conférences d’histoire locale du lycée de Domfront 9, 14–30, L’association pour la restauration du Château de Domfront Dhénin, M (1992), ‘Coins of William Longsword’, From Viking to Crusader, Exhibition catalogue (Paris/Berlin/Copenhagen 1992), no. 403, European Council



Dieudonné, A (1936), Manuel de numismatique française, IV. Monnaies féodales françaises, Picard, Paris Dolley, M and Morrison, K F (1963), ‘Finds of Carolingian coins from Great Britain and Ireland’, BNJ 32, 75–87 Dolley, M and Morrison, K F (1966), The Carolingian Coins in the British Museum, British Museum, London Dolley, M and Yvon, J (1971), ‘A group of tenth-century coins found at Mont-Saint-Michel’, BNJ 40, 1–16, BNS, London Doucet, L (1882), ‘Notice sur des monnaies Carlovingiennes trouvées à Juaye-Mondaye, arrondissement de Bayeux (Calvados), en 1870’, Mémoires de la Société d’agriculture, sciences et belles lettres de Bayeux 9 (1879/82) 215–258 Doyen, J-M (1991), Musées de Charleville-Mézières, catalogue des monnaies du Haut Moyen Age, Musées Charleville-Mézières Duby, P-A T (1790), Traité des monnoies des barons, Paris Ducarel, A C (1757), A Series of above two hundred Anglo-Gallic or Norman and Aquitain Coins of the antient kings of England, E. Withers & J. Scott, London Dumas, F (1971), Le trésor de Fécamp, Bibliothèque Nationale, Paris Dumas, F (1973), ‘Le début de l’époque féodale en France d’après les monnaies’, BCEN 65–77 Dumas, F (1978), ‘Monnaies carolingiennes’, Monnaies, médailles et jetons (MDA) 19–24, Musé départmental des Antiquités de la Seine-Maritime, Rouen Dumas, F (1979), ‘Les monnaies normandes’, RN, 84–140 Dumas, F (1980), ‘Les monnaies de Raoul, roi de France (923–936)’, Mélanges de numismatique, d’archéologie et d’histoire offert à Jean Lafaurie 215–222 Dumas, F (1984), ‘Nouvelles découvertes de monnaies normandes’, BSFN, 502–504 Dumas, F and Pilet-Lemière, J (1989), ‘La monnaie normande – Xe-XIIe siècle’, in H Galinié (ed), Les mondes Normands, 125–131, Société d’Archéologie Médiévale, Caen Duplessy, J (1985), Les trésors monétaires médiévaux et modernes découverts en France, Vol I (751–1223), Bibliothèque Nationale, Paris Fillon, B (1860), Collection Jean Rousseau. Monnaies féodales françaises, Jean Rousseau, Paris Galster, G (1959), ‘Carolingian coins found in Denmark’, Coins and History, Copenhagen, 65–78, ad hoc editorial board (Danish version in Nordisk Numismatisk Årsskrift 1951) Gariel, E (1883–84), Les monnaies royales de France sous la race carolingienne, I–II, G. Fischbach, Strasbourg Gaudichard (1931), [trésor de Haute-Isle], RN, 244 Grierson, P and Blackburn, M A S (1986), Medieval European Coinage, Vol 1. The Early Middle Ages (5th–10th centuries), Cambridge University Press Haeck, A (1996), Middeleeuwse muntschatten gevonden in België (750–1433), Cercle d’études numismatiques, Bruxelles Haertle, C M (1997), Karolingische Münzfunde aus dem 9. Jahrhundert, 2 Vols, Böhlau Verlag, Cologne Hatz, G and Jonsson, K (1976), ‘Vikingatidens myntimport’, Kulturhistorisk Leksikon for Nordisk Middelalder 20, col 25–35, Rosenkilde & Bagger, Copenhagen Hawkins, E (1842–43), ‘An account of coins and treasure found in Cuerdale’, NC 5, 1–48 and 53–104 Lafaurie, J (1965), ‘Deux trésors carolingiens: Saumeray (Eure-et-Loir), Rennes (Ille-etVilaine)’, RN, 262–305 Le Maho, J (1995), ‘Rouen à l’époque des incursions vikings (841–911)’, BCDA 42 (1994), 143–202 Le Maho, J (1996), ‘Autour d’un millénaire: l’oeuvre architecturale à Rouen de Richard Ier (1996)’, Groupe de recherche archéologique du Pays de Caux, Bulletin, 33–49



Le Maho, J (1998), ‘Un exode de reliques dans le pays de la Basse Seine à la fin du IXe siècle’, BCDA 46, 137–188 Le Roy, P and Moesgaard, J C (1999), ‘Monnaies normandes du Xe siècle inédites ou rares – le trésor de Saint-Ours de Soleure (Suisse)’ BSFN, 142–145 Longpérier, A de (1847), Notice des monnaies françaises composant la collection de M. J. Rousseau, Rousseau, Paris Longpérier, A de (1869–70), ‘Louis d’Outremer en Normandie. Trouvaille d’Evreux’, RN 1869/70, 71–85 Malmer, B (1966), Nordiska Mynt före år 1000, C W K Gleerups förlag, Lund Marquigny, J and B (1983), ‘Le monnayage d’Evreux du IXe au XXe siècle’, Connaissance de l’Eure, 50, 2–25, Société libre de l’Eure Metcalf, D M (1990), ‘A sketch of the currency in the time of Charles the Bald’, in M T Gibson and J L Nelson (eds), Charles the Bald – Court and Kingdom, 2nd ed, 65–97, Aldershot, Variorum, London Metcalf, D M (1996), ‘Viking-age numismatics 2. Coinage in the northern lands in Merovingian and Carolingian times’, NC 156, 399–428 Moesgaard, J C (1995), ‘Stray finds of Carolingian coins in Upper Normandy, France’, in I Leimus (ed), Studia Numismatica, Festschrift Arkadi Molvõgin 65, 87–102, Eesti Ajaloomuuseum/Estonian History Museum, Tallinn Moesgaard, J C (1997), ‘A hoard from the Blois Region and the proto-feudal coinage of Blois, c. 920/40’, NC 157, 196–205, pl 44 Moesgaard, J C (1998), ‘Renovatio monetae en Normandie à l’époque ducale?’, BSFN, 127–131, 206 Moesgaard, J C (2003), ‘Le trésor de Saint-Taurim à Evreux (Xe siècle)’, Cahiers Numismatique, 158, 23–40 Morrison, K F and Grunthal, H (1967), Carolingian Coinage, The American Numismatic Society, New York North, J J (1994), English Hammered Coinage, Vol 1, 3rd ed, Spink, London Poey d’Avant, F (1858–62), Monnaies féodales de France, 3 Vols, Rollin, Paris Prou, M (1896), Catalogue des monnaies françaises de la Bibliothèque Nationale. Les monnaies carolingiennes, Bibliothèque Nationale, Paris Renaud, J (1989), Les Vikings et la Normandie, éditions Ouest-France, Rennes Rey, M van (1972), ‘Die Münzprägung Karls des Kahlen und die westfränkische Königslandschaft’, Die Stadt in der Europäischen Geschichte (Festschift Edith Ennen), 153–184, Ludwig Röhrscheid, Bonn Salaün, G (2000), ‘Un denier de Louis IV (936–954) découvert à Saint-Aubin-Epinay (SeineMaritime)’, BSFN, 13–14 Skaare, K (1966), ‘Die karolingischen Münzfunden in Skandinavien und der Schatzfund von Hon’, Hamburger Beiträge zur Numismatik, 393–408, Muesum für hamburgische Geschichte, Hamburg Skaare, K (1976), Coins and Coinage in Viking-Age Norway, Universitetsforlaget, Oslo Völckers, H H (1965), Karolingische Münzfunde der Frühzeit, Vandenhoeck & Ruprecht, Göttingen Wiechmann, R (1996), Edelmetalldepots der Wikingerzeit in Schleswig-Holstein (Offa-Bücher 77), Wachholtz Verlag, Neumünster


Viking Economies: Evidence from the Silver Hoards Märit Gaimster

The purpose of this chapter is to make a case for using an anthropological approach towards the Viking silver hoards. Having said that, I do not necessarily mean using anthropological models, but rather, to paraphrase L P Hartley, to acknowledge the past as a foreign country, or as we archaeologists like to say – a foreign culture: ‘They do things differently there.’ Looking back over the past 15 years, it will be useful for me to reconsider briefly the questions that have been raised, and the methods and sources that have been employed, when discussing this subject. I shall then go on to demonstrate the method used by myself when studying the silver hoards from two different regions in Scandinavia, and to discuss some of the results of my analyses.

A GENERAL VIEW OF VIKING SILVER Why did the Vikings leave so much silver in the ground? In some parts of Southern Scandinavia you can hardly stick a spade into the earth without hitting silver. They were certainly violent times, but somewhere along the way the simplest explanation – that all the owners of this silver died in Viking raids before they could retrieve their riches – has been given up. But why did these Scandinavians not mint their own coins when they clearly had so much silver? And why do some hoards resemble the last processing stages of an, albeit very expensive, scrap-yard? On the surface, it is not difficult to understand how silver of this kind was used in Scandinavia and other parts of the Viking world. We are talking weight economies in which the silver was weighed, bent and pecked to control its fineness. This is also confirmed by the many finds of weights and scales from burials (Steuer 1987). Above all, the presence of hack-silver (cut-off and fragmented bits of coins, ingots and jewellery) is important as it demonstrates that silver was used as a means of payment. This fragmentation reflects a need for smaller units, which may be compared with the small change, the base-metal coins, of monetary economies (Hårdh 1996, 27). In this sense, a high degree of fragmentation in hoards could be described in terms of a more intensive use of silver for small and everyday transactions (cf. Kiersnowski 1956). This aspect of Viking silver has been investigated by several scholars (eg Kiersnowski 1956; Tabaczynski 1958; Hårdh 1976; 1996; Wiechmann 1996). An important finding, 123



particularly in hoards from south-west Scandinavia, has been the weight-correlation between hack-silver and coins. The variations in composition within datable hoards from this region show a sequence from mainly unfragmented silver, in the late eighth and early ninth centuries, through phases of increasing fragmentation, to hoards dominated by intact, and increasingly domestic, coins in the eleventh century. This has been interpreted as showing the changing function of silver as a means of payment, through time, with the hack-silver representing a phase in the development towards a monetary economy (Hårdh 1996, 104–107; cf. Hårdh 1976, 140–143). Besides coins, hack-silver, bars and ingots, another characteristic feature of Viking hoards is the inclusion of many arm- and neck-rings. Their presence has long been recognised as representative of something more than an economic function; they were personal ornaments and objects of status and symbolic value. In relation to the hack-silver hoards, regions – or periods – dominated by silver in this form may be regarded as less economically advanced (cf. Hårdh 1996, 168–170). But the general phenomenon of Viking silver hoards may also be explained in economic terms; even when silver had advanced functions as a means of payment, it may not have had a practical use in dailylife. This phenomenon has been argued for Finnish coin hoards from the eighteenth century (Sarvas 1967). By and large, therefore, the cessation of Viking silver hoards would seem to reflect the transition into a monetary society. But here it is necessary to stop and remember that such a linear development may be a rear-mirror construction of changes in the past.

SOME RESERVATIONS On the one hand, it is clear that the presence of broken-up jewellery and small units of precious metal does not necessarily lead to the development of a domestic coinage. From the Iberian peninsula there are Celtic hack-silver hoards, from the third to second centuries BC. Some of these hoards bear an uncanny similarity to the much later Scandinavian finds in their composition of foreign coins, hack-silver and domestic arm-rings (Raddatz 1969). In the late Roman period silver hoards are known from Scandinavia, the continent and the British Isles. The finds include coins, bars, ingots and broken-up fragments of both domestic objects and Roman silver plate (Gaimster 1992a, 5–6). Fragmented objects and foreign coins also appear in some of the many gold hoards that are characteristic of Southern Scandinavia in the fifth and sixth centuries AD (cf. Fonnesbech-Sandberg 1991; Hedeager 1991). It is true that the volume of earlier finds with hack-silver or fragmented gold objects is far smaller than during the Viking period. Nevertheless, it is important to recognise that these features, and the economic processes behind them, were not something new. Rather, such finds demonstrate that the non-monetary areas of early medieval Europe were also capable of more sophisticated systems of exchange. Ultimately, this demonstrates that coins are not necessary for economic transactions. On the other hand, the appearance of domestic coinages in Scandinavia shows that minted silver also had important socio-political functions. This is perhaps still best demonstrated in Anders Andrén’s (1983) study of the early Danish royal coinages. Here the output of mints was found to be in discordance with the more economically stable and prosperous parts of the Danish kingdom. The more voluminous minting in east Denmark



could be related to the different structures of royal control; unlike west Denmark this was based not on patrimonial estates, but on the crown demesne. The function of this coinage is best perceived in the administration of these estates, and in the establishment of royal supremacy in east Denmark. That early coinages were strongly motivated by social and political, rather than economic, functions is well recognised. One has only to think about the Roman and Byzantine empires, where the output of high-value gold and silver coins was determined both by administrative needs and by the needs of the emperor to express and demonstrate his power (Crawford 1970, 40–48; Reece 1988, 271–278). A dominating function as revenue, in the form of tax, tolls and fines, is also characteristic of the early Germanic precious-metal coinages (Gaimster 1992a, 7–8). Hence, the term ‘early cash’, developed by economic anthropology, refers to the introduction of coins in terms of a medium that is both uniform and controlled, and its connection with the formation of states (Dalton 1977). There are, however, also early medieval coinages that clearly appear to be motivated by trade. The continental series of eighth-century silver sceattas – as well as the earliest, and anonymous, south-western Scandinavian coinages – show a distribution that is associated with known trading-places, such as Ribe, Birka and Hedeby (Callmer 1980; 1984; Metcalf 1984; see also Malmer and Wiechmann, Chapters 2 and 3, in this volume).

VIKING SOCIETY I shall return to the question of different uses and meanings of precious metal, but it is surely agreed, to some extent at least, that between the two phenomena of silver hoards and a domestic monetary economy, or between the Viking period and the Middle Ages, lies a broad spectrum of social change. This includes new forms of regional control and social organisation, the development of towns and, not least, of an organised Christian church. The elements that make up these changes all have histories of their own; power and control over men and land was nothing new, neither was trade and trading-places. Coins and other means of payment and exchange already existed, and even Christianity was well known. What sets the Middle Ages apart from the Viking period is perhaps, above all, a change of mentalities. Following in the tradition of the French Annales School, this has been described succinctly by the Russian historian Aaron Gurevich. Among the categories that formed part of the image of the world, attitudes towards labour, property and wealth may be informative for our understanding of Viking silver. In contrast to the free citizen of the Classical world, who concerned himself with idleness and religious sacrifices as well as athletics, warriorship and feasting, there stands the Christian attitude towards idleness as a sin and the need to consider wealth in the light of salvation in the after-life (Gurevich 1985, 211–285). To place the Viking world within the Classical tradition is not as far-fetched as it may seem. The impact from the late Roman world on the development of the Germanic peoples is today well understood (cf. Webster and Brown 1997). Among the central elements that were operative in the early Middle Ages were cult, feasting and the ideal of the warrior. Warfare and allegiance to successful kings and chieftains, not toil on the land, were decisive factors in the formation of the new barbarian kingdoms. Phenomena such as



large-scale war-booties, to celebrate victories, clearly reflect attitudes towards wealth that differ from the medieval world. This is a mentality that we also meet in the Norse sagas where codes of behaviour in the Viking period may be studied. Following the French anthropologist Marcel Mauss, the now familiar institution of gift-giving and generosity has been illustrated by numerous scholars (Gurevich 1968; Durrenberger 1982; Vestergaard 1991). Here arm- and neckrings play a prominent role as a recognised medium of exchange, something that is also illustrated by Baugatal, the Icelandic law on blood-money, which literally translates as ‘the counting of rings’ (Gaimster 1991, 117–118). Basically, Norse literature, like the earlier Anglo-Saxon epic Beowulf, demonstrates that wealth in itself was not an ideal; it had to be used – and used in the correct way and in the right amounts. Rather, it was the creation and maintenance of social relationships – a gift demands a gift – that was central. But there was also a sacral and magical aspect to wealth, in that it incorporated the owner’s personal qualities, fortune and happiness. A gift, therefore, brought with it a bit of the donor’s success and social standing and, conversely, to be mean with ones wealth was to withhold its magical powers and good fortune. This attitude towards wealth is not least important in offering another explanation for silver left buried in the earth. In contrast to the practical-economic understanding of passive wealth, hoards may have had active sacral functions in preserving their owner’s success into the after-life. The sagas also contain references to this act of hoarding in the face of danger and death (Gurevich 1968).

APPROACHING VIKING ECONOMIES Having considered the significance played by socially motivated exchange in the early medieval period, there are plenty of reasons to acknowledge that Viking society was not a modern market-economy. To focus only on the practical-economic function of the silver would leave us with an understanding of Viking society that is linear and deterministic, as well as anachronistic. But is it enough to quote a few Norse or Anglo-Saxon poems to explain Viking silver hoards? I think not. The silver hoards provide a unique source for the Viking period. Unlike the literary sources, most of which were written down much later, and in a Christian context, they are contemporary remnants of Viking society. And also, unlike the literary sources, they represent large areas of Scandinavia and other parts of the Viking world, not predominantly Iceland and the West. Moreover, the silver in the hoards reveals a complexity of forms and functions, from the practical and economic to the social, political and sacral. This appears to me to be an important feature. Studies focused on gift-giving, outdoings and the destruction of wealth often ascribe the silver few if any practical functions. Its central function was semiological; it was a language of social status, authority, friendship and allegiance. Trade as such would instead have been carried out by means of barter and exchange in kind. The rosy image of Viking society characterised by social negotiations and barter, however, is long outdated (Samson 1991). Where anthropological approaches were initially based on studies of primitive societies with elaborate indigenous systems of exchange, later models have taken inspiration from more recent situations in the colonial period. Here the contact and



interaction between different economies and exchange systems is central, and differing expectations, demands, access and attitudes towards exchange have to be considered (cf. Smith 1976; Callmer 1977, 174). To me, this is a relevant approach for early medieval Europe with its long-term contacts with the Roman Empire, its development of Germanic kingdoms and longdistance contacts with the Arab world. In the Viking period there were many different ways for silver to be accumulated – the usual suspects being trade, plunder and tributes. On a local and regional level, there also needs to be added the social spectrum of gifts and various institutions of compensation, such as wergild, bride-wealth and fines. Rather than reflecting stages of development within an economic system, I would see the complexity of the silver hoards as meaningful, where the different categories such as foreign coins, hack-silver and personal ornaments may be understood in terms of a society with multiple and co-existing systems of exchange. Here the terminology developed by economic anthropology is a helpful, if limited tool. While terms such as ‘valuables’, ‘status objects’ and ‘socially motivated trade’ remain vague, however, the understanding of the precious metal as ‘special-purpose money’ must be considered significant. As a means of payment, silver in its different forms could clearly be motivated by social and political as well as economic interests, and this would have depended on the context and situation where exchange or accumulation took place. In what follows I shall demonstrate my own methodology for the study of Viking silver hoards, with the object of identifying the use of silver in different economic spheres; here the term ‘economy’ is used in its manifold and socially embedded sense. While my study confirmed that this is a meaningful approach to silver hoards, it also raises questions as to the relevance of silver to Viking economies. Ultimately, we need to confront the evidence or interpretations based on the hoards with other aspects of regional organisation, above all production and how this was organised.

SILVER HOARDS AND REGIONAL ECONOMIES The two regions under study were the Baltic island of Öland in south-east Scandinavia, and the area corresponding to medieval Denmark in the south-west ( Thurborg 1988; Gaimster 1992b). It should be stressed, of course, that there were differences between these two regions in terms of political development. In the south-west, the development of a more centralised power took place far earlier than in other parts of Scandinavia, and the region includes well-developed early trading-centres such as Ribe and Hedeby. In addition, as noted above, the minting of domestic coins also started earlier in Denmark than elsewhere in Scandinavia (cf. Hårdh 1996, 170–171). Regardless of these differences, the same method of study could be applied to the silver hoards in both regions. This was based on a division of the silver into four categories: (i) coins; (ii) bars and ingots; (iii) hack-silver and (iv) personal ornaments. Of the latter only the arm- and neck-rings, which are the most proliferate form throughout the Viking period, were studied separately. The presence of coins in the finds enabled a study of changes in the regional use of silver through time, but their varied provenance – including Arabic, West European and Scandinavian issues – also provides information about external contacts and the contexts




number of different media

of exchange. To what degree foreign coins circulated as a means of payment in Scandinavia is a difficult question, where features like peck-marks or the transformation of coins into pendants certainly suggest regional differences in the use of this medium. In regions with signs of a weak coin circulation, such as Gotland, this has been understood to confirm that the silver had a limited use in the regional economy and was deposited in the earth shortly after its importation (Malmer 1985). In this respect, however, it is significant that in all regions of Scandinavia the imported silver was transformed into various types of domestic media. The most obvious as a means of payment are anonymous forms, such as bars and ingots and ring- or spiralshaped silver bars. But the silver was also turned into elaborate arm-rings and other personal ornaments, such as penannular brooches and pendants. The phenomenon of hack-silver is a further element that, as we have seen, may be recognised as an important indication of economic activity. Using the terminology developed within economic anthropology, these different forms of silver may be described as ‘special-purpose money’, with predominantly economic or social functions. To investigate the silver further as separate media, the different individual categories were related to complexity within the hoards. That is, they were considered in terms of their degree of combination with other forms of money. A largely separate sphere of circulation could be seen for arm- and neck-rings in both regions (Figure 7.1). This fits well with our understanding of the social context and meaning of such rings in literary sources. In contrast, bars, ingots and hack-silver were found to be associated with more complex hoards (Figures 7.2 and 7.3). By and large, these differences may be understood in terms of a social versus an economic use of silver. However, considering the meaning behind the various types of

4 3 2 1 0





number of different media

percentage of hoards

4 3 2 1 0





percentage of hoards

Figure 7.1 The distribution of arm/neck-rings in hoards of different complexity: (a) hoards from Öland, tenth–eleventh centuries; (b) hoards from Denmark, ninth–twelfth centuries.


number of different media



4 3 2 1 0








number of different media

percentage of hoards

4 3 2 1 0



percentage of hoards


number of different media

Figure 7.2 The distribution of bars and ingots in hoards of different complexity: (a) hoards from Öland, tenth–eleventh centuries; (b) hoards from Denmark, c 970–1016 (for hoards c 800–970, and c 1016–1150, see Gaimster 1992b, fig 8).

4 3 2 1 0







number of different media

percentage of hoards

4 3 2 1 0




percentage of hoards

Figure 7.3 The distribution of hack-silver hoards of different complexity: (a) hoards from Öland, tenth–eleventh centuries; (b) hoards from Denmark, c 970–1016 (for hoards c 800–970, and c 1016–1150, see Gaimster 1992b, fig 9).



‘special-purpose money’, and adapting this to the context of Viking society, gives a more complex insight into the processes of exchange. This can be visualised as a series of transformations from intact rings to hack-silver, by which time the rings have lost all contact with their original function. Many hoards, in particular on the Baltic islands of Öland and Gotland, contain twisted and broken rings, or larger fragments. The distinction between the process of fragmentation and the original transformation of an object into hack-silver is important for our understanding of exchange at a regional level. As a phenomenon, hack-silver is primarily indicative of an economic sphere in which exchange was not continuous, of one in which personal valuables could be transformed into money whenever the need or opportunity arose. The presence of finely fragmented hack-silver within some hoards, however, is important. Rather than a general use of small change, this would seem to indicate a more specialised economic situation. In earlier or primitive societies, a balanced and impersonal exchange is characteristic of border situations, at the meeting-point between different regions or different economic systems. Not surprisingly, finely fragmented hack-silver has been found on trading-places, but there are also signs within hoards that this material circulated across long distances. A further element is its close association with other anonymous media such as bars and ingots. While the unminted silver reflects different uses within regional and supra-regional contexts of exchange, diverging spheres of circulation could also be identified among the coins. In this respect, however, there are marked differences between south-west Scandinavia and Öland. The Öland hoards fall naturally into two groups: the first consisting purely of Arabic coins; and the second, from the last quarter of the tenth century onwards, of West European issues. In the Danish hoards the picture is altogether more complex, with a more gradual inflow of West European coins and the development of a domestic coinage. There is also the interesting anonymous Nordic coinage, commencing in the ninth century (Malmer 1966, and Chapter 2, in this volume). At least initially, these coins, which were probably minted in Hedeby, appear to have a strong connection to trading-places and a function that was more pronouncedly economic (Callmer 1980). In Denmark the relationship of the coins to other forms of money shows a high degree of complexity, until the final phase of the silver hoards when there is a clearly discernible Danish minting under Cnut the Great. In this phase, the distribution of coins within the hoards resembles that of the arm-rings with a largely separate sphere of circulation (Gaimster 1992b, fig 5a). Again, this would seem to confirm the picture of a successful development into a monetary economy. However, as Anders Andrén’s (1983) study of the Danish minting as a medium for royal control and administration at this time showed, the coinage was not necessarily motivated by trade or a more widespread use of coins as a means of exchange. Looking at the geographical distribution of silver hoards in south-west Scandinavia, however, it is interesting that there appears to have been a contraction of finds to Scania and east Denmark already in the phase before the Danish minting (Gaimster 1992b, figs 2 and 3). This situation can be compared with Andrén’s finding, that the output of coins under Cnut and Hardacnut was largest in this part of Denmark, and it raises further questions as to the complex use of silver. How did social and economic spheres interact in the development towards a coin-using economy? How did the existing economic system of coastal trading-places relate to other elements of Danish state formation (cf. Ersgård 1986; Gaimster 1992b, 284)?



The coins in the Öland hoards yield further information about diverging economic spheres. On Öland, Arabic coins were deposited singly to a far greater extent than West European ones (Thurborg 1988, fig 6). The presence of Arabic coins on Öland is generally understood in terms of their accumulation outside Scandinavia, in a context of tributes and trade in slaves and furs. It is interesting to consider the deposited coins against arm- and neck-rings, for the two appear in almost mutually exclusive economic spheres. In this context, the small extent to which the Arabic coins were transformed into domestic media may indicate that the actors and activities further east had little influence on the regional economic system (Thurborg 1988, 320). Silver and its use on Gotland, by far the silver-richest region of Scandinavia during the Viking period, presents a particularly interesting problem in light of the vast quantity of bronze that was also being imported onto the island; this is testified by the jewellery that was continuously deposited in burials, and which shows a steady increase in size through time (Gaimster 1998, 245). This undoubtedly presents a problem if the silver is to be seen as the result of balanced trade. On the other hand, if the silver was accumulated through plunder and tributes, how does this economic sphere relate to an obviously stable regional organisation, able to maintain a continuous supply of raw-material? Reconsidering my own analyses of the Viking silver hoards, they seem to have been successful, even if they leave me with more questions than I had at the time. Rather than peaceful barter and gift-giving, I now end up with a picture of regional economies characterised by diversity, conflict and change. But perhaps this makes sense when considering that this was the period of transformation into the medieval world?

ACKNOWLEDGEMENT I am grateful to Chris Jarrett at Pre-Construct Archaeology for helping me with the graphics for this chapter.

References Andrén, A (1983), ‘Städer och kungamakt – en studie i Danmarks politiska geografi före 1230’, Scandia 49/1, 31–76 Callmer, J (1977), Trade Beads and Bead Trade in Scandinavia ca. 800–1000 A.D. (Acta Archaeologica Lundensia Series in 4°, no. 11), Rudolf Habelt Verlag, Bonn Callmer, J (1980), ‘Numismatics and archaeology: some problems of the Viking period’, Fornvännen 71, 175–185 Callmer, J (1984), Sceatta Problems in the Light of the Finds from Åhus (Scripta Minora), Gleerup, Lund Crawford, M H (1970), ‘Money and exchange in the Roman world’, Journal of Roman Studies 60, 40–48 Dalton, G (1977), ‘Aboriginal economies in stateless societies’, in T K Earle and J E Ericsson (eds), Exchange Systems in Prehistory, Academic Press, New York, 191–212 Durrenberger, E P (1982), ‘Reciprocity in Gautrek’s saga: an anthropological analysis’, Northern Studies 19, 23–37 Ersgård, L (1986), ‘Kämpinge och köpingarna – om tidigmedeltida handelsplatser i Sydvästskåne’, in A Andrén, M Anglert, U-B Ekstrand, L Ersgård, H Klackenberg, B Sundnér



and J Wienberg (eds), Medeltiden och arkeologin. Festskrift till Erik Cinthio (Lund Studies in Medieval Archaeology 1), Lund University: Institute of Archaeology, 313–323 Fonnesbech-Sandberg, E (1991), ‘Guldets funktion i aeldre germansk jernalder’, in C Fabech and J Ringtved (eds), Samfundsorganisation og Regional Variation. Norden i romersk jernalder og folkevandringstid (Jysk Arkaeologisk Selskabs Skrifter 27), Højbjerg, 233–244 Gaimster, M (1991), ‘Money and media in Viking Age Scandinavia’, in R Samson (ed), Social Approaches to Viking Studies, Cruithne Press, Glasgow, 113–133 Gaimster, M (1992a), ‘Scandinavian gold bracteates in Britain: money and media in the Dark Ages’, Med Arch 36, 1–28 Gaimster, M (1992b), ‘Premonetary media in Viking Age Denmark’, in J Okulicz-Kozaryn and W Nowakowski (eds), Studien zur Archäologie der barbarischen Völker aus der Ostseeküste und aus dem Weichselgebiet (BARBARICUM, Band 2), Instytut Archeologii Uniwerytetu Warszawskiego, Warszawa, 278–295 Gaimster, M (1998), Vendel Period Bracteates on Gotland: on the Significance of Germanic Art (Acta Archaeologica Lundensia Series in 8°, no. 27), Almqvist & Wiksell International, Lund Gurevich, A (1968), ‘Wealth and gift-bestowal among the ancient Scandinavians’, Scandinavica 7, 126–138 Gurevich, A (1985), Categories of Medieval Culture, Routledge, London Hedeager, L (1991), ‘Gulddepoterna fra aeldre germanertid – forsøg på en tolkning’, in C Fabech and J Ringtved (eds), Samfundsorganisation og Regional Variation. Norden i romersk jernalder og folkevandringstid (Jysk Arkaeologisk Selskabs Skrifter 27), Højbjerg, 203–212 Hårdh, B (1976), Wikingerzeitliche Depotfunde aus Südschweden. Probleme und Analysen (Acta Archaeologica Lundensia Series in 8° minore, no. 6), Gleerup, Lund Hårdh, B (1996), Silver in the Viking Age: a Regional-Economic Study (Acta Archaeologica Lundensia Series in 8°, no. 25), Almqvist & Wiksell International, Stockholm Kiersnowski, R (1956), ‘Glowne momenty rozwoju srodków wymiany na Pomorzu wczesnofeudalnym’, Wiadomocfci Archeologiczne 23, 229–251 Malmer, B (1966), Nordiska mynt före år 1000 (Acta Archaeologica Lundensia Series in 8°, no. 4), Lund Malmer, B (1985), ‘Circulation of monetary silver in the Baltic during the Viking Age’, in S-O Lindquist (ed), Society and Trade in the Baltic during the Viking Age (Acta Visbyensia 7), Gotlands fornsal, Visby, 185–194 Metcalf, D M (1984), ‘Monetary circulation in southern England in the first half of the eighth century’, in D Hill and D M Metcalf (eds), Sceattas in England and on the Continent (BAR, British Series 128), Oxford, 27–69 Raddatz, K (1969), Die Schatzfunde der Iberischen Halbinsel vom Ende des dritten bis zur Mitte der ersten Jahrhunderts vor Chr. Geb. (Madrider Forschungen 5), Walter de Gruyter & Co, Berlin Reece, R (1988), ‘Theory and practice in Roman coinage’, in P Kos and Z Demo (eds), Studia Numismatica Labacensia. Alexandro Jelocnik Oblata, Narodni Muzej, Ljubljana, 271–278 Samson, R (1991), ‘Economic anthropology and Vikings’, in R Samson (ed), Social Approaches to Viking Studies, Cruithne Press, Glasgow, 87–96 Sarvas, P (1967), ‘De finska myntskatterna från 1700-talet’, NNÅ, 23–146 Smith, C (1976), ‘Exchange systems and the spatial distribution of elites: the organization of stratification in agrarian societies’, in C Smith (ed), Regional Analysis, II. Social Systems, Academic Press, New York, 309–373 Steuer, H (1987), ‘Gewichtsgeldwirtschaften im frühgeschichtlichen Europa. Feinwaage und Gewichten als Quellen zur Währungsgeschichte’, in K J Düwel, H Jankuhn, H Siems and D Timpe (eds), Untersuchungen zu Handel und Verkehr der vor- und frühgeschichtlichen Zeit in Mittel- und Nordeuropa, Vandenhoeck & Ruprecht, Göttingen, 405–527 Tabaczynski, S (1958), Z badán nad wczesnosredniowiecznymi skarbami srebnymi Wielkopolski, Warszawa



Thurborg, M (1988), ‘Regional economic structures: Viking Age silver hoards from Öland, Sweden’, World Archaeology 20:2, 302–324 Vestergaard, E (1991), ‘Gift-giving, hoarding, and outdoings’, in R Samson (ed), Social Approaches to Viking Studies, Cruithne Press, Glasgow, 97–104 Webster, L and Brown, M (eds) (1997), The Transformation of the Roman World AD 400–900, British Museum, London Wiechmann, R (1996), Edelmetalldepots der Wikingerzeit in Schleswig-Holstein. Vom ‘Ringbrecher’ zur Münzwirtschaft (Offa-Bücher 77), Wachholtz, Neumünster


Oriental-Scandinavian Contacts on the Volga, as Manifested by Silver Rings and Weight Systems Birgitta Hårdh

The purpose of this chapter is to present some preliminary thoughts about the so-called Permian rings and a group related to them. These rings are made of rods, circular in crosssection and ornamented with a characteristic continuous striation. The terminals consist of a clasp formed as a hook and a faceted knob (Figure 8.1). The distribution of these rings is striking. Their main concentration is the taiga landscape north of the Volga bend, primarily the Vjatka and Perm gouvernements west of the Ural mountains, around the rivers Vjatka and Kama.1 However, there are also some in the Baltic area, with the emphasis being on both sides of the Gulf of Finland and on the islands of Gotland and Öland. Some scattered examples also occur on the south coast of the Baltic and, finally, there is a small, rather isolated, group in Jutland and Schleswig-Holstein (Figure 8.2). Occasionally they have been found with coins, primarily with Abbasid or Ummayad dirhams, and thus they belong to the ninth or early tenth century. They seem to have arrived in the Baltic area in connection with dirhams, but before the great influx of the Samanids.

TRADE CONNECTIONS IN THE ABBASID ERA The period from the eighth to the eleventh century has been called the ‘Golden Age of Islam’ by the French historian Maurice Lombard (1992). During these centuries the Islamic world was crisscrossed by trading routes that linked distant parts of the Old World into an advanced network. The Central Asian transit route, from Samarkand and Bukhara – via Baghdad – to Syria, created the driving force for civilisation and economy during the Abbasid era. The consequence of the Arabic conquest was that trading regions, which earlier had been isolated from one another, now became linked together into an immense network, including China and India, the Sudan-Sahara region and the vast forest areas around the great Russian rivers. Through this network goods of various kinds flowed: gold, textiles (as wool, cotton and silk), timber and slaves. From Samarkand to Córdoba, the Moslem world was remarkably homogenous with great flow of people, goods and ideas. It consisted of a series of islands, linked together by trading routes. Beyond this 135



Figure 8.1 ‘Permian’ rings found on Öland (© Birgitta Hårdh).

world there was a periphery, strongly influenced in part by the Islamic world and its civilisation (Lombard 1992, 19 et seq, 27 et seq). The establishment of Baghdad in 762 played an especially important role for economic growth during the Abbasid era, not the least for the trade via the Caspian Sea to Khazaria and Eastern Europe. Khazaria clearly worked as an important intermediary between the Arabs and Europe during the ninth and tenth centuries. Islamic merchants created trade ties to the Khazars and through them could have acquired, for example, furs and slaves for which a ready and affluent market already existed in Iraq (Noonan 1984, 263 et seq). Contacts with the northern forest regions had long traditions. During the sixth and early seventh centuries Persian silver vessels, as well as Persian and Byzantine silver coins, reached the regions around the Vjatka and Kama basins, through trade rather than through non-commercial activities. Several of the vessels, with stamps from ruling dynasties in Constantinople or Persia had probably, according to Franklin and Shepard, been exchanged for furs. These vessels were highly esteemed in the Kama region for ritual purposes, or as wealth accumulation. Sometimes graffiti were incised on them. Franklin and Shepard maintain that Persian and other Oriental tradesmen acted in the north (Franklin and Shepard 1996, 7 et seq). Darkevich mentions fifth- to sixth-century Sassanid and Byzantine vessels in the Ural region. However, he maintains that trade was still sporadic along the Volga route at the beginning of the seventh century and that the main share of the eastern vessels did not reach the Ural region until after AD 800 (Darkevich 1976, 147). Noonan stresses that Islamic relations with Eastern Europe between 650 and 800 were mainly defined by the Arab-Khazar relations in the Caucasus.

