Celtic Art in Europe: Making Connections. Essays in Honour of Vincent Megaw on His 80th Birthday 9781782976554, 1789253837, 9781789253832, 9781782976585

Table of contents :
Contributors ix
1. Introduction to Celtic Art in Europe: making connections / Chris Gosden, Sally Crawford and Katharina Ulmschneider 1
2. Once again, Herodotus, the Κελτοί, the source of the Danube, and the Pillars of Hercules / John T. Koch 6
3. The Sheffield origins of Celtic Art / John Collis 19
4. Theorie der keltischen Kunst. Ein Versuch / Felix Müller 28
5. Les codes de représentation visuelle dans l’art celtique ancien / Laurent Olivier 39
6. Hidden faces and animal images on Late Iron Age and Early Roman horse harness decorated using the 'champlevé' technique / Jennifer Foster 56
7. The human masks of unknown provenience / Mitja Guštin 68
8. Heads, masks and shifting identities: a note about some Danubian 'kantharoi' with anthropomorphic decoration / Mariana Egri 73
9. 'Off with their heads...!' once again: images of daggers and severed heads on an Iberian 'falcata' sword / Fernando Quesada Sanz 86
10. A Celtic severed head, or Lazarus in the arms of Abraham? / Natalie Venclová and Jan Royt 96
11. Zur Attachenzier der Schnabelkannen von Basse-Yutz / Otto-Herman Frey 101
12. The not so ugly duckling – an essay on meaning / Flemming Kaul 105
13. Fragments of a 'carnyx' from Leisach (Austria) / Paul Gleirscher 113
14. Between ruling ideology and ancestor worship: the 'mos maiorum' of the Early Celtic 'Hero Graves' / Thomas Stöllner 119
15. Alfred and Alexander / John Boardman 137
16. La fibule de Moscano di Fabriano: un jalon important de l’évolution de l’art celtique au IVe siècle avant J.-C / Luana et Venceslas Kruta 140
17. Zum Wenden: der Halsring aus Gehweiler-Oberlöstern im Saarland / Rudolf Echt 148
18. Late Hallstatt and Early La Tène gold and silver beads in southeast Pannonia / Hrvoje Potrebica and Marko Dizdar 152
19. 'East meets West...': The stamped pottery from the La Tène cemetery at Fântânele-Dealul Popii (Transylvania, Romania) / Aurel Rustoiu 159
20. A vessel with stamped decoration from the Želiezovce collection / Gertrúda Březinová 173
21. Balkan 'Kantharoi' / Petar Popović 177
22. La Tène and Przeworsk strap shield bosses from Poland / Tomasz Bochnak 183
23. De l’anneau en bronze à têtes de béliers de Chermignac (Charente-Maritime) et de quelques pièces de harnais. La Tène finale de Gaule de l’Ouest / José Gomez de Soto 196
24. A mould for Celtic-type rings from Sanzeno in the Valle di Non, Trentino / Franco Marzatico 206
25. 'Leopold Bloom I' and the Hungarian Sword Style / Paul Jacobsthal, with introduction by Katharina Ulmschneider and Sally Crawford 213
26. The Celtic mercenary reconsidered / Jan Bouzek 223
27. The Dragon from Oberleiserberg / Maciej Karwowski 234
28. A l'aube du IIIe s. av. J.-C.: les fourreaux d’épée à décor estampé sur fer / Thierry Lejars 239
29. '...to boldly go where no man has gone before.' Dedicated to Ruth and Vincent... / Boris Kavur and Martina Blečić Kavur 264
30. Art and Craftsmanship in elite-warrior graves: 'from Boii to Parisii and back again...' / Nathalie C. Ginoux and Peter C. Ramsl 274
31. Ascot hats: an Iron Age leaf crown helmet from Fiskerton, Lincolnshire? / Andrew Fitzpatrick and Martin Schönfelder 286
32. Snettisham swansong / I. M. Stead 297
33. The Iron Age open-air ritual site at Hallaton, Leicestershire: some wider implications / Colin Haselgrove and Vicki Score 304
34. Brit-art: Celtic Art in Roman Britain and on its Frontiers / Jody Joy 315
35. Art in context: the massive metalworking tradition of north-east Scotland / Fraser Hunter 325
36. The Torrs 'Chamfrein' or Head-piece: restoring 'A very curious relic of antiquity' / C. Stephen Briggs 341
37. Vincent, in appreciation / Mansel Spratling 356
J. V. S. Megaw: bibliography 361
Colour Plates 373

Citation preview

Celtic Art

in

Europe

Ruth and Vincent Megaw at the Palace of Culture, Târgu Mureş, Romania, 2009

CELTIC ART IN EUROPE making connections Essays in honour of Vincent Megaw on his 80th birthday

Edited by

Christopher Gosden, Sally Crawford and Katharina Ulmschneider

Oxbow Books Oxford & Philadelphia

Published in the United Kingdom in 2014 by OXBOW BOOKS 10 Hythe Bridge Street, Oxford OX1 2EW and in the United States by OXBOW BOOKS 908 Darby Road, Havertown, PA 19083 © Oxbow Books and the individual authors 2014 Hardcover Edition: ISBN 978-1-78297-655-4 Digital Edition: ISBN 978-1-78297-656-1; Mobi: ISBN 978-1-78297-657-8; PDF: ISBN 978-1-78297-658-5 A CIP record for this book is available from the British Library Library of Congress Cataloging-in-Publication Data Celtic art in Europe : making connections : essays in honour of Vincent Megaw on his 80th birthday / edited by Christopher Gosden, Sally Crawford and Katharina Ulmschneider. 1 online resource. Contributions in English, French, and German. Includes bibliographical references and index. Description based on print version record and CIP data provided by publisher; resource not viewed. ISBN 978-1-78297-656-1 (epub) -- ISBN 978-1-78297-657-8 (mobi (kindle)) -- ISBN 978-1-78297-658-5 ( pdf) -- ISBN 978-178297-655-4 1. Art, Celtic. I. Gosden, Chris, 1955- editor. II. Crawford, Sally (Sally Elizabeth Ellen) editor. III. Ulmschneider, Katharina, editor. IV. Megaw, J. V. S., honouree. N5925 704.03’916--dc23 2014023136 All rights reserved. No part of this book may be reproduced or transmitted in any form or by any means, electronic or mechanical including photocopying, recording or by any information storage and retrieval system, without permission from the publisher in writing. Printed in the United Kingdom by Berforts Information Press Ltd, Eynsham, Oxfordshire For a complete list of Oxbow titles, please contact: UNITED KINGDOM Oxbow Books Telephone (01865) 241249, Fax (01865) 794449 Email: [email protected] www.oxbowbooks.com UNITED STATES OF AMERICA Oxbow Books Telephone (800) 791-9354, Fax (610) 853-9146 Email: [email protected] www.casemateacademic.com/oxbow Oxbow Books is part of the Casemate Group Front cover: Oval-style massive armlet, Auchenbadie, Banffshire. © National Museums Scotland. Back cover: The statue from Glauberg, Hesse. Photograph courtesy of hessenARCHÄOLOGIE; Photo U. Seitz-Gray.

CONTENTS

Contributors ix 1. Introduction to Celtic Art in Europe: making connections Chris Gosden, Sally Crawford and Katharina Ulmschneider

1

2. Once again, Herodotus, the Κελτοί, the source of the Danube, and the Pillars of Hercules John T. Koch

6

3. The Sheffield origins of Celtic Art John Collis

19

4. Theorie der keltischen Kunst. Ein Versuch Felix Müller

28

5. Les codes de représentation visuelle dans l’art celtique ancien Laurent Olivier

39

6. Hidden faces and animal images on Late Iron Age and Early Roman horse harness decorated using the champlevé technique Jennifer Foster 7. The human masks of unknown provenience Mitja Guštin 8. Heads, masks and shifting identities: a note about some Danubian kantharoi with anthropomorphic decoration Mariana Egri

56 68

73

9. Off with their heads…! once again: images of daggers and severed heads on an Iberian falcata sword Fernando Quesada Sanz

86

10. A Celtic severed head, or Lazarus in the arms of Abraham? Natalie Venclová and Jan Royt

96

vi

Contents

11. Zur Attachenzier der Schnabelkannen von Basse-Yutz Otto-Herman Frey

101

12. The not so ugly duckling – an essay on meaning Flemming Kaul

105

13. Fragments of a carnyx from Leisach (Austria) Paul Gleirscher

113

14. Between ruling ideology and ancestor worship: the mos maiorum of the Early Celtic ‘Hero Graves’ Thomas Stöllner

119

15. Alfred and Alexander John Boardman

137

16. La fibule de Moscano di Fabriano: un jalon important de l’évolution de l’art celtique au IVe siècle avant J.-C Luana et Venceslas Kruta

140

17. Zum Wenden: der Halsring aus Gehweiler-Oberlöstern im Saarland Rudolf Echt

148

18. Late Hallstatt and Early La Tène gold and silver beads in southeast Pannonia Hrvoje Potrebica and Marko Dizdar

152

19. East meets West... The stamped pottery from the La Tène cemetery at Fântânele-Dealul Popii (Transylvania, Romania) Aurel Rustoiu 20. A vessel with stamped decoration from the Želiezovce collection Gertrúda Březinová

159 173

21. Balkan Kantharoi 177 Petar Popović 22. La Tène and Przeworsk strap shield bosses from Poland Tomasz Bochnak 23. De l’anneau en bronze à têtes de béliers de Chermignac (Charente-Maritime) et de quelques pièces de harnais. La Tène finale de Gaule de l’Ouest José Gomez de Soto

183

196

24. A mould for Celtic-type rings from Sanzeno in the Valle di Non, Trentino Franco Marzatico

206

25. ‘Leopold Bloom I’ and the Hungarian Sword Style Paul Jacobsthal, with introduction by Katharina Ulmschneider and Sally Crawford

213

26. The Celtic mercenary reconsidered Jan Bouzek

223

27. The Dragon from Oberleiserberg Maciej Karwowski

234

Contents 28. A l’aube du IIIe s. av. J.-C.: les fourreaux d’épée à décor estampé sur fer Thierry Lejars

vii 239

29. ‘…to boldly go where no man has gone before.’ Dedicated to Ruth and Vincent… 264 Boris Kavur and Martina Blečić Kavur 30. Art and Craftsmanship in elite-warrior graves: ‘from Boii to Parisii and back again…’ 274 Nathalie C. Ginoux and Peter C. Ramsl 31. Ascot hats: an Iron Age leaf crown helmet from Fiskerton, Lincolnshire? Andrew Fitzpatrick and Martin Schönfelder

286

32. Snettisham swansong I. M. Stead

297

33. The Iron Age open-air ritual site at Hallaton, Leicestershire: some wider implications Colin Haselgrove and Vicki Score

304

34. Brit-art: Celtic Art in Roman Britain and on its Frontiers Jody Joy

315

35. Art in context: the massive metalworking tradition of north-east Scotland Fraser Hunter

325

36. The Torrs Chamfrein or Head-piece: restoring ‘A very curious relic of antiquity’ C. Stephen Briggs

341

37. Vincent, in appreciation Mansel Spratling

356

J. V. S. Megaw: bibliography

361

Colour Plates

373

CONTRIBUTORS

Martina Blečić Kavur Faculty of Humanities and Social Sciences, University of Rijeka, Croatia

Mariana Egri Faculty of Ancient History and Archaeology, Babeş-Bolyai University of Cluj-Napoca, Romania

John Boardman Beazley Archive, University of Oxford, United Kingdom

Rudolf Echt Institut für Vor- und Frühgeschichte und Vorderasiatische Archäologie, Universität des Saarlandes, Saarbrücken, Germany

Tomasz Bochnak Institute of Archaeology, University of Rzeszów, Poland Jan Bouzek Faculty of Arts, Masaryk University, Czech Republic Gertrúda Březinová Institute of Archaeology, Slovak Academy of Sciences, Nitra, Slovakia C. Stephen Briggs formerly Royal Commission on the Ancient Historical Monuments of Wales, Aberystwyth, United Kingdom John Collis Department of Archaeology, University of Sheffield, United Kingdom Sally Crawford Institute of Archaeology, University of Oxford, United Kingdom Marko Dizdar Institute of Archaeology, University of Zagreb, Croatia

Andrew Fitzpatrick Wessex Archaeology, United Kingdom Jennifer Foster Department of Archaeology, University of Reading, United Kingdom Otto-Herman Frey Institut für Vor- und Frühgeschichte, Universität Marburg, Germany Nathalie C. Ginoux Institut d’art et d’archéologie, Université Paris-Sorbonne, France Paul Gleirscher Abteilung für Ur- und Frühgeschichte, Landesmuseum Kärnten, Klagenfurt, Austria José Gomez de Soto Laboratoire HeRMA, Université de Poitiers, France

x

Contributors

Chris Gosden Institute of Archaeology, University of Oxford, United Kingdom

Laurent Olivier Département des âges du Fer, Musée d’Archéologie nationale de Saint-Germain-en-Laye, France

Mitja Guštin Institute for Mediterranean Heritage, University of Primorska, Koper, Slovenia

Petar Popović Institute of Archaeology, University of Belgrade, Serbia

Colin Haselgrove School of Archaeology and Ancient History, University of Leicester, United Kingdom Fraser Hunter Department of Scottish History and Archaeology, National Museums Scotland, United Kingdom Jody Joy Department of Prehistory and Europe, The British Museum, London, United Kingdom Maciej Karwowski Institut für Ur- und Frühgeschichte, Universität Wien, Austria, and Instytut Archeologii, Uniwersytet Rzeszowski, Rzeszów, Poland Flemming Kaul Nationalmuseet, Copenhagen, Denmark Boris Kavur Institute for Mediterranean Heritage, University of Primorska, Koper, Slovenia John T. Koch Centre for Advanced Welsh and Celtic Studies, University of Wales, United Kingdom Luana Kruta Independent scholar, Paris, France Venceslas Kruta Centre national de la recherche scientifique (CNRS), Paris, France Thierry Lejars Ecole Normale Supérieure, Archéologies d’Orient et d’Occident et Science des Textes, Paris, France Franco Marzatico Castello del Buonconsiglio, monumenti e collezioni provinciali, Trento, Italy Felix Müller Abteilungen für Ur- und Frühgeschichte und Archäologie der Römischen Provinzen, Universität Bern, Switzerland

Hrvoje Potrebica Faculty of Humanities and Social Sciences, University of Zagreb, Croatia Fernando Quesada Sanz Universidad Autónoma de Madrid, Spain Peter C. Ramsl Austrian Academy of Science, Commission for Prehistory c/o. Dept. for Prehistory, Natural History Museum, Vienna, Austria Jan Royt Institute of the History of Christian Art, Charles University Praha, Czech Republic Aurel Rustoiu Institute of Archaeology and History of Art, University of Cluj-Napoca, Romania Martin Schönfelder Römisch-Germanisches Zentralmuseum Mainz, Germany Vicki Score University of Leicester Archaeological Services, Leicester, United Kingdom Mansel Spratling Independent scholar, Cambridge, United Kingdom I. M. Stead Independent scholar, York, United Kingdom Thomas Stöllner Institut für Archäologische Wissenschaften, Ruhr-Universität Bochum, Germany Katharina Ulmschneider Worcester College, University of Oxford, United Kingdom Natalie Venclová Institute of Archaeology, Academy of Sciences of the Czech Republic, Praha, Czech Republic

1 INTRODUCTION TO CELTIC ART IN EUROPE: MAKING CONNECTIONS Chris Gosden, Sally Crawford and Katharina Ulmschneider

This volume attempts to provide a map of the interests and personal connections of Vincent and Ruth Megaw in their pursuit of an understanding of Celtic art. Since 1963 Vincent has been regularly visiting Europe, on both sides of the former Iron Curtain, to see and talk Celtic art. Initially enrolled as a PhD student in Edinburgh, although then based in Sydney, Vincent embarked on travels useful both to him and everyone he encountered. Meeting him for the first time in (probably) 1979 on the Dürrnberg, Austria where one of us (CG) was working for an Austrian team, it became immediately apparent that Vincent knew everyone, had a story to tell about each and their latest work and results. Indeed, Vincent’s work on Celtic art and his personal style combined around a central theme of connections. The Megaws’ work has been invaluable in demonstrating links between so many of the key pieces of Celtic art across the continent that it has helped create a comparative framework that we all take for granted. Built on earlier foundations, most obviously that of Paul Jacobsthal (1944), the evolving work of the Megaws has allowed us all to chart the growing corpus of material over the last 50 years. The line up of authors in this book is a testimony to the Megaws’ connections and people’s willingness to contribute is due to the high regard, personally and academically, in which they are held. Much has changed over the last half century of work, but also many things have stayed the same. The corpus has grown and Europe has unified in a manner unimaginable in the 1960s. The Celts are still a preoccupation and this is linked to how far the art styles of the Early Iron Age derived from the Mediterranean. One way of reading the stress on the Celts was that they provided a means of insisting that the world of interior and western Europe had a set of styles and sensibilities separate and different

to the Classical tradition. As is evident in this volume, the degree and nature of this separation is still being debated. Individual motifs came from the south, but it is likely that the logic of construction of decoration and its perception were foreign. The lack of naturalism has long been recognized and in their best papers (such as Megaw 1970) the Megaws have traced the construction and possible reception of styles. New sets of connections are now being canvassed, which see Celtic art as a western element of Eurasian systems of style and signification (Wells 2012) and much of the future may be spent charting such links and the differences from the naturalistic art of the world of empires and states to the south. New technologies in the form of modern databases and mapping tools will allow for more systematic and wide-ranging comparisons within Europe and beyond. The increasing number of finds excavated under modern conditions, such as at the Glauberg and Hochdorf, allow us to link objects to their archaeological contexts. This is allowing for discussion about the political and cultural strategies behind the making, use and deposition of spectacular objects (Garrow and Gosden 2012). We are also becoming more aware that the La Tène styles did have some roots in late Hallstatt forms, while differing from them in key aspects of their sensibility. We feel that this volume is a fitting tribute to Vincent, as it is the first time in many years that such a wide geographical set of discussions on Celtic art have been brought together in one volume. It will give the reader a sense of the richness of the material and the main debates in which people are engaged. The papers chart the history of attempts to understand Celtic art and argue for novel approaches. Some fundamental questions continue to bedevil 21st century studies of Celtic Art – the first is the definition of ‘Celtic’, the second the definition of ‘Art’. The question

2

Chris Gosden, Sally Crawford and Katharina Ulmschneider

of whether these two words together have any meaningful relationship with the economy, society, identity and community of Europe outside the classical world hinges on the definition of ‘Celtic Art’. The Megaws have, through their scholarship, made a significant contribution to the continuing debates on these questions, and so it is fitting that the opening papers in this volume take up the baton of controversy. Both John Koch and John Collis point up, in their papers, that to understand the problems, it is necessary to understand the history of the discipline, to contextualize the original use of these contentious terms. Starting with the idea of ‘the Celts’, Koch re-investigates Herodotus’ famous references, and argues that there is a connection between Celtic ethnicity, language and art, that the geographical range covered by the term ‘Celtic’ should be widened, and that ‘La Tène’ art is only a small part of a much bigger ‘Celtic art’ picture. John Collis takes a closer look at the modern origins of the term ‘Celtic Art’, in particular the first published article on the subject by Victorian entomologist and artist John Obadiah Westwood. While the term seemed to fit the current state of knowledge in the mid-19th century, Collis argues that subsequent developments mean the term is no longer fit for purpose, if its purpose is to describe an art style confined to the Celts, or an art style used by all ‘Celtic’ people in antiquity. If the term ‘Celtic’ is problematic, the term ‘art’ for the decorative material culture of the Iron Age is equally contentious. Part of the problem, as Felix Müller elaborates, is that modern ideas about ‘art’ are founded on criteria which explicitly exclude the proto-historic Celtic world. Is it possible to have ‘art’ without an ‘artist’? Like Koch and Collis, Müller looks back at the development of the history of the subject. The first serious study of European Celtic art was by Paul Jacobsthal, first known for his study of Greek vase painting, and whose early love of modern art was reflected in his publication on the modern artist, Ingres (1929). Jacobsthal’s work on European Celtic art, as Müller points out, was predicated on the assumption that, as with Greek vases, it would be possible to identify a ‘master’ responsible for particular outstanding pieces of work. The limitations of an ‘art historical’ approach, ultimately based on classical models, sell Early Celtic art short. These constraints, Müller argues, particularly the anachronistic search for an ‘artist’, should be set aside. Laurent Olivier’s paper takes up the challenge posed by Müller in his closing paragraph by proposing a new way of interpreting Celtic art. Olivier argues that the modern eye is trained to view images in a particular, perspective-driven way, which is preventing us from seeing Celtic art with Celtic eyes. Borrowing a term from geometry – ‘rabatment’ – Olivier argues that Celtic artists were simultaneously communicating multiple viewpoints in their art, expressing the key characteristics of their object – lateral, horizontal and vertical views of a chariot, for example – rather than

restricting themselves to the parts of the object visible from a static viewpoint. Olivier’s thesis demands a significant revision of current readings of Celtic art. Vincent Megaw himself notably argued that Celtic imagery carries a range of meanings, in particular drawing attention to hidden faces within geometric patterns (Megaw 1970). In her paper, Jennifer Foster continues to build on her work of identifying these hidden faces, adding further persuasive zoomorphic images to the catalogue of elusive motifs. As Mitja Guštin’s review of three beautiful face masks in gold, silver and bronze indicates, the idea of the head inspired some of the finest pieces of Celtic art, even though the function and meaning of the objects may be obscure to archaeologists. Mariana Egri’s paper also focuses on faces, this time relating specifically to Danubian kantharoi. Egri correlates the imagery on these vessels to their archaeological contexts, arguing that, on kantharoi, the head carried a specific meaning relating to the ritual and personal use of the object – a significant step forward in decoding the ‘language’ of Celtic art. From the other side of Celtic Europe, Fernando Quesada Sanz considers the possibility of an Iberian ‘head cult’, but postulates that the decorative ‘severed heads’ found on a funerary dagger are as likely to represent trophy ‘kills’. There is no consensus on the meaning of the severed head in Celtic art, and the chronological span and geographical ubiquity of heads as images throws up further problems, as Natalie Venclová and Jan Royt’s fascinating paper reveals. They review the possible Celtic origin of a stone head relief sculpture from the Czech Republic. The authors conclude, after careful assessment of the evidence, that the head is medieval – one fewer example for Vincent to catalogue in his forthcoming Early Celtic Art Supplement. But the Celtic origin of this head could only be ruled out after careful research, and the exercise serves to underline the intriguing possibility that the reverse may also be true: there may be many stone heads in collections across Europe currently mislabelled as ‘Medieval’ when they may in fact be Celtic. Otto-Herman Frey and Fleming Kaul both turn their attention to iconic objects of the Celtic world, the Basse-Yutz flagons, well known through the Megaws’ detailed study of the burial (Megaw and Megaw 1990). At the time of their publication, the Megaws had little with which to compare the art on the Basse-Yutz flagons, but the recent discovery of the Glauberg flagon provides an impressive parallel, as Frey discusses. Kaul’s intention is to try to understand the symbolic meaning of the animal images on the flagon. Rather than seeing the flagon as a static work of art, Kaul thinks about the interaction between the object and the user, arguing that, as with Mariana Egri’s katharoi, the ‘artistic intent’ of the image is only fulfilled through action, in this case pouring the liquid from the mouth. For Kaul, the Basse-Yutz flagon needs to be ‘read’ in the context of movement: for Paul Gleirscher, who analyzes a

1.  Introduction to Celtic Art in Europe: making connections set of metal fragments from Leisach (Austria), the art on a carnyx is intimately connected to its use and to the sound it makes. For Gleirscher, the carnyx was both a weapon and a ritual object: its sound literally inspiring awe in the listeners. Celtic art has to be ‘experienced’, as well as viewed, for its meaning to be unlocked. If there is a relationship between art, culture and social structure – religious or political – then the adoption or adaptation of new art styles should be a red flag indicator of social change. This idea is explored by Thomas Stöllner in his paper on the rich ‘princely’ graves of the Late Hallstatt and Early La Tène culture north of the Alps. Stöllner argues that the influence of Greek art in Celtic production is intimately tied to an adaptation of the older Greek idea of hero worship. For Stöllner, the emergence of a new style of princely burial marks a social and religious change characterized by an appropriation of images and ideals from the Greek world to empower and authenticate an emergent elite. There are very close analogies for Stöllner’s thesis in the so-called ‘final phase’ burial ritual in Anglo-Saxon England. In the 5th and 6th centuries, grave goods in Anglo-Saxon burials signaled ethnicity and local group identity. But in the 7th century AD, Christianity was being introduced, and a new elite emerged who were keen to align themselves to powerful leaders on the continent and impose a new idea of kingship in their burials, rather than signal their local, ethnic identity. The change in social structure associated with the emergence of kingdoms found expression in a changing burial ritual. The new elite sought to validate their kingship, founded on an idea of imperium, by appropriating ideas of Romanitas in more or less accurate ways. The most well-known of these elite burials is that of the man buried at Mound 1, Sutton Hoo, whose grave goods were ostentatiously ‘Roman’ in character, including a unique burial costume imitative of images found on Roman sculpture and coins from previous centuries. The encompassing 7th century idea of Romanitas even included hanging bowls, typically with La Tène-style decoration, whose appearance in 7th century princely graves is best explained as part of an eclectic approach to expressing and recreating an Anglo-Saxon perception of the power of Rome in princely graves (Geake 2000, 17). The material culture of early medieval England provides a number of close parallels with the Celtic world which might provide a fruitful area of comparison in terms of understanding the processes by which new influences are appropriated into art styles, the correlations between changing art styles and changing social structure, and the ways in which the language of art was read and interpreted in a pre- and proto-literate society. In this context, John Boardman’s contribution on the Alfred Jewel offers further fuel for comparison. Boardman argues for the selective and deliberate adoption and adaptation by the 9th century Anglo-Saxon court of iconography referencing part of the

3

Alexander story, originating in the eastern Mediterranean. The image on the Alfred Jewel was a deliberate choice to convey a message which King Alfred knew his court and bishops, and possibly a wider audience, would be able to decode – it should hearten Celtic art scholars that its meaning still remains obscure to modern viewers, though Boardman offers a plausible explanation. A similar process of selective inclusion of foreign material can be seen in the decoration of the 4th century BC Moscano di Fabriano brooch. As Venceslas Kruta argues, this brooch underlines the idea that external images were adopted and transmitted over time because they chimed with specific pre-existing Celtic concepts which gave the image meaning: for Kruta, it is necessary to distinguish between the content of an image, which relates to ideological systems, and the form of the image, which relates to chronological or geographical factors. Other papers in this volume explore and test this contrast between ‘content’ and ‘form’ and their relationship to geography and chronology through studies of single objects, or micro-study of a particular type of object. Rudolf Echt argues that the decoration on a unique early La Tène reversible torc from Saarland helps to provide a chronological date for its manufacture and distribution, while Hrovje Potrebica and Marko Dizdar look at the geographical range of gold and silver beads in southeast Pannonia, finding that their distribution argues for links between the eastern Sava and the Požega Valley. Aurel Rustiou looks for parallels for the stamped pottery from a La Tène cemetery in Romania which demonstrate connections between the East and West in the 4th century: for Rustiou, as for Kruta, there is clear evidence in his study that Celtic art demonstrates local adaptation of symbolic language from external societies. But studies of stamped pottery decoration may lead to unexpected results, as Gertrúda Březinová shows in her examination of a single vessel with stamped decoration from Slovakia. The vessel has a stamped design which is relatively easy to parallel on other media, but is extremely unusual on pottery. What does the normative exclusion of this design on pottery signify, and what does its non-normative appearance on this exceptional pot indicate about the pot or the meaning of the stamp? The question remains open. Distribution maps of objects or decorative elements may indicate the movement or ideas, or of people. For Tomasz Bochnak and Petar Popovič, documentary and archaeological evidence combine in their respective case studies to argue positively for movement of people. In his study of Balkan Kantharoi, Popovič argues that this chronologically persistent form of pottery vessel, adapted from Greece across the Balkans, allows an insight into contacts and migration, which are the best explanation for the widespread distribution of this form. Bochnak’s La Tène and Przeworsk strap shield bosses indicate, through

4

Chris Gosden, Sally Crawford and Katharina Ulmschneider

different distribution of forms over time, that there were pre-migration contacts and exchange of ideas between the Przeworsk and Celtic cultures, followed by a westward migration of people from present-day Poland. Bochnak argues that there is an archaeologically-visible difference between people migrating away from a region, where the direction of movement may be erratic, widespread, and prolonged, compared to people making a deliberate decision to migrate to a chosen location, where the archaeological traces of movement will be deliberate and direct. José Gomez de Soto moves the spotlight to the far west of the Celtic continental world in his study of a bronze ring found in the Charente-Maritime. The ring is problematic, in that the circumstances of its discovery mean there is scant evidence for the date or context of its original deposition, and its heavily-abraded state suggests that it was in circulation for many years after its original date of manufacture – perhaps curated through the Roman period. Gomez de Soto analyses the form and ornamentation of the ring and its closest parallels – of which there are none known so far from Gaul – to test whether it was of local manufacture or came to western France from long-distance trade: though it is plausibly of local manufacture, the decoration of the ring points up the need for more investigation of the ways in which western Gaul engaged with the rest of the Celtic world. The question of local manufacture or import seems to have been solved to some extent for Celtic-type rings from Sanzeno, Italy, by the discovery of a stone mould, discussed by Franco Marzatico. Here, at least, there is evidence for the location of metalworking activity, usually elusive in a Celtic context. However, as Marzatico notes, while a stone mould is evidence for local manufacture, it leaves open the question as to whether the metalworkers themselves are local, foreign or itinerant smiths. Paul Jacobsthal, whose work is referenced in many of the papers in this volume, was confident of the existence of itinerant master smiths in the Celtic world, but frustrated by his inability to prove this theory on the basis of the archaeological evidence to which he had access. Instead he wrote a tongue-in-cheek paper hypothesising their lives to explain the links he identified between decorative elements on Iron Age sword scabbards from Ireland and Britain and sword scabbards in Hungary. Some 70 years after he originally wrote his prescient paper for private circulation, it seems appropriate to publish it for the first time in this volume for Vincent Megaw, in many ways Jacobsthal’s direct scholarly descendent. Jacobsthal’s putative smith worked for, and was com­ missioned by, warriors. Soldiers and mercenaries played an exceptionally influential role in commissioning, transmitting and transporting new styles in the early Celtic world, as Jan Bouzek demonstrates in his paper comparing changes in art in the archaeological evidence with written sources about the campaigns and subsequent migrations and settlements of

early Celtic mercenaries. Looking in detail at the fragment of a dragon’s head terminal from Lower Austria, Maciej Karwowski also finds evidence of links between armour, Celts, the early Romans and the specialized steel from which this weaponry was made. Karwowski teases out RomanoCeltic links from a fragmentary ornament: Thierry Lejars takes equally unpromising material – corroded fragments of stamp-decorated iron sword scabbard fragments – to show that careful examination and reconstruction using modern conservation methods can significantly add to our body of knowledge about Celtic material culture. Boris Kavur and Martina Blečić Kavur continue the weaponry theme, bringing dragon decoration and sword scabbards together in their examination of Early La Tène art, swords, and the transmission of material culture: again, they argue that Celtic mercenaries had a part in transporting a sword produced in Northern Italy to Croatia, a picture reinforced by Nathalie Ginoux and Peter Ramsl’s paper on art and craftsmanship in swords found in elite burials in the territories of the Boii and Parisii. While the previous papers have ranged across the full extent of Celtic presence in continental Europe, the final papers in this volume turn to that part of the Celtic world lying across the Channel which is the academic cradle of the birthday boy. Andrew Fitzpatrick and Martin Schönfelder begin with some of the earliest examples of insular Celtic art yet found, from Lincolnshire. Ian Stead follows with a review of the seminal excavations at Snettisham, on the west coast of Norfolk. Extended work at Snettisham – not all under archaeological control – has allowed Stead to identify a phenomenon of ‘multiple hoard’ deposition on one site over a period of time. Stead argues that there may be other examples of such ritual deposition, though they are obscured by the fact that hoard finds may be reported individually, many years apart, and even with different findspot names. Further review of the evidence might reveal additional examples of ‘multiple hoards’ in England and on the Continent. Ritual deposition in the form of coins is the subject of Colin Haselgrove and Vicki Score’s paper, which reviews the Iron Age open-air ritual site at Hallaton, Leicestershire. Haselgrove and Score review the site to assess what implications it may have for our understanding of links between Rome and Britain before the Invasion, and how this may affect interpretation of other archaeological evidence from this period. Jody Joy develops this theme on a wider chronological and regional scale, looking at the role of art in constructing and renegotiating identity on the frontiers of the Roman world. Fraser Hunter takes up one aspect of this theme, assessing the conception and purpose of the regionally-distinctive Massive Metalworking tradition of north-east Scotland, looking at the social and political factors at play in the development of this innovative style of prominent costume metalwork.

1.  Introduction to Celtic Art in Europe: making connections Many of the papers in this volume have demonstrated the research value of reviewing and reconsidering well-known finds, and it is appropriate that the research papers in this Festschrift close with a re-appraisal of the Torrs chamfrein, a piece central to any discussion of Celtic art in the British Isles. In his paper, Stephen Briggs presents forgotten information about the original discovery, and draws attention to subsequent evidence which has been largely neglected or overlooked, to argue that Atkinson and Piggott’s dominant interpretation of the artefact as a composite fake made from separate Iron Age pieces is wrong. Mansel Spratling brings proceedings to a close with a personal message for Vincent, voicing, on behalf of everyone who has been inspired by the work of Vincent and Ruth, congratulations to Vincent on his 80th birthday, thanks for a monumental contribution to Celtic art, and best wishes for more to come, not least the Early Celtic Art in Europe Supplement. Artefacts of rich form and decoration, which we might call art, require a considerable investment of skill and materials in their production, but also skills of discrimination in appreciation. Such artefacts provide a most sensitive set of indicators of key areas of past societies – their power, politics and transformations. Artistic productions are also demanding subjects for present day academics, in their subtlety and complexity. Vincent’s character has matched that of his subject matter, producing written work and a network of friendships of lasting value.

5

Envoi While this volume was in the proof stage, we learnt the sad news that Vincent’s wife, Ruth had died suddenly on 13th July 2013. The world of Celtic art owes a great debt to Ruth’s inspiration and scholarship. Though this volume is billed as a celebration of Vincent’s 80th birthday, it is also, of course, an hommage to a great academic partnership and is dedicated to Ruth’s memory.

Bibliography Garrow, D. and C. Gosden. 2012. Technologies of Enchantment? Exploring Celtic art 400 BC to AD 100. Oxford: Oxford University Press. Geake, H. 2000. When were hanging bowls deposited in AngloSaxon graves? Medieval Archaeology 43, 1–18. Jacobsthal, P. 1929. Ingres: dessinateur des antiquités. Gazette des beaux-arts 6/1, 75–80. Jacobsthal, P. 1944. Early Celtic Art. Oxford: Clarendon Press. Megaw, J. V. S. 1970. Cheshire Cat and Mickey Mouse: analysis, interpretation and the art of the La Tène Iron Age. Proceedings of the Prehistoric Society 36, 261–279. Megaw J. V. S. and Megaw, R. 1990. The Basse-Yutz find: masterpieces of Celtic Art. Society of Antiquaries of London. Wells, P. 2012. How Ancient Europeans Saw the World. Princeton: Princeton University Press.

2 ONCE AGAIN, HERODOTUS, THE ΚΕΛΤΟΊ, THE SOURCE OF THE DANUBE, AND THE PILLARS OF HERCULES John T. Koch

The passages The two famous references to Κελτοί in the Histories of Herodotus have always been difficult to interpret for two reasons. First, where exactly are the two locations where he says the Κελτοί lived? Second, what does this shared group name mean? In what sense are they one people in those two places? In recent years, as the concept of the Celts has come under more critical scrutiny, the uncertainty has intensified, and it has seemed increasingly likely that Herodotus suffered from some serious misunderstanding. However, it will be proposed in the present paper that these passages can be accepted, in the light of recent research, as coherent and factually accurate. This solution has not been noticed previously because the research is new, it involves discoveries and re-interpretations across a wide diversity of archaeological and historical linguistic topics, which have been published in various languages, and none of it has to do directly with Herodotus. §2.34. I am willing to believe that [the Nile] rises at the same distance from its mouth as the Ister [Ἴστρος ‘Danube’], which has its source amongst the Keltoí at Purēnē [Πυρηνη]1 and flows right through the middle of Europe, to reach the Black Sea at Miletus’s colony of Istria [Ἴστρια]. The Keltoí live beyond the Pillars of Hercules, next to the Kunēsioi [Κυνησιοι], who are the most westerly people of Europe. §4.49. . . . the Ister [Ἴστρος ‘Danube’], that mighty stream which, rising amongst the Keltoí, the most westerly, after the Kunētes [Κυνητες], of all the European nations, traverses the whole length of the continent before it enters Scythia (Herodotus: adapted from translations of de Sélincourt 1954).

The problem One consequence of the new work just mentioned (see

p.  13) is that we now know with more certainty and in more specific detail about the western neighbours of the Κελτοί, the Κυνησιοι or Κυνητες. (The latter form of the name, Romanized Cynetes, is better attested.) That group lived west of the lower river Anas (now Guadiana) in what is today south Portugal. Their territory extended westward to the extremity of the European mainland, the sacrum promontorium (Sagres point). They also lived on the upper Guadiana, where we find Κονιοι (Romanized Conii) as the byform of the group name (Alarcão 2001). These variants Κυνησιοι, Κυνητες, Κονιοι are Celtic and mean ‘houndlike, wolf-like men’ (De Bernardo Stempel 2008), an epithet of warriors/heroes. Geographically, then, ‘beyond the Pillars of Hercules’ and ‘next to the Κυνησιοι’ focuses very specifically on what is today Spain’s south Atlantic coast, including the ancient coastal emporia at Cádiz and Huelva. Greek imports from the 8th, 7th, and 6th centuries BC are known from this zone. Herodotus includes accounts of Greek mariners from Samos and Phocaea reaching the kingdom of Ταρτησσος, the core of which was very probably in this area, during the later 7th and mid 6th centuries BC, respectively (§4.152, §1.163; Freeman 2010, 314–315). Herodotus (§4.8) correctly describes the island of Ερύθεια (Erythea) near Γαδείρα (Phoenician Gadir), i.e. Cádiz, as being on the Ocean ‘beyond the Pillars of Hercules’ (. . . ἔξω στηλέων ‘Ηρακλέων), the same phrase applied to the Κελτοί. So he clearly had sources of information for southwest Spain. Similarly, the mention of the Greek colony of Ἴστρια at the mouth of the Danube most plausibly derives from merchant boatmen who regularly navigated the fluvial route between the Greek world and west-central Europe. The puzzling fact is that the passages refer to Κελτοί in two specific locations about 1700 km (1100 miles) apart ‘as the crow flies’. In Herodotus’s time (c.484–c.430 BC), these

2.  Once again, Herodotus, the Κελτοί, the source of the Danube, and the Pillars of Hercules two areas did not have much in common in material culture and, though the Greeks evidently had links with both, they were not in close contact with each other, nor were they earlier in the Iron Age. Their mutual isolation is not hard to understand in the light of the passages themselves, where one is described as being reached by a transcontinental river route and the other by a sea voyage on the Atlantic. The two areas evidently spoke closely related, but not identical languages (p. 13 below). Why were the people in both called by the one name Κελτοί by the Greeks and/or by themselves? And how did groups, geographically and culturally detached in this way, come to speak closely related languages? However that happened, the Greeks in general (and Herodotus in particular) were not that interested in the languages of barbarians, so it is unlikely that they would have given the name Κελτοί to both groups in recognition of their similar languages. One solution is to assume cultural and chronological priority for the Danubian Κελτοί. This approach allows a close identification of the group name Κελτοί and the Ancient Celtic linguistic evidence in west-central Europe with characteristically Hallstatt and La Tène material from Austria and south Germany. Such a Danubian focus has figured in the standard explanation of the Celts which forms the usual narrative introduction to Celtic studies: for example, Powell’s The Celts: ‘The archaeological evidence indeed goes far to show that the region of the Upper Danube was the Celtic homeland from which some of these people had spread to Spain, as they did somewhat later into Italy and the Balkans’ (Powell 1958, 14). To attempt greater specificity for this still prevalent view, it is most often understood to mean that the earliest Iron Age of west-central Europe (Hallstatt C1a or Hallstatt C0, c.800–c.750 BC) was the source of people(s) called Κελτοί, the undifferentiated Proto-Celtic language, and characteristically Celtic material culture: in other words, Ha C1a = P[roto‑]C[eltic]. However, the idea is seldom spelled out with such precision, but more often merely implied. In the Ha C1a = PC model, the Celtic languages in the Iberian Peninsula and peoples there called Κελτοί, Κελτικοί (Celtici), and Κελτiβερες (Celtibēres) result from movements across the Pyrenees, which were historically undocumented, unlike the movements to Italy, the Balkans, and Asia Minor (i.e. Brennos I and Brennos II, etc.). As groups secondary and marginal to the Celtic world in general, why these ‘HispanoCelts’ do not show characteristically Hallstatt or La Tène cultures has not been a question of central importance to most researchers outside Spain and Portugal. There are Late Bronze Age Urnfields in Catalonia, but confined mostly to the area where the non-Indo-European Iberian language was used in pre-Roman times and far away from the Κυνητες and Pillars of Hercules. Therefore, adopting the theory that the Celts originated and then dispersed from the upper Danube region leads to the conclusion that the Celts who

7

came to the Iberian Peninsula kept their language but lost their culture (Hoz 1992, 19; cf. Lorrio and Ruiz Zapatero 2005 for a general recognition of the archaeological problem of the Hispano-Celts). A different solution for Herodotus’s widely spread and culturally disparate Κελτοί is to suppose that he did not know the actual location of the source of the Danube, but thought it was far to the west. If so, the basis for equating Hallstatt and La Tène with the name Κελτοί is unsound, making terms such as ‘Celtic art’ misnomers except by concession to modern convention (Collis 1997; 2003 contra Megaw and Megaw 1996; 1998). A second implication of this ‘western’ interpretation is that Proto-Celtic possibly first branched off from Late Indo-European in the Atlantic façade region, then dispersed towards central Europe (Koch 2010). However, it would of course be preferable if we could interpret the passages without assuming a major misconception. The pivotal mention of Ἴστρια implies that Herodotus’s information came from Greek colonists long established on the Danube, who were not likely to have had completely fantastic ideas about the river’s source or the people living near it.

New work and its implications Proto-Celtic emerged as the ‘lingua franca’ of the Atlantic façade. This theory was introduced by Cunliffe in Facing the Ocean (2001) as a conclusion drawn from an overview of the longue durée of the later prehistory across the region (further developed in Cunliffe 2009; 2010). The theory rests largely on the continuity of cultural connectivity along the Atlantic façade at successive stages in later prehistory juxtaposed with the geographical distribution of Ancient Celtic linguistic evidence. It envisions the primary formation of the lingua franca in the zone of Atlantic maritime networks with secondary spread into central Europe along the principal navigable rivers. I have accepted this ‘Celtic from the West Theory’ as more consistent with the historical-linguistic evidence, particularly the evidence of Hispano-Celtic (Koch 2010). Having rejected as chronologically impossible Ha C1a = PC (Koch 2013a), I have concluded that an approximate correspondence between the Atlantic Late Bronze sociocultural area and the Proto-Celtic linguistic area is promisingly workable (Koch 2013b). Although not coming out for the ‘Celtic from the West’ model, Mallory (2013b, 243–286) does significantly have Ireland becoming Celtic at about this time – c.1000 BC – with the hillforts as a key domain for social assembly with linguistic implications. Dagmar Wodtko (2013) has explored a non-invasionist sociolinguistic model in which a single Indo-European dialect might have arisen as the

8

John T. Koch

lingua franca within a prehistoric landscape of small-scale, pre-state societies, which had formerly divided amongst several Indo-European and non-Indo-European languages on more or less equal footing. In the ‘Celtic from the West’ model, the time depth for the emergence of Proto-Celtic cannot be later than the Atlantic Late Bronze Age 1300/1250–1000/950 BC (Koch 2013b). A longer process is envisioned by Cunliffe (2001; 2010; 2013), traced through continuities between such commonalities embracing the region as the Beaker Copper Age and megalithic phenomena. The theory is broadly consistent with Renfrew’s (1987; 1990) Anatolian agricultural dispersal theory of Indo-European and is in key respects anticipated by it (Renfrew 2013). Mallory (2013a) does not see the new model as inherently incompatible with the more shallow (Late Neolithic/Copper Age) chronology of the so-called ‘Kurgan’ theory of the origin and spread of Proto-Indo-European; however, he thinks that an IndoEuropean language could not possibly have been spoken by the makers of the earliest beakers on the Tagus estuary in the 28th century BC. The impact of Celtic from the West on the interpretation of Herodotus is that it inverts the long-held assumption that the Κελτοί and Proto-Celtic came into being in west-central Europe and spread to the Atlantic façade from there. In other words, we must now consider the long-ignored possibility that the Κελτοί beyond the Pillars of Hercules and bordering the Κυνητες had priority. The loss of central-European culture by the Celts of the Iberian Peninsula thus becomes a non-problem.

Proto-Celtic can be explained as the Iberianization of Indo-European in south-west Europe The modifications of Indo-European *p – mostly weakenings and in most phonetic positions complete loss – have traditionally been given the status of defining the branching off of Proto-Celtic from Proto-Indo-European. Historical linguists working in diverse frameworks – McCone (1996, 43), Ballester (2004, 114–17), Koch (2009a, 21), and most recently Schrijver (2011) – have suggested that contact with non-Indo-European languages of south-west Europe that lacked p, such as Iberian, could have affected the sound system of the Indo-European spoken the in same region, leading to the loss of Indo-European *p and hence the definitive emergence of Celtic. It is likely that Aquitanian / Palaeo-Basque also shared this negative phonological characteristic; however, Basque is not fully attested until the 16th century AD, and even reconstructed ‘Pre-Basque’ can only be taken reliably back to the period of the earliest Latin loanwords about 2000 years ago (Trask 1997). Schrijver plausibly argues that the key sociolinguistic situation was a cultural expansion, which resulted in numerous speakers of Iberian learning Indo-European as a second language. To put

it in layman’s terms, Celtic is Indo-European spoken with an Iberian accent. The spread of prestigious metalwork of Atlantic type in the Iberian Peninsula and south-west Gaul during the Late Bronze Age (1300/1250–1000/950 BC) can be plausibly recognized as reflecting circumstances in which speakers of Iberian and Aquitanian were learning IndoEuropean as a high-status second language (Koch 2013b). This growing consensus implies that it was in south-west Europe that Indo-European evolved into Proto-Celtic and that it dispersed from there. In other words, it favours – or is at least consistent with – the Celtic from the West theory (p. 8 above). Returning to the evidence of Herodotus, the Pillars of Hercules and territory of the Κυνητες were very near the contact zone of Iberian and Indo-European at the proto-historic horizon, the upper Danube was not. Therefore, if Κελτοί refers to speakers of Celtic, the more westerly group were longer established in their historical homeland.

*Keltoi was the self-designation of the Proto-Celtic linguistic community In the opening section of T. G. E. Powell’s Celts, he wrote: ‘The name Celt was never applied to peoples of Britain and Ireland by ancient writers so far as is known, and there is no evidence that the natives ever used this name themselves’ (1958, 15). This position has long remained the standard view and has been emphasized as particularly significant by writers who are sceptical about the modern concept of a coherent Celtic identity widespread across ancient Europe (e.g. James 1999; Collis 2003). However, as a formulation concerning negative evidence, it can only be pushed so far. And the latter half, about the self-designations of the Britons and Gaels, has always been weaker, because the ancient Insular languages are very poorly attested and only become fully available following Romanization and Christianization processes which transformed the world views and identities of both groups (Koch 2003). In other words, there is no basis to state with confidence what the inhabitants of Iron Age Britain and Ireland did not call themselves. In his 2008 Myles Dillon Memorial Lecture, McCone demonstrated that the application of the standard methods of historical linguistic reconstruction showed that the group name *Keltoi must have existed in the common ancestor of all the Celtic languages where it must have functioned as the self-designation of the speakers of Proto-Celtic (see also Koch 2009b). In the first place, Κελτοί, Celtae, &c., is not a word of Greek or Latin origin. As well as occurring for groups of larger than tribal scale in ancient Gaul and Hispania, the stem is widespread in Celtic men’s names: for example, Celtillus father of Vercingetorix, Hispano-Celtic Celtius, Arceltius, and Celtienus, and Old Irish Celtchar. The stem Kelto- occurs and is used similarly in both Gaulish and Hispano-Celtic. Its attribution in form and meaning to ProtoCeltic follows, because these languages are on either side

2.  Once again, Herodotus, the Κελτοί, the source of the Danube, and the Pillars of Hercules of the primary split in the Celtic branch of Indo-European. There are uncertainties over the configuration of the family tree of the Celtic languages, but there is today general agreement that the first split divided Hispano-Celtic, on the one hand, from the proto-language of Gaulish-BrittonicGoidelic (McCone’s ‘Gallo-Insular’), on the other (Isaac 2005). A Gaulish-Brittonic-Goidelic commonality following the Hispano-Celtic split is implied by linguistic innovations limited to this subgroup, such as the uninflected enclitic relative pronoun *io. *Keltoi is therefore characteristic of the entire Celtic family: it was the self-designation of the ProtoCeltic linguistic community and then of the Hispano-Celtic and Gaulish-Brittonic-Goidelic communities that succeeded it. As McCone argues, the absence of the group name from attested Brittonic and Goidelic is unremarkable. All languages lose inherited vocabulary over time. And as noted above, national names might be especially vulnerable during periods when major political and cultural changes occurred. For McCone’s explanation, the evidence of Herodotus is confirmatory: it predicts the self-designation Κελτοί as a common inheritance in languages of both the Hispano-Celtic and Gaulish-Brittonic-Goidelic sub-branches. The latter linguistic area extended to the upper Danube at the horizon of the earliest written records.

The Gündlingen swords were one of several Atlantic- or Trans-Manche-derived types to spread to the upper Danube at Ha C1a (Fig. 2.1) In the early 20th century, the idea took hold that Gündlingen swords – often simply identified as ‘Hallstatt’ bronze swords – spread from the upper Danube region to northern Gaul, Britain, and Ireland with invading Celts (Gerloff 2004, 124). This scenario of Celticization was still widely believed in the later 20th century, retaining wide currency amongst Celtic philologists (e.g. Mac Eoin 1986), as well as archaeologists. Subsequently, new discoveries, improved relative and absolute chronologies, and detailed re-examination of the metalwork itself have effectively stood this interpretation on its head. The Gündlingen swords do not have antecedents in central Europe, but can be traced to a lineage of British (or British cum Trans-Manche) types: Ewart Park > Thames > Holme Pierrepont (Milcent 2009). Therefore, the Gündlingen type probably developed in south-east Britain in the later 9th century BC then spread from there to south-east Gaul and the upper Danube (Gerloff 2004; cf. Cunliffe 2013, 294–298). Gerloff similarly reverses the usual explanation of the winged chapes, a single-edged razor type, and the ‘Hallstatt’ buckets that appear in westcentral Europe at about the same time, showing that these had insular antecedents. She concludes from this, ‘It is likely that some of our “Hallstatt sword-bearers” had their roots in the Late Bronze Age Celtic or Proto-Celtic Atlantic west and introduced “Hallstatt forms” such as Gündlingen

9

swords [and] winged chapes ... to more central parts of Europe, where they or their kinsmen contributed to rise and wealth of the Hallstatt Culture during the earlier 8th century BC’ (Gerloff 2004, 147). Brandherm (2013a) identifies an extensive range of ‘Early Hallstatt élite status markers derived from Atlantic prototypes of the Ewart Park phase’ seeing these as ‘indicative of historical processes that might have been accompanied by language shift’. Returning to Herodotus, these reinterpretations of the metalwork show that an important element in the formation of the culture of the people called Κελτοί on the upper Danube had come from the Atlantic zone in the period c.800–c.750 BC. Covered under this same potentially linguistic explan­ ation, Brandherm also recognizes an earlier east-to-west movement of high-status forms from early western Urnfield to the south-western Iberian Peninsula: rilled ware from south Portugal, a Rixheim sword from Huelva, and crested helmets depicted on Late Bronze Age warrior stelae. Also datable broadly to the 13th century BC, Gerloff (2010) demonstrates the arrival of central European prototypes for the sheet-bronze cauldrons and buckets of the Atlantic Bronze Age. Therefore, a complex pattern of flow and ebb can be traced between west-central and Atlantic Europe through the Late Bronze Age, implicating both mobile metalworking specialists and their aristocratic patrons (cf. Koch 2013b).

Huelva swords and the Iberian Peninsula’s Bronze-Iron Transition (950–850/800 BC) One notable area of progress in recent years has been with the typologies and relative and absolute chronologies of Bronze Age metalwork. An important aspect of this trend is that fresh scrutiny has been applied to the concept of the Atlantic Bronze Age. New work has confirmed the widespread use of closely related types across the region (Ireland, Britain, Armorica, and the western Iberian Peninsula) in the period 1300/1250–1000/950 BC (Milcent 2012) and also similar practices for the use and deposition of particular types of artefacts, e.g. swords and cauldrons. Therefore, know-how, values, and other cultural information were shared across the Atlantic façade through intense and regular contacts (Gibson 2013). At this period, the region represents a common sociocultural area, and the likelihood of a regionwide lingua franca used by specialist metallurgists and their patrons is strong. An error in the earlier chronological schemes has been rectified in the light of a calibrated C14 date in the 10th century BC for the Huelva deposition (Ruiz-Gálvez 1995) and also the recognition that Huelva swords were significantly earlier forerunners of the principal ‘carp’s tongue’ type (Brandherm 2007). The combined evidence now shows that the heyday of the Atlantic Bronze Age in the western Iberian Peninsula occurred in Bronce Final 2

John T. Koch

10

# # # #

##

#

# # # # ## # ## # # # # ##

## # #

#

300 km

### ### ### ###

#

## ### #

{

12 unprovenanced Britain

##

G

# #

G

#

##

# #

G

#

# #

G G

# # ## # # ## # # ## # # ##

# # ##

G ## #

$$$$$$## # $

{

G #

$ $

#

600 km

$$ $$$ $ $$ $$$ $ #

##

G

#

# ## # ## $$ $#

# # ##### # ##### # #

16 unprovenanced Ireland

$$

$ $$ $$$$

$

#

$ $$

G G$ $

$$

$

$

$

$$# $

$ $$

$$$ $

$ $ $ $ $$ $ $$ $

# $$$ # $ $$ #$ $$$$ # $ # $ $$$ #$ # # #

$

# # # # #

G

hoard or probable hoard

#

river/single find

domaine d’origine des épées pistilliformes à languette (Milcent 2009, fig. 2)

$ burial or presumed burial extensions possibles de ce domaine (Milcent 2009, fig. 2)

Fig. 2.1. The distribution of Gündlingen swords (after Gerloff 2004) and the Trans-Manche zone of the prototype (after Milcent 2009)

2.  Once again, Herodotus, the Κελτοί, the source of the Danube, and the Pillars of Hercules (1150/1100–1000/950 BC in Burgess and O’Connor’s 2008 scheme), during which Mediterranean imports were common and iron was coming into use. At the next stage, Bronce Final 3, 950–850/800 BC, the range of bronze types decreases and evidence for contacts with the Atlantic façade north of Pyrenees declines. After that, 850/800–700 BC, there are no close correspondences in the Peninsula for the characteristic types of the Llyn Fawr Phase in Britain or Hallstatt C in central Europe. In other words, by this period, the Peninsula is completely out of the Atlantic system and in the Mediterranean Iron Age as a result of the Orientalization catalysed by the Phoenician colonies in the south (Burgess and O’Connor 2008). The western Peninsula and Atlantic zone north of the Pyrenees still appear as a meaningful sociocultural area at the time of the Huelva deposition, but much less so after c.900 BC. The corresponding linguistic area split when the Atlantic sociocultural area broke up as part of the ‘PreColonial Orientalization’ of Iberia’s Bronze to Iron Age transition. The primary split within the Celtic family tree is between Hispano-Celtic, on the one hand, and GaulishBrittonic-Goidelic, on the other, which fits neatly with Iberia’s departure from the Atlantic Bronze Age, soon after the end of Bronce Final 2 1000/950 BC. After c.900 BC, the evident contacts across the Pyrenees or Bay of Biscay would have been insufficient to ensure the sharing of linguistic innovations between the cognate Indo-European dialects on either side. As McCone (2008) explains, both HispanoCeltic and Gaulish-Brittonic-Goidelic had inherited *Keltoi as a self-designation of their speakers. The spread of the Gündlingen swords and other Ewart Park-derived high-status material to west-central Europe at Ha C1a (p. 9 above) occurred after the Iberian Peninsula’s departure from the Atlantic Bronze Age. Therefore, the linguistic area corresponding to this expanding sociocultural area would not have been the undifferentiated Proto-Celtic, but its Gaulish-Brittonic-Goidelic descendant. These successive developments – the break-up of the Atlantic Bronze Age and expansion of the northern Atlantic area to include west-central Europe – can be seen as cause and effect. Bronce Final 2 (1150/1100– 1000/950 BC) was characterized by abundant Atlantic and Mediterranean material in the western Peninsula (Burgess and O’Connor 2008). Clearly the trans-Gibraltar route was vigorously in operation. There were probably not many long-distance journeys from the northern Atlantic façade to the Mediterranean or the reverse. However, in the late 2nd millennium BC these macro-regions were connected in a world system of interlocking subsystems. This chain broke with Orientalization in the Iberian Peninsula. The subsequent spread of Atlantic élite forms to west-central Europe reflects the reinvigoration of the alternative (fluvial) routes to the Mediterranean by way of the Rhône, Danube, and Alpine passes to the Po (Koch 2013b).

11

The Celticity of the South-western (SW) or ‘Tartessian’ inscriptions Many scholars have now recognized substantial Celtic linguistic content in this corpus of approximately 100 inscriptions from south Portugal and south-west Spain. The presence of Celtic names and an overall Celtic interpretation of the language of the texts was proposed by Correa (1989; 1992), who subsequently revised his view to a hypothesis of Celtic names within an unknown, probably non-IndoEuropean, matrix language (1996). Untermann (1997, 161– 168) again identified Celtic names in the corpus and also showed that recurrent forms in the corpus were inflected as Indo-European verbs. Ballester (2004, 118–121) published an extensive list of forms from the South-West inscriptions with Celtic comparanda, not limited to names. Villar recognized Celtic in the SW corpus and noted specific dialect features differing from Celtiberian, suggesting that this was ‘a kind of “frühes Gallisch”’ (2004, 268). In the standard work on the Celtiberian language, Jordán tentatively classified the language of the SW inscriptions as ‘¿macrofamilia indoeuropea? ¿familia celta?’ (2005, 7). In the third volume of the report on the Medellín necropolis (see p. 13 below), Almagro-Gorbea, Lorrio, Mederos, and Torres published a list of Celtic names in the corpus as part of an argument for the Atlantic Celtic background of the population of the site (2008, 1050–52). Another point in their argument was to recognize the Celticity of the name of the Tartessian king ’Αργανθωνιος, as has been proposed many times, including recently Ballester (2004, 119) and De Bernardo Stempel (2006, 47): Proto-Celtic *arganto- ‘silver’, the principal export of Tartessos. ’Αργανθωνιος is mentioned twice by Herodotus (§1.161–163). His phenomenal 80-year reign would notionally have been c.625–c.545 BC. In reporting the September 2008 discovery of what is now the longest of the SW inscriptions, the Mesas do Castelinho stela, its discoverer Guerra (2009; 2010) identifies in the text an Indo-European third person plural present-tense verb teebaantii, and a personal name tiilekuur- which survives as TILLEGVS in early Roman Galicia (note also the Gaulish genitive TILLICI). In a substantial multidisciplinary work, Villar, Prósper, Jordán, and Fernández (2011, 100) say that the case for the Celticity of the language of the SW inscriptions (meaning not just borrowed elements, but the matrix language as well) was sufficiently strong that Tartessian should be removed, provisionally at least, from the list of non-Indo-European languages. Recent work on the decipherment and on Celtic classification is summarized by Broderick (2010, 304–6) and Maier (2012, 210–213). The SW inscriptions have been notoriously hard to date, although the situation has been clarified significantly lately (see below p. 13). A sizable minority of the stones that have identifiable archaeological contexts belong broadly within the Iberian Early Iron Age, 9th to 5th centuries BC.

John T. Koch

12

150km



Lepontic inscriptions, 6th–1st century BC

Fig. 2.2. The source of the Danube and the Lepontic inscriptions

2.  Once again, Herodotus, the Κελτοί, the source of the Danube, and the Pillars of Hercules As to the implications for the Celtic passages of Herodotus, the evidence of the SW corpus shows that Celtic was spoken beyond the Pillars of Hercules, in and around the country of the Κυνητες, during and before the 5th century BC. Writing this early has not been found along the upper Danube, i.e. the other place Herodotus locates Κελτοί. But at no great distance from there, over the Alpine passes near lakes Como and Maggiore, there are inscriptions in the Lepontic language, which is also Celtic, and the earliest of these date back to the 6th century BC. The Lepontic and Tartessian inscriptions show related, but distinct, Celtic languages. In other words, undifferentiated Proto-Celtic had divided into dialects before, or at least by, the Early Iron Age (Fig. 2.2).

The necropolis of Medellín-Konisturgis In sources pertaining to the beginning of the Roman period, peoples on the scale of ciuitates or ‘tribes’ living in both the north-west and south-west of the Iberian Peninsula are sometimes listed under the more extensive group name that was written Κελτικοί in Greek and Celtici in Latin. According to the Greek Strabo writing in the early 1st century AD and probably working from the lost text of Posidonios, Κονιστοργις (= Latin Conisturgis) was the most famous town of the south-western Κελτικοί. Konisturgis means ‘town of the Κονιοι/Conii’, and, as was mentioned above, this people’s name is generally understood as a variant of Κυνησιοι and Κυνητες, the names that Herodotus applied to the ‘westernmost people of Europe’, neighbours of the Κελτοί. As a result of a review of the ancient historical sources on the Κονιοι and Κονιστοργις, Alarcão (2001) came to the conclusion that the town was at the site of modern Medellín, Badajoz, Spain, on the upper Guadiana. This identification has since been generally accepted, most importantly by the team who published the major report on the Early Iron Age necropolis of Medellín, leading to their explanation that a major indigenous group, the KonioiKunētes, were established over a territory spanning more than 300 km, from Sagres to the upper Guadiana, during the Early Iron Age (Almagro-Gorbea, Lorrio, Mederos and Torres 2008, 1034–59). Three objects with Tartessian writing on them were recovered from the necropolis, and close archaeological dating was possible on these: a fragmentary funerary stela of 650–625 BC (J.57.1 = T1 86H/12), a ceramic vessel with graffiti of 625–600 BC (T2 86H/13-1), and another graffito on ceramic of the later 6th century BC (T3 86/TP-1). The stela shows not only the same epigraphic style as was in use far to the west in Portugal, but also the same language, containing the previously attested forms lokoon, keeloia, and naŕkee[. Therefore, a common writing system and common language were evidently in use across the extensive territory of the Konioi-Kunētes. That this language was Celtic, or at

13

least contained Celtic elements in the broadest consensus, is implied yet again by this particular example. lokoon at Medellín, as on the great Fonte Velha VI inscription from a necropolis in the western Algarve, is probably the same word as Celtic LOKAN, translated as Latin VRNVM, i.e. ‘funerary urn, interment’, on the bilingual Cisalpine Gaulish-Latin funerary inscription from Todi, near Rome. As to the implications of Medellín-Konisturgis for the chronology of the SW corpus overall, Almagro-Gorbea (2008, 751–771) argues that stela J.57.1 represents a mature second stage of south-western literacy, at which point the practice was widespread geographically and no longer limited to a small specialist élite. It follows that the tradition goes back significantly earlier than the Medellín example of 650–625 BC. Similar conclusions have lately been reached approaching the question from different angles. For example, Ruiz-Gálvez (2008; 2009) argues that the SW inscribed stones probably began before the last of the approximately 125 Late Bronze Age ‘warrior stelae’ or, at least, not long after. The categories overlap in a few stelae with both Tartessian texts and warrior images, such as Capote (J.54.1) and Cabeza del Buey IV. Brandherm (2013b) argues that the Early Iron Age stelae with writing should be understood as forming a continuum with Late Bronze Age warrior stelae and the ‘Alentejo’ stelae which depict Middle Bronze Age weaponry of c.1600–c.1300 BC. The latter have a similar geographical distribution to the stelae with writing, and use similar burial rites, i.e. stone cists cutting through circular or sub-circular pavements on the surface. Arguably, the incorporation of (Celtic) vernacular texts into a revived funerary rite signals that these both form part of a nativist reaction to the cultural upheaval of Orientalization. Though writing itself was new and foreign inspired, it is self-consciously presented as part of something old and indigenous. Brandherm also explains that the identifiable equipment (swords, shields, crested helmets, etc.) depicted on the Late Bronze Age warrior stelae includes no types later than the 10th century BC. In other words, the warrior-stelae phase was evidently moribund by c.900 BC, and the transition to the stelae with writing might start nearer that date (Fig. 2.3).

Synthesis Previously, it had been difficult to understand why Herodotus spoke of an identifiable people with the common name Κελτοί located both at the source of the Danube and in Europe’s extreme south-west. Either Herodotus’s seemingly specific geographical reference points were badly off, or the arrival of the Κελτοί in the south-west had no clear trail. Recent work suggests a solution that can be concisely summarized as follows. • *Keltoi was the self-designation of the speakers of Proto-Celtic.

John T. Koch

14

150 km

Fig. 2.3. The Pillars of Hercules and the south-western Iberian Peninsula in later prehistory and protohistory

•

•

•

Languages descended from Proto-Celtic were spoken in extreme south-west Europe and probably at the source of the Danube by Herodotus’s day (c.484–c.430 BC). Proto-Celtic probably emerged in the west as the result of contact between Proto-Indo-European and the region’s p-less non-Indo-European languages, Iberian and Aquitanian/Palaeo-Basque. The Atlantic Bronze Age formed a sociocultural area

•

embracing Britain, Ireland, Armorica, and the western Iberian Peninsula. This phenomenon is also most easily understood as forming a linguistic area. Proto-Celtic is the most likely candidate as the area’s lingua franca. As the result of the Iberian Peninsula’s precocious Bronze-Iron Transition and Orientalization, it left the Atlantic Bronze Age, and Proto-Celtic split into Hispano-Celtic and Gaulish-Brittonic-Goidelic by c.900 BC.

2.  Once again, Herodotus, the Κελτοί, the source of the Danube, and the Pillars of Hercules

western Iberian Peninsula from 1300/ 1250 BC

Huelva

Iron AGE

* Kel to i

Hispano-Celtic

Ταρτησσος

5th century BC Herodotus

Alpine westcentral Europe

* Kel to i

IBErIAn EArly

by c. 700 BC

from c. 600 BC

Atlantic Gaul

AtlAntiC lAtE BronzE AGE Gaulish-Brittonic-Goidelic

*Keltoi

650–625 BC

Britain

At l A n t i C l At E B r o n z E A G E Proto-Celtic

1000/950 BC

c. 800–c. 750 BC

Ireland

15

Gündlingen swords, &c. lATE DoWrIS BA

EAr l I EST Iro n AGE

HAllSTATT C1a

Gaulish-Brittonic-Goidelic * Kel to i

Tartessian literacy Stage 1

HAllSTATT C2

Tartessian stela at Medellín/ Konisturgis

HAllSTATT D Lepontic inscriptions

Ἀργανθωνιος IBErIAn lATE Iron AGE

Ηa D / lT A

Hispano-Celtic dialects

Κελτοί

Κυνητες

Early Gaulish

Early Gaulish

[Κελτοί]

Κελτοί

Fig. 2.4. Proposed relative and absolute chronologies for group names, languages, and sociocultural areas in the Atlantic zone and west-central Europe

•

At about 800–750 BC, Gündlingen swords and other ‘Atlantic’ (i.e. Ewart Park-derived) élite metalwork types appear in west-central Europe. Some, at least, of the makers and users of this material spoke the Gaulish-Brittonic-Goidelic dialect of Celtic and selfidentified as *Keltoi (Fig. 2.4).

Afterword Some may read the foregoing proposal as a vindication of the traditional, mid 20th-century concept of Celticity. That would be a backward-looking misunderstanding. The ‘Celtic from the West’ model shares little common ground with the old standard position exemplified by the quotations from

John T. Koch

16

Powell above. Was the whole ‘Celtosceptic’ controversy a time waster as Vincent Megaw now reflects (2012), or a ‘deconstructionist distraction’ as per Harding and Gillies (2005)? Some of us would have gained little from this debate had the Megaws (1996; 1998; 2000) not brought to it a perspective that was at once philo-Celtic, New World, multicultural, and outsider. What we have learned is to disaggregate Celtic categories carefully (Renfrew 2013) and allow them to tell their own stories – ethnonyms, languages, specific features of material culture, ancient cultural stereotypes, and modern national myths (cf. Karl 2010). Without this, no progress would be possible. La Tène art is still Celtic in its recurrent association with Celtic languages and with groups called – and calling themselves – Κελτοί and related names. If, as expected here, Celtic studies survives and widens its outlook to include the Palaeohispanic world and the European Bronze Age, La Tène and its resonances will decline as a percentage of the whole and other kinds of art will come to be appreciated as Celtic.

Note 1

Accents are not written on the Greek spellings of uncommon non-Greek names, because these cannot be expected to reflect the accent position in the source languages, such as Celtic.

Bibliography Alarcão, J. de 2001. Novas perspectivas sobre os Lusitanos (e outros mundos), Revista Portuguesa de Arquelogia 4/2, 293–349. Almagro-Gorbea, M. (ed.) 2008. La necrópolis de Medellín II: estudio de los hallazgos, Bibliotheca Archaeologica Hispanica 26–2/Studia Hispano-Phoenica 5–2. Madrid: Real Academia de la Historia. Almagro-Gorbea, M., Lorrio, A., Mederos, A. and Torres M. 2008. La ciudad-estado de Medellín-Conisturgis y la cuenca del Guadiana en el periodo orientalizante. In M. AlmagroGorbea (ed.) La necrópolis de Medellín III–V, Bibliotheca Archaeologica Hispanica 26/Studia Hispano-Phoenica 5, 1033–59. Madrid: Real Academia de la Historia. Ballester, X. 2004. Hablas indoeuropeas y anindo­europeas en la Hispania prerromana, Estudios de lenguas y epigrafía antiguas – ELEA 6, 107–38. Brandherm, D. 2007. Las espadas del Bronce Final en la Península Ibérica y Baleares, Prähistorische Bronzefunde IV, 16. Stuttgart: Franz Steiner Verlag. Brandherm, D. 2013a. Westward Ho? Sword-Bearers and all the rest of it .... In J. T. Koch and B. Cunliffe (eds.) Celtic from the West 2: Rethinking the Bronze Age and the Arrival of IndoEuropean in Atlantic Europe, Celtic Studies Publications 16, 147–56. Oxford: Oxbow Books. Brandherm, D. 2013b. Mediterranes, Atlantisches und Kontinentales in der bronze- und ältereisenzeitlichen Stelenkunst der Iberischen Halbinsel. In G. Kalaitzoglou and G. Lüdorf (eds.)