Figure 8.2 Distribution of complete ‘Permian’ rings: (1–2) Perm/Vjatka area; (3–4) Moscow area; (5) Finland; (6) Estonia; (7) Gotland; (8) Öland and (9) Jutland.




This trade had been carried out for centuries and Arabs and Khazars were new partners in an old trade relation. The Vikings first appeared, according to Noonan, at the stage when the Arabs and the Khazars had laid the foundations for this commerce (Noonan 1981, 52 et seq). However, there are also indisputable signs of contacts between the Middle VolgaUral area and the Baltic in both grave rituals and the spread of certain objects already from the seventh and eighth centuries (Callmer 1994, 13 et seq, 30, figs 1 and 16). From late eighth and early ninth century other types of finds may also be informative. Callmer has mapped the occurrence and composition of various types of Oriental beads. The distribution indicates a trade route through Eastern Europe along the rivers Volga and Dnjepr. The region around the Volga bend has a dense concentration of such beads (Callmer 1991, 33, fig 4). The distribution of the beads clearly indicates, as Callmer maintains, that the long-distance trade routes were regularly used and that the tradesmen are to be regarded as agents of trade in the period when the Abbasids were at the summit of their economic and political power (Callmer 1991, 35). Dirhams appear, according to Noonan, in Eastern and Northern Europe in the last quarter of the eighth century. Their presence is mainly due to trade between the Islamic world and the northern areas (Noonan 1984, 152 et seq). Recently, Sebastian Brather has, mainly through compiling works by Noonan and others, made a survey of how the network of contacts in the Viking Age might have appeared from finds of Arabic dirhams (Brather 1996, figs 4 and 7). First and foremost the Russian hoards seem to derive from coins circulating in the Near East. A contribution of Caucasian issues indicates that the Arabic coins came along a route via the Caucasus. This route can be appreciated in greater detail from written sources where Itil, on the Volga estuary, is also referred to as an important centre. However, this map does not present an indisputable picture of the Volga as the intermediate link. Rather, the distribution might indicate that either the Dnjepr or the Don was the main route up to the regions around the Upper Volga. The map shows hoards with a terminus post quem earlier than AD 900. The earliest dirham hoards in Russia have been found in Staraja Ladoga and at the tributaries of Upper Volga. Noonan maintains that the early dirham hoards at the Upper Volga indicate, that Slavonic tribes at this time obtained silver through fur trade (Noonan 1986, 342). There are several Islamic sources which report that Rus merchants exchanged furs and slaves against dirhams (Noonan 1986, 153). The earliest dirhams arrived in Russia, according to Noonan (1986, 340 et seq), through mediation by the Khazars. The Volga Bulghars supplemented eventually the Khazar activities in the Volga region. Noonan regards the Khazar-Islamic trade, and the inflow of dirhams into Northern Russia as the catalyst that brought the Vikings into Russia (Noonan 1984, 277 et seq). The contact between the Moslem world and the markets around the Upper Volga and its tributaries are marked by coin finds. Here the main share of dirhams has been found and here the oldest Russian towns arose. Various scholars have discussed the kind of goods that could have been obtained in this area. Amber, furs, wax and honey are known from written sources and were obviously important. Warnke (1987) has emphasised the importance of forest products. However, according to Haussig (1987, 530 et seq), only trade in slaves can be the explanation for the big silver deposits, whereas Martin (1986, 167) points out that the fur trade was a constant, ever present, economic factor in the Rus and mid-Volga lands throughout all the centuries considered here.



SILVER RINGS OF ‘PERMIAN’ TYPE What are ‘Permian’ rings? First of all, it should be noted that they are of standardised manufacture, above all according to weight. The table (Table 8.1) shows clearly that there are distinct weight groups. The largest and heaviest rings weigh about 400 g, with a large group weighing around 200 g and some around 100 g; there is also a small group of rings weighing about 300 g. A few rings differ from this standardised scheme, but it is worth noting that rings in this latter group are often made with less care than the others, with coarser and less regular striation. A reasonable interpretation is therefore that these rings are money, in large units. What is also obvious is that the heaviest rings are found in the East (Figure 8.3). The really heavy ones are not found in the Baltic region or in Scandinavia, but in Perm and Vjatka. Rings from the Baltic area belong to the lighter groups, weighing c 200 g or, more usually,

Table 8.1 Distribution of weights among ‘Permian’ rings (in g). Russia: Sweden: Denmark:

412–423 4

291–294 2 309 1

196–212 56 195–208 8 198–201 3

95–107 10 95–103 21

The weights of the following rings from Russia do not fit into these groupings: 72.5; 138.1; 171.7; 175.4; 177.6; 178.9; 179.8; 179.0; 182.2; 184.0; 189.9 grams

Figure 8.3 Weight grouping of the ‘Permian’ rings in Russia, Sweden and Denmark.



c 100 grams. However, they are made in total accordance with the rings found in Russia. Is it possible that the heavy rings were hoarded in Russia whereas the lighter ones were exported, according to some variety of Gresham’s law? Or was trade perhaps conducted on a larger scale in Russia than in the Baltic area, with larger units of payment involved? The clasps, consisting of a hook and a facetted knob, show that the rings were made as neck-rings, although they have often found wound-up into spirals, especially in the Baltic area. The facetted knobs have great similarities to the polyhedral weights, well known in Scandinavian Viking-Age contexts (Figure 8.4). The terminal knobs of the rings are also often decorated with dots or circles exactly like the weights. Steuer (1978) has demonstrated, in a study of weights and balances from the Iron Age, that Scandinavia and the Baltic area were strongly influenced by Arabic weight systems, especially in the ninth and early tenth century. At the end of the ninth century, spherical and polyhedral weights appear in the Baltic area together with balances. The prototypes of these balances and weights are to be sought in the Orient. The Khazars and Volga Bulghars probably participated in the distribution although their part is not elucidated. Ibn Fadlan reported that the money-changers sold cubes, spheres and dirhams, which probably means that they sold polyhedral and spherical weights together with dirhams. In the Khazar area, between the Don and the Donez, balances and weights for precious metals have been found in graves. Spherical weights are known from this area (Steuer 1997, 46). The polyhedral shape often occurs as ornaments on balance beams probably to express a connection with the weights. The polyhedral knobs on these rings may likewise demonstrate a connection between the rings and the weight systems.

Figure 8.4 Polyhedral weights found at Uppåkra, Skåne, Sweden (© B Almgren, LUHM).



Figure 8.5 Some terminal knobs with graffiti (scale 1:1): upper row, Russia; lower row, Gotland (© Birgitta Hårdh).

Another feature that connects rings of Permian type with means of payment is the graffiti that occasionally appears on the terminal knobs. So far I have been able to substantiate this on 26 rings from Russia and four from Gotland. The incised figures are to some extent the same as those that occur on Oriental coins found in Scandinavia, as well as in Russia. Sometimes they look like Scandinavian runes, but sometimes they are signs of other types: swastikas, crosses, stars and so on (Figure 8.5). Among complete Permian rings, the ones from the southern part of Jutland have the most south-western distribution, but fragments are known also from Schleswig-Holstein, Norway and the British Isles. The Danish rings are the same as those found on the Baltic islands and in Russia, in shape as well as in their weights. However, in Denmark as well as on Gotland, another group of rings has also been found, differing in both weight and execution, even if it is possible to claim a connection. These rings are striated and stamped like the others, but their terminals are much simpler, ending in hooks; above all their weights are different (Figure 8.6). These rings are clearly also weight-adjusted, but are light (c 100 or 50 g) in contrast to their heavier relatives (Munksgaard 1963). As far as I know, there are no indisputable occurrences of these rings in Russia, although there are fragments of a striated ring with the same type of hooks in a find from Vjatka (State Historical Museum, Moscow: inv no. 39942). It has been much debated whether these rings are South Scandinavian copies of the Permian rings or whether they too were imported from the East. Their uniform design and clear weight grouping indicate that they are value units, a kind of money. Worth noticing also is that they agree completely with the Russian ‘grivna’ weight (see next section). Whether they were made in Denmark or not, they show an indisputable connection to their heavier relatives.

WEIGHT SYSTEMS The weight of the Permian rings may be connected to the Old Russian pound of 408–410 g. A fraction of the pound, 1/8 (that is c 50 g), is called ‘grivna’ – meaning,



Figure 8.6 A small striated ring (after Stenberger 1958, Textabb 18).

moreover, a neck-ring. This is a weight standard corresponding to 20 dirhams. According to Nazarenko (1994, 76 et seq), this weight-money-unit was established for the needs of long-distance trade with Byzantium as well as with the Arabic world where gold formed the basis for the currency, whereas silver dominated in Russia: 1 dinar ⳱ 20 dirhams. Arne traces the origins of the Russian pound from the Iraqi pound of 408 g (⳱96 Sassanian drachmes); the Iraqi pound was probably in use in Mesopotamia and Persia immediately before the Arabic conquest (Arne 1914, 181). In Russia ingots of standardised shape and weight were used as means of payment during the period AD 1000–1400. These ingots are roughly formed and have only the standardised weight in common with ‘Permian’ rings. Ingots are also known from early hoards, from the ninth century, but these, according to Noonan, do not seem to be standardised by weight (Noonan 1987, 403). Otherwise, as pointed out by Kotljar (1994, 81), the means of payment in silver during this period was totally dominated by Arabic coins. However, in the ninth century we also find, beside the dirhams, the extremely wellmade, weight-adjusted ‘Permian’ rings. Fekhner, who has done a study of neck-rings from northern Russia maintains that they are not common in graves from the eighth–ninth centuries; not until the eleventh century, and then mainly in its later part, did they come into fashion (Fekhner 1967, 56). The weight-adjusted ingots originated in close connection to the old Russian trading centres of Kiev and Novgorod. The considerably older Permian rings have their main concentration in the woods north of Bulghar. They represent a special kind of money, in the shape of jewelry, as a convenient means of payment in big units, and are probably a result of the cosmopolitan trading network of the period and region. As to the weight of the pound there was, according to Nazarenko, an almost worldwide connection. The origins of the pound may be traced back at least to the Roman Empire, and the West European pound was, after the monetary reform by Charlemagne, connected to the weight of the Arabic and Byzantine gold units, the dinar and the nomisma (Nazarenko 1994, 24, 30). The mark-weight in the Scandinavian area (c 200 g) was connected to Frankish weight systems. As pointed out by Brøgger (1921, 102), the Scandinavian, Frankish and Anglo-Saxon weight-systems were all based on old units of c 4 and 8 g. The introduction in Russia of the grivna (c 50 g) was, according to Janin, in order to make it possible to connect with units used in the West European trade



(Nazarenko 1994, 21, with refs; cf. Leimus 2000, 126). Thus there were almost identical weight units in use in both the west and the east, all of them probably ultimately deriving from the Roman pound. On turning to South-West Scandinavia the question of weight-adjusted objects becomes complicated because different weight systems with units of similar weight meet. Most striking is the hoard from Vester Vedsted, not far from Ribe. This hoard consists of gold rings, filigree ornaments of gold, two cut gold rods, silver beads, hacksilver and Arabic coins; its terminus post quem is 913 (Skovmand 1942, 72 et seq). The weights of the rings are: 196, 103.5, 101, 97.5, 49, 27 and 24 g; and the cut rods weigh 51.1 g and 50.0 g respectively (Skovmand 1942, 72; Graham-Campbell 1999, 64). The rings are obviously of South Scandinavian manufacture (cf. Graham-Campbell 1999). The filigree pendants in the same hoard are probably imports and made by a Frankish craftsman (Eilbracht 1999, 103). The Norwegian Hoen hoard also contains weightadjusted gold rings of South Scandinavian manufacture, and likewise it has a Frankish connection. It belongs to the late ninth century, before the great influx of Kufic coins (Graham-Campbell 1999). Skovmand, as well as Brøgger, has pointed out that a large number of gold rings from the Viking Age, and earlier, were made according to fixed weights (Brøgger 1921, 23 et seq, 102 et seq; Skovmand 1942, 72). There seems to have been a long tradition, at least from the Migration period, that gold rings should be weight adjusted, in the Frankish area as well as in Scandinavia. Silver rings of Scandinavian manufacture in the Scandinavian hoards do not show anything like a clear weight grouping, although there does also seem to have been some concern about their weights (Hårdh 1996, 58 et seq). This means that there is an agreement in weights between East and West, making it difficult to claim with certainty, from the weight alone, that an object is of Eastern, Western or Scandinavian manufacture. The reason could be a common origin of the various weight systems in the Roman pound, but there was perhaps also an ambition to synchronise the various systems for the benefit of longdistance trade. During the early Viking Age in Scandinavia there seem to have been, first and foremost, golden objects made according to fixed weights, a phenomenon with Western European traditions. However, there exists also a group of weight-adjusted silver ingots from Gotland, south-western Denmark, Schleswig-Holstein and Holland, dating from the earliest part of the Viking Age (Hårdh 1996, 142, with refs; Wiechmann 1996; Besteman 1999, 257). These ingots will play an important role in future analysis of these weight systems – and in further discussion of the relations between East and West.

RINGS AS MONEY Elsewhere, I have claimed that ingots, as well as ‘Permian’ rings, represent money in large units thereby expressing trade on a large scale, mainly in raw materials; the rings thus fit well into the picture of trade in slaves, furs and other commodities (Hårdh 1996, 179). Hoards consisting of several rings together, for example in the Vjatka area, may be regarded as the result of trading expeditions or of the so-called poludia, that is the special form of tax collection described by Emperor Constantine in De administrando imperii. The rings may have been a reward to some local potentate who administered the collection of



the tax within his territory – and thus a successor to the Persian silver vessels of previous centuries. There are some particular aspects of the ring hoards worth commenting on. The rings do not seem to have circulated to any great extent in either Russia or Scandinavia. It is typical of hoards containing a number of rings that they are usually almost identical with each other, in the nature of their striation, in the form of their knobs and hooks and in their stamped decoration. They would seem to have been made by the same person – or by persons in close contact with one another. Thus they did not function as money to any great extent within the areas where they have been found; rather, they would seem to represent some kind of wealth accumulation. If one accepts the hypothesis put forward by Gurevitj (1979, 77), they might also have been used as prestigious gifts to allies. There are also some variations between regions worth noticing. First of all, there is the obvious variation in weights: the heaviest rings being those found in Russia (c 400 g, and above all c 200 g), whereas there are only some c 200 g rings in the Baltic area, with the majority weighing c 100 g. Stamped ornament occurs on the knobs in Scandinavia as well as in Russia, but stamped decoration on the ring itself occurs only on some rings from Scandinavia and around the Baltic, being unknown on the Russian rings. Also within Russia there are variations, for example between the Vjatka and Perm regions. The rings from Perm seem on average to be heavier than those from Vjatka. This is perceptible in all weight classes, but most clearly, of course, with the rings of c 200 g that are the most abundant (Figure 8.7). Further, there exists an obvious concordance in the details of several rings from Vjatka hoards, for example, the knobs are of medium height, rather broad and unstamped. These observations speak in favour of there having been different workshops, the products of which were deliberately directed towards certain receivers. The rings seem, as mentioned above, to be connected first and foremost with Abbasid dirhams and so belong to a period before the big influx of Samanids to Northern Europe.

Figure 8.7 Distribution of weights of rings weighing c 200 g from Perm and Vjatka.



These observations are true also for the smaller Danish or Gotlandic silver rings with hook clasps. In Denmark, but also on the Baltic islands, hoards with the two types of striated rings form a distinctive group, with special types of objects, ingots and spirals, which does not mix with forms typical of tenth-century mixed hoards. These hoards consist of Arabic coins, hack-silver and ornaments of indisputably Scandinavian manufacture. They are an obvious result of the large import of Arabic coins, mainly Samanid, at the beginning of the tenth century and the coins obviously form the main source of raw material for an extensive indigenous production of typical Scandinavian ornaments, such as twisted rings and filigree work. Returning to the ‘Permian’ rings and their smaller relatives, it can be seen that they were made according to quite other traditions. The striation is sometimes so accurate that the rings look almost machine-made (Figure 8.8). Russian colleagues have suggested that a striated rod was heated and then wound (Fekhner 1967, 56 et seq). This method has, as far as I know, not been used in Scandinavian silver-working during the Viking Age. I think it reasonable to claim that a distinct and independent silver craft was developed in Scandinavia from the beginning to the middle of the tenth century in connection with the rich access to silver. The spiral rings, with their foreign traditions, did not make any impact on Scandinavian craftsmanship and had no later successors. So, I regard them as an imported means of payment, in large units. Thus the light spiral rings from the Danish area might be regarded as the westernmost link in a long chain of contacts, the other end of which is to be sought in the Abbasid caliphate. So, who made them and why? This question is difficult or impossible to answer, but I think it is important to emphasise the monetary character of the rings. They are

Figure 8.8 Part of a ring from the Perm gouvernement (State Hermitage Museum, St Petersburg): note the incised star figure (© Birgitta Hårdh).



executed in a standardised way almost reminding one of coins issued by some authority. Their main concentration is in the forest areas of northern Russia, a main supply area of the desired furs. Moreover, the rings do not seem to have had a monetary function either on the Volga or in Scandinavia. They obviously did not circulate. The distribution of the rings shows first and foremost where desirable goods were to be obtained. Possibly, the standardised execution and, above all, the adjusted weights were of greater concern for those who mediated the rings than for those who received them. Accurate weight systems and octahedral weights were typical of the Arabs. It is not unlikely that the origin of the rings is to be sought considerably further south than their present distribution might indicate. The Russian rings, which do not fit into the strict weight grouping and which also are less accurately made, might then possibly be unauthorised copies. During the eighth and ninth centuries, south-west Scandinavia was an economic focus, with centres at Ribe and Hedeby, in close contact with the prosperous North Sea trade. There are also many indications of links to the East. Doubtless, it is not merely a coincidence that these eastern rings have a concentration in this area. As for the smaller average size and lighter weight of the rings, I am inclined to think that it has something to do with the type of trade carried out from this region.

Note 1. In Russian literature these rings are usually referred to as rings of ‘Glazow-type’, after the district in Vjatka where they are most frequent.

References Arne, T J (1914), La Suède et l’Orient, K W Appelberg, Uppsala Besteman, J C (1999), ‘A Viking Age silver hoard from Westerklief on the former Isle of Wieringen (Province of North Holland) in the light of Viking relations with Frisia’, in H Sarfatij, W J H Verwers and P J Woltering (eds), In Discussion with the Past. Archaeological Studies presented to W. A. van Es, Stichting Promotie Archeologie, Zwolle/Ammersfort, 253–266 Brather, S (1996), ‘Frühmittelalterliche Dirham-Schatzfunde in Europa. Probleme ihrer wirtschaftsgeschichtlichen Interpretation aus archäologischer Perspektive’, Zeitschrift für Archäologie des Mittelalters 23/24 (1995/96), 73–153 Brøgger, A W (1921), Ertog og øre. Den gamle norske vegt (Videnskapsselskapets Skrifter II, Hist-filos Klass 3), Kristiania Callmer, J (1991), ‘Beads as a criterion of shifting trade and exchange connections’, Studien zur Sachsenforschung 7, 25–38 Callmer, J (1994), ‘The clay paw burial rite of the Åland Islands’, Current Swedish Archaeology 2, 13–46 Darkevich, V P (1976), Khudozhestvennye metally Vostoka VII–XIII vv. Proizvedeniya vostochnykh torevtik na territorii evropeiskoy chasti SSSR i Zaural’pa, Moskva Eilbracht, H (1999), Filigran- und Granulationskunst im wikingischen Norden. Untersuchungen zum Transfer frühmittelalterlicher Gold- und Silberschmiedstechniken zwischen dem Kontinent und Nordeuropa (Zeitschrift für Archäologie des Mittelalters 11), Köln Fekhner, M V (1967), ‘Sheynye grivny’, Ocherki po istorii russkoy derevni X–XIII vv, Moskva, 55–87



Franklin, S and Shepard, J (1996), The Emergence of Rus 750–1200, Longman, London and New York Graham-Campbell, J (1999), ‘Rings and things: some observations on the Hon hoard’, in G Fellows-Jensen and N Lund (eds), Berettning fra attende tværfaglige vikingesymposium, Forlaget Hikuin, Højbjerg, 53–64 Gurevitj, A (1979), Feodalismens uppkomst i Västeuropa, Tiden, Borås Hårdh, B (1996), Silver in the Viking Age (Acta Archaeologica Lundensia Series in 8°, no. 25), Almquist & Wiksell International, Stockholm) Haussig, H-W (1987), ‘Die Praxis des Warenaustausches im Warägerhandel mit den chasarischen Märkten Sarkel und Itil’, in K Düwel, H Jankuhn, H Siems and D Timpe (eds), Untersuchungen zur Handel und Verkehr der vor- und frühgeschichtlichen Zeit in Mittel- und Nordeuropa, Göttingen, 528–541 Kotljar, N F (1994), ‘Severorusskye “Chernigovskie” monetnye grivny’, in Drevneyshie gosudarstva vostochnoy Evropy, Moskva, 80–142 Leimus, I (2000), ‘Russkiy denezhny shchët XII v. – svoy ilinzaimstvovannyy?’, in Vosmaya vserossiyskaya numizmaticheskaya konferentsiya, Moskva, 126–127 Lombard, M (1992), Blütezeit des Islam. Eine Wirtschafts- und Kulturgeschichte 8.-11- Jahrhundert, Frankfurt/M Martin, J (1986), Treasure of the Land of Darkness: the Fur Trade and its Significance for Medieval Russia, Cambridge University Press Munksgaard, E (1963), ‘Skattefundet fra Duesminde’, Aarbøger for Nordisk Oldkyndighed og Historie 1962, 94–112 Nazarenko, A V (1994), ‘Proiskhozhdeniye drevnerusskogo denezhno-vesovogo schëta’, in A P Novoseltsev (ed), Drevneyshie gosudarstva vostochnoy Evropy, Moskva, 5–79 Noonan, T S (1981), ‘Ninth-century dirham hoards from European Russia: a preliminary analysis’, in M A S Blackburn and D M Metcalf (eds), Viking-Age Coinage in the Northern Lands (BAR, International Series 122), Oxford, 47–117 Noonan, T (1984), ‘Why dirhams first reached Russia: the role of Arab-Khazar relations in the development of the earliest Islamic trade with Eastern Europe’, Archivum Eurasiae Medii Aevi 4, 151–282 Noonan, T (1986), ‘Why the Vikings first came to Russia’, Jahrbücher für Geschichte Osteuropas 34, 321–348 Noonan, T (1987), ‘The monetary history of Kiev in the pre-Mongol period’, Harvard Ukrainian Studies 11:3/4, 384–409 Skovmand, R (1942), De danske skattefund fra vikingetiden og den ældste middelalder indtil omkring 1150 (Aarbøger for nordisk Oldkyndighed og Historie), København Stenberger, M (1958), Schatzfunde Gotlands der Wikingerzeit,Vol I, Kungl. Vitterhets Historie och Antikvitets Akademien, Stockholm Steuer, H (1978), ‘Geldgeschäfte und Hoheitsrechte im Vergleich zwischen Ostseeländern und islamischer Welt’, Zeitschrift für Archäologie 12, 255–260 Steuer, H (1997), Waagen und Gewichte aus dem mittelalterlichen Schleswig. Funde des 11. bis 13. Jahrhunderts aus Europa als Quellen zur Handels- und Währungsgeschichte, RheinlandVerlag, Köln Tegengren, H (1962), ‘Valrosstanden i världshandeln’, Nordenskiöldsamfundets tidskrift 22, 3–37 Warnke, C (1987), ‘Der Handel mit Wachs zwischen Ost- und Westeuropa im frühen Mittelalter’, in K Düwel, H Jankuhn, H Siems and D Timpe (eds), Untersuchungen zur Handel und Verkehr der vor- und frühgeschichtlichen Zeit in Mittel- und Nordeuropa (Abhandlungen der Academie der Wissenschaften in Göttingen Philologisch-Historische Klasse, Dritte Folge, Nr 156), Vandenhoeck & Ruprecht, Göttingen, 546–569 Werbart, B (1998), ‘Archaeology and cultural identity: Russia – Crimea – Khazars. New archaeological discoveries and theories’, in Studia zdziejów cywilizacji, Warszawa, 241–248


The Form and Structure of Viking-Age Silver Hoards: The Evidence from Ireland John Sheehan

In this chapter some thoughts are offered on those Viking-Age silver hoards which contain non-numismatic material, whether accompanied by coins or not, and on the potential for using the form, composition and structure of such hoards as classificatory criteria. This exercise is largely based on the results of a recent case-study of the hoards of early VikingAge date from Ireland (Sheehan 2001a).1 It is hoped that a classificatory approach of this kind, which need not be particular to the hoards from Ireland and which could be applied to hoards from anywhere within the Viking world, may lead towards a new system of categorising and analysing hoards and that, in turn, this may result in a more developed understanding of both the social and economic implications of the data. The chapter concludes with a brief consideration of the phenomenon of hack-silver in Ireland. Prior to outlining the proposed classificatory scheme, and to applying it within the context of the silver hoards from Ireland, it is necessary to provide a brief introduction to these hoards.

IRELAND’S VIKING-AGE SILVER HOARDS Ireland’s Viking-Age silver hoards, like those from Scandinavia and elsewhere, may be classified along conventional lines into three categories: ‘coinless’ hoards (those consisting exclusively of non-numismatic material, ranging from complete ornaments and/or ingots to hack-silver); ‘mixed’ hoards (those consisting of coins combined with non-numismatic material) and coin hoards (those consisting exclusively of coins). Over 130 hoards are on record from Ireland of which a large proportion, comprising some 110 examples,2 may be assigned to the ninth and tenth centuries and are thus of relevance to the focus of this chapter. Some 53 of these are coinless hoards, 16 fall into the mixed category and the remaining 41 comprise coin hoards. Prior to considering the coinless and mixed categories some details of the coin hoards should be briefly noted. The types of coins that occur in these finds are predominantly Anglo-Saxon, though Arabic and continental issues as well as those of the Viking rulers of East Anglia and Northumbria are also represented. No less than 75 per cent of the total number of ninth- and tenth-century hoards were deposited after c 940. The majority of the finds are small in size and together they represent an insignificant amount of the overall silver wealth of the period in terms of their bullion value, as has been noted by Kenny (1987, 517). In this respect the evidence of the coin hoards serves to underline and 149



emphasise the significance of the hoards containing non-numismatic silver for the study of the early Viking Age in Ireland. Excluding ingots, most of the components of the coinless hoards are diagnostic in form and it is thus possible to suggest regional or cultural attributions for them. When this is done it becomes evident that much of this ornament material is of HibernoScandinavian or Irish Sea origin, with the most characteristic components being arm-rings of Hiberno-Scandinavian workmanship. Several different classes of such have been identified, but in numerical terms by far the most significant of these is the broad-band type (Graham-Campbell 1976, 51–53; Sheehan 1998a, 178–180). The HibernoScandinavian silver-working tradition was centred on Dublin and, to a significantly lesser extent, the Munster towns (Sheehan 1998b, 154–156). In addition to its products, there are also Scandinavian, Baltic, Scotto-Norse and native Irish elements represented in the coinless hoards from Ireland (Sheehan 1998a, 177–194). In most cases these hoards are composed exclusively of complete objects, with hack-silver being present in 19 of the 53 finds of this type. On the basis of the hoard-associated material, combined with the evidence of the related mixed hoards, it is evident that a significant majority of Ireland’s coinless hoards were deposited during the hundred years or so between c 850 and c 950. All 16 examples of early Viking-Age mixed hoards from Ireland were deposited during the tenth century (Sheehan 1998a, Table 6.1).3 Just over half of these contain hacksilver and most feature either ingots or ingot fragments in addition to the coins. These traits serve to distinguish these hoards from the typical types of coinless hoard, both in terms of their date-range and composition. Nevertheless, a few mixed hoards contain ornaments of Hiberno-Scandinavian type, or hack-silver derived from such ornaments, and this serves to link them to the majority of the coinless hoards. It is interesting to note that the depositions of only a dozen of the 69 hoards under discussion, that is of the coinless and mixed varieties combined, are clearly datable to after the mid-point of the tenth century. These finds are mainly composed either of ‘ring-money’ from Scandinavian Scotland or of ingots accompanied by coins and, as such, are not characteristic of Ireland’s ninth- and tenth-century hoards in terms of their composition. (It is also of interest that the most recently discovered mixed hoard from Ireland, that from Dunmore cave, Co Kilkenny, which was deposited in the 970s, is also of unusual composition and thus conforms in this sense to several other later tenth-century finds). It can be strongly argued, on the other hand, that at least 40 of these 69 hoards were deposited in the period before c 950, and the minimal nature of this figure is emphasised by the fact that some of the ingot-only hoards will also have been deposited during this period. It is clear, therefore, that the period encompassing the second half of the ninth and the first half of the tenth century is a critical one for the study of Scandinavian activity in Ireland. This core period is one when Ireland’s hoards were frequently composed of HibernoScandinavian ornaments together with, occasionally, ingots and native Irish material, as well as some imported objects from Scandinavia and the Baltic. The latter types of material, however, appear to be insignificant in terms of their quantity when measured against the total amount of Viking-Age silver on record from Ireland, especially when it is recognised that some of the simpler pieces might well be local copies of the imported objects. These imports, however, should be regarded as being illustrative of a much larger volume of material that was presumably consigned to the melting-pot, along with Arabic, continental and Anglo-Saxon coins, to provide the raw material for the HibernoScandinavian silver-working tradition (Sheehan 2001b, 54–57). It is probable that much



of this non-numismatic material arrived in Ireland from the Baltic region along with Kufic coins. These latter have been recognised in 21 hoards from Ireland and Britain, practically all of which were deposited prior to the 940s (Sheehan 1998a, Table 6.2). It appears, therefore, that the import of Arabic silver was predominantly a phenomenon associated with the core period noted above. As the Viking Age in Ireland progressed, the composition and structure of its silver hoards changed. A gradual but steady transformation is evidenced from the bullion economy of the late ninth- and early tenth-century, as represented by the coinless hoards, towards a more sophisticated economy in which imported coin began to be retained. This transition towards a monetary economy may be represented by the mixed hoards, with their coins and hack-silver. Presumably the retention of coins was for commercial purposes, and it is tempting to associate the mixed hoards with the foundation of the Hiberno-Scandinavian towns in the early decades of the tenth century. From the mid-point of the tenth century onwards the deposition of coinless hoards appears to decline sharply, whereas mixed hoards continue to be deposited; on the other hand, the number of coin hoards rises steadily to predominate. Following the establishment of a mint in Dublin, in c 997, the transformation of Ireland’s early Viking-Age bullion economy is largely complete.

CLASSIFICATION AND CATEGORISATION Silver hoards have frequently been categorised on the basis of whether or not they contain coins. Within such a system, as has been noted above in the context of the evidence from Ireland, hoards were categorised by Graham-Campbell as ‘coin’ hoards, ‘coinless’ hoards and ‘mixed’ hoards (1976, 46–51). One anomaly which may immediately be discerned with this system is that one of its three classes of find – the ‘coinless’ hoard – is defined in terms of the absence of a feature which hoards of this type cannot, by definition, actually exhibit. Another limitation in this approach is that it appears to give precedence to coined silver as a defining component of Viking-Age hoards, despite the fact that this is just one of the several types of component that may potentially occur in such hoards. It would be equally illogical to assign precedence to ingots, hack-silver or ornaments in such a manner, as this involves the conscious imposition of a form of value judgment on the material. Neither should the presence or absence of coins be the principal basis of assigning a hoard a particular economic role or status, as coins were variously regarded as bullion or money at different places and times throughout the Viking Age – just as silver arm-rings will have been viewed variously as ‘currency rings’ or status objects in different economic and social contexts. The most likely reason for this skew towards coins in the context of hoard definition and classification is that until recent decades in Scandinavia, as well as in Ireland and Britain, it was numismatists who carried out most research on Viking-Age hoards. It could thus be argued that the dominant perspective on hoards and on hoard-related issues was for long a numismatic one. Such a perspective, naturally, will define other phenomena in relation to its own concerns. It is the author’s contention that a more accurate impression of the degree of variation in those hoards which contain non-numismatic material, and which together comprise the clear majority of Ireland’s Viking-Age hoards, may be arrived at through the application of more developed classificatory methods and, recently,



he has proposed a re-classification of those Viking-Age hoards from Ireland which contain non-numismatic material (Sheehan 2001a). This scheme focuses primarily on the non-numismatic contents of the hoards because it is felt that such an approach allows the varying degrees of complexity present within the hoards to become more evident. The basis for the new classification is the actual composition of the hoards as reflected through the presence, absence or association of the three non-numismatic phenomena that may form part of Viking-Age hoards, namely ornaments, ingots and hack-silver. The presence or absence of coins is not deemed to be particularly significant in terms of this classificatory scheme, as coins could be variously regarded as bullion, ornaments or money during the Viking Age. Nonetheless, an inbuilt aspect of the system is that the conventional coinless/mixed hoard categorisation may be superimposed upon it. When this is done the results actually emphasise the compositional similarities rather than the differences between the coinless and mixed hoards. This, of course, undermines the value of the traditional categorisation and demonstrates how the numismatically driven approach, being focused on a single element, serves to separate what are clearly related phenomena. It is recognised that there must be a meaning and a purpose, beyond the mere practice, of an empirically defined type of exercise such as that outlined here. And there is, as is indicated by the fact that the five principal hoard categories that emerge from this procedure – when it is applied to the evidence from Ireland – have particular and reasonably discrete distributional patterns (Figure 9.1). This may be taken as an indicator that these groupings are not merely creations of the classificatory process, but that they actually reflect a Viking-Age reality. Furthermore, when the broad contextual information pertaining to these hoards – such as chronology, distribution and find location – is added to their proposed classifications it becomes possible to identify and appreciate their social and economic significance more readily (Sheehan 2004). Thus the proposed system facilitates the examination of the various roles that silver hoards and their components played in both Hiberno-Scandinavian and Irish society.

THE CLASSIFICATORY SCHEME This general system of classification focuses on the compositional structure of those hoards, which contain non-numismatic material. It is based on the occurrence, absence or combination of the three principle non-numismatic object-categories found in hoards, namely ornaments, ingots and hack-silver. A fivefold division of the hoards is advanced and the shared characteristics of each of these classes, whether they contain coins or not, are stressed. The non-numismatic components of Class 1 hoards consist of complete ornaments only; those of Class 2 hoards consist of complete ingots only; those of Class 3 hoards consist of a combination of complete ornaments and ingots only; those of Class 4 hoards consist of complete ornaments and/or ingots in association with hack-silver, while those of Class 5 hoards consist of hack-silver only. The principal classes in numerical terms are Classes 1, 2 and 4 (Figure 9.2), which together account for almost 84 per cent of the total number. The classificatory system is capable of accommodating further levels of refinement for the purpose of more detailed analysis. Thus, for example, a hoard which contains hacksilver (ie a Class 4 or 5 hoard) may be sub-classified into one of 12 permutations and



Figure 9.1 Distribution map of the early Viking-Age hoards from Ireland (Classes 1–5).

combinations on the basis of whether the hack-silver is derived from ornaments, from ingots or from both of these, combined with the form of its other non-numismatic contents, if any (Table 9.1). This number of potential permutations and combinations is doubled when coins are also present in the find. A summary account of the five principal classes of hoards that emerge from this classification, as represented in the hoards from Ireland, is presented below where each



Class 5 12% Class 1 39%

Class 4 29%

Class 3 4%

Class 2 16%

Figure 9.2 Pie-chart showing the relative proportions of hoards of Classes 1–5 amongst the Viking-Age finds from Ireland.

group of hoards is briefly discussed and interpreted. A fuller discussion of the classificatory scheme and its results has been published elsewhere (Sheehan 2001a), and it is intended to publish a definitive account, with complete lists of the hoards and their details, in due course.

Class 1 hoards The non-numismatic element of the characteristic type of Viking-Age hoard from Ireland, Class 1, which accounts for some 39 per cent of the total number, is composed exclusively of complete ornaments. In the majority of cases these ornaments are of arm-ring type and they usually vary in number from only two to four examples per hoard. Typical examples of finds of this type include those from Rooskey, Co Donegal (Raftery 1969), and Rathmooley, Co Tipperary (Sheehan 1992b). Interestingly, only one of the 27 hoards which comprise Class 1 – a poorly recorded find from the west of Co Kilkenny (Lindsay 1842, 125) – contains coins in addition to the ornaments.4 At least 18 of these hoards are composed in whole or in part of arm-rings of Hiberno-Scandinavian type. This would suggest that, in the main, Class 1 hoards belong to the period between c 850 and c 950, the dating brackets for the Hiberno-Scandinavian silver-working tradition. In terms of their distribution these hoards are reasonably well scattered throughout the country (Figure 9.1), though few of them occur in the central midlands/north Leinster region where a very large proportion of the coin hoards is found. The margins of overlap between the respective distributions of the coin hoards and the Class 1 hoards are minimal. Kenny interpreted these two contrasting distribution patterns in economic terms (Kenny 1987, 518). It is probably inappropriate, however, to interpret the Class 1 hoards in this manner. Given their character as finds which consist exclusively of complete ornaments (with the exception of the single example noted in the above paragraph), they are clearly to be distinguished from those hoards which contain hack-silver



and/or ingots (Classes 2–5). The latter should probably be interpreted economically, having the characteristics of silver used as a means of payment in exchange and trade, but ornament-only hoards seem best interpreted in social terms. Coinless hoards of the general type represented by the Class 1 finds from Ireland have been classified as ‘passive’ in commercial terms by Graham-Campbell, who noted that the contents of such hoards were ‘clearly not intended for everyday circulation, having been converted into artefacts that conferred status, whether to patron, donor or recipient’ (Graham-Campbell 1989, 54). More recently Hårdh has interpreted the occurrence of a group of hoards in Western Norway as reflecting ‘a system where silver … had a social function … Here the precious metals are important as a means for giving gifts and forming alliances, to build up social positions in areas with politically unclear conditions’ (Hårdh 1996, 178). Similar interpretations may be applied to the Class 1 hoards from Ireland which may be viewed as archaeological reflections of the various political alliances, such as are testified to in the historical sources, that were formed between the Scandinavians and the Irish from the middle of the ninth century onwards.

Class 2 and 3 hoards The non-numismatic elements of Class 2 and 3 hoards, which for the purposes of this chapter are dealt with together, comprise, respectively, finds containing complete ingots only and complete ingots combined with complete ornaments only. Hoards of these two classes together comprise 14 finds, 20 per cent of the total number under consideration. Eleven of this total are Class 2 hoards and the remainder contains combinations of ingots with ornaments, mostly arm-rings. Most examples of Class 2 finds contain fewer than five ingots, although some are exceptionally large. The latter includes the unusual find from Carrick, Co Westmeath, which comprised 61 large ingots and weighed a total of just under 32 kg (Ryan et al 1984, 335–336). In contrast with the Class 1 finds, a high proportion (43 per cent) of Class 2 and 3 hoards contain coins. With one exception the deposition of all of these hoards dates to the second half of the tenth century, indicating the likelihood of there being a general degree of chronological distinction between the Class 1 and Class 2/3 finds. A distinction between these hoards is also clearly evident when the distributional patterns of their find spots are examined (Figure 9.1). There is no clear convergence between the Class 1 and Class 2/3 finds, as might be expected, but a significant degree of overlap exists between the geographical distribution of the latter finds and of the coin hoards, especially in north Leinster and the midlands region. This suggests that there may have been some degree of correlation between the use of coins and of ingots during the second half of the tenth century, especially in these areas. This is supported by the fact that, as has already been noted, a high proportion of Class 2 and 3 hoards themselves contain coins. Class 2 and 3 hoards share an important characteristic with the Class 1 hoards, of course, in that they lack any hack-silver content. The presence of ingots in them, however, indicates that they should be distinguished from the commercially ‘passive’ ornament hoards as well as from the commercially ‘active’ hoards of Classes 4 and 5 (which both contain hack-silver). Class 2 and 3 hoards may, therefore, be classified as ‘potentially active’ in commercial terms as the silver in them, although possibly converted into ingot form for reasons of commercial convenience (or for the manufacture of ornaments), does not occur in the hack-silver form which is characteristic of truly ‘active’ hoards.



Class 4 hoards Class 4 hoards from Ireland comprise 20 finds, representing 29 per cent of the total number, and are characterised by the presence of hack-silver alongside complete ingots and/or ornaments. The recently discovered hoard from the excavations at Cloghermore cave, Co Kerry (Sheehan 2005), which consists of two ingots and four pieces of ornamentderived hack-silver (Figure 9.3), is an example of this class. Nine potential permutations and combinations exist for the compositional structure of the non-numismatic elements of these hoards, of which seven are actually represented in the find evidence. Thus there is a good deal of variety evident within the composition of Class 4 hoards. The find from Raphoe, Co Donegal, for instance, with its arm-rings, ingots and a solitary and rather large fragment of hack-silver (Graham-Campbell 1988), represents a type of hoard that is clearly at a considerable remove in terms of its composition and nature from the remarkable collection of coins, ingots and highly fragmented ornament- and ingotderived hack-silver contained in the Dysart Island (no. 4) hoard (Ryan et al 1984, 339–356). It is noteworthy that a sub-group, comprising almost one-third of the Class 4 hoards, consists of finds which are composed exclusively of ornaments combined with ornamentderived hack-silver. Examples of these hoards include those from Kilbarry, Co Cork (Sheehan 1998b, 155), and Cushalogurt, Co Mayo (Hall 1973). Such a distinctive composition suggests the likelihood that these finds represent ‘passive’ Class 1 hoards which were rendered commercially ‘active’ in response to some economic exigency. Significantly, as with all but one of the 27 Class 1 finds, each of these hoards is coinless. If it is accepted that these are related to the Class 1 finds in terms of their original function (which, it could be argued, was social rather than economic), then a large proportion of the total number of Viking-Age hoards from Ireland which contain non-numismatic material may be regarded as having been originally social in function and thereby ‘passive’ in commercial terms. Various hypotheses may be advanced to interpret Class 4 hoards, with their varying degrees of commercial complexity. Each of them, however, may be regarded to some Table 9.1 Sub-division of Class 4 and 5 early Viking-Age hoards from Ireland (after Sheehan 2001a, with additions) CLASS


4a 4b 4c 4d 4e 4f 4g 4h 4i 5a 5b 5c

Ornaments with ornament hack-silver Ornaments with ingot hack-silver Ornaments with mixed hack-silver Ingots with ingot hack-silver Ingots with ornament hack-silver Ingots with mixed hack-silver Ornaments and ingots with ornament hack-silver Ornaments and ingots with ingot hack-silver Ornaments and ingots with mixed hack-silver Ornament hack-silver Ingot hack-silver Mixed hack-silver




6 – – 2 2 3 1 – 1 – 1 3

– – 1 – 1 1 1 – 1 1 2 1

6 – 1 2 3 4 2 – 2 1 3 4



Figure 9.3 Viking-Age silver hoard of Class 4, from Cloghermore cave, Co Kerry (University College Cork Audio-Visual Services).

degree as evidence for the use of silver as a means of payment. Under Graham-Campbell’s proposals such hoards should be viewed as ‘active’ in commercial terms, ‘for silver bullion will have been so rendered for commercial purposes and not for reasons of status’ (Graham-Campbell 1989, 55). It is clear, however, that there are varying levels of ‘activity’ discernible in these hoards, as is evidenced by their variety of composition and structure. The fact that coins occur in only four of these 20 hoards is interesting, and may suggest that coined silver did not play an important part in the hack-silver economy. The association of ingots with the coins in each of these four cases may be taken as further evidence for the economic correlation between the uses of coins and ingots, which was noted above in the context of the Class 2 and 3 hoards.

Class 5 hoards The non-numismatic elements of Class 5 hoards consist exclusively of hack-silver. Such finds are of quite uncommon occurrence in Ireland, with only eight of the 69 relevant early Viking-Age finds belonging to this category. Seven of these contain ingot-derived hack-silver, some in association with ornament-derived hack-silver, and four of the total also feature coins in their composition. Examples of Class 5 hoards include that from Kilmacomma, Co Waterford, which consists of 10 ingot fragments, a rod fragment and an arm-ring fragment (Sheehan 1998b, 163; Figure 9.4), and the find provenanced only to ‘Co Antrim’, which is comprised of four arm-ring fragments, two ingot fragments and a single coin (Sheehan 1992a, 52; Graham-Campbell 1993, 80; Figure 9.5).



Figure 9.4 Viking-Age silver hoard of Class 5 from Kilmacomma, Co Waterford (© National Museum of Ireland).

Figure 9.5 Viking-Age silver hoard of Class 5 from ‘Co Antrim’ (© Ulster Museum).



HACK-SILVER AND IRELAND Classic hack-silver hoards are of relatively rare occurrence in Ireland. In fact, it is noteworthy that few of the Class 4 finds may be classified as ‘true’ hack-silver hoards in the sense in which this phenomenon is understood in Scandinavia. Hårdh, for instance, has defined hack-silver hoards as finds ‘where half or more of the objects are fragments, and where most of the objects weigh less than five grams’ (Hårdh 1996, 33). Under these terms many Class 4 hoards from Ireland, including those noted above from Raphoe, Cushalogurt, Kilbarry and Cloghermore, are excluded from the definition. Similarly, some of the Class 5 hack-silver hoards, such as the example provenanced to ‘Co Antrim’ (Figure 9.5), fall outside the terms of Hårdh’s definition. This exclusion is mainly due to the low-weight limitation imposed by Hårdh’s maximum 5 g stipulation. It must be assumed that she arrived at this figure on the basis of the comprehensive degree of hack-silver fragmentation that is evident in some southern Scandinavian hoards. However, these hoards are generally later in date than the hoards under consideration here, and this may mean that the two bodies of evidence are not directly comparable to one other. If the minimum weight limitation in Hårdh’s definition is doubled to 10 g in the case of the hoards from Ireland, then about half of the total of 28 finds that contain hack-silver (Classes 4 and 5 combined) may be regarded as ‘true’ hacksilver hoards. This represents just 20 per cent of the total number of hoards from Ireland that contains non-numismatic material. Hack-silver may be interpreted in terms of ornaments and ingots being absorbed into commercial circulation because the available quantity of coined silver was insufficient to meet the needs of a metal-weight economy (Lundström 1973, 11). The phenomenon is usually interpreted as an intermediary form of development between a bullion and a monetary economy, with the fragmentation of silver indicating that the demand for a means of making small payments was greater than the current supply of imported coins for use as weighed metal (Hårdh 1996, 86). The occurrence of highly fragmented hacksilver may be regarded as the penultimate stage in this process, preceding the introduction of minting (see Hårdh 1996, 84–130, for a full discussion of hack-silver). Unfortunately, however, the evidence from Ireland does not readily conform to this model. First, there is a general lack of hack-silver hoards from amongst the corpus of silver finds and this is especially true of the highly fragmented variety of hack-silver hoard. Second, on the basis of both the chronological and distributional evidence it seems that it may have been the retention and use of imported coins, rather than the process of reducing ornaments to hack-silver, that resulted in the late tenth-century development of indigenous minting in Ireland. Dublin’s mint was established c 997, following a period of about 50 years during which quantities of Anglo-Saxon coins were being retained and hoarded. The clear majority of the coin hoards deposited during this period is located within those parts of the country centered on Dublin, the north Leinster and central midlands regions (Kenny 1987, 511–514). On the other hand, the distributional patterns of those hoards which contain hack-silver as a component (Class 4) and of those composed exclusively of it (Class 5) are largely exclusive to these regions, with only marginal degrees of overlap evident (Figure 9.1). However, the patterns of hack-silver distribution generally conform to those of the Class 1 hoards, suggesting that there may have been some form of relationship between the hack-silver and the ornament hoards rather than between the



hack-silver and the coin hoards. The fact that only nine out of the total of 28 Class 4 and 5 hoards contain coins serves to further distinguish these types of finds from the coin hoards (and, incidentally, serves to distinguish them even further from the hack-silver hoards from Scandinavia which are only rarely found without coins). Thus, it seems unlikely that hack-silver had a significant role to play in the move towards monetarisation that occurred in later tenth-century Ireland. The interpretation of those few hack-silver hoards that are on record from Viking-Age Ireland is problematic. Could their occurrence outside of the Dublin sphere of influence indicate that a hack-silver economy began to develop and operate in parts of Ireland that were peripheral, both geographically and culturally, to the major Hiberno-Scandinavian settlements? The hoards cannot confirm this independently, for it is not possible to ascertain whether the fragmentation of their components took place before or after they came into native ownership. On the other hand, if hack-silver was a phenomenon normally associated with the economically more refined Hiberno-Scandinavian populations of the coastal towns, and with the native Irish in their vicinities, then why is there a virtual absence of it in the Dublin/north Leinster region? Finally, there are questions relating to the low frequency of occurrence of classic, economically ‘active’, Scandinavian-type hoards from Ireland. These are characterised by the presence of highly fragmented hack-silver, ingots, a variety of coins and the absence of complete ornaments. Typically, however, examples of this type from Scandinavia are later in date than the majority of the Irish hoards and, therefore, should not be used directly for comparative purposes. What is remarkable, however, is that a small number of HibernoScandinavian hoards, from both Ireland and Britain, do fall within this category and date to a period well in advance of when such hoards become common in Scandinavia. The classic examples in this regard are the finds from Cuerdale, Lancashire, deposited c 905, with its exceptionally large collection of hack-silver of Hiberno-Scandinavian origin (see GrahamCampbell 1987, 339–340), and of Dysart Island (no. 4), Co Westmeath, deposited c 907. These two classic hack-silver hoards, with their highly fragmented and heavily nicked components, may be interpreted – in accord with the Scandinavian model – as representing an intermediary stage of development between a bullion and a monetary economy. The implication of this is that the longphort settlement at Dublin may actually have been poised to commence minting, following York’s example, during the early tenth century. Indeed, it might even have done so had it not been for the traumatic events of 902, when the combined forces of Leinster and Brega forced an exodus of the Dublin Scandinavians (AU 902).

ACKNOWLEDGEMENTS The author wishes to thank Michael Connolly for granting permission to refer to the hoard from Cloghermore, Denise Maher for preparing Table 9.1 and Maeve Sikora for preparing Figures 9.1 and 9.2.

Notes 1. The ‘early Viking Age’ is defined in this context as comprising the ninth and tenth centuries. 2. This total is comprised of the 108 examples listed in Sheehan 1998a (Appendix 1) along with the recently discovered Class 4 examples from Dunmore cave, Co Kilkenny (Andy



Halpin pers comm, National Museum of Ireland) and Cloghermore cave, Co Kerry (Sheehan 2005). 3. The second hoard from Dunmore cave, Co Kilkenny, should now be added to those listed in Sheehan 1998a (Table 6.1). It was deposited during the 970s (Andy Halpin pers comm, National Museum of Ireland). Some details of the find are contained in the Sunday Times Magazine (16 July 2000), 46–51. 4. It has now been established by the author that this board is, in fact, from Kilkenny West, a barony in Co Westmeath.

References Graham-Campbell, J A (1976), ‘The Viking-Age silver hoards of lreland’, in B Almqvist and D Greene (eds), Proceedings of the Seventh Viking Congress, Dublin 1973, Royal Irish Academy, Dublin, 31–74 Graham-Campbell, J A (1987), ‘Some archaeological reflections on the Cuerdale hoard’, in D M Metcalf (ed), Coinage in Ninth-Century Northumbria: The Tenth Oxford Symposium on Coinage and Monetary History (BAR, British Series 180), Oxford, 320–344 Graham-Campbell, J (1988), ‘A Viking-Age silver hoard from near Raphoe, Co. Donegal’, in P F Wallace and G Mac Niocaill (eds), Keimelia: Studies in Medieval Archaeology and History in Memory of Tom Delaney, Galway University Press, 102–111 Graham-Campbell, J (1989), ‘The coinless hoard’, in H Clarke and E Schia (eds), Coins and Archaeology. Medieval Archaeology Research Group: Proceedings of the First Meeting at Isegran, Norway, 1988 (BAR, International Series 556), Oxford, 53–61 Graham-Campbell, J (1993), ‘A “vital” Yorkshire Viking hoard revisited’, in M Carver (ed), In Search of Cult: Archaeological Investigations in Honour of Philip Rahtz, Boydell Press, Woodbridge, 79–84 Hall, R (1973), ‘A hoard of Viking silver bracelets from Cushalogurt, Co. Mayo’, JRSAI 103, 78–85 Hårdh, B (1996), Silver in the Viking Age: A Regional-Economic Study (Acta Archaeologica Lundensia, Series in 8°, no. 25), Almquist & Wiksell International, Stockholm Kenny, M (1987), ‘The geographical distribution of Irish Viking-age coin hoards’, PRIA 87c, 507–525 Lindsay, J (1842), A View of the Coinage of the Heptarchy, etc, Cork Lundström, L (1973), Bitsilver och Betalningsringar: Studier i Svenska Depåfynd från Vikingatiden påträffade mellan 1900 och 1970, Stockholm Raftery, J (1969), ‘A hoard of Viking silver bracelets from Co. Donegal’, JRSAI 99, 133–136 Ryan, M, Ó Floinn, R, Lowick, N, Kenny, M and Cazalet, P (1984), ‘Six silver finds of the Viking period from the vicinity of Lough Ennell, Co. Westmeath’, Peritia 3, 334–381 Sheehan, J (1992a), ‘Coiled armrings: an Hiberno-Viking silver armring type’, Journal of Irish Archaeology 6 (1991–92), 41–53 Sheehan, J (1992b), ‘The Rathmooley hoard and other finds of Viking-Age silver from Co. Tipperary’, Tipperary Historical Journal, 210–216 Sheehan, J (1998a), ‘Early Viking-Age silver hoards from Ireland and their Scandinavian elements’, in H Clarke, M Ní Mhaonaigh and R Ó Floinn (eds), Scandinavia and Ireland in the Early Viking Age, Four Courts Press, Dublin, 166–202 Sheehan, J (1998b), ‘Viking-Age hoards from Munster: a regional tradition?’, in M A Monk and J Sheehan (eds), Early Medieval Munster: Archaeology, History and Society, Cork University Press, 147–163 Sheehan, J (2001a), ‘Ireland’s early Viking-Age silver hoards: components, structure and classification’, Acta Archaeologica (Supplementa II) 71, 49–63 Sheehan, J (2001b), ‘Ireland’s Viking-Age hoards: sources and contacts’, in A C Larsen (ed), The Vikings in Ireland, Vikingeskibsmuseet, Roskilde, 51–59



Sheehan, J (2004), ‘Social and economic integration in Viking-age Ireland: the evidence of the hoards’, in J Hines, A Lane and M Redknap (eds), Land, Sea and Home (Society of Medieval Archaeology Monograph 20), Maney, Leeds, 177–188 Sheehan, J (2005), ‘The silver hoard’, in M Connolly and F Coyne, Underworld: Death and Burial in Cloghermore Cave, Co. Kerry, Wordwell, Bray, 135–154

– 10 –

Trade and Exchange Across Frontiers Susan E Kruse

This chapter examines exchanges across frontiers during the Viking Age.1 The precise demarcation of a frontier is, however, an issue that will not be addressed here; for present purposes, the movement of goods from one cultural group to another will suffice. This chapter concentrates on the evidence for the way in which transactions occurred, rather than the archaeological or documentary evidence for what was exchanged. It is an issue that inevitably raises more questions than answers. Transactions across frontiers occurred for a number of reasons of which trade was once seen as the only one: witness the number of books and articles with this word in the title. Nowadays, however, the word ‘exchange’ is linked to ‘trade’ in many of these titles, as more attention is paid to the reasons why objects moved from one place to another. Philip Grierson’s important paper of 1959 still warrants going back to, as it clearly outlines a number reasons, apart from trade, why goods could have been exchanged. His attention to economic anthropological approaches has been followed by a number of people. Their results have added greatly to the study of exchanges, even if the issue occasionally gets very muddied indeed. These economic anthropological approaches have shown that even the concept of trade and markets needs careful attention, and a number of authors have usefully summarised different approaches and their relevance to the Viking Age (eg Hårdh 1978; 1996; Gaimster, Chapter 7, in this volume). A recent trend has seen these arguments taken to the extreme, with some authors seeming to argue that a market-trading society, with impersonal exchange, hardly existed at all for Scandinavians, or if it did, was only a feature during the last gasp of the Viking Age (eg Samson 1991a; 1991b). Grierson had perhaps spread the seed of doubt when he noted that, although centres such as Hedeby and Birka had traders, it could not be proven that they were Scandinavian traders (Grierson 1959, 128). More recently, Ross Samson has argued that the Vikings were not merchants, though occasionally the word ‘traders’ creeps back into his discussions (Samson 1991b, 127, 132). Here we enter into backlash economic interpretations: first, the Vikings were only raiders; then they were peaceful traders with a small rogue element and now they are not even really traders. Did foreign trade occur at all? Before providing yet another pendulum swing, it is useful to review the evidence so as to nip this in the bud. The evidence that can be brought to bear on the issue of foreign traders is both documentary and archaeological. 163



DOCUMENTARY EVIDENCE Documentary sources clearly show that objects were exchanged across frontiers for a number of reasons. Grierson assembled a range of documentary evidence describing types of exchange other than trade that can account for the presence of foreign goods and coins. These include gift-exchange, robbery, ransoms, marriages, mercenaries’ wages, tributes, exiles and bribes. A range of objects could be exchanged and, when silver was involved, it was often in large quantities (Grierson 1959). There may have existed implicit rules governing these non-trade exchanges. For example, a gift of a certain type might have required a gift of another type. However, there is little evidence that can be brought to bear in this discussion. One can note, however, that some of the letters between clerics are very explicit as to the type of gift they expect in return (eg EHD, nos 180, 185 and 188), suggesting that they were not leaving the reciprocal gift to chance. In looking for documentary evidence of trade, we must focus on the people the Scandinavians traded with. There is, of course, no contemporary Scandinavian documentary evidence for Viking trade, and recent attempts to project back saga evidence should be used with caution (eg Woolf 1999, with refs). The sorts of exchanges detailed in the sagas, with their element of gift-exchange throughout, may well have existed, but this does not mean that this sort of exchange was the only kind in operation. There is the tangential evidence of Ohthere, recorded by King Alfred in his translation of Orosius’ History of the World (Lund (ed) 1984). When Ohthere (or Ottar), a Norwegian traveller, visited the English court, he was carrying goods and was thus presumably exchanging, but the source is not explicit that this was a ‘neutral’ exchange, even if it is likely to have been so. ‘Neutral’ here means that the person exchanging goods did not expect a similar gift in return, that the exchange was purely an exchange of objects, and thereafter the matter was closed. It is worth noting that even if Ohthere had been intending to sell his goods in England, he had used other forms of exchange during his travels, from his northern home to Kaupang and Hedeby; indeed, many of his goods, he told Alfred, had been obtained as tribute paid by the Finns. Ohthere’s account of his travels is also of interest here for the explanation suggested by Christine Fell as to the reason that he gave for not having entered the territorial waters of the Beormas (in the White Sea area), which was ‘for unfriðe’: All previous translators have offered some variant on the phrase ‘for fear of hostility’. My examination of the evidence [Fell 1983] leads me to the conclusion that to have frið meant to have formal trading rights and agreements, and the absence of frið is more properly construed as the absence of an entry permit, than as enmity. (Fell in Lund (ed) 1984, 63)

Other Anglo-Saxon sources clearly demonstrate that, by the seventh century, within the kingdoms, trade by people outside the social group, crossing frontiers, did occur. The seventh-century laws of Hlothere and Eadric, kings of Kent, deal with traders and other men coming across the frontier; one clause deals with the importance of having witnesses if a man buys property across the frontier in London. The laws of the West Saxon king Ine (688–94) deal with selling countrymen across the seas, and the importance of having witnesses if a trader buys goods in the countryside. The laws of a century later also deal with traders going into the country (EHD, nos 30, 32 and 33). And so it goes from law code to law code. Such legislation occurred because the transactions were neutral, outside



of gift-exchange or other controlling mechanisms, and the king stood as surety (Sawyer 1978, 226). There is also one piece of direct evidence for trade between Scandinavian and English. The treaty drawn up between the Anglo-Saxon king Alfred and the Viking leader Guthrum, sometime between 886 and 890, is very short. The first clause deals with boundaries, the second with weregilds and the third with oaths. The last two clauses, however, concern issues of trade, the fourth dealing with how to ensure that purchases were properly witnessed and the final one stipulating that there should be no movement of people between the two armies, an understandable prohibition between opposing forces. However, the treaty adds that, if such movements occurred for cattle or goods, ‘it is to be permitted on condition that hostages shall be given as a pledge of peace and as evidence so that one may know no fraud is intended’. (EHD, no. 34). This is surely neutral trade – and it is important to note that it occurred at the end of the ninth century, a period when many scholars have argued that little market trade occurred at all. It is also worth noting here that ‘cattle or goods’ were not the only things exchanged between the English and Danelaw Scandinavians. A charter of 926 refers to land in Bedfordshire bought by an English thegn from a Dane for 10 pounds of gold and silver, by order of King Edward the Elder; this has been interpreted as part of a deliberate policy by Edward to regain lands by purchase, before he began his military conquest of the Danelaw (EHD, no. 103; Sawyer 1968, no. 396). Continental sources are no less clear. Indeed, the wealth of documentary evidence that has survived leaves no doubt as to the existence of markets and trade from an early time (Doehaerd 1978; Nelson 1992, 31 et seq). Some were internal, but others must have had foreign traders. Charlemagne and Offa corresponded at the beginning of the Viking Age concerning the protection of their own traders in the other’s lands (EHD, no. 197). Viking-Age documentation also survives from Ireland, but here the laws, which are such a useful source for late Anglo-Saxon England, are less so, due to their already having been written down by the second half of the seventh century (Doherty 1980, 70). Saints’ lives and other sources provide hints of trade and merchants dealing with luxury goods (Doherty 1980, 77–79) but it is the archaeological evidence which fleshes out this picture. Patrick Wallace has usefully listed the range of imported objects found in the Dublin excavations. These include not only complete objects, but also raw materials – such as over 4,000 beads, rings, pendants, waste chips and unworked lumps of amber, most of which surely came from the Baltic (Wallace 1987, 215–216).

THE ARCHAEOLOGICAL RECORD Archaeological evidence of foreign trade is, however, less explicit. Luxury goods are the easiest to identify within it, yet these are precisely the objects that could have been exchanged by other means, such as gifts. Nevertheless, if we look at foreign material found in quantity, where it can be identified, trade will surely have accounted for much of it. It would be difficult, indeed, to think of another explanation to account for much of the Dublin imports of amber, jet, ivory, silver, pottery, glass, soapstone objects, English disc brooches and textiles (Wallace 1987, 215 et seq). There are of course chronological distinctions, and it has sensibly been argued that the Viking Age as a whole was a period of transition – from limited exotic trade to more



intensive market-based transactions (eg Jones 1993, with refs). Different areas will have been at different stages, and the situation was ever changing. We may not be able to chart this in detail in many cases, but the general trend is clear. However, even this brief summary of the evidence shows that a Scandinavian trader would have had to deal with neutral trade at some of his potential markets, even at the beginning of the Viking Age. Thus to deny that market-based neutral exchange occurred in the Viking Age is simply to ignore some of the limited evidence that is at our disposal. Opportunities certainly did exist for those who wished to take them, for moving goods across frontiers in an exchange that was neutral. Such exchanges might have been undertaken by full-time traders, but they need not have been. How does one quantify this type of exchange? Too often finds of coinage or silver alone have been used to assess intensity of exchange. But, as noted above, not all coinage or silver can be viewed as the result of trade. The danegelds brought vast amounts of coin to Scandinavia; the wages of mercenaries in the eastern armies may account for a great deal of Byzantine coins in Scandinavia. The archaeological record is simply too imprecise to provide us with the reasons behind artefact movements. While 4,000 amber objects and pieces of raw material were found in Dublin, what exactly does this represent? A thriving trade or a few boatloads? We simply cannot quantify trade in a useful manner, beyond broad generalisations – and guesses – of no evidence, little evidence and good evidence. It may be accepted therefore that trade occurred, but there is no way in which to determine the extent of exchanges caused by trade alone. The rest of this chapter looks at how trade in neutral exchanges might have taken place. This does not deny the importance of other types of exchange, or assume that neutral exchange was more important. However, reasons such as wages, gifts, tributes, etc, are all of an intensely personal nature, with each party bargaining for its own acceptable level. The danegeld negotiated one year was not necessarily the same as another year. Only with neutral exchange is there a possibility that there may be enough evidence to provide clues to the way in which such exchange operated.

NEUTRAL EXCHANGE Although any group could readily be used as a focus for a consideration of neutral exchange, given that the emphasis is on crossing frontiers, the Vikings are well suited for us to concentrate on. This is because they were highly mobile and they crossed a range of economic situations. The question of logistics – as to just how such transactions occurred – has received little attention. Too often we are shown the picture of the Viking trader, holding his portable scales, with the assumption being that he just got on with it. The situation was surely far more complicated. There were different economic situations in the Viking world (and the situation was changing throughout the period). The first broad distinction to be made is between economies where transactions were based on weights of silver and those where a national coinage was used. In the former any silver was valid, including complete objects, hack-silver, ingots or even slag. In the latter, the national coin, regardless of weight or purity, was guaranteed to have a fixed buying power. Such an economy excluded foreign coin and silver. Even within these categories there are some uncertainties. Were weight standards regularised? Coined-money economies include the highly organised Anglo-Saxon and



Frankish kingdoms, as well as the Arabic empires. But how can we interpret the shortlived and limited circulation of coinages at Hedeby or in the Danelaw? How did bullion weights relate to coinage weights? Crossing both economic situations is barter. This has again received some critical attention from recent scholars, but mainly from those who argue limited neutral exchange took place in the Viking Age (eg Samson 1999, 87). Yet it must have been a factor where impersonal transactions occurred, regardless of economic situations (Blindheim et al 1999, 149). Indeed, barter occurs today in Britain despite the complex market economy we operate in: informal exchanges of goods and services often occur (even legally), and some communities have set up so-called LETS schemes that even formalise this.2

SOME POSSIBLE SCENARIOS Some of the possible scenarios that could have been encountered by a Scandinavian trader, male or female (Stalsberg 1991), during the Viking Age exchanging across frontiers include: 1) A Scandinavian trading with another Scandinavian (eg a man from present-day Denmark exchanging with a man from present-day Norway). This would involve one metal-weight economy to another, but whether based on the same standard is unclear. 2) A Scandinavian trader sailing to Ireland. He could certainly trade in Dublin, which until the very end of the Viking Age was a metal-weight economy. Perhaps he could trade directly with the native Irish, or perhaps such exchanges were filtered from the coastal trading areas. Native documents mention weights of silver in the course of the tenth and eleventh centuries, perhaps only as a common unit of account or perhaps referring to exchanges (Doherty 1980, 82 et seq). There is little evidence of scales and weights outside the trading towns, yet the references to values according to a metal weight suggest that there must have been some way to determine this (Wallace 1987, 213). Analysis of the types of silver hoards found throughout the country has shown that the patterns of economic exchange were quite complex and changed during the Viking Age (Ó Floinn 1998, with refs; Sheehan, Chapter 9, in this volume). 3) En route to Dublin, or perhaps directly, our trader could have visited Scotland or the Isle of Man. The Northern and Western Isles of Scotland were Norse outposts, but thus far no evidence has emerged to suggest that any trading towns were established there. Presumably these Scandinavians were using a metal-weight economy, but to what standard and how often? A few very large hoards have swayed some scholars towards postulating the existence of a flourishing metal-weight economy within Scandinavian Scotland (eg Crawford 1987, 133–134), but I have argued elsewhere that the evidence for this is not clear cut (Kruse 1993, 199–200). The silver hoards date mainly from the tenth and eleventh centuries (Graham-Campbell 1995), and their contents were clearly meant to be weighed. Some weights and a few balances have been found, but these are generally early in date; only one of the truncated-sphere type of weight, which dates from the tenth and eleventh centuries, has been found (Graham-Campbell and Batey 1998, 92, 98–99, 109, 119–20, 139, 140, 163, 227; Barrett et al 1998; Owen and Dalland 1999).3 The Norse settlers were clearly obtaining goods from the Viking world, but how much of this was the result of ‘neutral’ trade is impossible to determine. The economic relationship with the native populations in Scotland and Man is unclear.



4) The situation in Wales is equally vague. As in Ireland, there is evidence for valuation in silver weight, and there are isolated documentary references to internal purchase as distinct from gift-exchange, but very few weights or balances that might have been used (Redknap 2000). Indeed, the documentary evidence suggests that internal exchanges were often a matter of barter, finding an acceptable means of payment. There are isolated references suggesting exchanges with foreigners as well, but the scarcity of documentary and archaeological material permits little detailed discussion as to its nature (Davies 1982, 50 et seq). 5) Our trader might certainly have gone to Anglo-Saxon England or to the Carolingian (and later the Ottonian) empire, wealthy lands perhaps providing better returns for the risks. These areas had coined-money economies, however, and for purchases in silver would not accept, in theory, silver weighed on a balance. The Anglo-Saxon and Frankish laws clearly indicate that only one coinage – the king’s – was to be accepted in the land and, as time went on, this coinage was re-minted at regular intervals, excluding older issues, presumably to allow the king to gain revenue. The fact that so few foreign coins are found suggests that the policy was quite successful (Loyn 1986; Kruse 1988, 285). 6) Once the Danelaw was established, our trader could have traded there, presumably with people who spoke his language and understood his needs. But the economic situation of the Danelaw remains unclear. The Vikings who settled there inherited a land used to dealing in coin, not metal weight. The Danelaw rulers soon began to mint coins, but weights and balances have been found, as well as hack-silver hoards, which suggest that a metal-weight economy functioned to some extent (Kruse 1992; GrahamCampbell 1992; 2002; Blackburn 2001 and Chapter 5, in this volume). 7) Our trader might also have gone to the other Norse settlements in the Faroe Islands, Iceland and Greenland. Coins were not minted there, and there have been relatively few finds of silver hoards (Graham-Campbell 2005). As in Scandinavian Scotland, there were no towns. Unlike Scotland, however, there were no native populations to deal with either (assuming limited contact with the Inuit in Greenland). One would suspect that the new settlers brought with them, and continued to use, the methods of exchange from their homeland. However, the personal links of settlers may have led to exchanges that were entwined with gift-exchange. Certainly the later saga evidence points towards this (Woolf 1999, 67, with refs). 8) And of course, our trader could have gone east as well as west. Trading areas and colonies were situated along the Baltic, although our understanding of the economies of the areas in which they traded is still somewhat vague (Jansson 1987; Callmer 1994). At the furthest end of the route was the Arabic empire which had a vast coinedmoney economy, but also technological capability to produce accurate balances and weights, both of which were copied in Scandinavia (Steuer 1987, 476–477; Gustin 1997). How much of this trade with the Arabic empire was direct, and how much was via middlemen, is impossible to determine (eg Callmer 1994; Noonan 1994). Thus we picture our trader leaving Scandinavia to cross a frontier, carrying some goods to trade. As soon as he left his own kin group and social network he became vulnerable (Lebecq 1999, 235–237). Some have argued that he would have needed a trading contact (Bäck 1997, 152), and this would certainly have smoothed the way. Thus he can be envisaged sailing to the largest farm settlement in a place with dispersed settlements and



with few or no towns, such as Scotland, and forging or renewing contacts with the main landowner. Gift-exchange, bribes, etc, may have formed part of the exchange, but presumably more impersonal trade would have followed. In some areas, such as AngloSaxon England, there were laws for the protection of the foreigner, some of which have been mentioned above. But security may also have been found in trading communities, with other traders. There was safety in numbers, better access to goods and contacts and, for the ruler on whose land it was situated, presumably revenues as well. Christian traders also had the protection of the church to call upon (Lebecq 1999, 237).