Petasos: Festschift für Hans Lohmann zugeeignet von sein Schülern, Freunden und Kollegen zu seinem 65. Geburtstag, 131–48. Bochum: Ruhr-Universität. Broderick, G. 2010. Die vorrömischen Sprachen auf der iberischen Halbinsel. In U. Hinrichs (ed.) Handbuch der Eurolinguistik, 287–320. Wiesbaden: Otto Harrassowitz Verlag. Burgess, C. and O’Connor, B. 2008. Iberia, the Atlantic Bronze Age and the Mediterranean. In S. Celestino, N. Rafel and X.-L. Armada (eds.) Contacto cultural entre el Mediterráneo y el Atlántico (siglos XII – VIII ANE), La precolonización a debate, Serie Arqueológica 11, 41–58. Madrid: Escuela Española de Historia y Arqueología en Roma-CSIC. Collis, J. 1997. Celtic myths, Antiquity 71, 195–201. Collis, J. 2003. The Celts: Origins, Myths & Inventions. Stroud: Tempus. Correa, J. A. 1989. Posibles antropónimos en las inscripciones en escritura del SO. (o tartesia), Veleia 6, 243–52. Correa, J. A. 1992. La epigrafía tartesia. In D. Hertel and J. Untermann (eds.) Andalusien zwischen Vorgeschichte und Mittelalter, 75–114. Cologne: Böhlau. Correa, J. A. 1996. La epigrafía del Sudoeste: estado de la cuestión. In F. Villar and J. d’Encarnação (eds.) La Hispania prerromana: Actas del V Coloquio sobre Lenguas y Culturas Prerromanas de la Península Ibérica, 65–75. Salamanca: Ediciones Universidad de Salamanca. Cunliffe, B. 2001. Facing the Ocean: The Atlantic and its Peoples 8000 BC–AD 1500. Oxford: Oxford University Press. Cunliffe, B. 2009. A race apart: insularity and connectivity, Proceedings of the Prehistoric Society 75, 55–64. Cunliffe, B. 2010. Celticization from the West: the contribution of archaeology. In B. Cunliffe and J. T. Koch (eds.) Celtic from the West: Alternative Perspectives from Archaeology, Genetics, Language and Literature, 13–38. Oxford: Oxbow Books. Cunliffe, B. 2013. Britain Begins. Oxford: Oxford University Press. De Bernardo Stempel, P. 2006. Language and historiography of Celtic-speaking peoples. In S. Rieckhoff (ed.) Celtes et Gaulois, l’archéologie face à l’histoire: Celtes et Gaulois dans l’histoire, l’historiographie et l’idéologie moderne, 33–56. Glux-en-Glenne: Bibracte 12/1. De Bernardo Stempel, P. 2008. Linguistically Celtic ethnonyms: towards a classification. In J. L. García Alonso (ed.) Celtic and Other Languages in Ancient Europe, 101–18. Salamanca: Ediciones Universidad de Salamanca. De Sélincourt, A. (trans.) 1954. Herodotus: The Histories. Harmondsworth: Penguin. Freeman, P. 2010. Ancient references to Tartessos. In B. Cunliffe and J. T. Koch (eds.) Celtic from the West: Alternative Perspectives from Archaeology, Genetics, Language and Literature, Celtic Studies Publications 15, 303–34. Oxford, Oxbow Books. Gerloff, S. 2004. Hallstatt fascination: ‘Hallstatt’ buckets, swords and chapes from Britain and Ireland. In H. Roche, E. Grogan, J. Bradley, J. Coles, and B. Raftery (eds.) From Megaliths to Metals: Essays in Honour of George Eogan, 124–54. Oxford, Oxbow Books. Gerloff, S. (with J. P. Northover) 2010. Atlantic Cauldrons and Buckets of the Late Bronze and Early Iron Ages in Western Europe: With a Review of Comparable Vessels from Central

2.  Once again, Herodotus, the Κελτοί, the source of the Danube, and the Pillars of Hercules Europe and Italy, Prähistorische Bronzefunde II, 18. Stuttgart: Franz Steiner. Gibson, C. 2013. Beakers into Bronze: tracing connections between Western Iberia and the British Isles 2800–800 BC. In J. T. Koch and B. Cunliffe (eds.) Celtic from the West 2: Rethinking the Bronze Age and the Arrival of Indo-European in Atlantic Europe, Celtic Studies Publications 16, 71–100. Oxford: Oxbow Books. Guerra, A. 2009. Novidades no âmbito epigrafia pré-romana do sudoeste hispânico, Acta Palaeohispanica X / Palaeohispanica 9, 323–38. Guerra, A. 2010. Newly discovered inscriptions from the southwest of the Iberian Peninsula. In B. Cunliffe and J. T. Koch (eds.) Celtic from the West: Alternative Perspectives from Archaeology, Genetics, Language and Literature, 65–78. Oxford: Oxbow Books. Harding, D. W. and Gillies, W. 2005. Introduction: archaeology and Celticity. In W. Gillies and D. W. Harding (eds.) Celtic Connections: Papers from the Tenth International Congress of Celtic Studies, Edinburgh, 1995, 2, Archaeology, Numismatics, Historical Linguistics, 1–14. Edinburgh: University of Edinburgh Archaeology Monograph Series 2. Hoz, J. de 1992. The Celts of the Iberian Peninsula, Zeitschrift für celtische Philologie 45, 1–37. Isaac, G. R. 2005. Insular Celtic vs Gallo-Brittonic: an empirical and methodological question. In W. Gillies and D. W. Harding (eds.) Celtic Connections: Papers from the Tenth International Congress of Celtic Studies, Edinburgh, 1995, 2, Archaeology, Numismatics, Historical Linguistics, 190–202. Edinburgh: University of Edinburgh Archaeology Monograph Series 2. James, S. 1999. The Atlantic Celts: ancient people or modern invention? London: British Museum Press. Karl, R. 2010. The Celts from everywhere and nowhere: a reevaluation of the origins of the Celts and the emergence of Celtic cultures. In B. Cunliffe and J. T. Koch (eds.) Celtic from the West: Alternative Perspectives from Archaeology, Genetics, Language and Literature, 39–64. Oxford: Oxbow Books. Jordán Cólera, C. 2005. Celtibérico. Zaragoza: Ediciones del Departamento de Ciencias de la Antigüedad. Koch, J. T. 2003. Celts, Britons, and Gaels – names, peoples, and identities, Transactions of the Honourable Society of Cymmrodorion 2002, NS 9, 41–56. Koch, J. T. 2009a. (2nd edn. 2013) Tartessian: Celtic in the Southwest at the Dawn of History, Celtic Studies 13. Aberystwyth: David Brown. Koch, J. T. 2009b. On Celts calling themselves ‘Celts’ and related questions, Studia Celtica 43, 73–86. Koch, J. T. 2010. Paradigm shift? Interpreting Tartessian as Celtic. In B. Cunliffe and J.  T. Koch (eds.) Celtic from the West: Alternative Perspectives from Archaeology, Genetics, Language and Literature, 185–301. Oxford: Oxbow Books. Koch, J. T. 2013a. Prologue: Ha C1a ≠ PC (‘The Earliest Hallstatt Iron Age cannot equal Proto-Celtic’). In J. T. Koch and B. Cunliffe (eds.) Celtic from the West 2: Rethinking the Bronze Age and the Arrival of Indo-European in Atlantic Europe, Celtic Studies Publications 16, 1–16. Oxford: Oxbow Books. Koch, J. T. 2013b. Out of the flow and ebb of the European Bronze Age: heroes, Tartessos, and Celtic. In J. T. Koch and B. Cunliffe (eds.) Celtic from the West 2: Rethinking the Bronze Age and

17

the Arrival of Indo-European in Atlantic Europe, Celtic Studies Publications 16, 101–146. Oxford: Oxbow Books. Lorrio, A. J. and Ruiz Zapatero, G. 2005. Celtiberians: archaeology of the Celts in Iberia. In W. Gillies and D. W. Harding (eds.) Celtic Connections: Papers from the Tenth International Congress of Celtic Studies, Edinburgh, 1995, 2, Archaeology, Numismatics, Historical Linguistics, 33–55. Edinburgh: University of Edinburgh Archaeology Monograph Series 2. McCone, K. R. 1996. Towards a Relative Chronology of Ancient and Medieval Celtic Sound Change, Maynooth Studies in Celtic Linguistics I. Maynooth: Department of Old and Middle Irish, St Patrick’s College. McCone, K. R. 2008. The Celtic Question: Modern Constructs and Ancient Realities, Myles Dillon Memorial Lecture. Dublin: Dublin Institute for Advanced Studies. Mac Eoin, G. 1986. The Celticity of Celtic Ireland. In K.-H. Schmidt (ed.) History and Culture of the Celts, 161–74. Heidelberg: Winter. Maier, B. 2012. Geschichte und Kultur der Kelten. München: Verlag C. H. Beck. Mallory, J. P. 2013a. The Indo-Europeanization of Atlantic Europe. In J. T. Koch and B. Cunliffe (eds.) Celtic from the West 2: Rethinking the Bronze Age and the Arrival of Indo-European in Atlantic Europe, Celtic Studies Publications 16, 17–40. Oxford: Oxbow Books. Mallory, J. P. 2013b. The Origins of the Irish. London: Thames and Hudson. Megaw, J. V. S. 2012. The Flying Dutchman reaches port, Antiquity 86, 546–57. Megaw, J. V. S. and Megaw, M. R. 1996. Ancient Celts and modern ethnicity, Antiquity 70, 175–81. Megaw, J. V. S. and Megaw, M. R. 1998. ‘The mechanism of (Celtic) dreams?’: a partial response to our critics, Antiquity 72, 432–5. Megaw, M. R. and Megaw J. V. S. 2000. The Ancient Celts: the first European Community? In G. Evans, B. Martin, and J. Wooding (eds.) Origins and Revivals: Proceedings of the First Australian Conference of Celtic Studies, 559–97. Sydney: Centre for Celtic Studies, University of Sydney. Milcent, P.-Y. 2009. Le passage de l’âge du Bronze à l’âge du Fer en Gaule au miroir des élites sociales: une crise au VIIIe siècle av. J.‑C. ? In M. J. Roulière-Lambert, A. Daubigney, P.-Y. Milcent, M. Talon Marc, and J. Vital Joël (eds.) De l’Âge du Bronze à l’Âge du Fer en France et en Europe occidentale (Xe–VIIe siècle av. J.-C.); la moyenne vallée du Rhône aux âges du Fer, Actes du XXXème colloque international de l’AFEAF, co-organisé avec l’APRAB, Saint-Romain-en-Gal, 26–28 mai 2006, , 453–476. Dijon: Revue Archéologique de de l’Est, supplément 27. Milcent, P.-Y. 2012. Le temps des élites en Gaule atlantique: chronologie des mobiliers et rythmes de constitution des dépots métalliques dans le contexte européen (XIIIe–VIIe av. J.-C.). Rennes: Presses Universitaires de Rennes. Powell, T. G. E. 1958. The Celts. London: Thames and Hudson. Renfrew, C. 1987. Archaeology and Language: The Puzzle of Indo-European Origins. London: Cape. Renfrew C. 1990. Models of change in language and archaeology, Transactions of the Philological Society 87, 103–78. Renfrew, C. 2013. Early Celtic in the West: the Indo-European

18

John T. Koch

context. In J. T. Koch and B. Cunliffe (eds.) Celtic from the West 2: Rethinking the Bronze Age and the Arrival of IndoEuropean in Atlantic Europe, Celtic Studies Publications 16, 207–17. Oxford: Oxbow Books. Ruiz-Gálvez Priego, M. 1995. Cronología de la Ría de Huelva en el marco del Bronce Final de Europa occidental. In M. Ruiz-Gálvez (ed.) Ritos de paso y puntos de paso. La Ría de Huelva en el mundo del Bronce Final europeo, 79–83. Madrid: Editorial Complutense. Ruiz-Gálvez Priego, M. 2008. Writing, counting, self-awareness, experiencing distant worlds. Identity processes and free-lance trade in the Bronze Age/Iron Age Transition. In S. Celestino, N. Rafel, and X.-L. Armada (eds.) Contacto cultural entre el Mediterráneo y el Atlántico (siglos XII–VII ane). La precolonización a debate, 27–40. Madrid: Escuela Española de Historia y Arqueología en Roma. Ruiz-Gálvez Priego, M. 2009. ¿Qué hace un micénico como tú en un sitio como éste? Andalucía entre el colapso do los palacios y la presencia semita, Trabajos de Prehistoria 66/2, 93–118. Schrijver, P. 2011. Pruners and trainers of the Celtic family tree:

the rise and development of Celtic in the light of language contact (handout). XIVth International Congress of Celtic Studies, Maynooth. Trask, R. L. 1997. The History of Basque. London: Routledge. Untermann, J. (ed.) (with D. S. Wodtko) 1997. Monumenta Linguarum Hispanicarum IV: Die tartessischen, keltiberischen und lusitanischen Inschriften. Wiesbaden: Ludwig Reichert. Villar, F. 2004. The Celtic language of the Iberian Peninsula. In P. Baldi and P. U. Dini (eds.) Studies in Baltic and Indo-European Linguistics in Honor of William R. Schmalstieg, 243–74. Amsterdam: John Benjamins. Villar, F., Prósper, B. M., Jordán, C. and Fernández Álvarez, M. P. 2011. Lenguas, genes y culturas en la prehistoria de Europa y Asia suroccidental. Salamanca: ediciones Universidad de Salamanca. Wodtko, D. S. 2013. Models of language spread and language development in Prehistoric Europe. In J. T. Koch and B. Cunliffe (eds.) Celtic from the West 2: Rethinking the Bronze Age and the Arrival of Indo-European in Atlantic Europe, Celtic Studies Publications 16, 185–206. Oxford: Oxbow Books.

3 THE SHEFFIELD ORIGINS OF CELTIC ART John Collis

It was in 1993 when I first crossed swords with Ruth and Vincent over the matter of the Celts, at a conference in Cardiff organised by the Board of Celtic Studies. Though I had published sceptical comments before (Collis 1984; 1985), this was my first public lecture dealing specifically with the subject of the Celts. I was discussing the increasing precision of nomenclature in the second half of the first millennium BC, and re-iterating the well-known fact that no classical author ever called the inhabitants of Britain and Ireland ‘Celts’ (Collis 1997a). The Megaws had been asked to sum up the conference, but spent most of their time with a pre-written attack on my contribution and claiming that I had said that the Celts never existed (their slide of my vision of a ‘Celt’ was blank!). It is a claim that has often been reiterated not only applying to me, but all ‘Celtosceptics’, especially in the popular archaeological press in Britain and also in Spain and Germany. It is implied again in Vincent’s latest paper (Megaw 2012, 554), but I never said it! The next stage of the debate was at the International Celtic Congress (ICC) in Edinburgh in 1994 where the Megaws used the privilege of a plenary session, at which no questions or discussion were permitted, to again attack my supposed views on the Celts (Megaw and Megaw 1999). My own contribution at that conference was not particularly controversial, looking at the ideas and methodology of George Buchanan, the first person to suggest that the Irish and Scots originated from the Celts of Spain around the 5th century BC (Buchanan 1582; Collis 1999). The controversy was picked up by Dennis Harding and Willie Gillies (Harding and Gillies 2005), the two editors of the archaeological volume of that conference, when it finally appeared in 2005 (Gillies and Harding 2005). This mainly attacked Simon James (1999), and claimed that this ‘antiCeltic’ stance was confined to a small number of English

archaeologists, and had not affected either continental studies of the Celts or other facets of Celtic Studies such as linguistics, though they noted at the end the recent appearance of my book on the Celts in 2003. An attempt to provoke a discussion at the ICC conference in Cork in 1998, with papers by myself, Simon James, Patrick SimsWilliams and Gonzalo Ruiz Zapatero proved abortive as no-one was willing to argue the traditional case, though Vincent was in the audience, and it was not until the ICC conference at Bonn in 2008 that I was able to refute the points made by Harding and Gillies (Collis 2010). By the time of the ICC meeting in Maynooth in 2012 the critique of traditional studies had become more widely accepted, and the new president, Patrick Sims-Williams, in his paper at the conference, discussed the implications of the new ideas, especially the concept of ‘continuity’, and has subsequently presented some of his present views in writing (SimsWilliams 2012a; 2012b). The most public discussion, however, took place in the pages of Antiquity, launched by a paper by the Megaws (1996) which dealt with questions such as multiple identities, but which also claimed that the whole discussion was an exclusive domain confined to ‘Little Englander’ archaeologists who had an anti-Celtic agenda. In an emotive text they claimed that this was an academic equivalent of ‘genocide’. It provoked responses from Simon James (1998), Patrick Sims-Williams (1998b) and myself (Collis 1997b), though the then editor asked me to tone down my language, something he had not done with the original article! A subsequent editor has brought together the papers and other relevant discussion in Antiquity back to the 1930s (Carr and Stoddart 2002). In another paper Sims-Williams introduced the term ‘Celtosceptic’ to encompass those who reject the traditional views of the Celts (Sims-Williams

John Collis

20

1998a), while admitting recently that he personally is at least a ‘crypto-Celtosceptic’ (pers. comm.)! More recently, I have attempted to show that it is the traditionalists whose views can be construed as ‘anti-Celtic’ as they lean heavily on stereotypes taken from classical sources, claiming the Celts were illiterate, drunken, childish and warlike barbarians, and ignoring, for instance, the widespread evidence of literacy from the 7th–5th centuries BC onwards, and of Celts who achieved high office in the Roman world, including one who became emperor (Avitus) and another, Pompeius Trogus, who was certainly from a Celtic speaking family and probably a Celt, and who wrote a world history. The Celts of antiquity such as Pompeius Trogus, Martial and Sidonius Apollinaris chose to write in the more ‘universal’ and scholarly languages of Greek and Latin (Collis 2009).

Defining the Celts In any discussion of ‘Celtic Art’ we must start by defining our terms. Though on several occasions Vincent has queried the use of the term ‘Art’ in this context, I am not sure what the problem is. True that generally it appears as an ornamentation on functional objects, and is not like a painting or a sculpture which can often form an end in itself and so is genuinely ‘art for its own sake’. But in its craft-like character it is no different from Baroque Art, or Art Nouveau, or Art Deco, so I have no hesitation in using the term, and I object to modern critics who have high-jacked the term for their own ahistorical definition. As with the Celts, I prefer to go back to the original classical texts for what it means! More contentious is the adjective ‘Celtic’ which most archaeologists assume was applied because it was the art of the ancient Celts. This is decidedly not so, except by the most circuitous of routes, and it was not for 50 years after the term had been coined and defined that continental archaeologists linked it with objects found in the territories occupied by the ancient Celts, though British archaeologists had done so from the 1860s. The definition of the ancient Celts is normally assumed to be language, and we find this continuously repeated especially by linguists in the recent literature, for instance by Graham Isaac (2010, 165): Since it has no meaning to speak of the ancient or prehistoric ‘Celts’ independently of statements about the nature of the languages they spoke, there is no question of the origin of the Celts that is not by definition a question of the origins of the Celtic languages. Without language, there are no Celts, ancient or modern, but only populations bearing certain genetic markers or carriers of certain Bronze Age and Iron Age material cultures. The origin of the Celts therefore is the prehistory and the protohistory of the Celtic languages.

Equally emphatically Meid (2010) states on the cover of his book: ‘“Celtic” is, in the first instance, a linguistic concept,

and disregarding this linguistic foundation must lead to an impasse. It is the proven relationship of the so-called “Celtic” languages and their derivation from a common ancestor which justifies this scientific concept’. This view is also still reiterated by some archaeologists (e.g. Cunliffe 2010, 20). They could not be more wrong. The linguistic definition of the ‘Modern Celts’ is correct, for the Scots, Irish, Welsh, Bretons, Manx and Cornish, but it is not true of the ancient world; it was the people who gave their name to the language. So Celts spoke celtice, Gauls gallice and Britons britannice (or some such term as lingua or sermo accompanied by an adjective). There was no reliable classification of languages until the rise of the German school of Philology in the early 19th century. Both in the ancient world and in the Renaissance, supposed similarities between languages were no more than guesses, usually wrong and contradictory (e.g. Caesar versus Strabo on the similarity, or lack of it, of languages of Gaul). One or two got it right like Beatus Rhenanus (1485–1547; van Hal 2013–4) and George Buchanan (1582) and especially Edward Lhuyd (1707), though he seems to have changed his mind as in a letter of 1698 he links Irish with ‘Old Latin’ and British with Greek (Gunther 1945, 400). Linguists usually quote Joseph Scaliger (1599) as the first to list the ‘original’ European language groups (linguae matrices) spoken at the Tower of Babel, four major groups and seven minor, but he, despite his presumed contacts with Buchanan, failed to see the link between Irish and Welsh and he put them into separate groups. As I have argued elsewhere, it was the Abbé Pezron who mistakenly thought Breton was the last survival of the language spoken by Caesar’s Celtae and equated the people with the language (Collis 2003; Pezron 1703; Jones 1705). Had Lhuyd, rather than wrongly calling the group ‘Celtic’, followed Buchanan who called the language group ‘Gallic’, or had he recognised the insular origin of Breton and that it was a ‘Britannic’ language, the subsequent history of the British Isles might have been very different and there would be no ‘Celtic Studies’! But by the end of the 18th century because of the language it was generally assumed that the ancient inhabitants of the British Isles, Ireland and Armorica were ‘Celts’, and the Celtic-speaking people of western and northern Britain, of Ireland and of Brittany are their descendants (Collis 2007). The truth is that we do not know what the definition of a Celt was in the ancient world, and we have to look at the way the term was used on an author by author basis, from the generalised view of Ephorus for whom they were the people of the West, to Caesar who defines a specific group in central Gaul. In part it must be ethnic, but most discussion in the ancient literature just uses stereotypes in terms of physical appearances and habits, so one assumes that language and self-identification by the Celts themselves may have played a part, but it is also in part geographical,

3.  The Sheffield origins of Celtic Art and so the ‘polis keltike’ used to describe Narbo is almost certainly geographical (a town in Gaul) rather than ethnic as recently claimed by Cunliffe (2010, 18). The one source describing this area suggests the people were ‘Iberian’, and certainly this area of southern France was Iberian speaking (Gayraud 1981; Collis 2003, 127; 2010).

Naming Celtic Art In their 1989 overview of Celtic art the Megaws give a brief summary of the origin of the term, how it was originally coined in the 1850s and 1860s and how, on the continent, the major examples of what we now consider to be Celtic art like the Schwarzenbach bowl were considered to be imports from the classical world, from Greece, Italy or southern France. In this we are in general agreement, indeed I based my own ideas on theirs, and our differences are mainly ones of emphasis. We agree that the first person to identify the art style as pre- or protohistoric is John Kemble in his lecture in Dublin in 1857, and after his sudden and premature death a few weeks later, we owe the publication of this lecture to Augustus Franks in Horae Ferales and of the plates that Kemble had commissioned, which came out in 1863, though it is clear from the text that Franks originally pointed out the interest of the group of finds in the British Museum and so initiated Kemble’s study, the preparation of the plates and the text accompanying them are largely the work of Franks (Kemble, Franks and Latham 1863). But references to Kemble’s lecture had already started appearing in the Irish literature (Wilde 1861). Horae Ferales has been considered the first attempt to define the characteristics which we still assign to the prehistoric art style e.g. the ‘trumpet scrolls’, which are characteristics not found in contemporary classical art (for references to trumpet shapes, see Wilde 1861, 519, 567, 569). So it was assumed that they were of local insular origin, and as we have seen above, the inhabitants of Britain and Ireland at the period the art was produced were wrongly thought to be Celts, and so it was labelled ‘Celtic Art’. Where we disagree is in the relative importance of the scholars who developed the concept in the late 19th and early 20th centuries. The Megaws give prominence to Sir Arthur Evans, who in his Rhind lectures given in Edinburgh in 1895 pointed out the classical background to aspects of the art, but they were never fully published, except as summaries in the newspaper The Scotsman, and more extensively in a local journal in Liverpool (Evans 1897), so they are not much referenced in British, let alone continental, writings. Also the Mediterranean background he gives is very diffuse ranging from the Bronze Age of Greece and Crete to the classical period. In contrast the Megaws make no mention of the first book entitled Celtic Art, that by J. Romilly Allen in 1904, a book which has been reprinted on several

21

occasions in the 20th century, especially as a source of designs. In it he provides a good overview of the stage of scholarship and thought in Britain at that time. He uses the same nomenclature as his predecessors Kemble and Franks, that is that ‘Early Celtic Art’ belonged to the peoples of the Early Bronze Age whom he assumes to be ‘Goidelic Celts’, and whose arrival was marked by Beaker people (based on the evidence of craniometry). Spiral patterns (including those found on megalithic tombs) he links back to the Near East, following Evans, whom he cites, and are still to be found on Scandinavian metalwork contemporary with the Hallstatt period on the continent. Curvilinear ‘Late Celtic Art’ is dated from the Iron Age to post-Roman times, with a division between pagan and Christian phases, and was introduced by the Belgic ‘Brythonic’ peoples, along with La Tène features such as brooches, which started on the continent around 300 BC. He sees one of the Celtic characteristics both in the art and in the coinage as one of ‘borrowing’ rather than innovating new ideas, but placing a distinct Celtic stamp on the ornamentation. He also provides extensive lists and descriptions of prehistoric, Roman and post-Roman finds in Britain along with many illustrations of designs such as knotwork, from Irish manuscripts and crosses, and in this context he cites the work of Westwood (see below). Interestingly he also refers to the Hallstatt and La Tène ‘Cultures’, as does Evans, early examples of the use of the concept, as against Hallstatt and La Tène simply representing chronological phases. He also sees ‘La Tène’ as being a Swiss culture in contrast to the ‘Marnian’ in France, and, following Evans, he considers La Tène to have developed from the preceding Hallstatt Culture. The book forms a considerable contrast with the first French book on L’Art Celtique by Charles Roessler and published in 1908. Though he mentions British and Irish scholars such as Petrie, Greenwell, Westwood, Sir John Evans, as well as Romilly Allen (Roessler 1908, 27, ix, 26, v, and xii respectively) there is no coherent discussion of the art style, and the book consists largely of random notes and summaries along with some discussions contributed by British and French colleagues but in no special order or relevance, ranging from the palaeolithic female head from Brassempouy, and the Early Bronze Age burial from Upton Lovell in Wiltshire, to the oppida of northwest France, and Ogham script. The few elements which might be labelled as ‘Celtic Art’ include numismatics, the ‘Tara’ brooch, the burial from La Gorge Meillet (by the excavator Fourdrignier), helmets and the Book of Durrow. The title has little relevance to the contents, there is no historical narrative or coherent story, and the book provides no contribution to the subject, indeed it is a definite step backwards from Kemble, Franks and Romilly Allen. But for me the major omission from their overview is Joseph Déchelette (1914); they jump from Evans (1895) to Jacobsthal’s Early Celtic Art published in 1944. Déchelette’s

John Collis

22

Fig. 3.1. John Obadiah Westwood (1805–1893).

contribution is fivefold. First, he was the first scholar on the continent to accept the general definitions and nomenclature of the British scholars, and he cites both Horae Ferales and J. Romilly Allen. Secondly, he accepts a wider range of objects as being locally made in temperate Europe rather than as classical imports, and so to be included in the category of ‘Celtic Art’; thus he shifts the emphasis of discoveries from Britain and Ireland to Germany, Switzerland and France, at least for the Iron Age; thirdly, he also increases the number of motifs which are used to define prehistoric Celtic art. Fourthly, he develops his own chronology for the Iron Age (Hallstatt I, II, La Tène I, II, III) parallel with the system Reinecke was developing in southern Germany Hallstatt C, D, La Tène A, B, C, D). Finally, with this chronology, he was able to date the earliest finds to the 5th century BC using associations with imported classical objects (Attic Red Figure Ware, Etruscan bronze vessels), and to suggest the origin lay in the zone from northern France, southern Germany and Bohemia, especially on the central Rhine and lower Mosel. This is a basic framework which is still largely accepted today, and I would suggest his importance for developing ideas of Celtic art is comparable with that of Jacobsthal.

The origins of ‘Celtic Art’ As mentioned above, the concept of ‘Celtic Art’ and its first definition happened in Britain and Ireland in the 1850s and 1860s. In this section I wish to add a new name to the discussion as well as to explore the preconceptions of the

pioneers in a bit more detail. The name to add is that of John Obadiah Westwood (1805–1893; Foote 2004) (Fig. 3.1) who was born and educated in Sheffield (hence the title of this article), and he is the first person that I have been able to find, to write an article on the subject, published in 1856 (the year before Kemble’s lecture in Dublin). It appeared in a very influential book: The Grammar of Ornament, edited by Owen Jones (1809–1874), and published in 1856 (Figs 3.2, 3.3 and 3.4). The aim of the book was to provide coloured designs and patterns which could be employed by artists and architects, and taken from a wide range of cultures across the world: Arabian, Turkish and Persian ornament and especially Moresque which Jones had studied at first hand in Granada; also Chinese, Indian, and Turkish and ethnographic examples such as Maori art and more local styles from the Renaissance. The book was not only widely disseminated but it was also a seminal publication because Jones was pioneering the new printing method of chromolithography which allowed the high quality reproduction of coloured plates; it is a book which is still in print (Jones 2010). In his introduction Jones says: ‘Mr. J. O. Westwood having directed special attention to the Ornament of the Celtic races, has assisted in the Celtic Collections, and written the very remarkable history and exposition of the style’ (Jones 2010, 20). This can only mean that it is Westwood who devised the name some time before the book was published and he then wrote and illustrated the section on ‘Celtic Ornament’, though he also uses the term ‘Celtic Art’ in his text (Westwood 1856, 90). He is mainly known as an entomologist, and in 1861 was appointed to the new Hope Chair of Zoology in Oxford, but he was also at the time the leading expert on early medieval Anglo-Saxon, British and Irish Christian manuscripts. Linking his two interests was also a reputation as an artist of high quality. He had published an article in 1853 comparing the traits of some of these manuscripts, though the terminology he uses is ‘Irish’ and ‘British’ which he contrasted with the ‘Anglo-Saxon’ styles, so he started using the term ‘Celtic’ soon after that article appeared. In his 1856 article he states: The Origin of Celtic Ornament. – The various styles of ornament described above were practiced throughout Great Britain and Ireland from the fourth or fifth to the tenth or eleventh centuries; and as they appear in their purest and most elaborate forms in those parts where the old Celtic races longest prevailed we have not hesitated to give the Celtic as their generic name (Westwood 1856, 94: Jones 2010, 298).

Of its local insular origin he says: We have thus endeavoured to prove that, even supposing the early artists of these islands might have obtained the germ of their peculiar styles of ornament from some other sources than their own national genius, they had, between the period of the introduction of Christianity and the beginning of the eighth

3.  The Sheffield origins of Celtic Art 2

23

3

1 4

5

Fig. 3.2. Westwood 1856, Figure 63, Lapidary ornamentation. From the stone crosses of Aberlemno (1), St. Vigeans’s (2), Inchbrayoe (3), Meigle (4) and Eassie (5). Lithographer: Francis Bedford. The New York Public Library. www.victorianweb.org/art/design/jones24.html

century, formed several very distinct systems of ornamentation, perfectly unlike in their developed state to those of any other country (Westwood 1856, 96).

He lists a number of characteristics which he considers to be typical of various classes of object: the Christian manuscripts from Ireland and northern Britain (e.g. Lindisfarne), but also in continental contexts where there are clear Irish connections such as in the library of the monastery of St. Gall in Switzerland; on Irish stone crosses and their British derivatives; ‘circular objects of bronze of unknown use’ from Ireland; and on ‘small circular enamelled plates of early Anglo-Saxon work, found in different parts of England’ – what are now called escutcheons from ‘hanging bowls’ (Bruce-Mitford and Raven 2005). The motifs include ‘narrow ribbons interlaced and knotted’; ‘monstrous animals … extended into long interlacing ribbons intertwining together in the most fantastic manner’; a ‘pattern of diagonal lines, never interlacing; angulated lines … forming a series

Fig. 3.3. Westwood 1856, Figure 64. Interlaced style. Lithographer: Francis Bedford. The New York Public Library. www.victorianweb. org/art/design/jones25.html

of steps’; and the use of ‘red dots or points … as marginal ornaments of the great initial letters’. There were also negative characteristics, the lack of vegetal/floral motifs typical of Late Saxon art, for instance of the Winchester School. They are motifs which are largely confined to early Christian Irish art, but there is also one motif which was to become a characteristic of Kemble’s prehistoric Celtic art, the trumpet scroll: The most characteristic, however, of all the Celtic patterns, is that produced by two or three spiral lines starting from a fixed point, their opposite extremities going off to the centres of coils formed by other spiral lines… This pattern has been called the trumpet pattern, from the spaces between any two lines forming a long, curved, design, like an ancient Irish trumpet, the mouth of which is represented by a small pointed oval placed transversely at the broad end (Westwood 1856, 93).

Thus, for Westwood, the art was confined to the early Christian period and its production to Ireland and Britain.

24

John Collis

Fig. 3.4. Westwood 1856, Figure 65. Spiral, diagonal, zoomorphic, and later Anglo-Saxon ornaments. Lithographer: Francis Bedford.

3.  The Sheffield origins of Celtic Art The trumpet pattern he had already described in his earlier article in 1853, implying that it was already a widely accepted specialist term; it was its use to define ‘Celtic Art’ which was the innovation. It was also a key feature for Kemble: There is a peculiar development of the double spiral line, totally unknown to the Greeks and Etruscans, and the nations of the Teutonic North, which is essentially characteristic, not only of the Scoto-Keltic, but the Britanno-Keltic populations of these islands. If the lines are allowed to diverge, instead of following one another closely in their windings, they produce that remarkable pattern which since a few years we have been in the habit of calling the trumpet pattern. … You have a peculiar characteristic, a form of beauty which belongs to no nation but our own, and to no portion of our nation but the Celtic portion.   The engraved spiral line, with double winding, is found from America to the Baltic, from Greece to Norway, but the divergent spiral line (repoussé, in metal of a later date) and ornaments with champlevé enamel, is found in these islands alone, or in the neighbourhood of these islands; it vanishes in proportion to its distance from them. There is in all this not the slightest trace of the influence of what we call classical art. The trumpet pattern is neither Greek, nor Roman, nor Oriental. There is nothing like it in Etruscan art; there is nothing like it in German or Slavonic Art; there is little like it in Gallic or Helvetian art; it is indigenous, Gentlemen, – the art of those Keltic tribes which forced their way into these islands of the Atlantic, and, somewhat isolated here, developed a peculiar, but not less admirable system of their own (Kemble, Franks, and Latham 1863, 79, 80).

Other features which were considered peculiarly Celtic were the use of enamel and coral inlay; according to Philostratus this was feature found on the Atlantic coast (Kemble, Franks, and Latham 1863, 80). For Kemble there was little to compare with it on the continent: ‘There are traces of it, faint and poor, but sufficient for identification, among the Kelts of Normandy and the Keltic Helvetians’ (Kemble, Franks, and Latham 1863, 79). And likewise for Franks: But in the peculiar class of antiquities now to be considered, the British Isles stand unrivalled; a few ancient objects analogous in design, may be found in various parts of the Continent, and more extended work in local museums may bring others to light, but the foreign contributions to this section [a list of British and continental objects by category] are scanty when compared to those of our own country (Kemble, Franks, and Latham 1863, 172).

Kemble (Haigh 2004) whose wife was German, from Göttingen, was more familiar with the museums in northern Germany (Berlin, Schwerin, Hanover) and had actually excavated on the Lüneburg Heath while based in Hanover, and so he was more familiar with northern styles of antiquities (‘Teutonic’ and ‘Slavonic’) rather than those of central and southern Germany where we now envisage the origin and concentration of objects designated as ‘Celtic’

25

to lie. It is Franks who recognised the parallels in France, southern to central Germany and Switzerland. In this he was helped by a flurry of publications in the 1850s and early 1860s, such as the finds from Hallstatt (Gaisberger 1848; Simony 1851), Tiefenau in Berne (Bonstetten 1855), from the Swiss lakes (Troyon 1855), the first publications of finds from La Tène (Keller 1858; 1860), and of the finds from the Mosel-central Rhine area such as the burials from Weiskirchen (Lindenschmit 1858). There were also new British finds like the Battersea shield found in 1857, and published as ‘Celtic’ though this is not argued in terms of the art style, rather that is was associated with ‘Celtic’ skulls (Cuming 1857; 1858). In summary, Franks concludes: 1. That the antiquities in question have been found chiefly within the limits of Celtic occupation. 2. That the patterns on them differ from those of the Danes, Saxons and Romans. 3. That these patterns continued in use (with modifications) among the peculiarly Celtic races of Ireland; though not in a pure state after the introduction of Christianity. Their Celtic origin is further shown by the employment of coral; by the use of the boar as a symbol; by the prevalence of enamelled decoration; by the discovery of the remains of war chariots; the length and material of the swords; and by the presence of chain-mail (Kemble, Franks, and Latham 1863, 188).

Both Kemble and Franks assumed that most of the ‘Keltic’ objects they were dealing with were pre-Roman on the grounds that they showed little influence of Roman styles of art and were not generally found in association with Roman finds. Kemble deals with two specific sites. Lindenschmit had dated the dagger from Weiskirchen as 4th century AD on the grounds of the presence of a ‘Roman’ vessel, but as Franks points out, it is in fact Etruscan, one of several examples north of the Alps, and pre-Roman in date. Likewise Bonstetten had dated Tiefenau as Late Roman, but ignored the presence of 30 pre-Roman ‘Gaulish’ coins in the find, as well as an absence of Roman material. The only reason for a late dating in Britain was an assumption that the British were just backward barbarians who could not have produced such fine objects at such an early date. The Swiss finds such as those from La Tène were generally assumed to be pre-Roman, and this was likely to be the date of much of the material under discussion, though some elements did continue into the Christian period in Ireland. He suggested a date of around 200–100 BC, contemporary with the introduction of coinage into Britain (Kemble, Franks, and Latham 1863, 189).

Conclusions I think we can say fairly definitely that it was Westwood who invented the term ‘Celtic Art’ and was the first to employ it; however it is a misnomer. Firstly it is based on the false

26

John Collis

assumption by Pezron in 1703 that Breton is descended from the language of the ancient Celtae as described by Caesar. It is not – it was introduced from Britain in the late/post Roman period. Lhuyd then wrongly applied the term to the related languages in Britain and Ireland, and, following Pezron, wrongly used the language to name the people; in antiquity it was the other way round, the people gave their name to the language. The language group was first recognised by Buchanan in the 16th century, and before him the term Celtic had never been used for the inhabitants of Britain and Ireland; under the influence of Lhuyd, by the late 18th century it was assumed, wrongly, that the ancient inhabitants of Britain and Ireland were Celts because they spoke a ‘Celtic’ language. The name was then applied to the art in the mid-19th century by Westwood because it was the art of the ancient ‘Celts’ of Britain and Ireland, but only from the start of the 20th century was the name applied to La Tène Art on the continent. This was largely the work of Déchelette, and he also provided the first firm chronology for its date and opened up the possibility of a wider range of decorative motifs for the definition of the art style. Vincent and I have been arguing about this for the last 20 years, both the definition of the Celts and whether we are justified in using the term to describe the art. My position is that it is not an art style confined to the Celts (in this I and the Megaws agree), nor was it used in antiquity by all people called ‘Celts’ (e.g. in Spain), but by using the term we are imposing an interpretation on the archaeological data (that Celts can be defined by their art), while cutting out more useful lines of research (e.g. how much was the spread of the art and other features influenced by a shared language). Art cannot be used as a proxy for ethnicity, though it may, like language, have played a part in the definition of identity. I suspect we shall never agree, but such academic disagreements help greatly in clarifying our own ideas, and without the conflict the Celtic debate would never have got off the ground. But the written word actually hides a friendship between us that goes back many years!

Bibliography Allen, J. Romilly 1904 (first edition). Celtic Art in Pagan and Christian Times. London: Methuen and Co. Bonstetten, G., Baron von 1855. Recueil d’antiquités suisses. Berne: Mathey. Bruce-Mitford, R. and Raven, S. 2005. The Corpus of Late Celtic Hanging Bowls, with an account of the Bowls found in Scandinavia. Oxford: Oxford University Press. Buchanan, G. 1582. Rerum Scoticarum Historia. Edinburgh: Alexander Arbuthnet. Carr, G. and Stoddart, S. (eds.) 2002. Celts from Antiquity, Antiquity Papers 2. Cambridge: Antiquity Publications. Collis, J. R. 1984. The European Iron Age. London: Batsford. Collis, J. R 1985. Review of B. Cunliffe Danebury: an Iron Age

Hillfort in Hampshire (1984), Proceedings of the Prehistoric Society 51, 348–49. Collis, J. R. 1997a. The origin and spread of the Celts, Studia Celtica 30, 17–34. Collis, J. R. 1997b. Celtic myths, Antiquity 71, 195–201. Collis, J. R. 1999. George Buchanan and the Celts of Britain. In R. Black, W. Gillies and R. Ó Maolalaigh (eds.) Celtic Connections, Vol. 1, Proceedings of the Tenth International Congress of Celtic Studies, 91–107. East Linton: Tuckwell Press. Collis, J. R. 2003. The Celts: Origins, myths and inventions. Stroud: Tempus Publishing. (Second revised edition 2006). Collis, J. R. 2007. Die Entwicklung des Kelten-Konzepts in Britannien während des 18. Jahrhunderts. In H. Birkhan (ed.) Kelten-Einfälle an der Donau, 111–26. Vienna: Verlag der Österreichischen Akademie der Wissenschaften. Collis, J. R. 2009. An emperor in Rome and other famous Celts. In S. Grunwald, J. K. Koch, D. Mölders, U. Sommer and S. Wolfram (eds.) Artefact: Festschrift für Prof Sabine Rieckhoff zum 65. Geburtstag, Universitätsforschungen zur prähistorischen Archäologie Band 172, Teil 2, 409–16. Bonn: Dr. Rudolf Habelt GmbH. Collis, J. R. 2010. Redefining the Celts. In S. Zimmer (ed.) Kelten am Rhein, Proceedings of the Thirteenth International Congress of Celtic Studies, 23–27 July 2007 in Bonn, Part 2, Philologie: Sprachen und Literaturen, Beihefte der Bonner Jahrbücher 58/2, 33–43. Mainz: von Zabern. Cuming, H. S. 1857. On the discovery of Celtic crania in the vicinity of London, Journal of the British Association 13, 237–239. Cuming, H. S. 1858. On further discoveries of Celtic and Roman remains in the Thames off Battersea, Journal of the British Association 14, 326–330. Cunliffe, B. 2010. Celticization from the West; the contribution of archaeology. In B. Cunliffe and J. Koch (eds.) Celtic from the West: alternative perspectives from Archaeology, Genetics, Language and Literature, 13–38. Oxford: Oxbow Books. Déchelette, J. 1914. Manuel d’Archéologie Préhistorique, Celtique et Gallo–Romaine. II–3: Second Age du Fer ou Époque de La Tène. Paris: Librairie Alphonse Picard et fils. Evans, A. J. 1895. The origins of Celtic Art. Report on the Rhind Lectures, Lecture 2. The Scotsman, December 12, 1895. Evans, A. J. 1897. Greek elements in Ancient British art, Transactions of the Liverpool Welsh National Society. Twelfth session, 1896–1897, 24–38. Foote, Y. 2004. John Obadiah Westwood (1805–1893). In H. C. G. Matthew et al. (eds.) Oxford Dictionary of National Biography, 318–319. Oxford: Oxford University Press. Gaisberger, J. 1848. Die Gräber bei Hallstatt im Österreichischen Salzkammergute, Jahresbericht des Museums FranciscoCarolinum 10. Linz. Gayraud, M. 1981. Narbonne Antique des origines à la fin du IIIe siècle. Paris: Boccard. Gillies, W. and Harding, D. W. (eds.) 2005. Celtic Connections. Papers from the Tenth International Congress of Celtic Studies, Edinburgh, 1995. Volume 2: Archaeology, Numismatics and Historical Linguistics, Archaeology Monograph Series 2. Edinburgh: University of Edinburgh. Gunther E. T. 1945. Early Science in Oxford. Vol. XIV. Life and

3.  The Sheffield origins of Celtic Art Letters of Edward Lhwyd, Second Keeper of the Musæum Ashmoleanum. Oxford: privately printed. Haigh, J. D. 2004. John Mitchell Kemble (1807–1857). In H. C. G. Matthew et al. (eds.) Oxford Dictionary of National Biography, 153–5. Oxford: Oxford University Press. Harding, D. W. and Gillies, W. 2005. Introduction: Archaeology and Celticity. In W. Gillies and D. W. Harding (eds.) Celtic Connections, Vol. 2. Proceedings of the Tenth International Congress of Celtic Studies, 1–14. East Linton: Tuckwell Press. Isaac, G. R. 2010. The origins of the Celtic languages; language spread. In B. Cunliffe and J. Koch (eds.) Celtic from the West: alternative perspectives from Archaeology, Genetics, Language and Literature. Oxford: Oxbow Books. Jacobsthal, P. 1944. Early Celtic Art. Oxford: Clarendon Press. James, S. 1998. Celts, politics and motivation in archaeology, Antiquity 72, 200–209. James, S. 1999. The Atlantic Celts. London: British Museum Press. Jones, D. 1705. The Antiquities of Nations; more particularly of the Celtae or Gauls, taken to be originally the same people as our Ancient Britains, by Monsieur Pezron, englished by Mr. Jones. London: S. Ballard. Jones, O. 1856 (first edition). The Grammar of Ornament. London: Day and Son. Jones, O. 2010. The Grammar of Ornament. Lewes: Ivy Press (a reprint of the 2008 edition by The Herbert Press, London, with an introduction by Iain Zaczek). Keller, F. 1858. Pfahlbauten, 2. Bericht, Mitteilungen der antiquarischen Gesellschaft in Zürich 12/3, 116, 151–3, 155. Keller, F. 1860. Pfahlbauten, 3. Bericht, Mitteilungen der antiquarischen Gesellschaft in Zürich 13, 104. Kemble, J. M., Franks, A. W. and Latham, R. G. 1863. Horae Ferales. Studies in the Archaeology of the Northern Nations. London: Lovell Read and Co. Lhuyd, E. 1707. Archaeologia Britannica, giving some account additional to what has been hitherto publish’d of the Languages, Histories and Customs of the Original Inhabitants of Great Britain, from Collections and Observations in Travels through Wales, Cornwal, Bas-Bretagne, Ireland and Scotland. Oxford: Oxford Theatre. Lindenschmit, L. 1858. Die Alterthümer unserer heidnischen Vorzeit, nach den in öffentlichen und Privatsammlungen befindlichen Originalien zusammengestellt und herausgegeben von dem Römisch-Germanischen Centralmuseum in Mainz durch Ludwig Lindenschmit. Volume 1. Mainz: Victor von Zabern Verlag.

27

Megaw, R. and Megaw, V. 1989. Celtic Art, from its Beginnings to the Book of Kells. London: Thames and Hudson. Megaw, R. and Megaw, V. 1996. Ancient Celts and modern ethnicity, Antiquity 70, 175–181. Megaw, R. and Megaw, V. 1999. Celtic connections past and present. In R. Black, W. Gillies and R. Ó Maolalaigh (eds.) Celtic Connections, Vol. 1. Proceedings of the Tenth International Congress of Celtic Studies, 19–81. East Linton: Tuckwell Press. Megaw, V. 2012. The Flying Dutchman reaches port, Antiquity 86, 546–557. Meid, W. 2010. The Celts. Innsbruck: Institut für Sprachen und Literaturen der Universität. Pezron, P.–Y. 1703. Antiquité de la Nation et de la Langue de Celtes autrement appellez Gaulois. Paris: Prosper Marchand. Roessler, C. 1908. L’art celtique, avant et après Colomban: mémoires relatifs à l’archéologie, à la numismatique et au folklore celtique publiés. Paris: Durand. Scaliger, J. J. 1599. Diatriba de Europaeorum linguis. In I. Casaubon (ed.) 1610 Opuscula Varia antehac non edita, 119–122. Paris: Hadrian Beys. Simony, F. 1851. Die Alterthümer vom Hallstätter Salzberg. Vienna: Kaiserlich-königliche Hof- und Staatsdruckerei. Sims-Williams, P. 1998a. Celtomania and Celtoscepticism, Cambrian Medieval Celtic Studies 36, 1–35. Sims-Williams, P. 1998b. Genetics, linguistics and prehistory: thinking big and thinking straight, Antiquity 72, 505–27. Sims-Williams, P. 2012a. Bronze- and Iron-Age Celtic-speakers: what we don’t know, what we can’t know, and what could we know? Language, genetics and archaeology in the twenty-first century, Antiquaries Journal 92, 427–49. Sims-Williams, P. 2012b. Celtic civilisation: continuity or coincidence? Cambrian Medieval Celtic Studies 64, 1–44. Troyon, F. 1855. Habitations lacustres. Lausanne: Georges Bridel. van Hal, T. 2013–4. From Alauda to Zythus: collecting and interpreting Old Gaulish words in Early Modern Europe. Keltische Forschungen, 6, 225–83. Wilde, W. R. 1861. A Descriptive Catalogue of the Antiquities of Animal Materials and Bronze in the Museum of the Royal Irish Academy. Dublin: Hodges, Smith and Co. Westwood, J. O. 1853. On the distinctive character of various styles of ornamentation employed by the early British, Anglo-Saxon and Irish artists, The Archaeological Journal 10, 273–301. Westwood, J. O. 1856. Celtic Ornament. In O. Jones The Grammar of Ornament, 89–97. London: Day and Son.

4 THEORIE DER KELTISCHEN KUNST. EIN VERSUCH Felix Müller

Celtic art has never fitted into traditional histories of art and art theory, not least because it lacks named artists. In this paper, Celtic art will be reconsidered in terms of sociology, ideology and politics to show that there is a real sense in which Celtic art is ‘art’. Einleitung Wenn wir im fünften Buch des Livius den Sieg an der Allia, die Eroberung Roms durch die Gallier lesen, pflegen wir uns kaum klarzumachen, dass Brennus und seine Scharen sehr gut gekleidet waren, dass diese Kelten seit mehr als hundert Jahren in naher Berührung mit der Zivilisation des Südens standen. Und auch die hellenistischen Keltendarstellungen sind als kulturgeschichtliche Quellen schwach, sie pointieren sehr einseitig den heroischen Aspekt des gefährlichen Nordvolks, der Bedränger Delphis und Roms.

Dieses Zitat aus dem Jahre 1934 ist trotz seiner etwas schwerfälligen Formulierung als ein Bekenntnis zu verstehen, festgefahrene Meinungen über die Wertigkeit von Kulturen zu überdenken, gleichzeitig ist damit aber auch die Aufforderung verbunden, kulturgeschichtliche Phänomene in einem größeren Zusammenhang, d.h. über die akademischen Fächergrenzen hinweg, zu betrachten. Nur vor diesem geistigen Hintergrund war es dem Klassischen Archäologen Paul Jacobsthal möglich, der keltischen Kunst zu einem eigenständigen Gesicht innerhalb des antiken Kunstschaffens zu verhelfen und ihre Leistung aus weltgeschichtlicher Perspektive in den Rang eines ersten großen Beitrags des Nordens an die Gesamtkunst Europas zu setzten (Jacobsthal 1934). Zu einer solchen Sichtweise befähigten Jacobsthal nicht nur seine Kennerschaft der Klassischen antiken Kunst sowie deren Randgebieten, sondern offensichtlich verbarg sich hinter dem wissenschaftlichen Engagement der Wunsch, einen kulturpolitischen Kontrapunkt zu setzen zu der im

Dritten Reich auch in Wissenschaftskreisen aufkeimenden Germanentümelei (Crawford and Ulmschneider 2011). Darin hatte er in Max Verworn (1919) einen bemerkenswerten, heute leider in Vergessenheit geratenen Vorreiter. Auf dem Gebiet der Kunst hatten die Germanenerben jedenfalls kaum etwas Gleichwertiges entgegenzusetzen. Vielleicht ahnte Jacobsthal das Problem der eigenen wissenschaftstheoretischen Rechtfertigung, weshalb er bewusst – oder unbewusst – auf eine Definition seiner Early Celtic Art verzichtet hat. Die starke Kohärenz der keltischen Kunst und die daraus zu ziehenden Schlüsse hatte er allerdings klar vor Augen, als er schrieb: In my opinion the whole of Celtic art is a unit. It is the creation of one race, the Celts (Jacobsthal 1944, 160). Heute würde man zwar ein anderes als das zeitbedingte Vokabular verwenden, das damals selbst Jacobsthal gebrauchen musste, um sich Gehör zu verschaffen; inhaltlich steht diese Aussage jedoch in einem klaren Gegensatz zur unterdessen viel gehörten Meinung, dass es die Kelten gar nie gegeben habe. Dazu fehle es an einer geographisch und zeitlich übergreifenden Kulturparenthese. Aber: Gibt es einen anschaulicheren Beweis für eine kulturelle Identität als die gemeinsam verständliche Sprache der Kunst?

Kunsttheorien und ihre Anwendung auf die keltische Kunst Trotz bescheidener Ansätze existiert meiner Meinung nach

4.  Theorie der keltischen Kunst. Ein Versuch keine allgemein gültige Theorie der keltischen Kunst. Oft hört man von zünftigen Kunsthistorikern sogar die Meinung, dass es sich in diesem Falle nicht um Kunst, sondern um blosses Kunsthandwerk handle. Tatsächlich setzt die akademische Geschichte der Kunsttheorie erst mit der Antike ein oder mit anderen Worten, in dem Moment, wo Künstler selber – beziehungsweise deren Kommentatoren – sagen, was unter Kunst zu verstehen sei. Letztlich geht es immer um eine Rechtfertigung, oft im Sinne einer Statusfrage des Künstlers, aber in jedem Fall um Abgrenzungen im Sinne einer sozialen Distinktionslogik bzw. einer Rangordnung (Schneider 2011, 8; ferner Hauskeller 2000). So kommen für die Antike zum Beispiel Platon und Aristoteles als kritische Autoritäten zum Zug; in der frühen Neuzeit sind es dann Giorgio Vasari, der ’Vater der Kunstgeschichte’, und andere. Entscheidend ist zudem, dass in historischen Epochen auch Künstler mit ihren Namen bekannt sind. In Ermangelung unangefochtener Fürsprecher fehlen die paläolithische, die keltische und sogar die ägyptische Kunst oft in den großen Kompendien der klassischen Kunstgeschichte: Ohne zeitgenössische Kunstkritik, scheint es auch keine Kunst zu geben. Aus ähnlichem Grunde tat man sich Anfangs schwer mit den Kykladenidolen, den kleinen Scheusalen aus Marmorsplittern (Overbeck 1857, 41); sie entsprachen nicht dem Kanon des griechischen Kunstideals im 19. Jahrhundert. Desgleichen wurden zum Beispiel afrikanische und ozeanische Kunstwerke erst als solche akzeptiert, nachdem Autoritäten wie Picasso sagten, dass sie das seien. Auffällig ist jedoch, dass jede Epoche eine eigene Kunsttheorie beansprucht, und dass es schwer fällt, eine allgemeine, zeitübergreifende Definition zu formulieren. Der keltischen Kunst fehlen nicht nur authentische Kommentatoren, sondern auch die Namen von Künstlern. Dem versuchte Martyn Jope bis zu einem gewissen Grade Abhilfe zu schaffen, indem er einen fiktiven Waldalgesheim Master postulierte, für den sich die Forschung allerdings nie erwärmen konnte (Jope 1971; Driehaus 1971; Joachim 1995, 25–27). Aus den schriftlichen Quellen ist nur ein einziger keltischer Künstler mit Namen bekannt, nämlich Helico, der aus dem Lande der Helvetier stammend in Rom tätig war. Aber in einem Atemzug mit seiner Nennung verweist ihn Plinius auch gleich wieder ins Reich der Legende (Plin. nat. 12, 2, 5). Helico zeichnete sich in der ars fabrilis aus, was ihn in dieser Formulierung eher zum Kunsthandwerker im heutigen Sinne gemacht hätte. Dies ist jedoch nicht weiter von Bedeutung, da in der Antike Kunsthandwerk und Kunst nicht auseinander gehalten wurden: Die Begriffe techne im Griechischen und ars im Lateinischen umfassen beides und enthalten nicht nur, was die Archäologie unter Technik versteht, sondern auch alle Spielarten der darstellenden Kunst. Auf die Plastik und Malerei bezogen fallen Werke darunter, die einen Gegenstand möglichst naturgetreu abbilden oder die eine

29

materialisierte Umsetzung einer abstrakten Idee mit den dazu geeigneten Stoffen darstellen. Platons Theorie der Ästhetik versteht darunter das Hingezogensein zum Schönen im Allgemeinen und zum menschlichen Körper im Besonderen. Die Schönheit als der höchste Endzweck und als Mittelpunkt der Kunst sah das 18. Jahrhundert in vollkommenem Masse in der griechischen Plastik verwirklicht (Winkelmann 1764, 139). In diesem Falle wurde nur ein einzelner Aspekt der griechischen techne, der genau genommen einer Ästhetik des 18. Jahrhunderts entsprach, zum Dogma erhoben und der Rest des Begriffsinhaltes in der Bewertung deklassiert (Müller 2009a, 26; Schneider 2011, 30, 119). Das bedeutet, dass die Antike das, was wir heute mit der Kunst der Kelten bezeichnen, als techne oder ars akzeptiert hätte – wenn ihr eben nicht ein anderes Hindernis im Wege gestanden hätte, nämlich die Barbarenideologie. Selbst Winkelmann traute wirklich große Kunst nur den Völkern Griechenlands und Italiens zu (Winkelmann 1764, 39–45). Im Mittelalter Europas dient die bildende Kunst ausschließlich der Verherrlichung Gottes (Hauskeller 2000, 21–32). Daneben hat die ars im Sinne von Handwerkstechnik auch etwas zu tun mit der materiellen Seite eines Objektes, indem sie die Fertigkeiten umfasst, die Qualität von Stoffen zu nutzen, um etwas Neues herzustellen, das es in der Natur sonst nicht gibt, wie zum Beispiel Glas und Email. Ein solches Verständnis tritt uns im Lehrbuch De diversis artibus von Theophilus Presbyter aus dem 12. Jahrhundert entgegen (Brepohl 1999, 21, 39), würde jedoch der keltischen Kunst in dieser Ausschließlichkeit ebenfalls nicht gerecht. Nachdem die keltischen Künstler samt ihren Fürsprechern schon lange verstummt sind, und solange der keltischen Forschung ihr Winkelmann fehlt, bleibt eine Definition der keltischen Kunst auf diesem Wege ein Desideratum – falls man die Diskussion, ob Kunst oder Kunsthandwerk, nicht überhaupt als einen Streit um des Kaisers Bart begreifen will und eigentlich zu den Akten legen sollte. Ein wertfreies Verständnis für die Entstehung von Kunst im Allgemeinen beruft sich auf den menschlichen Spieltrieb. Von ihm ausgehend unterscheidet Barasch (1998, 235–37) eine physioplastische Kunst, die wiedergibt, was das Auge sieht, und eine ideoplastische Kunst, bei der sich eine geistige Auffassung Ausdruck verschafft. Der Unterschied der beiden Darstellungsarten beruht nicht auf unterschiedlichen Talenten und Fähigkeiten in der Ausdrucksweise, da every period, society, or culture is fully capable of expressing what it wishes to say. Ausschlaggebend sind overall human condition and nature, also eine grundsätzlich verschiedene Sichtweise auf die Dinge der Welt. Damit nimmt Barasch ein Konzept auf, das der Physiologe Max Verworn (1908) entworfen hat und dem auch Alois Riegls Kunstwollen  zugrunde liegt, bei dem sich immer eine bestimmte Geisteshaltung in allen Kunstgattungen Ausdruck verschafft (Riegl 1927). Die keltische Kunst mit ihrem Hang zum Phantastischen und

30

Felix Müller

ihrer hohen Befähigung zur Abstraktion gehört eindeutig zur zweiten Gruppe der ideoplastischen Kunst während die griechisch-römische Skulptur mit ihrer Tendenz zum Realismus zur ersten Gruppe gerechnet werden muss. Entscheidend für die frühe Anerkennung der herausragenden Stellung der griechischen Plastik war das Erkennen von Stilen und deren historische Entwicklung durch Winkelmann (1764, 207–237). In gleicher Weise verschaffte Jacobsthal 1944 der keltischen Kunst als erster mit der Umschreibung des Early Style, des Waldalgesheim Style und der Späten Stile ihre eigene Komplexität, deren chronologische Tiefe er in den Addenda and Corrigenda seines Hauptwerks nochmals untermauerte. Genau darin, in der Herausbildung von Stilen und einer länger dauernden Stilgeschichte, unterscheidet sich die keltische Kunst zum Beispiel von der iberischen, der thrakischen oder der skythischen Kunst (Jacobsthal 1944, 162f., 206–8).

Eigenheiten der keltischen Kunst Seit der spätarchaischen Zeit des 6. Jahrhunderts v.Chr. besaß die Kunst der Griechen eine mächtige Strahlkraft, der sich kaum ein Nachbarvolk zu entziehen vermochte (Boardman 1994). Allerdings verhielten sich diese ‘Randvölker’ unterschiedlich. Die keltischen Künstler setzten, indem sie einzelne Elemente nicht bloß imitierten, sondern mit einem neuen Inhalt versahen, eine eigene Kreativität in Gang, die sich in Zeit und Raum entwickelte und vervielfältigte. Neben den bekannten Stilen verklammern verschiedene Eigenheiten die keltische Kunst zu einer kulturgeschichtlichen Einheit.

Tradition und Kontinuität Ihre stärksten Momente hat die keltische Kunst in der Ornamentik, deren Grundprinzip in der Aneinanderreihung gleicher Motive zu einem unendlichen Rapport besteht. Diese Tradition von vielleicht ursprünglich textilen

Vorbildern entlehnt widerspiegelt sich in unterschiedlicher Intensität von der hallstattzeitlichen Gürtelblechen, über die Röhrenkannen von Reinheim und Waldalgesheim, die Keramik von Clermont-Ferrand (Fig. 4.1) bis zu den irischen Zierscheiben von Donore (Müller 2009a, Motive 4, 18, 86 und 100).

Vorbild und Variation Die keltische Kunst hat sich nur ausgewählter Elemente aus dem griechischen Motivschatz bedient und zwar in gewissen Fällen mit einiger zeitlicher Verzögerung, wenn der Funke nicht sofort sprang, sondern es eines Anstoßes bedurfte, dessen Ursache allerdings unbekannt ist. So gelangte das griechische Motiv der Herrin der Tiere zum Beispiel mit der Hydria von Grächwil (und selbstverständlich mit vielen anderen, uns nicht überlieferten Importstücken) schon in der Mitte des 6. Jahrhunderts v.Chr. nach Mitteleuropa, erhielt aber erst einige Generationen später seine abstrahierte keltische Version auf Gürtelhaken und vielleicht auch auf Schwertscheiden. Das Motiv lässt sich als symmetrische Komposition Menschengestalt-Tiere bis in die römische Zeit verfolgen (Müller 2009a, 167–169). Bisweilen scheinen mediterrane Vorlage und keltische Version in einem Grabensemble vereint wie bei den Attaschen von Situla und Schnabelkanne im Grabhügel Kleinaspergle oder beim Rosettenmotiv auf Bronzekessel und Goldhalsring in Waldalgesheim (Lenerz-de Wilde in Kimmig 1988, 236–238; Frey 1995, 203). In Weiskirchen haben die Sphingen auf einem Goldblech aus Grabhügel II ihre motivischen Vorbilder im skythischen Gebiet und dienten vielleicht selber als Inspiration, als ein Künstler den bekannten Gürtelhaken fertigte, der im Nachbarhügel zum Vorschein kam (Krausse 1996, 188–189). Einstweilen ein Gedankenspiel ist die Herleitung selbst des Hauptmotivs des Cheshire Style, nämlich das schemenhafte Gesicht, das sich wie auf der vielzitierten Schwertscheide

Fig. 4.1. Einer der Höhepunkte keltischer Ornamentik bildet der Dekor auf bemalter Keramik aus Clermont-Ferrand (nach Guichard 1994).

4.  Theorie der keltischen Kunst. Ein Versuch a

31

Fig. 4.2. Das mögliche Vorbild und seine Variation im Cheshire Style, von oben nach unter: Grabkeramik aus Apulien, Schwertscheide aus Filottrano, Scheibenhalsringe aus Trebur und Braunfels (a. nach Lohmann 1979; b. nach Verger 1989; c. und d. nach Müller 1989).

b

c

d

von Filottrano im Rankenwerk versteckt. Könnte es sich dabei nicht um eine Variation der Rankenköpfe zwischen Spiralbändern auf apulischen Grabgefäßen handeln, die dort als Symbole für eine Unsterblichkeitshoffnung gelten (Fig. 4.2)? Schließlich liegen Apulien und die Marken nicht weit voneinander entfernt. Und ließe sich nicht sogar eine Verkettung herstellen bis zur (beim Tragen unsichtbaren) Nackenzier mit Maske auf Scheibenhalsringen (Lohmann 1979; Müller 1989, 53)? Vielleicht ergibt sich sogar inhaltlich eine logische Assoziation zwischen der Grabkeramik im griechischen Apulien und dem Schmuck verstorbener Keltinnen im südlichen Mittelrheingebiet? Motivgeschichtlich läge ein Parallelfall zur Herrin der Tiere vor. Im Verlaufe des 3. Jahrhunderts begann die kreative Übernahme von südlichen Vorbildern zu schwinden (siehe Müller 2009b, 313–314).

Ortswechsel von Personen. Viele Beispiele stammen aus der dramatischen Phase der keltischen Geschichte im 4. und 3. Jahrhundert v.Chr. Sie sind Zeugen einer über weite geographische Gebiete gültige Formensprache. Exakt übereinstimmend sind die Wellenranken vom Typ A2 auf Trägern unterschiedlicher Funktion (Fig. 4.3). In Niort an der französischen Atlantikküste zieren sie einen Kesselgriff, in Bussy-le-Château in der Champagne einen Halsring und im inneralpinen Spiez ein Armband; leicht abgewandelt könnte man auch die Schwertscheide von Filottrano zu der noch weit umfangreicheren Gruppe zählen. Ein verwandtes Rankenmotiv vom Typ A1 findet sich auf zwei Schwertern in der Champagne (Fundort Epiais-Rhus und Saint-Germainmont) und im Hinterland von Ancona (Fundort Moscano di Fabriano), was seinen Grund in den historisch überlieferten Wanderungen der Senonen aus dem Senonais an die Adriaküste haben könnte (Verger 1987; Kruta 2008). Noch dichtere und weiträumigere Beziehungen widerspiegeln sich in der Verbreitung von Schwertscheiden, die mit zwei sich gegenüberstehenden Greifen verziert sind. Ihre Besitzer lebten zwischen Südengland und dem Karpatenbecken. An vier weit auseinander liegenden Orten

Interner Austausch und Kohärenz Gleiche oder beinahe identische Motive auf Gegenständen von weit auseinander liegenden Fundorten zeugen entweder von einem regen Güteraustausch oder von raschem

32

Felix Müller

a

b

c

Fig. 4.3. Die keltische Wellenranke gleichen Typs auf unterschiedlichen Trägern und in weit auseinander liegenden Regionen, von oben nach unten: Kesselgriff aus Nior am Atlantik, Halsring aus Bussy-le-Château in der Champagne, Armband aus Spiez in den Alpen (a. und b. nach Verger 1989; c. nach Müller 1998).

scheint das Motiv nahezu identisch wiedergegeben (Ginoux 2007; Szabo and Petres 1992, 30f.): In Montigny-Lencoup bei Paris, in Ameglia bei La Spezia, auf dem Monte Bibele bei Bologna und in Taliandörögd am Plattensee in Ungarn (Fig. 4.4). Eine eigene Bildsprache in der Wiedergabe von Fabeltieren offenbart sich an dem geweihtragenden Vierbeiner mit sonderbar riesigen Hasenohren, einmal in Holz geschnitzt aus Fellbach-Schmiden (Württemberg) und einmal auf Keramik gemalt aus Clermont-Ferrand (Auvergne) und beide Male unabhängig von einander um 120 v.Chr. datiert (Müller 2009a, 234–241).

Süden nach Norden, von Oberitalien über die Champagne nach Südengland, besonders schön zum Ausdruck (Fig. 4.6); das griechische Vorbild sowie die drei Motive im Waldalgesheimstil lassen sich mit Ausnahme des letzten mit ziemlicher Sicherheit ins 4. Jahrhundert datieren. Wie verschieden die Sichtweise auf die Natur der Dinge und deren Abstraktion sein kann, zeigt das Motiv des Pferdekopfes von der Quadriga, die heute das Dach des Markusdoms in Venedig ziert. Sie stand etwa zur gleichen Zeit auf einem Triumphbogen des Nero in Rom, als in Britannien die Applike des ‘Stanwick Horse’ entstand (Fig. 4.7).

Transformation und Abstraktion

Farbigkeit

Geradezu als Markenzeichen der keltischen Ornamentik kann man das Sezieren und Neukomponieren von klassischen Vorbildern betrachten. Bereits der Early Style zelebriert das – ganz im Sinne einer ideoplastischen Kunst – mit großer Virtuosität, wenn aus einem griechischen Fries mit Dreiblattpalmetten und Lotusblüten ein keltischer Eigenbau entsteht (Fig. 4.5), eindrücklich demonstriert am goldenen Trinkhornbeschlag von Schwarzenbach im Saarland (Frey 1971). Das Spiel mit der zunehmenden Abstraktion in mehreren Graden ließe sich an verschiedenen Beispielen zeigen, kommt aber in einer Abfolge von

Zur Farbgebung der Kunstwerke lässt sich nur wenig sagen, da entsprechende naturwissenschaftliche Untersuchungen noch weitgehend fehlen. Die kräftig farbigen Glasarmringe lassen immerhin erahnen, dass die keltische Welt mindestens so bunt war, wie die klassisch griechische. Darin stimmen die schriftlichen Quellen überein, wenn Diodor die keltischen Krieger beschreibt (Diod. 5, 30). Diese stecken in Hemden in vielerlei Farben. Die Mäntel tragen streifig angeordnete, dicht ineinander gesetzte, vielfarbige, eckige Muster und entsprechen damit weniger der schachbrettartig gemusterten Bekleidung von Troubadix,

4.  Theorie der keltischen Kunst. Ein Versuch a

c

b

d

33

Fig. 4.4. Das weit verbreitete Motiv der antithetischen Greife in seiner Detailausführung, von oben nach unter: Montigny-Lencoup bei Paris, Ameglia bei La Spezia, Monte Bibele bei Bologna, Taliandörögd am Plattensee (a. and d. nach Szabó und Petres 1992; b. und c. nach Vitali 2003).

dem Barden von Asterix, als dem schottischen Tartan. Es sind aber auch die logischen Muster, wie sie bei Arbeiten am Webstuhl entstehen. Auch werden die mannshohen Schilde als in eigentümlicher Weise verziert beschrieben. Mit ‘eigentümlich’ könnte dann in diesem Zusammenhang nicht nur ‘farbig’ gemeint sein, sondern vielleicht sogar ‘mit typisch keltischen Mustern’.

Qualität Aussagen über Qualitätskriterien, wie sie in keltischer Zeit galten, sind nur schwer anzubringen, da Riegls Kunstwollen im Einzelfall kaum abzusetzen ist vom individuellen Können eines bestimmten Künstlers. Ein Beispiel aus der Hallstattzeit scheint mir aber exemplarisch. Das wohl reichste Grab im Verbreitungsgebiet der Alb-

34

Felix Müller

Fig. 4.5. Das Motiv Dreiblattpalmetten und Lotusblüten in der keltischen Abstraktion auf dem Trinkhornbeschlag aus Schwarzenbach im Saarland (nach Joachim 1995).

a

d

b

c

Hegau-Keramik stammt aus Gomadingen; es enthält u.a. ein Pilzknaufschwert mit goldplattiertem Griff, worin sich ein hoher Repräsentationsanspruch manifestiert. Zwei Prunkschalen sind die absoluten Spitzenstücke ihrer Art, was Größe, Farbigkeit, Sorgfalt im Entwurf des Dekors und Präzision in der Ausführung betrifft (Zürn 1987, 124–133). Detailgenauigkeit und Regelmäßigkeit in der Konzeption, wie wir sie heute verstehen, scheint also durchaus schon damals ein Gradmesser für Qualität

Fig. 4.6. Die natürliche, klassische Wellenranke und ihre zunehmende, keltische Abstraktion in ihrer doppelten Lesbarkeit von Süden nach Norden, von oben nach unten: Apulien in Süditalien, Sanzeno im Südtirol, Prunay in der Champagne, Brentford in Südengland (a. nach Bird 2003; b. nach Joachim 1995; c. nach Charpy 1991; d. nach Duval 1978).

gewesen zu sein. Deshalb ist es wohl nicht verfehlt, in den Unterschieden der zwei hier vorgestellten Entwürfe von Engstingen-Grossengstingen und Albstadt-Tailfingen zwei unterschiedliche Könnerschaften zu identifizieren (Fig. 4.8). Im gleichen Sinne liesse sich das auf die berühmten Bronzebleche von Comacchio an der Adria und den Ritzdekor auf einer Keramik aus Alsopel in Ungarn übertragen, denen beiden die entwickelte Wellenranke zugrunde liegt (Fig. 4.9).

4.  Theorie der keltischen Kunst. Ein Versuch

35

Fig. 4.7. Zwei grundverschiedene Darstellungsweisen desselben Motivs: Die Pferdeplastik einer antiken Quadriga (heute in Venedig) und die Applike des ‘Stanwick Horse’ aus Britannien (© Branislav L. Slantchev, San Diego; © Trustees of the British Museum).

a

b

Fig. 4.8. Sichtbare Qualitätsunterschiede in der Motivkombination auf im Alb-Hegau-Stil verzierter Keramik: Engstingen-Grossenstingen und Albstadt-Tailfingen (a. und b. nach Zürn 1987).

Felix Müller

36 a

b

Fig. 4.9. Die entwickelte Wellenranke in unterschiedlicher Qualität, von oben nach unten: Bronzeblech aus Comacchio in Oberitalien, Keramik aus Alsopel in Ungarn (a. nach Joachim 1995; b. nach Lajos 1933).

Weiterleben

Funktion der keltischen Kunst

Die römische Kunstindustrie hat die keltische Ornamentik weitgehend überdeckt, ohne dass kaum etwas übernommen oder weiterentwickelt worden ist. Alleine in Britannien haben sich keltische Traditionen über längere Zeit gehalten und sogar weiter entfaltet. Noch kaum erforscht ist jedoch, wie die germanische und die keltische Ornamentkunst zueinander stehen. Eine allmähliche Annäherung, Übernahme oder Weiterentwicklung ist zwar kaum sichtbar, aber einzelne Motive könnten ihren Weg durchaus vom keltische Gebiet nach Germanien genommen haben. Ein offenes Problem in dieser Diskussion bleibt der große zeitliche Hiatus zwischen dem Ende der keltischen Kunst auf dem Kontinent und dem feststellbaren Beginn der germanischen Ornamentik. Im Germanischen dominieren eigene figürliche Motive und Darstellungsweisen auf jeden Fall, aber auch ein wohlbekannter Hang zum Phantastischen und zur Abstraktion ist vorhanden. Im Detail fehlt der Forschung eine motivgeschichtliche Linie, die bis ins Keltische hinab reicht und uns zum Beispiel die verblüffenden ikonographischen Übereinstimmungen der S-förmigen Fabelwesen auf frühlatènezeitlichen Gürtelhaken und den merowingerzeitlichen Kleinfibeln aufzeigen könnte, um nur einmal gerade dieses Motiv anzusprechen (Brather-Walter 2009).