WEIGHT STANDARDS If our trader sailed to a land with a metal-weight economy, how would he have operated? He had his balance and weights, but to what weight standard were they manufactured? Indeed, what were the weight standards? There have been a number of attempts, including by myself, to determine the weight standards in the Viking Age (Kruse 1988, with refs; Sperber 1989; Gustin 1997). My own feelings are that the evidence cannot provide the answer: we simply do not have enough weights or payment metal to be statistically valid, the conditions of the objects have almost always deteriorated, and we cannot know what sort of tolerance was accepted. Coins occur in numbers that allow statistical research, but they were not made for the metal-weight economies. Instead, in quantities unknown, they were used along with other bullion. Although a standard somewhere around 25 g – and the word ‘somewhere’ is used here in a broad sense – can be seen in much of the material, it is not possible to determine whether one is dealing with the same weight standard (whatever it is) imprecisely realised, or whether different standards were in operation at the same time. Brøgger (1921) believed there to be two standards in Scandinavia during the Viking Age, an early øre of c 26.5 g and a later one of c 24 g. Other analyses of weights found in Scandinavia have suggested modifications to these figures (see summary in Kruse 1988, 287). Analysis of weights in Dublin, still unfortunately unpublished, led Wallace to suggest a separate Irish unit in the tenth and eleventh centuries of 26.6 g (Wallace 1987, 212–214). My work on silver ingots also suggested something in the mid-20 g range, though it is notable that this was nowhere near as clear when concentrating on a single area like Scotland (Kruse 1993, 193–196). Analysis of other artefact groups, such as the Scottish ‘ring-money’ (Warner 1976), has also, to my mind, been inconclusive, with the exception of the Permian arm rings (Hårdh, Chapter 8, in this volume). However, Permian rings had a limited period of circulation, and their influence on Scandinavian metrological systems remains uncertain. Different types of weights were made during the Viking Age. Brøgger’s early weights are generally lead discs or rectangular in shape, sometimes with markings (Brøgger 1921). Probably dating mainly to the ninth century are weights with a lead base topped with ornamental metalwork. These appear to have been made in Britain and Ireland, but are found also in Scandinavia (Kruse 1992, 81–82). In tenth- and eleventh-century Scandinavia, a popular type of weight was a truncated sphere with an iron core and bronze mantle, a type that was derived from the Arabic world and that was copied in Scandinavia. Also derived from the east were bronze polyhedral weights (Kruse 1992, 79 et seq). From Anglo-Saxon England, where the documentary sources speak of official weights, a few lead weights with coins or coin impressions on top survive, dating from the time of Alfred



onwards; the style was also probably copied by Scandinavians (Archibald 1998; Williams 2000). But most common of all throughout the period, and in different places, are lead weights of various forms that allow no easy dating or provenance. Who made these weights? What standard would they have been measured against? There is evidence for the production of weights of various forms in Birka and Sigtuna, including truncated spheres (Gustin 1997, 170 et seq), and of lead weights in Dublin (Wallace 1987, 212 et seq), but we can only guess the answers to these questions. It is hard to see how a precise standard could have been replicated without any firm political control – nor any checks imposed against those who deliberately made false weights. Yet such firm political control in many of the areas mentioned belongs only to the period at the end of the Viking Age and beyond. Such issues clearly concerned contemporary Anglo-Saxon rulers – but these were men who had a much firmer grasp on economic situations, with control of coins, mints and weights. They had the means to try and impose control; there were standard weights kept in London and Winchester from at least the second half of the ninth century (Sawyer 1978, 212 et seq; EHD, nos 40, 50 and 92). Wallace assumes that similar official weights and even controllers existed in Dublin, even though no evidence survives of them, simply because it is difficult to see how the system was effective without them (Wallace 1987, 214). It is noteworthy that even amongst weights of the same design within weight sets, there is no obvious precise internal metrological system to be determined, and certainly none that correlates precisely with other sets. If there were, we would not still be searching for the metrological grail. A weight set found in a Viking grave at Kiloran Bay on Colonsay, in Scotland, contains a number of lead weights with metalwork caps, together with a balance. Similar weights were found in one or more Viking graves at Islandbridge in Ireland (Graham-Campbell 1980, nos 307 and 308). Yet these weights cannot be reconciled precisely, either internally or with each other. This may be due to their condition, or perhaps they are combinations to be used in different systems or perhaps they were never manufactured with great precision in the first place. Even if a perfectly preserved weight set, with good internal correlations, was to be found, would this be the standard that could be applied to all Viking-Age weights? Or only an example of one person’s attempt – or of one regional system – which when replicated would be distorted again? Similarly, who made the balances? Studies have suggested that some were capable of accurate weighing, as best as can be determined, particularly those in the Arabic tradition (Steuer 1987, 463; Sperber 1988). Unfortunately, the condition of most balances means that it cannot be known whether or not the manufacturers aimed towards the precision they were capable of, or compromised for something that was ‘good enough’. Thus when our trader went on to another area that also had a metal-weight economy, it cannot be assumed that he encountered the same standard. Nor can we know what accuracy would have been expected when weighing silver. Who would have guaranteed the purchase? Who would have declared the silver to be acceptably pure? Who would have trusted the balance of the other? One suggestion is that both traders pulled out their balances and weighed the silver involved in the purchase (Steuer 1987, 479). Clearly the chances of this resulting in the same weight were small indeed. Such discrepancies will therefore have been expected, but were they tolerated? Or maybe an agreed object was



weighed and the resultant discrepancy then formed the basis for the detailed haggling necessary to arrive at an acceptable silver payment. The line between exchange and theft can be a fine one, and a reading of contemporary documents such as the Anglo-Saxon laws, as well as later Norse sagas, shows just how much it exercised people at the time. Neutral trade by definition had no checks built in, and for this reason many of the Anglo-Saxon laws stressed – and kept stressing – the need to have witnesses (eg EHD, nos 30, 32, 33, 39, 35, 41, etc). Perhaps the same was true in metal-weight to metal-weight exchange – though if matters turned ugly, and there were calls of theft or fraud, the trader would be vulnerable. Altogether, these metal-weight to metal-weight transactions would seem to have been a situation fraught with danger. When our trader crossed the frontier to go into a coined-money economy, an altogether different situation occurred. The Anglo-Saxon laws and documentary sources, together with the fruits of numismatic research, provide helpful evidence, albeit onesided. If our trader wanted to buy with silver, he would have been forced to use the king’s coin. Thus anyone wishing to use silver in exchanges would either have possessed the appropriate coin, or would have exchanged a certain amount of bullion and received in return a number of coins. Such a system clearly operated at an early date in both England and the Frankish empire, since few foreign coins or hoards with non-numismatic material are found in the areas over which the king had control. Such a system presumably provided a useful source of revenue to the crown as its ‘fee’ for such conversion. It is not clear that foreigners could exchange their silver directly, but it is not unlikely. This was perhaps one of the duties of the reeve who went to check out any foreign ships (Sawyer 1978, 212, 214 et seq; Loyn 1986). The Anglo-Saxon laws make it clear that the vulnerability of the foreign trader was recognised, and laws were made to try and ensure his safety. All large purchases were to be in towns, where proper witnesses were available. The evidence of the treaty of Alfred and Guthrum, already referred to, provides further evidence of logistics (EHD, no. 34). The charter evidence makes clear that large purchases did occur in Anglo-Saxon England using silver – and perhaps gold as well. Some lands were sold, and the sums involved were large (Campbell 1989). In some cases, as noted above, lands were sold to people across frontiers (EHD, no. 103; Sawyer 1968, no. 396). How would our English thegn have paid his 10 pounds of gold and silver to the Danelaw Scandinavian? Whose tradition took precedence in this case? Indeed, one might well wonder whether all such large transactions, even those occurring within the English frontier, were really paid by counting thousands of coins. The scenarios outlined above imply payment in silver. Yet each situation has its drawbacks. In metal-weight to metal-weight exchange, the different standards for weighing silver provided complications. In the metal-weight to coined-money situation, our Scandinavian trader would lose some of his silver in the conversion. Perhaps that did not matter – it may have been an accepted part of the situation, just as bribes are in some exchanges today. But perhaps here barter had the advantage for some purchases. Barter is supply and demand driven. There is no fixed standard, but simply an agreed upon exchange. One wonders how the Anglo-Saxon kings reacted to barter, since it is far more difficult to get revenue from it. Did the laws requiring that only the king’s coin be accepted mean that silver only in coin form – and no barter – could be used in such exchanges? On the other



hand, it is known that from before the Viking Age kings were getting money from tolls, which meant that any trading ship had already paid for the privilege of exchanging across the frontier (Sawyer 1978, 87, 225–226). There may even have been some accepted barter transactions, for example, for some sorts of goods, another system might be involved. However, here again the sources are silent and one can only speculate. One also has to ask why there are so few weights and balances found in some trading areas even in countries with metal-weight economies. A number have been found in Birka, Hedeby and Dublin (Gustin 1997; Steuer 1987, 469; Wallace 1987, 212), yet very few in Kaupang or York. Is this an accident of survival and excavation – or an indication of lack of use? Hinton, in his discussion of finds from Hamwic, argued that the lack of weights and balances indicated that merchants did not deal there mainly in objects requiring weighing, but instead valued by eye and opinion (Hinton 1996, 60). In this respect, it is useful to note that attention to X-rays of metalwork found during excavations at Ipswich significantly increased the number of balances (Kruse 1992, 69). Metal-detecting finds are also dramatically increasing the numbers of weights. In particular, an unknown site (or combination of several sites) in the northern English Danelaw has produced large numbers of weights, though details of the site(s), types of weights and chronology are still unclear (Simmons & Simmons Gallery, Auction Catalogue, November 1999). It is also noteworthy that Anglo-Saxon sources mention different types of balance (Kruse 1992, 68), yet the archaeological finds are generally confined to small balances that could only have been used to weigh light amounts – presumably precious metal. The archaeological record is clearly only providing part of the evidence. Perhaps some of the confusion is also caused by assuming that weights and balances were used in all forms of exchange in metal-weight economies. They certainly provided a way to assess ones own wealth, for example, to ensure an equal distribution of loot, or to determine that one arm-ring was equal to the value indicated by a certain number of weights on the scale pan. But such considerations belong within ones own economic sphere. It is doubtful if they would have been useful across frontiers, where the different standards and economic situations complicated matters. It is noteworthy that among all the weights found in Dublin, few are of the contemporary truncated sphere type so popular in Scandinavia. There may even have existed very localised situations, where weights and balances were specific to exchanges in certain trading localities. Thus the weights at Birka, with their Arabic influence, may have been primarily designed to weigh out the coveted Arabic silver coins in return for the northern goods. Against this hypothesis, the fact that balances are found in a number of Viking-Age graves throughout Norway suggests more dispersed usage, unless this was the required equipment brought to a market like Kaupang, on yearly visits.

CONCLUSION Although the notion of there having been one common weight standard for Viking-Age Scandinavia, its colonies and its trading outposts is a neat one, it is not really supported by the evidence. Even if it had existed, who could have guaranteed the accuracy of the weights and balances? Weights and balances worked best within ones own economic



milieu, but were less useful when crossing frontiers. The idea of the foreign trader carrying his weights to measure out silver for purchase seems far less likely than trade based on barter, which used the scales and weights for subdivision of goods. Only when the political situation allowed the establishment of weight standards does it seem likely that weighing became common, just as had already occurred in many coined-money economies including Anglo-Saxon England. Gradually, as political control became more centralised in the latter part of the Viking Age, the possibility of regulating standards also occurred. Minting of coins also came gradually and, with that, came currency conversions more familiar to us today. Even then, balances and weights were still needed in exchanges, as shown by the numbers found in medieval Europe (Steuer 1997). The process was not necessarily a one-way development: barter and in-kind payments also continued to be used when necessary, for example during times of silver shortage (Spufford 1988, 96). Indeed, it is likely that during the post-Viking Age, with the increased number of centralised states, more weights and balances would have been used in exchanges than during the Viking Age.

Notes 1. I am very grateful to Professor James Graham-Campbell for nurturing this paper towards publication, pointing out new research since its writing and amending text where necessary. Having myself crossed frontiers to an area with few library resources, it could not have been completed without his assistance and patience. 2. Local Exchange Trading Systems (LETS) are local, non-profit networks for trading goods and services without money. Various such groups operate throughout the UK. 3. For this weight from Cleit, Westray, Orkney, see now Maleszka 2003, although she is incorrect in stating (Maleszka 2003, 287) that no weights of this type have been found in England. There are examples known from York, Yorkshire and Norfolk, as well as several unprovenanced metal-detector finds.

References Archibald, M (1998), ‘Two ninth-century Viking weights found near Kingston, Dorset’, BNJ 68, 11–20 Bäck, M (1997), ‘No island is a society. Regional and interregional interaction in central Sweden during the Viking Age’, in H Andersson, L Ersgaard and P Carelli (eds), Visions of the Past. Trends and Traditions in Swedish Medieval Archaeology (Lund Studies in Medieval Archaeology 19: Riksantikvarieämbetet Arkeologiska Undersökningar Skrifter, Nr 24), Central Board of National Antiquities, Stockholm, 129–161 Barrett, J, Simpson, I and Davis, T (1998), ‘Investigating the Viking Age to medieval transition in Norse Orkney’, Medieval Archaeology Newsletter 18, 8 Blackburn, M A S (2001), ‘Expansion and control: aspects of Anglo-Scandinavian minting south of the Humber’, in J Graham-Campbell, R Hall, J Jesch and D N Parsons (eds), Vikings and the Danelaw. Select Papers from the Proceedings of the Thirteenth Viking Congress, Oxbow Books, Oxford, 125–155 Blindheim, C, Heyerdahl-Larsen, B and Ingstad, A (1999), Kaupang-Funnene Bind II. Gravplassene i Bikjholbergene/Lamøya Undersøkelsene 1950–1957. Del B. Oldsaksformer. Kulturhistorisk tilbakeblikk. Del C. Tekstilene (Norske oldfunn XIX), Oldsaksamlingen, Oslo Brøgger, A W (1921), Ertog og øre. Den gamle norske vegt (Videnskapsselskapets Skrifter II, Hist-filos Klass 3), Kristiania



Callmer, J (1994), ‘Urbanization in Scandinavia and the Baltic Region c AD 700–1100: trading places, centres and early urban sites’, in B Ambrosiani and H Clarke (eds), Developments around the Baltic and the North Sea in the Viking Age (12th Viking Congress: Birka Studies 3), Riksantikvarieämbetet & Statens Historiska Museer, Stockholm, 50–90 Campbell, J (1989), ‘The sale of land and the economies of power in early England: problems and possibilities’, Haskins Society Journal 1, 23–37 Crawford, B E (1987), Scandinavian Scotland, Leicester University Press Davies, W (1982), Wales in the Early Middle Ages, Leicester University Press Doehaerd, R (1978), The Early Middle Ages in the West. Economy and Society, North Holland Publishing Co, Amsterdam, New York and Oxford Doherty, C (1980), ‘Exchange and trade in early medieval Ireland’, JRSAI 110, 67–89 English Historical Documents c. 500–1042 (D Whitelock (ed), 1955; 1968), Eyre & Spottiswoode, London Fell, C E (1983), ‘Unfrid: an approach to a definition’, Saga-Book of the Viking Society 21:1–2 (1982–83), 85–100 Graham-Campbell, J (1980), Viking Artefacts: a Select Catalogue, British Museum, London Graham-Campbell, J (1992), ‘The Cuerdale hoard: comparisons and context’, in J GrahamCampbell (ed), Viking Treasure from the North West. The Cuerdale Hoard in its Context (National Museums and Galleries on Merseyside Occasional Papers, Liverpool Museum No. 5), Liverpool, 107–115 Graham-Campbell, J (1995), The Viking-age Gold and Silver of Scotland (AD 850–1100), National Museums of Scotland, Edinburgh Graham-Campbell, J (2002), ‘The dual economy of the Danelaw. The Howard Linecar Memorial Lecture 2001’, BNJ 71 (2001), 49–59 Graham-Campbell, J (2005), ‘The Viking-age gold and silver of the North Atlantic region’, in S Varge and A Mortensen (eds), Viking and Norse North Atlantic. Select Papers from the Proceedings of the Fourteenth Viking Congress, The Faroese Academy of Sciences, Tórshavn, 125–140 Graham-Campbell, J and Batey, C E (1998), Vikings in Scotland: an archaeological survey, Edinburgh University Press Grierson, P (1959), ‘Commerce in the Dark Ages: a critique of the evidence’, TRHS 9, 123–140 Gustin, I (1997), ‘Islam, merchants, or king? Who was behind the manufacture of Viking Age weights?’, in H Andersson, L Ersgaard and P Carelli (eds), Visions of the Past. Trends and Traditions in Swedish Medieval Archaeology (Lund Studies in Medieval Archaeology 19; Riksantikvarieämbetet Arkeologiska Undersökningar Skrifter, N. 24), Central Board of National Antiquities, Stockholm, 163–177 Hårdh, B (1978), ‘Trade and money in Scandinavia in the Viking Age’, Meddelanden från Lunds Universitets Historiska Museum 1977–78, 157–171 Hårdh, B (1996). Silver in the Viking Age. A Regional-Economic Study (Acta Archaeologica Lundensia Series in 8°, no. 25), Almquist & Wiksell International, Stockholm Hinton, D A (1996), The Gold, Silver and other Non-ferrous Alloy Objects from Hamwic (Southampton Finds 2), Alan Sutton Publishing, Stroud Jansson, I (1987), ‘Communications between Scandinavia and Eastern Europe in the Viking Age’, in K Düwel, H Jankuhn, H Siems and D Timpe (eds), Untersuchungen zu Handel und Verkehr der vor- und frühgeschichtlichen Zeit in Mittel- und Nordeuropa. Teil IV. Der Handel der Karolinger- und Wikingerzeit (Abhandlungen der Academie der Wissenschaften in Göttingen Philologisch-Historische Klasse, Dritte Folge, Nr 156), 773–807 Jones, S R H (1993), ‘Transaction costs, institutional change, and the emergence of a market economy in later Anglo-Saxon England’, EHR 46, 658–678 Kruse, S E (1988), ‘Ingots and weight units in Viking Age silver hoards’, World Archaeology 20, 285–301 Kruse, S E (1992), ‘Late Saxon balances and weights from England’, Med Arch 36, 67–95



Kruse, S E (1993), ‘Silver storage and circulation in Viking-Age Scotland: the evidence of silver ingots’, in C E Batey, J Jesch and C D Morris (eds), The Viking Age in Caithness, Orkney and the North Atlantic (11th Viking Congress), Edinburgh University Press, 187–203 Lebecq, S (1999), ‘Long distance merchants and the forms of their ventures at the time of the Dorestad heyday’, in H Sarfatij, W J H Verwers and P J Woltering (eds), In Discussion with the Past: Archaeological Studies Presented to W A van Es, Stichting Promotie Archeologie, Zwolle, 233–238 Loyn, H (1986), ‘Progress in Anglo-Saxon monetary history’, in M A S Blackburn (ed), AngloSaxon Monetary History: Essays in Memory of Michael Dolley, Leicester University Press, 1–10 Lund, N (ed) (1984), Two Voyagers at the Court of King Alfred: the ventures of Ohthere and Wulfstan, together with the description of Northern Europe from the Old English Orosius, William Sessions Ltd, York Maleszka, M (2003), ‘A Viking Age weight from Cleit, Westray, Orkney’, PSAS 133, 283–291 Nelson, J (1992), Charles the Bald, Longman, London and New York Noonan, T S (1994), ‘The Vikings in the East: coins and commerce’, in B Ambrosiani and H Clarke (eds), Developments around the Baltic and the North Sea in the Viking Age (12th Viking Congress: Birka Studies 3), Riksantikvarieämbetet & Statens Historiska Museer, Stockholm, 215–235 Ó Floinn, R (1998), ‘The archaeology of the early Viking Age in Ireland’, in H B Clarke, M Ní Mhaonaigh and R Ó Floinn (eds), Ireland and Scandinavia in the Early Viking Age, Four Courts Press, Dublin, 131–165 Owen, O and Dalland, M (1999), Scar: a Viking Boat Burial on Sanday, Orkney, Phatassie Redknap, M (2000), Vikings in Wales: An Archaeological Quest, National Museums and Galleries of Wales, Cardiff Samson, R (1991a), ‘Economic anthropology and Vikings’, in R Samson (ed), Social Approaches to Viking Studies, Cruithne Press, Glasgow, 87–96 Samson, R (1991b), ‘Fighting with silver: rethinking trading, raiding, and hoarding’, in R. Samson (ed), Social Approaches to Viking Studies, Cruithne Press, Glasgow, 123–133 Samson, R (1999), ‘Illusory emporia and mad economic theories’, in M Anderton (ed), AngloSaxon Trading Centres: Beyond the Emporia, Cruithne Press, Glasgow, 76–90 Sawyer, P H (1968), Anglo-Saxon Charters: An Annotated List and Bibliography, Royal Historical Society, London Sawyer, P H (1978), From Roman Britain to Norman England, Methuen & Co, London Sperber, E (1988), ‘How accurate was Viking Age weighing in Sweden?’, Fornvännen 83, 157–166 Sperber, E (1989), ‘The weights found at the Viking Age site of Paviken. A metrological study’, Fornvännen 84, 129–134 Spufford, P (1988), Money and its Use in Medieval Europe, Cambridge University Press Stalsberg, A (1991), ‘Women as actors in North European Viking Age trade’, in R Samson (ed), Social Approaches to Viking Studies, Cruithne Press, Glasgow, 75–83 Steuer, H (1987), ‘Gewichtsgeldwirtschaften im frühgeschichtlichen Europa – Feinwaagen und Gewichte als Quellen zur Währungsgeschichte’, in K Düwel, H Jankuhn, H Siems and D Timpe (eds), Untersuchungen zu Handel und Verkehr der vor- und frühgeschichtlichen Zeit in Mittel- und Nordeuropa. Teil IV. Der Handel der Karolingerund Wikingerzeit (Abhandlungen der Academie der Wissenschaften in Göttingen Philologisch-Historische Klasse, Dritte Folge, Nr 156), Vandenhoeck & Ruprecht, Göttingen, 405–527 Steuer, H (1997), Waagen und Gewichte aus dem mittelalterlichen Schleswig. Funde des 11. Bis 13. Jahrhunderts aus Europa als Quellen zur Handels- und Währungsgeschichte, Rheinland-Verlag, Köln Wallace, P F (1987), ‘The economy and commerce of Viking Age Dublin’, in K Düwel, H Jankuhn, H Siems and D Timpe (eds), Untersuchungen zu Handel und Verkehr der vor-



und frühgeschichtlichen Zeit in Mittel- und Nordeuropa. Teil IV. Der Handel der Karolingerund Wikingerzeit (Abhandlungen der Academie der Wissenschaften in Göttingen Philologisch-Historische Klasse, Dritte Folge, Nr 156), Vandenhoeck & Ruprecht, Göttingen, 200–245 Warner, R (1976), ‘Scottish silver arm-rings: an analysis of weights’, PSAS 107 (1975–76), 136–143 Williams, G (1999), ‘Anglo-Saxon and Viking coin weights’, BNJ 69, 18–36 Woolf, A (1999), ‘The Russes, the Byzantines and Middle-Saxon emporia’, in M Anderton (ed), Anglo-Saxon Trading Centres: Beyond the Emporia, Cruithne Press, Glasgow, 63–75

– 11 –

Kingship, Christianity and Coinage: Monetary and Political Perspectives On Silver Economy in the Viking Age Gareth Williams

The period AD c 800–c 1050, which can conveniently be labelled as the Viking Age, was a period of major economic change in Scandinavia and, to varying degrees, in a number of the places occupied by Scandinavians during this period. These changes were manifest in a number of different ways. A key factor in the Viking expansion was the development and expansion of long-distance trade, and trade-routes stretched from the Black Sea and the Caspian Sea in the east to the new settlements in the Atlantic in the west. Expansion in the volume of trade led to the development of permanent markets, and thus to a form of urbanisation, although opinion remains divided as to precisely what constituted a ‘town’ in the Viking Age (Clarke and Ambrosiani 1995; Clarke 1998; Johanek 1999; Andersson 2003). Closely linked with both trade and urbanisation was the development of manufactured goods on a commercial scale, although this could take place in both urban and non-urban settings. The same period also saw a shift to production on a commercial scale in areas such as agriculture and fishing. Finally, and perhaps most importantly from a monetary perspective, the use of coinage became relatively widespread throughout the Viking world. It is important to note from the outset, however, that there was no single pattern of economic development across the Viking world. Both the pace and the nature of change varied considerably in different areas. Scandinavian settlers in England and the Frankish kingdoms found themselves part of societies which were in some ways more advanced economically (and certainly monetarily) than their homelands, and quickly adapted to fit in with those societies. By contrast, settlers in Iceland and Greenland, perhaps because of their remoteness, seem to some extent to have been left behind by developments within the Scandinavian homelands. Urbanisation is one of the defining features of Scandinavian settlement in Ireland, but no corresponding urban settlement has yet been discovered in Scandinavian Scotland. While Denmark, Norway and Sweden all experimented with issuing regal coinage shortly before the year 1000, the coinage developed in very different ways in all three countries over the next 50 years: Denmark started strongly, and maintained the development; Norway started weakly and only really began to develop coinage on any sort of scale in the mid-eleventh century; while the Swedish coinage began strongly but was already collapsing when the Norwegian coinage began to take off. Thus any study of Viking-Age economy must reflect the existence at 177



any given point of a variety of internal economic frontiers within the Viking world, as well as the changes which took place within this period (see Kruse Chapter 10, in this volume).

MONETARY AND NON-MONETARY ECONOMIES Any attempt at a comprehensive overview of Viking-Age economy would therefore be an extremely substantial work, and well beyond the scope of the present volume, still less this chapter. Instead, the focus is on that part of the economy which was based on precious metals. In fact, this is a much broader area than might immediately be apparent since, as several of the chapters in this volume explore, there was not a single ‘precious-metal’ economy, but a number of different and interlocking precious-metal economies. A distinction is often drawn between a ‘status’ economy on the one hand and a ‘monetary’ economy on the other. In the first, precious metals functioned as a means to express wealth and power, both through the conspicuous display of wealth and through the use of giftgiving (particularly the ring-giving found in Old Norse and Old English literature) to create social ties between the giver and receiver of gifts (Gaimster 1991; Hauken 1991; Samson 1991a; 1991b; 1999; Vestergaard 1991). Within this status economy, precious metal might take virtually any form, as long as it was seen to be desirable. Jewellery, ornamented weapons and other display items provided an obvious focus for this sort of status-based economy, but possession of precious metals in any form could provide status, since coins, ingots or hack-silver could easily be melted down to provide the material for the creation of new status items. Discussion of monetary economy in the Viking Age, by contrast, has often been narrowly focused on coinage, with a ‘bullion’ economy based on ingots and hack-silver seen as a kind of half-way step towards a fully monetary economy. Whether or not one can accept this view depends on how one interprets the concept of money, and the second step in this development is better defined as a ‘coin-based’ economy. From a purely economic perspective, the core functions of money must be that it provides both a medium of exchange and a medium for the storage of wealth. In some ways, these functions could be better provided in the form of a regulated coinage, in which the value of an individual coin could be immediately recognised and guaranteed. However, precious metal could function perfectly well both for exchange and for storage of wealth without the need for regulated coinage, given that those involved in exchange had the means to regulate both the quality and quantity of metal involved in any transaction. Quality of metal could be established through methods of testing such as nicking, pecking and bending (see Archibald, 1990 and this volume; Kilger 2006), to see whether the items concerned were consistent with the feel of ‘good’ metal. The very fact that such test marks appear so often on coins of the period clearly demonstrates that without complete faith in the integrity of regulated coinage, coins were not perceived as offering an advantage in this respect. This is demonstrated by an account in the saga compilation Morkinskinna that refers to the coinage of Harald harðráði in the mid-eleventh century. This relates that Harald’s coinage was mostly of copper, and at its best only half silver, so that the silver did not look fine. One of Harald’s warband, the Icelander Halldór Snorrason, refused to accept payment in this coinage (although others did so) on the grounds that he had never been treacherous in his service to Harald, so Harald had no business being treacherous in the matter of his payment. Harald was eventually persuaded to pay him 12 aurar (ounces) of



fine silver. This makes the point clearly that fine silver measured by weight (whether or not it came in the form of coins) was preferable to payment in coin if the coinage was not perceived as good. Although Morkinskinna was only written down much later, Kolbjørn Skaare (1976, 9–10) argues that the story is likely to preserve a genuinely early tradition, given that metallurgical testing demonstrates that Harald’s coinage was as base as the saga suggests, whereas the Norwegian coinage had been of good silver for a century or more by the time that the story came to be written down. Although coinage could provide standardisation, quantity of metal could be established simply by weight, and the large number of surviving weights and balances attest to this. Just as we have no real certainty as to what constituted a satisfactory ‘peck’ when testing a coin, there is considerable uncertainty concerning weight standards in the Viking Age (see Nightingale 1983; 1984; Coupland 1985–86; Kruse 1988; 1993; and this volume; Sperber 1996; Williams 1996, 207–214; 1999). Arguments may be made for separate heavy and light ‘ounces’ of 26.6 and 24 g respectively, or for a single imprecise ounce of c 25 g with a substantial margin of error on either side, or for a number of slightly different standards, each specific to a particular place and time, or for a lack of any real standardisation. To a great extent, from a monetary perspective, this does not matter. As long as both parties in a transaction were satisfied that an adequate quantity of metal of suitable quality had changed hands, complete standardisation was beside the point. Furthermore, it is unclear whether coins in fact offered any real advantage in this respect either. The Anglo-Saxon coinage of the early eleventh century is generally recognised as one of the strongest and most stable coinages of northern Europe in this period, but while some moneyers showed an astonishing degree of consistency in the weight of their coins, there are also instances where coins struck from the same dies vary in weight by 30 per cent or more. Although in some cases this may be because the dies used spanned a period in which a deliberate reduction in weight took place, in others the poor standard of striking suggests that coins were simply being produced somewhat carelessly. Like ingots and hack-silver, coins would need to checked carefully before being accepted in payment, and were probably similarly appreciated more for their intrinsic value as silver bullion than for any sense of their nominal value as coins, except in a state that was able to enforce acceptance of a face value which was significantly greater than the precious-metal value of the coins, which rarely seems to have been the case. Thus, on this interpretation of money, the ‘bullion’ economy which already existed in Scandinavia well before the beginning of the Viking Age was just as much a ‘monetary’ economy as the ‘coin-based’ economies found elsewhere. It is also worth noting that a bullion economy was likewise compatible with long distance exchange and monetary circulation. This is exemplified by the Cuerdale hoard which, in addition to small numbers of Byzantine and Danish coins, larger numbers of Islamic and Carolingian coins and a considerable number of Anglo-Saxon coins from south of the Danelaw, as well as coinage from both the northern and southern Danelaw, contained hack-silver and ornaments of Baltic, Scandinavian and Frankish origins as well as material from Britain and Ireland (Graham-Campbell 1992). It is also important to note that although some recent research has queried the extent to which one can talk about ‘trade’ in the early Viking Age, preferring to see exchange in this period almost exclusively in the light of social interaction (Samson 1999), this view has not met with widespread acceptance, and it is difficult to interpret fragmentary hacksilver, a common element in both hoards and productive sites, as having any economic



role which is not monetary (see Kruse, Chapter 10, in this volume), or at least on the same monetary level as coinage. Philip Grierson (1959) has pointed out that coinage could be used in a wide variety of contexts, including gift-exchange and the status economy as well as supporting trade, and in these respects coinage is not radically different from other forms of precious metal. More recently Peter Sawyer (1990) has questioned whether trade alone can account for the huge influx of foreign coins into Scandinavia, both from east and west, comparing the volume of imported coinage with the limited resources available for export in Viking-Age Scandinavia. This again raises the question of the viability of a rigid distinction between coinage having a ‘monetary’ function and not silver and gold in other forms.

NON-MONETARY FUNCTIONS OF COINAGE Coinage could, however, fulfil other non-monetary functions which ingots and hack-silver could not. Coins provided a means for rulers to exploit the economy in a number of ways. By insisting on the use of coinage for certain functions, and insisting that old or foreign coinage, like bullion, could only be exchanged at a discount, rulers could create income both for themselves and for the moneyers who issued the coins, as seems to have been the case in late Anglo-Saxon England. In societies in which coins were widely used, coinage also provided a useful medium for mass messages, whether political or religious. Since coinage was closely linked with kingship in this period, simply issuing coins at all could be seen as an expression of the legitimacy of an individual ruler’s kingship. A striking example of this can be seen in the coinage of the Viking leader Guthrum in East Anglia. Documentary sources record that under the terms of the treaty of Wedmore, Guthrum was baptised with the name Athelstan, with Alfred of Wessex as his godfather, and was then accepted by Alfred as king of East Anglia. Although most of Guthrum’s coins have blundered inscriptions, it is clear that he chose to use the title Athelstan Rex on his coins, while the design was copied from contemporary coins of Alfred (Blackburn 2001, 128). Thus Guthrum, by issuing coins of this type, was effectively proclaiming himself to be a Christian king who followed Anglo-Saxon regal practices. The relationship between coinage, kingship and Christianity will be discussed in more detail below, but it is worth noting at this stage that, with the exception of the very rare coinage in the name of ‘Sihtric Comes’, none of the coins issued by Vikings in Britain, Ireland or Scandinavia was issued in the name of a ruler other than a king, and that no coins, apart from anonymous imitations of Arabic coins, definitely carried exclusively non-Christian symbols or messages (although some of the images on the Danish coins of the ninth century may perhaps be interpreted as pagan symbols: see Malmer, Chapter 2, in this volume). Thus, while Thor’s hammer did appear on coins of the Danelaw, it was always coupled with explicitly Christian inscriptions and/or symbols (see below). The situation was slightly different in Normandy (see Moesgaard, Chapter 6, in this volume), where the Viking rulers did strike coins in their own name, first as counts of Rouen, and later dukes of Normandy, replacing the Frankish regal coinage which had been issued in the area prior to the grant of Rouen to the Viking Chieftain Rollo by Charles the Simple, c 911/12. No coins were issued in the name of Rollo himself although, as Moesgaard discusses, various imitations of Carolingian types may date from the early period of Viking rule in the area, and coins were certainly issued in the name of Rollo’s



son and successor, William Longsword (c 927–42) and subsequently in the name of Richard I (942/45–96). However, although issuing coins was technically a royal right in Carolingian Francia, by the early tenth century that right had largely devolved to a number of nobles and churchmen (Grierson and Blackburn 1986). The extension of the same right to the Viking rulers of Normandy, who held their lands by virtue of grants from the king, rather than directly by right of conquest, merely reflected Frankish practice at the time, although William was one of the first of the local rulers to place his own name on his coinage rather than just an initial, or keeping the royal name (Dumas 1971, 70; Moesgaard, Chapter 6, in this volume). The Norman coinage can thus be seen as devolution of the royal prerogative, rather than a departure from it. None of this excludes coinage having monetary functions as well as political or religious ones, and some coinages seem to have developed at least in part through commercial considerations. However, it is important to consider the introduction of coinage against a broader background of social and political change, rather than simply in economic terms if one is to understand how it fits in with other types of precious-metal economy in the Viking Age. Before moving on to look in more detail at the changes which took place, it is also important to note that two types of precious metal were in use during this period, and that both could function in a status economy, a bullion economy or a coinbased economy. As the title of the volume suggests, the precious-metal economy was heavily dominated by silver. However, as Mark Blackburn discusses elsewhere in this volume (see also Jørgensen and Vang Petersen 1998), gold was also important, although much rarer than silver. In addition to gold jewellery and ornaments, Blackburn cites the use of gold ingots and hack-gold, and notes that the Franks, Anglo-Saxons and Arabs all had gold coinage with which the Vikings presumably came into contact. Within the Anglo-Saxon and Frankish economies, gold and silver coinage seems to have fulfilled different functions, with the use of gold coins largely restricted to certain types of payments and transactions, generally linked with status or ceremony. A similar distinction is suggested for other forms of economy by the fact that gold and silver are seldom mixed in hoards. Silver hoards are comparatively common, gold hoards are rarer, but mixed hoards are rarer still. The distinction between gold and silver thus acts as a multiplier for the basic distinction between the three types of economy. Within the status economy, one can also add the category of gilt bronze, as a kind of poor man’s gold, for jewellery, ornamented weapons, etc. Gilding or silver-plating would obviously have less scope in either bullion- or coin-based economies, in both of which it would presumably have been seen as an abuse, if detected.