Es gibt nur wenige Anhaltspunkte, wozu die keltische Kunst hätte gedient haben können. Aufgrund der verwendeten Symbole und der verschiedenen Stile kann man von einer visuellen Sprache ausgehen, die wir allerdings nicht verstehen oder die uns mindestens nur zu einem kleinen und oberflächlichen Teil verständlich ist (Megaw and Megaw 1989, 19). Es bleibt letztlich nur der Versuch, Querverbindungen zu Bildthemen von zeitgleich existierenden literarhistorischen Kulturen des Mittelmeerraumes herzustellen (Eggert 2010). Wie tragfähig diese Brücken sind, muss im Einzelfall analysiert werden. Viele Forscher glauben, dass grundsätzlich alle urgeschichtlichen Bildwerke religiöse Bedeutung besitzen (Huth 2010) – unter der Voraussetzung, dass es sich eben um szenische Bilderzählungen handelt, die, gerade was die keltische Kunst betrifft, sehr rar sind. Ein gross angelegter Versuch legte Martin Guggisberg (2000, 189–203, 258–262) mit seiner Interpretation des Bilderzyklus auf den Erstfelder Goldhalsringen vor. Dort werden die ineinander verschränkten Doppelwesen mit dem Raubvogel als Begleiter oder als Kopfbekrönung in einen Zusammenhang gesetzt mit der mehrgestaltigen Potnia Theron, also dem mehrfach wiederholten und variierten Motiv der Herrin der Tiere. Diese Hypothese

4.  Theorie der keltischen Kunst. Ein Versuch beruft sich – zu recht oder zu unrecht – auf eine zeitgleiche, kulturübergreifende Bildsprache, was nur möglich ist, wenn auch ein Verständnis für den Inhalt vorhanden ist: Offenbar haben sich keltische Vorstellungswelten in den mediterranen Bildern wiedergefunden und sich in einer eigenen Ikonographie Ausdruck verschafft. Mit diesem Vorgehen ist mindestens ein dialektischer Anspruch in der Bildinterpretation eingelöst, der manch anderem Versuch von solchen Bildfriesen abgeht (Kruta 2007). Einer ähnlichen Logik folgt eine ikonographische Verkettung im Grab der Dame von Vix, der bis jetzt noch kaum Beachtung geschenkt worden ist (Lüscher 1991, 60). Dem gleichen mythologischen Zyklus entstammen die Gorgo (auf dem importierten Krater), der Amazonenkampf (auf der attischen Keramik) und der Pegasos, das Kind der Gorgo Medusa und ebenfalls Amazonenbezwinger (auf dem exklusiven, einheimischen Halsring). Ob das Zufall ist oder ob dahinter eine bewusste Auswahl der Auftraggeberin und Trägerin des Halsringes steckt? Auf jeden Fall wäre es verfehlt, die Kenntnis und das Verständnis mediterraner Kulturen der keltische Notabeln zu unterschätzen. Wenn die Deutung von Bildfriesen (als religiöse Botschaften) schon grösste Probleme bereitet, so ist es die Interpretation von Einzelmotiven erst recht, zumal nie sicher ist, ob das Dargestellte einen wirklichen Inhalt besitzt oder bereits schon zum Dekor verkommen ist, dem ein bloßer horror vacui zugrunde liegt. Im Moment scheint es ertragreicher, wenn man versucht, die keltischen Kunstwerke in einem soziologischen Zusammenhang auszudeuten, indem man fragt, zu welchem Zeitpunkt und woher welche Objekte, Motive und Stile stammen und als Beigaben in die Gräber gelangten. Zur Klärung dieser Fragen besitzt die vorgeschichtliche Archäologie die nötigen Instrumente und Methoden. Auf diese Weise wird offensichtlich, dass sich die hallstattzeitliche Oberschicht anders verhält als die frühlatène- oder spätlatènezeitliche, was man in Beziehung setzen kann mit dem Lauf der Geschichte (nicht mit historischen Einzelereignissen). Vor diesem Hintergrund wird die keltische Kunst zum Ausdruck der Herrschaftslegitimation einer tonangebenden Sozialgruppe mit einem identischen psychologischen Hintergrund. Insofern beinhaltet die keltische Kunst durchaus die eingangs erwähnte soziale Distinktionslogik Norbert Schneiders (2011, 8). In einem übergeordneten Sinn wird sie zum Abbild einer wenn auch nicht politischen, so doch ideologischen Identität einer größeren ethnischen Gruppe, und if art styles are accepted as evidence of self-image or cultural identity, then the term ‚Celtic’ is as valid as ‚European’ (Megaw and Megaw 1993, 221). In diesem Sinne wäre es angebracht, die keltische Kunst als eine Kunst zu bezeichnen, die – dann ohne den Begriff ’keltisch’ überstrapazieren zu müssen – am nordwestlichen Rand der antiken Welt entstanden ist. Ihre Identität unterscheidet sich unbestritten von

37

derjenigen der Mittelmeerkulturen. Im Verlaufe der zuerst kulturellen, dann politischen Durchdringung verwischen sich die Konturen in den letzten Jahrhunderten vor Christus allmählich – aus welthistorischer Sicht zu Ungunsten der keltischen Seite. Ob die keltische Kunst dennoch von den kulturhistorischen Wissenschaften als gleichwertig mit den mediterranen Hochkulturen akzeptiert wird, hängt weitgehend von der Intensität ab, mit der wir Forscher und Forscherinnen uns dafür einsetzen. Und vielleicht erwächst uns in der Zukunft mit der Neubearbeitung der Early Celtic Art sogar eine Art keltischer Winkelmann...

Acknowledgements Mein Dank geht an Jolanda Studer, Bernisches Historisches Museum, für Bildbearbeitung und an Arnd Kerkhecker, Universität Bern, für Übersetzungshilfe der Diodortexte.

Bibliographie Barasch, M. 1998. Modern Theories of Art, 2. From Impressionism to Kandinsky. New York: New York University Press. Bird, S. 2003. Greek designs. London: British Museum. Boardman, J. 1994. The diffusion of classical art in antiquity. London: Thames and Hudson. Brather-Walter, S. 2009. Schlange-Seewesen-Raubvogel? Die S-förmigen Kleinfibeln der älteren Merowingerzeit, Zeitschrift für Archäologie des Mittelalters 37, 47–100. Brepohl, E. 1999. Theophilus Presbyter und das mittelalterliche Kunsthandwerk. Band 1. Köln. Weimar, Wien: Böhlau. Charpy, J.-J. et al. (eds.) 1991. Le céramique peinte celtique dans son contexte européen, Actes du Symposium international d’Hautvillers, Octobre 1987, Mémoire de la Société Archéologique Champenoise 5, Suppl. au Bulletin Nr. 1. Reims: Société Archéologique Champenoise. Crawford, S. and Ulmschneider, K. 2011. Paul Jacobsthal’s Early Celtic Art, his anonymous co-author, and National Socialism: new evidence from the archives. Antiquity 85, 129–41. Driehaus, J. 1971. Zum Grabfund von Waldalgesheim, Hamburger Beiträge zur Archäologie 1, 101–13. Duval, P.-M. 1978. Die Kelten. München: C. H. Beck. Eggert, M. K. H. 2010. Hermeneutik, Semiotik und Kommunikationstheorie in der Prähistorischen Archäologie: Quellenkritische Erwägungen. In C. Juwig und C. Kost, (Hrsg.) Bilder in der Archäologie – eine Archäologie der Bilder?, 50–74. Tübinger Archäologische Taschenbücher, Münster, München, Berlin: Waxmann. Frey, O.-H. 1971. Die Goldschale von Schwarzenbach. Hamburger Beiträge zur Archäologie 1, 85–100. Frey, O.-H. 1995. Das Grab von Waldalgesheim. Eine Stilphase des keltischen Kunstandwerks. In H.-E. Joachim, Waldalgesheim. Das Grab einer keltischen Fürstin, 159–206. Pulheim: Rheinland-Verlag. Ginoux, N. 2007. Le thème symbolique de «la paire de dargons» sur les fourreaux celtiques (IVe–IIe siècles avant J.-C.). Etude

38

Felix Müller

iconographique et typologique, BAR International Series, 1702, Oxford: British Archaeological Reports. Guggisberg, M. A. 2000. Der Goldschatz von Erstfeld. Ein keltischer Bilderzyklus zwischen Mitteleuropa und der Mittelmeerwelt, Antiqua 32. Guichard, V. 1994. La céramique peinte des IIe et Ier s. avant J.C. dans le nord du Massif Central: nouvelles données, Études Celtiques 30, 103–136. Hauskeller, M. 2000. Was ist Kunst? Position der Ästhetik von Platon bis Danto, 5. Auflage. München: Beck. Huth, Ch. 2010. Früheisenzeitliche Bilderwelten – Eigenschaften und Aussagewert einer archäologischen Quellengattung. In C. Juwig und C. Kost (Hrsg.) Bilder in der Archäologie – eine Archäologie der Bilder?, 127–53. Tübinger Archäologische Taschenbücher, Münster, München, Berlin: Waxmann. Jacobsthal, P. 1934. Einige Werke keltischer Kunst, Die Antike 10, 17–45. Jacobsthal, P. 1944. Early Celtic Art. Oxford: The Clarendon Press. Joachim, H.-E. 1995. Waldalgesheim. Das Grab einer keltischen Fürstin. Pulheim: Rheinland-Verlag. Jope, E. M. 1971. The Waldalgesheim Master. In J. Boardman et al. (eds.) The European Community in Later Prehistory. Studies in honour of C. F. C. Hawkes, 167–80. London: Routledge & Kegan Paul. Kimmig, W. 1988. Das Kleinaspergle. Studien zu einem Fürstengrabhügel der frühen Latènezeit bei Stuttgart, Forschungen und Berichte zur Vor- und Frühgeschichte in Baden-Württemberg 30, Stuttgart: Theiss. Krausse, D. 1996. Hochdorf III. Das Trink- und Speiseservice aus dem späthallstattzeitlichen Fürstengrab von EberdingenHochdorf (Kr. Ludwigsburg), Forschungen und Berichte zur Vor- und Frühgeschichte in Baden-Württemberg 64, Stuttgart: Theiss. Kruta, V. 2007. La cruche celte de Brno. Dijon: Faton. Kruta, V. 2008. Les Sénons dans les Marches aux IVe et IIIe siècles avant J.-C. Etat de la question, Etudes Celtiques 36, 7–20. Lajos, M. 1933. Die Frühlatènezeit in Ungarn. Budapest: Archaeologia Hungarica XI. Lohmann, H. 1979. Grabmäler auf unteritalischen Vasen, Archäologische Forschungen 7, Berlin: Mann. Lüscher, G. 1991. Hallstattzeit: Zentren des Reichtums und der Macht. In A. Furger and F. Müller (eds.) Gold der Helvetier, 59–69. Ausstellungskatalog Schweizerisches Landesmuseum. Zürich: Schweizerisches Landesmuseum.

Megaw, J. V. S. and Megaw, M. R. 1989. Celtic Art. From the beginnings to the Book of Kells. London: Thames and Hudson. Megaw, J. V. S. and Megaw, M. R. 1993. Cheshire Cats, Mickey Mice, the New Europe and Ancient Celtic Art. In Ch. Scarre and F. Healy (eds.) Trade and Exchange in Prehistoric Europe. Oxbow Monograph 33, 219–32, Oxford: Oxbow. Müller, F. 1989. Die frühlatènezeitlichen Scheibenhalsringe. Römisch-Germanische Forschungen 46, Mainz am Rhein: von Zabern. Müller, F. 1998. Die Entwicklung des Waldalgesheimstils in Münsingen-Rain. In F. Müller (Hrsg.) Münsingen-Rain, ein Markstein der keltischen Archäologie. Funde, Befunde und Methoden im Vergleich. Akten des Internationalen Kolloquiums «Das keltische Gräberfeld von Mün­singen-Rain 1906–1996», Münsingen/Bern 9.–12. Oktober 1996. Schriften des Bernischen Historischen Museums 2, 71–83. Bern: Bernisches Historisches Museum. Müller, F. 2009a. Kunst der Kelten. Stuttgart: Belser. Müller, F. 2009b. The Celtic horror vacui and its demise on the continent. An essay on a La Tène fibula. In G. Cooney, K.Becker, J. Coles, M. Ryan and S. Sievers (Hrsg.), Relics of Old Decency. Festschrift for Barry Raftery, 305–14. Dublin: Wordwell. Overbeck, J. 1857. Geschichte der griechischen Plastik für Künstler und Kunstfreunde, Band 1. Leipzig: Hinrichs. Riegl, A. 1927. Spätrömische Kunstindustrie. Wien: Verlag der Öesterreichischen Staatsdruckerei. Schneider, N. 2011. Geschichte der Kunsttheorie. Von der Antike bis zum 18. Jahrhundert. Köln, Weimar, Wien: Böhlau. Szabo, M. and Petres, E. F. 1992. Decorated weapons of the La Tène Iron Age in the Carpathian Basin. Inventaria Praehistorica Hungariae 5, Budapest: Magyar Nemzeti Múzeum. Verger, S. 1987. La genèse celtique des rinceaux à triscèles, Jahrbuch Römisch-Germanisches Zentralmuseum 34, 287–339. Verworn, M. 1908. Zur Psychologie der primitiven Kunst. Jena. Verworn, M. 1919. Keltische Kunst. Berlin: Verl. der Sturm. Reprint Michigan 2012. Vitali, D. 2003. La necropoli di Monte Tamburino a Monte Bibele, Studi e scavi 19. Bologna: Gedit. Winkelmann, J. J. 1764. Geschichte der Kunst des Altertums. Nachdruck Wien 1934: Phaidon-Verlag. Zürn, H. 1987. Hallstattzeitliche Grabfunde in Württemberg und Hohenzollern. Forschungen und Berichte zur Vor- und Frühgeschichte in Baden-Württemberg 25, Stuttgart: Theiss.

5 LES CODES DE REPRÉSENTATION VISUELLE DANS L’ART CELTIQUE ANCIEN Laurent Olivier

Introduction: Les codes de representation visuelle a La Tène ancienne Réalisme intellectuel et codes de représentation visuelle En archéologie, autant il est essentiel, lorsqu’on étudie des objets manufacturés, de déterminer à quelle fonction technique ils étaient destinés et comment ils ont été fabriqués (Guillaumet 2003), autant il est indispensable, lorsqu’on étudie des représentations figuratives, de déterminer ce qu’elles représentent et comment elles sont visuellement construites. Or, si les figures zoomorphes ou anthropomorphes de l’art du Second âge du Fer ont été abondamment commentées du point de vue stylistique ou typologique, la plupart des auteurs ne se sont guère intéressés au problème de leur lecture, comme si celleci ne constituait pas un sujet de discussion. A la suite de Jacobsthal et de Paul-Marie Duval surtout, les spécialistes ont beaucoup disserté sur les processus de composition de motifs stylistiques et leur évolution typologique. En revanche, fort peu de chercheurs – à l’exception notable de Paul Jacobsthal lui-même – ont porté leur attention sur les procédés de représentation visuelle mis en œuvre pour produire les figures de l’art laténien. En effet, l’acte de représenter une figure quelconque nécessite qu’on la projette dans un espace défini par des dimensions spatiales spécifiques et qu’on la montre selon un point de vue particulier situé quelque part dans l’espace. La négociation de ces contraintes de représentation se traduit par la production de codes de représentation visuelle, qui conditionnent toute création de figures. Ces codes protohistoriques étant fondamentalement différents de ceux transmis par la tradition naturaliste de l’art occidental, les chercheurs du XIXe siècle en ont

inconsciemment déduit que ceux-ci étaient primaires, si ce n’est inconsistants. Cette approche a eu pour conséquence de repousser spontanément les figurations de l’art de la Tène aux confins de l’ornemental; en d’autres termes d’en faire une production purement décorative. Les deux propositions qui font de l’art de l’âge du Fer une manifestation ornementale, comme en particulier chez Joseph Déchelette ou chez Henri Hubert (Dechelette 1914, 1508–1509; Hubert 1932, 141–142), ou encore une manifestation «réfractaire aux images» (ou «aniconique», comme chez Raymond Lantier; Lantier 1946, 74) sont également fausses. Ces interprétations témoignent surtout d’une position idéologique informulée, qui place – dans une hiérarchie imaginaire des représentations artistiques – les figurations stylisées ou géométriques des sociétés «primitives» ou «barbares» à un stade inférieur, du point de vue du développement intellectuel des civilisations, à celui des figurations naturalistes des sociétés classiques (Semper 1861; Haddon 1895; Riegl 1923). Fondamentalement, l’acte apparemment simple de dessiner une figure sur un support revient à tenter de proposer une solution acceptable à un problème de représentation visuelle particulièrement ardu: il s’agit de déterminer les moyens de représenter un objet composé de plusieurs faces ou côtés visibles dans un espace – l’espace projeté sur le support du dessin – qui lui-même ne possède qu’une seule face. Dans un langage plus mathématique, on pourrait dire également qu’il s’agit de figurer sur un support à deux dimensions un objet développé en trois dimensions: dans ces conditions, comment rendre apparentes ces parties cachées de la figure (car situées derrière ou sur les côtés), qui sont masquées par le corps de la figure elle-même, selon l’angle d’observation visuelle choisi pour la représenter? A ce problème élémentaire, la représentation des scènes ajoute un degré de complexité supplémentaire, dans la mesure

40

Laurent Olivier

Fig. 5.1c. Arrivée d’un marchand dans un camp cheyenne (dessin d’un Amérindien adulte, fin XIXe à début XXe siècle). Exemple de représentation par rabattement. Les tipis, qui sont disposés autour d’un espace de forme rectangulaire, sont rabattus de manière à être tous représentés en élévation. Ce procédé s’applique aussi aux personnages qui sont placés dans le même plan que les rangées de tipis (d’après Luquet 1930, fig. 73).

Fig. 5.1a. «Avions de chasse et chars d’assaut» (dessin de Martin, 5 ans, 2002). Exemple de représentation en volume ignorant les conventions de la perspective. On notera que les avions et les chars sont dessinés en vue de dessus, afin que les ailes des avions et les chenilles latérales des chars soient clairement visibles.

Fig. 5.1b. «Sacs de petits pois» (dessin de Martin, 4 ans, 2001). Exemple de représentation par transparence. Les sacs sont représentés comme une enveloppe fermée, délimitée par un trait continu, qui tient enfermé son contenu. Les petits pois sont représentés groupés à l’intérieur sous forme de points circulaires, rendus visibles à travers la paroi de l’enveloppe des sacs.

où celle-ci exige de faire apparaître plusieurs figures qui, elles-mêmes, ne sont pas toutes complètement visibles simultanément dans le même plan, ou plus exactement sous le même angle de vue. Ici encore, la question posée est celle de déterminer comment rendre apparentes des parties cachées, qui sont directement masquées par l’étendue des figures que l’on a choisi de représenter selon un angle visuel déterminé. Il n’existe pas de solution complètement satisfaisante à ce problème: comme nous le savons, la vision naturelle (celle de nos yeux) comme la vision optique (celle des appareils photos ou des caméras) sont en elles-mêmes fragmentaires et incomplètes. On ne voit jamais qu’une partie de la réalité visible et ce qui est visible n’est jamais qu’une partie de ce qui existe. Schématiquement, il est possible d’adopter, face à ce problème, deux options différentes, qui sont situées aux extrémités d’une gamme de solutions visuelles diverses. On peut choisir, d’une part, de représenter le réel tel qu’il apparaît visuellement, ou plus exactement optiquement: c’est la voie prise par la tradition naturaliste occidentale. Ou bien, on peut choisir, à l’inverse, de le représenter tel qu’il est, ou plus exactement tel qu’on se le représente: c’est l’option retenue par la plupart des arts dits «primitifs», ainsi que par le dessin des enfants, lorsqu’il n’est pas encore soumis aux conventions de la représentation en perspective (Fig. 5.1a). A la suite des travaux de Georges-Henri Luquet sur le traitement visuel des formes dans l’art primitif (Luquet 1930), il est possible de distinguer ainsi des types de figurations qui procèdent soit d’un «réalisme visuel» (propre aux représentations naturalistes), soit au contraire d’un «réalisme intellectuel» caractéristique des productions dites «primitives». Ainsi que l’indique Luquet:

5.  Les codes de représentation visuelle dans l’art celtique ancien

41

Fig. 5.2. Gravure de char attelé de l’âge du Bronze scandinave (d’après Luquet 1930, fig. 78).

Le réalisme intellectuel de l’art primitif s’oppose donc au réalisme visuel de deux façons contradictoires: d’une part, le dessin contient des éléments du modèle qui ne se voient pas, mais que l’artiste juge indispensable; inversement il néglige des éléments du modèle qui sautent aux yeux, mais qui sont pour l’artiste dénués d’intérêt (Luquet 1930, 68–69).

Contrairement aux représentations qui procèdent du réalisme visuel, celles qui font appel au réalisme intellectuel mettent en œuvre des effets de distorsion optique, sans lesquels il ne serait pas possible de montrer les figures telles qu’elles sont en elles-mêmes, au delà de leur apparence visuelle immédiate. Ainsi, l’un de ces effets les plus fréquemment utilisés est celui de la transparence, qui permet de montrer des éléments essentiels de la figure qui se voient pas directement de l’extérieur (Fig. 5.1b). Un autre procédé très utilisé est celui du rabattement, auquel il est fait appel lorsque la figure nécessite plusieurs angles de vue simultanés pour être représentée complètement dans tous ses éléments constitutifs essentiels (Fig. 5.1c). Ce mode de représentation graphique est exploité lorsqu’il s’agit en particulier de représenter individuellement des objets ou des figures se faisant face, c’est-à-dire se masquant les unes les autres dans une représentation frontale. Comme le souligne en effet Luquet: Tout se passe comme si, l’intersection de chacun des deux plans verticaux contenant (ces objets ou ces figures) avec le plan horizontal du sol étant constituée par une charnière, on avait fait pivoter les plans verticaux autour de ces charnières pour les rabattre dans le plan horizontal, ce qui justifie le nom de rabattement que nous avons donné à ce mode de représentation graphique (Luquet 1930, 136).

Les conventions de représentation visuelle protohistoriques Le rabattement visuel est donc un procédé de représentation

Fig. 5.3a. Sublaines «Les Danges» (Indre-et-Loire). Char à quatre roues attelé de deux chevaux. Motif peint sur céramique (VIIIe s. av. J.-C.; d’après Cordier 1975).

Fig. 5.3b. Eberdingen-Hochdorf (Allemagne ; Bade-Wurtemberg). Char à quatre roues attelé de deux chevaux, portant un personnage. Motif latéral dessiné au repoussé du dossier de la banquette de bronze déposée dans la tombe à char (VIe s. av. J.-C.). La caisse suivie du timon et les roues du char, qui sont situées dans des plans différents, sont représentées par rabattement. Noter la position du personnage (féminin?), figuré de face au dessus de la caisse du char (d’après Biel 1985, fig. 54).

classique des modes de figuration «non rétiniens» – pour reprendre une expression de Marcel Duchamp – et cette technique a été largement utilisée durant la Protohistoire, en Europe continentale. Les représentations de chars

42

Laurent Olivier

attelés attribués à l’âge du Bronze des gravures rupestres de Scandinavie en fournissent une série d’exemples caractéristiques. Ici, les figures sont soumises à un procédé de rabattement qui s’articule autour de l’axe de symétrie longitudinal du couple char-attelage: ainsi, les graveurs protohistoriques ont-ils systématiquement représenté les deux roues latérales du char rabattues à plat de part et d’autre des extrémités d’essieu, tandis que les deux chevaux de l’attelage sont couchés à plat sur le côté de part et d’autre de l’axe du timon (Fig. 5.2). De cette manière, l’ensemble des éléments constituant la totalité de la figure du char attelé – qu’il serait impossible de représenter complètement dans une vision frontale, latérale ou surplombante – sont tous réunis sur le dessin à leur place morphologique. En l’occurrence, le dessinateur a clairement recherché à représenter la figure du char attelé telle qu’elle est constituée de chacun de ses éléments constitutifs, et non telle qu’elle se perçoit visuellement. La représentation graphique par rabattement est largement mobilisée également dans les figurations de la fin de l’âge du Bronze et du premier âge du Fer nord-alpin. A Sublaines «Les Danges» (Indre-et-Loire; Codier 1975), par exemple, on retrouve le procédé de figuration d’un char attelé par rabattement des deux plans symétriques de la figure, lesquels sont «dépliés» de part et d’autre de l’axe de symétrie longitudinal de l’ensemble (Fig. 5.3a). Sur l’urne en céramique peinte du «petit tumulus» (MAN 82974), la caisse du char, bien que schématisée, est manifestement représentée en vue de dessus, prolongée par son timon. Les quatre roues latérales du véhicule sont figurées à plat, comme si elles avaient été dépliées à l’horizontale à partir de leur position verticale initiale. Les deux quadrupèdes (chevaux?) attelés sont représentés chacun de profil, de part et d’autre de l’axe du timon, comme si chacun d’eux avait été couché à plat à partir de sa position verticale. Le vase sur lequel a été apposée cette figuration appartient à un contexte stylistique datable d’une période vraisemblablement située entre la fin du IXe siècle et le début du VIIIe siècle av. J.-C. Pour la période de la phase ancienne du premier âge du Fer, les figurations gravées sur céramique fournissent encore une série d’exemples de représentations par rabattement dont la construction – par «dépliage» de part et d’autre des axes de symétrie propres à la morphologie des figures – est analogue à celle du motif de Sublaines. On peut citer, par exemple, le motif de char de l’urne du Tumulus 87 de Fischbach-Schirndorf, en Allemagne (Bavière; Pare 1992, 207, fig. 143), ou encore ceux des vases en céramique des Tumulus 127 et 140 de la nécropole de Sopron, en Hongrie (Pare 1992, 208, fig. 144). La scène du char du dossier de la banquette en bronze de la tombe d’EberdingenHochdorf (Bade-Wurtemberg), qui appartient à un contexte archéologique attribuable à la fin du VIe siècle av. J.-C., montre que les représentations par rabattement visuel restent un mode d’expression graphique privilégié durant toute la

période du premier âge du Fer nord-alpin. Comme sur les exemples précédents, la caisse du char prolongée de son timon est clairement figurée en vue de dessus, tandis que les quatre roues sont rabattues à l’horizontale de chaque côté du véhicule. Les deux chevaux qui tirent l’ensemble sont parfaitement représentés de profil, à l’avant du véhicule, de part et d’autre du timon, comme s’ils étaient couchés sur le côté. Le personnage semble-t-il féminin1 qui se tient sur le char est nettement dessiné de face, se tenant à sa place; c’est-à-dire «au dessus» de la caisse visuellement rabattue à plat (Fig. 5.3b). De manière intéressante, la scène de char de Hochdorf montre l’existence de conventions de représentation visuelle différentes selon la morphologie des figures. Ainsi, les objets plats (comme ici la caisse ou les roues du char) sont-ils préférentiellement représentés relevés verticalement; c’està-dire à plat. En revanche, les quadrupèdes (comme ici les chevaux) sont figurés en vue de profil, ou de côté. Enfin, les êtres humains sont plutôt représentés en vue frontale, ou de face. On retrouve ces conventions de représentation notamment sur les figurations gravées de Sopron, qui associent au chars des animaux (chevaux) et des êtres humains (hommes et femmes?). On notera que, dans tous les cas, ces conventions visent à donner de la figure à représenter la vision la plus complète, avec tous ses éléments constitutifs visibles à leur place morphologique: il importe de représenter les chevaux avec leurs quatre pattes, leur tête et leur queue; c’est pourquoi ces animaux sont figurés de côté. Il importe également de les représenter avec leurs deux oreilles (ou leurs deux cornes, lorsqu’il s’agit, comme à Sopron, de bovidés ou de cervidés): c’est pourquoi, sur la scène du char de Hochdorf, les oreilles des chevaux – qui ne seraient normalement pas visibles ensemble sur une vue latérale de la tête de l’animal – sont figurées côte à côte, au sommet du crâne. De la même manière, sur la scène centrale de la «danse des épées», les guerriers sont représentés le corps et la tête de face; tandis que les bras et les jambes, qui sont en mouvement, sont figurés de côté. Là encore, on constate, à Sopron comme à Hochdorf, que les membres inférieurs sont généralement dessinés de profil, de manière à faire clairement apparaître les pieds, lesquels sont mal identifiables dans des vues frontales.

Procédés de rabattement visuel à La Tène ancienne Malgré l’appel à un style graphique très particulier, les représentations figurées de la période de La Tène ancienne ne mobilisent pas des codes de représentation visuelle fondamentalement différents de ceux de la fin de l’âge du Bronze et du Premier âge du Fer au Nord des Alpes. Comme leurs antécédents hallstattiens, le dessin de ces figures exploite en effet essentiellement des procédés de rabattement visuel. L’une des nouveautés introduites par les figurations du style de la Tène ancienne réside ici dans l’exploration de

5.  Les codes de représentation visuelle dans l’art celtique ancien

43

Fig. 5.4a. Vase de Bucy-le-Long «La Héronnière» (Aisne). D’après Coll. (2000) Archéologie en Picardie. Sépultures en carrières. Vallées de l’Aisne et de L’Oise. Amiens, SRA Picardie.

Fig. 5.4b. Vase caréné de La Cheppe (Marne). MAN 84643.

nouvelles combinaisons formelles que rendent possibles ces procédés de rabattement traditionnels: ceux-ci permettent en effet de décomposer les figures, de part et d’autre de leurs axes de symétrie propres, en sous-ensembles d’éléments, qui peuvent à leur tour être recomposés en de nouvelles figures, construites autour de symétries propres cette fois à la construction du dessin. Plusieurs exemples de motifs simples de frises d’animaux gravées sur céramique montrent comment ces plans de rabattement visuels conventionnels sont exploités à La Tène A pour en tirer de nouvelles potentialités graphiques. Sur le vase situliforme de la nécropole de la «Héronnière» à Bucy-le-Long (Aisne), par exemple, deux frises superposées de chevaux allant à droite, la tête baissée, sont gravées à la surface externe du récipient (Fig. 5.4a). Comme au premier âge du Fer, les animaux sont dessinés en vue de côté. Cependant, seule la moitié latérale des bêtes apparaît représentée, avec la queue, les pattes latérales droites et le profil droit de la tête, comme si la figure des chevaux avait été «découpée» les long de l’axe de symétrie constitué par la colonne vertébrale de la bête. Le motif gravé du grand vase caréné de la Cheppe (Marne; MAN 84643) révèle un mode de construction graphique tout à fait similaire à celui de la Héronnière. Ici, une frise gravée de quadrupèdes (chevaux?) allant à droite est figurée encadrée par une double bande de chevrons gravés. Les animaux, qui sont représentés la tête tournée en arrière et la queue légèrement relevée, apparaissent encore une fois réduits à leur moitié latérale droite: seule une des

deux pattes avant et arrière est figurée, le dessin indiquant nettement, pour la patte arrière, qu’il s’agit du membre latéral droit, faisant directement face à l’observateur (Fig. 5.4b). En multipliant à l’identique ces moitiés d’animaux les unes à la suite des autres, le potier (ou la potière) a obtenu un effet de rythme sous la forme d’une bande de S couchés que dessine la silhouette générale des bêtes se retournant toutes derrière elles.

La plaque de Cuperly et la question des monstres animaux bicéphales Revenons maintenant à la plaque au dragon de Cuperly, qui semble figurer à première vue – c’est-à-dire selon nos propres codes visuels – un monstre animal doté de deux têtes. L’examen de la construction du dessin montre que la figure représentée mobilise en fait des codes figuratifs propres à la représentation des animaux dans l’art de la Tène ancienne. En effet, la tête est représentée de profil, comme celle des animaux (quadrupèdes, oiseaux…). Le dessin de l’œil, figuré au moyen d’un cercle à point central, est également caractéristique des figurations animales champenoises. On notera également qu’une seule oreille est dessinée pour chaque profil représenté; ce qui correspond à un procédé de découpe «par moitiés latérales» typique, comme on l’a vu, des représentations animales de La Tène ancienne champenoise. Le fait que chacun des deux yeux est représenté d’une manière différente accrédite par ailleurs l’existence de deux faces différentes: la face gauche présente

44

Laurent Olivier

Fig. 5.6. Elément indéterminé de Saint-Etienne-au-Temple (Marne). MAN 12808 (d’après Jacobsthal 1969, pl. 63, 102).

Fig. 5.5. Plaque au dragon de la tombe à char de Cuperly «La Grammonerie» (Marne) MAN 27719.

ainsi un œil mis clos, alors que la face droite est dotée d’un œil normalement ouvert. Aussi, l’applique zoomorphe de Cuperly ne représente pas un animal à deux têtes, mais bien les faces latérales droite et gauche d’une même bête, possédant un corps serpentiforme unique. Sur la plaque de Cuperly, ce dernier apparaît ainsi «déplié» sous l’effet d’un procédé de rabatte­ ment visuel, qui s’organise de part et d’autre d’un axe de symétrie coïncidant lui-même avec l’axe de symétrie longitudinal de la palmette centrale; l’ensemble constituant l’axe de symétrie de la forme générale de la pièce. On observe précisément ce type de procédé de construction visuelle dans une série de pièces à figurations zoomorphes de la Tène ancienne de Champagne. En particulier, sur la paire d’éléments indéterminés à incrustations de corail des environs de Saint-Etienne-au-Temple (Marne; MAN 12808) est représenté un motif de double silhouette d’oiseau la tête tournée en arrière: celle-ci apparaît composée des deux moitiés latérales symétriques du corps qui sont «dépliées» de part et d’autre d’un axe de symétrie passant par la partie antérieure du corps, entre le poitrail et le cou (Fig. 5.6). Les torques champenois à motifs d’oiseaux montrent des types de représentation par rabattement visuel tout à fait similaires. Cependant, contrairement aux oiseaux des pièces de Saint-Etienne-au-Temple, l’animal de la plaque de Cuperly n’est pas une bête ordinaire, dans la mesure où, comme on l’a souligné, elle ne possède ni membres ni extrémités

(c’est le cas également des animaux fantastiques champenois auxquels elle est apparentée, comme les dragons du vase de La Cheppe, ou les «griffons» des disques de harnache­ ment de Semide). Plus précisément, l’axe de symétrie du rabattement visuel mis en œuvre pour représenter la figure est virtuel, dans le sens où celui-ci ne coïncide avec aucun axe de symétrie morphologique présent dans la figure ellemême, comme c’est le cas avec les autre animaux dont le corps est représenté complet. Le corps de la bête de la plaque de Cuperly n’a pas de fin; d’ailleurs, sa silhouette «dépliée» n’apparaît qu’indirectement, en filigrane, par l’effet de contraste créé par l’alternance de plages lisses et ajourées. C’est un monstre composite, qui tient à la fois de l’animal (il possède une tête, avec deux yeux et deux oreilles) et de l’immatériel (il ne possède pas réellement de corps physique). D’autres animaux, représentés par le même mode de rabattement visuel évoquant l’action de déplier les formes, apparaissent de la même nature que le monstre de Cuperly: c’est le cas, par exemple, des bêtes à tête de canard et à corps serpentiforme «déplié» de la bouterolle de fourreau des environs de Suippes (Marne; MAN 20249) (Fig. 5.7), ou des appliques de harnachement à bouton à bélière de la tombe à char de Semide (Lambot, Verger, Meniel 1995, fig. 66) (Fig. 5.8). Jacobsthal a bien montré que l’apparition des animaux doubles ou à deux têtes, particulièrement récurrente dans les figurations de La Tène ancienne, ne se limite pas à une simple question de représentation visuelle: en effet, la représentation simultanée des deux faces identiques d’un même animal n’apporte, fondamentalement, aucune information visuelle supplémentaire par rapport à une simple figuration latérale. C’est donc autre chose que ce mode

5.  Les codes de représentation visuelle dans l’art celtique ancien

45

Fig. 5.7. Bouterolle zoomorphe de Suippes (Marne). MAN 20249.

de figuration cherche à représenter. Selon Jacobsthal, ce dédoublement des figures vise plutôt à expliciter une qualité particulière, propre aux monstres composites: En fait, ces animaux doubles sont la contraction de deux bêtes uniques opposées l’une à l’autre. (…) L’interprétation de ces animaux doubles comme un effet de contraction de leurs éléments constitutifs individuels ne concerne que l’aspect formel du problème: en essence, ces animaux doubles sont frères et sœurs de ces animaux fantastiques qui possèdent un corps et deux têtes; c’est-à-dire des monstres qui combinent des membres d’êtres de différentes espèces. Еιλος πολΰγνιον signifie intensification du pouvoir, accumulation de facultés: une bêtes à deux corps est plus puissante et plus formidable qu’une autre qui n’a qu’un seul corps (Jacobsthal 1969, 50–51).

Conclusion et perspectives Animaux composites et figures hybrides Alors que les représentations figurées hallstattiennes ne montrent essentiellement que des êtres humains ou des animaux, celles de la période de La Tène ancienne introduisent de nouvelles catégories d’êtres, qui sont composites ou hybrides. L’introduction de ces nouvelles catégories dans le répertoire figuratif traditionnel de l’âge

Fig. 5.8. Applique à bouton à bélière de Semide (Ardennes; d’après Lambot Verger et Meniel 1995, fig. 66).

du Fer nécessite de trouver des solutions de représentation visuelle particulières, comme le montre l’exemple des sphinx. Les représentations classiques de sphinx qui parviennent en Europe continentale, comme par exemple celle de l’applique du tumulus du «Grafenbühl» à Asperg (Allemagne, Bade-Wurtemberg), montrent de profil ces êtres fantastiques à corps animal assis sur leurs pattes de derrière, leur visage anthropomorphe tourné vers le spectateur. Les pattes avant sont clairement individualisées dans des poses différentes, l’une passant devant l’autre, de même que la longue queue

Laurent Olivier

46

Fig. 5.9a. Sphinx du «Grafenbühl» (Allemagne).

Fig. 5.10. Agrafe de Somme-Bionne «L’Homme Mort» (Marne; d’après Jacobsthal 1969, pl. 170, 359).

Fig. 5.9b. Plaque de ceinture de la sépulture 1 du Tumulus 01 de Weisskirchen «Schanzenknöppchen» (Allemagne; d’après Reinhard 2003, pl. 100, 2).

recourbée sous le ventre, qui vient de la face arrière de la figure, non visible pour nous (Fig. 5.9a). De manière révélatrice, ces figures de sphinx étrangères sont adaptées sur la plaque de Weisskirschen «Schanzenknöppchen» (Allemagne, Sarre) sous la forme de deux «doubles» sphinx déployés de part et d’autre d’un masque humain représenté de face. On reconnaît immédiatement le procédé de construction par «dépliage» des deux moitiés latérales du corps animal de ces créatures chimériques: seules l’aile et les pattes latérales de chaque face sont représentées; les sphinx étant assis sur leurs pattes de derrière repliées. Les têtes hybrides de ces monstres – mi humaines, mi animales – sont figurées également sous

la forme de deux profil latéraux «dépliés», comme la tête composite du dragon de Cuperly. De manière caractéristique, les deux faces latérales des sphinx de Weisskirchen sont jointives au contact du cou et du poitrail ; ce qui montre que la forme générale des monstres est «découpée» le long de leur axe de symétrie longitudinal et qu’elle est «dépliée» à partir de l’arrière, la «pliure» jouant à l’avant de la bête (Fig. 5.9b). On retrouve ce procédé de «découpe» par moitiés latérales associé à un «dépliage» de part et d’autre de l’axe de la partie antérieure du corps sur les figurations dédoublées d’animaux, que l’on observe en particulier sur les agrafes de ceinture ajourées: dans ce cas, ce sont visiblement les deux faces symétriques d’un unique animal qui sont figurées, chacune présentant la même disposition des membres. Ici encore, les deux faces sont reliées au niveau du poitrail et du cou, notamment comme sur l’agrafe de «L’Homme Mort» à Somme-Bionne (Marne) (Fig. 5.10). Il est possible que les bronziers de La Tène ancienne de Champagne aient eu recours à ce procédé lorsqu’ils étaient confrontés à un support uniface, ou réduit à deux dimensions, qui ne permettait pas de représenter une figure dans son développement en volume, laquelle devait donc être «dépliée» ou «rabattue». En l’occurrence, nous trouvons ici plus qu’un simple effet de copie ou de transposition de

5.  Les codes de représentation visuelle dans l’art celtique ancien modèles méditerranéens, tels que les sphinx ou les griffons. Il s’agit plutôt d’une appropriation de figures étrangères dans des codes de représentation figurative particuliers à l’âge du Fer nord-alpin. Pour les artisans de l’âge du Fer, il importe en effet avant tout de ne pas altérer les formes propres aux figures à représenter; c’est pourquoi il n’est pas fait appel à la perspective, qui occulte et déforme les éléments constitutifs essentiels des figures.

47

Fig. 5.11a. Cheval de l’oenochoe de Waldalgesheim (Allemagne; d’après Joachim 1995, pl. 23, 5–6).

Les figurations en ronde bosse: une projection en volume des codes de rabattement visuel propres aux arts graphiques Plus nettement que les représentations projetées sur une surface à deux dimensions, les représentations en ronde bosse rendent apparente l’articulation dans l’espace de ces différents plans de rabattement visuel dans lesquels sont réparties les formes élémentaires des figures. La réalisation des figures en volume apparaît en effet appréhendée fondamentalement comme un processus de restitution à leur position initiale dans l’espace des plans de rabattement visuel, qui doivent nécessairement être «dépliés» sur les supports plats. La figure en relief du petit cheval de l’oenochoe de Waldalgesheim (Joachim 1995, fig. 26) montre bien, par exemple, comment est construite l’image des animaux. La pièce est constituée des deux moitiés latérales parfaitement symétriques (une gauche et une droite) d’un animal figuré de profil, qui sont directement appliquées l’une sur l’autre: de face la figure ne laisse apparaître qu’une seule jambe, à l’avant et à l’arrière. Vue de face, la jambe unique avant possède néanmoins deux sabots juxtaposés, figurés sous une forme triangulaire: ces derniers montrent bien que cette jambe unique procède en fait du «collage» de deux membres différents vus de profil (Joachim 1995, fig. 23, 7) (Fig. 5.11a). On retrouve donc bien ici, sur une pièce en haut relief, le mode de figuration «par moitiés latérales» propre aux animaux que l’on observe sur les représentations gravées. Le quadrupède à masque humain de l’oenochoe de Reinheim (Sarre; Echt 1999) a posé un problème de représentation particulier dans la mesure où il s’agit d’un monstre hybride. Comme avec le petit cheval de Waldalgesheim, on observe un procédé de construction de la partie animale de la figure par superposition des deux moitiés latérales du corps du cheval, lesquelles apparaissent parfaitement symétriques. Ici, contrairement au relief de Waldalgesheim, les paires de pattes sont individualisées, mais elles sont figurées exactement dans la même position anatomique sur chacune des moitiés latérales. En revanche, la face humaine du monstre, qui ne pouvait pas être représentée par le collage de deux moitiés latérales, comme les têtes d’animaux, a fait l’objet d’un traitement particulier:

Fig. 5.11b. Cheval de l’oenochoe de la tombe de Reinheim (Allemagne; d’après Frey 2007, fig. 10).

figurée nécessairement de face, elle a donc été apposée perpendiculairement à l’axe général du corps de l’animal; c’est-à-dire à 90° par rapport à la surface horizontale du support (Fig. 5.11b). On retrouve ce type de construction particulier aux êtres hybrides, mi-humains mi-animaux, sur les paires de monstres disposés de part et d’autre de l’ouverture de l’oenochoe du Glauberg (Hesse; Herrmann and Frey 1996, fig. 93–95). Ces créatures monstrueuses sont représentées tournant la tête en arrière vers le personnage cuirassé assis en tailleur à la naissance de l’anse: pour ce faire, on a placé la face humaine des monstres, représentée de face, perpendiculairement à l’axe du corps des animaux assis sur leurs pattes arrières, et à 90° par rapport à la surface horizontale du support. On remarque que le corps des animaux proprement dits est figuré de profil, par le procédé standard de collage des deux faces latérales symétriques (Fig. 5.12). Devant les pattes avant des monstres, est placée une tête humaine, également

48

Laurent Olivier

Fig. 5.13. Le monstre dévoreur de l’anse de l’oenochoe du Dürrnberg (Autriche ; d’après Herrmann et Frey 1996, fig. 108).

directement projetée dans le prolongement de l’échine de la bête, tandis que la tête humaine semble fixée directement sous son menton (Fig. 5.13). Fig.5.12. Un des sphinx de l’embouchure de l’oenochoe de la tombe du Glauberg (Allemagne; d’après Herrmann et Frey 1996, fig. 96).

représentée de face; c’est-à-dire perpendiculairement à l’axe de la bête. Pour lui conserver toute sa lisibilité, le bronzier l’a représentée non pas dressée verticalement, comme la tête du monstre, mais couchée horizontalement sur le support. C’est un mode de construction tout à fait similaire, propre aux êtres hybrides, que l’on reconnaît dans la figuration des monstres de l’oenochoe du Dürrnberg (Autriche; Jacobsthal 1969, I, 201, cat. n 382; II, pl. 184–187). Ici encore, le corps animal de la bête dévoreuse de l’anse est formé, de manière caractéristique, par le collage des deux moitiés latérales symétriques de la bête, dont les pattes sont couchées sur le support. En revanche, la gueule anthropomorphe du monstre est normalement figurée de face. Ici, la nécessité de représenter le monstre tenant une tête humaine entre ses pattes avant a posé un problème de figuration particulier, dans la mesure où cette tête coupée, comme la tête anthropomorphe du monstre lui-même, exigeaient également d’être montrées de face. Les deux têtes ont donc été figurées perpendiculairement à l’axe général du corps de la bête, représentée, elle, de profil. Cependant, la tête humaine coupée a été dressée verticalement, à 90° par rapport à la surface horizontale du support, tandis que la tête du monstre a dû être couchée à l’horizontale au dessus d’elle, formant ainsi un angle de 180° par rapport au support, de manière à ce que les deux têtes soient également visibles de face. Ces contraintes de représentation confèrent un aspect étrange à l’ensemble, dans lequel la face du monstre apparaît

Les codes de représentation visuelle de l’art celtique: formes et perspective L’univers figuratif de l’âge du Fer d’Europe continentale connaît donc des transformations importantes à partir du courant du Ve siècle avant notre ère. Schématiquement, on pourrait dire que les figurations hallstattiennes représentaient essentiellement des choses (comme des chars, des récipients, des outils ou des armes), des animaux (chevaux, bovidés, cervidés…) et des êtres humains (masculins ou féminins). Les représentations figurées du début du Second âge du Fer introduisent en l’occurrence une nouvelle catégorie, intermédiaire entre les hommes et les animaux: celle-ci est constituée par les êtres hybrides, mi-hommes mi-animaux, au premier rang desquels figurent les sphinx. On constate également – est-ce là une conséquence indirecte de ce qui précède? – que la figure humaine, représentée en son entier durant le Premier âge du Fer, n’est désormais plus figurée que partiellement. Les membres, qui constituaient généralement un élément graphique mineur des figurations hallstattiennes, disparaissent cette fois complètement, faisant perdre sa fonction au tronc, qui s’efface lui aussi. La figure humaine se trouve ainsi réduite essentiellement à la tête, laquelle n’était auparavant jamais vraiment développée, du point de vue graphique, au Premier âge du Fer. La grille de conventions de représentation visuelle qui s’instaure au début du Second âge du Fer peut donc être résumée de la manière suivante: Les animaux (comme en particulier les quadrupèdes) sont représentés en entier, le corps et la tête figurés de profil. Les êtres humains sont représentés de manière partielle

5.  Les codes de représentation visuelle dans l’art celtique ancien avec une réduction du corps à la tête, laquelle est représentée de face. La figure humaine prend alors la forme de masques. Les êtres hybrides (comme en particulier les sphinx) sont représentés en entier, leur corps animal de profil et leur visage humain de face. Ainsi, dans les représentations figurées de l’art de La Tène, les figures en mouvement sont représentées non sous la forme d’occultations ou de rotations anatomiques – comme dans la figuration naturaliste – mais sous l’aspect d’assemblages de plans, essentiellement perpendiculaires (comme des faces et des profils). C’est là une forme de pensée visuelle ignorant les effets déformants de la perspective, laquelle permet de représenter, à la condition d’une altération des formes, le déploiement dans l’espace des différentes faces d’un même objet ou d’une même figure. Comme le souligne Luquet: La perspective n’a pas seulement comme effet de supprimer pour la vue une partie plus ou moins grande d’un objet lorsqu’elle est masquée par un autre objet situé entre lui et le spectateur. Même pour un objet placé au premier plan, elle donne à certains de ses éléments une forme différente de leur aspect caractéristique. Le réalisme intellectuel tient au contraire à ce que tous ces éléments conservent dans l’objet dont ils sont des parties la forme exigée par l’exemplarité: chacun d’eux est figuré comme s’il était aperçu du point de vue d’où il présente cet aspect. (…) L’artiste primitif ne songe pas le moins du monde à signifier par la forme qu’il donne à chaque détail dans son dessin que si l’on regardait l’objet de tel point de vue, ce détail y apparaîtrait avec cet aspect; il veut simplement figurer ce détail tel qu’il le pense en soi. Si la forme que présente dans le dessin chaque élément d’un objet est forcément de nature visuelle, cette forme n’est que la traduction de son essence, c’est-à-dire d’une idée de nature intellectuelle; on pourrait dire que dans l’ensemble de l’objet, chacune de ses parties est rendue par un idéogramme (Luquet 1930, 152).

L’art du début du Second âge du Fer s’inscrit dans une étape ultérieure de ce processus de décomposition visuelle des figures, qui aboutit, comme le souligne Luquet, à l’obtention de formes élémentaires s’affichant comme des idéogrammes. Les créations complexes du style continu ancien, avec leurs procédés de découpe des motifs (comme celui, classique, de la palmette ou de la fleur de lotus) et de recomposition des formes élémentaires obtenues par des effets de symétrie en miroir, correspondent fondamentalement à une tentative de constitution «d’hyper-figures», se développant simultanément dans une multitude de plans. Ces formes idéelles peuvent être alors projetées sur des supports en relief, qui les étirent ou les déforment: on retrouve ici un procédé peu éloigné, au fond, de l’anamorphose. Sur le plan formel, on notera que ces procédés de pliage et d’assemblage des formes permettent de créer une multiplicité de combinaisons nouvelles de motifs et de figures: tout au long des VeIVe siècles avant notre ère, cette dynamique alimentera directement la production iconologique de ce que Jacobsthal

49

identifiait sous le terme «d’Art celtique ancien». Lorsqu’elle aura épuisé les potentialités du répertoire stylistique qu’elle exploite, cette dynamique de création stylistique conduira finalement à la phase baroque du style plastique, qui en est le prolongement.

Visual

representation codes in

Early Celtic

art

Introduction: Early La Tène visual representation codes Intellectual realism and codes for visual representation In archaeology, just as the study of manufactured items requires the determination of the technical function of an object and how it was made (Guillaumet 2003), so it is just as necessary, when studying figurative images, to determine what they represent and how they are visually constructed. However, if zoomorphic or anthropological figures from Late Iron Age art have recieved copious commentaries from the stylistic and typological point of view, most authors are hardly interested in the problem of visually reading them, as if this is not a proper subject for discussion. Following in the footsteps of Jacobsthal and Paul-Marie Duval above all, specialists have discussed at length the construction of stylistic motifs and their typochronological evolution. On the other hand, very few scholars – with the notable exception of Paul Jacobsthal himself – have turned their attention to the processes of visual representation used to produce the figures of La Tène art. In effect, the act of representing a figure necessitates in some way that one projects it into a space defined by specific spatial dimensions, and that it is shown according to a particular point of view located somewhere in space. The negotiation of these constraints on representation translates into the production of codes of visual representation, which condition every figural creation. These protohistoric codes being fundamentally different from those transmitted by the naturalistic tradition of classical art, 19th century researchers have unconsciously inferred that these were basic, if not inconsistent. The consequence of this approach has been to spontaneously push La Tène figurative art to the confines of the ornamental; in other words to turn it into a purely decorative production. The two propositions which have made La Tène art into a manifestation of ornament, as in particular according to Déchelette or Henri Hubert (Déchelette 1914, 1508–1509; Hubert 1932, 141–142), or of a manifestation ‘refractory to images’ (or ‘aniconic’) (Lantier 1946, 74) are equally false. Above all, these interpretations are evidence of

50

Laurent Olivier

an unformulated ideological position, which places – in an imaginary hierarchy of artistic representations – the stylized or geometric images of ‘primitive’ or ‘barbaric’ societies on a lower level, from the point of view of the intellectual development of civilizations, than the naturalistic representations of classical societies (Semper 1861; Haddon 1895; Riegl 1923). Fundamentally, the apparently simple act of creating a figure on a medium is to attempt to propose an acceptable solution to a particularly difficult problem of visual representation: it is to determine how to represent an object composed of several surfaces or sides simultaneously invisible in space – the space projected onto the drawing medium – which itself has only one face. In a more mathematical language, it could also be said that it is to project a three-dimensional object onto a two-dimensional surface: in these conditions, how to make visible the hidden part of the object (because located behind or on the sides), which is hidden by the body of the object itself, according to the visual angle of observation chosen for the view? The representation of scenes adds an additional level of complexity to this basic problem, since it requires that several figures are made visible which, themselves, are not all completely visible simultaneously in the same angle, or more precisely from the same point of view. Here again, the question is that of determining how to make visible the hidden parts of the scene, which are directly covered by the figures shown according to the particular visual angle chosen for constructing the picture. There is no completely satisfactory solution to this problem: as we know natural vision (our eyes), like optical vision (photographic equipment or cameras), are themselves fragmentary and incomplete. Only a part of the visible reality is ever seen; in other words, what is visible is only ever a fragment of that which stands in front of us. Schematically, faced with this problem, it is possible to adopt two different options, which are located at the extremities of a diverse range of visual solutions. On the one hand, there is the option of choosing to picture reality as it appears visually, or more exactly optically: this is the route taken by the Classical naturalistic tradition. Or alternatively, there is the option of showing it as it is, or more precisely as it is constituted: this is the option taken by the majority of arts called ‘primitive’, as well as by children’s drawings, when they have not yet submitted to the conventions of representation in perspective (Fig. 5.1a). Following the work of Georges-Henri Luquet on the visual treatment of forms in primitive art (Luquet 1930), it is thus possible to distinguish the types of representation which carry either a ‘visual realism’ (characteristic of naturalistic depictions), or on the contrary an ‘intellectual realism’ characteristic of so-called ‘primitive’ productions. As Luquet indicates: Thus the intellectual realism of primitive art is opposed to visual realism in two contradictory ways: on the one hand, the

drawing contains elements of the object which are not visible, but which the artist considers to be indispensible; inversely he ignores elements of the image which are clearly visible, but which are lacking in interest to the artist (Luquet 1930, 68–69).

Unlike representations which precede visual realism, those which involve intellectual realism implement the effects of optical distortion, without which it would not be possible to show the figures as they are in themselves, beyond their immediate visual appearance. Thus, one of the effects most commonly used is that of transparency, which allows us to show the essential elements of the figure which are not directly visible on the outside (Fig. 5.1b). Another frequently used method is that of rabatment, which is used when the figure requires several views simultaneously to be fully represented in all its constituent elements (Fig. 5.1c). This type of graphic representation is exploited especially when it comes to represent individual facing objects or figures, which is to say covering each other in a frontal representation. As Luquet essentially highlighted: Everything happens as if the intersection of each of the two vertical planes containing (these objects or these figures) with the horizontal surface plane being made of a hinge, the vertical planes are pivoted around these hinges, rabatting them into the horizontal plane, which justifies the name ‘rabatment’ which we have given to this method of graphic representation (Luquet 1930, 136).

The protohistoric conventions of visual representation Visual rabatment is therefore a classic method of representing ‘non-retinal’ figures – to use an expression of Marcel Duchamp – and this technique was widely used throughout protohistory in continental Europe. Representations of chariots on rock carvings attributed to the Bronze Age in Scandinavia provide a typical set of examples. Here, the figures are subject to a process of rabatment that revolves around the longitudinal axis of symmetry of the chariot: in this way, the protohistoric carvers systematically represented the two lateral wheels of the chariot folded out flat on either side of the axle ends, while the two chariot horses lie flattened out on either side of the yoke axis (Fig. 5.2). In this way, all the elements which make up the totality of the image of the chariot team – which it would be impossible to represent completely from a frontal, side or overhead perspective – are all brought together within the drawing in their morphological position. In this case, the artist has clearly sought to represent the figure of the horses and chariot as they are constituted in each of their constituent elements, and not as they are perceived visually. Graphical representation by rabatment is also widely used in art from the end of the Bronze Age and from the NorthAlpine Early Iron Age. At Sublaines ‘Les Danges’ (Indre-

5.  Les codes de représentation visuelle dans l’art celtique ancien et-Loire; Cordier 1975), for example, we find the process of representing a chariot team by rabatting the two symmetrical plans of the image, which are ‘unfolded’ on either side of the longitudinal axis of symmetry of the whole (Fig. 5.3a). On the painted ceramic urn from the ‘little tumulus’ (MAN 82974), the body of the chariot, although schematic, is clearly shown in plan, extended by its yoke. The four lateral wheels of the vehicle are represented flat, as if they had been folded horizontally from their original vertical position. The two harnessed quadrupeds (horses?) are each represented in profile, on either side of the yoke axis, as if each had been lying flat from its vertical position. The vase on which this image was drawn belongs to a stylistic context dateable to a period probably between the end of the 9th century and the beginning of the 8th century BC. For the period of the earliest phase of the Early Iron Age, drawings engraved on ceramic provide a further series of examples of drawing by rabatment whose construction – by ‘unfolding’ on either side of the inherent morphological axes of symmetry of the objects – is analogous to the Sublaines motif. These include, for example, the motif of a chariot from the Tumulus 87 urn of Fischbach-Schirndorf, in Germany (Bavière; Pare 1992, 207 fig. 143), or additionally those of the ceramic vases from Tumulus 127 and 140 from the cemetery at Sopron, Hungary (Pare 1992, 208 fig. 144). The chariot scene on the back of the bronze seat from the tomb at Eberdingen-Hochdorf (Bade Wurtemberg), which belongs to an archaeological context attributable to the end of the 6th century BC, shows that visual representations by rabatment remained the preferred method of graphic expression throughout the whole of the North-Alpine Early Iron Age. As in the previous examples, the extended yoke and body of the chariot are clearly drawn in plan, while the four wheels are folded horizontally on each side of the vehicle. The two horses pulling the equipage are perfectly represented in profile in front of the vehicle, on either side of the yoke, as if they were lying on their sides. The human image – apparently female – who stands on the chariot is clearly drawn from the front, standing in her place, which is to say ‘above’ the chariot, visually flat by rabatment (Fig. 5.3b). Interestingly, the Hochdorf chariot scene shows the existence of conventions for different visual representations according to the morphology of the images. Thus, the preferred representation of flat objects (as in this case the seat box or the chariot wheels) is vertical, that is to say flat. However, quadrupeds (like the horses here) are drawn in profile, or from the side. Finally, humans are more often represented in frontal view, or facing. These representational conventions are especially found on the engraved images at Sopron, which link animals (horses) and humans (men and women?). Note that in all cases, these conventions aim to provide the figure with the most complete visual representation, with all its constituent elements visible in

51

their morphological position: it is important to represent the horses with their four legs, their head and their tail: this is why the animals are shown from the side. It is equally important to show them with their two ears (or their two horns when it comes to cattle or deer, as at Sopron): this is why, on the Hochdorf chariot scene, the horses’ ears – which would not normally both be visible with a side view of the animal’s head – are drawn side by side, on top of the head. In the same way, in the central scene of the ‘sword dance’, the warriors are shown with head and body face on, while the arms and legs, which are moving, are shown from the side. Again, at Sopron as at Hochdorf, we find that the lower limbs are usually drawn in profile, in such a way as to give a clear view of the feet, which are unidentifiable in a frontal view.

Early La Tène visual rabatment methods In spite of its links to a very specific graphic style, figurative representations from the Early La Tène do not use visual representation codes fundamentally different from those of the Late Bronze Age and the Early Iron Age in the North Alps. Like their Hallstatt predecessors, the design of these figures exploits, in essence, the visual rabatment method. One of the innovations introduced by Early La Tène drawing style rests here in the exploration of new formal combinations which allow these traditional methods of rabatment: they make it possible to break down the figures on either side of their own axes of symmetry in subsets of elements, which in turn can be recomposed in new figures, this time built around the inherent symmetries of the drawing. Several examples of simple patterns of animal friezes drawn on pottery show how these conventional plans of visual rabatment were exploited in La Tène A to draw out these new graphic possibilities. For example, on the situliform urn from the ‘Héronnière’ cemetery at Bucy-leLong (Aisne), two superimposed friezes of horses going to the right, heads lowered, are engraved on the external surface of the container (Fig. 5.4a). As in the Early Iron Age, the animals are drawn in profile. However, only the lateral half of the animals is drawn, with the tail, the side legs straight and the right profile of the head, as if the figure of the horses had been ‘cut’ along the symmetrical axis created by the animal’s spinal column. The motif engraved on the large biconical urn from La Cheppe (Marne; MAN 84643). Here, a freize engraved with quadrupeds (horses?) moving to the right is drawn surrounded by a double band of incised chevrons. The animals, which are represented with their heads turned back and their tails slightly raised, once again appear reduced on their right lateral halves: only one of the two front and hind legs are represented, the design clearly indicating, for the back leg, that it is the lateral right limb, directly facing

52

Laurent Olivier

the observer (Fig. 5.4b). In multiplying these identical parts of the animals one after another, the potter has obtained a rhythmic effect in the form of a band of horizontal ‘S’s which describe the general silhouette of the animals looking back at each other.

The Cuperly plaque and the question of twoheaded monstrous animals Let us now turn to the dragon on the Cuperly plaque, which at first sight seems to depict – that is to say according to our own visual codes – a monstrous animal with two heads (Fig. 5.5). An examination of the construction of the drawing shows that the image in fact uses the figurative codes proper to the representation of animals in Early La Tène art. In effect, the head is represented in profile, like that of animals (quadrupeds, birds…). The drawing of the eye, shown by a circle with a central dot, is equally characteristic of animal drawings from the Champagne. Also notable is that a single eye is drawn for each profile; which corresponds to a process of cutting out the lateral sections typical, as we have seen, for animal images of the Early La Tène in Champagne. The fact that each of the two eyes is represented in a different way also acknowledges the existence of two different faces: so the left side has a half-closed eye, while the right side is given an open eye. In this way, the Cuperly zoomorphic appliqué does not represent an animal with two heads, but actually the right and left profiles of the same beast, which possesses a single serpentine body. So on the Cuperly plaque, this latter appears ‘unfolded’ according to the effect of a visual rabatment method, which is arranged on either side of an axis of symmetry itself coinciding with the general axis of symmetry of the piece’s shape. Precisely this type of visual construction process is observable in a series of objects with zoomorphic decoration from Early La Tène in the Champagne. In particular, a motif of a double silhouette of a bird with its head turned backward is represented on the pair of indeterminate objects with incrustations of coral from the area of Saint-Etienne-au-Temple (Marne; MAN 12808): these appear to be composed of the two lateral symmetrical halves of the body which are ‘unfolded’ on either side of an axis of symmetry passing through the anterior part of the body, between the breast and neck (Fig. 5.6). The Champagne torcs with bird motifs show very similar types of representation by visual rabatment. However, unlike the birds from the Saint-Etienne-auTemple pieces, the animal on the Cuperly plaque is not an ordinary beast, to the extent that, as has been stressed, it possesses neither limbs nor extremities (this is also the case with the related fantastic animals from Champagne, such as the dragons from the La Cheppe urn, or the ‘griffons’ from the Semide harness discs). More precisely, the axis of symmetry used for visually rabatting the image is virtual, in

the sense that is does not coincide with any morphological axis of symmetry present in the image itself, as is the case with the other animals where the entire body is represented. The body of the beast on the Cuperly plaque has no end; in addition, its ‘unfolded’ silhouette is only indirectly visible, in filigree, by the contrasting effect created by alternating flat and relief surfaces. This is a composite monster, encapsulating both the animal (it possesses a head with two eyes and two ears) and the spiritual (it does not actually have a physical body) at the same time. Other animals, represented by the same method of visual rabatment, evoke a sense of unfolding shapes, like the Cuperly monster: this is the case, for example, with the beasts with duck heads and ‘unfolded’ serpentine bodies on the scabbard chape from the area of Suippes (Marne; MAN 20249) (Fig. 5.7), or the appliqués from the ram harness button from the cremation tomb at Semide (Lambot, Verger and Meniel 1995, fig. 66) (Fig. 5.8). Jacobsthal clearly showed how the appearance of double animals or animals with two heads, particularly recurrent in Early La Tène images, is not limited to a simple question of visual representation: in effect, the simultaneous representation of the two identical faces of a single animal does not convey, fundamentally, any supplemental visual information compared to a simple profile image. So there must therefore be something else that this method of illustration seeks to represent. According to Jacobsthal, this doubling of the images is above all trying to pinpoint a quality specific to composite monsters: Actually they are the contraction of two single antithetic animals…The interpretation of the double animals as contraction of their single counterparts touches only the formal aspect of the problem: in essence they are the brothers and sisters of those fantastic animals with one body and two heads – monsters combining members of different species. Еιλος πολΰγνιον means intensification of power, accumulation of faculties: a beast with two bodies is stronger and more formidable than a one-bodied (Jacobsthal 1969, 50–51).

Composite animals and hybrid figures Although figurative images of the Hallstatt period essentially only show animal or human figures, those from the Early La Tène introduce new categories of beings, which are composite or hybrid. The introduction of these new categories to the traditional Iron Age figurative repertoire necessitates finding specific solutions to visual representation, as shown by the example of the sphinx. Classical representations of the sphinx which appear in continental Europe, for example like that on the applique from the ‘Grafenbühl’ burial at Asperg (Baden-Wurtemberg, Germany), show the profile of these fantastic beings with an animal body sitting on their back haunches, their

5.  Les codes de représentation visuelle dans l’art celtique ancien anthropomorphological face turned towards the viewer. The front legs are clearly individualized in different positions, one turning behind the other, while the long tail curls under the stomach, which comes from the back of the figure, not visible to us (Fig. 5.9a). Revealingly, these figures of the alien sphinx are adapted on the plaque from Weiskirchen ‘Schanzenknöppchen’ (Saarland, Germany) in the form of two ‘double’ sphinxes positioned on either side of a front view of a human face. The ‘unfolding’ construction process of the two lateral parts of the animal bodies of these chimerical creatures is immediately recognisable: only the wing and the lateral legs of each side are represented; the sphinxes are seated on their folded hind legs. The hybrid heads of these monsters – part human, part animal – are also drawn in the form of two ‘unfolded’ lateral profiles, like the composite head of the Cuperly dragon. In a characteristic manner, the two lateral faces of the Weiskirchen sphinxes are in contiguous contact at the neck and chest; which shows that the general form of the monsters is ‘cut out’ along their longitudinal axis of symmetry and is ‘unfolded’ at the back, the ‘fold’ running to the front of the beast (Fig. 5.9b). This method of ‘cutting’ lateral sections is found associated with an ‘unfolding’ on either side of the anterior axis of the body in images of doubled animals, which are particularly seen on openwork belt clasps: in this case these obviously represent the two symmetrical faces of a single animal, each with the same layout. Here again, the two faces are connected at the chest and neck, notably as on the clasp from the chariot burial at Somme-Bionne (Marne) (Fig. 5.10). It is possible that the Early La Tène bronzesmiths of Champagne had to resort to this method when they were confronted with a one-sided medium, or reduced to two dimensions, which prevents a figure from being shown in its full shape, which necessitates its ‘unfolding’ or ‘rabatment’. In this case, we find here more than a simple effect of copying or transposing Mediterranean models, such as sphinxes or griffons. Rather, it is about appropriating foreign figures into figurative representation codes specific to the North Alpine Iron Age. For craftsmen of the Iron Age, it is important above all to keep the forms proper to the figures represented, which is why there is no reference to perspective, which obscures and distorts the essential constituent elements of the figures.

Drawing in the round: three-dimensional projection of rabatment’s visual codes in graphic arts. Representations on curved surfaces emphasise even more sharply than representations on two-dimensional surfaces the spatial articulation of these different schemes of visual

53

rabatment in which the essential elements of the figures are distributed. Indeed, the three-dimensional realisation of figures appears to be understood fundamentally as a process of restoring visual rabatment schemes to their original position in space, which must necessarily be ‘unfolded’ on flat surfaces. The relief figure of a small horse from the Waldalgesheim oenochoe (Hessen, Germany: Joachim 1995, fig. 26) clearly shows, for example, how animal images are constructed. The object comprises the two perfectly symmetrical lateral sections (one left and one right) of an animal drawn in profile, which are directly applied to one another: straight on the figure only appears to have one leg at the front and back. Viewed from the front, the single front leg still possesses two juxtaposed hooves, drawn as a triangle: these last clearly show that this single leg is actually the result of a ‘collage’ of two different limbs seen in profile (Joachim 1995, fig. 23, 7) (Fig. 5.11a). Thus we find here, in an object in high relief, the method of representation ‘by lateral sections’ belonging to animals which are seen on engravings. The quadruped with a human face from the Reinheim oenochoe (Saarland, Germany: Echt 1999) posed a particular problem of representation in so far as it is a hybrid monster. As with the little Waldalgesheim horse, there is a process of constructing the animal part of the image by superimposing the two lateral halves of the horse’s body, which appear perfectly symmetrical. Here, unlike the Waldalgesheim relief, the pairs of legs are individualized, but they are drawn in exactly the same anatomical position on each of the lateral halves. In contrast, the human face of the monster, which could not be represented by bonding the two lateral halves, as with animal heads, has been given special treatment: of necessity drawn full face, it was therefore placed perpendicularly to the general axis of the animal’s body, that is to say at 90 degrees to the object’s horizontal surface (Fig. 5.11b). This special type of construction is found on hybrid beings, half-human half-animal, on pairs of monsters placed on either side of the opening of the Glauberg oenochoe (Hessen, Germany: Herrmann and Frey 1996, figs. 93–95). These monstrous creatures are shown turning their heads back towards the armoured character seated cross-legged at the beginning of the handle: to do this, the human face of the monsters, represented face-on, has been placed perpendicularly to the axis of the body of the animals sitting on their hind legs, and at 90 degrees relative to the horizontal surface of the medium. It is notable that the actual body of the animals is shown in profile, by the standard method of bonding the two symmetrical lateral faces (Fig. 5.12). A human head is placed in front of the forelegs of the monsters, also shown frontally, that is to say perpendicularly to the axis of the beast. To preserve its full legibility, the bronze sculptor has not shown it vertically, as with the head of the monster, but lying horizontally on the surface.

54

Laurent Olivier

This is a similar construction method, appropriate to hybrid beings, recognizable in the depictions of monsters on the Dürrnberg oenochoe (Austria: Jacobsthal 1969, I, 201, cat. n 382; II, pl. 184–187). Here again, the animal body of the voracious beast on the handle is formed in typical fashion by bonding the two lateral symmetrical halves of the animal, whose legs are fixed on the surface. On the other hand, the anthropomorphic mouth of the monster is shown normally in frontal view. Here, the need to represent the monster holding a human head between its front paws posed a particular problem, in that this severed head, like the anthropomorphic head of the monster itself, also needed to be shown in a frontal view. So the two heads were shown perpendicularly to the general axis of the beast’s body, which was shown in profile. However, the severed human head was drawn vertically, at 90 degrees to the horizontal surface of the object, while the monster’s head had to be lying horizontally above it, thus forming an angle of 180 degrees to the surface, so that the two heads are equally visible from the front. These constraints in representation confer a strange appearance to the whole thing, in which the face of the monster appears directly projected in the extended spine of the beast, while the human head seems to be fixed directly under his chin (Fig. 5.13).

Conclusion: Early Celtic art visual representation codes The figurative universe of continental Iron Age Europe experienced important transformations during the course of the 5th century BC. Schematically, one could say that the Hallstatt depictions essentially represent objects (like chariots, vessels, tools or weapons), animals (horses, bovines, cervids…) and human beings (male or female). Figurative representations from the beginning of the Late Iron Age introduce a new category, intermediary between humans and animals: this consists of hybrid beings, halfhuman and half-animal, chief of which are images of the sphinx. There is also – is this an indirect consequence of the above? – the fact that the human figure, represented in its entirety during the Early Iron Age, is now only shown in part. The limbs, which generally constitute a minor graphic element in Hallstatt depictions, completely disappear this time, losing their function to the body, which consequently also disappears. Thus the human figure is essentially reduced to the head, which had never really been developed before, from a graphic point of view, in the Early Iron Age. So the grid of visual representation codes which were established at the beginning of the Middle Iron Age can be summarized as follows: Animals (especially quadrupeds) are represented in full, the body and head shown in profile. Human beings are shown partially, with a reduction of

the body for the head, which is shown from the front. So the human figure takes the form of a mask. Hybrid beings (as in particular the sphinx) are shown in full, their animal body in profile and their human face in frontal view. Thus, in figurative representation in La Tène art, moving figures are not represented in the form of occlusions or anatomical rotations – as in naturalistic depictions – but by way of an assemblage of planes, essentially perpendicular (like faces and profiles). This is a form of visual thinking which ignores the distorting effects of perspective, and which allows the representation, conditional on the alteration of forms, of the deployment in space of different sides of a single object or figure. As Luquet emphasised: Perspective not only has the effect of hiding a greater or lesser part of an object from view when it is masked by another object situated between it and the viewer. Even for an object placed in the foreground, it gives certain parts of it a different shape from their characteristic appearance. Intellectual realism, by contrast, maintains that all elements in the object of which they are part should keep the shape demanded by the original: each of them is shown as if seen from the point of view which shows this aspect (…) The primitive artist is not at all concerned to signify by the form that he gives to each detail in his drawing that if one sees the object from this point of view, the detail would appear with this aspect: he simply wants to show the detail as he conceives it to be. If the shape which each element of an object presents in the drawing is necessarily visual in its nature, this shape is only the translation of its essence, that is to say an idea of an intellectual nature; it could be said that in the whole object, each of its parts is expressed by an ideogram (Lucquet 1930, 152).

After the 4th century BC, La Tène art is at a later stage in this process of the visual decomposition of figures, which results, as Luquet outlined, in achieving basic shapes which act as ‘ideograms’. The complex creations of the early continuous style, with their methods of cutting motifs (like those, classical, of the palmette or the lotus flower) and of recomposing the elementary shapes obtained by the effects of mirror symmetry, fundamentally corresponding to an attempt to create ‘hyper-figures’, developed simultaneously in a multitude of ways. These idealized forms can then be projected onto relief surfaces which stretch and deform them: here we find a process basically very similar to anamorphosis. In formal terms, it should be noted that these processes of folding and assembling shapes allows the creation of a multiplicity of new combinations of motifs and figures: throughout the 5th and 4th centuries BC, this dynamic would feed directly into the iconological production of that which Jacobsthal termed ‘Early Celtic Art’. When the potential of the stylistic repertoire which it exploited has been exhausted, this dynamic of stylistic creation would finally lead to the baroque phase of the Plastic Style, of which it is basically an extension. Translated by S. Crawford.

5.  Les codes de représentation visuelle dans l’art celtique ancien

Bibliography Bacault, M. and Flouest. J.-L. 2002. Schémas de construction des décors au compas de phalères laténiennes de Champagne. In O. Buchsenschutz (ed.) Décors, images et signes de l’âge du Fer européen, Actes du XXVIe colloque de l’Association française pour l’Etude de l’Age du Fer, Paris et Saint-Denis 2002, Revue archéologique du Centre de la France supplément 24, 145–70. Tours: FERACF. Biel, J., 1985. Der Keltenfürst von Hochdorf. Stuttgart: K. Theiss. Bretz-Mahler, D. 1971. La civilisation de La Tène I en Champagne. Le faciès marnien, XXIIIe supplément à Gallia. Paris: C.N.R.S. Cordier, G. 1975. Les tumulus hallstattiens de Sublaines (Indre-etLoire), I. Etude archéologique, L’Anthropologie 79.3, 451–82. Dechelette, J. 1914. Manuel d’archéologie préhistorique, celtique et gallo-romaine. II: archéologie celtique ou protohistorique. III second âge du Fer ou époque de La Tène. Paris: A. Picard. Echt, R. 1999. Das Fürstinnengrab von Reinheim, Studien zur Kulturgeschichte der Früh-La-Tène-Zeit, Saarbrücker Beiträge zur Altertumskunde 69. Bonn: Rudolf Habelt. Fourdrignier, E. 1880. Sur la découverte de deux casques gaulois à forme conique dans les sépultures de Cuperly et de Thuizy (Marne), Bulletin de la Société d’Anthropologie de Paris 3, 323–33. Guillaumet, J.-P. 2003. Paléomanufacture métallique: Méthode d’étude. Gollion: Infolio Ed. Haddon, A. C. 1895. Evolution in Art. London: W. Scott. Herrmann F.-R. and Frey O.-H. 1996. Die Keltenfürsten vom Glauberg. Ein frühkeltischer Fürstengrabhügel am Hang des Glauberges bei Glauburg-Glauberg, Wetteraukreis. Archäologische Denkmäler in Hessen 128/129. Wiesbaden: Archäologische Gesellschaft in Hessen. Hubert, H. 1932. Les Celtes et l’expansion celtique jusqu’à l’époque de La Tène, L’évolution de l’humanité XXI. Paris: La Renaissance du Livre.

55

Jacobsthal, P. 1969. Early Celtic Art, 2nd ed. Oxford: Clarendon Press. Luquet, G.-H. 1930. L’Art primitif, Bibliothèque d’Anthropologie 13. Paris: Gaston Douin. Lambot, B., Verger, S. and Meniel, P. 1995. Une tombe à char de La Tène ancienne à Semide (Ardennes). Mémoires de la Société Archéologique Champenoise 10. Reims: Société Archéologique Champenoise. Lantier, R. 1946. L’Art celtique (commentaire des ouvrages de Paul Jacobsthal Early Celtic Art 1944, and Fernand Benoit L’art primitif méditerranéen de la vallée du Rhône: la sculpture 1945), Journal des Savants, 67–77. Pare, C. F. E. 1992. Wagons and Wagon-graves of the Early Iron Age in Central Europe, Oxford Committee for Archaeology Monograph 35. Oxford: Oxford Committee for Archaeology. Riegl, A. 1923. Stilfragen: Grundlegungen zu einer Geschichte der Ornamentik, 2nd ed. Berlin: Schmidt. Schaaff, U. 1975. Frühlatènezeitliche Grabfunde mit Helmen von Typ Berru, Jahrbuch des Römisch-Germanischen Zentralmuseums Mainz 20 (1973), 83 and fig. 31.2. Schmit, E. 1926–1928. Répertoire abrégé de l’archéologie du département de la Marne des temps préhistoriques à l’An Mil, Mémoires de la Société d’Agriculture, Commerce, Sciences et Arts du Département de la Marne XXII, 125–204. Semper, G. 1861. Der Stil in den technischen und tektonischen Künsten oder Praktische Ästhetic. München: F. Bruckmann. Van Endert, D. 1987. Die Wagenbestattungen der späten Hallstattzeit und der Latènezeit im Gebiet westlich des Rheins, BAR International Series 355, 106–7 and pl. 68. Oxford: British Archaeological Reports. Verger, S. 1994. Les tombes à char de La Tène ancienne en Champagne et les rites funéraires aristocratiques en Gaule de l’est au Ve siècle avant J.-C, Unpublished PH.D thesis, vol. I, 172–4, fig. 79 and 81–83. Dijon: Université de Bourgogne.

6 HIDDEN FACES AND ANIMAL IMAGES ON LATE IRON AGE AND EARLY ROMAN HORSE HARNESS DECORATED USING THE CHAMPLEVÉ TECHNIQUE Jennifer Foster

Introduction Vincent Megaw’s work on Celtic art has been an inspiration, but none more so perhaps than in his analysis of the portrayal of the human face, for example in his iconic paper, ‘Mickey Mouse and Cheshire Cat’ (Megaw 1970). The term Cheshire Cat, based on the cat in Alice in Wonderland who gradually disappeared until only his smile was left, was first used by Jacobsthal (1941, 308) to describe the enigmatic faces in Celtic art. Megaw also pointed out two other aspects: the ability to change from human to animal and bird, and the humour in some of these pieces of art (the Mickey Mouse aspects), such as the face on the early La Tène Bad Durkheim gold fragments: a bearded old man who, when turned upside down, becomes a woman (1970, 272). The Celtic artist used caricature, reducing the image so much that it becomes absurd. The face motif begins in the early La Tène art and continues in Britain into the period after the Roman conquest. Megaw discusses the longevity of face motif over this period of c.500 years, describing, for example, the post-Conquest Westhall, Suffolk terret, decorated in red glass in the champlevé technique: ‘Here is a strongly archaic and continental formulation which like others in Iron Age British society seems to appear more or less out of the contemporary artistic subconscious’ (Megaw 1970, 275–6). So there is continuity from earliest Celtic art, leading to a style both progressive and archaic, with old motifs repeated but in a new style. In this paper I should like to explore further the hidden faces on the champlevé decorated harness equipment from Britain.

Champlevé decorated metalwork In the first century AD, a range of objects was developed

in Britain decorated with glass: brooches, horse harness, swords, tankard handles, mounts and bowls. They are particularly found in East Anglia and northern Britain, with a few in Wales (MacGregor 1976; Spratling 1972; the Seven Sisters hoard, Davis and Gwilt 2008). These objects were cast with recesses which were inlaid with glass, usually red, but with some white, blue or yellow. On pre-Roman objects the glass was warmed and softened until malleable, and then pushed into the recess (a technique known as champlevé). As it was not fused, the glass often fell out and is missing. Glass was imported into Britain in the Late Iron Age and was probably in short supply, so it was recycled; this can be seen in the bridle bit from Folly Lane, St Albans (Foster 1999a, 135) where one of the blue glass inlays has a yellow swirl, probably from melting a multi-coloured glass object such as a bead. Some post-Conquest objects (Davis and Gwilt 2008, 155–8) seem to have been inlaid with enamel, powdered glass that is fused by heat to the metal surface; the composition of the glass is different. It is difficult to distinguish between the two techniques without analysis, although the recesses for enamelling would have to be much shallower (Davis and Gwilt 2008, 155–8). Interestingly, on all these objects the metalworking techniques are very similar, particularly the use of incised lines to enhance the inlay. On all the champlevé examples these lines are carelessly executed, often with overlapping scratches. The designs on champlevé metalwork are simple, with large blocks of colour. They are developed from what went before, using motifs such as curved sided triangles, broken backed curves, spirals. Compared to earlier pieces of Celtic art such as the Battersea shield, possibly 3rd century BC in date, with their elaborate designs with many motifs (Stead 1985) the designs on these objects are sometimes seen as simplistic, even ‘degenerate’ (Kilbride-Jones 1980, 95).

6.  Hidden faces and animal images on Late Iron Age and Early Roman horse harness

57

Fig. 6.1. Harness brooches from Folly Lane, St Albans (1) and Polden Hill hoard, Somerset (2). Not to scale.

But in fact, closer examination has revealed some complex designs with hidden faces. The discussion in this paper will be confined to horse harness examples, as listed in Table 6.1. There is a range of images: human faces and animal designs such as horses and birds, e.g. birds of prey, owls, and particularly ducks. These images vary from examples which are very clear, while some are much more questionable, and some may never have been intended as images at all. Most of these designs have not previously been recognised.

The most complex objects are the harness brooches (Table 6.1, Nos 1–3), probably used to fasten a horsecloth to harness, and positioned just above the tail (Fox 1958, 124); the drivers of the chariots would therefore see the brooches on the two ponies just in front of them as the chariot was driven. No. 1 from Folly Lane, St Albans (Fig. 6.1.1) is the most closely dated object. It comes from a very late Iron Age grave (Foster 1999a, 139–42; Niblett 1999). The burial was in an elaborate funerary chamber under a mound, in the centre of a ceremonial rectangular ditched

Jennifer Foster

58

Table 6.1: List of champlevé harness objects discussed in the text 1

Harness brooch

Folly Lane, St Albans

Foster 1999, 139-42

2

Harness brooch

Polden Hill, Somerset (near Bridgwater)

Spratling 1972, no 241

3

Polden Hill, Somerset (near Bridgwater) Santon (Weeting), Norfolk

Spratling 1972 242

4

Harness brooch Strap union

5 6

Strap union Strap union

Santon (Weeting), Norfolk Norton, Suffolk

Spratling 1972, 207 Spratling 1972, no 204

7

Strap union

Westhall, Suffolk

Spratling 1972, no 208

8

Strap union

Polden Hill, Somerset (near Bridgwater)

Spratling 1972, no 205

9

Strap union

Unprovenanced, BM

Spratling 1972, no 214

10

Strap union

Alltwen, Neath

11

Strap union

12 13

Bridle bit Bridle bit

14

Terret

Maendy hillfort, Rhondda Cyfon Taf Birrenswark, Dumfriesshire Rise, Holderness, East Yorks Bolton Museum, Lancs

Davis and Gwilt 2008, 172 Davis and Gwilt 2008, 172 MacGregor 1976, no 2 MacGegor 1976, no 10

15

Terret

Richborough, Kent

Spratling 1972, no 70

16

Terret

Lakenheath, Suffolk

Spratling 1972, no 68

17 18

Terret Terret

Spratling 1972, no 75 MacGregor 1976, no 62

19 20

Terret Terret

Westhall, Suffolk Auchendolly, Kirkcudbright, Dumfries and Galloway Whaplode, Lincs Colchester, Essex

21

Terret

Runnymede, Surrey

Spratling 1972, no 71

22

Linchpin

Colne Fen, Cambs

Jope 2000, 300b and d

23

Linchpin

Colne Fen, Cambs

Jope 2000, 300a and c

Spratling 1972, no 206

Spratling 1972, Fig. 26

Spratling 1972, no 76 Spratling 1972, no 79

enclosure. About AD 55 a man was cremated and then buried with a selection of rich grave goods, most of which had been partially destroyed on the pyre, including harness equipment, fittings for furniture, two pairs of hobnailed shoes and a chain mail shirt. One of the grave goods was a champlevé decorated harness brooch (Fig. 6.1.1). In the

Horses, backed Sideways: smiley human faces Upside down: 2 human faces Eagle heads facing Side ducks (?) Face with O for mouth Upside down – eagle face Sideways turn – cat faces with one ear each ?Owl Man’s face shouting Back to back ducks Lower face ?Face with smile and whirling eyes Opposing faces, yellow eyes and cheeks Sideways: faces with sideways mouths (sneering or kindly) ?Owl faces Sideways turn: ?animal with ears ?grinning face with open mouth and whirling eyes which become ducks when turned sideways Sideways: possible grimacing face with teeth Opposing faces at top and bottom, helmeted? Sideways: Opposing ?hawk faces Opposing smiley faces at top and bottom Bird heads? in circles Figure with half closed eyes, smiling mouth and triangular feet Horse face Horse face Face with trails ending in duck heads 3 very similar from Portable Antiquities Scheme Face with trails ending in duck heads Eyes with trails ending in duck heads Mouth: the open hole of the terret Face with eyes, triangular mouth and hair trails Face with eyes, cheeks and open triangular mouth with hair trails Central eyes; the mouth is the terret opening Central eyes, triangular nose; the mouth is the terret opening Central eyes, smiley mouth. The base of the terret seems to be in shape of a ship with duck heads on either side Two facing ducks or a human face with large blue eyes and a grinning mouth below with pointed teeth Animal face (? boar), with frowning eyes, a snout below and two nostril holes, and spiral cheeks

centre of the brooch facing away from either side of the hole are shapes which can be interpreted as horse’s heads with turned up coiled noses (Fig. 6.1.1a), similar to horses’ heads on dragonesque brooches (Hunter 2008, 140). If the brooch is turned sideways it is possible to see a strange humanoid face on either side, with triangular mouth, yellow eyes and

6.  Hidden faces and animal images on Late Iron Age and Early Roman horse harness

59

Fig. 6.2. Harness brooch from Polden Hill Somerset (3) and strap unions from Santon, Norfolk (4–5). Not to scale.

blue cheeks (Fig. 6.1.1b). These are truly hidden faces as they cannot be viewed unless the brooch is turned sideways, which it would not be in use. If turned upside down there are two similar, though less clear, smiling faces side by side with triangular mouths and blue eyes (Fig. 6.1.1c). No. 2 from the Polden Hill hoard (Fig. 6.1) has two opposing heads of birds of prey (Fig. 6.1.2a), probably eagles though they are very stylised. The huge curved beak and rounded head crest are similar to those on a Golden Eagle or White-tailed Sea Eagle (Bruun 1980, 73), both of which could have been seen in Britain in prehistory. Other birds of prey have smaller beaks. At the base of the necks of the eagles are two possible ducks facing outwards (Fig. 6.1.2b) with pointed beaks, large circular eyes and a short neck with rounded back. These ‘ducks’ are very vague, but clearer if

compared to the image on the repoussé bronze plaque from Llyn Cerrig Bach, which has been variously been described as a duck, bird or puffin (Spratling 2008, 195–9). Spratling even tried to see a quadruped (ibid. 195), but a generic bird is more likely. On the Polden Hill harness brooch, in the corner below the ‘ducks’, are two incised ellipses, which at first sight are presumed to be there to fill in the corner space. The central portion at the base has a possible human face (Fig. 6.1.2c), with the hole as a tear-shaped nose, two curls for eyes on either side and an ‘O’ for a mouth. If this object is turned upside down (Fig. 6.1 reversed), the human face becomes an imposing eagle face (Fig. 6.1e), looking straight at the viewer with the hole forming its beak and its eyes formed of two spirals on either side. Like both the Golden and White-tailed Sea Eagle it has an imposing brow

60

Jennifer Foster

Fig. 6.3. Strap unions from Norton, Suffolk (6), Westhall, Suffolk (7) and Polden Hill, Somerset (8). Not to scale.

ridge above the eyes (Bruun 1980, 73). It looks just like the eagle on the beautiful linchpin from Manching (Megaw and Megaw 1989, fig. 224, left). On either side the ducks have turned into cat faces (Fig. 6.1d), with circular red eyes and a lopsided grinning mouth, and with those enigmatic engraved ellipses forming an ear for each. No. 3 (Fig. 6.2.3) could not be more of a contrast. Also from the Polden Hill hoard it has a very simple design. There may be an owl face at the base with the hole forming its beak and two red eyes set asymmetrically in recesses, looking like the facial disc of an owl (e.g. Little Owl: Bruun 1980, 173). The series of strap unions (which are assumed to be for joining the harness components for chariot ponies) are very varied in design. The most special is No. 4 from Santon,

Norfolk (Fig. 6.2). At the top is a man’s face shouting or wailing (Fig. 6.2.4a), with the hole forming his downturned mouth and his bulbous nose outlined with an engraved line. His two eyes, brows and cheeks are inlaid with red glass and at the sides of his face are hair curls. On the lower half of the strap union are two ducks with their backs to each other (Fig. 6.2.4b), with large red circular eyes and their beaks facing outwards; they are very similar in shape to those on No. 2. There is possibly another small hidden face between them (Fig. 6.2.4c) with two eyes formed from their wings and a bulbous nose outlined with an engraved line. If the strap union is turned upside down, the whole design is reversed, and the face becomes ducks and the ducks form the screaming face.

6.  Hidden faces and animal images on Late Iron Age and Early Roman horse harness

61

Fig. 6.4. Strap unions; unprovenced (9) and from Maendy hillfort, Rhondda (11) and Alltwen, Neath (10); bridle bits from Birrenswark, Dumfriesshire (12) and Rise, East Yorkshire (13). Not to scale.

Another strap union from this hoard (No. 5, Fig. 6.2) at first sight seems to be decorated with simple interlocking commas. It is possible, however, that there is a face (Fig. 6.2.5a), with mad twirling eyes in the centre panel, the curved sided triangle of the hole forming a nose and a thin smile below; two interlocking commas form the cheeks. If turned sideways (Fig. 6.2.5b) another possible face appears. Another complicated design is on the strap union from Norton, Suffolk (No. 6, Fig. 6.3). There is a face at the top with yellow eyes and cheeks and a triangular mouth (Fig. 6.3b). If turned upside down there is a mirror image face opposite. If turned sideways, however, on the left side of the object are two scowling faces squashed together (Fig.

6.3a), with sideways mouths formed by an engraved triangle infilled with punchmarks; the eyes are of yellow glass and are made to look angry by an engraved eyebrow line. The two faces on the right hand side seem to be mirror images, but these two look happy because the scowling brow lines are omitted. As on the Folly Lane brooch, these side faces would not be observed when the object was in use. Strap union No. 7 from Westhall, Suffolk (Fig. 6.3) has a very complicated design. There is possibly an owl face at the top (Fig. 6.3a), with red curves forming the facial disc and large round blue eyes. The beak is outlined by an engraved curved-sided triangle. The shape of the face is similar to that of a Tawny Owl (Bruun 1980, 177). When turned sideways

62

Jennifer Foster

Fig. 6.5. Terrets from Bolton Museum (14), Richborough, Kernt (15), Lakenheath, Suffolk (16), Westhall, Suffolk (17), Auchendolly, Kirkudbright (18), Whaplode Drove, Lincs (19), Colchester, Essex (20) and Runneymede, Surrey (21). Not to scale.

(Fig. 6.3b) a possible animal with pointed ears and blue eyes can be seen. If No. 8 from Polden Hill (Fig. 6.3) has hidden faces they are well disguised. The main body of the strap union has four circles which seem to contain pairs of facing ducks (Fig. 6.3a). These are much clearer and more complete than those on Nos 2 and 4, as both the head and body are outlined; they appear to be swimming. However, if the strap union is viewed as a whole, the top two ducks could be seen as twirling eyes, with the tear-shaped hole below as the nose and the triangular holes forming a smirking mouth; if it is an image of a strange animal it has pointed ears at the top, pointed feet at the bottom and curved paws to the sides.

If the object is turned sideways these curved paws change again (Fig. 6.3b) and appear to contain faces, although these are highly schematic: a bulbous nose, two eyes formed by a joined pair of comma shapes and a curved lower mouth with pointed teeth. Circles could represent cheeks. Another unprovenanced strap union, No. 9 (Fig. 6.4) appears to have helmeted heads with side curls, opposed at both top and bottom (Fig. 6.4a); alternatively these may be animal faces with a central snout and pointed ears. This face is very similar to one on a fragmentary terret found in Kent and recorded under the Portable Antiquities Scheme (Kent-91CFD2), but this can only be seen if the terret is

6.  Hidden faces and animal images on Late Iron Age and Early Roman horse harness

63

Fig. 6.6. Linchpins from Colne Fen, Cambs (22 and 23). Terrets from Whaplode Drove (19) and Colchester (20) with reins, looking like tongues. Not to scale.

turned upside down. The side panels on No. 9 seem to be the heads of birds of prey (Fig. 6.4b), with the eyes and beak picked out in red and the triangular beak outlined with an incised line. The small beak makes it look more like a small bird of prey such as a kestrel, rather than an eagle or kite (Bruun 1980, 91), although the ridge on the top of the head could indicate an osprey (Elphick and Woodward 2003, 211). A strange human face seems to peer out from the top of a strap union from Alltwen, Wales (No. 10; Fig. 6.4), with spirals for the eyes and a curved-sided triangle below forming a laughing mouth; this design is repeated opposed at the bottom. The shaft is like No. 8 with two wide panels forming the body; two circles seem to have back-to-back bird heads with curving beaks and comma shaped eyes. Finally, a very different enamelled strap union from Maendy hillfort, Rhondda Cynon Taf (No. 11; Fig. 6.4), has bright red and yellow decoration but unlike all the others is not in a balanced design; it is probably of Roman date (Davis and Gwilt 2000, 172). The figure has half closed eyes and a smiley mouth, balanced with triangular feet; when reversed it has wide open eyes with the same laughing mouth. There are two champlevé decorated bridle bits with possible faces on the side links; appropriately enough they appear to be horses’ heads. One from Birrenswark,

Dumfriesshire (No. 12, Fig. 6.4) has the eyes (a dot within a circle) and designs in the ears picked out in red; the junction with the centre link is swollen to form the nose. The other (No. 13; Fig. 6.4) from Rise, East Yorkshire, has a similar though less clear face, with the eyes picked out with a flower design in red and blue and a swollen nose as on No. 12, interlocking the side link to the centre link. Bridle bits often had one side link more highly decorated than the other: it is assumed that ponies were driven in pairs and the more decorative side would face outwards. Neither of these heads would be seen when the bridle was on the pony as they would be turned sideways. The terrets or reinrings (Nos 14–21; Fig. 6.5) are all very similar in design, with two circles in the centre, perhaps for eyes, and trails down either side, possibly representing hair, e.g. No. 14 from Bolton Museum and No. 17 from Westhall, Suffolk (Fig. 6.5). Some of these trails end in duck head shapes, e.g. No. 14; two very similar terrets were recorded under the Portable Antiquities Scheme, WMID-935475 and NIMS996511. With some, e.g. No. 19 from Whaplode, the terret hole appears to represent the mouth, with two open eyes above; here the terret hole is swollen as if to represent lips. Others seem have a possible mouth in the form of a curved sided triangle, e.g. No. 15 from Richborough and

64

Jennifer Foster

No. 18 from Auchendolly, with two yellow eyes and cheeks; like the harness brooch from Folly Lane this has circles for cheeks as well as eyes. In some, e.g. No. 20, a curvedsided triangle seems to represent the nose. The terret from Runneymede, No. 21, has a possible face with two circular red eyes and a curved-sided triangle for a smiling mouth, but the base of the terret appears to be in the form of a ship with birds heads at prow and stern. Two linchpins (Fig. 6.6) from Colne Fen, Cambs are possibly decorated with faces. No. 22 (as suggested by Jope 2000, 314), appears to have two facing ducks in the centre with round blue eyes, but look closer and this becomes a human face with blue eyes and a curved grinning mouth below with pointed teeth created by engraved lines. This object is a classic case of ‘now you see it, now you don’t’. No. 23 appears to have an animal face, perhaps that of a boar, with in-turning eyes divided by back-to-back trumpet shapes; these make the face look as if it is frowning, but is typical of the wrinkled snout of a male boar. Below is the snout with two nostril holes; the spirals on either side appear to be the cheeks of the animal or possibly the wrinkles of skin made by the tusks.

Discussion In conclusion, it is clear that there were many more hidden faces and animal images on champlevé metalwork than has hitherto been recognised. Only one (No. 22) was suggested by Jope (2000, 314). Some of these images are very clear (e.g. the human wailing face on No. 4). Some are carefully hidden, such as the eagles and hawk on No. 2 and the human face on No. 1. Some are much more enigmatic and may not have been intended as faces or animal images at all: these include most of the ducks, the cat on No. 2 and the faces with both eyes and cheeks of the same colour (e.g. Nos 6 and 18). It is important to point out that not all modern viewers can see these faces, and it is quite possible that designs on the terrets in particular described above were not intended to be viewed as faces. It is also clear that the carelessly-applied incised lines, previously supposed to be unimportant, are actually an essential part of the design, forming, for example, the cat ears on No. 2 and the bulbous nose on No. 4. The images are quite restricted in type: human faces, horses, ducks, birds of prey and owls. Human faces are very widespread in Celtic art, as are horses in British art, though birds of prey and owls are less common. The ducks are difficult to see, though the examples pointed out here suggest that they might be widespread. Other representations of ducks in Britain include a number of sites with duckdecorated pottery in the Welsh Marches in the Later Iron Age (e.g. Midsummer, Hill, Herefordshire: Stanford, 1981, 139–41). Ducks were very common in Late Bronze Age and

Early Iron Age art and it is postulated that they may have represented the sun chariot (Kaul 1998, 242–6), but are less common in later Iron Age art. There is one example of a cat in the objects represented here: a rare animal in Celtic art, though known in Wales (e.g. the early Roman pan handle from Snowdon: Redknap 2011, 75) and Ireland (e.g. the unprovenanced horsebit ring: Raftery 1983, fig. 43, no 129). The most difficult question, of course, is why the faces were included in the design. They are definitely the hidden faces of the Cheshire Cat. Because some are very obscure, they would not be recognised as faces without reference to the more obvious examples, of which they are apparently abbreviated or shorthand forms. The fact that these same elements recur again and again suggests that they could represent stories that the audience knew, such as mythical or folk tales. First Nation art in British Columbia, Canada (Kramer 1999) has the same animals and birds again and again, Thunderbird, Bear, Salmon, Beaver and particularly Raven, illustrating the important stories from myth. Stories about Raven show him to be a figure of both reverence and ridicule. He is helpful to humans, rescuing the sun, moon and stars and releasing the first salmon, but is also a trickster, thief and bounder. Like the figures in Celtic art, Raven is a sublime transformer, disguising himself as a baby, woman or animal, and then flying away from trouble in his own guise. This art is on everyday items: baskets and wooden artefacts such as boats and containers, and also on ritual artefacts associated with feasting and death, such as totem poles. There may have been other reasons why the images were included in the design. Perhaps these faces are for good luck, to encourage the forces of speed to take over the chariot or prevent it from turning over. For example the Polden Hill harness brooch (No. 8) has a heavily disguised grinning face and the Folly Lane faces (No. 1) could be interpreted as smiling. The faces on the terrets would have been particularly effective when the chariot was being driven, with reins moving through the mouths of the terrets like thrusting tongues (Fig. 6.6). However, some of the faces are far from pleasant, suggesting that they were not necessarily beneficial. That on the Santon strap union (No. 4) is of a man shouting or wailing in pain, whilst the sideways faces of the Norton strap union (No. 6) have frowning and sneering expressions. The arms of the creature on No. 8 when turned sideways appear to turn into grimacing faces with pointed teeth. Most Celtic art in Britain does not contain hidden faces and there are plenty of champlevé examples without, some from the same hoards as the ones described here. Why did some objects have faces and others not? Were those without faces made as copies by people who did not have the Celtic art tradition behind them, so did not recognise the hidden elements, or were they deliberately designed without? It is clear, however, that some designs (e.g. the Norton and Santon strap unions and the Polden Hill harness brooch) are

6.  Hidden faces and animal images on Late Iron Age and Early Roman horse harness so complicated they were definitely designed to have the faces. The theme of shape-changing is also evident in these objects; the trailing hair becomes a duck when turned upside down, the smiley face becomes a bird. Shape-changing, as in First Nation art from Canada, is a feature of the later Celtic folk tales, e.g. in the Irish tales of the Morrigan she appears as a raven and changes to eel, wolf and heifer (Green 1992, 150). Other figures become birds or boars; in the Mabinogion Trwrch Trwyth was changed from a king into a boar as a punishment (Jones 1975, 105–111). In the Welsh tale Math son of Mathonwy, Lleu becomes an eagle, and his wife, who was originally made out of flowers, is turned into an owl (Ross 1967, 274). Perhaps some of the designs illustrated here reflect earlier versions of these stories. These stories were written down many centuries after the making of the metalwork so their relevance is open to debate; however, there is some evidence that they contain traces of a much earlier oral tradition (e.g. Koch 1987). The question remains: why were these faces hidden? It is difficult to assess whether people in the Iron Age would have been able to see them. Everyone in British Columbia would have recognised the designs of First Nation art; though there were some specialist woodworkers, in general this art was (and still is) made by the people using it, though this is not the case with all anthropological examples (Joy 2011, 212). There is folk art of this type in Britain in the Iron Age, for example geometric designs on antler weaving combs, e.g. from Glastonbury and Meare (Bulleid and Gray 1911, 266–99; 1948, 61–88). It is possible, of course, that the Celtic art images on metalwork are abbreviated forms of images familiar to people in other media, such as in textiles, wood, house painting or body decoration, but if that is the case, they have not survived archaeologically. There are some domestic objects decorated with Celtic art but they are rare. Wooden vessels and pottery from Glastonbury and Meare do have Celtic art designs on them (Bulleid and Gray 1911, Chapter IX; 1917, 505–12; 1948, 15–88), and there are examples of stone objects, such as a steatite spindle whorl from Irby, Wirral (Foster 1999b). One antler weaving comb from Langbank, Renfrewshire in Scotland (MacGregor 1976, 275) even has a face on it when turned upside down. In complex societies, art is generally produced by and for a smaller proportion of the population. Roman art has more to do with status (e.g. wall paintings) and state imposition (of religion and monumental architecture); specialist manufacturers designed and created the art in separate workshops. Celtic art is also on status objects, rich metalwork and gold, and on objects associated with status: swords and feasting equipment and equestrian equipment; the presence of such objects in the rich grave at Folly Lane highlights this. As a result of the high-status value of Celtic art objects, most people in Late Iron Age society would not be able to use them. Does that mean that the people

65

viewing the objects did not know they had hidden faces? Did even the people using the artefacts know they were there? Interestingly as the champlevé decorated metalwork survives into the Roman period, it was used to decorate different classes of objects with more domestic associations, items such as jewellery, perhaps a reflection of the changing position of the art in the perception of people after the conquest (Hunter 2008, 136). Given the sophistication of Iron Age metalwork, it was obviously a society capable of sustaining specialist metalworkers. It is difficult to say how many producers there were. Based on the similarity of designs on Iron Age metalwork, Leeds originally proposed (1958, 144–5) that there were a very few centralised workshops, perhaps two in southern Britain and one in the north, while in contrast Joy recently suggested that many people would have been involved in the collection and processing of materials to produce one metal artefact such as a sword (Joy 2011, 208), perhaps involving the entire community. However, the small domestic scale of the metalwork-production at Gussage All Saints and Weelsby Avenue, Grimsby (Foster 1980; 1996) supports Leeds’ idea that very small numbers of people were involved, though perhaps not in centralised workshops. Weelsby Avenue, a small site with just two houses and a surrounding ditch full of discarded casting debris, seems to have been a permanent production site concentrating on terrets. The metal-working at Gussage, however, seems to have been a single event, perhaps conducted by a group of itinerant metalworkers bringing with them raw materials and designs. In both cases, the production was probably conducted by an extended family. It seems unlikely that the entire community was involved in what was a highly specialised profession. The suggestion that there were only a few workshops is also not supported by the distribution of metalwork decorated with Celtic art (Garrow 2008), which shows a widespread distribution across Britain, with a particularly interesting density in East Anglia. The distribution of this small group of champlevé objects (Fig. 6.7) is remarkably similar, with a larger number in East Anglia, but with a spread across southern England, into southern Wales and with a few sites in northern England and two in Scotland. This is not, of course, the distribution of production sites, but where these objects ended up, mostly in metalwork hoards. The final question is: who were the artists? It seems probable that in the Late Iron Age it was the metalworkers, the producers of the objects, who were also the designers. It is clear that the surviving champlevé objects were designed by a number of different artists; it is difficult to believe Nos 2 and 3 were designed by the same person, despite the fact that they came from the same hoard. The terrets illustrated here may have been, but are much simpler designs. It is possible that only one person held the secret of the faces, one man or one woman, the designer of the art, that

66

Jennifer Foster

Fig. 6.7. Distribution of champlevé-decorated harness eqipment.

6.  Hidden faces and animal images on Late Iron Age and Early Roman horse harness even the metallurgists did not understand the designs that they were making. This seems unlikely. It is also possible that the people using the objects did not see the hidden faces: human, animal and bird; this is more possible. A third alternative is that every person in Late Iron Age society saw these faces, and knew what they represented, that they were meaningful, frightening, special, or funny.

Acknowledgements Thank you first so much to Vincent Megaw for being such an inspiration, for seeing things and having the courage to acknowledge them in print. My thanks also to Ian Stead and Val Rigby for some excellent discussions about metal objects and to Mansel Spratling for his paper on the Cerrig y Drudion birds – I would not have seen some of these faces without that. Thank you also to my husband Martin Bell for his positive response to the faces.

Bibliography Bruun, B. 1980. Birds of Britain and Europe. London: Hamlyn. Bulleid, A. and Gray, H. S. G. 1911. The Glastonbury Lake Village Volume 1. Glastonbury: Privately printed. Bulleid, A. and Gray, H. S. G. 1917. The Glastonbury Lake Village Volume 2. Glastonbury: Privately printed. Bulleid, A. and Gray, H. S. G. 1948. The Meare Lake Village Volume 1. Glastonbury: Privately printed. Davis, M. and Gwilt, A. 2008. Material, style and identity in first century AD metalwork, with particular reference to the Seven Sisters hoard. In D. Garrow, C. Gosden and J. D. Hill (eds.) Rethinking Celtic Art, 146–84. Oxford: Oxbow. Elphick, J. and Woodward, J. 2003. RSPB Pocket Birds. London: Dorling Kindersley. Foster, J. 1980. The Iron Age moulds from Gussage All Saints. London: British Museum Occasional Paper. Foster, J. 1996. Metalworking in the British Iron Age: The evidence from Weelsby Avenue, Grimsby. In B. Raftery (ed.) Sites and sights of the Iron Age: essays on fieldwork and museum research presented to Ian Mathieson Stead, Oxbow Monograph 56, 49–61. Oxford: Oxbow. Foster, J. 1999a. The Funerary Finds. In R. Niblett (ed.) The Excavation of a ceremonial site at Folly Lane, Verulamium, Britannia Monograph Series 14, 133–69, 175. London: Society for the Promotion of Roman Studies. Foster, J. 1999b. Steatite spindle whorl. In R. Philpott and M.

67

Adams (eds.) Irby, Wirral: Excavations on a Late Prehistoric, Romano-British and Medieval Site, 1987–96, 104–6. Liverpool: National Museums Liverpool. Fox, C. 1958. Pattern and Purpose. Cardiff: National Museum of Wales. Garrow, D. 2008. The space and time of Celtic Art: interrogating the ‘Technologies of Enchantment’ database. In D. Garrow, C. Gosden, and J. D. Hill (ed.) Rethinking Celtic Art, 15–39. Oxford: Oxbow. Green, M. 1992. Dictionary of Celtic Myth and Legend. London: Thames and Hudson. Hunter, F. 2008. Celtic art in Roman Britain. In D. Garrow, C. Gosden, and J. D. Hill (ed.) Rethinking Celtic Art, 129–45. Oxford: Oxbow. Jacobsthal, P. 1941. Imagery in Early Celtic Art, Proceedings of the British Academy 37, 301–20. Jones, G. 1975. Welsh Legends and Folk-tales. London: Oxford University Press. Jope, E. M. 2000. Early Celtic Art in the British Isles. Oxford: Clarendon Press. Joy, J. 2011. ‘Fancy objects’ in the British Iron Age: Why decorate?, Proceedings of the Prehistoric Society 77, 205–30. Kaul, F. 1998. Ships on Bronzes: a study in Bronze Age Religion and Iconography. Copenhagen: National Museum of Denmark. Kilbride-Jones, H. E. 1980. Celtic Craftsmanship in Bronze. London: Croom Helm. Koch, J. T. 1987. A Welsh Window on the Iron Age, Cambridge Medieval Celtic Studies 14, 17–52. Kramer, P. 1999. Totem Poles. Alberta: Altitude. MacGregor, M. 1976. Early Celtic Art in North Britain. Leicester: Leicester University Press. Megaw, J. V. S. 1970. Cheshire Cat and Mickey Mouse: analysis, interpretation and the art of the La Tène Iron Age, Proceedings of the Prehistoric Society 36, 261–79. Megaw, R. and Megaw, J. V. S. 1989. Celtic Art. London: Thames and Hudson. Ross, A. 1967. Pagan Celtic Britain. London: Routledge and Kegan Paul. Raftery, B. 1983. A Catalogue of Irish Iron Age Antiquities. Marburg. Redknap, M. 2011. Discovered in Time: Treasures from Early Wales. Cardiff: National Museum of Wales. Spratling, M. 1972. Southern British Decorated Bronzes of the Late Pre-Roman Iron Age. PhD. Thesis, University of London. Spratling, M. 2008. On the aethestics of the Ancient Britons. In D. Garrow, C. Gosden, and J. D. Hill (ed.) Rethinking Celtic Art, 185–202. Oxford: Oxbow. Stanford, S. C. 1981. Midsummer Hill: an Iron Age hillfort on the Malverns. Hereford: Print Logic. Stead, I. M. 1985. The Battersea Shield. London: British Museum Press.

7 THE HUMAN MASKS OF UNKNOWN PROVENIENCE Mitja Guštin

Introduction On the occasion of this celebration of the high jubilee of Prof Vincent Megaw, a specialist on the beautiful La Tène style of the Celtic World, I would like to make a short contribution with the presentation of an outstanding collection of three human face masks from the collection kept in the Burgmuseum in Deutschlandsberg (Styria/Austria). The origin of these masks is unknown; two of them were given to the museum in the last decade of the 20th century, and the third one arrived a bit later through the antiquities trade after its distribution on the black market, a business that flourished as a result of the wars in the western Balkans. The first, the golden one (Fig. 7.2), has been part of the permanent exhibition of the Burgmuseum Deutschlandsberg for more than 15 years and was published in the first exhibition catalogue of the Gebrüder Steffan Fund (Kelten 1998, 36) with the subtitle ‘Ostkeltische Totenmaske aus Goldblech, 5.–4. Jh. v. Chr.’ (Eastern Celtic funeral mask made of gold sheet, 5th–4th century BC). Later it was featured in the well-known exhibition dedicated to Celtic culture and heritage in Castello di Gorizia (Italy), and published in its exhibition catalogue (Gli echi della Terra 2002, 87, fig. 86). Finally it appeared in a monograph in the Festschrift for Barry Raftery (Guštin 2009a). The second, in the permanent exhibition of the Burgmuseum Deutschlandsberg, is a remarkable bronze plate with a depiction of a human face. It was published for the first time in the same exhibition catalogue from Castello Gorizia (Gli echi della Terra 2002, 87, fig. 85). The plate was pointed out again in an overview of representations of human faces from the La Tène period in Slovenia (Fig. 7.3) (Guštin 2006, 127, 128, Abb. 6). The third mask, the silver one, came into the collection of the Burgmuseum Deutschlandsberg a few years ago and

was immediately exhibited in the museum hall together with a silver torc and a bronze buckler and scoop, suggesting a discovery ‘context’ of the objects or their association with the same time period. The mask is being presented to the public for the first time in this chapter (Fig. 7.4 and 7.5). The meaning of these face masks and their function accompanying notable dignitaries into the ‘Otherworld’ has been extensively discussed many times, especially with regard to the older Mycenaean and Egyptian golden masks. Lately they have been of special interest because of the numerous discoveries of various masks, especially funeral masks, in the Balkans (Fig. 7.1) (e.g. Behn 1955; Theodossiev 1998 and 2000; Guštin 2006). All in all, the making of life-sized human face masks from precious gold, and also bronze in some cases, was a frequent occurrence in the first millennium BC and has quite wide distribution and was a long term practice in continental Europe (Fig. 7.1).

The golden mask The mask is 14.7 cm high and 11.3 cm wide, made from carefully cut, beaten out and filed gold sheet metal. The face has round eyes with accentuated eyelids, a long, well-defined nose and tightly closed lips under which is a strongly defined chin. The arched eyebrows are marked out by numerous incised lines running from the middle to the sides of the face. The top of the mask is ornamented by small locks of hair. A full discussion of this golden mask from the Burgmuseum Deutschlandsberg (Fig. 7.2) was presented in honour of Barry Raftery a few years ago (Guštin 2009a). This chapter will simply review the period of its origin

7.  The human masks of unknown provenience

69

Fig. 7.1. Distribution of human masks in Europe. The circle shows the location of the Burgmuseum in Deutschlandsberg (after Guštin 2009a, produced by Andrej Preložnik).

and attempt to show the cultural environment in which the mask could have been in use, while briefly recognising the traditions of the use of face masks as funeral masks to symbolically depict the deceased, or in cult practice to symbolically depict heroes or gods. The holes indicate that the mask was affixed to something – perhaps to a coffin or shroud to depict the deceased, as was common in ancient Macedonia, or to some organic or metal surface for cult purposes, as is known from the western world. Nevertheless, in its artistic appearance, the elongated forms and the finely indicated brows and locks of hair, it differs considerably from the numerous Trebeništatype masks and accompanying decorative metal sheets. We see the mask from Burgmuseum Deutschlandsberg more as a distant echo of the workshops that produced the precious Šipka mask, the style of which can also be seen on the (much later) main human face on the Gundestrup cauldron. Perhaps this very mask is the last representative of their use in funeral and cult practice in the Balkan area (Guštin 2009a) (Fig. 7.2).

The bronze mask The bronze plate, 26.3 cm high and 21.5 cm wide, has a depiction of a human face in the style of a Greek comedy. It has a wide smile, slanted, almond shaped eyes, and a wide, well-designed nose in a broad face, with a twisted hairstyle and decorative volutes. The theatre person is wearing a typically Celtic twisted torc around its neck. It was published for the first time in the catalogue from the exhibition at Castello Gorizia and later became more widely known through archaeological discussion in connection with representations of human faces (Fig. 7.3) (Guštin 2006). This massive bronze plate is an important piece because it emulates perfectly the silver panels of the famous Gundestrup cauldron and leads us to confirm that such copper and silver sheet metal cauldrons with elaborate motifs on the rim were in wide circulation (Kaul and Martens 1995; Olmsted 2001, 58–60; see also Guštin 2006, 127, 128). If we can judge from the torc and the form of the plate, which is similar to that of the main human face

70

Mitja Guštin

Fig. 7.3. Burgmuseum Deutschlandsberg /Austria – the bronze plate. Fig. 7.2. Burgmuseum Deutschlandsberg /Austria – the golden mask.

depiction on the Gundestrup cauldron, it may indicate that this is a style characteristic for late Hellenistic art. After an examination of all the details, it seems that this bronze plate mask represents one of the latest products of this toreutic style and cauldron form. This supports the conclusions of those researchers who propose that workshops in the hinterland of the Black Sea were the centres where these prestigious metal toreutic products were made for ‘the barbarian Celts’ (Fig. 7.3).

The silver shield The shield is made from two pieces. The large piece is 39 cm high and 28.1 cm wide, made from carefully cut silver sheet metal beaten out to c.0.25 mm thin. It is decorated with two ridges with two lines of bosses between them, as well as some spiral ornaments on the upper and lower side. On the border of the plate there are numerous little holes, which were made to fix the plate to some other (possibly organic?) support. The second piece of the shield, made of another silver plate, is nailed to the centre of the large plate with quite strong rivets. The depiction of a human face takes up the

whole surface. The face has slanted eyes with accentuated eyelids, a well-defined nose and closed, sharply defined lips, above a strongly defined chin. The arched eyebrows and the moustache are marked out by numerous wide incisions running from the middle to the sides of the face. Both sides of the mask are framed by ears, and the top of the forehead is ornamented by the same locks of hair. The silver shield is described in the Museum’s collection as ‘ovaler Silberbeschlag eines Faustschildes mit Darstellung eines keltischen Schnurrbart. Der Schildrahmen bildet ein Dekor aus 8er-förmigen Schleifen und Buckeln. Späte Hallstatt- frühe Keltenzeit, c.500–400 v. Chr.’ (oval silver mounting of a little round shield with a representation of a Celtic man with a moustache. The framework of the shield has six spectacle loops in shape of figure eights and a double line of bosses. Late Hallstatt or Early Celtic period, c.500–400 BC). As support for this chronological determination of the shield are the artefacts displayed in the same showcase with the same period determination ‘Late Hallstatt or Early Celtic period, c.500–400 BC’, which include outstanding, twisted heavy silver torcs (‘Besonders schwerer Torques – Halsreif – aus Silber mit Drahtwicklung und vierfachen Spiralenden’) and a bronze situla (container for mixing wine) with a ladle

7.  The human masks of unknown provenience

71

Fig. 7.4. Burgmuseum Deutschlandsberg /Austria – the silver shield, front side.

Fig. 7.5. Burgmuseum Deutschlandsberg /Austria – the silver shield, back side.

and sieve (‘Situla – eimerförmiges Weinmischgefäß – aus Bronze und Siebschöpfer. Die Griffe des Situla und des Schöpfers sind tordiert gestaltet’), both with twisted handles. The group of objects in this showcase suggest to the visitor their common source from the antiquities market and also as a find/grave context (Fig. 7.4 and 7.5). The shield as an object and its decoration have had no known comparison until now. Some of the details are quite well-known in Celtic art, e.g. the convex ridges on the silver circle plate from Manerbio sul Mella (I Celti 1991, 466) or on the plate from Brno-Hořovičky (I Celti 1991, 141; Die Welt 2012, fig. 320) that in both cases follow the circle form of the plates. The face on the shield is decorated with typical ‘Celtic’ slanted eyes, arched eyebrows and a defined moustache. Earlier depictions of Celtic moustaches include the oinochoe of Waldalgesheim (I Celti 1991, 676), Kleinaspergle Hohenasperg (I Celti 1991, 179), and Ferschweiler (I Celti 1991, 162) as well as on Celtic wheelpins from Unterradlberg bei Sankt Pölten (I Celti 1991, 189) and Stradonice (I Celti 1991, 739, 767 (499)). These early examples, as well as the later ones like that on the Gundestrup caldroun (I Celti 1991, 539; also Hachmann 1990, Beilage 11: 1, 2) and the famous ‘Dying Celt’ from

Pergamon (I Celti 70, 71, 334), ‘Wounded Celt’ from Museo Nazinale Napoli (I Celti 1991, 66) and ‘Bearded Celt’ on knees (I Celti 1991, 68) confirm the tradition and identity of this type of moustache depiction in the Celtic world. The silver itself came into regular use in Celtic Europe when the Celts came into contact with silver Greek coins or with the silver material from mines in the hinterland of the Black Sea. At the end of the 4th and the beginning of the 3rd century BC, numerous silver bracelets and torcs spread from central Celtic Europe, which speaks to the popularity of silver jewellery with Celtic tribes in their core territory in the La Tène B2 phase (Guštin 2009b, 484). We can recognise the value and follow the expansion of silver products in 2nd and 1st century BC, not only in the traditional Celtic world, but also in the region of the Daci in the east and in the area of the Celtiberi towards the Atlantic coast (Guštin 2009b) with numerous phenomena of feasting equipment, like cauldrons and others, and robust forms of jewellery, including special torcs, which are the best representatives of prestige metal deposits (Fischer 1978; Furger-Gunti 1982; Eluère 1987), and heavy fibulae like those of the Jarak type from the Balkan peninsula, big belt plates (Guštin 2011), and even one of the helmets of the Novo mesto type (Otchir-Goriaeva 2002; Mihaljević and Dizdar 2007).

72

Mitja Guštin

When one considers the irregularities of the silver shield from the Burgmuseum Deutschlandsberg (unknown location and find circumstances, an unknown Iron Age form of the object without any known parallels) the idea that the shield may be a Celtic / La Tène object is hard to defend. We leave it open for further research because the presence of the characteristic ‘Celtic’ slanted eyes and ‘Celtic’ moustache may demonstrate some retention of the Celtic style in later epochs. Copyedited by Adrienne C. Frie.

Bibliography Behn, F. 1955. Vorgeschichtliches Maskenbrauchtum. Berlin: Akademie-Verlag. Die Welt der Kelten: Zentren der Macht – Kostbarkeiten der Kunst. 2012. Archaeologisches Landesmuseum Baden-Wuerttemberg, Konstanz: Jan Thorbecke Verlag. Eluère, C. 1987. Das Gold der Kelten. München: Hirmer. Fischer, F. 1978. Archaologie und Geschichte konkret. In Der Trichtinger Ring und seine Probleme, 9–37. Heidenheim an der Brenz: Heimat- und Altertumsverein. Furger-Gunti, A. 1982. Der Goldfundvon Saint-Louis bei Basel und ähnliche keltische Schatzfunde, Zeitschrift für Schweizerische Archäologie und Kunstgeschichte 39, 1–47. Gli echi della terra, 2002. Presenze celtiche in Friuli: dati materiali e momenti dell’immaginario. Catalogo. Pisa: Giardini. Guštin, M. 2006. Zu einigen Figuralmotiven im Gebiet der Taurisker. In W.-R. Teegen (ed.) Studien zur Lebenswelt der Eisenzeit: Festschrift für Rosemarie Müller, Ergänzungsbände zum Reallexikon der germanischen Altertumskunde, Bd. 53, 115–31. Berlin: De Gruyter. Guštin, M. 2009a. The golden mask from the unknown. In G. Cooney, K. Becker, J. Coles, M. Ryan, S. Sievers (eds.) Relics

of old decency: archaeological studies in later prehistory. Festschrift for Barry Raftery, 137–41. Dublin: Wordwell. Guštin, M. 2009b. Der Torques: geflochtener Drahtschmuck der Kelten und ihrer Nachbarn. In S. Grunwald, (ed.) ArteFact: Festschrift für Sabine Rieckhoff zum 65. Geburtstag, Universitätsforschungen zur prähistorischen Archäologie, Bd. 172, 477–86. Bonn: Habelt. Guštin, M. 2011. On the Celtic tribe of Taurisci: local identity and regional contacts in the ancient world. In M. Guštin and M. Jevtić (eds.) The Eastern Celts: the communities between the Alps and the Black Sea, Annales Mediterranei, 119–128. Koper: Založba Annales. Hachmann, R. 1991. Gundestrup-Studien. Untersuchungen zu den spätkeltischen Grundlagen der frühgermanischen Kunst. In: Bericht der Römisch-Germanischen Kommission. Bd. 71, 568–903. I Celti 1991. Direzione scientifica Sabatino Moscati, Comitato Amici di Palazzo Grassi, Catalogo, direttore editoriale Mario Andreose. Milano: Bompiani. Kaul, F. and Martens, J. 1995. Southeast European influences in the Early Iron Age of Southern Scandinavia. Gundestrup and the Cimbri, Acta Archaeologica 66, 111–61. Kelten 1998. Die Kelten im südweststeierschen Teil des Königreiches Norikum. Burgmuseum Deutschlandsberg. Mihaljević M. and Dizdar, M. 2007. Late La Tène bronze helmet, Vjesnik Arh. muz. Zagreb XL, 117–146. Olmsted, G. 2001. Celtic Art in transition during the first century BC. Budapest: Archaeolingua. Otchir-Goriaeva, M. 2002. Das sarmatische Grab von Jaškul, Kalmykien, Eurasia Antiqua 8, 353–87. Theodossiev, N. 1998. The dead with golden faces: Dasaretian, Pelagonian, Mygdonian and Boeotian funeral masks, Oxford Journal of Archaeology 17 (3), 345–67. Theodossiev, N. 2000. The dead with golden faces II: Other evidences and connections, Oxford Journal of Archaeology 19 (2), 175–209.

8 HEADS, MASKS AND SHIFTING IDENTITIES: A NOTE ABOUT SOME DANUBIAN KANTHAROI WITH ANTHROPOMORPHIC DECORATION Mariana Egri

Introduction Human heads and face-masks occupy a significant place in Celtic imagery, so their presence in a variety of contexts has attracted the interest of many archaeologists and historians seeking to understand their practical and symbolic meanings. The symbolic status of the human head in particular was mostly explored in the wider context of the appearance, use and manipulation of anthropomorphic representations in Celtic visual language (Megaw 1970; Megaw and Megaw 1998; 2001; Kruta 1992; Szabó 1965; 1992; Green 1997, 114–17; Aldhouse-Green 2004; Harding 2007 etc.), although some recent studies are also bringing into discussion the diverse roles played by this engaging but flexible motif in the life of different communities belonging to the ‘Celtic world’ (Arcelin and Rapin 2003; Rustoiu and Egri 2011, 75–99; Armit 2010). Vincent Megaw is one of the most successful explorers of this complex and fascinating visual universe populated by fantastic beasts, exuberant vegetation, fearless warriors and many other fluid, elusive creatures. A never-ceasing interest in the Celtic world of the Carpathian Basin first brought him in contact with the so-called Danubian kantharoi and their puzzling decorative repertoire many years ago. Some of these interesting artefacts found their place in his writings. It is therefore most appropriate to re-examine the theme, or at least a part of it, as a form of appreciation for his work and for his continuous interest in the eastern Celtic world. Thus the present article discusses a group of Danubian two-handled vessels decorated with human heads or masks placed on the rim, on the insertion point of the handles, and oriented towards the interior. The practical and symbolic functions of these vessels within the Celtic communities of the Carpathian Basin (Fig. 8.1) can be only deciphered by taking into consideration each of the briefly-listed morphological details, as well as their contexts of discovery.

The Danubian kantharoi with anthropomorphic decoration The Danubian kantharoi (Fig. 8.2) are characterised by a deep, carinated body, a more-or-less marked base or foot, and two vertical raised handles, symmetrically attached to the upper half of the body. They represent a particular ceramic form invented by potters from the Carpathian Basin, who combined Hellenistic, Celtic and earlier indigenous morphological and decorative elements. The earliest examples appeared in the LT B2a sub-phase, more likely due to the influence of the Hellenistic kantharoi, which were sometimes imitated by local potters. Three main variants of Danubian kantharoi were popular until the end of the LT C1 phase: only a few similar pieces are dated later. Their morphological evolution and particular distribution in the aforementioned region are closely related to a combination of key factors – the prior existence of a widely used indigenous two-handled beaker, later taken over by certain Celtic groups, and the contacts established by the Late Iron Age communities from the Carpathian Basin with the eastern Mediterranean world and the Balkans (Rustoiu and Egri 2011; see also Kruta and Szabó 1982 and Dizdar 2010). As already mentioned, three main variants of twohandled beakers have been identified in the Carpathian Basin (Rustoiu and Egri 2011, 20–52, fig. 4). The first one (Fig. 8.2.1) imitated, more or less closely, Hellenistic prototypes, whereas the remaining two variants (Fig. 8.2.2–3) were created by transforming pre-existing local ceramic forms (carinated bowls and large bi-truncated or ovoid jars), to fulfil the increasing demand of the local consumers for two-handled drinking vessels. This particular interest was determined not only by the increasing popularity of Hellenistic kantharoi, imported or locally made, but also by the earlier use of a traditional two-handled drinking vessel

74

Mariana Egri

Fig. 8.1. Map of the Carpathian Basin with the sites mentioned in text. 1. Balatonederics; 2. Belgrade-Karaburma; 3. Blandiana; 4. Deta; 5. Körösszegapáti; 6. Novo mesto; 7. Tiszafüred-Morotvapart; 8. Zalakomár.

belonging to the ceramic repertoire of the pre-Celtic local substratum, the so-called Illyrian-Pannonian kantharos. Both the potters and the consumers of the Carpathian Basin resorted to a variety of mechanisms through which the Mediterranean kantharoi were incorporated into the local practices. For example some vessels were decorated with incised or stamped geometric details, sometimes quite elaborated, belonging to the common repertoire of Celtic motifs. Others have the handles decorated with anthropomorphic or zoomorphic details of various origin and probably having different meanings. The anthropomorphic details – complete or nearly

complete human figurines, human heads and masks – appear mostly on kantharoi deriving from local ceramic forms (Rustoiu and Egri 2011, 52–3). The single example of an imitation of a Hellenistic kantharos probably decorated in this manner comes from grave no. 34 from BelgradeKaraburma, in Serbia (Todorović 1972, 20, pl. XIV/1, LII/1; Rustoiu and Egri 2011, 24, note 5). Similar decorations also appear on some single-handled beakers discovered in sites from the Great Hungarian Plain, as well as on a pseudokernos discovered in a dwelling at Tiszafüred-Morotvapart (Hungary), dated in the LT C2 (Kriveczky 1991). The latter vessels have to be included in the discussion concerning the

8.  Heads, masks and shifting identities

75

Fig. 8.2. Types of Danubian kantharoi: 1. imitations of Hellenistic forms; 2. kantharoi deriving from local bowls; 3. kantharoi deriving from local tall vessels (after Rustoiu and Egri 2011). Different scales.

76

Mariana Egri

presence of anthropomorphic decoration on two-handled beakers from the Carpathian Basin due to their relevant morphological and contextual similarities. Two morphological series (Fig. 8.3) have recently been identified within the repertoire of anthropomorphic elements decorating the handles of certain Danubian drinking-related vessels (Rustoiu and Egri 2011, 53–62, fig. 24). The first series consists of complete or nearly complete human figurines, which decorate or replace the handles and are outward-oriented. Their features were inspired by the kouros-shaped handles of some bronze vessels from the Mediterranean area, dated to the 6th to 5th centuries BC (Tarditi 1996, 155–8, 172–9, 189–190; Stibbe 2000, 22–6; Graells i Fabregat 2008, 201–4; Popović et al. 1969, 73–4, no. 42 etc.). Both masculine and feminine characters were represented in a more or less realistic manner, naked or costumed and displaying a variety of bodily gestures and ornaments. The second series of anthropomorphic decorations consists of human heads or masks placed on the rim, on the insertion point of the handles, and oriented towards the interior of the vessels. This type of decoration seems to have been inspired by the Greek bronze vessels (oenochoai) having a feminine or a Silenus head on their handle, which were also present in the northern Balkans and the northern Pontic region during the second half of the 4th century and at the beginning of the 3rd century BC (Teleagă 2008, 258; Boltryk, Treister and Fialko 2009). Within both series of anthropomorphic decorations the original Mediterranean motifs were taken over and transformed, interpreted and stylised by different local potters to comply with the local iconography and visual language, having particular meanings related to the ways in which these vessels were used.

Heads and masks on the Danubian kantharoi As mentioned above, the second series of anthropomorphic details decorating the Danubian kantharoi and a few other drinking-related vessels from the Carpathian Basin consists of human heads or masks placed on the rim, at the insertion point of the handle, and oriented inward. It is sometimes difficult to say whether the image is of a human face or of a mask, especially in the case of the more stylised ones. However, two examples convincingly depict masculine heads modelled in a quite realistic manner. One such head, having the typical short haircut and a thin moustache (Fig. 8.4.1), decorates the upper part of the handle of a fragmentary kantharos discovered at Balatonederics (Hungary) (Horváth 1987, 66–8, fig. 2–3). The face has a gloomy expression, with a downward-turning mouth and oval bulging eyes; the ears are anatomically modelled and the moustache and eyebrows are incised.

The hair is neatly back-combed in thick locks. The missing opposite handle was most likely decorated with another human head, perhaps a feminine one, given that the great majority of anthropomorphic decorations on similar vessels consist of a man / woman pair. The second example of a human head (Fig. 8.4.4) is also fragmentarily preserved and comes from Deta, in western Romania (Márton 1933, 79, fig. 20; Rustoiu 2012, 57–8, pl. II/1). The fragment belongs either to a kantharos, or more likely to a single-handled beaker, due to the morphology of the piece and the preserved traces of attachment to the handle (Rustoiu and Egri 2011, 60, fig. 23.4). The bald head has quite realistic features, being carefully modelled, with round sunken eyes having the large iris clearly marked, a prominent nose, a downward-turning mouth and a long, angular chin. Similar to the head from Balatonederics, it has a solemn, almost gloomy facial expression. A few more vessels have the handles decorated with opposing face-masks, more or less stylised. Perhaps the most intriguing example is the kantharos (Fig. 8.6) from Blandiana (Transylvania) (Aldea and Ciugudean 1985, 37, fig. 1.1, 2.1, 3.1–2, 4.1–2; Rustoiu and Egri 2011, 57, fig. 20–21.1), which has the handles modelled as a masculine and a feminine figurine respectively, whereas on the rim, towards the interior, each character is doubled by a human mask. The two outward-looking characters are almost identically modelled, displaying the characteristic backcombed hair, large round eyes having marked eyebrows in relief, downward-turning mouths and flagging ears. The masculine character also has a moustache consisting of fine incisions. The inward-oriented masks doubling both characters replicate some of these quite geometrical facial features and preserve the gloomy expression. The identification of the masks on this vessel is also supported by the presence of a thin incision above the forehead of both outward-looking characters. Another quite simply modelled face-mask decorates the handle of a beaker recovered from the cemetery at Körösszegapáti, in Hungary (Nepper 1976, 24, fig. 22–23). In this case the facial features (Fig. 8.4.3) are flattened and geometrical, being reduced to a few relevant details, so the eyes are represented as two incised dots, the nozzle is nearly rectangular and the mouth is an incised line. The hair seems to be again back-combed in a specific manner. The little oval face is surrounded by a series of finely incised lines, suggesting a beard, a sort of stripe or a veil, perhaps imitating the ‘string’ surrounding the human masks depicted on a series of early pieces, for example on the linchpin from St. Pölten or on the brooch from Slovenské Pravno (Megaw, Megaw and Neugebauer 1989, 506–7, fig. 14, 15.1 and 15.3). A pair of masks was attached not to the handles but to the body of a richly decorated kantharos (Fig. 8.4.5) from grave 40 at Novo mesto, in Slovenia: they were considered

Fig. 8.3. Two morphological series of the anthropomorphic decoration (after Rustoiu and Egri 2011). Different scales.

8.  Heads, masks and shifting identities 77

78

Mariana Egri

Fig. 8.4. Vessels with anthropomorphic decoration (second series). 1. Balatonederics; 2. Zalakomár; 3. Körösszegapáti; 4. Deta; 5. Novo mesto (after Rustoiu and Egri 2011). Different scales.

8.  Heads, masks and shifting identities

79

Fig. 8.5. Pseudo-kernos and agricultural tools from Tiszafüred-Morotvapart (after Kull 1997). Different scales.

to have an apotropaic role (Knez and Szabó 1980–81, 82–3, fig. 1–6; Megaw and Megaw 2001, 150–1, fig. 235–237). These masks have the same gloomy facial expression as the human heads from Balatonederics and Deta, with an angular chin and a downward-turning mouth, but the large round, bulging eyes and the rather simplified features are close to those of the masks decorating the kantharos from Blandiana. Lastly, two other vessels were decorated with muchstylised versions of a human head or mask. The first one is the large kantharos from the settlement at Zalakomár (Fig. 8.4.2), in Hungary (wrongly reconstructed in the original publication – Horváth 2008, 113–14, fig. 2–7). In this case two circles representing the eyes were incised on one handle, separated by a longitudinal line, while on the opposite handle the circles were separated by two lines, the intention being probably to suggest the presence of two different characters, perhaps a masculine and a feminine one, according to the iconographic scheme governing the first series of anthropomorphic decorations mentioned above. The second vessel decorated with stylised face-masks looking inward is the previously mentioned pseudo-kernos from Tiszafüred-Morotvapart, in Hungary (Fig. 8.5). In this case three human masks have each been placed on the insertion points of three little vessels attached to the rim of the main container (Szabó 1998, 67–8, fig. 66; 2003, 31, fig. 7–8). The vessel was discovered in a hut together

with agricultural tools (three coulters and three sickles), a fragmentary mill stone and a spindle whorl. It has been recently suggested that the entire assemblage might have been related to an agricultural or fertility cult, either as the northernmost evidence of the cult of Demeter, or perhaps in connection with some Central or Western European Celtic goddesses of fertility and maternity (Kull 1997, 358–9, fig. 76.1–2).

Shifting identities Realistic, caricatured or stylised human heads and facemasks adorned a wide variety of metal, stone or ceramic objects created by the artisans working in the Celtic environment. Their distinct stylistic features are frequently used to define one or another of the evolving phases of Celtic art, alongside other decorative patterns. There is a longlasting debate about the identity and meaning of the human characters depicted by Celtic artisans – deities, mortals or anthropomorphised symbols. A straightforward, universallyfitting answer cannot be easily provided since their practical and symbolic meanings and functions, as well as their place in the visual language of different communities, might have varied through time and from one region to another. Furthermore, these can be only reconstructed by interpreting

80

Mariana Egri

the use of anthropomorphic images in particular contexts (Ginoux 2003; Aldhouse-Green 2004, 2–7). From an anthropological point of view, it has to be noted that the option to depict the human being and the manner in which this is done are closely related to the way in which different individuals and communities chose to materialise their concepts and ideologies about the nature of personhood and its place in social dynamics. The process of representing these otherwise abstract notions always involves a commonly-acknowledged range of familiar visual tools. Thus, the realistic representations are more likely related to the perception of the self and of its corporeal projection as a single, bounded entity, which needs to be depicted as such in order to underline and preserve its individuality. A more stylised representation, involving the use of symbols and suggestions and the incorporation of the non-human and surreal, is perhaps related to the perception of personhood as a dividual entity, composed of multiple features, some transcending spiritual boundaries. Several of these composing features are seen as the result of different social interactions or of the impact of the real or imaginary environment, and the physical body is considered to be only a (sometimes disposable) part of this complex symbolic structure (Douglas 2004, 72–91; Fowler 2004, 3–5; Robb 2007; Egri 2012, 507–9). In an essay concerning the presence of human heads in early Celtic art, Ruth and Vincent Megaw (1998, 286) stressed that there is a significant difference between the representational manner in which the human image is depicted in Classical art and the abstract, symbolic and often fluid manner favoured in Celtic art. This particular preference is reminiscent of the manner in which the divine power and its actions were visualised during the Greek Dark Age through the use of non-human symbols, which were both familiar (to be easily read and acknowledged) and strange (to underline their ability to reach other spaces). In this case the most important element was the carefullycoded scenario based on the game of hiding and revealing, and on the personal relationship established between the divine power and the master of the sacred (Vernant 1995, 406–11). Sometimes the latter also employed masks in rituals, including human masks, in order to impersonate, or to proclaim possession, by a powerful force (Patton 2007, 361–2). In the Celtic environment, easily recognisable, albeit often cartoon-like, human images were favoured especially at the very beginning of the early LT period, very probably under the influence of certain Mediterranean objects, whereas the more realistic depictions mainly belonged to the late LT repertoire (Megaw and Megaw 2001, 69–74 and 164–73; 1998; Harding 2007, 51–7 and 221–4). Still, similar quite realistic modelled human figures appeared earlier in the Carpathian Basin, during the middle LaTène period (Rustoiu and Egri 2011, 52–67). On the other hand,

the so-called Plastic Style was characterised by highly stylised, elusive and multi-dimensional anthropomorphic depictions, in which only certain features were highlighted and intertwined with other relevant visual symbols (Megaw and Megaw 2001, 135–43; Harding 2007, 124–8). V. Kruta has considered that the abstracting process denotes a particular way of understanding the universe of the immortal gods, more precisely their ability to transform themselves and to appear in front of mortals in a variety of forms. This original evolution of Celtic imagery, based on elusive and undefined forms, peaked at the beginning of the 3rd century BC. This is the age of Brennos, the leader of the Celts who attacked the sanctuary of Delphi and who, according to Diodorus (XXII. 9), was surprised and amused by the mimetic, anthropomorphised manner in which the gods were shown in the Greek temples (Kruta 1992, 834, 841). Most of the current interpretations of the particular preference for the human head in Celtic art underline the symbolic potency of this motif, as it results from the analysis of ancient and medieval written sources. According to Classical authors and later Irish and Welsh vernacular literature, the Celts believed that the head was the receptacle of the soul and it retained its spiritual power even after being separated from the body. The severed head of certain heroes or of other divine-like characters were considered to have magical powers, being able to protect and actively support individuals and communities in their actions. The practice of head-hunting was also commonly related to these beliefs, whereas the very visible display of the severed heads of enemies was interpreted as part of a more complex process of expressing and reiterating military prowess. Some archaeological discoveries, including many stone representations of the human head, are usually interpreted as confirmations of the written sources (Cunliffe 1997, 209–10; Green 1997, 114–17, s.v. head and headhunting; Brunaux 2000, 202–4). Still, there are significant discrepancies concerning the chronological and geographic distribution of these categories of written and archaeological evidences, pointing to a variety of functions and meanings of the human head and of the associated practices, which changed through time and from one region to another (Armit 2010, 90–3, fig. 9.1). Returning to the Danubian vessels decorated with human heads and face-masks oriented inward, a series of key morphological and decorative elements, pointing to the individualised character of these particular objects, may contribute to the identification of their practical and symbolic functions. One important element is the frequency of kantharoi with anthropomorphic decoration, and it has to be noted that the total number of such finds is reduced in comparison with that of the undecorated similar vessels from the Carpathian Basin (less that 25% of the total number of finds). They are also quite sparsely distributed, so none of the analysed cemeteries and settlements produced more than

8.  Heads, masks and shifting identities two examples. Another important element concerns their dating – with a single exception (the pseudo-kernos from Tiszafüred-Morotvapart), all these vessels are dated to the LT C1 period (Rustoiu and Egri 2011, Annex 5). The great majority of the Danubian kantharoi come from funerary contexts (Rustoiu and Egri 2011, Annexes 1, 3–5), although this pattern might be the result of a slightly biased interest in funerary archaeology in the Carpathian Basin. However, their presence in completely investigated settlements from the Carpathian Basin is relatively scarce (Rustoiu and Egri 2011, 71–3). Amongst the vessels decorated with human heads and face-masks oriented inward, the fragment from Deta has an unknown context of discovery, whereas the kantharos from Zalakomár was recovered from a settlement and the pseudo-kernos came from a cult-related context. The remaining pieces come from funerary contexts, with or without weaponry in their inventories, which may suggest that the use of such vessels was not restricted to the dominant social category of the warriors, or to the men. Furthermore, with a single exception – the head from Deta – these vessels are complete or nearly complete. The kantharos from Zalakomár was slightly damaged in ancient times, but was carefully mended with a tar-like substance to remain in use as long as possible, despite the nearby presence of a ceramic workshop which could have easily replaced it (Horváth 2008, 113). This careful preservation of a vessel posing no technological challenges which might have deemed it irreplaceable, but having a particular, less common decoration, indicates that it must have been very important for the owner, both in life and in death, since it accompanied the deceased into the grave. Returning to the stylistic features of these decorative details, it has to be noted that no two heads or face-masks are identical, displaying a remarkable morphological diversity. This might partially be the result of the variable artistic ability of the craftsmen, albeit certain regional specificities of the visual language might have also played an important role, for example certain features had to be emphasized in a manner which was relevant and appropriate for one community (or for just one category of consumers) or another. Still, some of the heads and face-masks share a number of important visual characteristics, which might have had a symbolic relevance for the users or/and the beholders across a wider area. One of these characteristics is the depiction of the back-combed hair, a feature which belongs to a long-lasting iconographic tradition in temperate Europe. Similar human images appeared during the early La Tène, for example the masculine figures on the handle of a Schnabelkanne from Dürrenberg bei Hallein (Megaw 1970, No. 72; Megaw and Megaw 2001, 36, pl. V–VI; Moosleitner 1991, 171), those on the brooches from Schwieberdingen, Litér and Manětin-Hrádek (Megaw 1970, no. 31, 89; Binding 1993, fig. 17.4, pl. 10/2, 8/7), or on a

81

bronze openwork fittings (4th century BC) from Čižkovice in Bohemia (Kruta 1991, 205) etc. During the middle La Tène the human heads similarly modelled are present in the Carpathian Basin on the hilts of some short swords or daggers (probably used as sacrificial instruments, see the synthesis of opinions in Megaw 2002, 411), for example the piece from Dinnyés in Hungary (Szabó and Petres 1992, 57, 87, pl. 12; in general for the type and iconography see Megaw 2002). It has been already noted that the haircut, alongside other natural or carefully modified elements of the human figure, are visual agents of social communication having a powerful significance in particular social contexts (Arnold 2008, 375–9; Aldhouse-Green 2004, 300). Both archaeological and written sources seem to suggest that certain individuals from the Celtic environment had a particular style of haircut, probably as a differentiating visual marker of some social groups (Venclová 2002, 460–70). Another iconographic category identified within the aforementioned heads and face-masks consists of human images with quite elaborated facial details, but having bald or shaved heads. Their general morphology suggests that the sources of artistic inspiration should be sought in the eastern Mediterranean area, and the gold Janus-like pendant from the Museum of Schumen, in eastern Bulgaria, may offer a good stylistic analogy (Tonkova 2006, 266–7, pl. I/5; Rustoiu 2008, 59–60, fig. 24.4–5). A distinct group of inward-looking anthropomorphic details consists of face-masks. Genuine life-size Celtic masks are rare, but smaller versions do appear, mostly as decorative elements on different objects. It is considered that masks, perhaps made of wood or other perishable materials (textiles or leather), might have been used within different religious ceremonies (Green 1997, 144–5, s.v. mask). A small bronze statuette representing a nude masculine character, found at Ponická Huta in Slovakia, seems to wear a mask (Pieta 2008, pl. 41), whereas the two aforementioned characters decorating the handles of the kantharos from Blandiana also have masks attached to their heads. Other masks were probably fitted on various structures or objects (pillars, doors, walls etc.) or perhaps even over the face of some wooden statues. It has been recently suggested that some of the more stylised human heads carved in stone on certain architectural elements from the southern Gaulish sanctuaries are masks-reliquaries, serving as a visual mnemonic for the spirits of the ancestors and as a prized material reminder of a worthy lineage (Arcelin and Rapin 2003, 191–2; Armit 2010, 95; see also Aldhouse-Green 2004, 169). However, in many societies the masks are not simple images of dead people, but the material expression of their spirit, or of the divine power, which is only able to transcend boundaries and to act in the world of the living if it gets shape through personification. Sometimes the mask even

82

Mariana Egri

has its own spirit, which can be controlled only by initiated individuals. Some of these masks have exaggerated facial features – oversized eyes, nose, ears and mouth – and quite often a gloomy or even a menacing expression. In other situations the facial features are simplified and again, only the most important details – the eyes, nose and mouth – are carefully represented. In both cases the emphasizing of certain features, which provide expressiveness and facilitate visual contact and communication in a way which is familiar to all human beings, are a reminder that the mask is a means of action, and not an inert, mimetic effigy of a deceased individual. These spirits of the otherworld have to become visible from time to time, and to be made visible through the mediation of the masters of the sacred, but they cannot be physically identical to mortals, even if they once lived in a mortal body. As in the case of the idols of the Greek Dark Age mentioned above, they have both familiar and strange features, so they can be recognised but have to be also acknowledged as fundamentally different, benevolent or ill-tempered, or even dangerous if they were improperly contacted. Such concepts related to the visualization of the divine power may explain, for example, the large bulging eyes, dominating the face, and the gloomy expression accentuated by the downward-turning mouth and the angular chin, which have been also noted on many Celtic human images, including on those decorating the Danubian kantharoi. It has been noted already that the severed human heads or face-masks decorating the Celtic sanctuary from Roquepertuse might have been active visual mediators between the physical and the supernatural worlds (Armit 2010, 95). Since the masks allow the manipulation of appearances, the impersonation and the transgression of invisible spiritual boundaries by hiding, deceiving and transformation, they were commonly used in a variety of rituals which were meant to facilitate the connection with, or the safe passage to, the otherworld (Napier 1986, 15–17; Aldhouse-Green 2004, 150–69). The mask-wearers were able to switch identities and temporarily transform their persona by changing their appearance, and to enter other spaces which are normally out of reach for ordinary individuals. These shape-shifters (Aldhouse-Green 2004, 164) might have also used other complementary means of transgressing symbolic boundaries, for example certain substances having psychoactive properties, including alcoholic beverages. Returning to the functionality of the Danubian vessels decorated with human heads and face-masks oriented inward, it has to be noted that all of them are related to drinking or drink-sharing, very probably alcoholic beverages (beer, mead or wine). The ritualised consumption of alcohol plays an important role in the convivial practices of various communities, including the Celtic ones, due to

its psychoactive properties. These carefully coded social practices facilitate the negotiation, affirmation, reiteration or transformation of personal or group identities and status, but they may also contribute to the creation and maintenance of communal solidarities (Dietler 2006, 235–7; for the symbolic meanings of beer consumption in temperate Europe also see Hell 1991). In certain societies, alcoholic intoxication was also seen as a means of approaching the gods, the ancestors or the spirits of the otherworld. The Greek authors of Archaic and Classical times clearly differentiated the civilized consumption of wine epitomised by the symposium and the excessive drinking manners and preference for uncivilized beverages of different barbarian populations (for a summary of these contrasting consumption styles see Poux and Feugère 2002, 202–11; Nelson 2005, 38–44). Still, they also present wine as a substance which may produce an alteration of the mind and body, allowing the transgression of the normal social order. For example, some Homeric poems underline that only wine, excessively consumed in a pure state, can produce the ατη, a psychical stage which is normally attained only through divine intervention (Andò 1991, 167–9). Other ancient texts also mention the intoxicating properties of other alcoholic beverages, including the ‘inferior’ beer which may also cause mania (Nelson 2005, 34–7). Thus the psychotic properties of different alcoholic beverages might have also been taken into consideration for rituals requiring the temporary transformation of the self, alongside the use of masks, albeit the two means appealed to different human senses. In the Greek world masking and drinking were commonly associated with the orgiastic practices of the Dionysian cult, whereas the shamanic practices of many populations across the world also required this combination of shape-shifting means. As a consequence, it might be presumed that the presence of human heads and face-masks on certain drinking-related vessels from the Carpathian Basin is related to particular convivial practices of transformative nature. These images are oriented towards the interior of the mentioned vessels and seem to look through a symbolic gate, which can be crossed only by consuming the alcoholic beverage (wine is the mirror of the heart, according to Aeschylus, Frg. 384), a substance facilitating the temporary transformation of the self, the transgression of dangerous boundaries and the communication with the otherworld. While discussing the phialae and cups from Thrace decorated inside, on their tondo, with human busts or heads, Marazov (2000, 264–5) has noted that the local kings might have ceremonially drunk the wine to discover these images, which connected them with the ancestors or the divinity. The beakers from the Carpathian Basin bear human images on their handles, and perhaps they were reflected by the alcoholic beverage, upside-down and slightly distorted, thus producing a striking effect, especially after the consumption of a significant

8.  Heads, masks and shifting identities

83

allowed the sharing of a beverage by three consumers, or perhaps the mixing of three different ingredients into the main container. The vessel was accompanied by iron agricultural tools (three coulters and three sickles), the entire assemblage being related to an agrarian or fertility cult. The presence of the three little stylised masks on the rim may also support this interpretation since some aspects of the cult of the ancestors are related to the idea of rebirth and fertility (Armit 2010, 93–4). It might be therefore concluded that the aforementioned vessels were part of the ritual practices, which required the alteration of the self and the shifting of identities in order to transgress symbolic boundaries and to allow contact with the otherworld. They have a highly individualised character and were part of the symbolic being of the owner, which may explain their presence only in certain burials and the scarcity of settlement discoveries. The manipulation of such vessels was probably restricted to some initiated individuals (men or women), masters of the sacred who were able to correctly understand the message encrypted in the specific decoration, to perform the rituals, and to interpret the results for the community. Very probably the vessels could not have been re-used by others and their rare presence in settlements does not contradict this hypothesis. The latter vessels could have been seriously deteriorated during the ceremonies (or due to other more or less violent actions), making them useless, thus unsuitable even for funerary purposes. From the purely stylistic point of view, the decorative elements on the vessels are quite diverse, but all belong to a well structured visual language corresponding to a particular ideological code which was purposefully created by different communities. In some situations they might suggest the existence of similar practices and beliefs across larger areas of the Carpathian Basin. Nevertheless the finer symbolic meanings of these decorative details, as well as the circumstances in which such vessels were used, might have been slightly different from a community to another.