CO-EXISTENCE OF DIFFERENT TYPES OF ECONOMY Having established a distinction between the status-based economy, the bullion-based economy and the coin-based economy, it is important to note that these three types of economy were not necessarily mutually exclusive. Thus items of jewellery, with a clear role in the status-based economy could also serve as high-value monetary units within a bullion economy, if they were produced to set weight standards. This appears to have been the case with the so-called Permian rings from the eastern Baltic, and with various types of silver ‘ring-money’ from Denmark and Sweden (see Hårdh 1996; and Chapter 8, in this volume). A similar argument has been made for ‘ring-money’ in the Scandinavian



settlements in Scotland, where it has been proposed that arm-rings produced to an approximate weight-standard of 24 g may even have had a formal currency role (Warner 1974; Crawford 1987), although this view has been challenged by Susan Kruse (1993) on the basis that the variations on this 24 g ‘standard’ are so great as to call the standard itself into question. Birgitta Hårdh (Chapter 8, in this volume) also notes that gold rings appear to have been produced to approximate weight standards by several of the Germanic peoples well before the Viking Age, and James Graham-Campbell (1999) has noted that some of the gold rings from the Hoen hoard appear to be weight-adjusted. The interesting question with such bullion jewellery is whether it was intended to function separately within a status economy and a bullion economy, or whether it represents a ‘bullionisation’ of the status economy, in which the importance of the display of wealth was qualified by precise amount of precious metal involved. A partial modern analogy might be not simply wearing a designer watch, but making sure that everybody knows how much it cost. This interpretation is supported by Ibn Fadlan’s account of trading with the Rus, in which he tells us that when a Rus trader had acquired 10,000 dirhams through trade with the Arabs, he would have the coins melted down and turned into a neck-ring for his wife. Once he had acquired another 10,000 coins, he would have another neck-ring made (Simonsen 1981; Gaimster 1991). Such a neck-ring would have been incredibly heavy, and it is likely that Ibn Fadlan substantially exaggerates the number of coins involved, but this story highlights the different roles that coins could play in different types of economy. For the Arabs trading with the Rus, trade was a monetary transaction in which they purchased trade goods with coins. For the Rus, the trade involved exchanging their own goods for silver coins which could be used as raw material for jewellery, and the huge quantities of silver dirhams found in Scandinavia, compared with the relatively small number of other Islamic trade goods, suggests that silver as a raw material was the main purpose of trade in the east (Duczko 1998; Noonan 1999). While this may still be regarded as trade, it is not a monetary transaction. Ibn Fadlan’s account is interesting since the weight standards of the Permian rings are consistent as multiples of silver dirhams (albeit on a much smaller scale than Ibn Fadlan suggests), while the polyhedral terminals on the rings are reminiscent of Arabic polyhedral weights, which themselves became widely copied in Viking-Age Scandinavia. Furthermore, the word grivna, which became an important weight unit for silver bullion in medieval Russia, originally meant neck-ring (Gaimster 1991; Kruse, Chapter 10, in this volume. For discussion of the polyhedral weights, see Sperber 1996; Blackburn 2002). Coupland (1985–86) argues that Anglo-Saxon and Frankish coins were also melted down for silver within the Scandinavian homelands in the early Viking Age, which suggests a similar transition between different types of economy, although it is less clear that this had a direct effect on weight standards as the Islamic coinage in Russia. Another interesting example of interplay between different types of silver economy, which Graham-Campbell has drawn attention to (1982; Chapter 12, in this volume) can be seen in the account in Heimskringla of the presentation of a brooch to the skald Eyvind skáldaspillir: Eyvind wrote a poem about all the people of Iceland, and they rewarded him thus, that each farmer gave him a ‘tribute penny’. That was the weight of three pennies of silver, and white to the cut … A brooch was made from it … [which] weighed fifty marks. They sent it to Eyvind, but Eyvind had the brooch chopped up, and bought cattle with it.



This episode shows many features of silver currency in the Viking Age. Coins are used as units of weight for a payment. The silver is then turned into a high-status object, much too heavy to wear, and is then cut up into hack-silver for use as money. The phrase ‘white to the cut’ indicates testing by pecking. Thus pennies, valued by weight and silver quality rather than face-value, were converted into a brooch as a status symbol, which was then broken up to be used as a means of exchange again (Williams 1996, 206). Coins could also be transformed into jewellery directly by mounting them as brooches or pendants, whereby the coins lost their monetary status and became ornaments, despite retaining the physical form of coins. This was not uncommon in Viking-Age Scandinavia, as exemplified by the coin-pendants found in graves at Birka (see below, and Malmer, Chapter 2, in this volume) or the gold coins used to make up a necklace in the famous Hoen hoard (Skaare 1966; 1988; Graham-Campbell 1999). However, the reuse of coins as jewellery can also be observed in this period even in well-established coin-based economies, such as late Anglo-Saxon England (Williams 2001b). Equally, both coins and status items could be treated as bullion. This is apparent where coins with test-marks appear in mixed hoards, a common feature of Viking hoards of the early tenth century in England, and surviving rather later in the Scandinavian homelands (Archibald 1990; and Chapter 4, in this volume; Kilger 2003; 2006). It is also apparent in the inclusion of jewellery (whether intact or in the form of hack-silver) in mixed hoards. As John Sheehan has noted (2000; and Chapter 9, in this volume) with reference to the Viking-Age hoards from Ireland, different combinations of jewellery and bullion in hoards suggest very different economic purposes for hoarding, and the tendency towards particular hoard types in different areas suggest the dominant economic pattern in each area (see also Hårdh 1996, for regional patterns of hoard types within Viking Scandinavia; and Graham-Campbell 1989, for the ‘coinless hoard’ in general). With reference to the use of jewellery as a resource for bullion, it is also worth noting the splendid gold neck-ring from Tissø in Denmark (Graham-Campbell 1980, 60–61, 235; 1999; Jørgensen and Vang Petersen 1998, 300–302). With a diameter of over 30 cm and a weight of 1.83 kg (originally around 2.1 kg), its size and weight make it a colossal status item, especially in gold, but some of the gold strands have been unwound from one side of it, presumably for use as bullion (although possibly to provide material for another item of jewellery). Nevertheless, enough of the neck-ring survives intact that it remains an important status item. Fragments of similar twisted gold jewellery have been found at Torksey in Lincolnshire (Blackburn 2002; and Chapter 5, in this volume), and it is tempting to see this as evidence for the re-use of such material in a bullion economy. However, GrahamCampbell (2001a, 55–56) has also raised the question of whether finds of scrap gold and silver necessarily represent monetary bullion, rather than the raw materials used by jewellers to produce new items of jewellery. The same point can be made about ingots. Where such items occur in coin-producing areas, one may also question whether status items, bullion or imported coins may actually have been intended as the raw material for local coinage. The versatility of the raw material means that finds of precious metal have to be interpreted cautiously, against the background of the society within which they occurred. The three types of economy reflect different degrees of economic complexity in society, and the adoption of a monetary economy reflects changes in society beyond the purely economic sphere. Status economies could survive in societies with little formal



political organisation, and little in the way of organised trade, as long as there was some access to precious metals. Some transfer of wealth must have taken place, through social structures such as ring-giving, and perhaps through forms of gift-exchange, but a statusbased society could have a fairly static economy. Bullion economies likewise require little political authority, but (unless bullion is seen solely as a raw material for the creation of status items) the widespread use of a medium for exchange and storage of wealth implies a greater degree of circulation and transfer of wealth. This is especially true when bullion circulates in small units, as in the case of some hack-silver fragments, as this implies that it is used in transactions of fairly low value. Finally, an economy based on a controlled coinage (rather than simply using imported coin within a bullion economy) implies both political authority and control, as well as considerable circulation and transfer of wealth. Thus the shift towards issuing coins tends to coincide with the emergence of concrete states (however short-lived), and is often only one of a number of manifestations of increased royal power. Despite this, it would be a mistake to assume that a shift towards coin-based economies was inevitable or, for the majority of people engaged in monetary transactions, necessarily advantageous. A willingness to use coinage was important to those who wished to engage in trade with people from areas or states with well-developed monetary economies, and this could often lead to the production of imitative coinage for local use by those who were already familiar with foreign coin through overseas trade. There is a general tendency in monetary history for the introduction of coinage to any given area to imitate foreign coinage already accepted in that area (for numerous examples of this in the Middle Ages, see Grierson and Blackburn 1986; Grierson 1991), and only to develop independent designs as a secondary phase, and this is certainly true for the Viking world. The earlier phases of Danish coinage of the early ninth century imitated Carolingian models (see Malmer, Chapter 2, in this volume), the earlier coin types of the southern Danelaw imitated Anglo-Saxon models from East Anglia and Wessex (Blackburn 2001), while the coinage of the 990s in Dublin, Denmark, Norway and Sweden all imitated the English coinage of ÆthelrÌd II (Dolley 1966; Malmer 1997). Issuing coinage within a controlled economy could also produce financial benefits for those who controlled the economy, as discussed above (although the requirement to use a regulated coinage at a premium would be an additional cost for people actually using the coins), and this was no doubt an important factor in the decision by individual rulers to adopt coinage of their own. Nevertheless, many peoples familiar with coins chose not to use them. Although imported coins are known from Wales in the Viking Age (Dykes 1976; Boon 1986; Redknap 2000, 61–64; Besly 2006), coinage was first issued there by Anglo-Norman rulers and, with the exception of the Hiberno-Norse coinage, the same is true in Ireland, while it was only in the twelfth century that coinage was first issued in Scotland (Bateson 1997), although Anglo-Saxon (and other) coins were known long before in all these areas. Similarly, the introduction of locally produced coinage in Viking-Age Scandinavia does not simply reflect the availability of imported foreign coinage to serve as models. This suggests that the decision to introduce coinage in all these areas was motivated by something more than the supposedly self-evident benefits of coinage to facilitate trade and other monetary transactions. Clear models were certainly available close to hand. As Michael Metcalf argues elsewhere in this volume (Chapter 1), there was a healthy coin-based economy across



much of Anglo-Saxon England in the eighth century. The scale of production of the early pennies misleadingly known as sceattas was unmatched until the end of the Anglo-Saxon period, and the large number of dies known in the coinage of Offa suggests that a substantial coinage continued in the late eighth century (Blunt 1961; Chick 1997; Williams 2001a). On the continent a substantial coinage was also developed in the late eighth and early ninth centuries under Charlemagne, and continued and expanded by his successors. Unlike later Viking raids, which aimed more at conquest and settlement, the early raids focused rather on portable wealth. Coins in quantity were an immediately divisible form of portable wealth, and Metcalf (in Chapter 1) is probably right to suggest that the Anglo-Saxon coins themselves, not simply the wealth of the society which produced them, may have been a major target for Viking raids, although comparatively few early coins have survived from Scandinavia, suggesting that they were probably melted down for reuse in other media. Within the Carolingian empire, the apparent shift around the beginning of the Viking Age from an offensive to a defensive military stance has also been interpreted as a reflection of a shift from the Franks raiding others for portable wealth, to the Carolingian empire becoming a major source of portable wealth for external raiders (Reuter 1985; 1990).

THE RIBE CONTROVERSY AND THE NORTH SEA MONETARY ZONE IN THE EIGHTH CENTURY Contacts with western Europe (and thereby its coinage) were by no means exclusively hostile, and it was probably at least partly through more peaceful contact that the use of coins as coins, rather than as items of bullion, first developed in southern Scandinavia. The earliest example of this can be found even before the Viking Age, at Ribe in western Denmark. The issue of coin use at Ribe has been, and remains, a controversial one, as reflected by some of the views expressed in this symposium (see the opening chapters by Malmer, Metcalf and Wiechmann, in this volume, with their references to earlier debate on the subject), and I will briefly digress from the main theme of this chapter to address the issue. Large quantities of early pennies (sceattas) have been found in excavations in Ribe, of which a significant proportion are of a type (Series X) known as the ‘Wodan-Monster’ type, together with the so-called Porcupines which are generally now accepted as mostly being of Frisian origin, although some ‘Porcupines’ are still believed to be of Anglo-Saxon origin. The relationship between Anglo-Saxon and Frisian coinage in this period is complex, with widespread imitation of types, and although concentration in the distribution of particular types in some cases allows us to identify the area in which those types were produced, the distribution of many different types does not seem to have been limited by national boundaries, as demonstrated by the evidence of both hoards and single-finds (Metcalf 1993–94). The North Sea was an important route for transportation, not simply a barrier, and recent work has tended to emphasise the importance of contacts and influence around the North Sea in the centuries leading up to the Viking Age (eg Carver (ed) 1992; Kramer et al 2000). Monetarily, it seems possible to think of a North Sea economic zone in the early to mideighth century, with its heart in the coin producing areas of eastern England and Frisia, but extending up the east coast of Scotland as far as the Dornoch Firth, and up the coast of Jutland as far as Ribe (Blackburn 1999; Heidinga 1999).



In this context, the opposed views on the attribution of Series X are perhaps less far apart than the protagonists have acknowledged. Michael Metcalf has argued, with some justice, that the distribution of Series X, although it includes England and Frisia as well as Jutland, points to Series X having been issued at Ribe. Brita Malmer, by contrast, has argued that since Series X has so much in common with the other Frisian coin types that it must be considered Frisian, and has further argued that labelling Series X as ‘Danish’ is misleading, since she does not accept the possibility of a coin-issuing Danish state so early. Whether or not one accepts the possibility of anything which can be considered a Danish state in Jutland in the early eighth century, the most recent datings for the earliest phases of the Danevirke certainly suggest the existence of some sort of centralised power in southern Jutland in this period (Anderssen 1998, 9–11; for broader discussion of the formation of the Danish state, see Lund 1995; Axboe 1999; Näsman 1999; Olsen 1999; Skovgaard-Petersen 2003). The debate, ultimately, hinges on the fact that both Metcalf and Malmer see Ribe as a Danish town, and the plausibility or not of the attribution of Series X to Ribe has become almost secondary to the question of whether or not Series X is to be considered a ‘Danish’ type. A less ‘nation-based’ approach removes the problem. Ribe, then as now, was part of Jutland, and the line of the Danevirke strongly suggests that the border of whatever kingdom it formed part of was well to the south. Equally clearly, the town, or proto-town, could not have developed without the encouragement of the local ruler. To that extent, Ribe is therefore Danish, and the archaeological evidence suggests that a sizeable proportion of the population was probably reasonably local. However, as Ole CrumlinPedersen has pointed out (pers comm), from a maritime perspective the west coast of Jutland is simply an extension of the Frisian coastline, while the archaeological evidence (Jensen 1991; Welch 2000) indicates that the function of Ribe was to act as an interface between that part of western Jutland and the North Sea economic zone (effectively forming the north-eastern tip of that zone). As a trading centre and a centre for specialised manufacture, it fits a broader pattern of wics in both England and Frisia. If one interprets Ribe as a wic under the patronage of a local Danish ruler rather than simply as a ‘Danish town’, it is possible to reconcile Metcalf’s and Malmer’s positions, since this would allow Series X to be both typically Frisian and minted in Ribe. This would apply whether the coins were issued by Frisians in Ribe under local patronage, or by locals wishing to conform to the monetary practices of their neighbours and trading partners. This debate is somewhat incidental to the main theme of the chapter, but the significance here of the use of coinage in Ribe in the eighth century (whether or not it was locally produced) is that it shows both contact with the monetary practices of western Europe, and a willingness to assimilate those practices, although this was very much on a localised basis, since coins of this period are comparatively rare elsewhere in Scandinavia, and even in Denmark. Unfortunately, the absence of any historical documentation makes it impossible to tell whether this was purely driven by economic factors or whether there might have been political/cultural factors as well. It is striking, however, that despite the ‘Wodan-Monster’ name given to the type by earlier numismatists, the facing bust on Series X has two equal eyes (rather than the characteristic one seeing and one blind eye associated with Odin) and a cross to either side. There is thus nothing obviously to associate the iconography with Odin, while the Christian iconography is consistent with my broader pattern of linking coinage with a particular concept of Christian kingship.



DENMARK IN THE NINTH AND TENTH CENTURIES Iconography aside, a similar pattern of function and influence can be seen with the Danish coinage of the ninth and tenth centuries (discussed in more detail by both Brita Malmer and Ralf Wiechmann, in this volume), and for these coins there is at least a little more in the way of historical context. The Danish coinage began in the early ninth century, c 825 according to Malmer, but was not remotely a national coinage at this stage as circulation was largely limited to two trading centres and their hinterlands. The early coin types derived from Carolingian prototypes and from the eighth-century Wodan-Monster designs mentioned earlier. As mentioned above, this pattern of introducing a new coinage by imitating the designs of accepted coins from elsewhere is typical for the adoption of coinage both in early medieval Europe and more generally. The distribution of the ninth-century coins suggests two centres of coin production, with more or less mutually exclusive areas of circulation. Metcalf (1996a; 1996b) has identified the two centres of production as Hedeby and Ribe, and although Malmer (Chapter 2, in this volume) prefers to avoid such a definite attribution, her discussion of the distribution of both types is consistent with Metcalf’s interpretation. If the solution suggested above to the question of whether or not the earlier Wodan-Monster coins can be attributed to Ribe is accepted, this would also tend to support a Ribe attribution for the later coins of the same design. Since Hedeby and Ribe were the two main trading centres in Denmark in the ninth century, and since both the size and weight of the coins, as well as some of the designs, points to Carolingian influence, one could argue that these coins had a purely commercial function, and that coinage was first adopted in Denmark to facilitate trade with traders from overseas. Interestingly, coins of both the Hedeby and Ribe types have been found at another important early trading centre at Birka in Sweden. While it is certainly possible that they made their way there through trade (although they could also have arrived there as souvenirs, or loot or through gift-exchange), almost all of the early Danish coins from Birka are grave finds, and have been modified to turn them into pendants, indicating that they have moved from having a monetary function to some other function, whether as jewellery within a status economy, or possibly as some kind of amulet. Given that Birka, like Hedeby and Ribe, had trading links with western Europe, one might reasonably suppose that, if the benefits of coinage for a purely commercial function were self-evident, coinage would have been adopted there as well, but this seems fairly clearly not to have been the case. This is typical of a broader pattern in Scandinavia in the earlier Viking Age, with over 20 Carolingian coins and at least 15 early Anglo-Saxon coins converted into pendants in Denmark, Norway and Sweden (Coupland 1985–86, 20). The coinage should, however, be seen in a broader political context. We know from the Frankish Annals that Hedeby was founded by Godfred king of the Danes forcibly transplanting Slavic traders from Reric to the new settlement. Hedeby can thus be seen to be a settlement very much under royal control, which is consistent with a wider picture of royal control of trade and coastal emporia in the early Middle Ages (Kruse, Chapter 10, in this volume). It is also consistent with what else is known of Godfred from the Frankish Annals, in which he appears as a ruler powerful enough to defy Charlemagne, and in control of considerable resources. It is recorded that in 810 a Danish fleet of 200 ships attacked Frisia, and imposed a tribute of 100 pounds of silver. Such a tribute would probably have been paid largely in coin, and a large tribute in coins from Frisia would have helped to increase familiarity with Carolingian coinage, in addition to any contacts



through trade. The Carolingian types which provided a model for the Danish coinage were slightly earlier, predating the reforms of c 794, but could equally have been the product of either raiding or trading. If Malmer’s dating of the Danish coins to c 825 is correct, Godfred himself was already dead before their introduction, although if Blackburn’s suggestion (1985–86) that Malmer’s KG 2 is Scandinavian rather than Frisian is correct, then the earliest coinage could have been firmly in Godfred’s reign, with subsequent types produced under his successors, while Kirsten Bendixen (1981, 407), who also attributed the Hedeby coins to Godfred, suggested that he may have issued coinage for the same reasons that he took control of the merchants from Reric, to exploit trade to generate revenue for himself, which might have taken the form of customs duties, commercial taxes and, arguably, payments for minting rights. In any case, Godfred left a legacy of strong royal control, although this may have dissipated to some extent after his death, as a number of rivals fought for control of the Danish kingdom, but it is likely that his successors continued to wield considerable power within those areas which they controlled, and it is against the background of that rivalry that the coinage developed. Certainly Frankish sources continue to note the ability of Danish kings to raise substantial fleets in this period, and Danish rulers also seem to have managed to maintain some authority in southern Norway, despite attempts by local rulers to assert their independence (Lund 1995; Wamers 1995; 2002). Equally, Godfred left a legacy of raiding in Frisia which, alongside trade, characterised the relationship between Denmark and Frisia in the first half of the ninth century, with both plunder and trade probably providing sources for Carolingian coins in Scandinavia in the ninth and tenth centuries (Coupland 1985–86). Nelson (1997) notes that there was a substantial increase in the production of coinage at Dorestad in the early ninth century, of which the Danish raiders were probably the main beneficiaries. However, Coupland (1988) notes that the numismatic evidence suggests that Dorestad remained prosperous until around 840, with a gradual decline thereafter, and a sharp downturn in the 850s. For Dorestad’s prosperity to survive through repeated raiding in the 830s, it must have retained an important function as a trading centre (hence its attractiveness to raiders), and it is likely that part of that trade was with Scandinavia. Furthermore, the succession conflict following Godfred’s death was not a wholly internal matter. Godfred was initially succeeded by his nephew Hemming, but Hemming only survived briefly, and Denmark was then fought over by Godfred’s sons and a rival pretender called Harald Klak and his brothers. Harald commended himself to Louis the Pious in 814, receiving military support in return and eventually managed to establish himself as co-ruler in Denmark with two of Godfred’s sons in 819 (Coupland 1998, 88–89). The sources do not give us the details of this arrangement, but it seems likely that Harald ruled in one area, and Godfred’s sons in another, and this may have a direct bearing on the clearly distinct circulation areas of the different coin types. In addition to political and military support from Louis, Harald’s claim to the Danish kingship provided an opening for Frankish bishops to plan for the Christianisation of Denmark, and first Archbishop Ebbo of Rheims and Bishop Ansgar of Hamburg-Bremen were active in Denmark in the 820s. Harald himself was baptised in great splendour at Ingelsheim in 826, with Louis the Pious as his godfather (Lund 1995; Malmer 1966, 219; Nelson 1997, 22–23; Wamers 2002) and it is possible to interpret elements in the design of some coins of this period in the context of Harald’s baptism and the Christian mission (Malmer 2002; 2004; and Chapter 2, in this volume). Although the early Danish coins are anonymous,



one may perhaps draw a parallel with the later Viking ruler Guthrum, as mentioned above. Guthrum, after his defeat by Alfred, accepted baptism and settled down to behave in the manner expected of a Christian king, including striking coins based on Alfred’s own. Harald may have done much the same, although his rule in Denmark was short-lived and he was driven out in 827, taking refuge in the county of Rüstringen in northern Frisia, which had been granted to him as a benefice by Louis, possibly as a precaution against circumstances of that kind (Coupland 1998, 90). Throughout this period, both Harald and his rivals seem to have sought validation for their rule through recognition and support from the Carolingian monarchs, and this is likely to have been reflected in a willingness to be seen to adopt some elements of Frankish kingly practice, as well as in toleration of, if not active support for, Christianity (Lund 1995; Skovgaard-Petersen 2003, 173–174; Sawyer and Sawyer 2003, 147–149). The fact that a series of Carolingian rulers provided support and refuge to members of the Danish royal house in the ninth century (Coupland 1998, 70), may well in part reflect a conscious desire to bring Denmark under Frankish hegemony, as well as to curb Danish piracy in Frisia and up the Frankish river systems. This would certainly be consistent with the suggestion of the adoption of elements of Frankish kingly practice in Denmark in the course of the century. Following Harald’s expulsion in 827, Denmark seems to have fallen under the sole rule of Godfred’s son Horek, and while apparently never Christian himself, he was at least willing to engage in diplomacy with the Franks and to permit Ansgar to build a church in Hedeby. He may thus also have seen some benefit to his relations with Louis in behaving as he felt was expected of him as a king. However, in the latter part of his reign there seems to have been some sort of pagan reaction, and Horik himself led a raid on Ansgar’s church at Hamburg in 845 (Nelson 1997, 23). How far this was motivated by religion, how far it was simply a desire to raid, and how far Horik’s relationship with the Church reflect his relationship with the Frankish monarchy is difficult to be certain, but it does seem that the collapse of the Christian mission coincides with a lengthy gap in the coinage, since on Malmer’s dating her type KG 6 probably ceased to be produced in the mid-ninth century, while KG 7 was not produced before c 920. This gap also coincides with a period in which a Swedish royal dynasty, apparently without strong links either to the Frankish monarchy or the Church, established itself in southern Denmark, probably with their centre of power in the area around Hedeby (Lund 1980; Moltke 1985; Lund 1995, 211; SkovgaardPetersen 2003, 174). The revived coinage was still anonymous, but was now dominated by Christian symbolism, and survived until the introduction of an overtly regal coinage in the late 990s, with KG 8 and KG 9 falling comfortably into the reign of Harald Bluetooth (Malmer 1966, 209–226), who died c 986. Harald, as his memorial stone for his parents at Jelling declares, ruled Denmark and part of Norway, and made the Danes Christian (Skovgaard-Petersen 2003, 168), while the construction of the Trelleborg fortresses and the bridge at Ravning Enge show him commanding very considerable resources later in his reign, which may be interpreted as an attempt at centralised royal control of military organisation along Anglo-Saxon or Frankish lines (for discussion of this interpretation, see Lund 1995; 1996; Williams 2002, and forthcoming b). Since both raiding and trading continued throughout this period, as the more ‘economic’ stimuli to the use of coinage in southern Scandinavia, it seems likely that the decision by Danish kings whether or not to issue coins in the ninth and tenth centuries was at least partly motivated by a desire to conform to a broader European pattern of kingship. Similarly, the fact that the Danes issued coins at all during the ninth



and tenth centuries whereas the Norwegians and Swedes did not, despite having comparable trading centres to interface with western European traders (and with Carolingian coin finds recorded in both countries: Coupland 1985–86) appears to reflect greater Frankish cultural influence, probably based on geographical proximity, which led to a variety of contacts, peaceful and otherwise. This said, it is important to remember that the adoption of coinage in Denmark in the tenth century had only a limited effect. Its use was largely restricted to the regions around the two issuing centres, with no evidence to suggest widespread use elsewhere, and therefore contrasts very markedly with the impression of a growing ‘national’ kingship suggested by the fortification and building work. It is likely that this is because it was only ever intended for use in those areas, rather than as a planned national coinage, which failed to materialise, but that limited purpose itself contrasts strikingly with the ambition displayed in the inscription at Jelling and in the scale of Harald’s construction projects, although Coupland (1985–86, 22–23) suggests that expansion in the coinage from the 970s led to a wider circulation, so that coins were ‘an acceptable means of exchange in most if not all of the country. The adoption of local coinage did not lead to the total exclusion of imported coinage, although it was largely successful in excluding imported coinage from a small area around Hedeby (Wiechmann 1996; and Chapter 3, in this volume), while the issue, and to some extent even the use, of local coinage throughout the ninth and tenth centuries was intermittent. It is also relevant to note that, both in Denmark itself and more widely in Scandinavia, coin finds from the ninth and tenth centuries are dominated by Islamic dirhams and their Khazar, Bulghar and possibly Russian imitations, rather than by native Danish coins or their Frankish models (Hovén 1981; Kromann 1990; Jensen and Kromann 1998; Talvio 1998). Many of the dirhams are found in fragmentary form, and neither the weight nor the design had a great influence on Scandinavian coinage, despite the apparent availability of these coins as a model for the native coinage. This also contrasts strongly with the evidence for Byzantine and Islamic influence on weights and weight-standards in Viking-Age Scandinavia (Sperber 1996; Duczko 1998; Mikkelsen 1998; Hårdh, Chapter 8, in this volume). Furthermore, graffiti on a number of Islamic coins, in the form of Thor’s hammers and Christian crosses, suggests both awareness and a rejection of Islam (Mikkelsen 1998), although Duczko (1998) suggests that other graffiti on Islamic coins may indicate possible conversion to Islam. The Islamic coins functioned as bullion, while it would appear that Danish rulers consciously decided to imitate Carolingian models of coinage in the face of the obvious influence of Islamic coins, but on a comparatively small scale. This clearly demonstrates both the limitations of the power of the Danish kings, assuming that the interpretation of coinage as an expression of a particular model of kingship is correct, and that the benefits of coinage were rather less clear at the time than what later historians and economists, who have seen the adoption of coinage in terms of ‘progress’, might believe. While the Viking Age certainly was a period of economic transition, it was neither particularly quick nor smooth.

NORWAY This is apparent also with the adoption of regal coinage in Denmark, Norway and Sweden in the last decade of the tenth century. All three kingdoms introduced explicitly regal coinage around the same time, based on the CRVX coinage of Ethelred II of England, and



begun with the help of Anglo-Saxon moneyers, although production seems relatively quickly to have shifted to local workmen (Skaare 1976; Jensen 1995; Malmer 1997). Both Svein Forkbeard of Denmark and Óláf Tryggvason of Norway had direct familiarity with English ways, having raided extensively in England, and Óláf had certainly undergone conversion and baptism, providing another example of a newly Christianised Scandinavian king settling down to rule a kingdom, and attempting to rule in the manner of an Anglo-Saxon or Frankish king (Sawyer 1999). However, neither the coinage nor Óláf’s rule were particularly successful. This must be seen against the background of weak central kingship in Norway in the tenth century. Although historians long accepted the saga traditions of a powerful unified kingship under Harald hárfagri in the late ninth and early tenth centuries, which provided a firm precedent for his descendants in the late tenth and early eleventh centuries, recent thinking has questioned the extent of the authority of Harald and his sons Eric and Hákon, and has argued that the saga tradition that the claims of Óláf Tryggvason (995–1000) and Óláf Haraldsson (1015–28) to kingship over a united Norway was based on descent, and therefore inheritance from Harald is spurious (Krag 1989; 2003, 185–191; Helle 1993). Harald’s direct authority was largely confined to western Norway, as was that of his sons, and such nominal overlordship as they were able to claim in other areas was almost certainly heavily dependant on alliance with powerful local chieftains, such as the earls of Hlaðir and Mœre, rather than absolute authority (Krag 1990; Williams 2001c). Harald’s son Hákon, who was brought up as a Christian at the court of Athelstan, seems to have attempted to introduce a more Christian ‘Anglo-Saxon’ style of kingship, but he was unable to persuade the population as a whole to adopt Christianity and many of his reforms probably did not last. Notably, he seems never to have got as far as issuing coins, although it would have been consistent with what else we know of his style of kingship to have done so (Skaare 1976, 26–29, 58; Sawyer 1999; Williams 2001c), and the sons of Eirík who succeeded him seem to have struggled to exert much authority in the face of Danish territorial ambitions. Óláf Tryggvason faced similar problems, both in his inability to oppose his more powerful Danish neighbour successfully, and in the unwillingness of pagan chieftains to submit to his aggressive attempts to promote himself as a Christianising king of a more or less united Norway (Krag 2003, 191–193). Against this background, and on the basis of the surviving evidence, the coinage was on so small a scale that Skaare (1976, 60) comments that it ‘looks more like a short-lived experiment than a well-established institution’. Nevertheless, his ambitions to be seen as king of ‘Norway’ are clearly demonstrated by his use of that title on the coins. Óláf himself was killed after a reign of only five years, and Norway fell under the rule of the earls of Hlaðir, under the overlordship of Svein of Denmark. The Hlaðir earls did not issue coins, either because of their subordinate status, or their paganism, but perhaps also because Norway still lacked urban centres of any size as a focus for the use of coinage. Norway had a second attempt to introduce coinage under Óláf Haraldsson who, like the earlier Óláf, had been exposed to the Anglo-Saxon model of kingship and had adopted Christianity before seizing power in Norway (Sawyer 1999, 99). Although he survived longer as king than Óláf Tryggvason, he was ultimately similarly unsuccessful, and was first driven out of Norway in 1028 and subsequently killed when he attempted to regain his kingdom in 1030. Krag (2003, 195) suggests that resistance to Óláf ‘may have been caused by his earlier policy of establishing a more “modern” and European type of kingdom’. Like the earlier Óláf, his coinage appears only to have been on an extremely



limited scale, despite a considerably longer reign, and is unlikely to have circulated widely, suggesting that the act of issuing coinage was more important than its monetary function, and like the earlier Óláf he used coins to promote himself as king of Norway. The coins are also explicitly Christian, copying not just Anglo-Saxon regal busts, but crosses and the dove, symbolising the Holy Spirit. Following Óláf’s enforced departure into exile in 1028, there was a brief period of direct Danish rule, in contrast to the earlier periods of Danish overlordship in which the earls of Hlaðir had exercised the real authority in Norway. Cnut installed his young son Svein as king, with his English mother, Ælfgifu, as regent. This period saw a substantial enlargement of royal authority, remembered as tyrannical in later Norwegian tradition, and probably reflecting Ælfgifu’s Anglo-Saxon influence (Sawyer 1999, 99). However, it is striking that no coins appear to have been issued in Norway although Cnut had an extensive coinage in both England and Denmark. However, this may reflect Svein’s status as, effectively, a sub-king to his father rather than a fully independent king. The expansion of royal authority under Svein led to conflict with the Norwegian chieftains, as it had for Óláf, and he was unable to maintain his position in Norway. He was succeeded by Óláf’s son Magnús, who also managed to take control of parts of Denmark in the latter part of his reign. It is surprising in some ways that Magnús appears not to have issued coinage in Norway although he was a Christian king like his father, and certainly issued coins in Denmark (Becker 1981; Hauberg 1900; Jensen 1995). A possible explanation for this, if later saga tradition is to be believed, may have been the fate of his father. One of the factors which led to Óláf’s expulsion appears to have been his desire to expand royal authority in Norway, both geographically, and in terms of the rights and powers associated with kingship, such that he could be seen (or at least portrayed) by his political opponents as a tyrant. Magnús, according to the sagas, raised fears early in his reign that he might be equally tyrannical (and notably did not repeal all of Svein’s innovations), and was persuaded to moderate his style of kingship to fit what the Norwegian chieftains felt to be acceptable, subsequently becoming such a popular ruler that he earned the nickname Magnús the Good. Although there is no suggestion that it was issuing coins as such which led to Óláf’s downfall, if they were seen as an expression of the style of strong kingship and centralised power which had made Óláf unpopular and against which Magnús was warned, it may be that he avoided issuing coins in Norway because of the message that this might send out, rightly or wrongly, about other aspects of his kingship. It is also conceivable, however, given how few coins of the two Óláfs survive, that Magnús also issued coinage on a minimal scale, but that none has so far been discovered. Whatever the reasons for the lack of coinage under Magnús, however, it was in the reign of his successor, Harald harð ráð i that Norwegian coinage was introduced on a large scale. Harald had spent years travelling abroad, serving as a mercenary in Russia and the Byzantine empire, and would certainly have become familiar with coinage there. Indeed, the large quantity of Byzantine coin which he brought back with him to Scandinavia probably influenced the design of contemporary Danish coins, although interestingly not Harald’s own Norwegian coinage. Harald, as his nickname suggests, was not afraid of being seen as tyrannical, and was capable of ruthless brutality towards those who opposed him. He thereby apparently imposed the sort of strong kingship that his predecessors had been unable to sustain (Krag 2003, 198–199). As mentioned earlier, he was also canny enough to recognise the potential benefits of debasing the coinage, especially when



payments were being made from his own treasury. This debasement demonstrates the regulation of both the weight and silver content of the coins along Anglo-Saxon lines, together with an appreciation of the financial benefits to be derived from the introduction of a regulated coinage The Norwegian coinage continued to expand under Harald’s son Óláf the Peaceful, although the designs of all but the earliest of Harald’s coins and all of Óláf’s coins were crude, and inscriptions largely illiterate. It is perhaps significant that it was also in the reign of these two kings, in the late eleventh century, that urbanisation also took off, with the development (or in some cases redevelopment) of a number of planned towns, along with the establishment of permanent bishoprics based in the new towns and both had strong relations with the Church, although Harald occasionally had disputes with the archbishops of Hamburg-Bremen over primacy within Norway (Andersson 2003, 325–328; Krag 2003, 197–198; Sawyer and Sawyer 2003, 155). In the same period, locally produced coinage gradually replaced Anglo-Saxon and German coins, especially in Trondheim and Oslo, but also more widely, and the move towards exclusion of foreign coins has been interpreted as a sign of English influence, while the recoinage at the beginning of each reign from Harald onwards may be interpreted as a controlled renovatio monetae along English lines (Skaare 1976, 112; 1981; Gullbekk 1991). The use of coinage in late eleventh-century Norway fits into a broader pattern of kingship and state development, and Sawyer (1999, 102–104) argues for a broader anglicisation of Norwegian government in this period. It should also be noted that under Harald and Óláf the Norwegian coinage for the first time had an economic function outside Norway itself, as coins of both kings have been found in England and Scotland, reversing the earlier trend of coins from England to Scandinavia (Archibald 1991; Williams and Sharples 2003). This probably reflects growing trading links between England and Norway, with a Norwegian trading ship recorded at Grimsby in 1069, and a flourishing Anglo-Norwegian trade in place by the early twelfth century (Sawyer 1986; 1999, 98).

SWEDEN Whereas coinage began in Norway with a false start in the 990s and only took off half a century later, the pattern in Sweden was almost the opposite. Under Eric the Victorious and Olof Tribute-king, a strong centralised royal power developed in the late tenth century. For the first time this involved combining the separate kingdoms of the Svear in eastern Sweden and the Götar in southern and western Sweden (Sawyer 1988). One expression of this new strong kingship was the replacement in the 970s of Birka as a trading centre by the new planned town of Sigtuna, under firm royal control (Lindkvist 2003b, 221–223). Malmer has argued that even before the regal coinage of the late 990s, the later types of the anonymous tenth-century Scandinavian coinage were Swedish, and probably produced in Sigtuna (Malmer 1966). Sigtuna was also the centre for early Swedish regal minting, and also had a strong Church presence from an early stage. Olof, whose nickname probably indicates that he was subject to some form of Danish overlordship, was converted to Christianity and baptised at Husaby in Västergötland, far from his power base in the Sigtuna area. He is credited with establishing the bishopric of Skara and is seen as the founder of the Christian kingdom of Sweden (Lindkvist 2003, 223–224; Sawyer and Sawyer 2003, 154). This Christian emphasis is again reflected in the coinage. The designs largely copied Anglo-Saxon models, although there was also some Byzantine influence in the



eleventh century (Malmer 1981), but in addition to purely derivative (and blundered) inscriptions, there were also innovative and explicitly Christian inscriptions, such as IN NOMINE DOMINI (‘In the name of the Lord’), which occurred at an early stage in the Sigtuna coinage, probably within the first five years (Malmer 1995; 1997). Both the Christianising influence and the united kingdom of the Svear and the Götar continued under Olof’s son Anund (c 1020–c 50), who took the additional Christian name Jacob. Anund continued to have coins minted at Sigtuna, although this process may have terminated before the end of his reign. Like Óláf Haraldsson, Anund came into conflict with the expanding authority of the Danish king Cnut, and this may have disrupted his own authority in Sweden. Opinion is divided as to whether coins struck at Sigtuna in the name of Cnut indicate that he exercised authority in eastern Sweden, or whether the coins are simply local imitations of Cnut’s own coins which failed to change the name of the ruler, which is by no means unusual amongst imitative coins in this period (Sawyer 1989; Malmer 1995; Lindkvist 2003, 224). Whatever the truth of Cnut’s influence in Sweden, which can only have been extremely short-lived, the strong start to Swedish kingship and coinage did not outlast Anund Jacob. Whereas both Christianity and royal power were consolidated in Norway as the century progressed, with a corresponding influence on the coinage, in Sweden the unified kingship of the Svear and the Götar collapsed, and while the process of Christianisation saw some advances, with bishoprics probably established in Linköping and Sigtuna as well as Skara by the end of the eleventh century, the firm link between Christianity and kingship was lost, and Sweden faced a period of religious divisions as well as dynastic struggles (Sawyer 1988; Lindkvist 2003, 224–225; Sawyer and Sawyer 2003, 155). By no coincidence, the same period saw the collapse of the coinage, which only revived in the mid-twelfth century in the comparatively long and stable reign of Knut Eriksson (c 1167–95/96) (Jonsson 1995; Kilger 1995).