Fig. 8.6. The kantharos from Blandiana (photo A. Rustoiu).

quantity. A pair of mask-wearers is present on the beautiful kantharos from Blandiana (Fig. 8.6), a man and a woman illustrating two moments from the identity-shifting process – their daily, mundane personae gazing to the world of the living while about to perform the ritual, and their masked, transformed self, ready to enter another world through the symbolic inner gate. The unusually large dimensions of this vessel may suggest that it was perhaps used as a communal container, the beverage being shared by ritual performers. The shape of the pseudo-kernos from TiszafüredMorotvapart points to a particular use of this vessel, which

Bibliography Aeschylus. 2, Agamemnon, Libation bearers, Eumenides, Fragments (tr. H. W. Smyth and H. Lloyd-Jones), Loeb Classical Library 146, London, 1957, W. Heinemann. Aldea, I. A. and Ciugudean, H. 1985. Noi descoperiri celtice de la Blandiana, Apulum 22, 37–43. Aldhouse-Green, M. 2004. An archaeology of images. Iconology and cosmology in Iron Age and Roman Empire. London: Routledge. Andò, V. 1991. Vin et mania. In D. Fournier and S. D’Onofrio (eds.) Le ferment divin, 167–180. Paris: Éditions de la Maison des sciences de l’homme. Arcelin, P. and Rapin, A. 2003. Considérations nouvelles sur l’iconographie anthropomorphe de l’âge du Fer en Gaule méditerranéenne. In O. Buchsenschutz, A. Bulard, M.-B.

84

Mariana Egri

Chardenoux and N. Ginoux (eds.) Décors, images et signes de l’âge du Fer européen, Revue Archéologique du Centre de la France Supp. 24, 183–219. Tours: FERACF. Armit, I. 2010. Porticos, pillars and severed heads: the display and curation of human remains in the southern French Iron Age. In K. Rebay-Salisbury, M. L. Stig Sørensen and J. Hughes (eds.) Body Parts and Bodies Whole, 90–100. Oxford: Oxbow Books. Arnold, B. 2008. ‘Reading the Body’: Geschlechterdifferenz im Totenritual der frühen Eisenzeit. In C. Kümmel, B. Schweizer and U. Veit (eds.) Körperinszenierung – Objekt­ sammlung – Monumentalisierung: Totenritual und Grabkult in frühen Gesellschaften. Archäologische Quellen in kulturwissenschaftlicher Perspektive, Tübinger Archäologische Taschenbücher Bd. 6, 375–395. Münster: Waxmann Verlag. Binding, U. 1993. Studien zu den figürlichen Fibeln der Frühlatènezeit. Bonn: Rudolf Habelt. Boltryk, Y. V., Treister, M. Y. and Fialko, O. Y. 2009. Imported bronze vessels from the eastern grave of Berdyansk mound [in Russian], Arheologija – Kiev 2009.1, 40–52. Brunaux, J.-L. 2000. Les religions gauloises (Ve-Ier siècles av. J.-C.). Paris: Editions Errance. Cunliffe, B. 1997. The ancient Celts. Oxford – New York: Oxford University Press. Dietler, M. 2006. Alcohol: anthropological/archaeological perspectives, Annual Review of Anthropology 35, 229–49. Diodorus Siculus. Bibliotheca historica, Books 21–32 (tr. F. R. Walton), Loeb Classical Library 409, London, 1968, W. Heinemann. Dizdar, M. 2010. Kantharoi of autochthonous – ‘Pannonian’ origin from the La Tène culture cemetery in Zvonimirovo, Croatia. In S. Berecki (ed.) Iron Age Communities in the Carpathian Basin. Proceedings of the International Colloquium from Tg. Mureş, 9–11 October 2009, 297–307. Cluj-Napoca: Editura Mega. Douglas, M. 2004. Natural symbols. Explorations in cosmology (4th edition). London-New York: Routledge. Egri, M. 2012. ‘A warrior never dies’. The manipulation of tradition in early funerary contexts from Pannonia. In S. Berecki (ed.) Iron Age Rites and Rituals in the Carpathian Basin. Proceedings of the International Colloquium from Tg. Mureş, 7–9 October 2011, 503–29. Cluj-Napoca: Editura Mega. Fowler, C. 2004. The archaeology of personhood. An anthropological approach. London – New York: Routledge. Ginoux, N. 2003. La forme: une question de fond dans l’expression non-figurative des sociétés sans texte. In O. Buchsenschutz, A. Bulard, M.-B. Chardenoux and N. Ginoux (eds.) Décors, images et signes de l’âge du Fer européen, Revue Archéologique du Centre de la France Supp. 24, 259–272. Tours: FERACF. Graells i Fabregat, R. 2008. Vasos de bronce con asas ‘a kouroi’ en el occidente arcaico a la luz de un nuevo ejemplar procedente de Cuenca, Archivo Español de Arqueologia 81, 201–12. Green, M. J. 1997. Dictionary of Celtic Myth and Legend (2nd edition). London: Thames and Hudson. Harding, D. W. 2007. The Archaeology of Celtic Art. London – New York: Routledge. Hell, B. 1991. La force de la bière. Approche d’une recurrence symbolique dans les systems de representations de l’Europe nord-occidentale. In D. Fournier and S. D’Onofrio (eds.) Le ferment divin, 109–23. Paris: Éditions de la Maison des sciences de l’homme.

Horváth, L. 1987. The surroundings of Keszthely. In T. Kovács, É. Petres and M. Szabó (eds.) Corpus of Celtic Finds in Hungary I, 63–178. Budapest: Akadémiai Kiadó. Horváth, L. 2008. Kelta pszeudo-kantharos Zalakomárból (Zala m.), Zalai Múzeum 17, 109–29. Knez, T. and Szabó, M. 1980–81. Ein keltischer Kantharos aus Novo mesto, Archaeologia Iugoslavica 20–21, 80–8. Kriveczky, B. 1991. Régészeti ásatások Tiszafüred-Morotvaparton, Szolnok megyei múzeumi adattár 32, 65–96. Kruta, V. 1991. I Celti della prima espansione storica (IV secolo a.C.). In S. Moscati, O. H. Frey, V. Kruta, B. Raftery and M. Szabó (eds.) I Celti, 195–213. Milano: Bompiani. Kruta, V. 1992. Brennos et l’image des dieux: la représentation de la figure humaine chez les Celtes, Comptes rendus de l’Académie des Inscriptions et Belles-Lettres 136.4, 821–46. Kruta, V. and Szabó, M. 1982. Canthares danubiens du IIIe siècle avant notre ère. Une exemple d’influence hellénistique sur les Celtes orientaux, Études celtiques 19, 51–67. Kull, B. 1997. Tod und Apotheose. Zur Ikonographie in Grab und Kunst der jüngeren Eisenzeit an der unteren Donau und ihrer Bedeutung für die Interpretation von ‘Prunkgräbern’, Bericht der Römisch-Germanischen Kommission 78, 197–466. Marazov, I. 2000. Thracians and wine. Rousse: Dunav Press. Márton, L. 1933. Die Frühlatènezeit in Ungarn, Archaeologia Hungarica 11. Budapest: Magyar Törteneti Múzeum. Megaw, J. V. S. 1970. Art of the European Iron Age. A study of the elusive image. Bath: Adams & Dart. Megaw, J. V. S. 2002. A late La Tène anthropoid gripped sword in New York. In K. Kuzmová, K. Pieta and J. Rajtár (eds.) Zwischen Rom und dem Barbaricum. Festschrift für Titus Kolník zum 70. Geburstag, 407–18. Nitra: Archeologický ústav SAV. Megaw, M. R. and Megaw, J. V. S. 1998. The stone head from Mšecké Žehrovice: an essay on the human head in Early Celtic art. In N. Venclová (ed.) Mšecké Žehrovice in Bohemia. Archaeological background to a Celtic hero (3rd–2nd Cent. BC), 281–92. Sceaux: Kronos B. Y. Editions. Megaw, R. and Megaw, J. V. S. 2001. Celtic art. From its beginnings to the Book of Kells (2nd edition). London: Thames and Hudson. Megaw, J. V. S., Megaw, M. R. and Neugebauer, J.-W. 1989. Zeugnisse frühlatènezeitlichen Kunsthandwerks aus dem Raum Herzogenburg, Niederösterreich, Germania 67.2, 477–517. Moosleitner, F. 1991. Il complesso del Dürrnberg. In S. Moscati, O. H. Frey, V. Kruta, B. Raftery and M. Szabó (eds.) I Celti, 167–173. Milano: Bompiani. Napier, A. D. 1986. Masks, transformation and paradox. Berkeley – Los Angeles: University of California Press. Nelson, M. 2005. The barbarian’s beverage. A history of beer in ancient Europe. London – New York: Routledge. Nepper, M. I. 1976. Kelta temető Körösszegapáti határában, A Bihari Múzeum Évkönyve 1, 7–33. Patton, K. C. 2007. Discussion. In E. Csapo and M. C. Miller (eds.) The origins of theatre in ancient Greece and beyond: from ritual to drama, 361–375. Cambridge: Cambridge University Press. Pieta, K. 2008. Keltské osídlenie Slovenska. Mladšia doba laténska. Nitra: Slovenská Akadémia Vied. Popović, L. B., Mano-Zisi, D., Veličković, M. and Jeličić, B. 1969. Antička bronza u Jugoslaviji. Beograd: Narodni Muzej.

8.  Heads, masks and shifting identities Poux, M. and Feugère, M. 2002. Le festin miroir privilégié des elites celtiques de Gaule indépendante. In V. Guichard and F. Perrin (eds.) L’aristocratie celte à la fin de l’âge du Fer, Collection Bibracte 5, 199–222. Glux-en-Glenne: Centre archéologique européen du Mont Beuvray. Robb, J. 2007. Burial treatment as transformations of bodily ideology. In N. Laneri (ed.) Performing death. Social analyses of funerary traditions in the ancient Near East and Mediterranean, 287–97. Chicago: The Oriental Institute, University of Chicago. Rustoiu, A. 2008. Războinici şi societate în aria celtică transilvăneană. Studii pe marginea mormântului cu coif de la Ciumeşti. Cluj-Napoca: Editura Mega. Rustoiu, A. 2012. The ceramic human head from Deta (Timiş County). About the La Tène vessels with anthropomorphic decoration from the Carpathian Basin, Analele Banatului 20, 57–72. Rustoiu, A. and Egri, M. 2011. The Celts from the Carpathian Basin between Continental traditions and the fascination of the Mediterranean. Cluj Napoca: Editura Mega. Stibbe, C. M. 2000. The sons of Hephaistos. Aspects of the Archaic Greek bronze industry. Roma: L’Erma di Bretschneider. Szabó, M. 1965. A Celtic double head from Badacsony-Lábdi, Acta Archaeologica Academiae Scientiarum Hungaricae 17, 233–50. Szabó, M. 1992. Les Celtes de l’Est. Le Second Age du Fèr dans la cuvette des Karpates. Paris: Editions Errance. Szabó, M. 1998. Les Celtes de la cuvette des Carpates. In T. Kovács, T. Kemenczei and M. Szabó, A la frontiere entre l’Est et l’Ouest. L’art protohistorique en Hongrie au premier millénaire av. n. ère, 51–71. Glux-en-Glenne: Centre archéologique européen du Mont Beuvray.

85

Szabó, M. 2003. Les racines de l’art des oppida dans la region du Moyen-Danube. In A. Szabó and E. Tóth (eds.) Pannonica Provincialia et Archaeologia. Studia Sollemnia Autorum Hungarorum Eugenio Fitz Octogenario Dedicata, 27–46. Budapest: Magyar Nemzeti Múzeum. Szabó, M. and Petres, É. F. 1992. Decorated weapons of the La Tène Iron Age in the Carpathian Basin. Budapest: Magyar Nemzeti Múzeum. Tarditi, C. 1996. Vasi di bronzo in area Apula. Produzioni greche ed italiche di età arcaica e classica. Lecce: Congedo. Teleagă, E. 2008. Griechische Importe in den Nekropolen an der unteren Donau. 6. Jh. – Anfang des 3. Jh. v. Chr., Marburger Studien zur Vor- und Frühgeschichte Bd. 23. Rahden/Westf.: Verlag Marie Leidorf. Todorović, J. 1972. Praistorijska Karaburma I. Nekropola mladeg gvozdenog doba. Beograd: Muzej Grada Beograda. Tonkova, M. 2006. Influences réciproques dans l’orfèvrerie des Thraces et des Celtes au IVe–IIIe siècle av. J.-C. In V. Sîrbu and D. L. Vaida (eds.) Thracians and Celts. Proceedings of the International Colloquium from Bistriţa 18–20 May 2006, 265–78. Cluj-Napoca: Editura Mega. Venclová, N. 2002. The Venerable Bede, druidic tonsure and archaeology, Antiquity 76, 458–71. Vernant, J.-P. 1995. Mit şi gîndire în Grecia antică. Studii de psihologie istorică (Romanian translation by Z. Petre and A. Niculescu of Mythe et pensée chez les Grecs. Études de psychologie historique). Bucureşti: Editura Meridiane.

9 OFF WITH THEIR HEADS…! ONCE AGAIN: IMAGES OF DAGGERS AND SEVERED HEADS ON AN IBERIAN FALCATA SWORD Fernando Quesada Sanz

The Iberian cemetery at Coimbra and burial 48 The archaeological complex at Coimbra (Jumilla, Mucia) includes a settlement surrounded by fortifications or cliffs encircling c.54 ha (only a fraction of this area was actually a built-up area, however), a sanctuary and three cemeteries, all of them partially excavated. The site falls squarely within the non Indo-European area of the Peninsula; in fact, it is one of the prototypical ‘Iberian Culture’ sites in Contestania-Bastetania, a region that covers a good part of the southeastern quadrant of modern Spain (Fig. 9.1b). Between 1995 and 2004 a grand total of 160 cremation burials were excavated in the cemetery of ‘El Poblado’, the one closest to the main entrance of the oppidum. The time span covered by this particular cemetery is c.400 to c.175 BC, with a reasonable equilibrium by generations (García Cano 1997; García Cano et al. 2008). Burial 48 contained the cremated remains of a single adult; sex could not be ascertained by the examination of the remains. The ashes and cremated bone fragments were deposited while still very hot in the bottom of shallow rectangular pit (1.6 m. × 1.1 m. × 0.30 m.) without a cinerary urn, together with a substantial set of grave goods and some cremated ovicaprine bones. The pit was then covered with a rectangular platform made of undressed stones, a sort of one-step tumulus typical of this region. The grave goods included two black glaze Attic vases (one 40D L kantharos and one 21L bowl) dated to c.350–325 BC, a coherent set of weapons (one falcata sword, one soliferreum, one shield, two horse spurs, a snaffle bit), some partly melted down faience decorative elements (perhaps from the shield body), a bronze brooch of the so-called La Tène I type but of typically Iberian manufacture, some Iberian pottery, at least six small bone awls, and some other badly preserved small metal (iron and copper alloy) remains. The weapons were carefully placed

in the pit: first the soliferreum was bent in a figure-of-eight shape, the falcata was then laid over it, and finally the shield on top of them. Overall, it seems quite safe to assume that this was a masculine burial (further discussion on bone analysis, presence of weapons, sex identification and gender problems in Quesada Sanz 2011 and 2012). Tomb 48 is a very significant burial of above-average richness within the context of the cemetery. Grave goods include 17 objects, while the mean number of objects per burial in the cemetery is 8.1. Specifically, it is the third ‘richest’ burial deposited during the 4th century BC, quite above the statistical mode (9 objects) and mean, although the two really ‘rich’ burials of the 4th century BC contain 31 and 94 objects. No wealth-units analysis was performed (García Cano 1997, 93 ff.) but the qualitative analysis (multiple presence of imported pottery, complex set of weapons, horse harness) place this burial within the VIB (‘very important burials’) group, even if the structure of the grave itself is quite modest, not to be compared with the 94 objects and the imposing sculpted pillar-stele on top of the almost contemporary burial 70.

The falcata sword and its decoration The falcata mentioned above as part of the burial is a typical example of this type of cut-and-thrust sword (the asymmetrical shape of its blade is deceptive). It has a horseshaped side guard/pommel, and its shape and relative size (blade length 56 cm; total length 67 cm) is typical of these swords at this time and in this region (Quesada Sanz 1997, 83 ff. for typological details). The falcata is not the typical Iron Age sword of the Peninsula; there are many other characteristic models in

9.  Off with their heads…! once again: Images of daggers and severed heads on an Iberian falcata sword

87

a

b

Fig. 9.1. a: The two great linguistic and cultural areas of the Iberian Peninsula in the Iron Age combining different criteria (after J. Untermann 1995). b: Writing systems and diagnostic toponyms in ancient Iberia (c. 5th–1st century BC) (after P. Moret in Moret and Rouillard 1997, summarizing Untermann, Albertos and others).

88

Fernando Quesada Sanz

different regions (e.g. Celtiberia). In fact, it is not even a typical ‘Iberian sword’: there are many Iberian-culture regions where falcatas are very rare, while La Tène type swords are more frequent, such as in Catalonia. Over 86 percent of all falcatas (more than 700 are known) have been found in south-eastern Spain (Quesada Sanz 1997, fig. 16). A few of them come from Iberian regions to the north (Valencia, Aragón, Catalonia) and west (western Andalusia), and very few have been documented in Celtiberia or Lusitania. In fact, most falcatas found outside the provinces of Córdoba, Jaén and Granada (in eastern Andalusia), and Albacete, Murcia, Alicante and southern Valencia (south-east) are fairly rare exceptions that can be safely be considered as imports from the nuclear area of production. The Coimbra falcata (Fig. 9.2) was richly decorated with silver inlay motifs along the blade fullers, and also in the wider proximal part of the blade, just under the hilt. These are typical places for decoration, and in this case the motifs were designed in what I’ve denominated the ‘free style’, fairly crude from the technical point of view, but displaying unusual, very individual elements rarely if ever found in the more detailed, rigid and perfectionist ‘precise style’ (see Quesada Sanz et al. 2000). Silver inlay tends to disappear under layers of iron corrosion, but modern restoration techniques and more systematic use of metallurgical X-ray testing show that a much higher number of Iberian swords than previously known were decorated with silver inlay technique, reaching perhaps 25 to 30 percent of the total number of falcatas. Only fragments of the original decoration remain, but it is clear that two long parallel bands, linked spirals forming ‘waves’ and zigzag lines, were inlaid with silver wire along the fullers and the base of the blade in the shape of a long ‘L’ (Figs 9.2, 9.3, 9.4). These motifs (Quesada Sanz nos. 1 and 9) are very common in Iberian metalwork but they are rare, especially in combination, in the Iberia Celtica. Also at the junction of the two branches of the ‘L’ shape another typically Iberian motif (n. 28) was inlaid (see Quesada Sanz 1997, figs. 66 and 69 for details). Far more interesting, and unique, are the three frontal human faces and the triangular dagger inlaid in the surface of the base of the blade (Figs 9.4 and 9.5). They were originally inlaid in both sides, but the reverse is badly preserved, and only the dagger is now visible on that surface (Fig. 9.6). X-ray images show in many cases that identical themes were depicted on both sides of a blade (see examples in García Cano and Gómez Ródenas 2006, burials 160, 198, 386, 597 etc., from the cemetery at Cabecido del Tesoro). As far as we know, these heads are in fact the only known anthropomorphic representations inlaid in the surface of falcatas. Animals (felines, boars, wolves, birds, even fishes) are attested, but human figures are so far unknown, except for this case. There are indeed a very few examples of three-dimensional human heads modelled as parts of the

side-guards of hilts (e.g. La Serreta de Alcoi, burial 53, see Moltó and Reig 2000), but they are different in character and style. The three human faces – or heads – are placed together in a row. The one to the right is the best preserved (and the only one mentioned in the original publication: García Cano 1999, 44), but the other two are also discernible and seem to be almost identical (Fig. 9.4, insert). The faces measure about 0.8 cm in height (1.1 including beard) and are completely frontal. Head hair is not represented, but what looks very much like a beard marked by five lines is clearly visible. In fact, facial hair, although very uncommon in sculpture or reliefs, is occasionally shown in painted pottery (Ballester 1943 as dots or short lines). Nose, eyes and mouth are shown as very short straight lines, while eyebrows (or brow ridges) are clearly indicated as curved lines joined to the top of the nose. The ears are also very prominent, and are placed in a very high position. Very relevant to the interpretation of the row of heads is the dagger represented to their side (Fig. 9.5). It has a broad, triangular blade and the pommel is indicated by a short curved line, pointing upwards. This simplified depiction of a dagger (the blade is 1.1 cm, and the whole weapon is only 1.8 cm long) has some parallels in painted pottery. For example, the hero fighting a monster (in this case a sphinx) in a vase from Corral de Saus (Valencia) is shown wielding a spear in one hand, and a similar dagger in the other (Izquierdo 1995). But particularly, the vase from burial 400 at Cigarralejo (Cuadrado 1983), a cemetery very close both geographically and culturally to Coimbra, shows in black the silhouette of very similar daggers (Fig. 9.7). The vase can be dated to c.350 BC, was used as a cinerary urn in a warrior burial, and its decoration as a whole is unique, although the different icons appear elsewhere. All painted daggers in it show a similar broad triangular blade, but two pommel types can be distinguished, both easily recognizable in archaeological terms: the atrophied antennae and the frontón dagger (Quesada types IIA1 and IIB respectively, see Quesada 1997, 280 ff.). The broad, straight sided, triangular blades with either fronton or antennae pommels are typical of 5th to 4th century BC Iberian daggers. They have been found both in the Southeast and Andalusia, with some more isolated examples in Portugal (Alcacer do Sal, mouth of the river Tagus) and the western Meseta. They are also depicted in the great Porcuna monument (Negueruela 1990, plates XVIIIA, XXXVI).

Interpretation We agree with the principle that ancient craftsmanship, prehistoric or classical, was not ‘art’ for the sake of aesthetic pleasure in itself, even in those works that clearly show a defined ‘style’ and a specific interest in producing aesthetically pleasant or visually impressive results, be it a

9.  Off with their heads…! once again: Images of daggers and severed heads on an Iberian falcata sword

89

Fig. 9.4. Detail of the lower part of the hilt and proximal part of the blade, with anthropomorphic decoration (three human heads) and a dagger. (Author’s photo)

Fig. 9.2. Iberian falcata from burial 48 at the ‘El poblado’ cemetery, Coimbra del Barranco Ancho, Jumilla, Murcia, Spain, c. 350–325 BC. (Author’s drawing)

Fig. 9.3. Detail of the silver inlay decoration on the blade, along the fullers. (Author’s photo)

Fig. 9.5. Detail of one of the human faces and the triangular dagger. (Author’s photo)

90

Fernando Quesada Sanz

Fig. 9.7. Cinerary urn of burial 400 at El Cigarralejo (Mula, Murcia). (Photo Mula Museum).

Fig. 9.6. Reverse of the blade, showing the blade of the dagger. The rest of the decoration, mirroring that of the other side of the blade, has been lost. (Author’s photo)

biface, an Aphrodite or Hagia Sophia. Nor was ‘art’ conceived as a vehicle for expressing the artisan’s (or artist’s if we so wish) intentions, world view or desire to communicate. Minor or major objects that we now classify as works of art – and their decoration – were never purely ornamental, but were loaded with layers of abstract meaning that could and did accumulate one on of top of the other, both in daily life or in even more symbolically loaded contexts, such as funerary ritual. A boar inlaid in silver on the surface of a falcata (Quesada 1997, pl. IIID) could not only allude both to some physical characteristic of the weapon’s owner and to his bravery in battle, but also additionally play with his nickname, act as an heraldic symbol, and even add a funerary symbolism when the sword was deposited in his grave, or perhaps in his father’s or son’s. Of course much more sophisticated discussions can be added to this basic concept

but this will suffice here (in the case of ‘Celtic’ art, centering on metalwork, see recently the inspiring work by Garrow and Gosden 2012, 38 ff.). Of course, we can accept a working definition of art as the ‘elements of decoration beyond those necessary for functional utility though these elements represent a form of symbolic visual communication which is only partially accessible to us’ as the Megaws humbly and succinctly put it years ago (Megaw and Megaw 2001, 9). In our opinion, the combination of heads and dagger in the Coimbra falcata is basically a direct allusion to victory in combat, and probably to the number of vanquished enemies, killed by the sword’s owner, probably (but not necessarily) the one who was cremated with it and buried in Grave 48, a quite important burial as we have described above, worthy of a respected and probably rich warrior. The dagger complements the images of the severed heads; in fact this is the only image of a weapon on a real Iberian weapon, interestingly enough, as it seems to indicate that there is a specific connection with the human severed heads. We do not know if this particular falcata was specifically forged and decorated for the grave. In fact, we suspect it was not. We do know that in Antiquity many richly decorated weapons were carried into battle and were perfectly functional, as this falcata certainly was. So it is quite possible that its proud owner carried it many times before his death. It is even possible that it was originally forged without that particular decoration, or without any decoration at all. We do know that the smiths who forged swords were not necessarily those who

9.  Off with their heads…! once again: Images of daggers and severed heads on an Iberian falcata sword decorated them with fine silver motifs, and most probably they almost never were blacksmiths and silversmiths at the same time (Quesada et al. 2000). Our impression is that the geometric motifs along the fullers could have been inlaid first, and that heads and dagger could have been added later, the weapon being displaced to the right for lack of space (Fig. 9.5). This phenomenon has parallels elsewhere (see examples in Celtic swords in Garrow and Gosden 2012, 132–133). The idea of ‘counting heads’ as proof of victory in History, from ancient Egypt to Vietnam, from Assyria to the Americas, is of course so common and well-known it needs no further discussion here; it has nothing particularly ‘Iberian’ or even ‘Celtic’ in it (Sterckx 2005, 53–103 proves the point while concentrating of the Celtic variants of this practice). The Iberians were probably not alien to this custom: Diodorus Siculus describes the siege of Selinous in Sicily in 409 by a mercenary Carthaginian army. Hannibal launched successive assaults by different contingents, first the Campanians (13.55.7) and other peoples, but it was the Iberians who finally took the city (13.56.6). The conquerors, ‘according to the practice of their people, mutilated even the dead, some carrying bunches of hands about their bodies and other heads which they had spitted upon their javelins (saunia) and spears (gaesa)’ (13.57.3). It is however true that our source does not specifically mention that it was the Iberians who did this, and it seems likely (contra Sterckx 2005, 89; Marco 2006, 201) that he had the Gauls in mind, as he had already mentioned this custom in his discussion of their customs (5.29.4) see also Aguilera 2013). Some would also cite the names of the weapons that Diodorus mentions as additional evidence for this. However, the idea of displaying images of specific numbers of weapons or of body parts as some kind of ‘statistic’ is not so common, but is also documented. We should recognize however that it is weapons, and not human parts, that are usually employed for ‘body counts’. In the case of the Iberians, Aristotle specifically states (Politics 7.2.11 – 1324b) that ‘among the Iberians, a warlike people [Ibersin, ethnei polemikoi], they fix spears [obeliskoi] in the earth round a man’s grave corresponding in number to the enemies he has killed’. Admittedly, Aristotle never went to Iberia, but his master Plato did travel to Sicily at least three times between 387 and 361 BC, and met Iberian mercenaries in Syracuse (Plato Epistles 7.348a). A curious man, he could have obtained some first-hand information there and told about it later in Athens. This literary reference is not our only hint: there is evidence of small numbers of spearheads stuck vertically in a number of Iberian (and Celtiberian) burials. Last but not least, there is a series of Late Iberian incised stone stelae that depict rows of spears or spearheads, although it is unlikely that they were meant to show the exact number of enemies killed in combat (see Quesada 1997, 424 ff. and figure 250; most recently, Marco and Royo 2012, 315–16). In this respect, the Iberian funerary pillar-stele from Caspe

91

(López Monteagudo 1983; Pérez Rojas 1983) is worth a brief comment. On top of the monument a stone lion sculpted partly in the round, partly in relief, crowns a square section pillar that displays a rather long but fragmentary text in Iberian script, unreadable – as all of them are – but perhaps originally mentioning two to four personal names. Between the inscription and the lion, four shields (three caetrae and one oval thureos) are represented. These shields could belong to the people mentioned in the inscription (supposing the interpretation is correct, but see de Hoz 2011, 321 ff.) but another possibility is available. In epigraphic cultures whose texts we can read, things become somewhat easier: for example, the great commemorative pillar from Xanthos in Lycia shows a row of seven aspides (hoplite round shields), a fallen warrior and a standing one. The long inscription on the northern side of the pillar tells the story: seven Arcadian hoplites were defeated by the Lycian prince, and the weapons represent them (Demargne 1958, 79–103). As we have mentioned above, this basic meaning could have been complemented and enriched with other added meanings, probably even in the funerary context in which the sword finally met its resting place. Among them, an apotropaic symbolism seems likely (see below).

The Peninsula: Falcatas, Iberians and Celticspeaking peoples The palaeoethnology of the Iberian Peninsula during the Iron Age is a very complex affair (Almagro and Ruiz Zapatero 1992), but epigraphy and its associated disciplines toponymy and onomastics have long proved (see Untermann 1963 refined many times afterwards, Fig. 9.1a) the existence of two main linguistic areas, marked by a roughly diagonal, undulating buffer zone running northeast to southwest, basically along the mountain ranges separating the Mediterranean and Atlantic watersheds. To the west and north of this line different variants of Indo-European languages are attested. This area corresponds with regions inhabited by different pre-roman peoples known to ancient writers as Celtiberi, Vaccaei, Vettones, Celtici and a long etcetera. To the east and south of this linguistic – but also archaeological – frontier (Catalonia, Lower Ebro valley, Mediterranean coastal regions and most of Andalusia), the different pre-Roman inhabitants spoke a non-Indo-European language (or languages) collectively known as Iberian and probably Turdetanian in western Andalusia. There is a long list of known ethnonyms for these peoples (including, from south to north, the Turdetani, Bastetani, Contestani, Edetani, Ilercavones, Ilergetes, Laietani and Indicetes), which are conveniently grouped into two archaeological ‘cultures’: Turdetanians in western Andalusia and Iberians all along the Mediterranean façade and into southern France. With all its limitations, this basic division still stands as a useful

92

Fernando Quesada Sanz

instrument for basic research and description (e.g. Lorrio and Ruiz 2005a; Lorrio 2011; Moret and Rouillard 1997) (Fig. 9.1b). Some – perhaps most – languages in the Central, Western and Northern parts of the Peninsula were thus of IndoEuropean lineage, and some of them are undoubtedly Celtic in character (e.g. Celtiberian), while there is still much discussion on others (e.g. Lusitanian, see de Hoz 2011, 563 ff.). On the other hand, all serious linguists agree that Iberian is certainly not an Indo-European language (e.g. Untermann, 2001; de Hoz 2010; 2011). Also, almost all specialists maintain that the dates and geographical extension of Iberian epigraphy closely match what archaeologists call ‘Iberian Iron Age culture’ from the point of view of material remains, including architecture, pottery, weapons and other artifacts, together with more elusive concepts such as economy or patterns of territorial control. Thus, the Iberian Culture is basically neither Indo-European nor Celtic in language or material culture.

Têtes coupées, Celts and Iberians How should we then interpret the severed heads in the Coimbra falcata? As an Iberian element, or as proof of strong Celtic influence? Most Celtic scholars allow for the Mediterranean origin (mostly Etruscan) of many of the images of faces and heads in early Celtic Art (e.g. Megaw and Megaw 2001, 70), or at least pay lip service to that iconographical debt before moving on into a truly ‘Celtic’ head-and-face fixation. In the case of Coimbra, the cultural and archaeological context, and the weapon itself, are purely Iberian. But some scholars (e.g. Lenerz de Wilde 1986) have seen a strong La Tène influence not only in some decorative patterns in Celtiberian metalwork, but also in clearly Iberian weaponry, including some motifs such as linked lyres and palmettes, or ivy leaves, that are rare or absent in Celtiberian weapons. It is certainly possible to view these elements as ‘Celtic’, but it is a much more economical hypothesis to acknowledge – as most scholars do – that Celtic art is an amalgam of many elements, many of them originally Etruscan or Italian, interpreted according to its own logic (e.g. Megaw and Megaw 1989, 20ff. 70 ff.; Garrow and Gosden 2012, 40). Iberian art was also subject to the same influences, and the falcata type itself has an original Italic provenance (Quesada 1997, 123 ff.). But the problem of severed heads as a custom remains, leaving aside the subject of certain ornamental motifs. Most European archaeological literature tends to see the ‘tête coupée’ phenomenon during the Iron Age as an almost purely Celtic cultural trait. It is perhaps striking that some of the early papers on the subject emphasize the ‘Mediterranean connections’ (Benoit 1949; 1969), while later works insistently remark the words ‘rite celtique des têtes coupées’ (Chassaing

1976 among many others). In many cases it has acquired the rank of a distinguishing feature of the Gauls in particular. Most manuals on the Celts display specific entries on ‘heads’ in their indexes, often with specific subheadings in which some emphasis is put on ‘severed heads’ (e.g. Megaw and Megaw 1989; James 1993; Green 1993; Cunliffe 1997 etc.). But these entries hide an often bewildering variety of sources (literary, iconographic, osteological), and of subjects (references to decapitation of fallen enemies in literary sources, exhibition of crania in sanctuaries or even the façades of houses, crowned heads, stylized decorative motifs in the form of human heads, sculptures in which heads, severed or otherwise are shown as part of a more complex theme, and a long etcetera). These objects, images or literary citations may or may not have a related meaning, except in that their central theme is the human head or the human face, disembodied but not necessarily the severed head of a particular individual. At the bottom of every single explanation, of course, lies the fact, already explained by Plato, that the human head is the most divine part of us (Plato, Timaeus 44d). The apotropaic virtues of preserved heads or crania have been a preferred explanation for ritual decapitation (Sopeña 1987, 103 ff.), and this apotropaic explanation is often extended to human faces in many different Celtic and Celtiberian contexts, from jewellery to monumental stone sculpture (Harding 2007, 54; Alfayé 2011, 200). In some of these cases, human heads could have been those of heroes, ancestors or gods, a very different proposition (see recently on the human sculptures in Southern France and their meaning, Py 2001, 119 ff.; McCartney 2012, 74 ff. in which the contradiction between interpretation as sculpted ancestors and decapitated enemies is very much in evidence). Finally, the idea of a ‘head worship’, a cult centered on the human head as such (e.g. Llanos 2007), should be treated with healthy scepticism, in the Keltike and elsewhere (see a summary of different views on this in Alberro 2003–4, 217–18 and a shorter one in McCartney 2012, 82). On the other hand, the specific explanation of the beheading of defeated enemies as a means of symbolically killing his soul and identity, of destroying his hopes of life in the netherworld and of controlling his spirit (e.g. Brunaux 1986, 110–111; Dedet 2011, 288), has also long been prevalent, and some authors have even unfairly reproached classical authors for leaving the ritual aspect of this practice unexplored (Richtie and Richtie 1995, 54). But some recent work is also returning to what many ancient sources actually say about mutilation of enemies as a straightforward means of inflicting heavy indignity on his remains (Alfayé 2004, 71 ff.). The amputation of hands is in this respect closely related to decapitation (e.g. Sopeña 2008; Alfayé 2004; Green 2006, 298 ff.; Marco 2006, 201). A certain economic value, a ‘valeur marchande’ of severed heads as proof of victory or of treason has also been suggested (Brunaux 2004, 112). The same author, using ancient literary sources, has recently criticized the idea that

9.  Off with their heads…! once again: Images of daggers and severed heads on an Iberian falcata sword the Gauls inflicted humiliation on the severed heads of their defeated enemies, as the Romans or Scythians did; to Brunaux, ‘la tête connaît un sort gratifiant’ (2012, 107). In the Iberian Peninsula there is a long tradition of studies on the tête coupée, mostly – but not always – from a Celtic perspective (Taracena 1943; Balil 1956; Blázquez 1958; 1961; Pujol 1979–80; López 1987; Sopeña 1987, 99 ff.; Díaz 1989; Almagro and Lorrio 1989; Alberro 2003–4; Llanos 2007). Only a perceptive – and still unpublished – paper criticizes systematically the excessive focus placed on Celtiberia while at the same time discarding the opposition between ‘Iberian’ and ‘Celtic’ (Aguilera 2013). Some human crania, originally fixed with strong iron nails to walls, have been found in Iberian sites such as Ullastret, Illa d’en Reixac or Molí de Espígol (Gerona), close to Southern Gaul, and have therefore been interpreted according to the Celtic tradition of defeated enemies (see Hermary 2003; Agustí, Lara and Martín 2010; Ciesielki et al. 2011; Codina et al. 2011). Other crania from Numantia in Celtiberia, far from Catalonia, are however obviously related to it (Taracena 1943). Recent work believes that some of these crania might have been not only those of enemies, but also of ancestors, obviously to be placed in different contexts (Ciesielki et al. 2011, 116). Their interpretation has reached a high degree of complexity and even confusion, leading to a very specific Conference on the subject, aptly titled ‘Crânes trophées, cranes d’ancêtres et autres pratiques autour de la tête’ (Boulestin et al. 2012). Images of disembodied human heads also appear in Celtiberian horse-fibulae. These are often shown just in front of the miniature horses that form the body of the brooch, as if they were suspended from the horse’s neck (Almagro and Torres 1999, 72 ff.). Of course, this is just what Diodorus says Gauls did to their victims (see above), but it is also exactly what Assyrian cavalry did to theirs in Ashurnasirpal’s II time (Yadin 1963, 385). Although Occam’s razor counsels to choose the influence that is close in time and space, and it is reasonable to consider these horse-fibulae with trophy heads as ‘one of the most characteristic subjects of Celtic Art in the Iberian Peninsula’, the fact that this seems to be an almost universal practice should not be forgotten. An article by E. Knauer has shown indeed how frequent the custom of suspending heads from the neck of horses has been (Knauer 2001). Miranda Green has recently suggested that these brooches could be perhaps also understood within the realm of a ‘mythologised conflict’ as it seems to be the horses that take iconographic precedence (Green 2010, 197), an example of these possible ‘layers of meaning’ we have mentioned above. Therefore we should not use these fibulae – or any other Celtiberian object with a similar motif – to argue that the severed human heads on the Coimbra falcata reflects ‘Celtic’ influence of any kind, other than a common perception of the significance of this part of the human body in many different

93

contexts. In fact, some authors have recently pointed out that most of the most explicit images of decapitation come from Iberian – Aragonese – and not from Celtiberian contexts. This is the case of the stone stelae from El Palao (Alcañiz) and La Vispesa (Huesca) (Marco Simón 2006, 201) both displaying explicit scenes of victory. Aguilera has recently added that the evidence of human crania in other Iberian sites such as La Alcudia (Elche) renders a purely Celtiberian or Celtic interpretation of the ‘têtes coupées’ in the Peninsula very questionable.

Conclusion We have argued elsewhere (Quesada 2005): (a) that it is a serious mistake to apply La Tène categories, types and chronologies to the Iberian Peninsula affairs in general, and to weapons in particular (see also recently García Jiménez 2012). The Peninsula has a strong and distinctive personality that does not allow for simplistic extrapolations (see also Lorrio and Ruiz 2005b, 44). (b) That – due to the history of research – it has often been argued that the Celtic areas of the Iberian Peninsula, supposedly more warlike, heavily influenced Iberian weapon design and evolution, while in fact the opposite is generally true. Both these arguments are relevant to our present case, and while widely accepted, are often forgotten. I would particularly emphasize that while some La Tène-type objects or decorative elements may appear from time to time in Iberian contexts, they are mostly isolated cases (except in northeastern Catalonia) and should not be taken as proof of anything like strong Celtic presence or influence. In the case of our falcata, it might be tempting to see the images of the severed heads as part of the supposedly Celtic (or even Gaulish) ‘tête coupée’ phenomenon. I believe that this would be a mistaken approach, and that the decoration should be interpreted in purely ‘Iberian’ terms. Beheading vanquished enemies, the display of their heads and crania with both an apotropaic meaning and the intention to humiliate the memory of the fallen enemy; the depiction of heads as a means of boasting and calling attention to one’s own accomplishments, these are all activities close in meaning and aspect to similar Celtic and Mediterranean customs and traditions, but can be more easily explained in terms of convergence rather than of external influence.

Acknowledgements I am grateful to Dr. Jose Miguel García Caro (Universidad Murcia) for his kind permission to study the weapons even when they were originally still unpublished, and for the time spent discussing Iberian cemeteries. Any mistakes that might remain are exclusively mine.

94

Fernando Quesada Sanz

Bibliography Aguilera Durán, T. 2013 (forth.). El rito celta de las cabezas cortadas en Iberia: revisión de un tópico historiográfico. In F. Burillo (ed.) VII Simposio sobre los Celtíberos. Nuevos hallazgos. Nuevas interpretaciones. Daroca: Marzo 2012. Agusti Farjas, B., Lara, L. and Martín, A. 2010. Restauració i reestudi antropològic de quatre cranis del Museu d’Ullastret, Cypsela 18, 307–19. Alberro, M. 2003–2004. El paradigma céltico de las cabezas cortadas y su conexión con la diosa de guerra ornitomórfica y los ritos funerarios de los celtíberos. Homenaje a R. Lucas Pellicer, Kalathos 22–23, 195–249. Alfayé Villa, S. 2004. Rituales de aniquilación del enemigo en la Estela de Binéfar (Huesca), Actas del XXVII Congreso Internacional Girea-Arys IX. Jerarquías religiosas y control social en el mundo antiguo, Valladolid, 2002, 63–74. Valladolid. Alfayé Villa, S., 2011 Imagen y ritual en la Celtica peninsular. Noia, ed Toxosoutos. Almagro-Gorbea, M. and Lorrio Alvarado, A. 1989. Represent­ aciones humanas en el arte céltico de la Península Ibérica. Actas del II Symposium de Arqueología Soriana, I, 409–52. Soria. Almagro-Gorbea, M. and Ruiz Zapatero, G. (eds.) 1992. Paleoetnología de la Península Ibérica, Actas de la reunión celebrada en la Facultad de Geografía e Historia de la Universidad Complutense de Madrid, 1989, Complutum 2–3. Madrid. Almagro-Gorbea, M. and Torres Ortiz, M. 1999. Las fíbulas de jinete y de caballito. Aproximación a las élites ecuestres y su expansión en la Hispania céltica. Zaragoza: Institución Fernando el Católico. Balil, A. 1956. Representaciones de cabezas cortadas y cabezastrofeo en el Levante español, IV Congreso Internacional de Ciencias Prehistóricas y Protohistóricas, 871–79. Zaragoza. Ballester Tormo, I. 1943. Las barbas de los Iberos. Notas sobre la cerámica de San Miguel de Liria, Ampurias 5, 109–16. Benoit, F. 1949. L’aire mediterranéenne de la tête-coupèe, Rivista di Studi Liguri XV, 243–55. Benoit, F. 1969. Gorgone et tête coupée, du rite au mythe, Archivo Español de Arqueología 42, 81–93. Blázquez Martínez, J. M. 1958. Sacrificios humanos y representaciones de cabezas en la P. Ibérica, Latomus 19, 27–48. Blázquez Martínez, J. M. 1961. Cabezas cortadas inéditas del Castro de Yecla (Salamanca), VII Congreso Nacional de Arqueología, 217–26. Zaragoza. Boulestin, B. and Gambier, D. H. (eds.) 2012. Crânes trophées, crânes d’ancêtres et autres practiques autour de la tête: problèmes d’interpretation en archéologie, BAR International Series 2415. Oxford: Archaeopress. Brunaux, J. L. 1986. Les Gaulois. Sanctuaires et rites. Paris: Errance. Brunaux, J. L. 2004. Guerre et religion en Gaule. Essai d’anthropologie celtique. Paris: Errance. Brunaux, J. L. 2012. Du prix et des usages de la tête. Les données historiques sur la prise du crâne en Gaule. In B. Boulestin and D. H. Gambier (eds.) Crânes trophées, crânes d’ancêtres et autres pratiques autour de la tête: problèmes d’interprétation en archéologie, BAR Interational Series 2415, 107–116. Oxford: Archaeopress. Chassaing, M. 1976. Du rite celtique des têtes coupées et de sa

survivance dans l’iconographie gallo et germano-romaine, XX Congrés Prehistorique de France 1974, 69–81. Paris: Société Préhistorique Française. Ciesielki, E., Duday, H., Girard, B., Roure, R., Martin, A. and Agusti, B. 2011.La pratique des têtes coupées et les dépôts d’armes en Gaule méditerraneenne et dans le nord-est de la Péninsule Ibérique. In R. Roure and L.Pernet (eds.) Des rites et des homme:. Les pratiques symboliques des celtes, des Ibères, et des Grecs en provence, en Languedoc et en Catalogne, 112–145. Paris: Errance. Codina, F., Martin, A., Prado, G. de and Agustí, B. 2011. Ullastret. Baix Empordà, Catalogne. In R. Roure and L.Pernet (eds.) Des rites et des hommes. Les pratiques symboliques des celtes, des Ibères, et des Grecs en provence, en Languedoc et en Catalogne, 158–163. Paris: Errance. Cuadrado Díaz, E. 1983. Una decoración excepcional en la cerámica ibérica. In Homenaje al Prof. Martín Almagro Basch III, 57–67. Madrid. Cunliffe, B. W. 1997. The Ancient Celts. Oxford: University Press. De Hoz Bravo, J. 2010. Historia lingüística de la Península Ibérica en la Antigüedad, I. Preliminares y mundo meridional prerromano, Anejos de Emerita L. Madrid: CSIC. De Hoz Bravo, J. 2011. Historia lingüística de la Península Ibérica en la Antigüedad, II. El mundo ibérico prerromano y la indoeuropeizació, Anejos de Emerita LI. Madrid: CSIC. Dedet, B. 2011. La tête coupée, symbole de mise à mort suprême en Gaule méridionale? Des textes anciens aux données de l’archéologie. In P. Gruat and D. Garcia (eds.) Stèles et statues de l’Age du Fer; Documents d’Archéologie Méridionale 34, 281–9. Demargne, P. 1958. Fouilles de Xanthos I. Les piliers funéraires. Paris: Institut Français d’Archéologie d’Istambul. Díaz Sanz, M. A. 1989. Sacrificios humanos en la Celtiberia Oriental: las ‘cabezas cortadas’. II Encuentro de Estudios Bilbilitanos, Calatayud 1986, I, 33–41. Calatayud. García Cano, J. M. 1997. Las necrópolis ibéricas de Coimbra del Barranco Ancho (Jumilla. Murcia). I. Las excavaciones y estudio analítico de los materiales. Murcia: Universidad de Murcia. García Cano, J. M. 1999. Las necrópolis ibéricas de Coimbra del Barranco Ancho (Jumilla. Murcia). II. Análisis de los enterramientos, catálogo de materiales y apéndices. Murcia: Universidad de Murcia. García Cano, J. M. and Gómez Ródenas, M. A. 2006. Avance al estudio radiológico del armamento de la necrópolis ibérica del Cabecico del Tesoro (Verdolay, Murcia) I. Las falcatas, Gladius 26, 61–92. García Cano, J. M.; Page, V.; Gallardo, J.; Ramos, F.; Hernandez, E. 2008. El mundo funerario ibérico en el Altiplano JumillaYecla (Murcia): la necrópolis de El Poblado de Coimbra del Barranco Ancho. Investigaciones de 1995–2004. Vol. II. Murcia. García Jiménez, G. 2012. El armamento de influencia la Tène en la Península Ibérica (siglos V–I a.C.), Monographies Instrumentum 43. Montagnac: Eds. M. Mergoil. Garrow, D. and C. Gosden. 2012. Technologies of Enchantment? Exploring Celtic art 400 BC to AD 100. Oxford: Oxford University Press. Green, M. 1993. The Gods of the Celts 2nd ed. Stroud: A. Sutton.

9.  Off with their heads…! once again: Images of daggers and severed heads on an Iberian falcata sword Green, M. 2006. Semiologies of subjugation: the ritualisation of War-prisioners in later European Antiguity. In T. Otto, H. Thrane and H. Vandkilde (eds.) Warfare and Society. Archaeological and Social Anthropological perspectives, 281–304. Aarhus: University Press. Green, M. 2010. La religión celtibérica desde la religión céltica. In F. Burillo (ed.) Ritos y mitos, VI simposio sobre Celtíberos, 189–204. Zaragoza. Harding, D. W. 2007. The Archaeology of Celtic Art. London: Routledge. Hermary, A. 2003. Grecs et Barbares cloueurs de têtes: compléments au temoinage de Poseidonios, Hommage a Guy Barruol. Suppl. Revue Archéologique de Narbonnaise 35, 525–30. Izquierdo Peraile, I. 1995. Un vaso inédito con excepcional decoración pintada procedente de la necrópolis ibérica de Corral de Saus (Moixent, València), Homenatge a Milagros Gil-Mascarell, Saguntum 29, 93–104. James, S. 1993. Exploring the world of the Celts. London: Thames and Hudson. Knauer, E. R. 2001. Observations on the ‘Barbarian’ custom of suspending the heads of vanquished enemies from the neck of horses, Archaologische Mitteilungen aus Iran 33, 283–332. Lenerz de Wilde, M. 1986. Art celtique et armes ibériques, Aquitania Suppl. 1, 273–80. López Monteagudo, G. 1983. La estela de Calpe y los pilares-estela ibéricos, Archivo Español de Arqueología 56, 261–5. López Monteagudo, G. 1987. Las ‘cabezas cortadas’en la Península Ibérica, Gerión 5, 245–52. Lorrio Alvarado, A. J. 2011. Los celtas en el Occidente de Iberia. In G. Ruiz Zapatero and J. Alvarez Sanchís (eds.) Castros y Verracos, 45–100. Avila: Diputación provicial. Lorrio Alvarado, A. J. and Ruiz Zapatero, G. 2005a. The Celts in Iberia: an overview. E-Keltoi 6, 167–254. (=http://www4.uwm. edu/celtic/ekeltoi/volumes/vol6/6_4/lorrio_zapatero_ 6_4.pdf). Lorrio Alvarado, A. J. and Ruiz Zapatero, G. 2005b. Celtiberians: Archaeology of Celts in Iberia. In W. Gillies and D. W. Harding (eds.) Celtic Connections, Papers from the Tenth Intrnational Congress of Celtic Studies, Edinburgh 1995, Volume 2, 33–55. Edinburgh: University of Edinburgh. Llanos Ortiz, A. 2007. El rito de las cabezas cortadas en el poblado de La Hoya (Laguardia, Alava), Homenaje a I. Barandiarán, Veleia 24–25, 1273–81. Marco Simón, F. 2006. Intimidación y terror en la época de las Guerras Celtibéricas. In G. Urso (ed.) Terror et pavor, 197–213. Pisa: ETS. Marco Simón, F. and Royo Guillén, J. I. 2012. Iconografía entre la Primera Edad del Hierro y la Romanización: nuevos documentos y nuevas lecturas. In C. Belarte et al. (eds.) Iberos del Ebro, 305–20. Tarragona: ICAC. Megaw, M. R. and Megaw, J. V. S. 2001. Celtic Art: From its beginnings to the Book of Kells, 2nd ed. London: Thames and Hudson. Moltó Gisbert, S. and Reig Seguí, C. 2000. La falcata de la sepultura 53 de la necròpolis de la Serreta i el seu context arqueològic, La falcata ibérica de La Serreta, 31–43. Alicante. Moret, P. and Rouillard, P. 1997. Les Ibères. In Les Ibères, Dossiers d’Archèologie 228, 4–13. Dijon. McCartney, M. 2012. Warfare and Violence in the Iron Age of Southern France. BAR International Series 2403. Oxford: Archaeopress.

95

Negueruela, I. 1990. Los monumentos escultóricos ibéricos del Cerrillo Blanco de Porcuna (Jaén). Madrid: Ministerio de Cultura. Pérez Rojas, M. 1983. La estela ibérica de Caspe: introducción a su estudio lingüístico, Archivo Español de Arqueología 56, 269–85. Pujol Puigvehi, A. 1979–80. Los cráneos de Ullastret y su posible significado, Pyrenae 15–16, 267–75. Py, M. 2011. La sculpture gauloise méridionale. Paris, Errance. Quesada Sanz, F. 1997. El armamento ibérico. Estudio tipológico, geográfico, funcional, social y simbólico de las armas en la Cultura Ibérica (siglos VI–I a.C.), Monographies Instrumentum 3. Montagnac: Ed. Monique Mergoil. Quesada Sanz, F. 2005. Patterns of Interaction: ‘Celtic’ and ‘Iberian’ weapons in Iron Age Spain. In W. Gillies and D.W. Harding (eds.) Celtic Connections. Papers from the Tenth Intrnational Congress of Celtic Studies, Edinburgh 1995, Volume 2, 56–78. Edinburgh: University of Edinburgh. Quesada Sanz, F. 2011. Au-delà du guerrier: signification et fonction de l’armement dans les tombes féminines et d’immatures dans la Péninsule ibérique au second âge du Fer. In L. Baray, M. Honneger and M. H. Dias-Meirinho (eds.) L’armement et l’image du guerrier dans les societés anciennes. De l’objet à la tombe, Actes Table Ronde, Sens-CEREP, 4–5 Juin 2009, 337–54. Dijon: Eds. Universitaires. Quesada Sanz, F. 2012. Mujeres, amazonas, tumbas y armas: una aproximación transcultural. In L. Prados Torreira (ed.) La arqueología funeraria desde una perspectiva de género, 317–64. Madrid: UAM. Quesada Sanz, F.; Zamora, M.; Requena, F. 2000. Itinerant smiths in the Iberian Iron Age? (6th–2nd centuries BC). In M. Feugère and M. Gustin (eds.) Iron, Blacksmiths and Tools, Monographies Instrumentum 12, 15–19. Montagnac: Ed. Monique Mergoil. Richtie, J. N. G. and Richtie, W. F. 1995. The army, weapons and fighting. In M. J. Green (ed.) The Celtic World, 37–57. London-New York: Routledge. Sopeña Genzor, G. 1987. Dioses, ética y ritos. Aproximaciones para una comprensión de la religiosidad entre los pueblos celtibéricos. Zaragoza. Sopeña Genzor, G. 2008. Acerca de la amputación de la mano diestra como práctica simbólica. El caso de Hispania en época de las guerras celtibérico-lusitanas. In F. Cadiou, M. A. Magallón and M. Navarro (eds.) La guerre et ses traces, Saldvie 8, 271–83. Sterckx, C. 2005. Les mutilations des ennemis chez les Celtes préchrétiens. Paris: L’Harmattan. Taracena Aguirre, B. 1943. Cabezas-trofeo en la España céltica, Archivo Español de Arqueología 16, 150–171. Untermann, J. 1963. Estudio sobre las áreas lingüísticas prerromanas de la Península Ibérica. Archivo de Prehistoria Levantina 10, 165–192. Untermann, J. 1995. Lengua y poblamiento prerromano en el territorio celtibérico. In F. Burillo (ed.) Poblamiento celtibérico, III Simposio sobre los Celtíberos, Daroca, 7–24. Zaragoza: Inst. Fernando el Católico. Untermann, J. 2001. La toponimia antigua como fuente de las lenguas hispano-celtas, Palaeohispanica 1, 187–218. Yadin, Y. 1963. The Art of Warfare in Biblical Lands in the Light of Archaeological study. London: Weidenfeld and Nicholson.

10 A CELTIC SEVERED HEAD, OR LAZARUS IN THE ARMS OF ABRAHAM? Natalie Venclová and Jan Royt

A few years ago Tomáš Durdík†, castellologist and medieval archaeologist, pointed out to one of the authors the find of an unusual relief sculpture of what looked like two human heads from Kněžice in the borderland between Bohemia and Moravia. He was concerned that the sculpture might be of Celtic origin, perhaps connected to the ‘head cult’. This contribution considers different concepts in Celtic and medieval (Romanesque) sculpture. The term ‘Celtic’ is used here as a synonym for ‘La Tène’ to be more intelligible to non-archaeologists.

Finds context In the 1980s, during reconstruction work on his house, no. 48 in Kněžice (Třebíč district, Czech Republic), Libor Šlouf found a re-used stone block with a relief sculpture on one side. The origin of the stone block was unknown to him. The house is located in the village known for its St James the Elder (Greater) Romanesque and Gothic church. The relief depicts a human figure of which just head and arms are visible, holding another, smaller, human head in its arms. The lower parts of the arms are missing. The lower edge of the stone block is very irregular and it could have been broken up, so that it is not clear whether it was originally of a bust or whole figure. Kněžice is located in the area of Brtnice heights, part of the Bohemian-Moravian highlands on the border of the Třebíč and Jihlava regions (SW Moravia), on the River Brtnice in altitude around 540 m. The hilly region has provided only scarce traces of prehistoric activities, mainly represented by silicite artefacts (Neolithic?) and single finds of Neolithic polished stone artefacts (Diviš and Grepl 1984, 14; Koštuřík et al. 1986, 197). Iron Age – La Tène period settlement has not been recorded there with

certainty, although ‘Celtic’ presence related to the River Brtnice, renown for its gold panning, has been proposed. Activities from this period, 3rd or early 2nd century BC, have nevertheless been recorded by a single find in the nearby town of Brtnice, before 1892, of a silver, gold-plated stater of the Nike type, an imitation of an Alexander III of Macedon coin. The find context of the coin is unknown. If the coin was indeed found in the area, then it could be related to the assumed prehistoric long-distance S-N route connecting Lower Austria with western Moravia and the East Bohemian Elbe region (Měřínský 1988, 15–16). Medieval settlement in the area was attested by archaeol­ ogical finds in the mid-13th century. Corroborative evidence is provided by history. The first historical report concerning Kněžice dates from AD 1222, when Herich (Dětřich) from Kněžice, from the Hrutov family, was mentioned. The church of St James the Elder (Greater) was built around AD 1240; its Romanesque apse was demolished in the 14th century and a Gothic presbytery built in its place. The church has been re-built several times during the following centuries. Kněžice profited from its location on the so-called Habry (Haberská) route and Lovětín route, which were important long-distance medieval routes through the Bohemian-Moravian borderland (Měřínský 1988, 24–26, 30–39). It follows that there is some (although slightly dubious) evidence for La Tène activity at the site, and at the same time a well recorded medieval setting for the stone sculpture whose origin and date are the focus of this paper.

La Tène (Celtic) art and anthropomorphic sculpture In a Festschrift dedicated to a Celtic art connoisseur it seems needless to go into details of the nature of this art; let us

10.  A Celtic severed head, or Lazarus in the arms of Abraham?

97

Fig. 10.1. Kněžice (Třebíč district, Czech Republic). A stone fragment with a pair of heads (photo N. Venclová).

just quote that ‘among the most noticeable features of Celtic art is its determined non-narrative quality’, and – as far as the subject of our contribution is concerned – ‘the human head is omnipresent, frequently mask-like ... often reduced to cartoon-like abstraction ... It has to be presumed that the head had major symbolic importance, and that there may

even have been some kind of taboo on the representation of human form ...’ (Megaw and Megaw 2001, 20–21). Indeed, La Tène art, so abundant in the decoration of a wide range of artefacts or in small metal figurines, is much less represented by large anthropomorphic sculptures in stone (wooden sculpture, however, mostly escapes archaeological

98

Natalie Venclová and Jan Royt

evidence). Up to now, just a few dozen stone statues are known from La Tène Europe of the last five hundred years BC, even if finds are slowly increasing in number. They are statues of whole life-sized human figures, busts, or images in relief. The life-sized statues represent standing warriors (or heroes, see Megaw 2003a), such as the well-known sculpture from Glauberg in Hessen, or, as a specific category, the sitting cross-legged figures occurring in south France (‘personnages accroupis’) but exceptionally also elsewhere in west Europe (Frey 2002; Guillaumet 2003; Arcelin and Rapin 2003, 204–205).  Remarkable anthropomorphic sculpture appears in the Celto-Iberian area (Höck 2002). Busts recently ascribed to druids, or heroes and ancestors, are well represented by the notorious stone head from the enclosed elite residence at Mšecké Žehrovice in Bohemia (Venclová 1998), with parallels from enclosed sites in France (cf. Menez et al. 1999; Venclová 2002 with references). Relief sculpture appears on burial stelae as early as the Early Iron Age / Hallstatt period (Frey 2002, 211–212). Besides that, remarkable depictions in low and high relief, or sometimes only incised, showing a single head or heads in rows, occurred later on architectural blocks and pillars in the Celto-Ligurian area (as demonstrated, e.g., by the best known examples from Entremont: Arcelin and Rapin 2003; Arcelin and Gruat 2003). La Tène (Celtic) sculpture can be dated according to the archaeological find context and, of course, to its art style. The sculpted heads are always more-or-less stylised: their faces are flat, with almond eyes, roughly designed trapezoidal nose, mouth only slightly indicated by an incised line, with flattened ears, and created in low relief, and with some sort of coiffure or a head-band. Figures have arms folded on their abdomen or holding some object. The abstract La Tène style is manifested, among others, in the spirally executed brow or moustache tips, or in the ornamentally depicted ear interpreted as a lotus bud, which cannot be considered a Celtic element, but which appears frequently in La Tène art, overtaken from or inspired by the Mediterranean world. A theme of its own is the hair style, in particular the ‘band’ haircut, interpreted as a type of tonsure, possibly druidic (Venclová 2002), and perhaps also some types of beard or moustache. The dating of single sculpted objects without a find context can only be based on ‘Celtic’ attributes. Classified as such can be parts of personal equipment, including the ‘leaf crown’ on the head, a torc, weaponry and gear including armour, personal ornaments or a lyre, which are analogous to objects known from well dated La Tène period archaeological finds. And what about the severed heads, or the Celtic head cult, source of much excitement to the general public? It is a fact that human skulls are occasionally found in settlement contexts, e.g. at the gates of some of the La Tène oppida

(Drda and Rybová 1995, 140), at Montmartin and Glanum in France or at Koben-Gondorf in Germany (Brunaux 2012; Haeussler 2012). In contrast, deposits of bones, possibly of warriors, at ritual sites are known, but with skulls missing, as demonstrated by the finds from the sanctuaries of Gournay and Ribemont in northern France (general overview in Brunaux and Malagoli et al. 2003). Besides heads in relief found on the stone architecture in the south of France, there are also head-shaped niches where actual skulls or mummified heads were perhaps originally exhibited, as shown by the examples from Roquepertuse or Nîmes (Arcelin and Rapin 2003, 191, fig. 6–8). Severed heads are rarely illustrated in stone sculpture: the monster from Noves holds two heads in his claws (Arcelin and Rapin 2003, 191 with references), and according to the reconstruction from Entremont, a sitting figure held several severed heads in its hands (Arcelin and Gruat et al. 2003, fig. 123). Heads/skulls could have been war trophies, or could have been related to the cult of ancestors. Of a quite different character is the cross-legged sitting person holding a head in its arms from Bouriège, perhaps dated to the late Hallstatt period (Arcelin and Rapin 2003, 206, fig. 30), but in this case it has been considered doubtful whether a trophy or some other symbol is intended. Considering the examples named above, the Kněžice heads do not show typical traits of Celtic anthropomorphic art, at least as far as their round eyes or prominent ears are concerned; and there are no Celtic attributes visible on the sculpture.

Celtic tradition in later periods Celtic traditions in the Roman period can be traced in the former Celtic territories of the Roman Empire, be it according to Celtic attributes or based on their non-Roman style (Wilburger 2012). Sitting cross-legged figures continue into the Gallo-Roman period; they can differ from the La Tène ones e.g. by their dress (Guillaumet 2003). But even genuine La Tène attributes can be of a long duration: heads or figures with torcs are still being depicted even in the Gallo-Roman period (Arcelin and Gruat 2003, 212–213, fig. 124). Celtic roots are exhibited in early medieval art, especially in the British Isles, not only in sculpture including stone crosses, but also in illuminated manuscripts (as is well expressed by the title of the volume by R. and V. Megaw Celtic art from its beginnings to the Book of Kells (1990, reprinted 2001). Celtic roots are less apparent in continental Europe, western or central. No date can be established for the bust from Gotha in Thuringia holding a head in its right hand, and the same is true for further sculpted heads both from the continent and the British Isles (Megaw 2003b, 275, fig. 5). Nevertheless, it is tempting to see some archaic elements, including Celtic, in Romanesque sculptures. But

10.  A Celtic severed head, or Lazarus in the arms of Abraham? are the superficial similarities caused by the simple execution of some sculpted works, especially in a rural milieu? Could the Kněžice sculpture represent a person holding a severed head, based on a Celtic tradition?

Figural sculpture in the Romanesque art Relatively few examples of figural sculpture have been found in Romanesque Bohemia and Moravia compared to e.g. Germany, Austria or Poland. The formal stylistic and comparative analysis is therefore quite difficult in the case of the Kněžice sculpture. The low relief was created by a sculptor working with the abstraction of human figure and face, accentuating elements important for communication with the observer. It is clear that two people are depicted, where the bigger and possibly older bearded man holds a smaller, beardless figure in his arms (in his lap?). The prominently-bearded bald head of the older man is of a triangular shape. Characteristic are big eyes, originally protruding, slightly raised superciliary ridges, and a relatively wide and short nose, as can be judged in spite of its damage. To the characteristic features of the face belong the relatively big ear lobes and wide mouth with inconspicuous lips. The second head resting in the arms or lap of the older man is smaller. It has also a triangular shape and protruding eyes. The sculpture can be approximately dated according to known parallels. Similar shape and stylistical features of the faces can be found on the figures of the saints on the tympanum fragments from Předhradí (Lapidarium of the National Museum in Prague, inv. no. 1180–1200; Merhautová and Třeštík 1983, 183), and namely on the tympanum of the St Venceslas church in Hrusice from the early 13th century (Merhautová and Třeštík 1983, 245), or in the St Gunther’s face on the tombstone from the second quarter of 13th century in the Prague-Břevnov St Margaret monasterial church. It can therefore be reasonably assumed that the Kněžice sculpture was created at the turn of 12th to 13th century. The questions are, what is the represented theme and where was the sculpture installed? Presumably the theme could be ‘Lazarus in the arms of Abraham’ (Merhautová and Třeštík 1983, 298) inspired by the evangelical parable about the rich man and Lazarus (Luke 16, 19). Relief sculptures depicting the bearded Abraham with the young and beardless Lazarus are relatively frequently part of the decoration of French Romanesque and Gothic churches (e.g. the relief sculpture on the western facade of the St Peter abbatial church in Moissac, AD 1115–1135; or the relief on the capital in the St Dennis cathedral crypt in Paris, early 13th century). The theme also occurred from the early Middle Ages in illuminated manuscripts abroad (e.g. The Codex Aureus

99

from Echternach, AD 1043–1046, Nürnberg, Germanisches Nationalmuseum) as well as in Bohemia (initial ‘S’ in Mater verborum, before the middle of 13th century, Library of the National Museum in Prague; Krása 1990, fig. 8), where it illustrates the mentioned parable. The two heads from Kněžice most probably come from the decoration of the late Romanesque apse of the local St James the Greater church, destroyed in the second half of the 14th century. The Romanesque church was built around AD 1240, as is revealed by the preserved fragments of its portal (Samek 1999, 145–146). Kněžice belonged at that time to the Želiv monastery, but it is difficult to find any common traits in the architecture, because almost nothing survived from the medieval period at Želiv. It has been suggested (Kudělka 1977–1978, 41) that the church could have been built by the same workshop which was involved in the building of the Benedictine St Procopius church in Třebíč. However, nothing of the progressive architectonic morphology representing the phase of the so-called ‘transition style’ can be found at Kněžice. The artistic stylisation of the preserved architectonic detail inside the church is in accord with the stylisation of the heads on the Kněžice fragment. This therefore could be dated to around AD 1240. If we accept that the fragment comes from the St James church in Kněžice, then it should be considered which part of the demolished apse would be most suitable for it. Similarly to the St Procopius church in Třebíč, it could have been placed outside. The Old Testament motives applied to the outer side of the apse can be paralleled with, e.g. those on the Romanesque church in the Austrian Schöngrabern (Pippal 1991), a village located on an important route to Bohemia. Although the pair of heads from Kněžice do not seem to be of Celtic origin, it does not diminish their significance. They enlarge the meagre list of Romanesque sculpture from Bohemia and Moravia.

Conclusion We could be disappointed that no new Celtic severed head sculpture was identified in Kněžice, but it is not too surprising. To recognise a single sculpture as Celtic, at least some of the appropriate cultural or art attributes should be present. However, if there are no iconographic clues, it makes you wonder: on the one hand, how many Lazaruses are in fact hidden in the sculpture assumed to be Celtic or with Celtic roots; and on the other hand, how many Celtic heads may be hidden in the assumed medieval sculpture?

Acknowledgement The research was supported by the RVO 67985912.

100

Natalie Venclová and Jan Royt

Bibliography Arcelin, P. and Gruat, P. 2003. La France du Sud-Est. In P. Arcelin and J.-L. Brunaux (eds.) Dossier – Cultes et sanctuaires en France à l’Ấge du Fer, Gallia 60, 169–241. Paris: CNRS. Arcelin, P. and Rapin, A. 2003. Considérations nouvelles sur l´iconographie anthropomorphe de l´âge du Fer en Gaule méditerranéenne. In O. Buchsenschutz, A. Bullard, M.-B. Chardenoux and N. Ginoux (eds.) Décors, images et signes de l’âge du Fer européeen, Actes du XXVIe colloque de l’Association Française pour l’Étude de l’Ấge du Fer, 183–219. Tours: FERACF. Brunaux, J.-L. 2012. Köpfe rollen. Der Kult um den Schädel. In D. Beilharz, T. Hoppe, and R. Röber (eds.) Die Welt der Kelten. Zentren der Macht – Kostbarkeiten der Kunst, 292–4. Ostfildern: Thorbecke. Brunaux, J.-L. and Malagoli, C. et al. 2003. La France du Nord. In P. Arcelin and J.-L. Brunaux (eds.) Cultes et sanctuaires en France à l’Ấge du Fer, Gallia 60, 9–73. Paris: CNRS. Diviš, J. and Grepl, E. 1984. Nálezy křišťálové a pazourkové industrie z Kněžic, Přehled výzkumů 1982. Brno: Archeologický ústav ČSAV 14. Drda, P. and Rybová, A. 1995. Les Celtes de Bohême. Paris: Errance. Frey, O.-H. 2002. Menschen oder Heroen? Die Statuen vom Glauberg und die frühe keltische Grossplastik. In H. Baitinger and B. Pinsker (eds.) Das Rätsel der Kelten vom Glauberg: Glaube – Mythos – Wirklichkeit, 208–18. Stuttgart: Theiss. Guillaumet, J.-P. 2003. Les personnages accroupis: essai de classement. In O. Buchsenschutz, A. Bullard, M.-B. Chardenoux and N. Ginoux (eds.) Décors, images et signes de l’âge du Fer européeen, Actes du XXVIe colloque de l’Association Française pour l’Étude de l’Ấge du Fer, 171–82. Tours: FERACF. Haeussler, R. 2012. Hero cults between Iron Age and Principate. In P. Anreiter, E. Bánffy, L. Bartosiewicz, W. Meid and C. Metzner-Nebelsick (eds.) Archaeological, cultural and linguistic heritage, 249–64. Budapest: Archaeolingua. Höck, M. 2002. Die “Lusitanischen Kriegerstatuen” in Nord­ portugal. In H. Baitinger and B. Pinsker (eds.) Das Rätsel der Kelten vom Glauberg: Glaube – Mythos – Wirklichkeit, 229–31. Stuttgart: Theiss.

Koštuřík, P., Kovárník, J., Měřínský, Z. and Oliva, M. 1986. Pravěk Třebíčska. Brno: Muzejní a vlastivědná společnost and Třebíč: Západomoravské muzeum. Krása, J. 1990. České iluminované rukopisy 13.–16. století. Praha: Odeon. Kudělka, Z. 1977–78. Kněžice, kostel sv. Jakuba. In Výzkum románské architektury na Moravě I, Sborník Filosofické fakulty brněnské univerzity F 21–22, 41. Megaw, J. V. S. 2003a. Celtic foot(less) soldiers? An iconographic note, Gladius 23, 61–70. Megaw, J. V. S. 2003b. Where have all the warriors gone? Some aspects of stone sculpture from Britain to Bohemia, Madrider Mitteilungen 44, 269–86. Megaw, R. and Megaw V. 2001. Celtic art from its beginnings to the Book of Kells. London: Thames and Hudson. Ménez, Y. et al. 1999. Les sculptures gauloises de Paule (Côtesd´Armor), Gallia 56, 357–414. Merhautová, A. and Třeštík, D. 1983. Románské umění v Čechách a na Moravě. Praha: Odeon. Měřínský, Z. 1988. Počátky osídlení Brtnicka a nejstarší dějiny obce. In J. Janák et al. (eds.) Dějiny Brtnice a připojených obcí, 13–49. Brno: Muzejní a vlastivědná společnost. Samek, B. 1999. Umělecké památky Moravy a Slezska, vol. 2. Praha: Academia. Pippal, M. 1991. Die Pfarrkirche von Schöngrabern: eine ikonologische Untersuchung ihrer Apsisreliefs, Veröffent­ lichungen der Kommission für Kunstgeschichte, Bd. 1. Wien: Verlag der Österreichischen Akademie der Wissenschaften. Venclová, N. 1998. Mšecké Žehrovice in Bohemia: Archaeological background to a Celtic hero. 3rd–2nd cent. B.C. Sceaux: Kronos Editions. Venclová, N. 2002. The Venerable Bede, druidic tonsure and archaeology, Antiquity 76, 458–71. Wilburger, N. 2012. Neue Herren, alte Sitten. Das Fortleben keltischer Traditionen in römischer Zeit. In D. Beilharz, T. Hoppe, and R. Röber (eds.) Die Welt der Kelten. Zentren der Macht – Kostbarkeiten der Kunst, 443–50. Ostfildern: Thorbecke.

11 ZUR ATTACHENZIER DER SCHNABELKANNEN VON BASSE-YUTZ Otto-Herman Frey

In their ‘Masterpieces of Celtic Art’, the Megaws have dealt at length with the spouted flagons from Basse-Yutz (Megaw und Megaw 1990). The relative brevity of the section discussing parallels of form and decoration for the flagons indicates how unique these objects are. The discovery of the rich undisturbed graves from the Glauberg in Hesse (Germany; Baitinger und Pinsker 2002) now provides an impressive parallel to the motives on the lower handle attachments of the two flagons.

Dem Ehepaar Megaw wird die eingehende Untersuchung der beiden Schnabelkannen aus Basse-Yutz in Lothringen: ‘Masterpieces of Celtic Art’, verdankt (Megaw und Megaw 1990). Aus der relativ kurzen Behandlung von Vergleichen zur Formgebung und zum Dekor ist ersichtlich, wie singulär die Fundstücke sind. Durch die Entdeckung der ungestörten reichen Gräber vom Glauberg in Hessen (Baitinger und Pinsker 2002) gibt es jetzt aber eine schlagende Parallele zum Motiv der unteren Henkelattachen der beiden Kannen, auf die hier näher eingegangen werden soll (Fig. 11.1). Bei dem Glauberger Komplex geht es um die der Röhrenkanne aus Grab 2. Ihre untere Attache nimmt ein Gesicht ein, nach dem Schnurrbart zu urteilen das von einem Mann (Baitinger und Pinsker 2002, Kat.-Nr. 2.1, Abb. 253). Nach oben wird die Stirn horizontal von einem astragalierten Band begrenzt, das an den Enden zu den ‘Ohren’ abknickt. Daran hängt je eine durch die Musterung wie geflochten wirkende Locke, die unten nach außen eingerollt ist. Dort werden beide in einer Dreiblatt-Palmette zusammengeführt. Das ganze Gesicht wird dadurch eng umfangen. Über dem Kopf hängt ein Paar von Blättern, ihre Spitzen sind auseinandergespreizt, die Stängel wieder nach außen eingerollt. Sie begegnen sich in einem weiteren, von einer Kugel überragten Blatt, das zwischen ihnen aufwächst. Dieses Bildschema wiederholt also fast genau dasjenige der Lothringer Kannen. Die Augen der Gesichter sind allerdings

nicht mandelförmig, sondern für Einlagen wie kreisrunde Näpfe geformt; das gleiche trifft auf den Ort der Palmette unterhalb der Locken zu, die durch eine entsprechende Fassung für Einlagen ersetzt ist. Etwas differiert die Bekrönung über den Häuptern, welche allerdings eine einander entsprechende Bewegung ausdrückt. Paul Jacobsthal sah auf den Henkelattachen ein Haupt mit Hathor-Locken, ein Motiv, das aus Ägypten via Phoenikia in die orientalisierende Phase der etruskische Kunst einging. Dort erscheint beispielsweise die ganze Göttin zusammen mit dem Löwentöter, doch ebenfalls als einzelne weiblichen Gestalt oder nur als Kopf (Jacobsthal 1944, 15f., 21). Solche Darstellungen sind aber fast zweihundert Jahre älter als unsere einander ähnelnden keltischen Beispiele (Bonfante 1975, 70f. Man vergleiche z. B. die Figuren der Goldohrringe aus der tomba Regolini – Galassi, Cerveteri, Bonfante 1975, fig. 56). Denn in Etrurien wird die Frisur der vollen Locken schon früh im 6. Jahrhundert in einzelne herabfallende Strähnen oder offene Haare aufgelöst. Als ein Modell für die keltischen Künstler, das den Zeitabstand von fast zweihundert Jahren überbrücken sollte, hatte Jacobsthal auf die Kopfdarstellungen eines etruskischen Halsschmucks in New York hingewiesen, bei dem die Haarangabe entfernt vergleichbar wirkt. Ein treffenderes Beispiel erwarb B. B. Shefton für das Museum in Newcastle-upon-Tyne, einen Henkel mit einem weiblichen Kopf mit dicker Haarlocke, den ich bereits in dem Katalog der Keltenausstellung von 1980 in Hallein wiedergeben durfte (Frey 1980; vgl. auch Megaw und Megaw 1990, Pl. 32. Shefton zog bereits Verbindungen zu keltischen Werken wie dem Goldfingerring von Rodenbach. Da ich damals nach Übersendung eines noch unfertigen Manuskripts für die Ausstellung einen schweren Unfall hatte, wurde dieses ohne die nötigen Anmerkungen und Nachweise gedruckt). Leider gelang es

102

Otto-Herman Frey

Fig. 11.1. Henkelattachen der Schnabelkannen von Basse-Yutz.

mir bislang nicht, dazu weitere Parallelen in Mittelitalien zu finden. Wichtiger als eine formale Ableitung des Motivs ist mir aber die Frage: was für eine Vorstellung konnten bei den Zeitdifferenzen die Kelten mit dem ins Männliche umgedeuteten Bild verbinden? Als ein mitteleuropäisches Beispiel für die Lockenfrisur nannte Jacobsthal bereits den Fingerring aus dem ‘Fürstengrab’ von Rodenbach, worin ihm die Megaws folgten. Diese bildeten dazu auch den Goldschmuck aus

Ferschweiler ab (Megaw und Megaw 1990, Pl. 29b; siehe auch Jacobsthal 1944, no. 30a). Wahrscheinlich ließen sich noch einige weitere, doch nicht so klare Zeugnisse aus der keltischen Welt anschließen, u. a. die untere Attache der Schnabelkanne aus dem Kleinaspergle mit allerdings kürzeren Locken(?) (Jacobsthal 1944, no. 385; Megaw und Megaw Pl. 31b), auf die und andere Jacobsthal ebenfalls hinwies. Immer handelt es sich um Köpfe von Männern und nicht um solche von Frauen. Und insgesamt ist die

11.  Zur Attachenzier der Schnabelkannen von Basse-Yutz Liste, wenn auch durch den Beleg vom Glauberg vergrößert, kurz. Ist mit solchen Bildern aus reichen Funden etwas besonderes gemeint? Jacobsthal ging es bei seinen Untersuchungen stets zuerst um greifbare formale Analysen, von denen aus er vorsichtig die Ausdeutung möglicher Inhalte vornahm. Jedoch sprach er bei ganzen Mischwesen oder bei Köpfen, ohne gewisse Unterschiede zu verfolgen, allein von ‘beasts and masks’ (Jacobsthal 1944, 161f.). Solche Häupter finden sich, z. T. nur angedeutet, auf verschiedenen Objekten in der ganzen frühen keltischen Welt von der Champagne bis hin zur ungarischen Grenze; dabei stammen die ältesten Belege aus einem weiten Kreis um den Mittelrhein (Am häufigsten sind Ringe mit solchen ‘Masken’, vgl. Lenerz-de Wilde 2006. Gewisse Vorarbeiten zu dem Thema stammen bereits von Vincent Megaw 1965–6, 1967 und 1970). Jüngere Autoren haben vor allen die apotropäische Bedeutung der Bilder hervorgehoben (vgl. besonders Pauli 1975; siehe auch Echt 2004. Lenerz-de Wilde, 2006, 350 denkt ebenfalls an apotropaia). Jacobsthal sah durchaus auf Grund ihrer Herkunft gewisse Abweichungen in den Darstellungen (Jacobsthal 1944, 22ff.). Etliche ‚Masken’ tragen z. B. spitze Tierohren. Wird dadurch ihre Abkunft von antiken Silensköpfen bezeugt? (Dazu Jacobsthal 1944, 21f., der zu der Attache der Kanne aus dem Kleinaspergle solche importierter etruskischer Stamnoi vergleicht. Ebenso Megaw 1965–6 und andere). Hatte der keltische Künstler, als er auf der Glauberger Kanne Grab 1 vor der Figurengruppe vom Mündungsrand auf jeder Seite einen solchen Kopf platzierte, an eine Ausgestaltung ähnlich etruskischen Werken gedacht? (Baitinger and Pinsker 2002, Kat.-Nr. 1.1, Abb. 236). Und verband er mit solchen ‘Masken’ bestimmte Ideen? Klarer könnten La Tène A Figuren mit einer Blattkrone interpretierbar sein (Jacobsthal 1944, 4, 8f, 15ff. Ausführlich hat sich mit dem Zeichen Lambrechts 1954 auseinandergesetzt. Haffner 1999 konnte auf Grund der Darstellung eines Hauptes als Knauf eines Schwertes aus Bescheid wieder für eine besondere Frisur plädieren). Wir kennen das Zeichen von einer Anzahl steinerner Statuen. Im Grab 1 vom Glauberg ist dieses ‘vergöttlichende Symbol’ auch als eine reale Beigabe des Toten aus Draht, Stoff und Fell nachgewiesen (zuletzt Herrmann 2008–9). In der Kleinkunst erscheint es mehrfach etwa bei Köpfen auf Gürtelhaken (Jacobsthal 1944, no. 354), aber bisher nur einmal an einer Maskenfibel (Kruta 1989). Bei früherer Gelegenheit hatte ich auf das über­ einstimmende Konzept des figürlichen Schmucks aus ‘beasts and masks’ der drei Röhrenkannen aus Reinheim, Waldalgesheim und ebenfalls vom Glauberg hingewiesen (zuletzt dazu Frey 2012, 37f.). Bei allen drei spiegelt sich darin ein deutlich bestimmbares Programm und keine allgemeine Aussage. Trifft entsprechendes nicht auch auf die sogenannten Hathorlocken von unseren keltischen Werken zu, auch wenn wir dieses spezielle Zeichen noch

103

nicht richtig enträtseln können? Wenn man solche frühen Darstellungen als eine ‘Bilderschrift’ oder ‘Bildersprache’ auffasst, dann sollte man auch Vergleiche zur ‘Semantik’ als Teildisziplin der Linguistik ziehen, in der die besondere, präzise Bedeutung eines Ausdruckes, bzw. hier bildlichen Gedankens angesprochen wird. Damit hätten wir schon die Schwelle zum genaueren ‘Lesen’ unserer Bilder erreicht, auch wenn wir noch um eine tiefergehende Deutung kämpfen müssen! Unmißverständlich ist die Botschaft anderer Zeichen: evident ist etwa bei dem sogenannten Keulenmotiv, welches Haffner im Raum um den Rhein herausgestellt hat, dass die zugehörigen Häupter nur göttliche Wesen sein können (Haffner 1992, 99ff.). Und vor allem sei die Wiedergabe des ‘Herrn der Tiere’ hervorgehoben, wie er auf der Schnabelkanne vom Glauberg als ganze Figur klar identifizierbar ist (zuletzt dazu Guggisberg 2010). Wir haben für ihn aus Mitteleuropa mehrere Belege auf Gürtelhaken, Ringen und weiteren Objekten, bei denen der Gott allerdings wieder nur auf einen Kopf begrenzt ist. Ferner kennen wir, doch aus einem anderen Bereich, nämlich vom Südalpenrand, bzw. aus Oberitalien, Gürtelhaken, bei denen eine solche Gottheit nur durch ein relativ kleines ‘Hallstatt’– Männchen symbolisiert wird. Wenn wir demnach annehmen, dass der Attachenzier der Kannen von Basse-Yutz bereits eine dezidierte Vorstellung zu Grunde liegt, dann müssen wir auch fragen: was ist am gleichen Objekt der Sinngehalt der Figuren auf dem Mündungsrand und vom Henkel? (für die ganze Komposition vom Vogel am Schnabelende bis zum Henkeltier zusammen auf einem Foto siehe Jacobsthal 1944, no. 381, Pl. 181 unten). Müssen wir an persische Vorbilder denken? Die lange Liste von Wissenschaftlern, die sich mit der Frage auseinandergesetzt haben, wird ausführlich von den beiden Megaws referiert (Megaw und Megaw 1990, 43, 54ff.). Sie selbst lehnen aber eine solche Übernahme ab und begründen das wie D. R. Castriota (1981) mit der ‘absence of any specific ... evidence for contacts between Persians or Scythians and Celts at the beginning of the La Tène period’. Sie denken wie er nur an eine keltische Komposition mit Anregungen aus dem Orient via Italien. Entsprechendes sollte für die Übernahme von Motiven aus dem thrakischen Gebiet zutreffen. Kürzlich hat aber Teleaga bereits eine Frühstufe der thrakischen Kunst im 5. Jahrhundert herausgearbeitet (Teleaga 2012). Entsprechend möchte Pare Details und einzelne Motive innerhalb der frühen keltischen Kunst aus Thrakien herleiten (Pare 2012, 26). Allerdings will ich der Einschätzung der beiden Megaws insofern Recht geben, als auch ich in der ruhigen Kontrastierung der verschiedenen Tiere eine keltische Komposition sehe. Was meinen die kleine Ente vorne auf dem Ausguss und die Hunde (oder Wölfe) auf dem Rand, die in deutlich entspannter Haltung den Vogel beäugen? Soll die Ente einen Abkömmling des ‘Hallstattvogels’ symbolisieren? Ist es nur die Konstellation

104

Otto-Herman Frey

verschiedener aussagekräftiger Gestalten, die hier gleichsam zu einer Geschichte zusammengerückt sind? Zu einem solchen Präsentation kann ich nur ein weiteres Beispiel nennen. Auf der mit Gravuren verzierten Schwertscheide aus Grab 2 vom Glauberg sind Hunde und wilde Raubtiere einander gegenübergestellt (z. B. Frey 2004, fig. 4). Sie stürzen hier aufeinander zu; die wütigen Raubtiere mit aufgerissenen Mäulern und spitzen Zähnen sind verrenkt und verzerrt. Es kann sich nur um die Darstellung zweier Kräfte handeln, von denen die eine mit den Menschen verbunden ist, die andere die ungebändigte Natur spiegelt. Es wird in dieser Komposition von keinem Mythos berichtet, wie ja auch die Bilder der importierten, erzählenden griechischen Vasen im Norden keine Nachahmungen fanden. Es ist allein ein beherrschender Gegensatz wiedergegeben, der in der spirituellen Welt der Kelten eine Rolle spielte. Es sind solche Bilder, die den vollzogenen Wandel zur vorausgehenden Hallstattkultur markieren, welche sich merkwürdigerweise aber auch nicht in der Latènekultur kontinuierlich weiterentwickeln. Warum? Das kann hier nicht ausführlich hinterfragt werden. Doch hoffe ich, dass meine wenigen Interpretationsanregungen zu dieser ersten Phase der Latènekultur das Interesse von Vincent Megaw als eine Gratulation zu seinem Geburtstage treffen werden.

Bibliography Baitinger H. and Pinsker, B. (eds.) 2002. Das Rätsel der Kelten vom Glauberg. Glaube – Mythos – Wirklichkeit, Ausstellungskatalog Frankfurt 2002. Stuttgart: Theiss. Bonfante, L. 1975. Etruscan Dress. Baltimore/London: Johns Hopkins University Press. Castriota, D. R. 1981. Continuity and Innovation in Celtic and Mediterranean Ornament: A Grammatical-syntactic Analysis of the Processes of Reception and Transformation in the Decorative Arts of Antiquity. Columbia: Columbia University Press. Echt, R. 2004. Äußerer Anstoß und innerer Wandel – drei Thesen zur Entstehung der Latènekunst. In M. Guggisberg (ed.) Die Hydria von Grächwil. Zur Funktion und Rezeption mediterraner Importe in Mitteleuropa im 6. und 5. Jahrhundert v. Chr. Akten Internat. Kolloquium Bern 2001, 203–15. Bern: Bernisches Historisches Museum. Frey, O.-H. 1980. Die keltische Kunst. In L. Pauli (ed.) Die Kelten in Mitteleuropa: Kultur, Kunst, Wirtschaft, Salzburger Landesausstellung Keltenmuseum Hallein 1980, 76–92, bes. 81, Abb. 11–12. Salzburg: Amt der Salzburger Landesregierung. Frey, O.-H. 2004. A new approach to Early Celtic Art, Proceedings of the Royal Irish Academy 104 C, Nr. 5, 107–29. Frey, O.-H. 2012/13. Keltisches Denken im Spiegel der

frühkeltischen Kunst. Zu den Kleinfunden vom Glauberg, Berichte der Kommission für Archäologische Landesforschung in Hessen 12, 31–56. Guggisberg, M. 2010. The Mistress of Animals, the Master of Animals: Two Complementary or Oppositional Religious Concepts in Early Celtic Art? In D. B. Counts und B. Arnold (eds.) The Master of Animals in Old World Iconography, 113–36. Budapest: Archaeolingua. Haffner, A. 1992. Die frühlatènezeitlichen Fürstengräber von Hochscheid im Hunsrück. Trierer Zeitschrift 55, 25–103, bes. 99ff. Haffner, A. 1999. Ein Frühlatèneschwert mit anthropoidem Knauf von Bescheid, Landkreis Trier-Saarburg. In B. Chaume, J. P. Mohen und P. Périn (eds.) Archéologie des Celtes. Mélanges à la mémoire de R. Joffroy, Protohistoire Européenne 3, 123–9. Montagnac: Merigoil. Herrmann, F. R. 2008–9. Profanes und Sakrales vom Glauberg, Berichte der Kommission für Archäologische Landesforschung in Hessen 10, 245–53. Jacobsthal, P. 1944. Early Celtic Art. Oxford: The Clarendon Press (reprint 1969). Kruta, V. 1989. La fibule ‘à masque’ du gué de Port-à-Binson (Marne), Études Celtiques 26, 7–22. Lambrechts, P. 1954. L’exaltation de la Tète dans la pensée et dans l’art des Celtes, Dissertationes archaeologicae Gandenses 2. Burges: De Tempel. Lenerz-de Wilde, M. 2006. Frühlatènezeitliche Ringe mit Maskenzier, Germania 84, 307–68. Megaw, J. V. S. 1965–6. Two La Tène finger rings in the Victoria and Albert Museum, London: an essay on the human face and Early Celtic Art, Prähistorische Zeitschrift 43–4, 96–166. Megaw, J. V. S. 1967. Ein verzierter Frühlatène-Halsring im Metropolitan Museum of Art, New York, Germania 45, 50–9. Megaw, J. V. S. 1970. Further Early La Tène rings and other material of the ‘Hochheim’ and ‘Andernach’ classes, Germania 48, 126–30. Megaw, J. V. S. und Megaw, M. R. 1990. The Basse-Yutz find: Masterpieces of Celtic Art, Reports of the Research Committee of the Society of Antiquaries of London 46. London: Society of Antiquaries. Pare, Ch. 2012. Eastern relations of Early Celtic Art. In Ch. Pare (ed.) Kunst und Kommunikation. Zentralisierungsprozesse in Gesellschaften des europäischen Barbarikums im 1. Jahrtausend v. Chr., RGZM-Tagungen 15, 153–78. Mainz: Verlag des Römisch-Germanischen Zentralmuseums. Pauli, L. 1975. Keltischer Volksglaube. Amulette und Sonderbestattungen bei Hallein und im eisenzeitlichen Mitteleuropa. München: Beck. Teleaga, E. 2012. Der Beginn der figuralen thrakischen Kunst im 5. Jahrhundert v. Chr. In Ch. Pare (ed.) Kunst und Kommunikation. Zentralisierungsprozesse in Gesellschaften des europäischen Barbarikums im 1. Jahrtausend v. Chr., RGZM-Tagungen 15, 123–51. Mainz: Verlag des Römisch-Germanischen Zentralmuseums.

12 THE NOT SO UGLY DUCKLING – AN ESSAY ON MEANING Flemming Kaul

Introduction It is a great honour and a pleasure to contribute to this Festschrift on the occasion of the 80th birthday of J. Vincent Megaw. Together with his wife, Ruth, he has made great contributions to our understanding of Celtic art. In what follows we shall take a closer look at the famous Basse-Yutz flagons which for decades have fascinated the Megaws as well as other prominent researchers of Celtic art. When it comes to the meaning of prehistoric art, different opinions have been expressed through the history of research. Let us take a look at the Scandinavian Bronze Age situation, also of relevance when considering early La Tène art. The leading Danish archaeologist J. J. A. Worsaae (1821–1885) regarded the images of ships on bronze objects and on rock carvings, as well as circular designs, fish, snakes and horses connected with the ships, as religious images related to a solar cult (Worsaae 1882, 93–96). Worsaae’s successor at the National Museum of Denmark, S. Müller (1846–1934), was much more restrained in interpreting the imagery of the Bronze Age. He did not consider such pictures as symbolic representations with a religious content (Müller 1897, 419–420). When it came to miniature art on bronze objects, Müller presented some arguments based on specific concepts of art. The art on bronzes was called decorative art or ornamental art (Danish: prydkunst). And decorative art in this context had no meaning, apart from being decorative, being ornament. When a ship is seen on a razor, then it is because the ship shape is simply a nice decoration (Müller 1897, 352–354; Kaul 1998,

69). Furthermore Müller underlines that when a pictorial motif, such as the ship, became increasingly stylized then it had totally lost any meaning, merely being an ornament. The same goes for horse motifs and zoomorphic motifs: when a horse figure had totally disintegrated as a result of repeated artistic treatment it had no meaning. (Müller 1920, 139–140). Also when a motif is transferred from one area to another it could lose its (original) meaning. When the aquatic bird in Period IV of the Nordic Bronze Age (1100–900 BC) found its place on a razor’s handle it had totally lost the religious significance which it had in the South, for instance in Greece. In the North, these birds became merely decorative motifs (Müller 1897, 346–347). By claiming that by stylization or transfer of a motif the art became meaningless, being solely decorative, Müller consequently dissociated himself from religious interpretations. When the Chariot of the Sun (c.1400 BC) came to light in Trundholm Bog, Zealand, Denmark in 1902, things changed a bit, and Müller wrote that the find threw new light on spiritual life and religion in the distant past (Müller 1903, 303). The Chariot of the Sun rendered intelligible the idea that the non-personified sun was pulled over the heavens by a divine horse. The myth (of the sun horse pulling the sun) had been put on wheels in order to represent the movement of the sun in a ritual (Müller 1903, 114; Kaul 2010, 523–524). It is important to note that the horse figure was considered ‘real’ art – a true plastic piece of sculpture. And Müller maintained his ideas of the meaninglessness of what he considered as decorative art long after the find of the Chariot of the Sun

106

Flemming Kaul

– on, for instance, razors. This find could not change his notions of decorative art being merely decorative. In the 1930s an interest in religious interpretations was increasing. For instance, the Danish archaeologist J. Brøndsted reached the conclusion that beliefs about the journey of the sun and its various means of transportation were significant features in Bronze Age iconography as expressed on razors (Brøndsted 1938, 81–100). In the 1930s a corresponding interest arose in Germany in understanding the pictures on the bronzes as being of a religious nature. The interplay between ship and horse as the sun’s means of transport was the focus (JacobFriessen 1934; Sprockhoff 1936). Jacob-Friessen’s work was an iconographical analysis of the motifs on some richly decorated razors from Niedersachsen, and the religious and mythological background for the pictures was considered as obvious. He suggested that a comprehensive treatment of all the decorated razors should give further insight into the mythology of the voyage of the sun (Jacob-Friesen 1934, 375). In the works of Sprockhoff (1954) and Gelling and Davidson (1969), ideas about the voyage of the sun were further developed and discussed. Due to studies by the present author, where more than 400 Danish Late Bronze Age objects carrying ship images have been analyzed, it has been possible to reveal what seems to be a full cyclical myth of the daily and nightly voyage of the sun – the horse working together with other zoomorphic agents as well as the ship (Kaul 1998; 2004; 2010). During the 1970s, 1980s and also parts of the 1990s, interest in religious interpretations of the Nordic icono­ graphical material seemed to have been dwindling. More materialistic approaches were favoured, and ritual and religion were subjects of low scientific value on the archaeological agenda (Jennbert 2000, 127). For instance the Swedish archaeologist M. Malmer preferred to use non-mythological explanations for the ship-motif on rock carvings, which he first and foremost considered as a symbol of wealth and power (Malmer 1989, 96).

Early La Tène art Even though the present author does not agree with S. Müller’s idea (see above) that the aquatic bird, when being transferred to the North during the Bronze Age, lost its mythological meaning, it is still worth considering a possible loss of meaning when a motif ‘travels’ from one cultural region to another. However, it is not so much a total loss of meaning we should consider, but rather a change or addition of meanings when a motif or a pattern was transferred. Such discussions are still most relevant today, also regarding Early La Tène art. For instance when dealing with the possibilities of interpreting figural motifs of the early La Tène Schnabelkanne from Dürrnberg, Austria,

the question is asked whether the Celts saw the figural representation on imported objects merely as interesting exotic motifs – or whether also a transfer of myths and beliefs took place (Moosleitner 1985, 92) – an imitation with or without transfer of meaning? A similar question was asked by Paul Jacobstahl, though he stated that we cannot answer it (1944, 36). Beginning with the gold rings from Erstfeld, Switzerland, it has been suggested that some of the motifs and com­ positions of Early La Tène art could reflect a transmission of the Mediterranean / Eastern tradition of the master or mistress of the animals, the Potnia Theron. But the requirements of a free Celtic creativity mean that the prototypes, Etruscan or Greek, have been subject to certain alterations. The figures were partly ‘cut up’ and put together in a new way, and the Celtic artist found quite different expressions of style (Guggisberg 2000, 258–259). It is very unusual among the Celts to find the motif of ‘the mistress of the animals’ in its classical symmetrical or antithetic appearance. Celtic figural art should reflect its own world of beliefs, where only certain elements were adopted (ibid. 262). When we find a number of similarities, with Etruscan art in particular, then we should not see these similarities as reflecting a huge wave of diffusion, but as a deliberate process, where certain elements were picked up, according to Celtic needs or wishes, whilst other elements were rejected. L. Pauli, among others, has taken up these questions, also considering ‘the master of the animals motif’ as something still having a deep meaning when appearing remodelled in Early Celtic art. Pauli does not exclude that other gods were taken over by the Celts at this changeable time in European history (Pauli 1992, 135–140). The plastic horse figure standing on the lid on of the Reinheim flagon, South Germany, yields a fine example of how elements of Mediterranean iconography where reshaped into something particularly Celtic. The horse figure with a human head (also carrying the Celtic leaf-crown) resembles the Greek Centaur, but it is incorrect to view it as a variant of this, since the Greek centaur has a full torso and extra arms in human shape. Even though both are mixed horse-human creatures: ‘Le monstrueux cheval à tête human est une invention purement celtique’ (Kruta 2012, 44). According to Kruta, the horse figure had deep European roots, being a representation of a solar divinity of cyclical nature. In the Hallstatt culture in particular, the horse figure appears in connection with the aquatic bird, another animal to be considered as a representation of the solar divinity. When the horse, or the horse in association with a human head and floral ‘ornaments’, appears later on Celtic coins, in certain cases on coins imitating Greek Apollo coins, then we are still facing a manifestation of something profoundly divine (ibid. 45 ff.). And when the horse motif became stylized, almost beyond recognition, then there was seemingly no loss of meaning (Kaul 2002).

12.  The not so ugly duckling – an essay on meaning Also, what could be considered as ‘ornaments’ could have a meaning, or other meanings may be added in the process of transfer. For instance, certain lotus or palmette ornaments were somewhat remodelled by the Celtic artist, showing almost elusive human-like masks (Kruta 1986). This would probably have given this motif a new and widened meaning, perhaps with reference to divinities in an unseen world. In other cases the lotus ornament seems to form a muchstylized full humanlike figure, or the lotus ‘ornament’ may form the body for a humanlike mask, as on a mounting from an Early La Tène burial at Weiskirchen, Saarland, Germany (Rapin 2002). The possibility should not be excluded that the lotus shape here could indicate a divine manifestation or presence, partly floral, partly anthropomorphic. But let us not forget the old (basic) meaning of the lotus, even though this meaning might have been ‘lost in translation’ while being used as a decorative ornament in the Mediterranean world for centuries. In ancient Egyptian mythology the lotus was a symbol of rebirth, and in particular the rebirth of the sun every morning, but also the birth of the sun in creation myths. This solar connection may have derived from the characteristic tendency of the lotus to grow out of the water, opening its petals in the morning and closing them again at night (Shaw and Nicholson 1995, 164–165). In the creation myths, the leaves opened and revealed the sun, which sat as a newborn baby in the lotus flower. The young sun god can be depicted as a child in a lotus flower. Perhaps such meanings related to the rebirth of the sun, and rebirth in a general sense, should be considered even when the lotus in a most stylized shape occurs as an ornamental border, for instance in Etruscan burial paintings. The discussion as to the possible meaning of decoration is ongoing. When considering the decoration of later British La Tène bronze mirrors it has recently been pointed out that the abstract motifs did not represent anything else, they did not have an underlying symbolic meaning. The hatched motifs were placed there in order to, in a decorative manner, underline the values of a mirror, reflecting light (Joy 2008, 94). Here, it seems however also to be a matter of choice which art concept one uses. Where is the borderline between abstract art and figural art, when the figural motifs by their artistic treatment have been stylized at or beyond the limits of recognition – when it seems possible to find anthropomorphic or zoomorphic motifs, such as the aquatic bird or a bird of prey, hidden in the decorative scrolls of the mirrors? Did the ‘abstract art’ then not incorporate a specific meaning, or layers of meanings, a visual communication of certain religious ideas? Where do we find J. V. S. Megaw – or the Megaws – in such discussions? Megaw does not belong to those who deny that most of La Tène art had a meaning far beyond aesthetics. That includes what we could call abstract or decorative art, though it must be admitted that the borderlines between different art concepts do not seem

107

to work well when considering prehistoric art (Megaw and Megaw 1989, 16). For Megaw, Celtic art is basically religious (Megaw 2002, 57). But the main problem is that we cannot grasp the meaning of the art, simply because we are not in direct communication with its makers. Even when accepting that most of La Tène art decorative scrolls had a religious meaning – such motifs representing a form of visual communication – this communication is only partially accessible to us. We cannot tell the precise meaning of even some of the commonest motifs (Megaw and Megaw 1989, 19; 2008, 41). When dealing with figurative art, with humanlike – or partly anthropomorphic – motifs, it may be difficult to find a firm division between deities and mortals, princes and gods, princesses and goddesses, though the human head or mask – often being mysteriously disguised – could be regarded as a symbol of divinity (Megaw 1970b, 269; 2002, 61). Even though it is stated that (more specific) interpretations must remain speculative, it is not excluded that for instance the rings from Erstfeld may have represented myths or stories, or played some exorcizing role, since for the Celts there were obviously beings both to worship and to fear (Megaw and Megaw 1989, 93). Megaw seems to be unconvinced by attempts to interpret Early La Tène on the basis of the much later ‘Pantheon’ of Celtic gods from the Roman period and later (Megaw and Megaw 1989, 22–23; Megaw 2002, 61). This is also the case when it has been suggested that Early Celtic art could demonstrate a take-over of larger belief systems from the Italo-Greek world (Megaw 2002, 61). I fully agree that one should be extremely cautious when using much later and geographically distant material for a direct explanation of religious systems. When using chronologically much later material it seems like the interpreter works with time as though time actually went backwards. On the other hand, according to Megaw, we should not fall into hopeless scepticism – we should not be moving towards a position which maintains that matters of meaning in Prehistory are beyond the bounds of possibility (ibid.). We shall still try to explain the iconography, not leaving the study of Celtic art to ultimately sterile stylistic analyses, to lists and distribution maps (ibid. 63). Below I shall try to follow that explanatory path.

Basse-Yutz. The not so ugly duckling – reading it the Bronze Age way The Basse-Yutz find with the two famous flagons (c.450–400 BC), among the highlights of the British Museum, comes from a princely grave found in Lorraine, France, in 1928. The little ducklings, on each of the spouts of the flagons, a rare example of Celtic realism, will be the focus of consideration here (Fig. 12.1). In the comprehensive publication of the Basse Yutz

108

Flemming Kaul

Fig. 12.2. Aquatic bird stems of bird-sunships on a Late Bronze Age Hajduböszörménu-bucket found at Siem, Jutland, Denmark. Photo F. Kaul. Fig. 12.1. The little duckling on the spout of one of the Basse-Yutz flagons. The eyes were originally filled with coral studs. Photo F. Kaul.

find by Vincent and Ruth Megaw it is pointed out that the ducks belong to a bird imagery which extends from Hallstatt and Villanovan Bronze Age prototypes (Megaw and Megaw 1990, 43). Accepting that the duckling belongs to the European aquatic bird tradition gives consequential implications as to the understanding of its rôle in the La Tène period. The aquatic bird was among the most popular iconographical motifs of Late Bronze Age Europe. From Scandinavia in the North to Italy and Greece in the South we meet this waterbird in a many compositions, on flat surfaces of bronzes, or as plastic figures (Sapouna-Sakellarakis 1978; Kaul 1998; Kossack 1999; Sestieri and Macnamara 2007). The aquatic bird does not need to represent one specific species, though sometimes it may be identified as a swan with S-shaped neck, a duck, or perhaps a grebe. From its iconographical context it is clear that it is related to the sun and the eternal voyage of the sun. It is seen close to solar representations, or it is itself adorned with solar images. It is often seen as an integral part of the solar barque, being the stems of this divine ship. On, for instance, the Central European Hajduböszörménu-buckets (Fig. 12.2) the sunships with aquatic bird stems are carrying the sun (Patay 1990). Whereas in Scandinavia and North Germany the rich Bronze Age iconography on the bronzes disappeared at the beginning of the Pre-Roman Iron Age, around 500 BC, a new, rich iconography emerged in central and western Europe – Early La Tène art. Even though many elements of La Tène art are quite different from Late Bronze Age and Hallstatt Period art, some surviving elements can be traced. The old bird-sunship (Vogel-Sonnenbarke) and the aquatic bird are among these motifs. The old sun-ship motif is most clearly seen on a neck-ring from Attencourt

in Haute Marne, northern France (Fig. 12.3), the stems of the ship seemingly being a duck, the ship itself carrying the sun (Sprockoff 1955; 1969; Kruta 1986). On other Early La Tène rings, from northern France – Catalauni, Breuvery and Pogny, all Marne – plastic birds and ducks occur related to sun images (Jacobsthal 1944, nos. 239 and 240). Sprockhoff (1969, 24) concludes that the bird-sunship of the Urnfield Culture appears in the classical Early La Tène culture of the Marne area in such an unmistakable form that the ideas behind it should be considered as still perfectly alive. Also in the southern periphery of the La Tène culture we find the survival of the sun-ship. At the lower attachment of the handle of a beaked flagon from Valeria di Borgovico near Como, North Italy, the sun ship with stems in the shape of bird’s heads is seen, in this case perhaps a cock, a new animal in Central European iconography (Pauli 1992, 140). A monstrous animal (head) seems to threaten the ship. The flagon is regarded as a local imitation of Etruscan flagons, belonging to the 4th century BC, and the motif is understood as a sun ship – barca solare (Rapi 2009, 32). Other similar flagons from the same South Alpine region, such as the flagon from Ticino, Switzerland, show such stylized ship images in the same place (Jacobsthal and Langsdorff 1929, 57 and nos. 123–126). It is with this background that we can allow ourselves to interpret the motifs of the Basse-Yutz flagons, and we should also consider that Megaw and Megaw (1989, 78; 1990, 43) had pointed out that the duck is a representative of the waterbird whose artistic ancestry stretches back to the Bronze Age. In 1970 J. V. S. Megaw (1970a, 69) mentioned that the little duck was just a simple expression of a neatly observed bit of nature. Later Megaw and Megaw gave more room to interpretative remarks. It was mentioned that we could be dealing with even a half-incubated duckling – from its large eyes high crown and thin bill – and the punch-marks on the body may represent the papillae from which the feathers will

12.  The not so ugly duckling – an essay on meaning

109

Fig. 12.3. Detail of an Early La Tène neck ring from Attancourt, Haute Marne, decorated with bird-sunships, and detail of a neck ring from Catalauni, Marne, with aquatic birds related to a sun image. After Sprockhoff 1955.

eventually emerge. An interpretative remark is important: ‘The positioning of an embryo duck on a river of wine could perhaps be seen as an intriguing reference to the lifegiving powers of alcohol…’ (Megaw and Megaw 1990, 43). Whether being an embryo duck or a newly hatched duckling we are seemingly dealing with a ‘new-born’ creature, a most important observation. According to Megaw and Megaw the three canines (dogs?) behind the duckling could be pursuing it or perhaps guarding it (ibid. 44). But – ‘Unconcerned by the crouching canines, a cheerful duck swims on the river of wine in the spout’ (Megaw and Megaw 1989, 78). Recently the same ‘scene’ on top of the flagons has been treated along similar interpretative lines: ‘All three dogs have their attention focused on a tiny bronze duck that sits right at the end of the spout… When somebody poured from the flagon, it would look as though the duck was swimming on a stream of wine, mead or beer’ (MacGregor 2010, 181). Such primary interpretations, being close to a description of the pictorial evidence, make a good point of departure for further analysis. It should be underlined that we are dealing with a young animal. Furthermore the duckling immediately gives an impression of being something positive, being ‘cheerful’ (Megaw and Megaw 1989, 78; 1990, 43) or ‘funny’ (MacGregor 2010, 181). On the other hand the three canines, dogs, if not wolves, could express something negative, beastly and unfriendly. They are pursuing the little duckling. But it seems like the duckling can swim happily away, unharmed by the attacking beasts. And it is at this point of our interpretative process that we realize that the whole design, shape and function of the flagons in a masterly and most creative way melt together with the pictorial narrative. When the flagons were used, probably for pouring wine or some other liquid into a beaker, the user would play a decisive role, himself becoming an

actor in the legend told on the spout of the flagons. Just in front of the duck the spouts have an ‘open section’, and when pouring something from the flagons the liquid would have been visible, streaming out just in front of the duck. A stream here in the spout thus provided an escape route from the three threatening animals. The duck could swim away from the danger – but helped by ‘human interaction’: the user would provide a helping hand to the survival of the duckling. Without further interpretations we might believe that we are witnessing a sort of hunting scene, where a duckling escapes its predators helped by the users of the flagons. Without references to and the understanding of the aquatic bird’s role in Bronze Age iconography a more specific, mythological interpretation of the scenario will not be possible. As we have already seen, the aquatic bird, sometimes in the shape of a duck, was the most important central and western European helper of the sun during its eternal cyclical voyage over the heavens at daytime and through the underworld during the night. Since the birdsunship of central Europe – with stems ending in birds’ heads – is symmetrical, it should be regarded as working both during day and night. When the aquatic bird stands alone, without any ship, it should still be perceived as the helper or representation of the sun. Sometimes these birds carry solar images themselves, or solar symbols are shown close-by. This is also the case with the duckling on the Basse-Yutz flagons, where each side of the spouts is decorated with a row of concentric circles. The aquatic bird could even be seen as a manifestation or incarnation of the sun or sun-god (Kruta 2012). The intentional rendering of a very young duckling underlines the sun-symbolism of the Basse-Yutz flagons, the duckling being a representation of the new-born sun, the rising

110

Flemming Kaul

morning sun. There is nothing curious in appreciating the sun-rise as the time of the rebirth of the sun after the voyage through the sinister underworld related to darkness and death. The duckling – the sun – has just successfully left the underworld. In or at the gates of the underworld there are monster-like creatures that can prevent the movement of the sun towards the horizon and the world of the living. Using the punctuated line of four rectangular pieces of coral inlay on the spout it is possible to follow the escape route of the duckling away from monsters. The route of the sun is marked by concentric circles between the coral inlay. Were it not for the stream of liquid, which the duckling is ready to jump into in order to escape in its best element – by swimming – the sun, the duckling, might have been stopped and would stay in the underworld. By the interference of the users of the flagons, this would not happen. While feasting and drinking – ritual use – the eternal return of the sun was secured. The path of the duckling goes from the mouth of the flagons to its position close to the spout. On the mouth of the flagons there are stoppers. Their domed tops carry a decoration of inserted red enamel which form four stylized lotus buds, each joined at its two extremities to its neighbours to form a symmetrical design (Megaw and Megaw 1990, 26). Here once again we should consider whether the lotus pattern was just a decorative motif without any religious meaning, or whether it did carry connotations as to lotus symbolism related to (ultimately Egyptian) notions of rebirth of the sun. A votive object exhibited in the Archaeological Museum in Delphi, Greece, may provide important information as to a connection between the symbolism of the aquatic bird and the lotus. Here a sort of votive stand from the 8th century BC, deposited in one of the sanctuaries of Delphi, terminates in a plastic lotus, and in the very flower sits a duck or a duckling. When regarding the duck as a manifestation or representative of the sun – and here sitting in a lotus – as the reborn sun, we have a clue that the lotus in this case has not lost its original meaning. On an Early La Tène linch-pin, c.400 BC, from Unterradlberg at St. Pölten, Austria (Megaw, Megaw and Neugebauer 1989, 506 ff.), a stylized lotus is rising as a crown on a human-like head between a couple of monster-like creatures with eagles’ (?) heads. Here the lotus should be more than just an ornament. With its dominant position it must have had a meaning, perhaps partly its original meaning. It would be tempting to see this lotus as a representative or manifestation of the rising sun, which, by its natural force, could press itself up and unfold, having just escaped the guardian monsters of the underworld. Considering these two examples of the lotus and the implications as to a possible understanding of the motif, would it then be going too far to imagine that the duckling

Fig. 12.4. Small plastic aquatic bird figure decorating a complex bronze belt, Thorshøj, North Jutland, Denmark. Late Pre-Roman Iron Age, local manufacture. Photo F. Kaul.

on the spout of the Basse-Yutz flagons has just jumped out of the lotus (decoration) on the stoppers at sunrise – to continue its voyage as the new born sun on a river of wine, and helped by humans? The three monsters on the Basse-Yutz flagons, wolves or dogs, probably pursuing the duckling, belong according to the present interpretation to the realm of the underworld, representing chaos-powers or eschatological powers trying to stop the eternal voyage of the sun. The monster-like creatures do not fit into the general vocabulary of Bronze Age religious art. Wolves or dogs seem to emerge in Early Etruscan art and in ‘situla art’ (Jacobsthal and Langsdorff 1929; Lucke and Frey 1962, Nr. 44; Frey 1966, 54–58). Such creatures belong to something new, emerging in Early La Tène art, though their origin has been disputed (Megaw and Megaw 1990, 54 ff.). Since there are three of these creatures, it is tempting to mention the three-headed monster dog Cerberus of Greek mythology as a sort of structural parallel – Cerberus being Hades’ guard dog of the underworld. Cerberus is related to Heracles and his twelve labours of Greek mythology, and the exploits of Heracles (Hercle) were also very popular motifs among the Etruscans (Bonfante and Swaddling 2006, 34 ff.). When these splendid flagons were used, probably for feasts, we must remember that any feast in Prehistoric times was a feast with a religious background (as are many feasts today). By feasting and drinking, using the BasseYutz flagons, cosmological order could be secured, and the reborn sun could rise in the morning, unharmed by the dangers of the underworld. Even though the Basse-Yutz flagons belong to the La Téne period, they can be seen as representing a Bronze Age world view, the solar duckling being the central motif.

12.  The not so ugly duckling – an essay on meaning

111

Fig. 12.5. Double aquatic birds, probably bird-sunships, wagon fittings, Late Pre-Roman Iron Age, import or local manufacture? After Jensen 1982.

Late birds of the North A late revival of the aquatic bird motif can be seen outside ‘the Celtic world’, in what should be considered as Germanic areas. In the 1st century BC the duck reappears in Jutland, Denmark, as a result of Celtic influences. It occurs in particular as decoration on heavy bronze belts (Madsen 1999), also as seen on a belt from Thorshøj (Fig. 12.4) recently acquired as treasure trove by the National Museum of Denmark (Kaul and Klingenberg 2013). In its full body plastic form it reminds us of the duckling from Basse-Yutz, even though not as detailed and much corroded. In the case of these decorated belts there seems to be a reference to religion and ritual: According to the Greek geographer Strabo, chain belts were a specific accessory for Germanic priestesses (Kruta 2012, 54). On a number of bronze carriage mountings from Fredbjerg, Jutland, also dating to the last century BC, the aquatic bird appears as double birds (Jensen 1982; Kaul 2009) (Fig. 12.5). These double birds could be considered as a representation of the bird-sunship, where the protruding rivet hole with its decorated rivet amidships could be regarded as representing the sun. Here in the North, around the birth of Christ, we are close to the last quirk of the venerable duck and the bird-ship with deep roots going back to the Bronze Age.

Bibliography Bonfante, L. and Swaddling, J. 2006. Etruscan Myths. London: British Museum Press. Frey, O. H. 1966. Der Ostalpenraum und die antike Welt in der frühen Eisenzeit, Germania 44, 48–66. Brøndsted, J. 1938. Bronzealderens Soldyrkelse. København: Gyldendalske boghandel, Nordisk forlag. Gelling, P. and Davidson E. 1969. The Chariot of the Sun and other Rites and Symbols of the Northern Bronze Age. London: J. M. Dent & Sons. Guggisberg, M. A. 2000. Der Goldschatz von Erstfeld. Ein keltischer

Bilderzyklus zwichen Mitteleuropa und der Mittelmeerwelt, Antiqua 32. Basel: Veröffentlichung der Schweizerischen Gesellschaft für Ur- und Frühgeschichte. Jacob-Friessen, K. H. 1934. Verzierte Bronzerasiermesser aus Niedersachsen und ihre kultische Bedeutung, Altschlesien 5, 364–75. Jacobsthal, P. 1944. Early Celtic Art. Oxford: Clarendon Press. Jacobsthal, P. and Langsdorff, A. 1929. Die Bronzeschnabelkannen. Ein Beitrag zur Geschichte des vorrömischen Imports nördlich der Alpen. Berlin-Wilmersdorf: Verlag Heinrich Keller. Jennbert, C. 2000. Archaeology and Pre-Christian Religion in Scandinavia, Current Swedish Archaeology 8, 127–39. Jensen, S. 1982. Fredbjergfundet. En bronzebeslået pragtvogn på en vesthimmerlandsk jernalderboplads, Kuml 1981, 169–216. Joy, J. 2008. Reflections on Celtic art: a re-examination of mirror decoration. In D. Garrow, C. Gosden and J. D. Hill (eds.) Rethinking Celtic Art, 78–99. Oxford: Oxbow Books. Kaul, F. 1998. Ships on Bronzes, A Study in Bronze Age Religion and Iconography. Copenhagen: Publications from the National Museum vol. 3 (1 & 2). Kaul, F. 2002. The horse, the ship, the sun and the wheel in the Nordic Bronze Age: Reality and abstraction, symbols and metaphors – conceiving figures of sacred images. In Z. Karazová and M. Licka (eds.) Figuration et Abstraction dans l’Art de l’Europe Ancienne (VIII’ème–I’er S. a. J.-C.), Actes du Colloque International du Praque, Musée National, 13.–16. Julliet 2000, Acta Musei Nationalis Pragae, Series A – Historia, LVI/2002, 143–54. Prague: Národní Muzeum. Kaul, F. 2004. Bronzealderens religion. Studier af den nordiske bronzealders ikonografi, Nordiske Fortidsminder, Serie B, Bind 22. Copenhagen: Der Kongelige Nordiske oldskirftselskab. Kaul, F. 2009. Kan vi finde guden bag vognene fra Dejbjerg og Fredbjerg. In K. M. Boe, T. Capelle and C. Fischer (eds.) Tollundmandens verden. Kontinentale kontakter i tidlig jernalder, 69–76. Silkeborg: Wormianum & Silkeborg Kulturhistoriske Museum. Kaul, F. 2010. The sun image from Trundholm (‘The Chariot of the Sun’) – a commented history of research. In H. Meller and F. Bertemes (eds.) Der Griff nach den Sternen. Wie Europas Eliten zu Macht und Reichtum kamen. Internationales Symposium in Halle (Saale) 16.–21. Februar 2005, Tagungen des Landesmuseum für Vorgeschichte Halle (Saale) Band 05,

112

Flemming Kaul

2010, 521–536. Halle: Landesamt für Denkmalpflege und Archäologie Sachsen-Anhalt. Kaul, F. and Klingenberg, S. 2013, Bælter of bronze, Skalk 2013 no 2, 18–24. Kossack, G. 1999. Religiöses Denken in dinglicher und bildlicher Überlieferung Alteuropas aus der Spätbronze- und frühen Eisenzeit (9.–6. Jahrhundert v. Chr. Geb), Bayerische Akademie der Wissenschaften, Philosophisch-Historische Klasse, Abhandlungen, Neue Folge, Heft 116. München: Verlag der Bayerischen Akademie der Wissenschaften. Kruta, V. 1986. Le corail, le vin et l’arbre de vie: observations sur l’art et la religion de Celtes du V’e au 1’er siècle avant J.-C, Études Celtiques 23, 7–32. Kruta, V. 2012. La place et la signification du cheval dans l’imagerie celtique, Études Celtiques 38, 43–59. Lucke, W. and Frey, O. H. 1962. Die Situla in Providence (Rhode Island). Ein Beitrag zur Situlenkunst des Osthallstattkreises, Römisch-Germanische Forschungen Band 26. Berlin: Verlag Walter de Gruyter & Co. MacGregor, N. 2010. A History of the World in 100 Objects. London: British Museum. Madsen, O. 1999. Hedegård – a rich village and cemetery complex of the Early Iron Age on the Skjern River. An interim report, Journal of Danish Archaeology 13, 57–93. Malmer, M. 1989. Principles of non-mythological explanation of North European Bronze Age rock art. In H. Å. Nordström and A. Knape (eds.) Bronze Age Studies, Transactions of the British-Scandinavian Colloquium in Stockholm, May 10th–11th 1985, The Museum of National Antiquities Stockholm, Studies 6, 1989, 91–99. Stockholm: Statens Historiska Museum. Megaw, J. V. S. 1970a. Art of the European Iron Age. Bath: Adams & Dart. Megaw, J. V. S. 1970b. Cheshire Cat and Mickey Mouse: Analyses, Interpretation and the Art of the La Tène Iron Age, Proceedings of the Prehistoric Society 36, 261–79. Megaw, J. V. S. 2002. Figuration and Abstraction in Early Celtic Art: who is Smiling Now? In Z. Karazová and M. Licka (eds.) Figuration et Abstraction dans l’Art de l’Europe Ancienne (VIII’ème–I’er S. a. J.-C.), Actes du Colloque International du Praque, Musée National, 13.–16. Julliet 2000, Acta Musei Nationalis Pragae, Series A – Historia, LVI/2002, 57–64. Prague: Národní Muzeum. Megaw, J. V. S. and Megaw, M. R. 1989. Celtic Art. From its Beginnings to the Book of Kells, London: Thames & Hudson. Megaw, J. V. S. and Megaw, R. 1990. The Basse-Yutz Find. Masterpieces of Celtic Art, Reports of the Research Committee of the Society of Antiquaries of London No. 46. London: Thames & Hudson. Megaw, J. V. S. and Megaw, M. R. 2008. A Celtic mystery: some thoughts on the genesis of insular Celtic art. In D. Garrow, C. Gosden and J. D. Hill (eds.) Rethinking Celtic Art, 40–58. Oxford: Oxbow Books.

Megaw, J. V. S., Megaw, M. R. and Neugebauer, J. W. 1989. Zeugnisse frühlatènezeitlichen Kunsthandwerks aus dem Raum Herzogenburg, Niederösterreich, Germania 67, 477–517. Moosleitner, F. 1985. Die Schnabelkanne vom Dürrnberg. Ein Meisterwerk keltischer Handwerkskunst. Salzburg: Salzburger Museum Carolino Augusteum. Müller, S. 1897. Vor Oldtid. Copenhagen: Det nordiske Forlag. Müller, S. 1903. Solbilledet fra Trundholm, Nordiske Fortidsminder, I, 6, Copenhagen. Müller, S. 1920. Billed- og Fremstillingskunst i Bronzealderen, Aarbøger for Nordisk Oldkyndighed og Historie 1920, 125–61. Patay, P. 1990. Die Bronzegefässe in Ungarn, Prähistorische Bronzefunde, Abteilung II, Band 10. München: C.H. Beck. Pauli, L. 1992. Quellen zur keltischen Religionsgeschichte. In R. Bech, H. Jankuhn and R. Wenskus (eds.) Germanische Religionsgeschichte, Quellen und Quellenprobleme, Ergänzungsbände zum Reallexikon der Germanischen Altertumskunde, Band 5, 118–44. Berlin: Walter de Gruyter. Rapi, M. 2009. La seconda Età del Ferro nell’area di Como e dintorni. Como: Materiali La Tène nelle Colezioni del Civico Museo Archeologico P. Giovio. Rapin, A. 2002. Nos concepts modernes d’abstraction et de figuration. Sont-ils adaptés aux analyses de l’art laténien. In Z. Karazová and M. Licka (eds.) Figuration et Abstraction dans l’Art de l’Europe Ancienne (VIII’ème–I’er S. a. J.-C.), Actes du Colloque International du Praque, Musée National, 13.–16. Julliet 2000, Acta Musei Nationalis Pragae, Series A – Historia, LVI/2002, 45–56. Prague: Národní Muzeum. Sapouna-Sakellarakis, E. 1978. Die Fiebeln der griechischen Inseln, Prähistorische Bronzefunde, Abteilung XIV, Band 4. München: C. H. Beck. Sestieri, A. M. B. and Macnamara, E. 2007. Prehistoric Metal Artefacts from Italy (3500–720 BC) in the British Museum, British Museum Research Publications Number 159. London: British Museum. Shaw, I. and Nicholson, P. 1995. British Museum Dictionary of Ancient Egypt. London: British Museum. Sprockhoff, E. 1936. Sonnenwagen und Hakenkreuz im Nordischen Kreis, Germania 20, 1–9. Sprockhoff, E. 1954. Nordische Bronzezeit und frühes Griechentum, Jahrbuch des Römisch-Germanischen Zentralmuseums Mainz 1, 28–110. Sprockhoff, E. 1955. Central European Urnfield Culture and Celtic La Tène: An outline, Proceedings of the Prehistoric Society 21, 257–81. Sprockhoff, E. 1969. Mitteleuropäische Urnenfelderkultur und keltisches Latène, Bremer Archäologischer Blätter 5, 21–47. Worsaae, J. J. A. 1882. The Industrial Arts of Denmark. From the Earliest Times to the Danish Conquest of England. London: Chapman and Hall Limited.

13 FRAGMENTS OF A CARNYX FROM LEISACH (AUSTRIA) Paul Gleirscher

This short essay is dedicated to Vincent Megaw who has studied so many relicts of Celtic culture, amongst them Celtic and Roman carnyces, too (Megaw 1968; 1991, 645– 647). Thus, of course, he has dealt with the extraordinary carnyx from Deskford (Scotland) with its well preserved boar-head (Megaw 1968, 350; 1970, 160 no. 272; recently in detail Hunter 2001, 77–86). It was found in 1816 during peat digging and thus is to be interpreted as a votive find, like the lost carnyx-finds at Tattershall Bridge (Lincolnshire) in 1768 (Hunter 2009, 86f.). Considering stylistic arguments, Megaw has proposed a production date for the Deskfordcarnyx in about AD 50 somewhere in Belgium. Following him, Morna Simpson has considered if the Deskford-carnyx could have been used in the battle against Agricola who left Britain in AD 85. Modern analytical results have shown that the bronze of the Deskford-carnyx contains a high level of zinc which is explained by Fraser Hunter with the use of remelted Roman alloys after AD 80 in a local factory (Hunter 2009, 78–80). The Burgstall in Leisach near Lienz (Tyrol, Austria) had already caught the eye of Sepp Kalser in 1982, when he mentioned the remains of ruins and terraces there. It is a small naturally-fortified hill with a favourable strategic position in the Drava-valley in the south-western area of Lienz. Though the ruins of the Burgstall had been noticed in the Atlas Tyrolensis by Peter Anich and Blasius Hueber in 1774, they have not received any adequate attention in regional archaeological research work up to now, and nor have the associated ruins of the medieval castle of Neuenburg – also called Rauberschlössl or Rabaschlössl – to the west (Kalser 2012, 3–4, fig. 1 and 4). Over recent years, Kalser, together with his friend Peter Egartner, has searched through both places using a metal-detector. As a result, they have brought to light a lot of medieval and Roman objects

from the ruins of the Neuenburg (Kalser 2012, 11–41) and objects from Late Iron Age and Roman times from the Burgstall (Kalser 2012, 51–75). The particular significance of the finds from the Burgstall in Leisach is that they are votive offerings from a Late Iron Age and Roman sanctuary. Thus one could mention several firedogs en miniature ending in bulls’ heads, a bronze Rhaetian wand, several votive plates of italic-alpine type, or several figurines of Roman deities made of tin. Traces of burnt animal give evidence of a so called Brandopferplatz going back to about 500 BC and continuing by changing its structure in Celtic and Roman times (in general Gleirscher, forthcoming). The four bronze fragments to be discussed now are part of the extraordinary votive offerings from the Burgstall of Leisach (Kalser 2012, 54–57, fig. 58). They were found close together and might have been part of a deposit within the sanctuary. All of them have been deformed intentionally. Fragment 1 (Fig. 13.1, A4 and B4): fissured tube in bronze (length 17.5 cm; aperture 2.2 cm), cone shaped tube (aperture 2.6 cm; height up to 3 cm) with solid sleeve-ring (aperture 3.5 cm; height 0.5 cm) and bronze tube (length 3.5 cm; width 0.8 cm), in the upper part quite damaged. Fragment 2 (Fig. 13.1, A2 and B2): deformed bronze sheet with solid bronze stick, getting narrower (length 7.9 cm). Fragment 3 (Fig. 13.1, A1 and B1): deformed bronze sheet with solid bronze rod, getting narrower (length 4.1 cm). Fragment 4 (Fig. 13.1, A3 and B3): bent bronze sheet (about 11 × 12.1 cm). In any case the tube in bronze with the solid sleeve-ring (Fig. 13.1, A4 and B4) is to be interpreted as fragment of a carnyx paralleled by Sanzeno in Val di Non, Italy (Roncador

Paul Gleirscher

114

4

4

Fig. 13.1. Leisach/Burgstall (Austria). One or more fragments of a Celtic carnyx. Scale 1:1, ‘bronze’ (Photo: K. Allesch, drawing: P. Gleirscher, Landesmuseum für Kärnten).

13.  Fragments of a carnyx from Leisach (Austria)

115

Fig. 13.2. Find spots of carnyces, amalgamated and updated after Hunter (2009, 80–2, fig. 11) and Veres (2009, 241–2) (Drawing: H. Mühlbacher, Landesmuseum für Kärnten).

and Mellini 2010, 160–161, figs. 19–20). It is not certain that the two deformed bronze sheets with the solid bronze rod (Fig. 13.1, A2/B2 and A1.B1) are to be interpreted as the ears of a boar-head – as seems to be probable – but the more bent of the two bronze-sheets could have been part of the bell of a carnyx (Fig. 13.1, A3/B3). The carnyces from the eastern Alpine region, from Sanzeno as well as from Leisach, seem to date to pre-Roman times, thus to the 2nd/1st century BC. The carnyces from Sanzeno have been analyzed recently in detail by Rosa Roncador (Roncador 2009; Roncador and Melini 2010, 160–161). Sanzeno was one of the main settlements of the non-Celtic Rhaeti (Marzatico 1999, 472–484), and also the upper Drava-valley with Leisach was part of the Rhaetian area between the late 6th and 3rd centuries BC (Metzger and Gleirscher 1992; cfr. Megaw and Megaw 1998, 398), when Celtic tribes coming from the lower Danube region started settling in

south-eastern Alpine areas (Gleirscher 1985, 720–721; 1987, 244–245; – cfr. Stadler 1992). Honorary inscriptions from the Magdalensberg (Carinthia) dating to early Augustan times mention a Celtic tribe named Lainaci around Lienz and another one named Savates westward around Bruneck (Gleirscher 1996; 2009, 19–22). The discussion as to whether and to what extent the middle part of the Adigevalley around Trento became Celtic within the 3rd/2nd century BC is still in progress (Marzatico 1992a; 1992b, 225–227; 1999, 491–494; 2000, 479–492, 527–532). Thus, the carnyces from Sanzeno – probably votive offerings – seem to derive from a Rhaetian and non-Celtic culture area, the one from Leisach from a Celtic sanctuary that goes back to Rhaetian times. In this sense they conform to the feature that all the known carnyces have been found in waterlogged contexts or as a deposit within a sanctuary (Hunter 2009, 81; Maniquet et al. 2011, 108). Thus they have been more or less

116

Paul Gleirscher

Fig. 13.3. Similarties between Celtic carnyx and standard after Hunter 2009, 75–9, fig. 5.

damaged or deformed. Also of note in eastern Celtic areas are the depiction of a carnyx-player from Kondoros (kom. Békés, Hungary) as well as two Roman silver fragments, probably fragments of carnyces, from Săliştea de Sus (jud. Maramureş) and Islaz (jud. Teleorman) in Romania (Veres 2009, 231–234, fig. 1,1 and 2; 242 and 245). The most important insight into carnyces of late Celtic times is based on the recent excavations of the Gallic-Roman temple of Les Arènes near Tintignac-Naves, dép. Corrèze (France). It dates to the 2nd/3rd century AD and continues a sanctuary from late Celtic times, which is charaterized by two quadratic surrounding ditches (about 24 × 25 m). In 2004, an extensive deposit containing about 500 mostly highly fragmented objects in metal was found within a shallow fosse (Maniquet 2008; Maniquet et al. 2009; 2011 – cf. Roncador and Melini 2010, 159, or Armbruster 2012). These date from the 4th to the 2nd century BC and may be interpreted as votive offerings and/or equipment for rituals. Of particular note are weapons, attachments in the form of animal figures (boar and horse), elements of harness, and an iron cauldron, besides the fragments of seven carnyces, all of them damaged intentionally and deposited incompletely,

dating probably to the 2nd/1st century BC. While six of them end in the head of a boar, one of them seems to end in the head of a snake. These carnyces have been analyzed not only by stylistic considerations but also in a technological sense. The mouth-pieces and the sleeve-rings emerged as cast elements based on an alloy containing copper, tin and lead. All the other elements, however, are made of coldforged sheets based on an alloy containing copper, tin and bronze. Soldering by tin as well as plugging and squeezing have been used to connect the single elements. Finally, the grooves were sealed. Only highly developed techniques enabled Celtic craftsmen to construct wind instruments with a length of 1.8 m, consisting of at least 28 single elements and weighing only 1.6 kg. Carnyces (cf. Piggott 1959; Hunter 2001, 86–97; 2009, 82, figs. 11–12; 2012 or Maniquet et al. 2011) are still classified, particularly in pre-Roman Europe, as the war trumpets of the Celts. They are first mentioned in written sources at the plunder of Delphi in 279 BC. Thus the oldest archaeological finds of carnyces are dated to the beginning of the 3rd century BC (La Tène B2). The supporting archaeological evidence is based on some uncertain fragments from the

13.  Fragments of a carnyx from Leisach (Austria) oppidum of Manching (Hunter 2009, 80–81, fig. 9) and the dating for the oldest carnyx from Tintignac proposed by Christophe Maniquet (Maniquet et al. 2011, 109). Within the long passage of time between the 3rd century BC and the 3rd century AD, carnyces were used by different tribes and peoples. Besides the finds of carnyces and their depictions concentrated within the Celtic area of La Tène culture (Fig. 13.2) pictures on coins from southern England, central Gaul and central Greece should be mentioned, as well as reports and depictions by the Romans. For them the carnyx was a symbol of barbarians, after the Gallic War particularly for the Celts but also for the Bretons and Germans (Cimbri). Carnyces became a symbol of Germans and Dacians within the later 1st and 2nd centuries AD in the lower Danube region, within the 3rd century AD only of Germans and Caledonians in Scotland (Hunter 2009, 81–82, figs. 11–12). The body structure of carnyces is known in general by archaeological finds as well as by depictions. Classification into types and chronology has started based on the finds from Tintignac. It seems that the long shank ended in a straight mouth-piece. The bell seems to have been formed mostly as a head of a boar representing male fighting power, well known by the Celts as a scabbard decoration, too (so called dragons), thus probably a symbol of Teutates. Thus carnyces may be derivates of the Etruscan lituus which was developed into a zoomorphic bell by the Celts north of the Alps. The carnyx gives a deep sound reaching to the core. It was designed to create confusion in battle as well as inspiring an awed atmosphere in cultic ceremonies, as recently underlined by Maniquet (et al. 2011, 107–108) by the fact that quite a lot of carnyces have been found in sanctuaries. The use of carnyces in ritual ceremonies is documented particularly on plate VI of the Gundestrup cauldron from Denmark (misinterpreted by Hunter 2001, 97; cf. Gleirscher 2008, 146–147 – for recent work on the cauldron see e.g. Kaul 1999; on its derivation Falkenstein 2004 or Veres 2009, 244–245). The confusing impact of the sound of carnyces within battle is referred to by Fabius Pictor concerning the battle of Talamone in 225 BC (related by Polybius, Histories 2, 29, 6). So carnyces could be classified as psychological weapons, too (Roncador and Melini 2010, 158). Thus, the question arises whether the carnyces found in sanctuaries or ritual deposits are to be classified as ritual equipment or as arms, kinds of deposition well known in late Celtic Europe as well as in the Alpine area (cf. e.g. Müller 2002, 93–158; Zemmer-Plank and Sölder 2002; Egg 2002; 2012). Hunter (2009, 79–80, fig. 6) has analyzed the attached and repeatedly oversized ears of the boar-heads of the Celtic carnyces. He distinguished two groups: Group 1 (type Tintignac) is characterized by ears with a length between 30 and 45 cm, group 2 (type Mandeure) by ears with a length between 10 and 15 cm. The attached ears – particularly the longer ones – not only enlarged the frightening appearance of the boar-heads, but also enhanced their frightening

117

sound. Doubtlessly also, fragments of these ears are to be interpreted as parts of carnyces. The smaller ears, however, and particularly those which were attached not by tubes but by solid bronze rods as in Leisach, can be interpreted as fragments of carnyces as well, but could also have belonged to standards or – less probable – to figures (Fig. 13.3). Further similarities between Celtic carnyces and Celtic standards, which are the configuration of the mouth and of the crest, have been outlined by Hunter (2009, 75–79, fig. 5). Maniquet has compared the oversized ears of the boar-heads of the Celtic carnyces with the crowns of leaves of the stelae of Celtic rulers who became heroes in death (Maniquet et al. 2011, 108–109). These crowns of leaves, however, are crowns of mistletoe (e.g. Frey and Hermann 1997, 480) and thus not to be compared with the boars’ ears of the carnyces. The only similarity is their size.

Acknowledgements I’d like to thank Rosa Roncador (Trento) and Martin Schönfelder (Mainz) for discussion.

Bibliography Armbruster, B. 2012. Wildschweine mal anders – Der Metallhort von Tintignac-Naves. In Die Welt der Kelten. Zentren der Macht – Kostbarkeiten der Kunst, 416–17. Ostfildern: Jan Thorbecke Verlag. Egg, M. 2002. Eisenzeitliche Waffenweihungen im mittleren Alpenraum. In L. Zemmer-Plank and W. Sölder, (eds.) Kult der Vorzeit in den Alpen. Opfergaben – Opferplätze – Opferbrauchtum, 961–84. Bozen: Athesia-Verlag. Egg, M. 2012. Die eisenzeitlichen Waffenweihungen in den Ostalpen. In W. Meighörner (ed.) Waffen für die Götter. Krieger Trophäen Heiligtümer,100–08. Innsbruck: Tiroler Landesmuseum Ferdinandeum. Falkenstein, F. 2004. Anmerkungen zur Herkunftsfrage des Gundestrupkessels, Praehistorische Zeitschrift 79.1, 57–88. Frey, O.-H. and Hermann, F.-R. 1997. Ein frühkeltischer Fürstengrabhügel am Glauberg im Wetteraukreis, Hessen, Germania 75.2, 459–550. Gleirscher, P. 1985. Zur zeitlichen und kulturellen Stellung der Schale aus dem Brandgrab von Lavant in Osttirol, Der Schlern 59, 712–21. Gleirscher, P. 1987. Die Kleinfunde von der Hohen Birga bei Birgitz. Ein Beitrag zur Fritzens-Sanzeno-Kultur, Berichte der Römisch-Germanischen Kommission 68, 181–351. Gleirscher, P. 1996. Die Kelten im Raum Kärnten aus archäologischer Sicht. Ein Forschungsstand. In E. Jerem, A. Krenn-Leeb, J.-W. Neugebauer and O. H. Urban (eds.) Die Kelten in den Alpen und an der Donau, Archaeolingua 1, 255–66. Budapest/Wien: Archaeolingua. Gleirscher, P. 2008. Menschen- und Tieropfer aus der DurezzaSchachthöhle über Warmbad Villach? In E. Lauermann

118

Paul Gleirscher

and P. Trebsche (eds.) Heiligtümer der Druiden. Opfer und Rituale bei den Kelten, 142–53. Asparn/Zaya: Verlag des Niederösterreichischen Landesmuseums. Gleirscher, P. 2009. Noreia – Atlantis der Berge. Neues zu Göttin, Stadt und Straßenstation. Klagenfurt: Hermagoras-Verlag. Gleirscher, P. (forthcoming). Vorrömerzeitliche Naturheiligtümer und die Frage ihres Fortwirkens in die Römerzeit. Fallbeispiele aus dem Ostalpenraum. In K. Sporn, S. Ladstätter and M. Kerschner (eds.) Natur – Kult – Raum. Wien: Verlag des Österreichischen Archäologischen Instituts. Hunter, F. 2001. The carnyx in Iron Age Europe, The Antiquaries Journal 81, 77–108. Hunter, F. 2009. Une oreille de carnyx découverte à La Tène. In M. Honegger, D. Ramseyer, G. Kaenel, B. Arnold and M.-A. Kaesar (eds.) Le site de La Tène: bilan des connaissances – état de la question, Archéologie neuchâteloise 43, 75–85. Neuchâtel: Office et musée cantonal d’archéologie. Hunter, F. 2012. Carnyx. In S. Sievers, O.-H. Urban and P. C. Ramsl (eds.) Lexikon zur keltischen Archäologie, Mitteilungen der Prähistorischen Kommission 73, 302–03. Wien: Verlag der Österreichischen Akademie der Wissenschaften. Kalser, J. 2012. Die vergessene Burg. Von der Neuenburg zurück in die römische Zeit und zu den verschollenen Opferkultplätzen unserer Ahnen. Leisach: Eigenverlag. Kaul, F. 1999. Gundestrup. In J. Hoops (ed.) Reallexikon der Germanischen Altertumskunde 13, 195–211. Berlin-New York: Walter de Gruyter. Maniquet, Ch. 2008. Le dépôt cultuel du sanctuaire gaulois de Tintignac à Naves (Corrèze), Gallia 65, 273–326. Maniquet, Ch., Adamski, F., Armbruster, B., Drieux, M., Espinasse, L., Lejar, Th. and Mora, P. 2009. Les guerriers gaulois de Tintignac. Limoges: Editions Culture et Patrimoine en Limousin. Maniquet, Ch., Lejars, Th., Armbruster, B., Pernot, M., DrieuxDaguerre, M., Mora, P. and Espinasse, L. 2011. Le carnyx et le casque-oiseau celtiques de Tintignac (Naves-Corrèze). Description et étude technologique, Aquitania 27, 63–150. Marzatico, F. 1992a. I Galli abitanti nel Trentino preromano? Revisione della vecchia tesi alla luce delle attuali conoscenze archeologiche. In N. Vicenzi (ed.) Per Aldo Gorfer, 619–51. Trento: Provincia Autonoma di Trento. Marzatico, F. 1992b. Il gruppo Fritzens-Sanzeno. In Metzger and Gleirscher 1992, 213–46. Marzatico, F. 1999. I Reti in Trentino: il gruppo FritzensSanzeno. In G. Ciruletti and F. Marzatico (eds.) I Reti/Die Räter, Archeologia delle Alpi 5.1, 467–504. Trento: Provincia Autonoma di Trento.