REGAL COINAGE IN DENMARK The regal coinage in Denmark was more successful than in either Norway or Sweden. As in Sweden, the coinage got off to a much stronger start than in Norway, but unlike Sweden it did not collapse. Furthermore, minting took place in a number of towns across the kingdom. This difference derives from a number of factors. First, as discussed above, coinage had been issued, if only intermittently, by Danish kings since the early ninth century, so the idea of using coins as coins, rather than as bullion, was more established in at least some parts of the kingdom. Furthermore, the underlying factors which had permitted the development of the earliest Danish coinage were still in place. Royal power in the late tenth century was strong, Christianity was spreading, as shown by a growing number of bishoprics, and the number of towns also increased. This trend continued in the eleventh century, and for quarter of a century Danish kings ruled England as well as Denmark, bringing direct contact with the more sophisticated administrative structures of Anglo-Saxon England. In particular, there was a strong Anglo-Saxon influence on the Danish Church under Cnut, with the introduction of Anglo-Saxon bishops, and a firm ecclesiatical organisation, founded on permanent urban-based bishoprics, was in place by c 1060 (Skovgaard-Petersen 2003, 177–178; Sawyer and Sawyer 2003, 151, 155). Political, economic and ecclesiastical contacts with Germany also remained strong.



The network of towns, mints and bishoprics which developed across the whole of Denmark in the eleventh century shows considerable similarity with the late AngloSaxon state (Randsborg 1980), although without the same degree of centralised power and standardisation, and probably with less royal authority (Williams 2002). It is also noticeable that Danish kings, unlike their Anglo-Saxon (and, to some extent, Norwegian) counterparts, were unable to prevent the circulation of foreign coinage in the kingdom alongside their own issues, and Danish hoards of the eleventh century were dominated by Anglo-Saxon and German coins, although the proportion of Danish coins increased as the century progressed. Furthermore, despite coin hoards becoming increasingly dominant, mixed hoards containing bullion as well as coins continued throughout the eleventh century, although largely in peripheral areas, showing that the transition to a coin-based economy was not absolute (Bendixen 1981, 411; Jensen et al 1992). In terms of royal authority, it is also interesting that coin production in Denmark was not strongest in the heartland of the Danish kingdom in Jutland, nor in the most established trading centres, but in newer towns in outlying areas, with mints like Lund, rather than Hedeby or Ribe, emerging as the main centres of production (Andrén 1983). This suggests that coinage was deliberately used as an instrument in the expansion of royal authority across the kingdom, and that the political functions of coinage were more important than the economic functions. This is not to deny that the coinage had economic functions as well, especially given the large scale of coin-production at Lund, but it is hard to believe that there was a strong economic demand only in the periphery of the kingdom, and not in the core. As elsewhere, the coinage is explicitly Christian, with a variety of Christian images derived from Anglo-Saxon, German and Byzantine designs, including a variety of crosses, the image of Christ, the hand of God the Lamb of God and the dove of the Holy Spirit. As in Sweden, there were also some innovative, but specifically Christian, inscriptions including one type from Slagelse which carries the opening of St John’s Gospel rather than a regal inscription or mint signature. As elsewhere, an initial phase of imitative coinage was followed by the adoption of more distinctively ‘national’ designs but imitative designs continued alongside Danish designs throughout the Viking Age. Coins are divided between those which give the kings their Danish titles, those which simply copy the inscriptions of the coins they were imitating, those which combine Danish rulers with English titles and still others which have blundered inscriptions. Matters were confused still further under Cnut and Harð acnut, since both were kings in both Denmark and England, and issued coinage with both titles. It can be particularly difficult to distinguish between coins of Cnut from London and Lund, given the similarity of the mint signatures. An initial attempt by Cnut to introduce a national coinage based on English designs was followed by a second reform, c 1030, with different designs for each mint, but common weight standards (Hauberg 1900; Jensen 1995). The coins of this period included a number of innovative designs based on Scandinavian motifs. The link between coinage and kingship became sufficiently well established in Denmark under Cnut, that Magnús of Norway issued coins in Denmark in the 1040s, during which period he was attempting to establish himself as king of Denmark as well (Becker 1981; Jensen 1995, 70), although (see above), he did not issue coinage on any sort of scale, if at all, in Norway. It is hard to imagine that this difference results solely from different economic pressures although, as noted, urbanisation had developed more quickly in Denmark than in Norway by this time, and towns provided a focus for the monetary use



of coinage. Nevertheless, it seems likely that in Denmark Magnús had to strike coins if he was to be seen as acting in a kingly fashion, whereas in Norway, this might have been interpreted as trying to extend royal rights further than was appropriate or welcome.

THE DANELAW If the development of a monetary economy was more advanced in Denmark than in the other Scandinavian kingdoms, it was more advanced still in the Danelaw. Here, of course, the bullion-using settlers came into direct contact with a population with established traditions of royal authority, Christianity, coinage and, to a lesser extent, urbanisation, and they seem to have been quick to adapt in many ways, both in the kingdom of Northumbria and in the southern Danelaw. It is also pertinent that many members of the micel here (‘great raiding army’) which conquered and settled in England in the late ninth century had also raided extensively on the continent, and were thus familiar with Frankish as well as Anglo-Saxon coinage, having taken extensive gelds as well as looting on both sides of the English Channel, as reflected in large numbers of Anglo-Saxon and Frankish coin hoards of the period (Armstrong 1998; Brooks and Graham-Campbell 2000). Furthermore, given the maritime background of the Scandinavian settlers, the North Sea continued to provide a link between the Danelaw and the continent, rather than a barrier. The introduction of coinage did not immediately follow the conquest of East Anglia, Mercia and Northumbria, and even after the new Scandinavian rulers began to issue coinage there was a transitional phase, in which a controlled local coinage existed alongside the use of imported coin and other forms of bullion. This transitional phase has recently been discussed by both Blackburn (2001; 2002) and Graham-Campbell (2001a), and forms the background to Archibald’s chapter in this volume (Chapter 4). With two recent major surveys by Blackburn (2001; 2004) of the coinage north and south of the Humber, detailed discussion of the coins themselves is unnecessary here, and I will instead concentrate on the main aspects of change and development. The first of these is a gradual transition from bullion to locally issued coinage. It is clear from documentary sources that the micel here received substantial payments from both Anglo-Saxon and Frankish rulers. The documentary evidence is supported by a group of weights with Anglo-Saxon coins set into the top, but with a distribution that suggests an Anglo-Scandinavian rather than an Anglo-Saxon origin for the weights. It has been suggested that these weights may have been used by the micel here for weighing out payments in coin (Archibald 1998; Williams 1999), thus representing the use of coins within a bullion economy. The majority of the weights of this type incorporate coins which would have been current while the micel here was active in the 860s and 870s, and only one example has so far been found which dates from after the battle of Edington in 878. This is a coin of the tenth century, which is in such poor condition that it cannot be precisely dated, but stylistically it seems most likely to be a coin of Edward the Elder, suggesting a shift away from this type of payment, and perhaps therefore from a bullion economy, after Edward’s reign. The use of coins as bullion is also suggested by the use of test marks on the coins. Testing by bending may appear as early as the Gravesend hoard, c 871, while the Croydon hoard of c 872 contains kufic dirhems with edge-nicking. This may have been done on the continent, as the hoard also contains Anglo-Saxon coins with no test marks, although test



marks may also appear on hack silver from the hoard, and possibly on the single Carolingian denier (Archibald 1990; Graham-Campbell 2001a). The earliest clear evidence for pecking comes in the Stamford hoard of c 890, followed by the Ashdon hoard of c 895 (Blackburn 1989; 2001; Archibald 1990; Graham-Campbell 2001a), and pecking is extremely common in the Cuerdale hoard of c 905 (Archibald, Chapter 4, in this volume). Testing by another method is apparent in a small mixed hoard found near Carnforth in Lancashire, provisionally dated to c 920 (Graham-Campbell 2001b), although it has not yet been fully published (a preliminary report is provided by Youngs and Porter 1999). This contains three kufic dirhems, one of which has a small group of indentations which from their shape appear to be tooth-marks, while nicking is also visible on hack-silver from the hoard, although it is again difficult to be certain whether this was done in England or elsewhere. The latest recorded instance of pecking is a single find of a St Peter sword type silver penny, dating from the 920s, found near Louth in Lincolnshire (Blackburn and Bonser 1986, no. 113; Blackburn 2001). Similarly, the continued use of bullion points to a mixed economy. Mixed hoards of Viking character are known from both the north and south of England in the ninth century, and a number of these have been specifically linked with the activities of the micel here (Brooks and Graham-Campbell 2000). While single-finds of ingots and hack-silver are known from the southern Danelaw, it is notable that while mixed hoards continued in the tenth century north of the Humber, none has yet been discovered south of the Humber (Blackburn 2001). To date, the latest of the northern hoards is Bossall / Flaxton, deposited c 927 (Graham-Campbell 2001a; 2001b). A similar pattern also occurs with the presence of imported coins, with Carolingian coins represented in Ashdon, Stamford and Cuerdale, and in the productive site at Torksey. Dirhems occur as single finds, and also in hoards as late as Goldsborough (N Yorks) and Carnforth, both c 920, and Thurcaston (Leics), c 925. The extent to which these are representative of the local economy, rather than recent importations from the Irish Sea area, may be debated (Blackburn 2001; Graham-Campbell 2001a; 2001b). Nevertheless, the general picture of the survival of bullion, test marks and imported coinage combines to suggest that the transition from a bullion to a coin-based economy was only fully completed by the late 920s, although this transition appears to have taken place more quickly in the south than in the north. Even in the northern Danelaw this development may have taken place at different rates, depending on proximity both to centres of royal power, such as York, and to trade routes to the Irish Sea, since the Vikings in Ireland were much slower to adopt coinage (see below). The proximity to centres of royal power, and particularly to towns in which coins were minted, is important both because towns provided an urban marketplace in which coins were useful for economic purposes, and because, as with the adoption of coinage in Scandinavia, the adoption of coinage was closely linked with the exercise of Christian kingship and ideology. In particular, the Christian aspect of the Danelaw coinage comes through very strongly. The Vikings of the micel here were probably predominantly pagan, and certainly their victims thought of them as such. There is also little doubt that they had done major damage to the Church in the Danelaw by the time they settled, but it is less clear how quickly the process of conversion and Christianisation took place (Abrams 1991a; 1991b). Julian D Richards (2001) has argued that, as early as 873–74, divergent burial practice in the Repton area suggests that one part of the micel here was beginning to assimilate with Anglo-Saxon ways, while another part retained their traditional customs, while the political ‘conversion’ of Guthrum and his leading followers after the



battle of Edington in 878 has already been mentioned. It is notable that Guthrum’s East Anglian coinage in his baptismal name of Athelstan is one of the earliest substantive coinages of the Danelaw (and the earliest not simply imitating Alfred’s coinage even to the extent of imitating his name), emphasising a Christian influence on the coinage right from the start. The Christian tradition continued with the St Edmund Memorial type, begun c 895 and closely imitating the coinage of Edmund of East Anglia, who had been killed by the micel here in 869. However, the coins now gave Edmund the title of saint as well as king, and thereby represent the earliest surviving evidence of the cult of St Edmund. This indicates fairly clearly a deliberate policy of support for the cult by the rulers in the southern Danelaw who were responsible for issuing the coinage. Nor was Edmund the only saint to be commemorated on coinage of the Danelaw, with coins produced at York in the name of St Peter, and at Lincoln in the name of St Martin (Dolley 1965; Stewart 1967; Blunt 1969; Stewart and Lyon 1992; Blackburn 2001; 2004). The Christian message was also reinforced by the use of religious inscriptions on some of the earlier coins issued at York, such as D[omi]N[u]S D[eu]S REX (‘The Lord God [is] King’), D[omi]N[u]S D[eu]S O[mnipotens] REX (‘The Lord God Almighty [is] King’), and a quotation from Psalm 98, MIRABILIA FECIT (‘He has done marvellous things’) (Lyon and Stewart 1961; Dolley 1965; Blackburn 2004). In addition, the early regal coinage struck at York, c 895–905, uses a variety of crosses in the design, while all of the Danelaw coinage follows the Anglo-Saxon practice of beginning the inscriptions with an initial cross. Although crosses are not the only images to be used on coins of the Danelaw, it is safe to say that there is some Christian symbolism or reference on virtually all of the Danelaw coins and, while some coins show pagan symbols as well, there are no coins which have exclusively pagan symbols. Thus, although the triquetra design on coins of Viking Northumbria from the 940s has sometimes been interpreted as a pagan symbol, there are other possible interpretations of the triquetra (Skaare 1976, 70), one of which is the Trinity. The bird on the coins attributed to Anlaf Guthfrithsson has usually been interpreted as a raven, but it may equally well be an eagle. Like the raven, the eagle has associations with the god Odin, but it is also an evangelist symbol, although it is fair to note that other evangelist symbols do not appear on the coinage. If the bird is an eagle, it could also be religiously neutral, as a symbol of power and authority rather than a religious symbol. While Thor’s hammer appears on York coins of the 920s, it appears only on coins in the name of St Peter, possibly as a deliberate means of promoting assimilation between the two religions. Interestingly, the Thor’s hammer appears on coins of the St Peter Sword type, and the sword is associated with St Peter’s defence of Christ at Gethsemane. As Thor was the defender of the Norse gods, so St Peter was the defender of Christ. Similarly, St Peter is associated with fishing and the sea, as was Thor, with fishing playing a repeated role in his battle against the forces of evil represented by Loki and the World Serpent, as well as Thor’s associations with the weather. Thus this particular coinage may reflect a deliberate element of religious syncretism in the conversion process. With regard to Thor, it is also interesting to note that a Thor’s hammer was found in the Cuerdale hoard, which represents a traditional mixed bullion hoard, even though it contains Christian coins from the Danelaw. Whether the hammer reflects the religion of the compiler of the hoard or whether it is simply scrap for bullion is, of course, an open question. However, there was also a ambiguous pendant representing a Thor’s hammer or a cross (or both) in the Goldsborough mixed hoard (Wilson 1957), and a Thor’s hammer



was also found near the Thurcaston hoard, one of the latest hoards to include Islamic coins (Blackburn 2001; Graham-Campbell 2001a). Thus one may argue that there is a pagan context to the survival of mixed hoards within the economy of the Danelaw, but this has to be seen alongside the explicitly Christian message of the coins. As Blackburn (2001) argues: ‘That Christians were in a position to influence the design of these coinages is interesting enough, but their sustained use over so long a period surely indicates that the promotion of a Christian message in this context was official policy’. A further point of interest in the Danelaw coinage is the gradual progression from imitation to innovation. As mentioned above, the normal pattern for the introduction of coinage in an area or culture in which coins had not previously been issued was an initial phase imitating a coinage from elsewhere, which was already widely accepted, followed more gradually by the introduction of more distinctively ‘local’ designs. The Danelaw basically follows this pattern, but with some interesting minor differences, reflecting the peculiar circumstances of coin use in the Danelaw. Prior to the conquest of the Danelaw, East Anglia and Mercia had issued silver pennies, and Northumbria had issued smaller copper-alloy coins, known to numismatists as stycas. This meant that the native population were already familiar with coinage. The main influence after the conquest was Alfred’s Wessex, which recovered much of its former wealth, as well as assuming overlordship of part of Mercia, after the pause in fighting following the Treaty of Wedmore. Alfred was able first to improve the quality of his coins, c 875, and then to introduce a new weight standard c 880, making the coins slightly larger and heavier than the coinage of the first few years of his reign (Blackburn 2003). In addition to the influence of Wessex, however, many of the Vikings who settled in the Danelaw had also raided the Franks, and were therefore also familiar with Frankish coinage, while there was also some familiarity with coins from even further afield, with small numbers of Islamic and Byzantine coins found in Viking contexts in the Danelaw. The first phase, which probably began c 880, draws on a variety of influences. The designs were copied from neighbouring Wessex, imitating the short-lived ‘London monogram’ type, a local ‘Oxford’ type and the more substantial ‘Two-line’ type, which covered the bulk of the latter part of Alfred’s reign. However, as mentioned, Alfred had introduced a new, higher weight standard in Wessex, but coins of the Danelaw followed the weight-standard of Alfred’s pre-reform coinage, which was also the standard which had been followed in East Anglia and Mercia before their conquest (Blackburn 2001; 2003). This raises interesting questions. Was the decision to follow the earlier weight standard due to pressure from the local population to strike coins to the standard they were used to, despite a short gap in coin production in Mercia, and a longer gap in East Anglia? Or was it due to the fact that the Vikings had obtained substantial gelds paid in coinage of the earlier standard (from Wessex, as well as from East Anglia and Mercia), and therefore took as their norm the weight standards of the coins they already possessed. Although the Christian influence on the coinage could point to some involvement of the Anglo-Saxon population, it is notable that the bulk of the recorded moneyers of the early Danelaw coinage seem to have been Franks, apparently imported specially for this purpose (Blackburn 2001). This would suggest complete discontinuity in the coinage, with the new Frankish moneyers striking coins to the standards set down by their patrons on the basis of their existing loot. That the weight standards were set by the Viking rulers, rather than by local pressure for continuity, is also suggested by the complete break in the



coinage of Northumbria. There the gap between the collapse of the pre-Viking coinage and the beginning of the new coinage was longer, and the difference more substantial. Rather than revive the base styca coinage, the new rulers chose to introduce silver pennies to the same weight standard as the southern Danelaw, suggesting a conscious view that this in some way the ‘proper’ weight for silver pennies, and that these were the ‘proper’ form for coinage. The next phase of the coinage was more original. Some imitation continued, with Anglo-Saxon, Frankish and possibly even Byzantine coins imitated in the late 890s and early tenth century (for the earlier imitation of Frankish gold, see Blackburn, Chapter 5, in this volume; for the view that the Cross-on-steps design is not derived from Byzantine coins, but is a purely local design deriving from Northumbrian stone crosses, see Blackburn 2004). However, the coinage now largely moved to more innovative designs, as well as shifting from imitating Alfred’s coinage more or less directly, to regal coinage in the names of Danelaw rulers, or anonymous coinage in the names of saints, or with religious inscriptions, as described above. Interestingly, there was a tendency towards innovative design from an early stage in the northern Danelaw since, although the York coins of the late 890s show a cross, which is a typical Anglo-Saxon coin-motif, the form of the cross is not imitated directly from Anglo-Saxon or Frankish designs. The greater innovation of the northern compared with the southern part of the Danelaw may reflect the lack of a pre-Viking Northumbrian silver coinage, since the form of the stycas made them unlikely to inspire imitation in the way that the old East Anglian coinage inspired the St Edmund Memorial type. The difference may also reflect a greater mixture of imported coinage in the north, and the lack of a single powerful coinage in a neighbouring kingdom to act as a model, which was probably a factor in the imitations of Alfred’s types in East Anglia and the south-east Midlands. Whatever the reason, however, the trend towards innovation continued and, while the cross was a tried and tested coin motif, as were two-line inscriptions (as seen on the early St Peter coinage) and monograms, later types from Northumbria have designs which are completely independent of Anglo-Saxon and Frankish traditions, including a bow and arrow, a sword, a triquetra, a standard or wind-vane, and a raven or eagle. In the case of the last three, the use of these distinctively non-Anglo-Saxon designs, together with the practice of giving the royal title in Old Norse (eg ANLAF CVNVNC ⳱ Anlaf konungr ⳱ King Óláf ), rather than in Latin, shows a marked contrast with coins in the names of Anglo-Saxon kings struck at York, and also with other coins from York in the names of Anglo-Scandinavian kings struck in more typically Anglo-Saxon style. These differences provide an interesting counterpart to the Anglo-Saxon Chronicle as an index of the struggle for power in Northumbria from 937 to 954. The coins do not necessarily tell the same story as the Anglo-Saxon Chronicle, while Scottish, Irish and later Scandinavian sources also call the chronology of the Chronicle into question, and this is an area which would repay further research (Sawyer 1995; Williams 1996; Woolf 1998). A final important aspect of the Danelaw coinage is the extent to which it fits the general trend associating coins with kingship. The Anglo-Saxon Chronicle makes it clear that the micel here was an association of different warbands, with different leaders, and implies that when the different groups banded together, the leadership of the combined force was shared. Some of these leaders are described as kings, others as earls, and occasionally even individuals with the lesser rank of hold are mentioned (Abels 2003; Williams, forthcoming b). Given the lack of landed wealth, at least during the raiding



phase, it is likely that these differences in status reflect lineage rather than any difference in power. Anyone of ‘royal’ blood could claim to be a king as long as others were willing to accept that claim, irrespective of whether he possessed a territorial kingdom. Those who came from families with the status of earls similarly inherited that status, and it is interesting that neither the earls of Hlaðir, who at times dominated if not actually ruled Norway, nor the earls of Orkney, who at times ruled large parts of northern Scotland, ever seem to have taken the title of king. If the Chronicle is to be believed, there were probably at least as many ‘earls’ at the head of the micel here as ‘kings’, and when the micel here broke up into smaller groups to settle the Danelaw, some of these groups appear to have been led by earls, such as the earl ‘thurferth’ who submitted to Edward the Elder at Passenham in 920 (ASC, sub anno 920). However, these earls do not generally appear on the coinage, the sole exception being ‘Sihtric Comes’, who issued coins at ‘Sceldfor’, probably Great Shelford in Cambridgeshire rather than the Nottinghamshire Shelford (Hart 1995; Blackburn 2001, 132), which would suggest that he was the leader of a group based in and around Cambridge. Sihtric’s coinage was comparatively small, although not necessarily any smaller than that of some of the ‘kings’ who were his contemporaries, such as Guthfrith, who is recorded as king of Northumbria, but who apparently had a small coinage produced somewhere in the Midlands (Blackburn 2001). By contrast, a number of specifically regal coinages do survive, and it is hard to see why there are not more surviving coins in the name of earls, unless we assume that there was some understanding that coinage was a specifically regal right, as suggested more generally in this chapter. An interesting possibility is raised, however, by the existence of the various anonymous coinages, including those issued in the names of saints, or with religious inscriptions, as well as those which simply imitated Alfred’s coinage directly. Some of these anonymous types were struck in very large numbers. Blackburn (2001, 132–133, 136) has raised the interesting possibility that the reason for the lack of a regal title on some of these coins might have been that particular coin types were jointly issued by several rulers, and circulated across an area over which there was no single ruler. This seems quite plausible since, as Blackburn points out, various coin types seem to have been produced in areas which fall across the boundaries of what appear on the basis of the Chronicle to have been distinct territories. A case in point is the St Edmund Memorial type, which has an obvious association with East Anglia, but also appears to have been issued in the south-east Midlands. Although it is not susceptible to proof, a further possible refinement of this idea is that some of the anonymous coinage may have been issued by earls, who wanted to participate in issuing coins as an aspect of being seen to be rulers, but were persuaded that it was a strictly kingly prerogative to have their names on the coins. A partial parallel to this would be the imitations of Byzantine coinage produced during the Migration Period, at which point it seems to have been considered acceptable for kings to put their names on silver and copper coins, while having a name on the gold coinage remained a strictly imperial prerogative (Grierson 1991). What strengthens the case for the Danelaw earls is that the Chronicle tells of relatively few kings, once the period of settlement began, and most of those known to history did issue coinage in their own name. In a number of cases, kings based in Northumbria seem specifically to have struck regal coinage south of the Humber (Blackburn 2001), and the only recorded ‘kingdoms’ were in Northumbria and East Anglia, leaving the rulership in the Midlands to be accounted for.



This is, of course, highly speculative, but it does fit the available evidence, as well as the broader theory presented here on the link between coinage and Christian kingship. What is certainly true is that there remains scope for further work on the interpretation of the relationship between coinage and political structures in the Danelaw.

IRELAND Altogether different patterns are visible in Ireland and Scotland. No coins were issued in Ireland before the Viking Age, nor was there much use of imported coin. There was therefore less of an immediate model for the Scandinavian settlers. However, the new urban settlements which grew up along the Irish coast derived their wealth from trade, including trade with Anglo-Saxon England, and this led to the importation of AngloSaxon coinage in the tenth century. As Sheehan discusses in this volume (Chapter 9), the coinage made little impact in inland areas, and it is likely that many of the coins were melted down for reuse in other forms, with different types of hoards in different areas reflecting different approaches to precious metal economies. This shows some similarities with the situation described by Ibn Fadlan at the other end of the Viking world, with trading going on around the Irish Sea using coins, with the Anglo-Saxons and AngloScandinavians using the coins as coins, but with the Irish and Hiberno-Norse using them as bullion for exchange, or even as raw material to be traded. It seems likely that at least some of the splendid jewellery from the Irish Sea area was made from melted down coins, converting the silver from a coin-based economy to a status-based economy, while so-called ring-money may, like the Permian rings, have straddled status economy and bullion economy. It was only at the very end of the tenth century that the Hiberno-Norse king of Dublin, Sihtric Silkbeard, introduced coinage of his own which, like the contemporary coinage in Denmark, Sweden and Norway, was closely copied from Anglo-Saxon models (Dolley 1966). It is particularly interesting that this development only took place so late in Dublin, since several of Sihtric’s predecessors in Dublin had also ruled in York, and a number of them had issued coins there. Given familiarity with coinage in York, the availability of silver in Viking Dublin, and given how easy it would have been on a technical level to produce coins, this suggests most strongly that for some reason the rulers of Dublin throughout the tenth century felt that it was unnecessary or inappropriate to issue coinage in Dublin. As elsewhere, this suggests in turn that the economic superiority of regulated coinage over imported coinage and bullion was less than obvious, and that the decision to introduce a Dublin coinage in the 990s was motivated by more than purely economic factors.

SCOTLAND AND THE ISLE OF MAN The pattern in Scotland is different again. As in Ireland, there was no coinage issued before the Viking Age, nor was there much use of imported coinage, with the exception of ninth-century Northumbrian coinage in the Borders, which should be seen within the context of Anglian Northumbria rather than the emerging Scottish kingdom, although that area falls within the borders of modern Scotland (Graham-Campbell 1995; Bateson 1997).



As mentioned earlier, the east coast as far as the Dornoch Firth may also have formed a peripheral part of the North Sea monetary zone in the early to mid-eighth century. There is a marked change in the course of the Viking Age, beginning in the late ninth century. No coins were issued locally, but imported coins circulated alongside bullion items. Interestingly, this becomes apparent earlier in the single find evidence (with finds beginning in the ninth century) than in the hoards (beginning in the early tenth century), indicating that the two types of evidence lead to different conclusions. A distinction between hoard evidence and the evidence of single finds is a broader pattern which appears elsewhere, and has been applied with particular effect to the study of monetary circulation in Anglo-Saxon England. It is now widely argued that the distribution of single finds may give a more accurate impression of the circulation of coinage than hoards alone, while hoards tell us more about the storage of wealth; several of the chapters in this volume use single-find evidence prominently. Unfortunately, studies of monetary circulation in Scotland have tended to focus on the hoards with, I believe, a misleading effect, while the most detailed recent study (Metcalf 1995) focused almost exclusively on what the evidence of the Scottish hoards tells us about the sources of the Anglo-Saxon coins in Viking Scotland, rather than on circulation within Viking Scotland itself. The hoards are dominated by bullion material in various forms, including hack-silver, ingots and intact items of jewellery. Both the first and the last category include the so-called ring-money (see above), which it has been argued may have been intended as a form of weight-adjusted currency on a fairly large scale. Of the hoards listed in GrahamCampbell’s catalogue, 11 are coinless hoards, 16 are mixed hoards and only seven are coin hoards with no recorded bullion. As yet, the sort of comparative regional analysis of hoard types discussed by John Sheehan in the Irish context has not been carried out for Scotland, and such a study might yield interesting results, although the limited number of hoards might mean that the results would not have much statistical validity. Of the mixed hoards and coin hoards, only five appear to have contained more than 100 coins, compared with probably nine containing fewer than 20 coins. This does suggest very clearly that coins were not the preferred medium for the storage of wealth in Viking Scotland, especially when one compares the small proportion of coins in large hoards such as Skaill or Burray with hoards from the Danelaw and Scandinavia, and it is likely that many of the coins that came into Scotland were melted down as the raw material for jewellery. The surviving coins are predominantly Anglo-Saxon, with a limited number of continental deniers (and up to three imitation gold solidi) and Islamic dirhams (and one gold dinar), together with Hiberno-Norse pennies, both as single-finds and in hoards. Finally, from the end of the Viking Age, there are four single-finds of Norwegian silver pennies, two each of Harald harð ráð i and Óláf kyrri (Williams and Sharples 2003; Williams, forthcoming b). The mixture shows clearly that this was not a closely regulated monetary economy. Viking hoards, whatever their contents, are themselves less common in Scotland than in Ireland, and this has led to the assumption that Viking Scotland was significantly poorer than Ireland, while the small total number of coins in total, and the limited importance of coinage within the hoards, has led to the view that coinage circulation in Viking Scotland was minimal (eg Crawford 1987; Williams 1996; Graham-Campbell and Batey 1998). This view may be questioned, on the basis of the single-finds, especially those from excavated sites. Certainly the number of coin finds is low, especially seen from an English perspective, but this is hardly surprising since no coins were being produced locally, and since Scotland also lacked towns to provide centres for monetary trade, in addition to



having a much lower density of population than England. Nevertheless, it is notable that there are coin finds from a significant number of the known settlement sites excavated using modern techniques, while the absence of coin finds from older excavations may reflect excavation techniques as much as a genuine absence of coins. It is particularly noticeable that there is a concentration of coin finds from areas of extensive settlement such as Whithorn and Birsay, but the distribution of site finds extends across most of the area of Norse settlement in Scotland. Most of them are Anglo-Saxon, and Robert Stevenson (1986) concluded that Anglo-Saxon coins may have had ‘a potentially monetary or exchange function, which coins from distant places did not have’, and more recent work on the single finds has also argued for a more extensive circulation of coinage than has traditionally been accepted. (Bateson 1997; Williams and Sharples 2003; Williams, forthcoming b). It is also relevant to note the existence of land assessment units known as ‘ouncelands’ and ‘pennylands’ across areas of Scotland which correlate fairly closely with what we know from other sources to be the areas dominated by the earls of Orkney in the late 10th and 11th centuries. Interpretations of the ounceland and pennyland assessments vary, including the dating of the assessment units and the extent to which they should be seen as a single system (Crawford 1987; 1993; MacGregor and Crawford 1987; Thomson 1987; 2002; Andersen 1991; Williams 1996; 2003 and 2004), but the most recent substantial study suggests that an origin for both the ounceland and pennyland assessments in the late Viking Age is likely. An assessment in ounces points clearly to valuation by weight, while assessment in pennies is more ambiguous, since penny in this context might conceivably relate directly to the silver penny, or to a unit of account based on the penny, or simply to a penny weight of silver within a bullion economy. However, whichever interpretation of the ‘penny’ in the pennyland is followed, it does at least suggest that pennies as coins were sufficiently common at the very least to indicate a recognised unit of weight, and possibly to affect directly a comparatively sophisticated system (or systems) of land assessment. Coupled with the single-find evidence, this suggests that the importance of coinage in Scotland in the later Viking Age has been underestimated. This raises the important question of why Scandinavian rulers in Scotland did not introduce coinage of their own. Viking settlers in England and Normandy had done so comparatively quickly, and while it took rather longer for coinage to take hold in Scandinavia, the kings of Denmark, Norway and Sweden had all attempted to introduce coinage by the end of the tenth century, as had the king of Dublin. The influence of Dublin is apparent around the Irish Sea area, and even extending into Scotland, with HibernoNorse pennies found in a number of hoards from southern Scotland, as well as two site finds from Whithorn in Galloway (Metcalf 1995; Williams, forthcoming a). Irish influence also apparently led to the creation of a coinage in the Isle of Man in the early eleventh century. This idea was proposed by Michael Dolley (1976), although he produced little evidence to support it. A more detailed study by Kristin Bornholdt (1999), based on stylistic analysis and distribution of finds, has strengthened the case for a Hiberno-Manx coinage considerably, although the evidence is not strong enough to provide conclusive proof. This development has to be seen in the context of the domination of the kingdom of Isle of Man by Dublin in this period (Duffy 1992), with Man repeatedly falling under the domination either of whoever was ruling Dublin at the time, or of rivals for the Dublin kingship who could use Man as a powerbase from which



to press their claims. Man, in turn, exercised some influence over the Western Isles, although precise details for the extent of Man’s authority in this period are few and far between. One might therefore have expected that Irish influence would have led to the adoption of coinage up the whole of the western seaboard of Scotland, and that from there it might have spread to other areas of the Scandinavian settlements in Scotland, and particularly into those dominated by the powerful earls of Orkney. The late tenth and eleventh centuries saw the height of the power of the Orkney earldom, with power and authority equal to, if not greater than, the kings of Dublin and of Man (Crawford 1987; Williams 1996; 2004). Furthermore, although fewer than the Irish hoards, the Scottish hoards show that considerable wealth was available to local rulers, while the presence of coins in both hoards and excavations shows that models were available. Why then did rulers in Viking Scotland, and especially the powerful earls of Orkney, not participate in the general trend and issue their own coins? Again, I would argue that this supports the contention that the introduction of locally issued coinage was not primarily economically driven, but tied in with both Christianity and kingship. It seems clear that Christianity survived amongst the pre-Viking population in northern and western Scotland, and that a substantial proportion of the Scandinavian settlers had been converted by the end of the tenth century. However, unlike England and Normandy, the pre-existing population did not have a coin-based economy for the settlers to assimilate, and unlike Dublin in the late tenth century, the earls of Orkney did not make a point of being seen to be Christians at this time. Saga traditions record that Óláf Tryggvasson forcibly converted Sigurð Hlöðvisson, earl of Orkney, in 995, but that he quickly apostasised once Óláf’s threat was removed, and it was not until 1048 that Sigurð’s son orfinn made Christianity a matter of state policy with a pilgrimage to Rome and the establishment of a permanent bishopric at Birsay (Crawford 1983; Barrett et al 2000). Thus, while Christianity may well have been a weapon in struggles between different elite factions within the earldom much earlier, it was only at the very end of the Viking Age that it is likely to have directly influenced earldom policy. A second point is that the earls of Orkney, for all their power, were earls rather than kings, and that the production of coinage, with rare exceptions (especially within the Frankish kingdoms) was seen as a royal prerogative. Although, according to saga tradition, orfinn was happy to defy the kings of Norway in his younger days, by the time that he settled down as a responsible ruler and established the bishopric, he was apparently more willing to accept Norwegian overlordship. It is also notable that this period coincides with the establishment of a lasting coinage in Norway, examples of which have been found in Scandinavian Scotland (see above). On this interpretation, the failure of the earls of Orkney to fit the pattern in one respect thus fits the general pattern in other ways.