Marzatico, F. 2000. La seconda età del Ferro. In M. Lanzinger, F. Marzatico and A. Pedrotti, Storia del Trentino I. La preistoria e la protostoria, 479–573. Trento: Società editrice Il Mulino. Megaw, J. V. S. 1968. Problems and non-problems in palaeoorganology: a musical miscellany. In J. M. Coles and D. D. A. Simpson (eds.) Studies in Ancient Europe. Essays presented to Stuart Piggott, 333–58. Leicester: Leicester University Press. Megaw, J. V. S. 1970. Art of the European Iron Age. New York/ Evanston: Harper & Row. Megaw, J. V. S. 1991. La musica celtica. In S. Moscati (ed.) I Celti, 643–48. Milano: Bompiani. Megaw, J. V. S. and Megaw, M. R. 1998: Cheshire Cats in the Tyrol. In A. Müller-Karpe, H. Brandt, H. Jöns, D. Krauße and A. Wigg (eds.) Studien zur Archäologie der Kelten, Römer und Germanen in Mittel- und Westeuropa. Alfred Haffner zum 60. Geburtstag, Internationale Archäologie, Studia honoraria 4, 389–400. Rahden/West: Verlag Marie Leidorf. Metzger, I. R. and Gleirscher, P. (eds.) 1992. Die Räter/I Reti. Bozen: Athesia-Verlag. Müller, F. 2002. Götter, Gaben, Rituale. Religion in der Frühgeschichte Europas. Kulturgeschichte der antiken Welt 92. Mainz: Philip von Zabern. Piggott, S. 1959. The carnyx in early Iron Age Britain, The Antiquaries Journal 39, 19–32. Roncador, R. 2009, La riscoperta del carnyx di Sanzeno (Val di Non, Trentino). Storia degli studi e inquadramento culturale. In S. Grunwald, J. K. Koch, D. Mölders, U. Sommer and S. Wolfram (eds.) ARTeFACT, Festschrift für Sabine Rieckhoff zum 65. Geburtstag, Universitätsforschungen zur prähistorischen Archäologie 172.2, 547–55. Bonn: Verlag Rudolf Habelt. Roncador, R. and Melini, R. 2010. Il karnyx di Sanzeno (Val di Non, Trentino): ritrovamento, indagini e ricostruzione. In M. Carrese, E. Li Castro and M. Martinelli (eds.) La musica in Etruria, Atti del convegno internazionale di Tarquinia, 155–76. Tarquinia: Comune. Stadler, H. 1992. Die eisenzeitlichen Gräber im Virgental und die Frage der Ostausdehnung der Fritzens-Sanzeno-Gruppe. In I. R. Metzger, and P. Gleirscher (eds.), Die Räter/I Reti, 551–65. Bozen: Athesia-Verlag. Veres, J. 2009. The depiction of a carnyx-player from the Carpathian Basin. A study of two Celtic bronze statuettes from eastern Hungary, Archäologisches Korrespondenzblatt 39, 231–49. Zemmer-Plank, L. and Sölder, W. (eds.) 2002. Kult der Vorzeit in den Alpen. Opfergaben – Opferplätze – Opferbrauchtum. Bozen: Athesia-Verlag.

14 BETWEEN RULING IDEOLOGY AND ANCESTOR WORSHIP: THE MOS MAIORUM OF THE EARLY CELTIC ‘HERO GRAVES’ Thomas Stöllner1

‘There can be no heroes if there are no monsters to fight and overcome’ (Vernant 1996, 58).

Instead of an introduction: the elaborate graves of the Glauberg The rich ‘princely’ graves of the Late Hallstatt- and Early La Tène culture north of the Alps have been a topic of discussion for some time now; the change from the Late Hallstatt period into the Early La Tène period, which manifests especially through the rite of these ‘princely’ graves, is counted as an old but still very current research question (e.g. Kimmig and Rest 1954; Pauli 1978; Verger 1995; Echt 1999; Megaw and Megaw 2009). The relations to the Mediterranean World that were associated with this transition raised the question, which social and religious changes determined the early Keltiké during a time which some even call the ‘Mediterranisation’ (e.g. Krausse 2004, esp. 193 pp.)? When starting to research a newly discovered burial mound in the area of the Glauberg near Büdingen in Hesse around twenty years ago, no one could even imagine how this find would fundamentally extend our knowledge of the Early La Tène culture (see especially Frey and Herrmann 1997; Baitinger and Pinsker 2002; Glauberg 2008; Baitinger 2010). I remember our beloved jubilarian when he was travelling once a year to the Wiesbaden laboratories to learn more about these inspiring findings. He himself called it a pilgrimage! The initial routine of the excavation soon developed a different quality, when the rich grave 1 with a golden torc and a bronze Celtic beaked flagon from the 5th century BC was discovered. The research was extended. Step by step the mysteries of the find spot were revealed: grave 2 was discovered in 1995, the following year the first

of four stone sculptures. Geophysical research clarified the monumentality of the grave: a grave-mound of 48 m in diameter circled by a ring ditch, ending in an approximately 10 m wide access ramp, facing south east. This one ended after 350 m in an extensive ditch system, enclosing the whole foreland of the Glauberg (e.g. Baitinger 2010). Some of the ‘breaks’ in the ditch system could be due to erosion or included already existing monuments. Thus the gap in the south, where an older, barrow cemetery from the Hallstatt period was discovered by the latest research. They may have been respected as silent witnesses of the ‘ancestors’ in the 5th century BC, maybe even the monuments were included because of a still vivid memory of this place (L. Hansen in Glauberg 2008, 21–34). In the area of the burial mound, another mound with a warrior burial (F.-R. Herrmann in Glauberg 2008, 89–107) as well as a rectangular enclosure and a strange ditch annexe with some post holes enclosed were discovered. The ring ditch of the burial mound ended in this annexe and the most complete of the four warrior statues was discovered exactly there (F.-R. Herrmann in Baitinger and Pinsker 2002, esp. 104 ff., 262 ff.). The statues are almost identical – the leafcrown, which is reminiscent of ears (the ‘famous forebears’ of Vincent’s Mickey Mouse ears!) and surrounds the head like a bonnet, is especially distinctive (Fig. 14.1.1). The fully plastic worked warriors are equipped with compound armour, an oval fencing shield with a spindle-shaped shield boss, a short sword from the Early La Tène period as well as a torc and a ring. The back of the armour, decorated with a lotus-leaf design imitating the original compass ornamentation, is quite striking and seems almost attached to it. Noteworthy is the similarity of the grim face expression, the massive chin, seemingly divided into individual parts, and the fixed, almost blank stare as well. The turned down

120

Thomas Stöllner

Fig. 14.1. Figural stone scultpures in the 6th to 4th century BC in Central and Western Europe, 1: Glauberg, Hesse (after Baitinger and Pinsker 2002), 2: Roquepertuse (after Lescure 1995), 3 Seurre, Burgundy (after Bonenfant and Guillaumet 1998), 4 Hischlanden, BadenWürttemberg (after Zürn 1964), 5 Vix, Burgundy (after Chaume et al. 1995).

14.  Between ruling ideology and ancestor worship: the mos maiorum of the Early Celtic ‘Hero Graves’ mouth makes the face look mask-like and deformed. Here it becomes clear, that the artist wants to express a certain ideology. Four times in the same way. Where exactly the statues were erected, remains unclear. Maybe some of the post hole positions and the exact position of the finds in the graves of the annexe can give some indication of a ‘holy district’ in this area (in such a sense especially Herrmann 2002, esp. 95 ff.; 2005, esp. 23 f). The inside of the graves discovered in the burial mound provides further information: in the centre there was only an empty cenotaph, whereas the best furnished grave 1 was discovered in the NW- extension. Grave 2 was situated directly at the entrance of the so called ‘procession-route’ and due to that position is supposed to be of an younger age than grave 1 (Frey and Herrmann 1997; Herrmann 2002, 98–101). Grave 1 is especially important, being completely investigated and wholly restored (see in general Baitinger and Pinsker 2002). The burial equipment corresponds for the larger part with the one of the statues: a golden torc with balusters, a ring, a fencing shield, a sword of the Early La Tène period and even more surprising: remnants of a head bonnet depicting presumably a leaf crown. The belt fittings, decorated with numerous plates and rattle rings are not less brilliant. In addition, the fibulae, the not yet interpreted and studded profiled bars as well as the throwing-lances and the bow and quiver were found, too. The sitting figurine, fixed between two predators on the opening crown of the Celtic beak-spouted ewer, is especially striking: wearing a compound armour like the stone figures and due to its positioning between two predator-like creatures, it can be interpreted as ‘Lord of the animals’ – being of the same iconographic composition as the burial equipment and the statues, thus indicating a comparable ideological context (Frey and Herrmann 1997; Frey 1998; 2008). Two points must be emphasized: first the close connection between the burial equipment and the iconography of the statues and accordingly the display of the sitting warrior on the beak-spouted ewer. It stands to reason to interpret them as the buried man himself – especially due to the latest find, a leaf crown in grave 1, wire netting in the shape of a leaf crown covered with leather (Herrmann 2005, 23, fig. 10). The second focus lies on the close bond to the equipment of the second grave – although a cremation grave, a luxurious ewer, now a so-called spouted ewer, was added. Stylistic and creative details show that both vessels are closely related (Frey and Herrmann 1997, 496–7; M. Bosinski and S. Martens in Baitinger and Pinsker 2002, 139 ff., 143 ff.; Frey 2008, esp. 36 ff.). That can also be demonstrated through the weaponry with sword and belt, only being a little bit richer in grave 1, but overall being similar in many details (Th. Flügen and M. Bosinski in Baitinger and Pinsker 2002, 151 ff.; also Stöllner 2010, esp. 278–282). The weapon belt differs only a little in the composition of its decorative fittings. The fibulae as well

121

as the weapon belt can only be assigned to one workshop or one craftsman, who worked around 430/420 BC (for the chronology see comments in Stöllner 2010, 279 note 3). The similarities emphasize that both deceased lived in a certain social relation. The deceased of grave 1 seemingly held a special position due to his close connection to the images. The monument with the ‘holy’ district seems to be raised on his account. How much later the deceased of grave 2 has been buried, we do not know for sure – but a contemporary burial seems likely. This situation recalls the records of Homer’s epics, where it says that Achilles and Patroclos, being connected as brothers-in-arms heroes, were buried together in a golden urn. The friend Antilochus, whom Achilles honoured ‘before all other friends’, got a separate burial in the same grave monument (Odyssey 24, 76–79) (in general Egg 2003; this interpretation for the Glauberg see Stöllner 2010, 279–280). But does such an analogy really enable us to understand the social and ritual context of our graves and the Early La Tène ‘princely’ tombs in general? The ‘habitus’ of the buried shows all in all a certain kind of warrior ideology.2 The warrior aspects became common in Early La Tène. It is something that research has regarded as typical for that time period (see below). But there are more aspects that we have to consider which have been added and which emphasize the social and ritual role of the deceased. These are next to the imposing grave monument, the ditch system, the burial equipment and especially the relation of the deceased from grave 1 to the stone statues. One could assume that these were positioned in a sort of sanctuary that held the memory of the ancestors, perhaps something like a heroon. However the next question follows – are the stone sculptures meant to be the deceased or is he positioned in a row with ‘heroic’ ancestors on purpose? It is even more interesting that there is no evidence for such an ideological construction that could be reconstructed by graves at least in the surroundings of the Glauberg. Can we assume that the person of grave 1 was seen in an imagined row of ancestors? Getting a burial with the adequate attributes quasi at last? This would include a following of brothers-in-arms. If in this context the leaf crown is a real ruling attribute or if it is just a means for the dead person’s glorification for the passing into the afterlife, remains open. The fact in itself is nevertheless noteworthy. It could be fruitful to take another look and ask ourselves, how it came to this development of a grave culture, how it can be integrated into contemporary events and which ritual meaning it could have.

The graves of the Glauberg in the contemporary context of Early La Tène period monuments First, we take a look at the burial equipment: clearly in the warlike elements in the grave – together with the statues – the deceased is shown as a warlike ‘hero’. But which

122

Thomas Stöllner

aspects link the graves to some kind of warrior ideology and which more to a heroic element? It is certainly difficult to deduce heroic ideas just from weapons which primarily represent warriorship and a newly arisen warrior ideology in Early La Tène society. But we may ask if the compound of iconography depicted on burial goods such as the flagons and swords and in the statues was also once seen as a heroic model, an ideal concept for Celtic warrior society. This would imply also some correlation with the life-world of those societies in which heroic persons played a dominant role, not only in terms of their martial skills but also in their relations to the ancestors and perhaps also in their curative aspects, which are displayed by the ‘master of the beasts’ idea (Stöllner 2010). These were in principle the burial rites of the Early La Tène period, especially in an area extending from the Champagne and Eastern France in the west towards Bohemia and Eastern Austria (Lorenz 1978, esp. 114–127; in terms of cultural zones: Dehn and Stöllner 1996). Suddenly there are numerous graves with weaponry in contrast to the Late Hallstatt period. Magnificent weaponry became the representative content of outstanding burials. Not only single weapons displaying status, but complete armour, swords, parts of the armament and shields as well as other war weapons were included in contrast to the Hallstatt period of the 6th and the early 5th century BC. The area with such burials is regionally different from those of the old ‘princely’ burials of the Hallstatt period. It connects with the old periphery of the Hallstatt area though (e.g. a map in Stöllner 2010, 297 ff., fig. 12). Outstanding male burials got a two-wheeled, sometimes splendidly decorated chariot, as well as luxurious flagons and paired drinking-horns, a richly decorated helmet and some Etruscan and Greek imported goods (for the princely graves in Early La Tène: e.g. Verger 1995; Echt 1999; Schönfelder 2003). The rich ornaments, a certain impracticality of the weapons, as well as connections to Mediterranean ideas, show that weapons were not only meant to be functional, but also had a symbolic use. The addition of obeloi – cooking equipment – for instance, would not have been possible without relevant originals from Italy of the 8th–6th century BC (first discussed by Déchelette 1913, esp. 191–243; Frey 1992, esp. 377). One may note that they appear only in a certain area during the 5th century and prove a special connection between Champagne tribes (‘Senoni’) and the Upper Adriatic Sea. The exclusive burial of cremated bones in a metal urn under a tumulus seems to have a similar meaning (Verger 1995, 441–2, fig. 46). It recalls a mos maiorum similar to that recorded by Homer for the burials of the heroes of Troy. Such burials show, especially during the 5th century north of the Alps, a spread of ideas to an area extending from Burgundy to the Central Rhine area and further to Bohemia and the Dürrnberg in the east. There are some examples of double burials of a warrior or man’s grave with grave goods and dating connected

to a primary grave of the Early La Tène period. Only the famous grave goods of La Gorge-Meillet (Morel 1876) in the Champagne or the grave of La Motte Saint Valentin in Courcelles-en-Montagne are to be mentioned here (Déchelette 1913, 101–151, pl. 34, fig. 13–22; Verger 1995, 353 ff., fig. 6–7): a simultaneous warrior grave together with a rich woman’s grave were positioned exactly above the central ‘metal urn’-grave in Courcelles, whereas in La Gorge-Meillet a considerably simpler warrior grave was positioned above with the same orientation, maybe focussed on the original grave ceiling. If we compare these finds of the Early Iron Age of ancient Europe, it becomes clear, that the above mentioned finds were prominent examples of a more widespread custom. Double or multiple burials of warriors with the same or similar equipment were quite common. One may think, they once mirrored personal war experiences but also displayed a common identity within a broader cultural background (Egg 2003; Stöllner 2010, 279–280). In the last mentioned context, a burial rite founded on a ‘brother-in-arms’ ideology can be another background of a heroic notion of warriordom. The cult around the divine twins, the Dioscuri, can be seen as one appropriate but analogous ideological model, as they were seen ‘as inseparable couple, whom was commemorated in every misery, especially in fight and storm, in an auxiliary way’ (Bethe 1903,7 1087–1123; v.Geisau 1967, 92–93); but I would not deny, that relevant critical research for the Early Iron Age cultures of ancient Europe is still not available. It becomes clear that the graves of the Glauberg mirror a time phenomenon, which is finally based on an amended version of elitist grave rites – that it is based on an amended version of a religiously founded ideology as well, becomes apparent thanks to the stone statues (in general Frey 2000; 2002). Comparable finds can be found in the cultural environment of the Rhine-Neckar-area (Fig. 14.2). Unfortunately a context for the almost identically made stone statues does not exist, so their original positions remain unsolved. Only an older stone statue from Hirschlanden in Württemberg gives some clues (Zürn 1964) (Fig. 14.1.4). Like the ones from the Glauberg, it was found next to a burial mound. Older reconstructions positioned the figure on the hilltop like a stele, but the findings from the Glauberg suggest a position next to the mound as more likely. The relationships between the graves of the mound are not yet definite, although the figure does quote all the important symbols of power of the late Hallstatt period: the dagger, the golden torc and the hat – equipment that was discovered in the Hochdorf grave. In comparison to the sculptures of the Early La Téne period, the nudity with erect phallus, perhaps a fertility symbol must be emphasized, which connects them with similar finds of the Upper Adriatic Sea and Italy. A corresponding grave is missing in Hirschlanden, whose attributes would equal those of the statue. So they probably did not even mean a certain real life person, rather an

14.  Between ruling ideology and ancestor worship: the mos maiorum of the Early Celtic ‘Hero Graves’

123

Fig. 14.2. Figural stone sculptures in the 6th to 4th century BC in Central and West Europe, mapping: Th. Stöllner, Bochum (after Bonenfant and Guillaumet 1998), with additions.

imagined type of ancestor or even a numen. Such a context is revealed by a similar wooden statue of the same age, which was found in a ford of the Saône near Seurre (Bonenfant and Guillaumet 1998, 21 ff.) (Fig. 14.1.3). Here the round shaped statue is not connected with a specific grave cult or a specific buried person. It guarded the crossing of the river. There is a second example for the ritual practices at graves from the early Celtic cultural area: let us take a look at Vix at the foot of Mont Lassois: a long-known grave area of the Hallstatt period lies at the foot of the site. The grave of ‘the lady of Vix’, discovered in 1949, with its chariot, golden torc and especially the 1.64 m tall bronze Vix crater is outstanding among the older burial mounds in the surrounding area (recently Rolley 2003). Only about 200 m away, a FrenchGerman research project analysed a rectangular enclosure with a lateral length of 23 m in 1992/93. Inside of the sidewalls at the only entrance two beheaded statues were

discovered – a sitting woman with a torc with pompons like on any other torc found in the grave of ‘the lady of Vix’ as well as a sitting warrior holding a shield, sword, tunic and leg pieces (Fig. 14.1.5). The iconographic relationship of the female statue with the close connection by an elaborate grave as well as the positioning of this statue in a temenos, or better a sekós, connects this find with the Glauberg. If there was a second grave that once had a counterpart to a heroic warrior grave is unknown. But the situation does show the importance of communal remembrance of female and male ancestors or even heroes as important transmitters to the after world. Glauberg and Vix as examples show better than others that we simply are not dealing with elite graves, but with ritually significant burial rites that were important for the disposal of elite persons. Even the finds from the ditch system could be seen as a symbol for cult rituals: among the animal bones, skulls are very prominent and the exclusive

124

Thomas Stöllner

Fig. 14.3. Salamis, tomb 1, base map (after Karageorghis 1975).

appearance of drinking bowls points to a type of ‘libationcult’ (Chaume et al. 1995; 2000). Even the well-known finds from the small oppidum of Roquepertuse in southern France, north-east of Marseille, give new evidence for the type of cult. There, excavations by Henri de Gérin-Ricard between 1919 and 1927, revealed a group of buildings, which were first seen as a ‘sanctuary’ of the 3rd to 1st century BC (Gérin-Ricard 1927, 19 ff.; Benoit 1955). Two sitting warrior statues as well as a porticus decorated with human faces being the decisive factor here: the original position of the two statues as well as the two-headed Hermes under the porticus influenced a lot of archaeologists and drew a certain picture: new excavations under Philippe Boissinot and the latest reconstruction of the existing stone fragments by Brigitte Lescure drew a completely different one (Lescure 1995, 76–77; B. Lescure in Baitinger and Pinsker 2002, 320 ff.): The settlement seems to be remodelled on a larger scale mainly in the 3rd century. This may have influenced the statues’ position close to the main gate or respectively the wall. Column bases in front of the large wall give evidence that they originally had been

positioned there. Today we have to revise the dating and the positioning as well. All in all there may have been eight sitting warriors statues – their compound armour with its raised back not dissimilar to the ones from the Glauberg, allow them to be dated into the 5th/4th century, instead of the originally thought 3rd century. The close connection with the statues of the Glauberg is not only shown through the compound armour (Fig. 14.1.2): at the head of the so called ‘Hermes’, a Janus face may have been present in the two characteristic upward pointed leaves of a leaf crown, which were later, during a remodelling in the 3rd century, removed (Jacobsthal 1944, 3–7 pl. 1–5; Lescure 1995, 76, fig. 73). In Roquepertuse the imagines maiorum were probably commemorated while entering the compound and their meaning for the settlement and the collective were honoured. Roquepertuse is not alone: finds from Entremont and some of the oppidum of Glanum (Saint Rémy-de-Provence) show a similar local tradition (Jacobsthal 1944; Benoît 1955; Bonenfant and Guillaumet 1998). It seems to have been common in the hinterland of Greek emporia and colonies. Thus an important semantic background can be recognized

14.  Between ruling ideology and ancestor worship: the mos maiorum of the Early Celtic ‘Hero Graves’

125

Fig. 14.4. Chiusi, tomba della Pania, picture scenes of the ivory pyxis (after Cristofani 1971).

for this kind of ‘hero worship’: their very special bond and their important cultic function for larger settlements. In summary: in view of the graves and stone statues from the Glauberg it became apparent, that in the Early Celtic area idealised ancestors and a grave cult were closely connected. This does not mean that a certain person had to be idealised, but that the representation used a rather common picture of a transmitting hero, either female or male. The examples from Vix, Hirschlanden and the Glauberg show, that it was a newly introduced iconography that differs in its human picture immensely from everything known until then. This iconography is introduced at the end of the 6th century. That is especially obvious in Hirschlanden. The younger warrior statues from the Glauberg are clearly influenced by the former iconography of the Late Hallstatt period (also Frey 2000; 2002). Nevertheless, differing from the Late Hallstatt period, the representation of the warlike ‘hero’ in the grave rituals, the images and even in the sanctuaries does not appear until the Early La Téne period. These are the real new aspects. This amended version takes a lot of iconographic adoptions from the Mediterranean area. Especially during the Late Hallstatt period, models from the Italian-Adriatic area are evident – particularly in regard to the iconographic aspect, if we think of the ‘nudity’, the positioning of the hands or even the ithyphallic presentation. That sometimes imports of Greek kouroi from the 6th century were found in Adriatic Italy, I will only mention here (‘Kouroi Milani’: Luni and Cardone 1998; Frey 2002, 213 ff.). The ‘warrior of Capestrano’ in the centre of an Abruzzi necropolis should be mentioned: it was discovered there without any connection to a burial mound (Moretti 1936; Baitinger and Pinsker 2002, 316–317, no. 127). According to the inscription on the side, he is supposed to be King Nevius Pompulidenus. Like many other goods found in Central Italian elaborate

graves, this statue demonstrates the ritual and political dominance of the deceased. In addition, Etruscan elaborate graves show a distinctive ancestor worship, which manifests in grave statues and imagines maiorum in a broad regional variation (Colonna 1992; Prayon 1998; Colonna 1999; Koch 2008) (Fig. 14.6). But most of the time these images were integrated into the grave itself. The imagines maiorum seem to be mostly an element of ancestor worship for the ruling families. The grave complex seems rather stately, where memorial worship is monumentalised with a special grave architecture and they get a certain use as a demonstration of a powerful ancestry. We may understand the stairway on the burial mound Melone di Camuccia Seconda in Cortona in this aspect (Marzi 1998; Zamarchi-Grassi 1998). This outside staircase displays a hero’s fight with a lion on each side of the balustrade. The staircase was possibly used for a memorial worship celebrated at the foot of a burial mound. The sculptures from the Early Celtic period can be connected in many aspects with the sculptures from the Early Iron Age of Tyrrhenian and Adriatic Italy (Etruria, Picenum) (Frey 2002). But as clear as the iconographic background may seem, just as different is its social context, and therefore the meaning, in which they evolved. We want to take a closer look at the underlying social praxis and try to use it for a comparison of similar social structures. Because we gain nothing with labels like a ‘hero’s grave’, ‘elaborate grave’ or ‘princely grave’, and it is important to know which social ideal is meant: the ‘noble under equals’ or the ‘regulus’, the ‘prince’ or the ‘religious king’ standing above all (summarizing recent positions e.g. Schweizer 2008; 2012). Can we get information about the type of grave rituals, or better, how close is the ‘ritualisation’ of a special deceased connected with certain social developments? How is the burial performance of a ‘hero’ expressed in different kinds of social historical contexts?

126

Thomas Stöllner

In between methodical access and the need for interpretations The short summary concerning the amended version of ‘heroic’ martial equipment at the beginning of the La Tène period, directed us into certain needs for interpretation. The interpretation of ‘heroic’ rites and ideological patterns is often interwoven with the materiality of the elaborate graves themselves. This point of view often ends in a static view in comparing phenomena, without being able to approach the present myth or the variety of religious beliefs behind. This problem can surely be solved by taking a detailed look at the performative attributes of the burial, but there are limits as well (e.g. Veit 2005; also different essays in Kümmel et al. 2008). It may seem even more difficult to find the ‘Homeric’ burial ideal or even to try to suggest the transmission of such rites to other cultural areas and their burial motives. The transfer, as done substitutionally by S. Verger (1995, 352–353 (‘un rite étranger’)) or V. Karageorghis (1975, esp. 54 ff.; 1998, 18–41; see also Rupp 1988)3 shows that modern theories about the ways in which transmission of ideas between cultures might have been interpreted depend on the historically-determined traditions within different academic disciplines. Celtic archaeologists such as S. Verger, influenced by the idea of a Mediterranising of the North, argue for a ‘fascination for the South’ in the Celts (the idea of a Mediterranising of the North: see e.g. Brun 1994; Krausse 2004), while Cypriot archaeologist V. Karageorghis instead tries to connect the graves in the city of Teukros (Salamis: e.g. Karageorghis 1975) culturally and ethnically with a Greek, ‘Homeric’ tradition. Verger’s interpretation is still influenced by the image of the ‘barbaric primitive’ culture as a primarily a recipient culture of southern – in the widest sense Mediterranean – ideas and images. Karageorghis notes the parallels between the influence of the neighbouring Greeks on Early Iron Age material culture with the modern political idea of ‘enosis’ (the connection with the Greek ‘homeland’), materialized in the modern Greek dominance on the island of Cyprus. Apart from the historical and political dimensions of such interpretations, the question remains, what characterises a ‘Homeric’ grave, and can it ever be reconstructed due to the archaeological find contexts? Is it the whole bundle of burial rites stated in the Iliad during the burial of Achilles and his companion Patroclos, or was it just parts of it? Did Antique populations realize it in this way? Could one think, however, that the more it felt canonical, the greater the distance in time from the conditions of the ‘Homeric’ in the Early Iron Age? It is mainly the cremation burial in a magnificent metal vessel as well as human sacrifices that were seen as the constructing element for a ‘Homeric’ ritual (e.g. Guggisberg 2008; for Homer’s epic see also Raaflaub 1998; Ulf 2012). Achilles was buried together

with his closest ‘companion’ Patroclus under a tumulus in a magnificent metal vessel: ‘Your mother gave a two-handled jar of gold. She said it was a gift from Dionysus, something made by illustrious Hephaestus’ as can be found in the 24th song of the Odyssey. Without following this argument any further, it should be hinted, that an evaluation of the Homeric ideal of a ‘heroic- or hero’s grave’, if there is not a further literary evidence of a reference to Homer’s epic , is difficult even in the Greek cultural area. It is nearly impossible for the Celtic influenced western- and central Europe. It will be smarter, to call burials, which are ostensibly connected with the described ritual, less ‘Homeric’ than ‘heroic’,4 if further aspects are evident and derived through the burial itself and its picture- and narrative tradition. What should now be the main aspects that would enable us to describe this phenomenon? First there is the political, social and ritual constellation, which was responsible for the emergence of a grave rite or hero worship. M. Guggisberg described the re-appearance of burials in a ‘Homeric’ style in Athens during the 5th century with a deliberate construction due to political reasons or emotional romantic attitude of the past (Guggisberg 2008, 299 ff. 307). The ritual recourse during the 5th century culminates in the ‘heroisation’ of the Tyrannicides Harmodios and Aristogeiton and continues, though in another way, with the ‘state funeral’ of the 192 fallen warriors during the battle of Marathon in the tumulus right there. In both cases they are surely a political manifestation of fledgling Athenian democracy but also a break with the familiar traditions of the Athenian aristocracy (Whitley 1994). Both even connect the massive revival of hero worship in the Greek poleis in the home country as well as in the colonies with the political and social changes of the 8th century (Stupperich 1977; Guggisberg 2008, 301). Second, the question, which precise ancient customs connect the elaborate grave with the hero worship in the graves? To get buried the mos maiorum way, means a conscious devotion to traditional ‘burial rites’, but at the same time a renunciation of the usual rites, for example an inhumation burial instead of the usual cremation. Here, the ‘hero’s grave’ can certainly meet the overall appearance of an elaborate burial custom – whose main attributes were the ritualistic separation from other burials. That could include outstandingly rich burial equipment or the implementation of objects and customs being integrated from a foreign cultural level that was regarded as superior and admirable (Kossack 1974; Rowlands 1999; Steuer 2006). ‘Breaking the rules’ from the perceived normative customs can be seen as a social manifestation as well. But two problems arise: first, besides the manifold problems of the evidence, is there really an ‘apparent’ cult forming at the graves, maybe immediately after or some time later? This depends a lot on the settlement development and the ideological continuity

14.  Between ruling ideology and ancestor worship: the mos maiorum of the Early Celtic ‘Hero Graves’

127

Fig. 14.5. Anthropomorphic statues in graves in Central Italy (7th–6th century BC), mapping: Th. Stöllner, Bochum (after Colonna 1992; Prayon 1998), with additions.

of the societies involved (for Athens e.g. Morris 1987). And second: was there a common ritual language, using elements of ‘heroic’ burial rites in different ways, without just being ordered ‘by the state’ to honour the actions of people and to exploit them politically like in the Greek cultural circle? Essential as a third topic is the central narrative, which can be revealed through the myths of a cultural circle (important aspects see Lévi-Strauss 1980; Vernant 1995, Assmann and Assmann 1998; Burkert 1999). Oral or written myths, iconography and abbreviated picture codes are closely related and complement each other in cult and burial practices (Dumézil 1949, 64; e.g. on the example of Verucchio: Kossack 1992). The background becomes clear to archaeologists through the pictorial and written account, sometimes even by analogy with neighbouring cultures. One

may distinguish between a deeper layer of older lore and younger inclusions, as we can learn from the literary evidence in Homer’s epic tales or the works of Hesiod (e.g. Vernant 1995, 55 ff.; Schweizer 2012, 467 ff.). It is evident, that stories can encourage ‘hero worship’ and elaborate burial rites. Finally there are different ways in which historical traditions of ‘epic’ myths and burial rites and their represent­ ation are transmitted and adopted as native or foreign traditions, and these have to be researched. The transmission and appropriation of external traditions into the beliefs of certain population groups, for example the elite, is a very complex process, not just an adaption of foreign concepts. It is rather an exchange and transformation of narratives. I would follow and agree with in this aspect the concept of myth as it was argued by W. Burkert (Burkert 1999, 13).

128

Thomas Stöllner

Fig. 14.6. Anthropomorphic statues in graves in Central Italy (7th–6th century BC), 1: Vetulonia, tomba della Pietrera; 2: Chiusi, Canope di Dolciano, 3: Cortona, tumulo II del Sodo di Cortona, the altar-terrace with a lion-fight by a warrior-hero (1–3 after Principie truschitra Mediterraneo ed Europa. Catalogo del Mostra Bologna, Marsilio: Venezia, 142, 175); 4: Ceri, Tomba delle statue (after Prayon 1998).

As easy as these four topics seem to be as a guideline for a research on ‘heroic patterns’, there are as many pitfalls – especially considering the background of shared ideological concepts, which can only be made accessible for the archaeologist through similar phenomena like symbolic imagery, grave goods and archaeological context. Looked at ostensibly, many phenomena are similar through time and space. If we put them eclectically as an analogy in a row, they would only give the conclusion of similar views of

the afterlife and ritual concepts. Then we would not be far from a general religious- and mythological-concept, but it would be doubtful, if they would reflect more than just the researcher’s own perspective. But for the situation north of the Alps it is important to consider, that a ‘heroic’ burial cult very possibly had different social rules. If we take a look at the seemingly ‘Homeric’ cauldron burial from the ‘princely graves’ of Salamis (Fig. 14.3), we detect similar elements, but now

14.  Between ruling ideology and ancestor worship: the mos maiorum of the Early Celtic ‘Hero Graves’

129

Fig. 14.7. 1–4 Lefkandi, Toumba, heroic burials beneath the building, 1 base map of the double-grave, 2 metal amphora, in situ position and reconstruction, 4 reconstruction of the building (after Popham et al. 1993; Popham 1994), 5: Kourion, Kaloriziki, grave 40, metal amphora, decorated handle and decorated rim-plate (after McFadden 1954).

130

Thomas Stöllner

Fig. 14.8. ‘Hero worship/cults’ and ‘homeric’ cremation burials between the 11th and the 7th century BC, mapping: Th. Stöllner, Bochum (after Coldstream 1976; Morris 1999; Guggisberg 2008), with additions.

paired with ostentatious grave pomp. We can discover the same for Etruscan Central Italy, where the burial rites have been completely different since the Villanovan time: the simple warrior disposal of the 9th and 8th century disappears and makes way for an overreaching pompous grave rite with ostentatious, purely symbolic weaponry (Tarquinia, Selciatello, Grotta Gramiccia and Regolini-Galassi: e.g. Putz 2007). In the graves of the Early Oriental period cremation burials in metal vessels were discovered interestingly mostly in Campania (Pontecagnano) (D’Agostino 1977; Schweizer 2003). Together with the ability to write also pictured stories, i.a. with reference to the Homeric myth circle, found their way into the grave cult (Chiusi, tomba della Pania:

Cristofani 1971) (Fig. 14.4). Surely this is good evidence that the adventures of the Homeric heroes were popular? But how they were understood: certainly there are depictions of the same stories, but if we consider their ritually different context, they were probably differently appropriated. The same is true if we compare the death cult that included the imagines maiorum exemplified by the large statues. It was part of a concept of mos maiorum probably in all areas of Etruria (Colonna 1992; Prayon 1998; Koch 2008) (Fig. 14.5–6). These cult concepts seemingly also influenced the surrounding areas all the way to Central Europe. In Central Italy one can only suggest that – not unlike the Greek cultural area – massive social-economic changes influenced this

14.  Between ruling ideology and ancestor worship: the mos maiorum of the Early Celtic ‘Hero Graves’

131

Fig. 14.9. Eretria, ‘heroon’ at the western gate, 1: heroon from east (Bérard 1970, pl. 1,1); 2: grave 6 with cauldron with lid (Bérard 1970, 14, fig. 2).

development. In Etruria it resulted in the rise of the Etruscan urban societies. At the beginning, particularly in the Eastern Mediterranean, there were only single and exceptional burials (e.g. Toumba: Popham et al. 1993; 1982; Popham 1994; Kaloriziki Grab 40: McFadden 1954; general also Morris 1999, 61) (Fig. 14.7), which certainly displayed not a canonical cult or ritual concept but more variety of ritual and social practice; this perhaps became different with the social and political changes during the 8th century: heroic cult practice in ‘princely graves’ increased, but are now, as has to be shown, in a completely different context (e.g. the Greek colonisation; the rise of urban societies). Hero worship flourished especially in distant colonies and was perhaps partially amended by aristocratic or leading families for political purposes, either in urban Greek or Etruscan contexts or by indigenous groups from the hinterland. This is different from the constitutional kings of geometric time, the ‘basileis’, who came into their function through their ancestors.5 But especially in Greece we are missing elaborate graves in an oriental sense, a proper king’s grave. If a magnificent burial in Argos, an inhumation burial with parade armour and andirons and obeloi can be interpreted in this direction, stays unclear, because typical elements of the ‘heroisation’ of the dead are missing especially in that example (e.g. funeral monument, libations, offerings: Courbin 1957). According to such examples the deceased warriors resembled the ‘hero’, who, magnificently dressed,

meets his ancestors and peers in the afterlife. For that purpose also elements of what may have been an ostensibly strange world of Eastern Mediterranean myths and epics were used and transformed in a new ritual context. The Celts probably also encountered such stories and myths of heroic deeds and persons during their early migrations even before the Great Migration of the 6th century in Southern France and Upper Italy. So it is possible that tribal nobilities integrated them for a specific ‘warrior ethos’ into their grave rituals. It is noteworthy to stress again the connection of this ethos with a different social class during the Early La Téne period than during the Hallstatt period. It seems that the ‘king of an army’ (according to German word ‘Herzog’) became important for an increasingly militant society (Dobesch 1996).

Parallels and occurrences in Greece during the Early Iron Age? An area which offers some help for answering the question how social and religious practices did interfere in the context of the ‘hero’s grave’ and accordingly the ‘hero cult’, is Geometric to Early Archaic Greece. Picture, written and material relics are found in such a large concentration, that a more detailed discussion is possible at last. Moreover many of the above mentioned grave rites can be closely associated to the Greek world, so it may be that they were

132

Thomas Stöllner

conducted in neighbouring areas or they had at least an indirect connection with Central Europe. Especially the Homeric epics tell us about the grave rites of heroes in the Greek cultural circle: but these stories also have influenced our modern views on a wider scale, so it seems difficult to separate them: Homer’s epic tales have influenced Greece and the western Mediterranean world since the 8th century. His epics also reached Central Italy during the early 7th century. Since then we have to review many questions about ‘heroic’ or more specific ‘Homeric’ burial rites in a receptive historical way: the flourishing and the spreading of ‘Hero’s cults’ during the early 8th century has been closely connected by many scholars with the tradition and circulation of the Homeric epic itself (in general Coldstream 2003; Calligas 1988; Catling 1995; Morris 1999) (Fig. 14.8). The works of Homer and Hesiod are of course based on much older traditions and we accept today, that Geometric Greece in particular had an exceptional influence on social order and the rise of interconnected cult practice. That is the reason why we are able to compare graves like the so-called Heroon of Lefkandi to the oldest prototypes of a burial rite, later on also mentioned by Homer. The double grave under the apsidal building of Toumba in the second half of the 11th century BC is the best mirror for aspects of these burial rites (Popham et al. 1993; Popham 1994) (Fig. 14.7.1–4): the cremation burial of the man took place in an exceptional bronze vessel, at the time of the burial in a probably old Cypriote amphora, an heirloom (e.g. Guggisberg 2004). An inhumation burial of a richly appointed female was right next to him. Four horses were given to the dead in a separate grave. Whether the apsidal building can be interpreted as a memorial one (heroon), or if it was erected even earlier – in the sense of a residential building – unfortunately cannot be solved by its stratigraphic context. After a certain time, a monumental mound was raised and the use of an also richly appointed secondary burial necropolis was started. The burial of Lefkandi is so special, that it is difficult to identify similar graves: they are very rare until the 8th century (for example on Crete and Cyprus: e.g. McFadden 1954; Catling 1995, fig. 7). Geometrical burials are usually unadorned inhumation and cremation burials, sometimes with weapons, which emphasize a possibly martial element of the ‘Dark Ages’. Hero worship had its peak especially during the late 8th century as pointed out by Nicolas Coldstream (1976) and Ian Morris (1987; 1999), and this is consistent with the detected acceptance of ‘Homeric’ cremation burials in cauldrons. It had been asked if the cult about the epic ancestors as well had – like the burial rites – its roots in the late Mycenean and Early Geometrical period (Coldstream 1976; Morris 1987; 1999). Their monuments were still visible during later periods and thus evoked hero worship, because the contemporary period was known as the iron one, following the fourth era, that, according to

Hesiod, was ‘better and fairer’, a time ‘of the heroes of divine origin, those who lived before us on earth’s endless plains, known as demigods by today’s generation’ (Hesiod, Works and Days 158–160). The amended version and random growth of hero worship during the late 8th century, a time of radical changes in the whole cult practice of the early Greeks, seems to be essential. James Whitley (1994) distinguishes different types of hero worship: those who became heroes through their actions, those heroes known through Homer’s work, and also memorial cults celebrated at the graves. We will take a closer look at the kalós thánatos and accordingly at those deceased defined through their actions (Vernant 1995). The heroon at the west gate of Eretria is a well known example: there the Swiss School discovered during their excavations in the 1960s beneath the Hellenistic town fortification a small necropolis with 20 graves from the 8th and 7th centuries (Bérard 1970) (Fig. 14.9). The six male burials show a surprisingly homogeneous grave rite, because of the cremation, the burned weapons, heirlooms and an almost identical setting of the graves. The context is – during a time without weapon grave goods in Greek necropoleis – even more remarkable. A platform was erected later above those burials. Claude Bérard interpreted this as an heroon, which the polis of Eretria built as an honour for the fallen heroes of the battle against Chalkis. This puts the heroon of Eretria in the same line as the emerging hero worships, in honour of those ‘mortals’, who earned outstanding power in the afterlife through their special efforts in life (e.g. also Morris 1987; Whitley 1994; Guggisberg 2008). The cult secured their demigod status and was an important identification attribute for the polis. Therefore a new kind of hero evolved during the 8th century – who matches the ‘ideal of society’ as mentioned in the Homeric epic, of a warrior hero among equals, for instance like the literal sense for hero given by Homer.

Conclusion The hero worship that developed in Greece during the 8th and 7th centuries obviously was independent of ‘leadership’ or royal status; a basileios who was not necessarily fitted with heroic attributes; or better: the construct and imagination of a hero was not necessarily connected with the king as a sacral mediator between the here and now and the afterlife – the kingdom of Sparta being the exception, which held on to the old traditions. Compared to the contemporary appearances in Levante, in Etruria, but also north of the Alps, two kinds of ‘heroic’ grave rites and grave cults can be distinguished. Although it always became connected with the special person and their actions, grave cults of this kind were interwoven into the social construct of a

14.  Between ruling ideology and ancestor worship: the mos maiorum of the Early Celtic ‘Hero Graves’ constitutional royal, princely or sacral kingdom, once again into an oligarchic society of warriors or aristocratic elite of warriors. The changes in Greek ‘hero cults’ certainly were a mirror of the political, economic and social upheavals. What do we have to expect for the area of the Early Keltiké of the 5th century? The richly equipped elaborate graves of the Late Hallstatt period stand, with all their royal and sacral attributes, at the end of a long ideal continuity, which goes back into the Late Bronze period of the 13th century; some single elements of the burial rites can be followed back till this period. Their discontinuity with the Early La Téne mirrors a radical social and religious change. The elaborate graves of this period, found in the northern periphery of the Hallstatt area, seem to be bound to new ideals. It is a warrior ethos that apparently absorbed aspects of heroism from the mos maiorum, fitting for an elite, which established itself by conflict and competition with the southern Hallstatt centres. To this extent, these elements were producing a new identity for their bearers. The ritual aspects are easy to explain, they even seem to be chosen specifically for integration into the burial rites north of the Alps. The image of the aristocratic ‘warrior king’, who defies dangers with his comrades in war, was seemingly much more appropriate for the ruling elites on the eve of the Celtic Migration. That these elements of self-manifestation are identifying attributes, which particularly in the coming 150 years of the Celtic Migration were crucial and led to a ‘Celtic’ identity beyond the tribe, we can only imply here.

Notes 1

2 3 4

5

The article consists of several studies that were assembled for a lecture first delivered in 2005. It is a reworked and added version of this lecture. I particularly want to thank Dr. Constance von Rüden, Heidelberg, for her critical and inspiring discussion on the topic. In general the concept of ‘habitus’ as it had been introduced by N. Elias and P. Bourdieu: Bourdieu 1982; in a more specific sense on behalf of wear-customs and clothes: Rummel 2007. Especially when constructing some sort of ideological entity: e.g. a Celtic, or an Etruscan or an Aegean one: for Aegean Archaeology e.g. Catling 1995, 123–135. Heroic will be used in a wider sense than warrior heroic, even more in a sense of persons being eminent for their societies according to their achievements but also according to their ritually significant origin. In Linear B-texts we know the word ‘p/qa-si-re-su’ as someone who is controlling metal and craft production: Schmidt 2006; Morris 1999, 64–65 which points to a second class of elites that become more important in the Early Iron Age, see also Hamilakis 2002 for late Bronze Age Crete.

133

Bibliography Assmann, A. and Assmann, J. 1998. Mythos. In H. Cancik et al. (eds.) Handbuch religionswissenschaftlicher Grundbegriffe 4, 179–200. Stuttgart u.a.: Kohlhammer. Baitinger, H. and Pinsker, R. (eds.) 2002. Das Rätsel der Kelten vom Glauberg: Glaube-Mythos-Wirklichkeit Ausstellungskatalog Frankfurt 2002. Stuttgart: Theiss. Baitinger, H. 2010. Der Glauberg – ein Fürstensitz der Späthallstatt-/ Frühlatènezeit in Hessen. Mit Beiträgen von P. Kresten, B. Zickgraf, I. B. Wagner und G. A. Wagner. Materialien zur Vor- und Frühgeschichte von Hessen 26, Glauberg-Studien 1. Wiesbaden: Landesamt für Denkmalpflege Hessen. Benoît, D. 1955. L’art primitif méditerranéen de la vallée du Rhône, Aix-en-Provence. Gap: Ed. Ophrys. Bérard, C. 1970. L’hérôon à la porte de l’ouest. Eretria. Fouilles et recherches III. Bern: Francke. Bethe, E. 1903. Dioskuren. In: G. Wissowa, Paulys Real­ encyclopädie der classischen Altertumswissenschaft 9, 1087– 1123. Stuttgart: Druckemüller. Bonenfant, P. P. and Guillaumet, J. P. 1998. La statuaire anthro­ pomorphe du premier âge du Fer, Ann. Litt. Univ. de Franche Comté 667, Sér. Arch. et Préhist. 43. Besançon: Presses Univ. Franche-Comté. Bourdieu, P. 1982. Der Sozialraum und seine Transformationen. In P. Bourdieu (ed.) Die feinen Unterschiede – Kritik der gesellschaftlichen Urteilskraft, 171–210. Frankfurt am Main: Suhrkamp. Brun, P. 1994. From the Hallstatt to La Tène Period in the Perspectives of the Mediterranean World Economy. In K. Kristiansen and J. Jensen (eds.) Europe in the First Millenium B.C., Sheffield Archaeological Monographs 6, 57–65. Sheffield: J. R. Collis Publications. Burkert, W. 1999. Antiker Mythos – Begriff und Funktion. In H. Hofmann (ed.) Antike Mythen in der europäischen Tradition, 11–26. Tübingen: Attempo. Calligas, P. G. 1988. Hero-Cult in Early Iron Age Greece. In R. Hägg and N. Marinatos (eds.) Early Greek Cult Practice, 229–34. Stockholm: Åström. Catling, H. 1995. Heroes returned? Subminoan Burials from Crete. In J. B. Carter and S. P. Morris (eds.) The Ages of Homer. A Tribute to Emily Townsend Vermeule, 123–35. Austin: University of Texan Press. Chaume, B., Olivier, L. and Reinhard, W. 1995. Das keltische Heiligtum von Vix. In A. Haffner (ed.) Heiligtümer und Opferkulte der Kelten, Archäologie in Deutschland, Sonderheft. Stuttgart: Theiss. Chaume, B., Olivier, L. and Reinhard, W. 2000. L’enclos hall­ stattien de Vix ‘Les Herbues’: Un lieu culturel de type aristocratique? In Mailhac et le premièr âge du Fer en Europe occidentale. Hommages à Odette et Jean Taffanel. Actes du Colloque International de Carcassonne 1997, 311–27. Lattes: CNRS. Coldstream, J.N. 1976. Hero Cults in the Age of Homer. Journal of Hellenic Studies 96, 8–17. Coldstream, J.N. 2003. Geometric Greece 900–700 BC. 2nd edition, London/New York: Routledge.

134

Thomas Stöllner

Colonna, G. 1992. Apportie truschi al l’orientalizzante ‘piceno’: il caso della statuaria. In M. Dardari (ed.) La civiltà picena nelle Marche. Studi in onore di G. Annibaldi, Ancona 1988, 92–127. Ripatransone: Maroni. Colonna, G. 1999. Die Steinplastik. In L. Franchi Dell‘Orto (ed.) Die Picener – Ein Volk Europas, Catalogue of exposition Schirn 1999/2000, 104–8. Roma: DeLuca. Courbin, P. 1957. Une tombe géométrique d’Argos, Bulletin de Correspondance Hellénique 81, 322–86. Cristofani, M. 1971. Per una lettura della Pisside della Pania, Studi Etruschi 39, 63–89. D’Agostino, B. 1977. Tombe ‘principesche’ dell’orientalizzante antico da Pontecagnano, Monumenti antichi 49, Serie Miscellanea 2. Roma: Accad. Nazionale dei Lincei. Déchelette, J. 1913. La collection Millon. Paris: Geuthner. Dehn, W. and Stöllner, Th. 1996. Fußpaukenfibel und Drahtfibel (Marzabottofibel). Ein Beitrag zum kulturhistorischen Verständnis des 5. Jh. v. Chr. In Th. Stöllner (ed.) Europa celtica. Untersuchungen zur Hallstatt- und Latènekultur. Veröffentlichnugen des Vorgeschichtlichen Seminar Marburg, Sonderband 12, 1–55. Marburg: Leidorf. Dobesch, G. 1996. Überlegungen zum Heerwesen und zur Sozialstruktur der Kelten. In E. Jerem et al. (eds.) Die Kelten in den Alpen und an der Donau. Akten Internat. Symp. St. Pölten 1992. Budapest: Archaeolingua. Dumézil, G. 1949. L’Hértiage indo-européen à Rome. Paris: Gallimard. Echt, R. 1999. Das Fürstinnengrab von Reinheim. Studien zur Kulturgeschichte der Früh-La-Tène-Zeit, Saarbrücker Beiträge zur Altertumskunde 69. Bonn: Habelt. Egg, M. 2003. Eine ungewöhnliche Bestattung der Frühlatènezeit in Novo Mesto in Slowenien. Jahrbuch des RömischGermanischen Zentralmuseums Mainz 46 (1999), 317–56 Frey, O.-H. 1992. I Gallinel Piceno. In M. Dardari (ed.) La civiltà picena nelle Marche. Studi in onore di G. Annibaldi, Ancona 1988, 364–81. Ripatransone: Maroni. Frey, O.-H. 1998. The Stone Knight, the Sphinx and the Hare: new aspects of early figural Celtic Art, Proceedings of the Prehistoric Society 64, 1–14. Frey, O.-H. 2000. Keltische Großplastik. In J. Hoops (ed.) Reallexikon der Germanischen Altertumskunde 16, 395–407. Berlin/New York: De Gruyter. Frey, O.-H. 2002. Menschen oder Heroen? Die Statuen vom Glauberg und die frühe keltische Großplastik. In H. Baitinger, and R. Pinsker, (eds.), Das Rätsel der Kelten vom Glauberg: Glaube-Mythos-Wirklichkeit Ausstellungskatalog Frankfurt 2002, 208–18. Stuttgart: Theiss. Frey, O.-H. 2008. Fund vom Glauberg und die erste Phase der ‘Keltischen Kunst’. In Glauberg 2008, 35–56. Frey, O.-H. and Herrmann, F.-R. 1997. Ein frühkeltischer Fürstengrabhügel am Glauberg im Wetteraukreis, Hessen, Germania 75, 459–550. Geisau, H.v. 1967. Dioskuroi. In K. Zigeler und W. Sontheimer (eds.), Der kleine Pauly. Lexikon der Antike, 92–94. Stuttgart: Druckenmüller. Gérin-Ricard, H. de 1927. Le sanctuaire préromain de Roquepertuse à Velaux (Bouche-du-Rhône). Marseille: Soc. de Statistique, d‘Histoire et d‘Archéologie de Marseille et de Provence. Glauberg 2008. Hessisches Landesmuseum Darmstadt und

Landesamt für Denkmalpflege Hessen (eds.) Der Glauberg in keltischer Zeit. Zum neuesten Stand der Forschung, Öffentliches Symposium 14.–16. September 2006, Fundberichte aus Hessen, Beiheft 6. Darmstadt: Landesamt für Denkmalpflege Hessen. Guggisberg, M. 2004. Keimelia: Altstücke in fürstlichen Gräbern diesseits und jenseits der Alpen. In M. Guggisberg (ed.) Die Hydria von Grächwil. Zur Funktion und Rezeption mediterraner Importe in Mitteleuropa im 6. und 5. Jahrhundert v. Chr. Schriften des Bernischen Historischen Museums 5, 175–192. Bern: Bernisches Historisches Museum. Guggisberg, M. 2008. Gräber von Bürgern und Heroen: ‘Homerische’ Bestattungen im klassischen Athen. In C. Kümmel, B. Schweizer and U. Veit (eds.), Körperinszenierung – Objektsammlung – Monumentalisierung: Totenritual und Grabkult in frühen Gesellschaften, Archäologische Quellen in kulturwissenschaftlicher Perspektive, Tübinger Archäologische Taschenbücher 6, 287–318. Waxmann: Münster. Hamilakis, Y. 2002. Too Many Chiefs? Factional Competition in Neoplatial Crete. In J. Dreissen, I. Schoep and R. Laffineur (eds.) Monuments of Minos. Rethinking The Minos Palaces, 179–99. Liège: Univ. de Liége. Herrmann, F. R. 2002. Der Glauberg. Fürstensitz, Fürstengräber und Heiligtum. In Baitinger and Pinsker 2002, 90–107. Herrmann, F. R. 2005. Glauberg – Olympia des Nordens oder unvollendete Siedlung? In J. Biel and D. Krausse (eds.) Frühkeltische Fürstensitze: Älteste Städte und Herrschaftszentren nördlich der Alpen? Internationaler Workshop zur keltischen Archäologie in Eberdingen – Hochdorf 12. und 13. September 2003, Archäologische Information aus Baden – Württemberg 51, 18 – 27. Stuttgart: Landesamt für Denkmalpflege. Jacobsthal, P. 1944. Early Celtic Art. Oxford: The Clarendon Press. Karageorghis, V. 1975. Salamis: Die zyprische Metropole des Altertums. Lübbe: Bergisch Gladbach. Karageorghis, V. 1998. Greek Gods and Heroes in Ancient Cyprus. Athens: Commercial Bank of Greece. Kimmig W. and Rest, W. 1954. Ein Fürstengrab der späten Hallstattzeit von Kappel am Rhein. Jahrbuch des RömischGermanischen Zentralmuseums Mainz 1, 179–216 Koch, L. 2008. Menschengesichtig, menschengestaltig – menschenähnlich? Die Anthropomorphisierung von Leichenbrandbehältern während des 8. und 7. Jahrhunderts v. Chr. in Etrurien. In In C. Kümmel, B. Schweizer and U. Veit (eds.), Körperinszenierung – Objektsammlung – Monumentalisierung: Totenritual und Grabkult in frühen Gesellschaften, Archäologische Quellen in kulturwissenschaftlicher Perspektive, Tübinger Archäologische Taschenbücher 6, 495–516. Waxmann: Münster. Kossack, G. 1974. Prunkgräber. Bemerkungen zu Eigenschaften und Aussagewert. In G. Kossack (ed.) Studien zur Vor- und Frühgeschichtlichen Archäologie. Festschrift für Joachim Werner. Münchner Beiträge zur Vor- und Frühgeschichte Ergänzungsband 1, 3–33. München: Beck. Kossack, G. 1992. Lebensbilder, mythische Bilderzählung und Kultfestbilder. Bemerkungen zu Bilderszenen auf einer Thronlene von Verucchio. In A. Lippert and K. Spindler (eds.) Festschrift zum 50jährigen Bestehen des Institutes für Ur- und Frühgeschichte der Leopold-Franzens-Universität Innsbruck, 231–46. Bonn: Habelt. Krausse, D. 2004. Komos und Kottabos am Hohenasperg?

14.  Between ruling ideology and ancestor worship: the mos maiorum of the Early Celtic ‘Hero Graves’ Überlegungen zur Funktion mediterraner Importgefässe des 6. und 5. Jahrhunderts aus Südwestdeutschland. In M. Guggisberg (ed.) Die Hydria von Grächwil. Zur Funktion und Rezeption mediterraner Importe in Mitteleuropa im 6. und 5. Jahrhundert v. Chr. Schriften des Bernischen Historischen Museums 5, 193–201. Bern: Bernisches Historisches Museum. Kümmel, C. Schweizer, B. and Veit, U. (eds.) 2008. Körperins­ zenierung – Objektsammlung – Monumentalisierung: Totenritual und Grabkult in frühen Gesellschaften, Archäologische Quellen in kulturwissenschaftlicher Perspektive, Tübinger Archäologische Taschenbücher 6, 495–516. Waxmann: Münster. Lescure, B. 1995. Das kelto-ligurische ‘Heiligtum’ von Roquepertuse. In A. Haffner (ed.) Heiligtümer und Opferkulte der Kelten, Archäologie in Deutschland, Sonderheft, 75–84. Stuttgart: Theiss. Lévi-Strauss, C. 1980. Myth and Meaning (Mythos und Bedeutung. Fünf Radiovorträge. Gespräche mit C. Lévi-Strauss, ed. by A. Reif). Frankfurt am Main: Suhrkamp. Luni, M. and Cardone, M. 1998. I ‘Kouroi Milani’ a Osimotra seicento e settecento, Atti della Accademia nazionale die Lincei, Rendiconti Classe di scienze moralistoriche e filologiche 9/4, 669–706. Lorenz, H. 1978. Totenbrauchtum und Tracht. Untersuchungen zur regionalen Gliederung der frühen Latènezeit. Berichte der Römisch-Germanischen Kommission 59, 1–380. McFadden, G. H. 1954. A Late Cypriote III tomb from KourionKaloriziki Nr. 40, American Journal of Archaeology 58/2, 131–42. Marzi, M. G. (ed.) 1998. Il tumulo di Camucia e il carteggio François-Sergardi. Montespertoli: Garlatti & Razzai Ed. Megaw, J. V. S. and Megaw, M. R. 2009. A style without genesis? The elusive birthplace of early Celtic Art. In G. Cooney et al. (eds.) Relics of Old Decency: archaeological studies in later prehistory. Festschrift for Barry Raftery, 291–306. Dublin: Wordwell. Morel, L. 1876. La Champagne souterraine. Reims. Moretti, G. 1936. Il guerriero italico di Capestrano, Opere d’Arte 6. Roma: Ist. Poligrafico dello Stato. Morris, I. 1987. Burial and ancient society. The rise of the Greek city-state, New Studies in Archaeology. Cambridge: Cambridge University Press. Morris, I. 1999. Iron Age Greece and the meanings of ‘princely tombs’. In P. Ruby (ed.) Les princes de la protohistoire et l’émergence de l’état. Actes de la table ronde inter. Naples 1994, 57–80. Naples-Rome: Centre Jean Bérard. Pauli, L. 1978. Der Dürrnberg bei Hallein III, Münchner Beiträge zur Vor- und Frühgeschichte. München: Beck. Popham, M. R. 1994. Precolonisation: early Greek contact with the East. In G. R. Tsetskhladze and F. De Angelis (eds.) The Archaeology of Greek Colonisation: Essays dedicated to Sir John Boardman. Oxford University Committee for Archaeology Monograph 40, 11–34. Oxford: Oxford University Committee for Archaeology. Popham, M. R., Touloupa, E. and Sackett, L. H. 1982. The hero of Lefkandi, Antiquity 56, 169–74. Popham, M. R., Calligas P. G. and Sackett, L. H. (eds.) 1993. Lefkandi II. The Protogeometric Building at Toumba. Part 2. The Excavation, Architecture and Finds. Oxford: British School Archaeology Athens.

135

Prayon, F. 1998. Die Anfänge großformatiger Plastik in Etrurien. In P. Schauer (ed.) Archäologische Untersuchungen zu den Beziehungen zwischen Altitalien und der Zone nordwärts der Alpen während der frühen Eisenzeit Alteuropas, Coll. Regensburg 1994. Regensburger Beiträge zur prähistorischen Archäologie 4, 191–207. Bonn: Habelt. Putz, U. 2007. Früheisenzeitliche Prunkgräber in Ober- und Mittelitalien. Archäologische Forschungen zur Entstehung temporärer Eliten. Regensburger Beiträge zur Prähistorischen Arch. 15. Regensburg: Universitäts-Verlag. Raaflaub, K. 1998. A historian’s headache. How to read ‘Homeric Society’. In N. Fischer and H. van Wees (eds.) Archaeic Greece. New Approaches and New Evidence, 169–93. London: Duckworth/Press of Wales. Rolley, C. (ed.) 2003. La tombe princière de Vix, Société des Amis d Musée du Châtillonais. Paris: Picard. Rowlands, M. 1999. The cultural economy of sacred power. In P. Ruby (ed.) Les princes de la protohistoire et l’émergence de l’état. Actes de la table ronde inter. Naples 1994, 165–172. Naples/Rome: Centre Jean Bérard. Rummel, P. v. 2007. Habitus Barbarus: Kleidung und Präsentation spätantiker Eliten im 4. und 5. Jahrhundert. RGA Ergänzungsband 5. Berlin, New York: de Gruyter. Rupp, D. W. 1988. The ‘Royal Tombs’ at Salamis (Cyprus): ideological messages of power and authority, Journal of Mediterranean Archaeology 1/1, 111–39. Schmidt, M. 2006. The semantics of anax and basileus in Homer. In S. Deger-Jalkotzy et al. (eds.) Ancient Greece, 439–47. Edinburgh: Edinburgh University Press. Schönfelder, M. 2003. Das frühlatènezeitliche Grab eines Reiters und Wagenfahrers aus Châlons-en-Champagne, Dép. Marne, Jahrbuch des Römisch-Germanischen Zentralmuseums Mainz 50, 231–78. Schweizer, B. 2003. Zur Repräsentation sozialen Raums. Die Fürstengräber von Pontecagnano 926 und 928. In U. Veit et al. (eds.) Spuren und Botschaften. Interpretationen materieller Kultur, Tübinger Archäologische Taschenbücher 4, 319–45. Münster u. a.: Waxmann. Schweizer, B. 2008. Archäologie und historischer Vergleich. Fürstengräber und Urbanisation im frühen und mittleren 1. Jt. v. Chr. In D. Krausse (ed.) Frühe Zentralisierungsund Urbanisierungsprozesse. Zur Genese und Entwicklung frühkeltischer Fürstensitze und ihres territorialen Umlandes. Colloque Blaubeuren 2006, 399–414. Stuttgart: Theiss. Schweizer, B. 2012. Fürsten, Chiefs und Big Men. Oder: Dorophagoi – Basileis als Gabenfresser. Zu Eliten in den Altertumswissenschaften und Elitenkritik in der Antike. In T. L. Kienlin and A. Zimmermann (eds.) Beyond Elites. Alternatives to Hierarchichal Systems in Modelling Social Formations, Universitätsforschungen zu prähistorischen Archäologie 215, 461–82. Bonn: Habelt. Steuer, H. 2006. Fürstengräber, Adelsgräber, Elitegräber: Methodisches zur Anthropologie der Prunkgräber. In C. CarnapBornheim, D. Krausse and A. Wesse (eds.) Herrschaft-TodBestattung. Zu den vor- und frühgeschichtlichen Prunkgräbern als archäologisch-historische Quelle, Universitätsforschungen zur Prähistorischen Archäologie 139, 11–25. Bonn: Habelt. Stöllner, Th. 2010. Kontakt, Mobilität und Kulturwandel im Frühlatènekreis – das Beispiel Frühlatènegürtelhaken. In E.

136

Thomas Stöllner

Jerem, M. Schönfelder and G. Wieland (eds.) Nord-Süd, OstWest Kontakte während der Eisenzeit in Europa, Akten der Internationalen Tagungen der AG Eisenzeit in Hamburg und Sopron 2002, 277–319. Budapest: Archaeolingua. Stupperich, R. 1977. Staatsbegräbnis und Privatgrabmal im klassischen Athen. Dissertation Universität Münster. Ulf, C. 2012. Der Streit um Standards: die homerischen Epen als Diskussionsforum. In T. L. Kienlin and A. Zimmermann (eds.) Beyond Elites. Alternatives to Hierarchichal Systems in Modelling Social Formations. Universitätsforschungen zu prähistorischen Archäologie 215, 471–82. Bonn: Habelt. Veit, U. 2005. Kulturelles Gedächtnis und materielle Kultur in schriftlosen Gesellschaften. Anthropologische Grundlagen und Perspektive für die Urgeschichtsforschung. In. T. L. Kienlin (ed.) Die Dinge als Zeichen. Kulturelles Wissen und materielle Kultur – Perspektiven einer kulturwissenschaftlichen Synthese, Internationale Fachtagung, Frankfurt/Main 2004. Universitätsforschungen zur prähistorischen Archäologie, 23–40. Bonn: Habelt.

Verger, S. 1995. De Vix à Weiskirchen. La transformation des rites funéraires aristocratiques en Gaule de Nord et de l’Est au Ve siècle avant J.-C. Mél. École Françaises Rome 107, 335–458. Vernant, J.-P. 1995. Mythos und Religion im alten Griechenland. Frankfurt, New York: Campus. Vernant, J.-P. 1996. Death with Two Faces. In S. Schein (ed.) From Reading the Odyssey: Selective interpretive essays, 55–62. Princeton: Princeton Univ. Press. Whitley, J. 1994. The monuments that stood before Marathon: tomb cult and hero cult in archaic Attica, American Journal of Archaeology 98, 213–30. Zamarchi-Grassi, P. 1998. Un edificio per il culto funerario. Nuovi dati sul tumulo II del Sodo a Cortona, Rivista di Archeologia 22, 19–26. Zürn, H. 1964. Eine hallstattzeitliche Stele von Hirschlanden, Kr. Leonberg (Württbg.). Vorbericht, Germania 42, 27–36.

15 ALFRED AND ALEXANDER John Boardman

Fig. 15.1. The Alfred Jewel. Oxford, Ashmolean Museum.

The Alfred Jewel (Fig. 15.1) is not Celtic, but it is not all that far removed from Celtic style and technique, and the subject of my paper is located roughly half way between the home of the Jewel in Oxford’s Ashmolean Museum and the home of our honorand in Australia. In 1956 the distinguished mediæval scholar, David Talbot Rice, made the suggestion that the figure on the Jewel was that of Alexander the Great, viewing the world from the heavens, a subject treated in a similar manner or more explicitly on many sacred buildings and in many media in Europe from about the time of the Jewel onwards. The suggestion has been studiously ignored by Ashmolean writers (e.g. Hinton 2008), and may be wrong, but I would like to recall it, not least to add something to the study of the origin of details of the representation in its earliest days. The story is that of the Alexander Romance which provided an after-life of adventure for Alexander in the east, rather than a return to Babylon and death. It involved many of the stranger eastern phenomena which occupied the interest of scholars from Ctesias in the 5th century BC (Nichols 2011), through Sir John Mandeville, to Sindbad and Marco Polo. It was a world of, for instance, men with one large foot with which they shielded themselves from the sun (skiapods – recalling a fakir pose) to giant birds that might lift a camel, which had a long pedigree even in Chinese lore. The Romance had reached something like a final form by the 3rd century AD but was much embellished thereafter and has been the subject of several excellent studies by Richard Stoneman (1991; 1994). Our episode was one in which Alexander was desirous of surveying the whole world. He visited the depths of the ocean in a diving bell, and for the heights of heaven he harnessed two large birds to a basket or throne, and persuaded them to fly upwards by holding pieces of meat on sticks above their heads, always just out of reach (like the proverbial donkey and the

138

John Boardman

Fig. 15.2. Detail of mosaic in Otranto Cathedral. After Schmidt.

Fig. 15.3. Sasanian plate. Boston, Museum of Fine Art. Author’s sketch.

carrot). It proved to be a subject which, with a number of variants, notably preferring griffins to ordinary birds, was much loved, from Europe to later Persia (where ‘Iskander’ was well-liked) and even to Mughal India. It has been, for its European aspect, well studied and illustrated by Schmidt (1995), with a plentiful harvest of examples. The abbreviated version, with just the Alexander bust holding the two floral stalks, not regal sceptres, and with or without fruits, is very common; or the figure and the lifting creatures are simply set as a threesome without any details of attachment. The point of the subject on the Jewel would have been that Alexander was embodying the sharpness of sight which the Jewel abetted, by being the holder for a pointer (aestel) used while reading documents. The slightly later Fuller Brooch (Hinton 2008, 21, 63) made a similar point, with a bust holding two floral stalks and signifying Sight, among personifications of four other senses, and is likely to derive from the same Alexander iconography. As for the Alfred Jewel, doubt has sometimes been cast on its function as the handle for an aestel pointer to aid reading, and on the identity of the ‘Alfred’ (Hinton 2008, 18–20, 25–9; Bakka 1966), but both seem to most scholars reasonably secure. To illustrate the fuller version of the air lift, as it appeared in Europe, I figure the group on a 12th-century Otranto mosaic (Fig. 15.2) (Schmidt 1995, fig. 33), where the figure is labelled, the motive power is supplied by griffins and Alexander is on a stool. This was, broadly, the standard iconography and had been, rather surprisingly, accepted into European ecclesiastical art which might also show griffins in their right function, or just birds and with Alexander

variously accommodated by throne or seat or even cage. The origins of the standard iconography for the scene have not been much discussed, and I wish here to review various existing iconographic motifs of the early centuries BC in the east which might have contributed to its creation and the way it was then seen in Europe. Roman art, especially that of the eastern empire, must be an obvious source. There, the motif of any figure, usually divine (an emperor or Bacchus), carried in a frontal fourhorse chariot, was commonplace. Very often the figure is shown centrally, full length, while the chariot itself is summarily represented below and the horses or other beasts climb up at either side of the group. This is especially true when the field is circular, as was often the case. When the charioteer is Bacchus, lions form the team. It was a composition which was carried far east also and eventually served the Indian sun god, Surya. Talbot Rice also, in passing, remarked on the possible relevance of Sasanian representations of their Moon God being lifted to heaven in a shrine or on a throne, pulled by a team of bulls, sometimes winged, which are shown climbing up the sides of the group, and he remarks the ‘Sasanian costume’ of one of the European examples (Talbot Rice 1956, 215–7). The scenes with the Moon God are on gilt silver plates, as Fig. 15.3 (Frye 1962, pl. 85). The Sasanian empire was as extensive and powerful as that of Rome and often had the better of its western neighbour. There is, however, another Sasanian motif, not shared with Rome, which may be relevant. The king is often shown sitting frontally on a broad stool. The ends of the stool

15.  Alfred and Alexander

139

Fig. 15.4. Sasanian plate. British Museum. Photo, author.

are supported by animal figures, a rôle on furniture in the east which goes back to Achaemenid times, at least. The animals all have raised wings – birds, horses, or griffins in the classical form which had been generally adopted in the arts of the griffin-infested lands of Central Asia. Prudence Harper comments on the Sasanian kings’ ascensions but is not sure whether the winged figures on stools also imply flight, illustrating one with birds with raised wings (Harper 1979, 56, pl. 2). But on an example on a plate in the British Museum (Fig. 15.4; Dalton 1964, pl. 38) the griffins have lifted their forelegs as if to rise. There is also a scene of the king seated at the edge of a square dais, at the front corners of which lions are leaping upwards (Ghirshman 1953, fig. 11; in St Petersburg; my Fig. 15.5). One begins to think of stories of flying carpets, and we may reasonably suspect that the Sasanian king’s regal seat could be air-borne, in which case we have a closer parallel to the western treatment of the Alexander motif than the Roman world has provided. And as for Alexander himself as semi-divine carrier of bounty and food for the wild, there are parallels enough for divine figures earlier in the east to justify the eastern pedigree of a comparable Alexander. Thus, on a 15th-century BC relief from Assur, the mountain deity stands frontally holding branches with fruit at his shoulders, with rams grazing upon them (Orthmann 1975, pl. 194), just as Alexander does for his griffins centuries later.