CONCLUSION At the beginning of the Viking Age, Scandinavia already had contacts in a number of directions, and with a number of peoples. Many of these, including the Byzantine Empire and the Arabic caliphates in the east, and the Anglo-Saxon and Frankish kingdoms in the west, already had well-established coin-based economies, even though the latter societies especially were not fully monetised by modern standards. Scandinavia already had two forms of precious-metal economy, one based on the ostentatious display of wealth, the



other on bullion as a medium for exchange and for the storage of wealth. The inhabitants of Scandinavia had been familiar with imported coinage since the Roman Iron Age, but had never chosen to issue coins of their own, with the debatable exception of the so-called Wodan-Monster coins from Ribe in the early to mid-eighth century. The use of precious metal for ostentatious display in a status economy survived the Viking Age (and one may argue that it still survives today), but with a reduced importance within society. At the beginning of the Viking Age, possession of great wealth and the ability to distribute it to followers was closely linked to this sort of display, although even then wealth could take other important forms, such as land and livestock. By the end of the Viking Age, however, a fundamental change had taken place in the other strand of the precious-metal economy, and for monetary purposes the bullion economy had largely been replaced in many areas by a coin-based economy. This process was gradual, and took place at different speeds in different areas. It was also by no means complete by the end of the Viking Age. Nevertheless, by the mid-eleventh century, Scandinavian rulers in Denmark, Norway, Sweden, England and Ireland had all begun to issue coinage. This meant a change in the monetary role of silver, as coins replaced bullion as the preferred means for both exchange and the storage of wealth. At the same time, however, the adoption of coinage also reflected an important change in the nature of royal power. While before, wealth and power was simply expressed in the status economy, now it was displayed even more clearly by the ability to control the coinage. Enforcing the use of locally controlled coinage as a standard of value showed power, as did the ability of Harald harð ráð i to debase the coinage systematically and, largely, get away with it. The rulers of late VikingAge Scandinavia may not have had the degree of regulatory authority over the coinage that their Anglo-Saxon contemporaries enjoyed, but they were moving in that direction, and this represents a substantial change from the situation 250 years earlier. This reflects a broader change in the nature of royal power. The expansion during the Viking Age brought external contacts, and with those contacts came ideas. Particularly important was a concept of kingship broadly shared at that time by most rulers in western Europe. This was a form of Romanised Christian kingship in which Church and state worked closely together, and provided a structured and law-bound society, with a literate machinery of government, and the social and political structure moved from an emphasis on lordship over people into lordship over territories. As Thomas Lindkvist (2003a) put it, ‘Ideologically Christianity and its clergy provided new and more effective ways of legitimising kingship.’ Issuing coinage was simply one facet of this. The Roman element in this was as important as the Christianity, as demonstrated by the fact that Christian rulers in Ireland, Scotland and Wales, with fewer links to Rome, did not fit the same pattern as Anglo-Saxon and continental rulers, despite having in many ways (like the Scandinavians) a comparable level of material culture. Although Ireland, Scotland and Wales therefore presented a slightly different model of Christian kingship, the more Romanised Christian kingship represented the European mainstream, and it was the latter model that the newly Christianised Scandinavian kings sought to emulate. Thus, when Scandinavian rulers in York or Dublin or Sigtuna chose to introduce coinage, they were not simply facilitating economic development within their respective kingdoms, nor were they simply finding a new means of expressing their royal power to their subjects, although they were doing both things. At the same time, however, they were sending a strong message to the rulers of other Christian kingdoms that they were to be



regarded as kings on the same terms as those whom they now regarded as their peers. Scandinavia remained geographically remote, but the differences between Scandinavian society and the rest of Europe were steadily diminishing by the end of the Viking Age, and those characteristics (economic or otherwise) that allow us to distinguish the ‘Vikings’ from their victims at the beginning of the Viking Age had largely disappeared by the end. The chapters in this volume clearly demonstrate that there was no single ‘silver economy’ in the Viking Age, any more than there was a single group that can be categorised as ‘Vikings’. Instead, there were many different silver economies, which can be distinguished both chronologically and geographically. Several of these can be considered ‘monetary’ economies, including some which were not based primarily, if at all, on coinage. The volume as a whole provides a warning, if one were needed, against a single monolithic view of the Viking Age, and equally against a monolithic view of economic structures in north-western Europe in the early Middle Ages. Nevertheless, it does present some recurrent themes in monetary development. The combination of archaeology, anthropology, history and numismatics brings a broader perspective than any one of these disciplines could present by itself, which was my intention and hope when the symposium which led to this volume was first proposed. At the same time, these collected chapters show most clearly that there are many more questions still to be asked.

ACKNOWLEDGEMENTS I am grateful to Marion Archibald, Mark Blackburn, Jens Christian Moesgaard and James Graham-Campbell, who all made helpful comments on earlier drafts of this paper. That should not be taken, however, to suggest that they necessarily share all of my interpretations, and any factual errors remain, of course, my own responsibility.

References Abels, R (2003), ‘Alfred the Great, the micel hæð en here and the viking threat’, in T Reuter (ed), Alfred the Great, Papers from the Eleventh-Century Conferences, Aldershot, Ashgate Publishing Ltd, 265–280 Abrams, L (2001a), ‘The conversion of the Danelaw’, in J Graham-Campbell, R Hall, J Jesch and D N Parsons (eds), Vikings and the Danelaw. Select Papers from the Proceedings of the Thirteenth Viking Congress, Oxbow Books, Oxford, 31–44 Abrams, L (2001b), ‘Edward the Elder’s Danelaw’, in N J Higham and D H Hill (eds), Edward the Elder, 899–924, Routledge, London/New York, 128–143 Andersen, P S (1991), ‘When was regular, annual taxation introduced in the Norse islands of Britain? A comparative study of assessment systems in north-western Europe’, Scandinavian Journal of History 16, 56–68 Anderssen, H H (1998), Danevirke og Kovirke. Arkæologiske undersøgelser 1861–1993, Moesgård Andersson, H (2003), ‘Urbanisation’, in K Helle (ed), The Cambridge History of Scandinavia, Vol 1. Prehistory to 1520, Cambridge University Press, Cambridge, 312–342 Andrén, A (1983), ‘Städer och kungamakt – en studie i Danmarks politiska geografi före 1230’, Scandia 49/1, 31–76



Archibald, M M (1990), ‘Pecking and bending: the evidence of British finds’, in K Jonsson and B Malmer (eds), Sigtuna Papers. Proceedings of the Sigtuna Symposium on Viking-Age Coinage, 1–4 June 1989, (Commentationes de Nummis Saeculorum IX–XI in Suecia repertis Nova Series 9, Stockholm, 11–24 Archibald, M M (1991), ‘Mot strømmen: Myntbevegelse fra Skandinavia til de Britiske øyer i Vikingtiden’, NNF-NYTT no. 1, 13–22 Archibald, M M (1998), ‘Two 9th-century Viking weights found near Kingston, Dorset’, BNJ 68, 11–20 Armstrong, S (1998), ‘Carolingian coin hoards and the impact of the Viking raids in the 9th century’, NC 158, 131–164 Axboe, M (1999), ‘Towards the Kingdom of Denmark’, in T Dickinson and D Griffiths (eds), Anglo-Saxon Studies in Archaeology and History 10, 109–118 Barrett, J et al (2000), ‘What was the Viking Age and when did it happen? A view from Orkney’, Norwegian Archaeological Review 33:1, 3–39 Bateson, J D (1997), Coinage in Scotland, Spink, London Becker, C J (1981), ‘The coinage of Harthacnut and Magnus the Good at Lund, c.1040–c.1046’, in C J Becker (ed), Studies in Northern Coinages of the eleventh Century, Det Kongelige Danske Vindenskabernes Selskabs Historisk-Filosofiske Skrifter 9:4, København, 119–173 Bendixen, K (1981), ‘The currency in Denmark from the beginning of the Viking Age until c.1100’, in M A S Blackburn and D M Metcalf (eds), Viking Age Coinage in the Northern Lands. The Sixth Oxford Symposium on Coinage and Monetary History (BAR International Series 122), Oxford, 405–418 Besly, E (2006), ‘Few and far between: mints and coins in Wales to the middle of the thirteenth century’, in B Cook & G Williams (eds), Coinage and History in the North Sea World c. 500–1250: Essays in Honour of Marion Archibald, Leiden, Brill, 701–720 Blackburn, M A S (1985–86), ‘Two imitations of Charlemagne’s Dorestad coinage – strays from the Krinkberg hoard?’, NNÅ, 6–10 Blackburn, M A S (1989), ‘The Ashdon (Essex) Hoard and the currency of the Southern Danelaw in the late ninth century’, BNJ 59, 13–38 Blackburn, M A S (1999), ‘An 8th-century coin from the Glebe Field’, in M O Carver et al, Tarbert Discovery Programme Bulletin 4 (online publication, 1999 for 1998). Blackburn, M A S (2001), ‘Expansion and control: aspects of Anglo-Scandinavian minting south of the Humber’, in J Graham-Campbell, R Hall, J Jesch and D N Parsons (eds), Vikings and the Danelaw. Select Papers from the Proceedings of the Thirteenth Viking Congress, Oxbow Books, Oxford, 125–155 Blackburn, M A S (2002), ‘Finds from the Anglo-Scandinavian site of Torksey, Lincolnshire’, in B Paskiewicz (ed), Moneta Mediævalis: Studia numizmatyczne i historyczne ofiarowne Profesorowi Stanislawowi Suchodolsiemu w 65. roznice˛urodzin, Instytut Archeologii i Etnologii Polskiej Akademii Nauk, Warszawa, 89–101 Blackburn, M A S (2003), ‘Alfred’s coinage reforms in context’, in T Reuter (ed), Alfred the Great, Ashgate, Aldershot, 199–218 Blackburn, M (2004), ‘The coinage of Scandinavian York’, in R A Hall et al, Anglo-Scandinavian York (Archaeology of York, Anglo-Scandinavian York 8/4), York, 325–349 Blackburn, M and Bonser, M (1986), ‘Single finds of Anglo-Saxon and Norman coins, 3’, BNJ 56, 64–101 Blunt, C E (1961), ‘The coinage of Offa’, in R H M Dolley (ed), Anglo-Saxon Coins: Studies presented to F.M. Stenton, Methuen & Co Ltd, London, 39–62 Blunt, C E (1969), ‘The St Edmund memorial coinage’, Proceedings of the Suffolk Institute of Archaeology 31, 234–255 Blunt, C E, Lyon, C S S and Stewart, B H I H (1989), Coinage in 10th-Century England, British Academy/Oxford University Press, Oxford Boon, G (1986), Welsh Coin Hoards, National Museum of Wales, Cardiff



Bornholdt, K A (1999), ‘Myth or mint? The evidence for a Viking-age coinage in the Isle of Man’, in P Davey (ed), Recent Archaeological Research on the Isle of Man (BAR, British Series 278), Oxford, 199–220 Brooks, N P and Graham-Campbell, J (2000), ‘Reflections on the Viking-Age silver hoard from Croydon, Surrey’, in N P Brooks, Communities and Warfare 700–1400, Hambledon Continuum, London, 69–92 Carver, M O H (ed) (1992), The Age of Sutton Hoo, Boydell, Woodbridge Chick, D (1997), ‘Towards a chronology for Offa’s coinage: an interim study’, Yorkshire Numismatist 3, 47–64 Clarke, H and Ambrosiani, B (1995), Towns in the Viking Age, 2nd ed, Leicester University Press, London / New York Clarke, H B (1998), ‘Proto-towns and towns in Ireland and Britain in the 9th and 10th centuries’, in H B Clarke et al (eds), Ireland and Scandinavia in the Early Viking Age, Four Courts Press, Dublin, 331–380 Coupland, S (1985–86), ‘Carolingian coinage and Scandinavian silver’, NNÅ, 11–32 Coupland, S (1988), ‘Dorestad in the 9th century: the numismatic evidence’, Jaarboek voor Munt-en Penningkunde 75, 5–26 Coupland, S (1998), ‘From poachers to gamekeepers: Scandinavian warlords and Carolingian kings’, Early Medieval Europe 7:1, 85–114 Crawford, B E (1983), ‘Birsay and the early Earls and Bishops of Orkney’, in Birsay: A Centre of Political and Ecclesiastical Power (⳱Orkney Heritage 2), 97–118 Crawford, B E (1987), Scandinavian Scotland, Leicester University Press, Leicester Crawford, B E (1993), ‘Norse earls and Scottish bishops in Caithness: a clash of cultures’, in C E Batey et al (eds), The Viking Age in Caithness, Orkney and the North Atlantic, Edinburgh University Press, Edinburgh, 129–147 Dolley, R H M (1965), Viking Coins of the Danelaw and of Dublin, British Museum, London Dolley, R H M (1966), Sylloge of Coins of the British Isles, Vol 8. The Hiberno-Norse Coins in the British Museum, London Dolley, M (1976), ‘A Hiberno-Manx coinage of the 11th century’, NC 136, 75–84 Duczko, W (1998), ‘Viking Age Scandinavia and Islam: an archaeologist’s view’, in E Piltz (ed), Byzantium and Islam in Scandinavia. Acts of a Symposium at Uppsala University, June 15–16 1996, Jonsered, 107–115 Duffy, S (1992), ‘Irishmen and Islesmen in the Kingdoms of Dublin and Man, 1052–1171’, Ériu 13, 93–133 Dumas, F (1971), Le trésor de Fécamp, Bibliothèque Nationale, Paris Dykes, D W (1976), Anglo-Saxon Coins in the National Museum of Wales, National Museum of Wales, Cardiff Gaimster, M (1991), ‘Money and media in Viking Age Scandinavia’, in R Samson (ed), Social Approaches to Viking Studies, Glasgow, 113–122 Graham-Campbell, J A (1975–76), ‘The Viking-age silver and gold hoards of Scandinavian character from Scotland’, PSAS 107, 114–134 Graham-Campbell, J (1980), Viking Artefacts: a Select Catalogue, British Museum Publications, London Graham-Campbell, J (1989), ‘The coinless hoard’, in H Clarke and E Schia, Coins and Archaeology (BAR, International series 556), Oxford, 53–61 Graham-Campbell, J (1992), ‘The Cuerdale hoard: comparisons and context,’ in J GrahamCampbell (ed), Viking Treasure from the North West: the Cuerdale hoard in its Context (National Museums and Galleries on Merseyside Occasional Papers, Liverpool Museum No. 5), Liverpool, 107–115 Graham-Campbell, J (1993), ‘The northern hoards of Viking-age Scotland’, in C E Batey et al (eds), The Viking Age in Caithness, Orkney and the North Atlantic, Edinburgh University Press, Edinburgh, 173–186



Graham-Campbell, J (1995), The Viking-Age Gold and Silver of Scotland, National Museums of Scotland, Edinburgh Graham-Campbell, J (1999), ‘Rings and things: some observations on the Hon hoard’, in G Fellows-Jensen and N Lund (eds), Beretning fra attende tværfaglige vikingesymposium, Aarhus, 53–64 Graham-Campbell, J (2001a), ‘The dual economy of the Danelaw. The Howard Linecar Memorial Lecture 2001’, BNJ, 49–59 Graham-Campbell, J (2001b), ‘The northern hoards: from Cuerdale to Bossall/Flaxton’, in N J Higham and D H Hill (eds), Edward the Elder, 899–924, Routledge, London / New York, 212–229 Graham-Campbell, J and Batey, C E (1998), Vikings in Scotland: an Archaeological Survey, Edinburgh University Press, Edinburgh Grierson, P (1959), ‘Commerce in the Dark Ages: a critique of the evidence’, TRHS 9, 123–140 Grierson, P (1991), The Coins of Medieval Europe, Seaby, London Grierson, P and Blackburn, M A S (1986), Medieval European Coinage, Vol 1, Cambridge University Press, Cambridge Gullbekk, S H (1991), ‘Some aspects of coin import to Norway and coin circulation in the late Viking Age’, NNÅ, 63–88 Gullbekk, S H (1992–93), ‘Renovatio monetae i Norge i middelalderen’, NNÅ, 52–87 Gustin, I (1997), ‘Islam, merchants, or king? Who was behind the manufacture of Viking Age weights?’, in H Andersson, L Ersgaard and P Carelli (eds), Visions of the Past. Trends and Traditions in Swedish Medieval Archaeology (Lund Studies in Medieval Archaeology 19; Riksantikvarieämbetet Arkeologiska Undersökningar Skrifter, N. 24), Central Board of National Antiquities, Stockholm, 163–177 Hårdh, B (1977–78), ‘Trade and money in Scandinavia in the Viking Age,’ Meddelanden från Lunds Universitets Historiska Museum, 157–171 Hårdh, B (1996), Silver in the Viking Age. A Regional-Economic Study (⳱Acta Archaeologica Lundensia Series in 8°, No. 25), Stockholm Hammarberg, I and Rispling, G (1985), ‘Graffiter på vikingatida mynt’, Hikuin 11 (Festschrift for Brita Malmer), 63–78 Hart, C (1995), ‘The Aldewerke and minster at Shelford, Cambridgeshire’, Anglo-Saxon Studies in Archaeology and History 8, 43–68 Hauberg, P (1900), Myntforhold og Udmyntninger i Danmark indtil 1146, Det Kongelige Danske Vindenskabernes Selskabs Historisk-Filosofiske Skrifter 9:4, København Hauken, Å D (1991), ‘Gift-exchange in early Iron Age Norse society’, in R Samson (ed), Social Approaches to Viking Studies, Cruithne Press, Glasgow, 105–112 Heidinga, A (1999), ‘The Frisian achievement in the first millennium AD’, in T Dickinson and D Griffiths (eds), Anglo-Saxon Studies in Archaeology and History 10, Oxford, 11–16 Hovén, B E (1981), ‘On oriental coins in Scandinavia’, in M A S Blackburn and D M Metcalf (eds), Viking Age Coinage in the Northern Lands. The Sixth Oxford Symposium on Coinage and Monetary History (BAR, International Series 122), Oxford, 119–128 Jensen, J S (ed) (1995), Tusentallets Danske Mønter fra Den konglige Mønt- og Medaillesamling. Danish coins from the 11th century in The Royal Collection of Coins and Medals, Nationalmuseet, København Jensen, J S et al (1992), Danmarks middelalderlige skattefund, c.1050– c.1550. Denmark’s Medieval Treasure Hoards, c.1050– c.1550, Nationalmuseet, København Jensen, J S and Kromann, A (1998), ‘Cufic coins in Denmark’, in E Piltz (ed), Byzantium and Islam in Scandinavia. Acts of a Symposium at Uppsala University, June 15–16 1996, Jonsered, 71–76 Jensen, S (1991), ‘Dankirke – Ribe: Fra handelsgård til handelsplads’, in P Mortensen and B M Rasmussen (eds), Fra Stamme til Stad i Danmark, 2. Høvdingesamfund og Kongemagt (Jysk Arkæologisk Selskabs Skrifter XXII: 2), Aarhus, 73–88



Johanek, P (1999), ‘Merchants, markets and towns’, in T Reuter (ed), The New Cambridge Medieval History III, c.900 – c.1024 Cambridge University Press, Cambridge, 64–94 Jonsson, K (1995), ‘Från utländsk metall till inhemskt mynt’, in K Jonsson, U Nordlind and I Wiséhn (eds), Myntingen i Sverige, 995–1995, Numismatiska Meddelanden, Stockholm, 43–62 Jørgensen, L and Vang Petersen, P (1998), Guld, Magt og Tro: Danske skattefund fra oldtid of middelalder. Gold, Power and Belief (Danish Gold Treasures from Prehistory and the Middle Ages), København Kilger, C (1995), ‘Myntfynd och den vikingatida silverhandeln’, in K Jonsson, U Nordlind and I Wiséhn (eds), Myntingen i Sverige, 995–1995, Numismatiska Meddelanden, Stockholm, 27–42 Kilger, C (2003), ‘Pecks och böjningar på vikingatida mynt’, NNUM (No 1–2 Februar, 2003), 3–12 Kilger, C (2006), ‘Silver-handling traditions during the Viking Age – Some observations and thoughts on the phenomenon of pecking and bending’, in Cook and Williams (see above), 449–446. Krag, C (1989), ‘Norge som odel i Harald Hårfagres ætt’, Historisk Tidsskrift 68, 288–301 Krag, C (2003), ‘The early unification of Norway’, in K Helle (ed), The Cambridge History of Scandinavia, Vol. I, Prehistory to 1520, Cambridge, Cambridge University Press, 184–201 Kramer, E, Stoumann, I and Greg, A (eds) (2000), Kings of the North Sea, AD 250–850, The Hague Kromann, A (1990), ‘The latest Cufic coins finds from Denmark’, in K Jonsson and B Malmer (eds), Sigtuna Papers. Proceedings of the Sigtuna Symposium on Viking-Age Coinage, 1–4 June 1989 (CNS IX–XI, New Series 6), Stockholm/London, 183–196 Kruse, S E (1988), ‘Ingots and weight units in Viking Age silver hoards’, World Archaeology 20:2, 285–301 Kruse, S E (1992), ‘Late Saxon balances and weights from England’, Med Arch 36, 67–95 Kruse, S E (1993), ‘Silver storage and circulation in Viking-age Scotland’, in C Batey et al (eds), Caithness, Orkney and the North Atlantic, Edinburgh University Press, Edinburgh, 187–203 Lindkvist, T (2003a), ‘Early political organisation. (a) Introductory survey’, in Helle (ed), Cambridge History (see above), 160–167 Lindkvist, T (2003b), ‘Kings and Provinces in Sweden’, in Helle (ed), Cambridge History (see above), 221–34 Lund, N (1980), ‘Svenskevældet i Hedeby’, Aarbøger for Nordisk Oldkyndighed og Historie, 114–125 Lund, N (1995), ‘Scandinavia, c.700–1066’, in R McKitterick (ed), The New Cambridge Medieval History II, c.700–c.900, Cambridge University Press, Cambridge, 202–227 Lund, N (1996), Lið, leding og landeværn, Roskilde Museums Forlag, Roskilde Lyon, C S S and Stewart, B H I H (1961), ‘The Northumbrian Viking coins in the Cuerdale hoard’, in R H M Dolley (ed), Anglo-Saxon Coins: Studies presented to F.M. Stenton, Methuen & Co Ltd, London, 96–121 Macgregor, L J and Crawford, B E (eds) (1987), Ouncelands and Pennylands, St John’s House Papers, St Andrews Malmer, B (1966), Nordiska mynt fore år 1000 (Acta Archaeologica Lundensia, Series in 8°, no. 4), Lund Malmer, B (1981), ‘Imitations of Byzantine Milaresia found in Sweden’, in C S Becker (ed), Studies in Northern Coinage of the 11th Century, Det Kongelige Danske Vindenskabernes Selskabs Historisk-Filosofiske Skrifter 9:4, Copenhagen, 9–28 Malmer, B (1989a), The Sigtuna Coinage c.995–1005 (Commentationes de Nummis Saeculorum IX–XI in Suecia repertis Nova Series 4), Stockholm/London Malmer, B (1989b), ‘A note on the coinage of Sigtuna, at the time of Anund Jacob,’ in Festskrift til Lars O. Lagerkvist (⳱ Numismatiska Meddelanden 37), Stockholm, 259–262



Malmer, B (1995), ‘Från Olof till Anund. Ur sigtunamyntingens historia’, in K Jonsson, U Nordlind and I Wiséhn (eds), Myntingen i Sverige, 995–1995, Numismatiska Meddelanden, Stockholm, 9–26 Malmer, B (1997), The Anglo-Scandinavian Coinage c.995–1020 (Commentationes de Nummis Saeculorum IX–XI in Suecia repertis Nova Series 9), Stockholm Malmer, B (2002), ‘Karolingisk mission och nordiska mynt’, NNUM (Nos. 9–10, December 2002), 182 Malmer, B (2004), ‘Kristna symboler på danska mynt’, in N Lund (ed), Kristendommen i Danmark før 1050, Roskilde Museums Forlag, Roskilde, 75–85 Metcalf, D M (1993–94), Thrymsas and Sceattas in the Ashmolean Museum, Oxford (RNS Special Publication 276), Vols 1–3, London Metcalf, D M (1995), ‘The monetary significance of Scottish Viking-age coin hoards, with a short commentary’, in J Graham-Campbell, The Viking-Age Gold and Silver of Scotland, National Museums of Scotland, Edinburgh, 16–24 Metcalf, D M (1996a), ‘Viking Age numismatics, 2. Coinage in the northern lands in Merovingian and Carolingian times’, NC 156, 295–335 Metcalf, D M (1996b), ‘Kan Hedeby mønter også præget Ribe og Birka?’, NNUM 5, 74–77 Mikkelsen, E (1998), ‘Islam and Scandinavia during the Viking Age’, in E Piltz (ed), Byzantium and Islam in Scandinavia. Acts of a Symposium at Uppsala University, June 15–16 1996, Jonsered, 39–52 Moltke, E (1985), ‘Det svenske Hedebyrige og Danmarks samling’, Aarbøger for Nordisk Oldkyndighed og Historie, 16–28 Näsman, U (1999), ‘The ethnogenesis of the Danes and the making of a Danish kingdom’, in T Dickinson and D Griffiths (eds), Anglo-Saxon Studies in Archaeology and History 10, 1–10 Nelson, J L (1997), ‘The Frankish Empire’, in P H Sawyer (ed), The Oxford Illustrated History of the Vikings, Oxford University Press, Oxford, 19–47 Nightingale, P (1983), ‘The ora, the mark, and the mancus: weight-standards and the coinage in eleventh-century England. Part 1’, NC 143, 248–257 Nightingale, P (1984), ‘The ora, the mark, and the mancus: weight-standards and the coinage in eleventh-century England. Part 2’, NC 144, 234–248 Noonan, T S (1993), ‘The Vikings in the East: coins and commerce’, in B Ambrosiani and H Clarke (eds), Developments around the Baltic and the North Sea in the Viking Age (Twelfth Viking Congress; Birka Studies 3), Stockholm, 215–235 Noonan, T S (1999), ‘European Russia, c.500– c.1050’, in T Reuter (ed), The New Cambridge Medieval History III, c.900 – c.1024, Cambridge University Press, Cambridge, 487–513 Olsen, O (1999), Da Danmark blev til, København Randsborg, K (1980), The Viking Age in Denmark, Duckworth Press, London Redknap, M (2000), Vikings in Wales: an Archaeological Quest, National Museum and Gallery of Wales, Cardiff Reuter, T (1985), ‘Plunder and tribute in the Carolingian Empire’, TRHS 5th ser 35, 75–94 Reuter, T (1990), ‘The end of Carolingian military expansion’, in P Godman and R Collins (eds), Charlemagne’s Heir: New Perspectives on the Reign of Louis the Pious (814–840), Clarendon Press, Oxford), 391–405 Richards, J D (2001), ‘Boundaries and cult centres: Viking burial in Derbyshire’, in J GrahamCampbell, R Hall, J Jesch and D N Parsons (eds), Vikings and the Danelaw. Select Papers from the Proceedings of the Thirteenth Viking Congress, Oxbow Press, Oxford, 97–104 Samson, R (1991a), ‘Economic anthropology and Vikings’, in R Samson (ed), Social Approaches to Viking Studies, Cruithne Press, Glasgow, 87–96 Samson R (1991b), ‘Fighting with silver: rethinking trading, raiding, and hoarding’, in R Samson (ed), Social Approaches to Viking Studies, Cruithne Press, Glasgow, 123–133 Samson, R (1999), ‘Illusory emporia and mad economic theories’, in M Anderton (ed), AngloSaxon Trading Centres: Beyond the Emporia, Cruithne Press, Glasgow, 76–90



Sawyer, B and Sawyer, P H (2003), ‘Scandinavia enters Christian Europe’, in K Helle (ed), The Cambridge History of Scandinavia, Vol 1. Prehistory to 1520, Cambridge University Press, Cambridge, 147–159 Sawyer, P H (1986), ‘Anglo-Scandinavian trade in the Viking Age and after’, in M A S Blackburn (ed), Anglo-Saxon Monetary History, Leicester University Press, Leicester, 185–200 Sawyer, P H (1988), The Making of Sweden, Alingsås Sawyer, P H (1989), ‘Knut, Sweden and Sigtuna’, in S Tesch (ed), Avstamp – för en ny Sigtunaforskning, Sigtuna, 88–93 Sawyer, P H (1990), ‘Coins and commerce’, in K Jonsson and B Malmer (eds), Sigtuna Papers. Proceedings of the Sigtuna Symposium on Viking-Age Coinage, 1–4 June 1989 (Commentationes de Nummis Saeculorum IX–X, in Suecia repertis, New Series 6), Stockholm/London, 283–288 Sawyer, P H (1995), ‘The last Scandinavian kings of York’, NH 31, 39–44 Sawyer, P H (1999), ‘Engelsk innflytelse på den norske rikssamligen / English influence on the development of the Norwegian kingdom’, in O Skevik (ed), Kongemøte på Stiklestad: Foredrag fra seminar om kongedømmet i vikingtid og tidlig middelalder, Stiklestad, 97–104 Sheehan, J (2000), ‘Ireland’s early Viking-Age silver hoards: components, structure, and classification’, in S Stummann Hansen and K Randsborg (eds), Vikings in the West (Acta Archaeologica 71), 49–63 Simonsen, J B (1981), Vikingerne ved Volga. Ibn Fadlans rejsebeskrivelse, Højbjerg Skaare, K (1966), ‘Die karolingischen Münzfunden in Skandinavien und der Schatzfund von Hon’, Hamburger Beiträge zur Numismatik, 393–408 Skaare, K (1976), Coins and Coinage in Viking-Age Norway, Universitets Forlag, Oslo, Bergen and Tromsö Skaare, K (1981), ‘Coins and coinage in Viking-Age Trondheim’, in M A S Blackburn and D M Metcalf (eds), Viking Age Coinage in the Northern Lands. The Sixth Oxford Symposium on Coinage and Monetary History (BAR, International Series 122), Oxford, 479–486 Skaare, K (1988), ‘Der Schatzfund von Hon und seine Münzen’, Commentationes Numismaticae 1988. Festgabe für Gert und Vera Hatz zum 4. Januar 1988 dargebrach, Hamburg, 51–61 Skovgaard-Petersen, I (2003), ‘The making of the Danish Kingdom’, in K Helle (ed), The Cambridge History of Scandinavia, Vol 1. Prehistory to 1520, Cambridge University Press, Cambridge, 168–83 Sperber, E (1996), Balances, Weights and Weighing in Ancient and Early Medieval Sweden (Stockholm University, Theses and Papers in Scientific Archaeology 2), Stockholm Steuer, H (1987), ‘Gewichtsgeldwirtschaften im frühgeschichtlichen Europa – Feinwaagen und Gewichte als Quellen zur Währungsgeschichte’, in K Düwel, H Siems, and D Timpe (eds), Untersuchungen zu Handel und Verkehr der vor- und frühgeschichtlichen Zeit in Mittel- und Nordeuropa, Teil IV. Der Handel der Karolinger- und Wikingerzeit (Abhandlungen der Academie der Wissenschaften in Göttingen Philologisch-Historische Klasse, Dritte Folge, No. 156), 405–527 Steuer, H (1997), Waagen und Gewichte aus dem mittelalterlichen Schleswig. Funde des 11. Bis 13. Jahrhunderts aus Europa als Quellen zur Handels- und Währungsgeschichte, Köln Stevenson, R B K (1986), ‘The Anglo-Saxon silver penny and its context’, in C D Morris and N Emery, ‘The chapel and enclosure on the Brough of Deerness, Orkney: survey and excavations, 1975–77’, PSAS 116, 301–74, at 339–41 Stewart, I (1967), ‘The St Martin coinage of Lincoln’, BNJ 36, 46–54 Stewart, I and Lyon, S (1992), ‘Chronology of the St Peter coinage’, Yorkshire Numismatist 2, 45–73 Swanton, M J (ed and trans) (1990), An Anglo-Saxon Chronicle, Weidenfeld & Nicholson, Exeter Talvio, T (1998), ‘Islamic Coins found in Finland’, in E Piltz (ed), Byzantium and Islam in Scandinavia. Acts of a Symposium at Uppsala University, June 15–16 1996, Jonsered, 77–84



Thomson, W P L (1987), History of Orkney, Mercat Press, Edinburgh Thomson, W P L (2002), ‘Ouncelands and Pennylands in the West Highlands and Islands’, Northern Scotland 22, 27–43 Vestergaard, E (1991), ‘Gift-giving, hoarding, and outdoings’, in R Samson (ed), Social Approaches to Viking Studies, Cruithne Press, Glasgow, 97–104 Wallace, P (1987), ‘The economy and commerce of Viking Age Dublin’, in K Düwel, H Siems and D Timpe (eds), Untersuchungen zu Handel und Verkehr der vor- und frühgeschichtlichen Zeit in Mittel- und Nordeuropa, Teil IV. Der Handel der Karolinger- und Wikingerzeit (⳱Abhandlungen der Academie der Wissenschaften in Göttingen Philologisch-Historische Klasse, Dritte Folge, No. 156), 200–245 Wamers, E (1995), ‘The symbolic significance of the ship-graves at Haidaby and Ladby’, in O Crumlin-Pedersen and B Munch Thye (eds), The Ship as Symbol (PNM Studies in Archaeology and History 1), Copenhagen, 148–159 Wamers, E (2002), ‘The 9th century Danish-Norwegian conflict: maritime warfare and state formation’, in A Nørgård Jørgensen, J Pind, L Jørgensen and B Clausen (eds), Maritime Warfare in Northern Europe: Technology, Organisation, Logistics and Administration 500 BC–1500 AD (PNM Studies in Archaeology and History 6), Copenhagen, 237–248 Warner, R (1975–76), ‘Scottish silver arm-rings: an analysis of weights’, PSAS 107, 136–143 Welch, M (2000), ‘Trade and trading places around the North Sea’, in E Kramer, I Stoumann and A Greg (eds), Kings of the North Sea, AD 250–850, The Hague, 67–78 Wiechmann, R (1996), Edelmetalldepots der Wikingerzeit in Schleswig-Holstein. Vom ‘Ringbrecher’ zur Münzwirtschaft (Offa-Bücher 77), Neumünster Williams, D G E (⳱Williams, G) (1996), Land Assessment and Military Organisation in the Norse Settlements in Scotland, c.900–1266 AD, unpublished PhD thesis, University of St Andrews Williams, G (1999), ‘Anglo-Saxon and Viking coin weights’, BNJ 69, 19–36 Williams, G (2001a), ‘Mercian coinage and authority’, in M P Brown and C A Farr (eds), Mercia: an Anglo-Saxon Kingdom in Europe, Leicester University Press, London/New York, 211–228 Williams, G (2001b), ‘Coin brooches of Edward the Confessor and William I’, BNJ 71, 60–70 Williams, G (2001c), ‘Hákon Að alsteins fóstri: aspects of Anglo-Saxon kingship in 10th-century Norway’, in T R Liszka and L M Walker (eds), The North Sea World in the Middle Ages: Studies in the Cultural History of North-Western Europe, Four Courts Press, Dublin, 108–126 Williams, G (2002), ‘Ship-levies in the Viking Age – the methodology of studying military institutions in a semi-historical society’, in A Nørgård Jørgensen, J Pind, L Jørgensen and B Clausen (eds), Maritime Warfare in Northern Europe: Technology, Organisation, Logistics and Administration 500 BC–1500 AD (PNM Studies in Archaeology and History 6), Copenhagen, 293–308 Williams, G (2003), ‘The dabhach reconsidered: pre-Norse or post-Norse?’, Northern Studies 37, 17–34 Williams, G (2004), ‘Land assessment and the silver economy of Norse Scotland’ in G Williams and P A Bibire (eds), Sagas, Saints and Settlement, Brill, Leiden, 65–104 Williams, G (forthcoming a), ‘Monetary economy in Viking-age Scotland in the light of single finds’, in J C Moesgaard and H Horsnæs (eds), Single Finds – the Nordic Perspective (⳱ NNÅ 2000–02), Copenhagen Williams, G (forthcoming b), Viking Warfare and Military Organisation, Routledge, London Williams, G and Sharples, N (2003), ‘Et nytt myntfunn fra Olav Kyrre fra Hebridene’, NNUM (Nos. 3–4, Maj 2003), 55–56 Wilson, D M (1957), ‘An unpublished fragment from the Goldsborough hoard’, Antiquaries Journal 37, 72–73 Woolf, A (1998), ‘Eric Bloodaxe revisited’, NH 34, 189–193 Youngs, S M and Porter, V (2000), ‘No 84. Warton, Carnforth, Lancashire: small Viking hoard of Cufic coins and cut silver’, in R Bland (ed), Treasure Annual Report 1998–1999, London, 49–51

– 12 –

Reflections on ‘Silver Economy in the Viking Age’ James Graham-Campbell

Shortly after the London symposium on ‘Silver Economy in the Viking Age’, at which versions of the articles to be considered here were delivered and discussed, I gave a paper to the ‘Nittende tværfaglige vikingesymposium’, at the University of Aarhus, on the subject of ‘The power of silver’ (Graham-Campbell 2000). This developed one of the main themes that had been under consideration in London, although the starting point for this Danish paper was a general introduction to the subject based on an earlier lecture on the overall archaeological importance of Viking-Age silver hoards (Graham-Campbell 1982).

SOME REFLECTIONS ON ‘DISPLAY’ AND ‘BULLION’ ECONOMIES The theme developed in Aarhus was ‘the changing power of silver as it moved from the “display economy” into the “bullion economy” ’ (Graham-Campbell 2000, 13). In elaborating on this topic, for an interdisciplinary audience, it seemed appropriate to refer to some documentary sources, of differing reliability, which had been passed over by the contributors in London, no doubt because members of that particular gathering could have been expected to be familiar with them. All the same, their interest is such that they merit having attention drawn to them in this volume; indeed, the first source to be mentioned here is one also commented on by Gareth Williams in his concluding paper (see above, Chapter 11). This is the report by Ibn Fadlan of his meeting by the River Volga with a party of Rus traders, and their wives, in the 920s. Although his account of their dress and rituals is doubtless exaggerated in places (or at times lacking in understanding), it has the merit of being the evidence of an eye-witness and, as such, he provides us with a clear description of the ‘display economy’ in action: Each woman carries on her bosom a container made of iron, silver, copper, or gold – its size and substance depending on her man’s wealth … Round her neck she wears gold or silver rings; when a man amasses 10,000 dirhams he makes his wife one gold ring; when he has 20,000 he makes two; and so the woman gets a new ring for every



JAMES GRAHAM-CAMPBELL 10,000 dirhams her husband acquires, and often a woman has many of these rings. (Brøndsted 1965, 265)

This is ‘family wealth’ in action, but another important side of the ‘display economy’ is the affirmation of power through gift-giving, and here Viking-Age poetry can be brought into play as an additional form of evidence, as in this half-stanza of skaldic verse, preserved in Snorri’s Edda: The land-bold prince of Lund harms [gives away] joint-brands [ornaments for the arms]. I do not think the ruler’s men will lack Rhine’s rock [gold].

Addressed to a tenth-century Danish king, this praise poem is attributed to the Icelandic poet Einarr Skálaglamm (Jesch 2000, 25). A generous ruler is a ‘disburser of gold’, and as Judith Jesch went on to observe at the Aarhus seminar: The skaldic poems themselves make frequent reference to the payment of the poet, mostly in gold, or something else more valuable than the usual Viking Age currency of silver. Whether poets really were paid in gold rather than silver is hard to say, but that is at any rate what they claimed in their poems. (Jesch 2000, 35)

This makes all the more interesting an episode in the life of another tenth-century Icelandic poet, Eyvindr Finnsson, which is recounted in Snorri’s Heimskringla. Despite the lateness of this source, it serves well to illustrate the ‘bullion economy’, or rather the movement of silver between the two: ‘Of the Icelanders and Eyvind Skaldaspiller’ Eyvind composed a poem about the people of Iceland, for which they rewarded him by each bonde giving him three silver pennies, of full weight and white in the fracture. And when the silver was brought together at the General Thing, the people resolved to have it purified, and made into a shoulder-pin, and after the workmanship of the silver was paid, the shoulder-pin weighed some fifty marks [Editor’s note: 25 lbs]. This they sent to Eyvind; but Eyvind had the shoulder-pin broken into pieces, and with the silver he bought a farmstead for himself. (Foote 1961, 125–126)

The manner in which the silver was collected as coin, the concern for its purity, and the payment for it to be worked into a ‘shoulder-pin’, only for it to be converted into hacksilver for the purchase of land, are all matters of note and have been discussed elsewhere (Graham-Campbell 1982, 32–33; 2000, 13; Williams 2004, 80). The key points for note in the present context are (i) that, in the ‘display economy’, a large silver ornament was regarded as a suitable reward, but (ii) that such a prestigious artefact nevertheless remained a large lump of bullion, available for use as hack-silver in the ‘bullion economy’. On the other hand, it is far from certain that there were so many coins in circulation in tenth-century Iceland for the above procedure to have been carried out in practice (cf. Graham-Campbell 2005). Indeed, Gaimster (see Gaimster, in Chapter 7, p 126) makes the point that the Icelandic law on blood-money, Baugatal, translates literally as ‘the counting of rings’ (Gaimster 1991, 117–118). There was no particular need to pursue methodological matters any further at the Aarhus symposium, but it is appropriate here to refer back to a paper given at the ‘Coins and Archaeology’ meeting held in Norway (1988), when I was asked to consider ‘The



coinless hoard’. It was on this occasion that the concept of ‘passive’ and ‘active’ hoards was introduced (Graham-Campbell 1989, 54–55), the intention being to draw a distinction between those coinless hoards that consist exclusively of complete ornaments and those that consist exclusively (or mostly) of ingots and hack-silver. In other words, in this terminology, complete ornaments = ‘passive’ hoards (representing the ‘display economy’), and hack-silver, etc = ‘active’ hoards (representing the ‘bullion economy’). On the other hand, the story of Eyvindr’s reward demonstrates how one might readily become the other, and the distinction then drawn was ‘not intended to be a hard and fast one’, with the suggestion being made that, as the ‘bullion economy’ was developing, a ‘passive’ hoard might be thought of as being in an ‘intermediate’ or ‘pending’ category (Graham-Campbell 1989, 55), or in John Sheehan’s more elegant terminology, ‘potentially active’ (Sheehan 2001, 60). This ‘active/passive’ terminology has since been developed by Sheehan (see Sheehan, Chapter 9, in this volume), but it has perhaps now served its purpose, and it may be worth considering how one might move on from it on the basis of some of the contributions to this volume. The overall concept of a ‘display economy’ accords well with what Birgitta Hårdh, writing of Western Norway, has described elsewhere as: ‘a system where silver … had a social function … Here the precious metals are important as a means for giving gifts and forming alliances, to build up social positions in areas with politically unclear conditions’. This is the quotation aptly chosen by Sheehan (Chapter 9, p 149), from Hårdh’s seminal volume, Silver in the Viking Age: a Regional-Economic Study (Hårdh 1996, 178), for application to the situation pertaining in Ireland during the early Viking Age, with respect to his Class 1 hoards. This socio-political role of precious metal may also be extended to a religious and/or legal function, for it must surely have been the case that the ‘oath-rings’ encountered in the documentary sources would have been of silver, if not of gold. In ‘The power of silver’ (Graham-Campbell 2000, 7), I commenced by drawing attention, in this connection, to the passage in Landnámabók where it is stated that: A ring of at least two ounces should lie on the altar of every main temple. Every priest should have such a ring on his arm at all legal moots that he had to inaugurate himself. First it must be reddened in the blood of the cattle that he himself had sacrificed there. Every man who needed to take part in pleading before the court must first swear an oath on the ring and give the names of two or more witnesses. (Page 1995, 174)

Having observed that ‘arm-rings of two or more ounces of silver (that is at least 50 g) are common enough finds from the Viking Age’, it was noted (Graham-Campbell 2000, 7) that, Eyrbyggiasaga informs us, on the altar-like stone of the supposed temple-building at Hofstaðir, there ‘lay a penannular ring weighing twenty ounces’; this also served as an oath-ring of which it is said that ‘the temple priest had to wear it on his arm at all formal meetings’ (Page 1995, 29–30). It should be noted that the conventional translations of ‘temple’ and ‘priest’ are followed here without any implications as to the nature and use of Viking-Age cult buildings in Iceland (cf. Friðriksson 1994, 48–74). This increase in supposed weight of the ‘oath-ring’ (from two to 20 oz) can be taken to be one of those exaggerations to which such medieval Christian sources for Iceland’s pagan Viking past are inevitably liable, for ‘arm-rings weighing 0.5 kg are truly exceptional’, although the existence of ‘oath-rings’ themselves is not in any doubt, given



that the contemporary Anglo-Saxon Chronicle, for the year 876, records the Vikings swearing an oath ‘on the holy ring which they had never before done for any other people’ (Graham-Campbell 2000, 7; quoting this passage from Page 1995, 30).

SOME REFLECTIONS ON THE STUDY OF VIKING-AGE SILVER HOARDS When I commenced work on the Viking-Age silver hoards of Ireland, at the beginning of the 1970s (Graham-Campbell 2001, 50), the term ‘coin hoard’ was then in general use (as still today) both of hoards that consist solely of coins and of those that contain a mixture of coins and non-numismatic silver (whether ingots, ornaments and/or hacksilver). At that time, research was focussed on the coins themselves, so as to establish a chronology for the deposition of the coin hoards and also to enable the study of their geographical distribution, against an historical background (cf. Dolley 1966). My initial studies of ninth/tenth-century silver brooches from Ireland was directed in particular to those coin hoards that contain example of silver brooches and/or their fragments (whether or not found in Ireland), in an endeavour to refine the chronology for their manufacture. The term ‘mixed hoard’ seemed appropriate for the particular category of hoard then under discussion, which led to my use of the term ‘coinless hoard’ for the remainder. At that time, it was a matter of importance to highlight the existence of a substantial body of archaeological material, in the form of precious metal, which was not finding its place in the discussion of the distribution and use of silver during the Viking Age in Ireland, simply because coins were lacking from their contents (Graham-Campbell 1976). Today, of course, this point no longer needs to be laboured. This simplistic three-fold division of the hoard material does, however, still seem to me, at any rate, to remain a useful working-tool, for the preliminary stage of assessing the totality of the available evidence, even if a more sophisticated approach is essential for its detailed analysis, as has now been proposed by John Sheehan for Ireland. Sheehan’s classificatory system (see Sheehan, Chapter 9, in this volume) has previously been presented by him in greater detail elsewhere (Sheehan 2001, 57–59), although it has yet to be commented on in print. It is based on a fivefold division of 67 coinless and mixed hoards, based on the presence/absence of ornaments, ingots and hacksilver (Gaimster’s ‘special-purpose money’), regardless of whether they contain coins or not. There is therefore no place in this scheme for the 41 hoards of ninth/tenth-century date from Ireland containing ‘coins only’, given that it deliberately embraces only those hoards belonging to the ‘display economy’ (Sheehan Class 1) and to the ‘bullion economy’ (Sheehan Classes 2–5). The status of a ninth-century hoard consisting solely of coins (when Ireland was not coin-using), such as that from Mullaghboden, Co Kildare (deposited c 847), which is comprised entirely of ‘loot from Aquitaine’, is thus not addressed, although in this case it clearly cannot be considered as a ‘currency hoard’ (Dolley 1961; Graham-Campbell 1976, 48, 63). Sheehan’s scheme may be summarised as follows: Class 1: complete ornaments only Class 2: complete ingots only Class 3: complete ornaments and complete ingots only



Class 4: complete ingots and/or complete ornaments, with hack-silver Class 5: hack-silver only ‘Only’ is thus a relative term in this context, given that 15 of these hoards also contain coins, but then this element in their composition is seemingly not considered relevant to gaining ‘a more accurate impression of the degree of variation in those hoards which contain non-numismatic material’ (as above). Sheehan acknowledges, however, the complex role played by coins during the ninth and tenth centuries in Ireland, commenting that they ‘could variously be regarded as bullion, ornaments or money’. Gaimster’s approach to the classification of the Viking-Age hoards from two different parts of Scandinavia (as above, in Chapter 7, in this volume) differs somewhat from Sheehan’s in being based on a division of the silver into the following four categories: (i) coins (ii) bars and ingots (iii) hack-silver (iv) personal ornaments The information to be gained by including coins in such a study is succinctly stated by her (in part) as follows (pp 127–128): The presence of coins in the finds enabled a study of changes in the regional use of silver through time, but their varied provenance … also provides information about external contacts and the contexts of exchange. To what degree foreign coins circulated as a means of payment in Scandinavia is a difficult question, where features like peck-marks or the transformation of coins into pendants certainly suggest differences in the use of this medium.

It seems to me therefore that Sheehan, by omitting coins altogether from his classificatory system, might be considered to have thrown out the baby with the bathwater. The problem is therefore how best they might be integrated into his scheme. It is obviously possible to identify those coins that were converted into ornaments and thus to classify them accordingly; likewise, randomly cut or mutilated coins can readily be classified as a further type of hack-silver. But what then of complete coins (or, indeed, halves) in good condition? If not being treated as money, they are bullion and thus, in practice, constitute small ingots. However, ingots are specially manufactured (frequently by melting down coins) and thus represent a clear-cut category of bullion on their own. Given their light weight compared with ingots proper, it would probably be most practical for Sheehan to regard the relevant coins (other than those converted into ornaments) as a form of hacksilver, whether complete or fragmentary, when they occur in the 15 ‘mixed’ hoards which need to be accommodated in his classificatory system, while remaining alert to the numismatic strictures of Kristin Bornholdt Collins (see p 214), which necessitate her taking a diametrically opposed approach to the same material. Doubtless their two different avenues of exploration will lead to the emergence of some form of consensus in due course. Meanwhile, it remains as potentially misleading for an archaeologist to discuss the economic role of silver in Viking-Age Ireland (or Britain) without reference to coins and coin hoards, as it has been for numismatists to do so without reference to the ingots, ornaments and hack-silver, which form part of their ‘coin hoards’. It is for this very reason



that Mark Blackburn’s contribution to this volume, discussing the significance of gold in later Anglo-Saxon England (Chapter 5), makes such a valuable advance in that he combines numismatic and archaeological approaches to a single end, as he has earlier done by initiating debate on the nature of ‘the dual economy in the Danelaw’ (Blackburn 2001; 2002; Graham-Campbell 2001).

SOME REFLECTIONS ON THE VIKING-AGE SILVER HOARDS OF IRELAND It is hardly surprising therefore that the first critique of Sheehan’s approach to studying the ninth/tenth-century hoards from Ireland has come in the PhD thesis of Mark Blackburn’s research student, Dr Kristin Bornholdt Collins (2003), for which she researched the Viking-Age hoards from the Isle of Man. The following two passages reflect her current concerns (and those of some others, including myself): Sheehan demonstrated that in Ireland most of the coinless hoards and mixed hoards occur in native Irish contexts, far away from Scandinavian Dublin (2001, figure 9). They overshadow what by contrast appears to be only a small amount of numismatic material from all of Ireland, at least in terms of its value by weight (Kenny 1987, 517–518). Unfortunately, this has resulted in a somewhat distorted view of the coin hoards, which tend to get dismissed since they are less remarkable in character and quantity. What Kenny, Sheehan and others have misunderstood is the fact that coins are present in so many Irish finds, but in relatively small numbers, is evidence in itself that they had a purpose beyond the melting pot. The present study rejects therefore the popular assumption that coins in this context were automatically regarded as bullion, soon to be transformed into ingots or ornaments. This is not viable when so many coins have survived whole, i.e. in a condition that tells a different story. (Bornholdt Collins 2003, 210)

Accepting that Class 1 hoards represent ‘a manifestation of the social function of wealth’, Bornholdt Collins proceeds to comment (in part, as follows) on the second sphere of silver circulation: [This] reflects commercial developments, and signals the beginning stages of regular coin use. Naturally the latter required smaller quantities of silver and the finds are less impressive in terms of their total weight, but this is important evidence in itself of the purpose coins served and the economic thinking which put them to use. It is therefore misleading when Sheehan concludes that, based on the magnitude of the coinless hoards, a ‘core era for the study of Scandinavian activity’ is that spanning the second half of the ninth century and the first half of the tenth (2001, 54). The succeeding period is equally important, but its archaeological manifestation is less dazzling and tells a different story, providing crucial evidence of a fundamental shift in the role of silver. (Bornholdt Collins 2003, 211)

There is no doubt, however, that Sheehan’s classificatory system, as it stands, has enabled him to develop important new insights into the interpretation of coinless hoards, in particular, but these are clearly issues for further consideration when he comes to complete his ‘definitive account’ of the Irish hoard material. At the same time, it is as well to remain aware of the overall complexity of his scheme when taken to its limits. For instance, his Class 4 (Chapter 9, in this volume, Table 9.1) provides 18 possible



subdivisions for a total of 18 hoards. It is hardly surprising therefore that five of these subdivisions are each represented by a single hoard – and that seven are left vacant.

REFLECTIONS ON SOME FURTHER TOPICS Topics that were not specifically addressed at the symposium on ‘Silver Economy in the Viking Age’ are various. Little attention was given to the sources of silver (other than by Birgitta Hårdh), or to the reasons why so many hoards were deposited and never recovered, although the question is raised in Märit Gaimster’s paper (cf. Samson 1991, 128–131; Williams 2004, 80–82). Likewise, there is the question as to what else, besides ornaments of precious metal and hack-silver, would have served as ‘special purpose money’, although this was a theme of James Barrett’s paper (being published elsewhere). After all, for the Late Norse period in the Earldom of Orkney: One should also consider that tribute was probably not paid exclusively in silver … but also in produce of equivalent value to the amount of silver assessed. (Williams 2004, 102–103)

Finally, there is Susan Kruse’s chapter on ‘Trade and exchange across frontiers’ (Chapter 10, in this volume) to be highlighted for its thoughtful – and thought-provoking – attempt to address the central, but usually neglected, question: how did the Vikings, in practice, actually manage to spend their ‘special-purpose money’? This should be stimulus enough to provide the central topic for another such symposium (cf. Hårdh 1978). Such should also concern itself with the closely related matter, only touched upon by Kruse, of the nature of the goods traded and exchanged. This remains one of the outstanding questions: what did the Vikings spend their ‘money’ on? It is one that was addressed by David Wilson, and others, during the 1970s/80s (Wilson 1980; 1982), partly in response to a debate initiated by Peter Sawyer in his chapters on ‘Treasure’ and ‘Towns and Trade’ in the first edition of The Age of the Vikings (1962, 83–116, 168–192). In particular, Sawyer suggested that ‘the rarity of English and Frankish coins’ in ninth-century Scandinavia was because ‘the raiders did not take their winnings home, but rather used them as a sort of capital with which to settle’ (1962, 98; 1971, 100; cf. Wilson 1982, 259–160). The investment of silver in the purchase and/or capitalisation of land was clearly an important element in the economy of the Viking Age, but Sawyer was perhaps over hasty in dismissing the alternative explanation (1962, 98; 1971, 100). that the early raiders were unused to coin and therefore quickly converted any coins they may have acquired into ornaments or ingots. This seems very improbable. If the raiders were only interested in the weight of their silver, it would not matter much to them if the silver was coined or bullion.

The attention drawn in this volume to the ‘display economy’ requires such ideas to be revisited. Indeed, the questions highlighted here are amongst those currently being posed in the context of the post-excavation analysis of ‘the kaupang at Skiringssal’ (cf. the sections on ‘What is trade?’ and ‘Business in the kaupang’, in Skre and Stylegar 2004, 52–58). Here then is newly excavated evidence to bring to bear on these matters, with the first volume



to appear in the Kaupang publication series (scheduled for 2006) being that on The Means of Exchange.

References Blackburn, M (2001), ‘Expansion and control: aspects of Anglo-Scandinavian minting south of the Humber’, in J Graham-Campbell, R Hall, J Jesch and D N Parsons (eds), Vikings and the Danelaw. Select Papers from the Proceedings of the Thirteenth Viking Congress, Oxbow Books, Oxford, 125–142 Blackburn, M (2002), ‘Finds from the Anglo-Saxon site of Torksey, Lincolnshire’, in B Paszkiewicz (ed), Moneta Mediævalis. Studia numizmatyczne i historyczne ofiarowane Profesorowi Stanislawowi Suchodolskiemu w 65. rocznice˛ urodzin, Instytut Archeologii i Etnologii Polskiej Akademii Nauk, Warsaw, 89–101 Bornholdt Collins, K (2003), ‘Viking Age Coin Finds from the Isle of Man: a Study of Coin Circulation, Production and Concepts of Wealth’, unpublished PhD dissertation, University of Cambridge Brøndsted, J (1965), The Vikings, Penguin, London Dolley, R H M (1961), ‘The 1871 Viking-age find of silver coins from Mullaghboden as a reflection of Westfalding intervention in Ireland’, Universitetets Oldsaksamlings Årbok (1960–61), 49–62 Dolley, R H M (1966), Sylloge of Coins in the British Isles, Vol 8. The Hiberno-Norse Coins in the British Museum, British Museum, London Foote, P (ed) (1961), Snorri Sturluson, Heimskringla, Part 2: Sagas of the Norse Kings (translated by Samuel Laing), Everyman’s Library, Dent, London Friðriksson, A (1994), Sagas and Popular Antiquarianism in Icelandic Archaeology (Worldwide Archaeology Series 10), Avebury, Aldershot Gaimster, M (1991), ‘Money and media in Viking Age Scandinavia’, in Samson (ed), 113–122 Graham-Campbell, J (1976), ‘The Viking-age silver hoards of Ireland’, in B Almqvist and D Greene (eds), Proceedings of the Seventh Viking Congress, Dublin, 1973, Royal Irish Academy, Dublin, 39–74 Graham-Campbell, J (1982), ‘Viking silver hoards: an introduction’, in R T Farrell (ed), The Vikings, Phillimore, London and Chichester, 32–41 Graham-Campbell, J (1989), ‘The coinless hoard’, in H Clarke and E Schia (eds), Coins and Archaeology. Medieval Research Group: Proceedings of the First Meeting at Isegran, Norway, 1988 (BAR, International Series 556), Oxford, 53–61 Graham-Campbell, J (2000), ‘The power of silver’, in Roesdahl and Sørensen (eds), 7–20 Graham-Campbell, J (2001), ‘The dual economy of the Danelaw: the Howard Linecar Memorial Lecture 2001’, BNJ 71 (2001), 49–59 Graham-Campbell, J (2005), ‘The Viking-age gold and silver of the North Atlantic region’, in S Arge and A Mortensen (eds), Viking and Norse North Atlantic: Select Papers from the Proceedings of the Fourteenth Viking Congress, Tórshavn, 19–30 July 2001, The Faroese Academy of Sciences, Tórshavn, 125–140 Hårdh, B (1978), ‘Trade and money in Scandinavia in the Viking Age’, Meddelanden från Lunds universitets historiska museum 1977–78 (New Series, 2), 157–171 Hårdh, B (1996), Silver in the Viking Age: a Regional-Economic Study (Acta Archaeologica Lundensia, Series in 8°, no. 25), Almquist & Wiksell International, Stockholm Jesch, J (2000), ‘The power of poetry’, in Roesdahl and Sørensen (eds), 21–39 Kenny, M (1987), ‘The geographical distribution of Irish Viking-age coin hoards’, PRIA 87c, 507–525



Page, R I (1995), Chronicles of the Vikings, British Museum, London Roesdahl, E and Sørensen, P Meulengracht (eds) (2000), Beretning fra nittende tværfaglige vikingesymposium, Forlaget Hikuin, Højbjerg Samson, R (1991), ‘Fighting with silver: rethinking trading, raiding and hoarding’, in Samson (ed), 123–133 Samson, R (ed) (1991), Social Approaches to Viking Studies, Cruithne Press, Glasgow Sawyer, P H (1962), The Age of the Vikings (2nd ed, 1971), Edward Arnold, London Sheehan, J (2001), ‘Ireland’s early Viking-age silver hoards: components, structure and classification’, Acta Archaeologica (Supplementa II) 71, 49–63 Skre, D and Stylegar, F-A (2004), Kaupang: the Viking town, University of Oslo Williams, G (2004), ‘Land assessment and the silver economy of Norse Scotland’, in G Williams and P Bibire (eds), Sagas, Saints and Settlement, Brill, Leiden, 65–104 Wilson, D M (1980), Economic Aspects of the Vikings in the West – the Archaeological Basis (The 2nd Félix Neubergh Lecture, 1978), Gothenburg University Wilson, D M (1982), ‘The Vikings and their use of wealth in the ninth and tenth centuries’, Saga og sed (Kungl Gustav Adolfs Akademiens årsbok 1982), Uppsala, 252–261

Index Abbasid dinars 38-39, 59-61, 144-45; Abbassid era, trade connections in 135-38 Ælfgifu 190 Æthelred II 40, 56, 64-66, 86, 182 æstels 73, 78 Alfred 55-58, 63, 65, 73, 164-65, 169-71, 180, 189, 198-201 Almoravid dinars 59-60 Andrén, A 124, 128 Angeln 35-36, 43; Viking-age finds in 4345 Angers coins 116 Anglo-Saxon coins/objects 49-51, 58-67, 70, 80-87, 126, 149-50, 159, 179-206; AngloSaxon gold jewellery and artefacts 73-75. Anglo-Saxon weight systems 142 anonymous coins see silent coins Ansgar, Bishop 23, 25, 188-89 Aquitaine mints 51-52, 68, 218 Arabic coins see Islamic coins Archibald, M M 61, 70-71 Arne T J 142 Ashdon hoard 53, 197 Baugatal, the Icelandic law on blood-money 126, 216 Bayeux coins 100-106, 115 Bendixen, K 188 Birka 20-25, 40, 125, 163, 170, 172, 187, 193 Blackburn, M A S 181, 188, 196, 199, 201 Blunt, C E 1, 49, 62 Bornholdt Collins, K A 70, 204, 220 Bornholm, Arabic gold coin from 60 Brather, S 138 Brøgger, A W 143, 169 Brooks, N P 56 ‘bullion economy’ 178-84, 215-17 Byzantine gold solidus 38-39, 59 Byzantine silver coins 136 Byzantium 142 Callmer, J 138 Carolingian coinage 15, 18, 22, 51, 72, 79, 99-118, 188 Caucasus 136, 138 Chadwick, H M 58 Charlemagne 18, 142, 165, 185, 187

Charles the Bald 3, 99, 101-103, 108, 115116 Charles the Simple 51, 99, 102-103, 108, 180 Charvet, J 103-104 Chester 7-8, 75, 90 Chick, D 5 Cloghermore Hoard 156-159 Cnut 8, 51, 56, 74, 130, 192, 194-95 Codex Aureus 57 coinage see also individual entries: English gold coinage (675-1250) 61-65; gold coinage in western and northern Europe 59-61; importance of 24; non-monetary functions of 180-81; regal coinage in Denmark 194-98; and silver economy in the Viking Age 177-207; south Scandinavian coinage in the ninth century 13-26 coin-based economy 178-79, 195, 197, 202, 205-06 see also coined money economies; types of, co-existence 181-85 coined money eceonomies 166-67, 171, 173 see also neutral exchange coin hoards 61, 68, 124, 149-60, 195-96, 201, 218-220 ‘coinless hoards’ 149, 151, 216, 220 coin production: and currency, in Normandy 99-121; coin production (864-77) 99-101; coin production (877-911/24) 102-109 coins see also individual entries: and currency and state formation 117-118; as jewellery 183 Combination-Groups (KG 1-6), in Scandinavian coinage 15-22, 16-17, 29; KG 7, 19, 36, 38 Coudres hoard 102, 109, 111, 116 ‘Co Antrim’ hoard 157, 161 Co Cork, gold ingots from 76, 156 Co Donegal 154-56 Co Kerry 156-57 Co Kilkenny 150, 154 Co Westmeath 155; gold arm-rings from 76 Coupland, S 68, 182, 188, 190 Coutances coins 100-102 Cross and Lozenge coinage 9, 50 Crumlin-Pedersen, O 186 CRVX coinage 190




Cuerdale hoard 9, 19-20, 89-90, 109; 179, 197-207; pecking on coins from 49-53, 50 Curtisasonien coins 100-102, 109, 115-116; minting of 109 Danelaw 7-9, 50, 52, 75, 89, 93, 161-168, 179-180, 184; and silver economy in Viking Age 196-202 Danish coinage 23, 40, 52, 182, 185-186, 192 Danish sceattas 14, 18 Darkevich, V P 134 Delfjizl hoard 68, 72 Denmark 6, 14, 25-26, 44, 89, 91-92, 128130, 139-144, 184-192, 202, 204; armrings from 91-92; ‘half-bracteates’ from 23; medieval Denmark 127; minting in 42, 124-125; in the ninth and tenth centuries 187; regal coinage in 177, 194-196; silver ‘ring-money’ from 181 Deux-Jumeaux coins 99-101 Devroey, J-P 7 ‘display economy’ 215-218 Dodwell, C R 74-75 Dolley, M 204 Dorestad 2, 5, 14-18, 22-25, 29, 40, 188 Dublin 7, 150, 167, 172; High Street excavations 76; gold ingots from 76-77; minting at 160; Wood Quay excavations 76 Ducarel, A C 112 Duczko, W 190 Duplessy, J 59-60 East Anglia 2, 6, 8, 50-52, 149, 180, 184, 196-202 Eadgar 74; coinage reform of 7-8 Eadmund 108 Eadred 55-58, 65-67, 79 Edward the Elder, coins of 50, 64-66, 87, 165, 196, 201 Elgin 69, 71 England: gold in, during the ‘age of silver’ 55-98; gold ornaments of Scandinavian character from 75; sceattas from 3-5 Eudes 102 Europe 1, 6, 9, 14, 29, 40, 42, 66-69, 74, 78, 136-138, 179, 185-191; early medieval Europe 124-127; gold coinage in 59-61; West European coins 130-131, 142, 185 Evreux coins 99-102, 109-10,112 Evreux/Saint-Taurin hoards 109-111, 116117 Exeter excavations 70

Fatimid quarter-dinars 60 Fécamp hoard 109, 115 Fekhner, M V 142 Fell, C E 164 foreign trade, archaeological record of 165166 France 2, 59, 68, 70, 78 ‘Francia’/Frankish coins 59-60, 68, 99-121 Frankish weight systems 142 Franklin, S 136 Frisia 14, 68-69, 185, 189 Frisian runic sceattas 5 Frisian sceattas 18 Gaimster, M 218-222 Gainsborough 70, 72, 76 Gariel, E 104 GDR-coins 99-109, 115-117 Germany 40, 52, 59 Godfred 187-188 gold 181-183 see also gold coins; AngloSaxon gold jewellery 73-75; Byzantine gold solidus 38-39; in England, in the ‘Age of Silver’ 55-98; gold finger-rings 73-77; Scandinavian gold ornaments, ingots and hack-gold 75-77; in the Scandinavian world 75-77 gold coins/coinage: in Britain, distribution map of, 72; in the British Isles, histogram of, 79; British finds of 80-85; English gold coinage 61-67; gold coinage in western and northern Europe 59-61; in western and northern Europe 59-61; in status economies 181-183 Gotland 60, 89, 92, 128-130, 135-137, 141144 graffiti 38, 137, 190 Graham-Campbell, J 49, 76-77, 91-93, 151, 155, 182-183, 203 Grierson, P 18, 57, 68-69, 163, 180 ‘grivna’ 141-143, 182 Grubenhaus 38, 43 Gurevitj, A 144 Guthrum 165, 171, 180, 189, 198 hack-gold; in Scandinavian world 75-77, in Britain and Ireland 89-93 hack-silver 35, 42-43, 76, 89, 91 123-130, 145, 149-161, 179-184, 197, 203, 216-221 ‘Half-bracteates’ 23, 37 Hamwic 2, 5, 69, 172 Harald Bluetooth, coinage of 23, 189

INDEX Harald Hardråde, coinage of, 39, 40, 178180, 192-194, 203, 206 Hårdh, B 24, 75, 155, 159, 182, 217 Hare Island hoard 76 Haussig, H-W 138 Haute-Isle, coins of 109 ‘Hebrides’ hoard, gold ornaments from 76 Hedeby coins 20, 23-24, 29-46, 117, 130, 146, 163-167, 171, 187-190 Hiberno-Norse/Scandinavian coinage 150152, 154, 160-161, 184, 202-204 Hinton, D A 58, 74, 172 Hlaðir earls 191-192, 200 hoard-finds and their structures 34-37 Hoen hoard 60, 68-69, 78, 81, 91, 142, 182183 Ibn Fadlan 140, 182, 202 Iceland 126, 177, 216, 218 Ilisch, L 57, 59, 62 Imbleville hoard 101, 115 Ipswich 2, 172 Ireland 7, 60, 77, 87, 91, 149-161, 165-170, 177, 202, 217-220 Islamic coins 9, 35, 38, 42, 57, 59-62, 67, 7980, 82, 85, 125, 130-31, 135, 138-45, 14951, 170, 180, 190, 203 Islandbridge, Viking graves at 170 Isle of Man 69, 71, 77, 202-205, 220 Italy 51-52, 59, 61-62, 67 Itil 138 Jesch, J 216 Jonsson, K 49 Juaye-Mondaye hoard 109, 115 Jutland 2, 5, 22, 186 Kama region 135-136 KAROLVS-monogram 99, 114-115 Kaupang 164, 172, 221 Kenny, M 149, 154 Kent 2, 63, 111 Khazaria/Khazars 18, 60, 136-140 Kilmacomma hoard 157-158 Kislaya 18, 20 Kosel, Viking-Age settlements of 43 Kotljar, N F 142 Kruse, S E 182 Kufic coins 49, 52, 143, 151 see also Arabic/Islamic coins Lafaurie, J 102-104 Le Maho, J 115 Le Mans mint 115-116


Le Talou mint 99-101 Lewes mint 65, 68 Lincolnshire 8, 70, 183, 197 Lisieux coins 100-102, 116 Loire 101, 114-116 Lombard, M 135 London 62-65 Longpérier, A de 111 Louis IV 110-111, 113-114, 116 Louis the Pious, coins of 14, 18-22, 57, 64, 82, 87, 109, 116-117, 188; and their imitations 67-73 Lunette pennies 52 Lyon, C S S 58 Magnús 192, 195-196 Malmer, B 29, 32, 49, 185-189, 193 ‘mancus’ 57-59, 93n1 Martin, J 138 Mauss, M 126 McCormick, M 57 Mercia 2, 50, 52 Merovingian plated coins 70 Metcalf, M 13-15, 18, 29, 101, 114, 184-186 ‘mixed’ hoards 149-151, 218 Moesgaard, J C 60, 68 monetisation 1-2, 5-9, 14, 23, 111, 205 Morkinskinna, saga compilation 178-179 Morrison, K F 19 Nazarenko, A V 142-143 Nelson, J L 188 Netherlands 51-52, 78 ‘Neustria’ 101, 114-115 neutral exchange, in foreign trade 166-169 Nightingale, P 58 Noonan, T S 137-138, 142 Norfolk 59, 70, 75, 89-90 Norman hoards 60, 71, 99-121 181 Normandy: coin production and currency in 99-121; Viking rulers in 180 North Sea regions: in the Pre-Viking and Viking Ages 1-9; and Ribe controversy 185-186 Northumbria 5, 18, 39-40 Norway 14, 23, 60, 69; and silver economy in Viking Age 190-193 Norwegian penny 39-40 Norwich excavations 75 obols 19 Odo coins, pecking pattern of 51 Offa 5, 57, 61-62, 66, 165, 185 Óláf Tryggvason 191-192, 205



Olof Skötkonung 44, 193-194 Öland: hoards 127-131; Arabic coins on 131; ‘Permian’ rings found on, 132 Pagan, H E 62, 65, 69, 73 pecking, coin testing by 49-53 ‘Permian’ rings, 132-133, 135-141, 143-146, 181 Persia 136, 142 polyhedral weights, 136, 141 Poey d’Avant, F 105 porcupines 185 pound, origins of 142 Rennes hoard 103, 115 Rhine valley 5-6, 40, 51 Ribe 2, 5, 13-20, 22, 44, 127; Ribe controversy and North Sea monetary zone 185-186; Post-Office excavation 21-22 Richard I 111, 114, 116 181 Richards, J D 197 Rollo 99, 109, 180 Roman period silver hoards 124 Rouen coins 99-117 Russia 138, 141, 144 Rus trade 138, 182, 215 Samanid, Arabic coin 145 Samson, R 163 Sawyer, P H 57, 180 Scandinavia 8-9, 52, 69, 130, 144 see also individual entries below Scandinavian coinage; chronology of 18-19; design in 16-18; dies, links and output in 20; in the ninth century, survey of 15-22; South-western Scandinavian mints 22-26; weight of 19-20 Scandinavian gold ornaments, ingots and hack-gold 75-78, 181-183 Scandinavian weight systems 141-143, 169171 sceattas 3-5, 14-18, 21-22, 38, 185-187, 206 Schleswig-Holstein 22-23, 34-37, 40, 141 Schuby 40, 43

Schwansen, Viking-Age finds in 43-45 Scotland 7, 69, 76, 182; and the Isle of Man 202-205 Sheehan, J 183, 202-203, 217-220 Shepard, J 136 Siefred coins 51 Sigtuna 43-44, 170, 193 ‘Sihtric Comes’ 180, 201 silent coins 13, 20-22, 127-131, 177-207, 215-221 Skaare, K 177, 191 Skovmand, R 143 ‘special-purpose money’ 78, 127-131, 218, 221 spherical weights 140 St Edmund coinage 9, 52 Steinfeld hoard 36, 38 Steuer, H 24, 140 Stevenson, R B K 204 Sweden 40, 43-44, 60, 193-194 Torksey 71, 73, 76, 77, 183 Tours-Chinon coins 116 Trewhiddle hoard 73, 75 Vjatka 139, 141, 143-144 Volga region 60, 133-144 Wallace, P 165 Warnke, C 138 weight systems 141-143, 169-171 Wessex 2, 8, 52, 70, 111, 180, 186, 199 Wiechmann, R 20, 24, 40 Wigmund, Archbishop 64, 66-67, 70, 72, 87 William I of England 9 William I of Normandy, 109-114, 116-118, 180 Williams, G 1, 215 Wodan-Monster sceattas 5, 14-18, 21-22, 38, 185-187, 206 Woodhead, P 112-113 York coins 3, 7-9, 51, 60, 64-65, 69, 197-206

James Graham-Campbell & Gareth Williams - Silver Economy in the Viking Age