Fig. 15.5. Sasanian plate. Hermitage Museum. Author’s sketch.

Bibliography Bakka, E. 1966. The Alfred Jewel and Sight, Antiquaries Journal 46, 277–82. Dalton, O. M. 1963. The Treasure of the Oxus. London: British Museum. Ghirshman, R. 1953. Notes iraniennes V, Artibus Asiae 16, 51–76. Frye, R. N. 1962. The Heritage of Persia. London: Weidenfeld and Nicolson. Harper, P. O. 1979. Thrones and enthronement scenes in Sasanian art, Iran 17, 49–64. Hinton, D. A. 2008. The Alfred Jewel. Oxford: Ashmolean Museum Handbook. Nichols, A. 2011. Ctesias on India. Bristol: Classical Press. Orthmann, W. 1975. Der alte Orient. Berlin: Propyläen Kunstgeschichte. Schmidt, V. M. 1995. A Legend and its Image. Groningen: Egbert Forsten. Stoneman, R. 1991. The Greek Alexander Romance. London: Penguin Classics. Stoneman, R. 1994. Legends of Alexander the Great. London: Everyman. Talbot Rice, D. T. 1956. New Light on the Alfred Jewel, Antiquaries Journal 36, 214–217.

16 LA FIBULE DE MOSCANO DI FABRIANO: UN JALON IMPORTANT DE L’ÉVOLUTION DE L’ART CELTIQUE AU IVE SIÈCLE AVANT J.-C. Luana et Venceslas Kruta

La tombe de guerrier de Moscano di Fabriano Une très riche tombe isolée de guerrier sénon fut découverte fortuitement, en 1955, à une quinzaine de kilomètres de Moscano di Fabriano, dans une position stratégique de la haute vallée de l’Esino, sur une voie qui relie la côte adriatique des Marches à la vallée du Tibre. Bouleversée antérieurement à la découverte par des travaux agricoles, la grande fosse rectangulaire contenait les résidus d’un individu armé accompagné d’un cheval harnaché et d’un très riche service de banquet, comportant aussi bien de nombreuses céramiques attiques que des récipients en bronze (Landolfi 1987; 1991; 1998). Elle n’a malheureusement pas été jusqu’ici l’objet d’une publication exhaustive et ses matériaux restent largement inédits. Dans la grande fosse rectangulaire furent recueillis, outre quelques ossements humains et équins, des pièces du harnachement parmi lesquels figurent deux phalères et un chanfrein dont la partie principale est constitué d’un fil de bronze ondulé, analogue à celui des bracelets laténiens attestés en nombre dans différentes régions d’Europe. Il s’agit de l’une des formes de parures annulaires les plus caractéristiques de la phase Duchcov-Münsingen des dernières décennies du IVe siècle et du début du siècle suivant (Kruta 1979). On a relevé également la présence d’un fragment de strigile en bronze et du récipient-balsamaire «a gabbia» qui lui est associé dans les soins corporels de la palestre. Ces instruments, liés aux exercices physiques du gymnase, sont attestés également dans d’autres tombes des aires sénone et boïenne et témoignent de l’adhésion des Celtes installés nouvellement en Cispadane à cette pratique hellénistique (Kruta 1981; 2008, Jolivet 2008). Les ustensiles du banquet comportaient plus d’une vingtaine de céramiques attiques de formes diverses, à figures rouges ou à vernis noir, les vases en bronze du service

à boisson – un stamnos qui pourrait avoir été produit dans un atelier celto-italique, un trépied, une situle prénestine (Landolfi 1998, fig. 43), une œnochoé, un bassin, une kylix, une passoire, ainsi que des fragments appartenant probablement à un cratère – un petit tonneau en bois garni de tôle de bronze ornée au repoussé, une râpe et un assortiment de coutelas. Deux pions en pâte de verre attestent la présence d’un jeu de damier, connu également d’autres riches tombes masculines de la Cispadane (Kruta 1980; 1981). Les nombreuses céramiques d’importation fournissent d‘après Maurizio Landolfi, excellent connaisseur de cette catégorie d’objets, une datation très sûre vers 360 avant J.-C. pour la mise en terre de l’ensemble (Landolfi 1991; 1998). Il s’agit donc bien d’un mobilier funéraire caractéristique de la phase initiale des nécropoles sénones, avec la riche tombe féminine de Santa Paolina de Filottrano et les plus anciennes des sépultures de Montefortino (Kruta 1981). La seule parure personnelle recueillie est la fibule laténienne en bronze qui sera examinée plus en détail dans cet article. L’armement est constitué par un casque en bronze de type dit celto-italique au cimier de fer (Landolfi 1991; 1998, fig. 48), analogue à des exemplaires trouvés dans d’autres nécropoles sénones. L‘épée laténienne, plus particulièrement son fourreau en fer et bronze, est indiscutablement l’objet le plus connu de la tombe de Moscano. Elle fut considéree pendant longtemps comme une arme courte gaînée sur les deux faces de tôle de bronze (Frey 1971; Megaw 1978; 1982; Landolfi 1987). La nature bimétallique du fourreau – plaque de revers en fer et droit revêtu de tôle de bronze – et son appartenance à une épée de longueur normale furent confirmées après restauration (Rapin 2008). La plaque en tôle de bronze décorée par estampage au repoussé constitue désormais une référence pour l’ornementation de

16.  La fibule de Moscano di Fabriano: un jalon important de l’évolution de l’art celtique cette catégorie d’objet (Kruta 1988a; 1992; 2000; Ginoux 1994). Elle est d’autant plus intéressante que le décor paraît avoir été exécuté avec le même poinçon qui fut utilisé pour les bandes de tôle de bronze, probablement réemployées, appliquées transversalement au droit du fourreau découvert dans une tombe de la nécropole d’Épiais-Rhus, à une trentaine de kilomètres au nord-ouest de Paris (Kruta et al. 1984). Il est donc évident que le fourreau de Moscano di Fabriano occupe une place privilégiée dans la séquence qui associe des exemplaires décorés du IVe siècle avant J.-C., cisalpins et transalpins, notamment ceux qui présentent une plaque de revers en fer et un revêtement du droit en tôle de bronze ouvragée. Cependant, l’examen plus détaillé de cet aspect particulier n’entre pas dans le cadre du présent article.

La Fibule Son centre d’intérêt est la fibule en bronze, un des rares objets de la tombe dont l’image a été publiée (Landolfi 1991). Nous avons eu toutefois l’occasion de l’examiner et d’en réaliser un relevé complet en décembre1994, grâce au surintendant archéologique des Marches de l’époque Luigi Malnati (Fig. 16.1). L’illustration publiée par Maurizio Landolfi avait en effet étayé l’hypothèse qu’il pourrait s’agir «d’une variante celto-italique des fibules campaniennes» (Kruta 1982, 42, n. 36) et nous avait donc incité à effectuer un examen autoptique de l’objet, jalon potentiel de la diffusion d’un certain type de décor végétal des fibules laténiennes, étudié précédemment sans connaître cet exemplaire autrement que par de brèves mentions (Kruta 1976/1977). Malgré son état fragmentaire, la fibule de Moscano, conservée dans les collections du Musée national des Marches sous le n° 20816, a effectivement livré des informations d’un grand intérêt. Elle est malheureusement incomplète: seuls nous sont parvenus un peu plus de la moitié postérieure de l’arc, avec le ressort bilatéral de deux fois deux spires – brisé et privé de sa corde externe – , ainsi que l’ardillon, amputé de sa pointe mais avec l’autre moitié du ressort. Son intérieur recèle un élément en fer, destiné probablement à palier une ancienne brisure, située à la fin de la de la première spire. L’objet avait donc été en usage bien avant son dépôt dans la sépulture. Les mesures du fragment conservé – longueur 4,8 cm, hauteur 2,3 cm – indiquent une longueur d’au moins une dizaine de centimètres pour l’objet complet. Il s’agit donc d’une fibule de grande taille, fréquente surtout dans les tombes d’hommes armés. La forme générale en arc de cercle est celle du type dit pré-Duchcov (Kruta 1979), illustré par exemple par la fibule des environs de Morat (Fig. 16.2b) (Kruta 1976/1977, fig. 4/1) ou les deux fibules de la tombe no. 48 de Münsingen-Rain (Hodson 1968, pl. 21/787, 788). L’une d’elles présente d’ailleurs un décor quadripartite dont l’un des champs est orné d’un motif analogue à celui du fourreau de Filottrano. Une des caractéristiques de ces

141

fibules est le ressort à deux fois deux spires et corde externe, remplacé successivement par un ressort à deux fois trois spires, sur les formes classiques du type dit de Münsingen, tel l’exemplaire remarquable de la tombe 49 (Hodson 1968, pl. 22/797). Ces deux fibules de la nécropole suisse permettent d’envisager la restitution hypothétique du pied de la fibule de Moscano, très probablement discoïdal, avec un cabochon de corail analogue à celui qui devait se trouver au milieu de l’arc. L’utilisation de cette matière est très vraisemblable non seulement par comparaison avec les fibules transalpines, mais également compte tenu de la proximité des pêcheries campaniennes et des contacts des Sénons des Marches avec cette région. Le corail, connu et utilisé par les Celtes depuis le VIe siècle avant J.-C., n’avait pas pour eux une signification particulière seulement à cause de sa couleur rouge, dont le lien avec la vie n’a rien d’anecdotique, car il remonte aux racines de l’idéologie indoeuropéenne. Cette couleur – avec le blanc et le noir l’une des trois couleurs cosmiques – est liée au ciel crépusculaire auroral et vespéral du parcours solaire ainsi qu’au feu (Jouët 2012, 269, 870). Dans le cas particulier du corail, la couleur rouge se trouve associée à son origine marine et à sa forme qui rappelle celle d’une flamme. Il n’est donc pas surprenant qu’il devint pour les Celtes la matérialisation idéale de l’idée du «feu dans l’eau», c’est-à-dire de la force vitale qui transforme l’eau morte de l’Autre Monde en la source de vie et d’énergie qui tombe du ciel ou jaillit des profondeurs, quelquefois même avec la chaleur résultant de son contact avec le feu. La présence de corail sur cette catégorie de fibules n’est donc pas uniquement décorative mais elle leur confère une charge magique très significative. Le médaillon central de l’arc de la fibule de Fabriano trouve son analogie la plus proche, chronologiquement et géographiquement, dans une fibule de la tombe n° 49 du cimetière de Münsingen-Rain, déjà évoquée précédemment (Fig. 16.2a) (Hodson 1968, pl. 22/798). Ce médaillon central est flanqué d’une palmette à cinq feuilles qui devait être répétée symétriquement de l’autre côté. La feuille médiane, nettement plus grande, arrive jusqu’au ressort, tandis que les feuilles latérales enveloppent l’arc et se rejoignent par leur extrémités sur son dessous. On trouve un équivalent à cette très rare disposition du décor sur la fibule des environs de Morat mentionnée ci-dessus, où le champ ondulé qui contient un double rinceau divergent s’enroule autour de l’arc et l’enveloppe complètement (Fig. 16.2b). Le fait que les feuilles latérales présentent la forme de la «feuille de gui» (Kruta 1986; 1988b) n’est certainement pas fortuit. On peut d’ailleurs constater dans la composition développée, aussi bien à partir du dessus que du dessous (Fig. 16.3), que ce motif constitue la composante principale d’évocations de masques proches de ceux que l’on peut observer sur d’autres œuvres du IVe siècle appartenant à la mouvance celto-italique, tels que le fourreau de Filottrano

142

Luana et Venceslas Kruta

(Kruta 2007, 28, 107) ou le casque d’Agris (Gomez de Soto 1986; Kruta 2000, 395). Le lien entre l’image de la palmette et un visage allusif, qui n’est pas nouveau, connaît alors une vogue qui connaîtra son apogée au siècle suivant (Kruta 1987; 1994).

Le dédoublement, de part et d’autre du médaillon central, illustre ainsi un thème attesté dès le Ve siècle av. J.-C. par la bague de Rodenbach (Jacobsthal 1944, no. 72) ou celle des environs de Hořovice en Bohême (Kruta 2007, 24, 103), celui des «têtes jumelées» (Kruta 2007, 62–68). Il s’agit de l’une des formes imagées données au concept de principes opposés, indissociables et alternés dans un cycle répétitif. Ils sont identifiés dans ce cas aux personnages qui en constituent l’incarnation divine. On peut suivre ce thème sur de nombreux objets du IIIe siècle, par exemple le bracelet de la tombe n° 31 de Brno-Maloměřice, où les deux visages allusifs sont le résultat de la transformation de doubles palmettes (Fig. 16.4c) (Kruta 1987, fig. 3–4; 2007, 34, 111). L’étude comparative des fibules ornées de palmettes, disposées généralement de part et d’autre d’un motif central, indique clairement qu’il s’agit de variantes issues de prototypes italiotes différents mais structurés de manière analogue. Le cabochon de corail peut y être remplacé par une esse – signe symbolique du parcours du soleil entre deux solstices d’hiver –, comme sur la fibule de la sépulture de guerrier no. 9 de Bučovice en Moravie (Fig. 16.4a) (Kruta 1976/1977, fig. 12), ou un «zigzag de feuilles au compas», attesté à cet emplacement sur l’exemplaire de la tombe no. 97 du Dürrnberg (Fig. 16.4b) (Moosleitner et al. 1974, Tf. 164/1; Kruta 1976/1977, fig. 11/3). Dans un cas comme dans l’autre, il s’agit d’images liées au même concept d’alternance cyclique qui apparaît sur les fibules sous des formes variées et quelquefois très simplifiées (Kruta 2010). Quant à la question de l’inspiration formelle de la fibule, il ne paraît pas nécessaire de modifier l’hypothèse déjà ancienne d’un modèle fourni par les fibules campaniennes, proposée il y a de cela plus d’une trentaine d’années (Kruta

Fig. 16.1. Moscano di Fabriano (Marches, Italie): fibule fragmentaire en bronze de la tombe découverte en 1955; longueur 4,8 cm; vers 370 av. J.-C.; Museo archeologico nazionale delle Marche, Ancona, inv. no. 20816 (dessin Hana Velímská d’après le relevé de V. K.).

a

b

Fig. 16.2. a. Münsingen-Rain, tombe no. 49: fibule en bronze à cabochons de corail; longueur c.7 cm; vers 350 av. J.-C.; Bern, Historisches Museum (d’après Hodson 1968). b. Environs de Morat (canton de Fribourg, Suisse): fibule fragmentaire en bronze; longueur 7 cm; vers 370/360 av. J.-C.; Genève, Musée d’art et d’histoire (d’après Kruta 1976/1977).

16.  La fibule de Moscano di Fabriano: un jalon important de l’évolution de l’art celtique Fig. 16.3. Moscano di Fabriano: développement du décor de l’arc de la fibule, à partir du dessus et du dessous (dessin Hana Velímská d’après le relevé de V. K.).

1978). Au contraire, le cas de la fibule de Moscano di Fabriano la renforce, en fournissant un jalon significatif sur le sol italien et en étoffant nos connaissances des liens privilégiés qui unissaient alors le monde celto-italique au plateau suisse, carrefour principal des trafics transalpins. D’une manière générale, le cas de la fibule de Fabriano souligne une fois de plus le fait que l’adoption et la transformation de modèles extérieurs au monde celtique n’est pas le résultat de leur adoption fortuite, mais celui d’une sélection de sujets qui pouvaient être facilement adaptés à la représentation de thèmes bien définis. Leur permanence dans le temps et dans l’espace ne s’explique pas seulement par la transmission de l’image, mais surtout par la présence de l’idée qui lui donne un sens. Ce dernier prévaut donc sur la forme, la substance de l’image prime sur son apparence. Ainsi, le contenu exprime l’appartenance à un système idéologique aux lignes générales communes et durables, la forme délimite des regroupements chronologiques ou géographiques, des faciès dont la cohérence culturelle ne se reflète pas seulement dans les œuvres d’art, mais également dans d’autres catégories de vestiges.

The Moscano di Fabriano brooch: an important 4th century BC milestone in the evolution of Celtic art The Moscano di Fabriano warrior grave A very rich isolated Senonian warrior burial was discovered by accident, in 1955, about 15 km from Moscano di Fabriano, occupying a strategic position at the top of the Esino Valley, on a route which links the Adriatic Marches

143

coast to the Tiber valley. Already damaged by agricultural activity before its discovery, the large rectangular pit contained the remains of an armed individual accompanied by horse-trappings and a very rich collection of banqueting equipment, including numerous Attic pottery vessels and bronze vessels (Landolfi 1987; 1991; 1998). Unfortunately, the discovery has not yet been the subject of an exhaustive study and the artefacts remain largely unpublished. Besides several fragments of human and horse bone, pieces of harness were recovered from the large, rectangular grave pit. Amongst these there were two phalerae and a chamfrein, whose main part consists of corrugated bronze wire, analogous to those of La Tène bracelets attested in a number of different regions of Europe. It would appear to be one of the most characteristic forms of annular jewellery of the Duchcov-Münsingen phase of the last decades of the 4th century and the beginning of the following century (Kruta 1979). In addition, a fragment of bronze strigil and an ‘a gabbia’ balm vessel were discovered, both associated with body care in the palestra. These tools, linked to physical exercise in the gymnasium, are also known from other tombs in Senones and Boii territories, and are evidence for the way in which Celts, newly-located in Cispadania, maintained this Greek practice (Kruta 1981; 2008, Jolivet 2008). The banqueting equipment consisted of more than two dozen Attic pottery vessels in a variety of forms, red- and black-figured, and bronze vessels from a drinking service – a stamnos which could have been produced in a CeltoItalic workshop, a tripod, a situla from Praeneste-Palestrina workshop (Landolfi 1998, fig. 43), an oinochoe, a basin, a kylix, as well as fragments probably belonging to a crater – a small wooden barrel ornamented with sheets of bronze decorated in repoussé, a grater and an assortment of knives. Two glass paste gaming pieces attest to the presence of a board game, also known from other rich male Cispadanian tombs (Kruta 1980; 1981). According to Maurizio Landolfi, great expert in this category of objects, the many imported ceramics give a very secure date of c. 360 BC for the interment of the assemblage (Landolfi 1991; 1998). It appears to be, then, a furnished grave characteristic of the first phase of Senonian cemeteries, along with the rich female grave from Santa Paolina di Filottrano and the earliest burials from Montefortino (Kruta 1981). The only item of personal jewellery found was the bronze La Tène style brooch which will be examined in more detail in this paper. The weaponry consisted of a bronze helmet (of the so-called Celto-Italic type) with an iron crest (Landolfi 1991; 1998, fig. 48), analoguous to examples found in other Senonian cemeteries. The La Tène sword, and more particularly its iron and bronze scabbard, is indisputably the best-known artefact from the Moscano tomb. For a long time it was thought to be a short weapon with both faces covered

144 a

Luana et Venceslas Kruta c

b

Fig. 16.4. a. Bučovice (Moravie, Rép. Tchèque), sépulture de guerrier no. 9: fibule en bronze avec concrétions de fer oxydé sur le pied discoïdal et à l’intérieur du ressort; longueur 6,4 cm; vers 340/320 av. J.-C.; Brno, Moravské zemské museum (d’après Kruta 1976/1977). b. Dürrnberg près de Hallein (Autriche), sépulture de guerrier no. 97: fibule en bronze à pied discoïdal et arc décoré; longueur 8,2 cm; vers 350 av. J.-C.; Hallein, Keltenmuseum (d’après Moosleitner et al. 1972). c. Brno- Maloměřice (Moravie, Rép. Tchèque), sépulture no. 31: bracelet en bronze à deux oves creux, orné de deux masques issus de doubles palmettes, séparés par des rinceaux divergents simplifiés dont la partie médiane dessine une esse; diamètre 8 × 8,5 cm; vers 280–240 av. J.-C.; Brno, Moravské zemské museum (d’après Kruta 1987).

16.  La fibule de Moscano di Fabriano: un jalon important de l’évolution de l’art celtique in bronze sheet (Frey 1971; Megaw 1978; 1982; Landolfi 1987). The dual-metal nature of the scabbard – reverse plate of iron and front dressed in bronze sheet – and its recognition as a normal-length sword was confirmed after restoration (Rapin 2008). The bronze sheet plate decorated by repoussé stamps is now a reference for ornamentation of this kind of object (Kruta 1988a; 1992; 2000; Ginoux 1994). It is even more interesting that the decoration seems to have been executed with the same stamp used for the bands of bronze sheet, probably re-used, applied transversely to the front of the scabbard discovered in a tomb from the Épiais-Rhus cemetery, some 30 km to the north-west of Paris (Kruta et al. 1984). It is therefore evident that the Moscano di Fabriano scabbard occupies an important place in the sequence which links decorated examples from the 4th century BC, Cisalpine and Transalpine, notably those which have a reverse plate in iron and a front plate decorated with a finely-worked bronze sheet. However, further consideration of this particular aspect is not within the scope of this paper.

The brooch The focus of interest is the bronze brooch, one of the few objects from the tomb whose picture has been published (Landolfi 1991). We have also had occasion to examine it and to produce a complete account in December 1994, thanks to Luigi Malnati, archaeological superintendent of Marches at the time (Fig. 16.1). The illustration published by Maurizio Landolfi essentially supported the hypothesis that it could be ‘a Celto-Italic variant of the Campanian brooches’ (Kruta 1982, 42, note 36), and we have therefore determined to make an autoptic examination of the object, marking the potential for the diffusion of a certain type of vegetal-style decoration on La Tène brooches, previously studied without knowing of this example other than from brief mentions of it (Kruta 1976/1977). In spite of its fragmentary state, the Moscano brooch, kept in the collections of the National Museum of the Marches as inventory number 20816, has actually provided us with very interesting information. Unfortunately, it is incomplete: only a little over half of the bottom part of the bow survives, with the bilateral spring of two doubled coils – broken and without the external wire, like the pin, which has lost its point, but still has the other part of the spring. There is some iron in the interior, probably intended to mend an old fracture, situated at the end of the first spiral. So the object had already seen considerable wear before it was placed in the grave. The measurements of the surviving fragment – 4.8 cm in length, 2.3 cm in height – indicate a total length of at least 10 cms for the complete object. It would therefore appear to be a large brooch, most common above all in the graves of weaponed males. The general form – of the arc of a circle – is that of the so-called pre-Duchcov type (Kruta 1979), illustrated for example

145

by the brooch from the area of Morat (Fig. 16.2b) (Kruta 1976/1977, fig. 4/1), or the two brooches from grave 48 at Münsingen-Rain (Hodson 1968, pl. 21/787, 788). One of these, however, had a quadripartite ornamentation on which one of the fields was decorated with a motif analoguous to that of the Filottrano scabbard. One of the characteristics of these brooches is the spring with two double coils and an external wire, replaced successively by a spring with two triple coils, on the classic so-called ‘Münsingen’ type, like the remarkable example from grave 49 (Hodson 1968, pl. 22/797). These two brooches from the Swiss cemetery allow us to postulate the hypothetical reconstruction of the foot of the Moscano brooch, very probably discoidal, with a cabochon of coral analogous with that which would have been at the centre of the bow. The use of this material is very probable, not only by comparison with transalpine brooches, but also taking into account the nearby fisheries in Campania and the contacts of the Senonians of the Marches with this region. Coral, known and used by the Celts from the 6th century BC, did not only have a special meaning for them because of its red colour, whose link with life has nothing anecdotal about it, because it develops from the roots of Indo-European ideology. This colour – with white and black one of the three cosmic colours – is linked to the course of the sun in the twilight sky at dawn and dusk as well as to fire (Jouët 2012, 269, 870). In the particular case of coral, the red colour is associated with its marine origin and its shape, which recalls a flame. It is therefore not surprising that it became, for the Celts, the ideal materialization of the idea of ‘fire in water’, that is to say the vital force which transforms the dead water of the Other World into the source of life and energy which falls from the sky or springs out of its depths, sometimes even with the heat resulting from its contact with fire. The presence of coral in this category of brooches is therefore not simply decorative, but also it confers a very significant magical charge. The central medallion on the bow of the Fabriano brooch finds closer analogies, chronologically and geographically, in a brooch from grave 49 from the Münsingen-Rain grave, already discussed (Fig. 16.2a) (Hodson 1968, pl. 22/798). This central medallion is flanked by a five-leaved palmette which is repeated symmetrically on the other side. The central leaf, distinctly larger, reaches to the spring, while the flanking leaves envelop the bow and their edges meet beneath it. An equivalent to the very unusual layout of the decoration is to be found on the brooch from the area of Morat mentioned above, where the undulating field which contains a diverging double tendril twists around the bow and completely envelops it (Fig. 16.2b). The fact that the lateral leaves take the form of a ‘mistletoe leaf’ (Kruta 1986; 1988b) is clearly not accidental. It is also possible to say in the developed composition, both on the top and underneath (Fig. 16.3), that this motif comprises the main

146

Luana et Venceslas Kruta

component evoking masks similar to those visible on other 4th century Celto-Italic art, like the Filottrano scabbard (Kruta 2007, 28, 107) or the helmet from Agris (Gomez de Soto 1986; Kruta 2000, 395). The link between the image of the palmette and an elusive face, which is not new, was part of a fashion which reached its apogee in the following century (Kruta 1987; 1994). Separating the motif on each side of the central medallion also illustrates another theme attested from the 5th century BC by the Rodenbach ring (Jacobsthal 1944, no. 72), or that from the Hořovice area in Bohemia (Kruta 2007, 24, 103), the one with the ‘twin heads’ (Kruta 2007, 62–68). It appears to be one of the pictorial forms given to the concept of opposite principles, inseparable and opposing in a repetitive cycle. In this case they are identified as the characters constituting the divine incarnation. This theme can be followed through a number of 3rd century objects, for example the bracelet from grave 31 from Brno-Maloměřice, or the two allusive faces resulting from the transformation of double palmettes (Fig. 16.4c) (Kruta 1987, fig. 3–4; 2007, 34, 111). Comparative study of brooches decorated with palmettes, generally arranged on either side of the central motif, clearly indicates that these are variant issues of different but similarly structured Italic prototypes. The coral cabochon can be replaced here by an ‘S’ – symbolic sign of the path of the sun between two winter solstices – as on the brooch from warrior grave 9 from Bučovice in Moravia (Fig. 16.4a) (Kruta 1976/1977, fig. 12), or a ‘zigzag of compass leaves’ attested in this position on the example from grave 97 at Dürrnberg (Fig. 16.4b) (Moosleitner et al. 1974, Tf. 164/1; Kruta 1976/1977, fig. 11/3). In both cases, the images are linked to the same concept of cyclical alternation which appears on brooches in a variety of forms, and sometimes is highly simplified (Kruta 2010). As for the question of the formal inspiration for the brooch, it does not seem necessary to modify the longstanding hypothesis, put forward over 30 years ago, of a model provided by Campanian brooches (Kruta 1978). On the contrary, the case of the Moscano di Fabriano brooch reinforces it, establishing a significant marker on Italian soil and so expanding our knowledge of the close ties which united the Celto-Italic world to the Swiss plateau, the main crossroads of transalpine traffic. In general, the case of the Fabriano brooch underlines once more the fact that the adoption and transformation of external models within the Celtic world is not the result of accidental adoption, but of the selection of subjects which could be easily adapted to represent very specific themes. Their permanence in time and space cannot only be explained by the transmission of the image, but above all by the presence of an idea which gave it a meaning. Therefore the latter prevails over its form, the substance of the image takes priority over its appearance. In this way, the content expresses affinity with an

ideological system with common and durable general lines, the form defines chronological or geographic groupings, the features whose cultural coherence is not only reflected in works of art, but also in other categories of surviving evidence. Translated by S. Crawford.

Bibliography Duval, P.-M. and Kruta, V. (éds.) 1979. Les mouvements celtiques du Ve au Ier siècle avant notre ère. Paris: Éd. du C.N.R.S. Duval, P.-M. and Kruta, V. 1982. L’art celtique de la période d’expansion: IVe et IIIe siècles avant notre ère. Genève: Droz. Frey, O.-H. 1971. Das keltische Schwert von Moscano di Fabriano, Hamburger Beiträge zur Archäologie I-1, 173–179. Ginoux, N. 1994. Les fourreaux ornés de France du Ve au IIe s. avant J.-C., Études celtiques 30, 7–86. Gomez de Soto, J. 1986. Le casque du IVe s. avant notre ère de la grotte des Perrats à Agris (Charente), Archäologisches Korrespondenzblatt 16, 79–183. Hodson, F. R. 1968. The La Tène Cemetery of Münsingen-Rain. Catalogue and Relative Chronology, Acta Bernensia V. Bern: Stämpfli. Jacobsthal, P. 1944. Early Celtic Art. Oxford: The Clarendon Press. Jolivet, V. 2008. Monte Bibele et l’Italie centrale. Quelques questions posées par l’étude des strigiles. In D. Vitali and S. Verger (éds.) Tra mondo celtico e mondo italico: La necropoli di Monte Bibele, 77–94. Bologna: Univ. di Bologna, Dipartimento di Archeologia. Jouët, P. 2012. Dictionnaire de la mythologie et de la religion celtiques. Fouesnant: Yoran Embanner. Kruta, V. 1976/1977. Les fibules laténiennes à décor d’inspiration végétale au IVe siècle avant notre ère, Études celtiques 15, 19–47. Kruta, V. 1978. Celtes de Cispadane et Transalpins aux IVe et IIIe siècles avant notre ère: données archéologiques, Studi etruschi, 46, 149–74, pl. XL–XLI. Kruta, V. 1979. Duchcov-Münsingen: nature et diffusion d’une phase laténienne. In P.-M. Duval V. and Kruta (éds.) Les mouvements celtiques du Ve au Ier siècle avant notre ère, 81–115. Paris: Éd. du C.N.R.S. Kruta, V. 1980. Les Boïens de Cispadane. Essai de paléoethno­ graphie celtique, Études celtiques 17, 7–32. Kruta, V. 1981. Les Sénons de l’Adriatique d’après l’archéologie (prolégomènes), Études celtiques 18, 7–38. Kruta, V. 1982. Aspects unitaires et faciès dans l’art celtique du IVe siècle avant notre ère: l’hypothèse d’un foyer celto-italique. In P.-M. Duval V. and V. Kruta. L’art celtique de la période d’expansion: IVe et IIIe siècles avant notre ère, 35–49. Genève: Droz. Kruta, V. 1986. Le corail, le vin et l’Arbre de vie: observations sur l’art et la religion des Celtes du Ve au Ier siècle avant J.-C., Études celtiques 23, 7–32, fig. 1–18. Kruta, V. 1987. Le masque et la palmette au IIIe siècle avant J.C.: Loisy-sur-Marne et Brno-Maloměřice, Études celtiques 24, 13–32.

16.  La fibule de Moscano di Fabriano: un jalon important de l’évolution de l’art celtique Kruta, V. 1988a. I Celti. In G. Pugliese Caratelli (éd.) Italia omnium terrarum alumna. La civiltà dei Veneti, Reti, Liguri, Celti, Piceni, Umbri, Latini, Campani e Iapigi, 261–311. Milano: Scheiwiller. Kruta, V. 1988b. L’art celtique laténien du Ve siècle avant J.-C.: le signe et l’image. In Les princes celtes et la Méditerranée, 81–92. Paris: Documentation Française. Kruta, V. 1992. Materiali senonici del Piceno e arte celtica. In M. Dardari (éd.) La civiltà picena nelle Marche. Studi in onore di Giovanni Annibaldi, 388–401. Ripatransone: Maroni. Kruta, V. 1994. Brennos et l’image des dieux: la représentation de la figure humaine chez les Celtes. Académie des Inscriptions & Belles-Lettres, Comptes rendus des séances de l’année 1992 novembre–décembre (paru en 1994), 821–43. Kruta, V. 2000. Les Celtes. Histoire et dictionnaire. Des origines à la romanisation et au christianisme. Paris: Laffont. Kruta, V. 2007. La cruche celte de Brno. Chef-d’œuvre de l’art, miroir de l’Univers. Dijon: Faton. Kruta, V. 2008. Les Sénons dans les Marches aux IVe et IIIe siècles avant J.-C. État de la question, Études celtiques 36, 7–20. Kruta, V. 2010. La question de l’art géométrique des Celtes, Ktéma 35, 243–52. Kruta, V. et al. 1984. Les fourreaux d’Epiais-Rhus (Val-d’Oise) et de Saint-Germainmont (Ardennes), Gallia 42, 1–20. Landolfi, M. 1987. Presenze galliche nel Piceno a sud del fiume Esino. In D. Vitali (éd.) Celti ed Etruschi nell’Italia centro-

147

settentrionale dal V secolo a.C. alla romanizzazione, 443–68. Bologna: University Press. Landolfi, M. 1991. La tomba di Moscano di Fabriano. In I Celti (version fr. Les Celtes, angl. The Celts), catalogue de l’exposition du Palazzo Grassi à Venise, 287. Milano: Bompiani. Landolfi, M. 1998. Fabriano-loc. Moscano, contrada Serroni. In Museo archeologico nazionale delle Marche 1998, 159–62. Megaw, J. V. S. 1978. The decoration on the sword-scabbard from Jenišův Újezd, inhumation grave 115. In J. Waldhauser et al. (éds.) Das keltische Gräberfeld von Jenišův Újezd (Langugest) in Nordböhmen II, 106–113. Teplice: Krajské Muzeum. Megaw, J. V. S. 1982. Finding proposeful patterns: further notes towards a methodology of pre-roman celtic art. In P.-M. Duval V. and V. Kruta. L’art celtique de la période d’expansion: IVe et IIIe siècles avant notre ère, 213–29. Genève: Droz. Moosleitner, F., Pauli, L. and Peninger, E. 1974. Der Dürrnberg bei Hallein II. München: Beck. Museo archeologico nazionale delle Marche. Sezione protostorica: I Piceni, 1998. Ancona: Istituto poligrafico e Zecca dello Stato, Libreria dello Stato. Rapin, A. 2008. Les Celtes et leurs voisins septentrionaux: nouveaux outils d’analyses pour l’armement laténien du sud de l’Europe aux Ve et IVe s. av. J.-C. In D. Vitali and S. Verger (éds.) Tra mondo celtico e mondo italico: La necropoli di Monte Bibele, 237–68. Bologna: Univ. di Bologna, Dipartimento di Archeologia.

17 ZUM WENDEN: DER HALSRING AUS GEHWEILEROBERLÖSTERN IM SAARLAND Rudolf Echt

A detailed analysis of an Early La Tène reversible torc from the richest female grave of the western Hunsrück-Eifel culture (at Gehweiler-Oberlöstern), discussing contexts and parallels for its unusual decoration. Für gewöhnlich kann man sie tragen, wie man will. Denn die meisten Halsringe der Früh-La-Tène-Zeit kennen keine Vorder- und Rückseite. Sie sind symmetrisch gestaltet. Was ein Gegenüberstehender sieht, ist auch dem Körper der Trägerin oder des Trägers zugewandt. Das gilt, um nur wenige Beispiele zu nennen, für den Halsring mit Maskenzier von Reinheim (Echt 1999, Taf. 1), für den rankenverzierten Pufferhalsring von Waldalgesheim (Joachim 1995, Abb. 37) oder den palmettengeschmückten Dreiknotenring von Bad Dürkheim (Echt und Thiele 1994, Abb. 21.2). Das ist auch in der Hunsrück-Eifel-Kultur nicht anders, wie die maskenverzierten Halsringe aus den Hügeln 4, 7, 9 und 10 von Losheim zeigen (Groß und Haffner 1969, Abb. 11.2, 14.1, 17.2, 19.1). Für gewöhnlich jedenfalls. Einen ganz ungewöhnlichen Halsring hat W. Reinhard 2001 im Waderner Ortsteil Oberlöstern ausgegraben. Hügel 5 einer kleinen Nekropole barg in einer mit Steinen umstellten Holzkiste das bis heute reichste Frauengrab der westlichen Hunsrück-Eifel-Kultur. Wie üblich, war das Skelett im kalkarmen Boden vergangen. Doch aus der Lage der Fuß-, Finger-, Arm- und Halsringe ließ sich eine NWSO-orientierte Rückenstreckerbestattung rekonstruieren (Reinhard 2003, 76 und Abb. 11). Zwei Eisenfibeln vom Früh-La-Tène-Schema, auf beiden Schultern getragen, vervollständigen das Trachtzubehör. An Beigaben enthielt das Grab ein langes Eisenmesser und sechs Tongefäße: zwei große Fußschalen vom Typ Otzenhausen und vier kleine Schalen. Davon soll zunächst nicht weiter die Rede sein, auch nicht von dem Trachtschmuckensemble. Nur davon,

dass die Tote, ungewöhnlich genug, zwei Halsringe trug: einen dünnstabigen aus Eisen mit Bronzeverschluss im Nacken und bronzener Knotenzier auf der Brust und den prächtigen Bronzehalsring, der Gegenstand dieser Zeilen ist. Der massiv gegossene, offene Halsring mit Pufferenden setzt sich – rein optisch – aus fünf Teilen zusammen. Den glatten, rundstabigen Nackenteil verbinden zwei ovale Knoten mit den beiden ebenfalls rundstabigen Seitenteilen, die jedoch im Gegensatz zum Nackenteil mit gewendelt eingepunzten Linien verziert sind, was diese Abschnitte tordiert erscheinen lässt. Die pseudotordierten Abschnitte enden in schwach plastischen Menschengesichtern, die jenen Darstellungen des Menschengesichts entsprechen, welche der Jubilar vor vielen Jahren als ‘Andernach class’ beschrieben hat (Megaw 1967, 57; 1970). Gleichsam von den Köpfen getragen schließen sich die beiden Endteile an, deren Querschnitt binnen kurzem nicht mehr rundstabig, sondern platt ist und erst wieder an den kleinen, unverzierten Endpuffern Volumen erhält (Abb. 17.1). Diese Endteile sind nicht allein durch ihre flache Form außergewöhnlich – sie tragen auch außergewöhnlichen Dekor. Auf der Vorderseite – das ist die bei der Auffindung nach oben weisende Seite – sind rechtes und linkes Endstück identisch verziert (Reinhard 2003, Abb. 9). Schmale Bronzestege bilden zwei gegenläufige ‘S’-Schwünge, von denen der innere sich am Ende in zwei Ranken teilt. Die eingetieften hornförmigen Felder zwischen den Stegen erzeugen, mit rotem Email gefüllt, einen Bicoloreffekt. Auf der Rückseite bilden allein eingepunzte Linien und Punkte das Muster, und zwar auf jedem Endstück ein anderes: auf dem einen sind ein geschwungenes ‘V’, zwei leierartige Figuren und eine ‘C’-Volute miteinander verkettet, auf dem anderen hat sich eine Wellenranke in zwei ‘C’-Voluten und verschiedene Bogenschnörkel aufgelöst (Abb. 17.2).

17.  Zum Wenden: der Halsring aus Gehweiler-Oberlöstern im Saarland

149

Abb. 17.2. Vorder- und Rückseite der flachen Endstücke des Bronzehalsrings Abb. 1. Landesdenkmalamt des Saarlandes, Am Bergwerk Reden 11, D-66578 Schiffweiler.

Abb. 17.1. Offener Bronzehalsring mit Bügelknoten, Pseudotorsion, Maskenzier und emailverzierter Vorderseite der flachen Endstücke mit kleinen Petschaftenden aus Gehweiler-Oberlöstern, Hügel 5. Landesdenkmalamt des Saarlandes, Am Bergwerk Reden 11, D-66578 Schiffweiler.

In diesem Fall war es also keineswegs gleich, wie der Ring getragen wurde. Er hat seine zwei Seiten, und man könnte seiner Phantasie freien Lauf lassen, könnte über Tagund Nachtseite, Sommer- und Winterseite oder Diesseitsund Jenseitsseite spekulieren, wüsste man nicht, dass damit keine Erkenntnis zu gewinnen ist. Erkenntnis erhoffen darf man sich indes von Form-, Motiv- und Mustervergleichen. Wie eingangs gesagt, kann man bei den meisten Halsringen der Früh-La-Tène-Zeit keine Vorder- und Rückseite unterscheiden. Streng genommen gilt das nur für die Periode La Tène A. Von den wenigen Prototypen in La Tène A abgesehen, kommt in La Tène B mit den vornehmlich am Oberrhein verbreiteten Scheibenhalsringen etwas Neues hinzu: Sie haben eine betonte Schauseite, und die Scheiben tragen Einlagen aus roter Koralle oder rotem Glas (Müller 1989, 13–18). Damit ist eine Unterscheidung in Vorder- und Rückseite gegeben. Da die Rückseite stets unverziert ist, sollte der Ring vernünftigerweise immer so angelegt werden, dass die Schauseite einem Gegenüberstehenden zugewandt ist. Die Wahl, einmal diese, einmal jene Seite zur Schauseite zu machen, hatte die Trägerin eines Scheibenhalsringes nicht. Nun gibt es eine Anzahl Scheibenhalsringe, die zusätzlich mit rotem Email verziert sind, das, wie bei dem Halsring von Oberlöstern, im Grubenschmelzverfahren in vertiefte Felder zwischen hochragenden Stegen eingefüllt worden ist (Müller 1989, 14f). Ein Beispiel dafür ist der Scheibenhalsring aus Hügel 3 von Reiningen (Reiningue) im Elsass, wo die

emaillierten Felder ähnliche S-Ranken einfassen wie auf unserem Halsring (Challet 1993, 49f, Abb. 13). Dieselbe Technik hat V. Defente an Münsinger Fibeln der Stufe La Tène B1 aus dem Gräberfeld Münsingen ‘Rain’ beobachtet, etwa in den Gräbern 49, 80 und 96 (Challet 1992, 219, Abb. 4). In diesen Zusammenhang gehören schließlich auch zwei Meisterwerke keltischer Kunst, denen der Jubilar 1990 eine meisterliche Studie gewidmet hat: die beiden Schnabelkannen aus Basse-Yutz, knapp 60 km von Oberlöstern entfernt (Megaw und Megaw 1990). Neben der Korallenzier am Fuß, auf dem Rand und unter dem Schnabel tragen die beiden Kannen auf Deckel, Henkel und den beiden seitlich auf dem Rand liegenden Tierfiguren Emaildekor (Challet 1993, 57–61, Abb. 18 a.b). Vor allem das Rückenornament der Henkeltiere zeigt dieselben ‘C’Voluten wie das Rückseitenornament des Halsrings von Oberlöstern, und der geschweift-‘V’-förmige Abschluss der emaillierten Mähnen der Tiere kehrt als Abschluss der Ornamentfigur auf der Rückseite des linken Endteils unseres Halsrings wieder. Die flachen Endteile des Oberlösterner Halsrings sind selten. Vergleichbares kenne ich einmal aus der mittelrheinischen Hunsrück-Eifel-Kultur. In Braubach im Rhein-Lahn-Kreis wurde, zusammen mit einer hochhalsigen, gitterglättverzierten Flasche, ein dünnstabiger, offener Halsring mit umgelegten, vogelkopfartigen Enden gefunden. Die Endteile des Ringstabes sind flach und breit, auf der Rückseite glatt, auf der Vorderseite randgekerbt und mit Kreisaugen verziert (Joachim 1977, 1, Abb. 1.1). Kreisstempelverzierung trägt auch ein fragmentarischer Halsring mit platten Endteilen, wohl aus einem zerstörten Grab bei der Klammreis-Kapelle auf dem Dürrnberg (Joachim 1977, 2, Abb. 1.3; Moosleitner, Pauli und Pen­ ninger 1974, 83, Taf. 181.A1). Ein Loch im Zentrum der jeweils vier Kreisaugen weist darauf hin, dass ehedem dort etwas aufgenietet war. Dieses Merkmal verbindet den

150

Rudolf Echt

Halsring vom Dürrnberg mit einem offenen Scheibenhalsring im Museum Székesfehérvár mit der Fundortangabe ‘Rácalmás-Kulcstelep’ (ECA Nr. 227). Der aus Bronze gegossene Ring ist nur im Nackenteil rundstabig, den großen, schälchenförmigen Endpuffern zu aber flach und mit jeweils drei Scheiben versehen, von denen eine noch die ursprünglich aufgenietete Korallenauflage trägt (Müller 1989, Taf. 72). Lediglich an den Enden abgeplattet und gewellt ist ein unverzierter, dünnstabiger Bronzehalsring aus Hügel 26 der Maegstub-Nekropole im Hagenauer Forst (Joachim 1977, 3, Abb. 2.1). Es ist bemerkenswert, dass nur der Scheibenhalsring aus Ungarn Pufferenden besitzt. Alle anderen hatten, wie das Braubacher Exemplar, umgelegte Enden. Die Kreisstempel des Rings vom Dürrnberg waren ursprünglich wie die Scheiben der Scheibenhalsringe andersfarbig gefüllt. Vielleicht darf man im Kreisstempelmuster des Braubacher Rings den Versuch sehen, wenn schon nicht in der Farbgebung, so wenigstens der Form nach den viel aufwändigeren Scheibenhalsringen nachzueifern? Mit Pufferenden kombiniert sind die flachgeschmiedeten und zudem mit eingepunzten Riefen pseudotordierten Endstücke des offenen Halsrings aus BaumholderBreungenborn, Hügel 34, Grab 14 (Haffner 1975, 39, Abb. 14). Das am Rand des Hügels angelegte Grab war stark gestört und nicht in die Stratigraphie des Grabhügels einzuordnen. A. Haffner konnte seinerzeit für den Ringtyp keine Parallele nennen. Eine Parallele ist auch der Oberlösterner Halsring nicht. Aber die Übereinstimmung in den beiden seltenen Merkmalen plattes Endstück und Pseudotorsion durch eingepunzte Riefen lässt an einen Werkstattzusammenhang denken – zumal das Nahetal die Verbindung mit dem knapp 35 km entfernten Baumholder erleichtert haben dürfte. An der Zugehörigkeit der Breungenborner Gräber zur Hunsrück-Eifel-Kultur wird nicht gezweifelt, obschon die Sitte, in einem Hügel zahlreiche Nachbestattungen anzulegen, dieser Kultur eigentlich fremd, im südlichen Saarland und in Lothringen dafür üblich ist. Der Bronzehalsring aus Hügel 5 von Wadern-Oberlöstern vereinigt in sich Merkmale, die aus unterschiedlichen Quellen zusammengeflossen sind. Die platten, nach außen verbreiterten Endstücke teilt er mit dem Scheibenhalsring aus Rácalmás-Kulcstelep, dem ‘Scheibchenhalsring’ vom Dürrnberg, dem Kreisaugenhalsring aus Braubach, dem unverzierten Halsring aus dem Hagenauer Forst und dem Pufferhalsring aus Baumholder. Mit letzterem hat er die Pseudotorsion gemeinsam, die ansonsten in der Champagne auftritt – allein aus dem Gräberfeld von Marson sind mit der Sammlung Morel vier Beispiele ins British Museum gelangt (http://www.britishmuseum.org/research/search_ the_collection_database). Auch die Ornamente auf der Rückseite lassen sich am ehesten nach Westen verfolgen. Das Muster auf dem rechten Endstück geht auf Vorlagen zurück, wie sie auf

den Helmen von Berru und Canosa di Puglia (ECA PP 464–466) vorkommen und auf einer tönernen Fußflasche aus Prunay (Kruta 1991, Abb. S. 201), die nun aber nicht mehr voll verstanden werden: Aus den einem Halbkreis eingeschriebenen dreiblättrigen Palmetten, in Berru in den fortlaufenden Schwung des Musters einkomponiert, sind hier isolierte ‘C’-Schwünge mit eingerollten Enden geworden. Darin wird sich neben dem Qualitätsunterschied wohl auch eine spätere Zeitstellung ausdrücken. Das Kettenmuster auf der Rückseite des linken Endstücks besteht aus einem ‘C’-Schwung und zwei herzförmigen Figuren, die jeweils entstanden sind aus der spiegelbildlichen Gegenüberstellung von massigen Kommamotiven. Dasselbe Motiv, nebeneinander gereiht, begegnet in Kombination mit gegenläufigen ‘S’-Spiralen auf drei lunulaförmigen Bronzebeschlägen aus Étréchy in der Champagne (ECA Nr. 376) – dort, wo auch das Verketten von Leiermotiven gern geübt wurde, etwa bei den durchbrochenen hornförmigen Beschlägen aus dem Wagengrab von La Bouvandeau (ECA Nr. 171) oder den beiden Bronzephaleren aus Saint-Jean-surTourbe (ECA Nr. 184). Allerdings könnte man die Urform für horizontal verkettete herzförmige Motive auch ganz in der Nähe suchen. Nur etwa 9 km von Oberlöstern entfernt, kam in Schwarzenbach, Grab 1, eine großgriechische Zungenamphora zutage, die als Bauchornament ein umlaufendes Band trägt. Darin bilden Palmetten und fortlaufende Wellenspiralen ein Muster, das sich auch als Verkettung herzförmiger Leiern lesen lässt (Haffner 1976, Taf. 145, 146). Bemerkenswert ist, dass keines der Motive und Muster auf der Rückseite des Oberlösterner Halsrings bei der Verzierung von Scheibenhalsringen eine Rolle gespielt hat. Im Gegensatz dazu sind die Motive und das Muster der Vorderseite einem Repertoire entnommen, aus dem sich auch die Meister oberrheinischer Scheibenhalsringe gern bedient haben. Gleichartige Wellenranken zieren etwa einen Scheibenhalsring aus Dornach in der Schweiz (Müller 1989, Taf. 43), aus Uffheim im Elsass (Müller 1989, Taf. 39), aus Schopfloch-Unteriflingen im Schwarzwald (ECA Nr. 233) und aus Bad Nauheim in der Wetterau (Müller 1989, Taf. 56). Auch Fibeln wie die von Stettlen-Deisswil im Kanton Bern wurden mit diesem Motiv dekoriert (Müller 2009, Abb. 270c). Die Verbreitung reicht von der Westschweiz bis ins Rhein-Main-Gebiet. Ebenfalls mit einer Anzahl Scheibenhalsringen, die vor allem in der Schweiz und im Oberrheintal, in geringerer Zahl am Neckar verbreitet sind (Müller 1989, SHR 6–32), ist die Technik gemeinsam, rotes Email in Gruben- oder Furchenschmelz einzusetzen, um einen Bicoloreffekt zu erzielen. Die Maskenzier im Stil der Andernach-Gruppe schließlich ist vornehmlich auf Hals- und Armringen anzutreffen, die von der Westschweiz durch das Elsass bis ins Mittelrheingebiet verbreitet (Megaw 1967, 57f; Megaw 1970, 126–128, Abb. 2), aber auch der westlichen Hunsrück-

17.  Zum Wenden: der Halsring aus Gehweiler-Oberlöstern im Saarland Eifel-Kultur nicht fremd sind, wie der Pufferhalsring aus Losheim, Hügel 4, Grab 1 (Groß und Haffner 1969, Abb. 11.2) und das Beinringpaar aus Rascheid, Hügel D3, Grab 1 (Haffner 1976, Taf. 115,10) verdeutlichen. Mit dem Ring von Oberlöstern kommt nun ein Meisterwerk hinzu, wie um zu bestätigen, was der Jubilar weitsichtig schon vor über 40 Jahren schrieb: ‘…while in the early La Tène period there was a clear cultural link between western Switzerland and the Hunsrück-Eifel, it is the latter area which because of the concentration of objects – as for example in the Losheim cemetery – must have been the centre on which the craftsmen responsible for such products based themselves’ (Megaw 1970, 128). Zu der Frage, wann dies gewesen sein könnte, äußerte sich der Ausgräber so: ‘Aufgrund der Mitgabe der beiden Fußschalen vom Typ Otzenhausen kann das Frauengrab aus Hügel 5 von Gehweiler/Oberlöstern nach HEK IIA3 datiert werden’ (Reinhard 2003, 85), wollte aber eine Datierung bereits in die Stufe HEK II A2 nicht völlig ausschließen (Reinhard 2003, 86). Die vornehmlich von A. Haffner begründete Forschungsmeinung synchronisiert HEK IIA3 überwiegend mit der Spätphase der Periode La Tène A und dem Beginn der Periode La Tène B. Den Gegenstand dieser kleinen Studie überregional bereits in die Stufe La Tène B1 einzuordnen, legt der Grabkontext nahe. Zwar helfen die beiden eisernen Drahtfibeln nicht, weil die Füße weggerostet sind, aber die beiden Knotenarmringe (Reinhard 2003, Abb. 12,7.8) vertreten Typen, die außerhalb der Hunsrück-Eifel-Kultur in Frauengräbern der Periode La Tène B vorzukommen pflegen. Diese Datierung ergibt sich aber auch aus dem Halsring selbst. Zum ersten daraus, dass der Rückseitendekor den Stil des Helms von Berru voraussetzt, der in La Tène A3 eingestuft werden kann, zum zweiten aus dem Wellenrankendekor der Vorderseite, der ein frühes Beispiel für den Waldalgesheimstil der Periode La Tène B abgibt, und zum dritten aus der Emailverzierung mit Grubenschmelz, die sich zum ersten Mal auf den Kannen von Basse Yutz nachweisen lässt, welche, stilkritisch datiert, noch La Tène A3 zugewiesen werden können, deren Blütezeit aber die Perioden La Tène B und C waren. V. Defente hat Recht, wenn sie schreibt: ‘La technique particulière de l’émail se développe, en Europe, dans le courant du IVe siècle avant J.-C.’ (Challet 1992, 217). Im Laufe des 4. Jahrhunderts v. Chr. muss auch der Bronzehalsring von Oberlöstern entstanden sein. Als so oder so zu tragender ‘Wendehalsring’ ist er ein Unikat geblieben.

151

Bibliography Challet, V. 1992. L’émail, innovation technique. Les débuts de l’émail sur le plateau Suisse d’après l’analyse du cimetière de Münsingen-Rain (canton de Berne). In G. Kaenel and P. Curdy (eds.) L’Âge du Fer dans le Jura. Actes du XVe colloque AFEAF tenu à Portalier et Yverdon-les-Bains les 9–12 mai 1991, Cahiers d’Archéologie Romande 57, 217–26. Lausanne: Bibliothèque historique vaudoise. Challet, V. 1993. Les Celtes et l’émail. Documents Préhistoriques 3. Paris: Editions du Comite des travaux historiques et scientifiques. ECA= Jacobsthal, P. 1944. Early Celtic Art. Oxford: Clarendon Press (reprint 1969). Echt, R. 1999. Das Fürstinnengrab von Reinheim : Studien zur Kulturgeschichte der Früh-La-Tène-Zeit. Saarbrücker Beiträge zur Altertumskunde 69. Bonn: Habelt. Echt, R. und Thiele, W.-R. 1994. Von Wallerfangen bis Waldalgesheim. Ein Beitrag zu späthallstatt- und frühlatènezeitlichen Goldschmiedearbeiten. Saarbrücker Studien und Materialien zur Altertumskunde 3. Bonn: Habelt. Gross, N. und Haffner, A. 1969. Ein Gräberfeld der jüngeren Hunsrück-Eifel-Kultur von Losheim, Kreis Merzig-Wadern. Bericht der Staatlichen Denkmalpflege im Saarland 16, 61–103. Haffner, A. 1975. Zwei Grabhügel der Hunsrück-Eifel-Kultur aus Breungenborn-Baumholder, Kreis Birkenfeld. Trierer Zeitschrift 38, 21–56. Haffner, A. 1976. Die westliche Hunsrück-Eifel-Kultur. RömischGermanische Forschungen 36. Berlin: De Gruyter. Joachim, H.-E. 1977. Ein frühlatènezeitlicher Halsring mit Vogelkopfenden von Braubach, Rhein-Lahn-Kreis, Nassauische Annalen 88, 1–8. Joachim, H.-E. 1995. Waldalgesheim; das Grab einer keltischen Fürstin. Katalog Rheinisches Landesmuseum Bonn 3. Pulheim: Rheinland-Verlag. Kruta, V. 1991. Les Celtes de la première expansion historique. In S. Moscati (ed.) Les Celtes. Cat. Expo. Venise, 195–213. Milano: Bompiani. Megaw, J. V. S. 1967. Ein verzierter Frühlatène-Halsring im Metropolitan Museum of Art New York, Germania 45, 50–9. Megaw, J. V. S. 1970. Further Early La Tène rings and other material of the ‘Horchheim’ and ‘Andernach’ classes, Germania 48, 126–30. Megaw, J. V. S. and Megaw, M. R. 1990. The Basse-Yutz find. Masterpieces of Celtic art. The 1927 discovery in the British Museum. Reports of the Research Committee of the Society of Antiquaries of London 46. London: Society of Antiquaries. Moosleitner, F., Pauli, L. und Penninger, E. 1974. Der Dürrnberg bei Hallein 2. Münchner Beiträge zur Vor- und Frühgeschichte 17. München: Beck. Müller, F. 1989. Die frühlatènezeitlichen Scheibenhalsringe. Römisch-Germanische-Forschungen 46. Mainz: von Zabern. Müller, F. (ed.) 2009. Kunst der Kelten 700 v. Chr. – 700 n. Chr. Stuttgart: Belser. Reinhard, W. 2003. Die Siedlungskammer von Oberlöstern / Gehweiler mit reichen Funden von der Früh-La-Tène-Zeit bis in die römische Epoche, Saarbrücker Studien und Materialien zur Altertumskunde 9, 75–124.

18 LATE HALLSTATT AND EARLY LA TÈNE GOLD AND SILVER BEADS IN SOUTHEAST PANNONIA Hrvoje Potrebica and Marko Dizdar

Introduction During the late phase of the Early Iron Age and the beginning of the Late Iron Age, in the territory of southeast Pannonia, among the grave goods accompanying rich female burials, there appeared gold and silver beads. These beads, made in various techniques, have been attributed to influences from the southern part of the Balkan peninsula. In the same territory, graves from the late phase of the Early Iron Age have been attributed to the group of sites lying between Lake Balaton and the River Sava (Teržan 1977), that is, to the Syrmium group (Garašanin 1973; Majnarić-Pandžić 1981; Vasić 1982; 1987; 1989; Medović and Hänsel 2006; Ljuština 2010), dated to the period between the late 7th century and the second half of the 4th century BC. These are flat cemeteries consisting of inhumation burials, with rare incineration graves (Szentlörinc) (Jerem 1968). Typical of this group are skeletal female burials, with grave goods consisting of attire and jewellery, while male graves feature items of weaponry (Guštin and Teržan 1976; 1977). The attire includes bronze astragal belts and various types of fibulae, the most frequent among them being variants V and XIII of fibulae of the Certosa type. The women wore necklaces consisting of glass beads of various shapes, and in several graves gold and silver beads have also been found, as well as beads made of amber and coral. Beads made of precious metals have also been discovered in sites dated to subsequent periods: for example, to the beginning of the Late Iron Age, such as Klasje near the village of Gradac in the south-eastern part of the Požega Valley in central Slavonia.

Dating The 1961 excavation of the Klasje site resulted in finds

belonging to the La Tène Culture: for example, fragments of Early La Tène bronze bracelets, a fragment of an Early La Tène bronze fibula and a large number of varicoloured glass beads of diverse shapes (Potrebica and Balen 1999, 11–12, 28, Pl. 8, Pl. 10, 3). Among other finds, there were also three small and two larger beads made of gold sheet and fragments of gold wire (Potrebica and Balen, 1999, 28). In view of such finds, it was assumed that these could have originated from a La Tène Culture cemetery. In the same year, a test excavation was done on the nearby site of Babišnjača, resulting in the discovery of items originating from various periods of prehistory, including some fragments of Late La Tène pottery (Potrebica and Balen 1999, Pl. 6, 3–4). The two larger beads made of gold sheet, discovered in Klasje, were partially deformed, but originally of an oval shape, with longitudinal ribs and cylindrical extensions of the central hole. Three smaller beads of gold sheet, partially damaged, had been of the same shape, with a ribbed tubelike extension. At the same findsite, there were also two spirally coiled or twisted pieces of gold wire (Fig. 18.1). The closest analogies to the beads from Klasje can be found in Osijek where, in the zone of the Late Hallstatt and Early La Tène cemetery at Zeleno Polje, four biconical beads have been found, made of hammered gold sheet, with a cylindrical extension of the central hole and decorated with longitudinal ribs (Fig. 18.2). On two of these beads, there are visible burning marks (Spajić 1954, 15, pl. V, 34; Šimić 1997, 14, fig. 19), possibly indicating that they originate from a La Tène Culture cremation grave, bearing in mind that the Late Hallstatt graves contained inhumation burials, while the Early La Tène and Middle La Tène graves were bi-ritual. Three more biconical gold beads have been found in the territory of southeast Pannonia – they were discovered in

18.  Late Hallstatt and Early La Tène gold and silver beads in southeast Pannonia

153

Fig. 18.1. Gold beads and wire from Klasje (archive of the Archaeological Museum Zagreb). Fig. 18.3. Gold necklace from grave I in Srijemska Mitrovica.

Fig. 18.2. Gold beads from Osijek (after Šimić 1997).

1901, in the Late Hallstatt inhumation grave I in Srijemska Mitrovica. The beads originated from a necklace, and they used to be separated by four thin tubes made of spirally coiled gold wire, or saltaleone. One of the beads is slightly larger than the other two, and they are all made of gold sheet, and decorated, on each half, with rosettes made of double wire using filigree technique (Fig. 18.3). In the same grave, there were also beads made of amber, a bronze astragal belt with a three-eyed buckle, and two bronze fibulae of the Certosa type with three protrusions on the bow (Brunšmid 1902, 75–77, fig. 36; Vinski 1960, 60, fig. 2, 5). The grave has been attributed to the Syrmium group, and the fibulae have been dated to the late 6th and the early 5th century BC (Guštin and Teržan 1976, 195, fig. 3, 1). Analogies to the gold beads with filigree decoration from Srijemska Mitrovica can be found in princely graves and

similar finds dating from the late phase of the Early Iron Age in the territory stretching from the central Balkans to Chalkidiki: for example, in the tumuli in Atenica, Pećka Banja and Novi Pazar, and among the finds from Kruševica. In the central grave of tumulus I in Atenica, there was a necklace composed of thin tubes made of ribbed gold sheet (imitating spirally coiled wire) and biconical beads made of gold sheet with transversal lines and circles (Djuknić and Jovanović 1965, 8, 13, pl. XV, 16–17; Palavestra 1984, 35; Vasić 1987b, 647; Vasić 2004, 24). Gold and silver decorative plates, and amber and glass beads, were also found in the same grave (Djuknić and Jovanović 1965, 8–9, pl. XV–XVII). The grave used to be dated to the late 6th and early 5th century BC (Guštin and Teržan 1976, 195; Vasić 2004, 21), but in recent papers addressing this topic it has been dated to the second quarter of the 6th century BC (Teržan 1998, 518, fig. 58). As with the grave in Srijemska Mitrovica, the finds from the peripheral grave in tumulus I included two thin tubes made of spirally coiled gold wire (Djuknić and Jovanović 1965, 10, pl. XIX, 12–13). A half of a gold-sheet bead was also found in the central grave of tumulus II (Djuknić and Jovanović 1965, 10, pl. XXI, 4). These biconical gold beads from the tumuli in Atenica are considered to have been produced in a local workshop, inspired by Greek models (Djuknić and Jovanović 1965, 13). In a tumulus in Novi Pazar, also dated to the early 5th century BC, three oval-shaped gold beads, decorated with spirals made using filigree and granulation techniques, have been found. In addition, there were eight small beads made using granulation technique, and thin tubes consisting of spirally coiled wire (Palavestra 1984; Vasić 1987b, 645,

154

Hrvoje Potrebica and Marko Dizdar

pl. LIX, 6, 10), similar to those discovered in the grave in Srijemska Mitrovica. The Kruševica hoard (Srejović and Vukadin 1988, 13), also described as a grave in some publications (Jevtović 1990, 188), has also been dated to the first quarter of the 5th century BC. In addition to other pieces of jewellery, a bronze oinochoe and a ceramic skyphos, the hoard also contained gold and silver beads decorated using granulation technique (Srejović and Vukadin 1988, 13). Two of the gold beads are of oval form, with ring-shaped extension made of filigree wire (Srejović and Vukadin 1988, 7, pl. I, 4–5). A silver gilded bead has also been found. It is biconical, with eight vertical ribs made using filigree technique. A further two beads of the same form are somewhat smaller in size (Srejović and Vukadin 1988, 7–8, pl. IV, 4–6), and similar to the bead discovered in grave 1 in Beremend (Jerem 1973, fig. 5, 5). The Kruševica find also contained two silver beads of an oval shape, ornate with triangular motifs executed in granulation technique (Srejović and Vukadin 1988, 8, pl. IV, 7–8), two silver beads of an oval shape with a largediameter threading hole (Srejović and Vukadin 1988, 8, pl. IV, 9–10), 11 thin tubes made of spirally coiled silver wire with meandering ornament in the middle section, and an amber bead (Srejović and Vukadin 1988, 8, pl. V, 1–4). Slightly larger silver tubes of an identical shape have been found in the princely grave – probably destroyed – in Pećka Banja, dated to the second half of the 6th century BC on the basis of imports of Greek black-figure pottery (Jevtović 1990, 59, 187, fig. 137/11). Although beads made of precious metals have not been found in this grave, the discovered inventory fully corresponds to contemporary princely graves: amber beads, silver fibulae of the Novi Pazar type, a pair of silver bracelets featuring snakes’ heads, a pair of silver bracelets of the Mramorac type, silver omega pins, a ribbed bead made of blue glass paste, and also Greco-Illyrian helmets, a bronze pendant for a whetstone and trilobate arrows (Lucci 1998, 577–586). Trilobate arrows, a bronze pendant for a whetstone of the same type and nearly all forms of jewellery have also been found in Atenica, while Greco-Illyrian helmets have been discovered in the famous necropolises of Trebenište and Sindos. The Sindos graves contained an abundance of gold jewellery, richly decorated with filigree and granulation. Such decoration is present especially on various pendants and necklace beads, such as that from grave 67, which also contained one of the four gold masks discovered in the necropolis (Despini 1985, 182–205, 208–209, figs. 322–323). The grave has been dated to the end of the 6th century BC. The same forms of jewellery and weaponry have been found in the exceptionally richly furnished graves in the necropolis of Trebenište near Ohrid, also dated to the late 6th and the early 5th century BC.

Locally-made or imported? It would appear that the workshops which produced this jewellery, including the gold beads with filigree and granulation decoration, were located somewhere in the area of Chalkidiki, possibly in the vicinity of Sindos. From there, these items spread towards the north, marking their way in the elite graves of the period in the central Balkans. Interestingly, in grave 29 of the cemetery in SopronKrautacker, a necklace has been discovered consisting of glass and amber beads and cowrie shells, and also some tiny gold beads. On the basis of the bronze Pontic earrings coated with electrum, the grave has been attributed to a woman who had come to northwest Pannonia through exogamy, and was buried in line with the local rite (Jerem 1981, 114, fig. 7, 1). These gold beads were not domestically produced in the territory of Pannonia either, but they arrived here from a different direction, which is reflected in their shape, different from the rest of the gold beads discovered in sites in southeast Pannonia. Generally, E. Jerem believes that beads made of precious metals and other similar forms of jewellery are characteristic of the material heritage of Thracians and Illyrians, and considers them, where present in southern and eastern Alpine sites, as imports (Jerem 1973, 81). The discoveries of silver beads, often decorated using filigree and granulation techniques, have also been attributed to southern influences. The best example is the biconical silver bead from grave 1 in Beremend, similar to the abovementioned items from the Kruševica hoard. In the same grave, there were also bronze fibulae of Certosa types I and XIIIa, an astragal belt and glass beads, allowing the grave to be dated to the late 6th or the early 5th century BC (Jerem 1973, 66, fig 5, 5). Fairly different silver beads have been found in the nearby cemetery of Szentlörinc. In each of graves 41 and 44, one biconical silver bead was found, made of horizontally coiled wire (decorated with filigree) (Jerem 1968, 186, fig. 25,41/1, fig. 26,44/2). The graves contained fibulae of Certosa type XIIIh, and fibulae with twisted tip of the foot, indicating that the graves can be dated to the late 5th and early 4th century BC. Beads made of horizontally coiled silver wire have also been discovered in Umčari in northern Serbia (Garašanin 1961, 88, fig. 8) and in Nikinci in Syrmium, and dated to the period between the 6th and 4th century BC (Vasić 2005, 68, fig. 12–16). In addition, two silver rings have been found in Nikinci, featuring a rhomboid plate decorated with a rosette motif (Vasić 2005, 68, figs. 21–22). Analogies for these items can be found in the Čurug hoard, containing three silver rings with oval plates decorated with triangular and circular motifs, and considered to be southern influences (Vasić 1995, 84, 87, fig. 1, 8). The Čurug hoard included some exceptionally

18.  Late Hallstatt and Early La Tène gold and silver beads in southeast Pannonia valuable pieces of female jewellery, such as silver rings and earrings, richly decorated using filigree and granulation techniques. Other pieces of jewellery have also been found there: for example, diverse bracelets and beads, and silver, bronze and glass elements of female attire (fibulae, parts of the belt). All the finds from this eponymous hoard have been attributed to the Čurug chronological phase, dated to the end of the first half and the second half of the 4th century BC (Božič 1984; Vasić 1988; Popović 1996; Vasić 1995; Ljuština 2010; Rustoiu 2012). Unlike the gold and silver jewellery from Srijemska Mitrovica and the Atenica and Novi Pazar tumuli, the finds from Nikinci and Čurug have been linked to the local south-Pannonian workshops. Silver and gold belts of the Mramorac type – the greatest number of which have been discovered in sites in the Morava region (Stojić 2007a; 2008) – are considered to be of local origin, and dated to the late 6th and the early 5th century BC, and the silver spirally coiled bracelets and earrings appearing in both Late Hallstatt and Early La Tène graves are local southPannonian forms, attesting to a longer period of use of some of the forms of jewellery made of precious metals (Jovanović 1994; 2007; Majnarić-Pandžić 1995, 36; Dizdar and Potrebica 2002; Rustoiu 2012). Silver bracelets whose ends might finish in snake-like heads (Bogdanovci, Čurug, Baranja) or in trapezoid forms (Čurug, Srijemska Mitrovica II, Sombor and several sites in northwest Bulgaria) can be taken as local products, too. The workshops in which they were produced have been placed in southeast Pannonia and northwest Bulgaria, and the models for their style have been identified in Macedonian bracelets with snake-like heads (Vasić 1995, 83–86; 2001, 26–27). On the other hand, the discovered gold and silver boatshaped earrings decorated using filigree and granulation techniques testify to the fact that not all items made of precious metals in the late Early Iron Age were produced in local workshops. In the territory of southeast Pannonia, two such silver earrings have been found in Erdut (Fig. 18.4) in eastern Slavonia, and one gold earring originates from each of Srijemska Mitrovica and Putinci in Syrmium. The earrings have been dated to the first half of the 4th century BC and are considered imports from Macedonian workshops, although the possibility that they, too, were produced locally cannot be ruled out (Vinski 1960; Vasić 1991; 2001, 26). However, even if these were local products, the idea must have come from the south, which is best attested to by the pair of silver earrings discovered in a grave from the first half of the 4th century BC in Ždanec (Sokolovska and Pašić 1975, 236–238, pl. III, 1–2), or those from the contemporary grave 1 in tumulus IX at Rudine (Zotović 1991, 85, fig. 6, 3). A similar silver earring was discovered in Radolište (Popović 1994, 143, fig. 134), and a further silver pair of

155

Fig. 18.4. Silver earrings from Erdut (archive of the Archaeological Museum Zagreb).

unknown origin is kept in the National Museum in Belgrade (Popović 1994, 143). Yet another argument in favour of the southern origin of the earrings is provided by a very similar gold piece from the destroyed grave in Ulpiana, dated to the end of the 5th and the beginning of the 4th century BC (Fidanovski 1998, 591, fig. 231). Another import that came from the south, that is, from the territory of the Balkans, was silver omega pins of the IIId type, with heads of a circular (Dalj, Vineyard L. Poštić, grave 89) or rectangular cross-section (Bogdanovci). Such pins have been found in south-Pannonian sites and dated to the 5th and 4th century BC (Vinski-Gasparini 1973, 163; Vasić 1988, 170; 2003, 122–123). The small silver box from Bogdanovci is probably an import which came from the same direction (Brunšmid 1909), and direct analogies for that item can be found in the Ždanec grave, mentioned above, from the early 4th century BC (Sokolovska and Pašić 1975, 241, pl. IV, 2). The small gold box from the Tremnik hoard is also similar, and it has been dated to the end of the 4th or the beginning of the 3rd century BC (Mitrevski 2011, 202, fig. 5).

156

Hrvoje Potrebica and Marko Dizdar

Southeast Pannonia and the Southern Balkans The discovered items made of precious metals referred to above, mostly elements of female attire and jewellery, suggest that, in the period between the 6th and 4th century BC, there were some complex ties between the territory of southeast Pannonia and the southern Balkans. The gold and silver beads decorated using filigree and granulation techniques found at the sites of southeast Pannonia and dated to the end of the Early Iron Age were probably imports from the south, that is, from Macedonia. Workshops which produced such jewellery were located in the area of Chalkidiki, as demonstrated by richly furnished graves in the necropolis of Sindos. From there, the communication network spread northwards, which is attested to by exceptionally rich elite graves from the end of the 6th and the beginning of the 5th century BC, such as those in Trebenište, Pećka Banja, Novi Pazar, Kruševica and Atenica. Soon thereafter, such items emerged in the southeast of the Pannonian Valley, as shown by the beads discovered in the richly furnished female graves in Srijemska Mitrovica and Beremend, also dated to the end of the 6th and the beginning of the 5th century BC. The communication and exchange definitely did not go in only one direction, which is demonstrated by the trilobate arrow-heads and necklaces made of unprocessed amber discovered at the site of At, near Vršac (Vasić 1987, 564–565), and also in the princely graves in Atenica and Pećka Banja. Both princely graves in Atenica (Djuknić and Jovanović 1965, 20, T. XIII, 10, T. XVIII, 3–5, T. XXV, 5–6) also contained bridle-bits of the Vekerzug type considered to be of Scythian origin. Such bits have been found regularly in the necropolises of Beremend and Szentlörinc (Jerem 1968; 1973). In the territory of southern Pannonia, in addition to beads, silver was used to produce other elements of attire and jewellery, such as fibulae, rings, earrings and belts. The typological diversity of such items and their greater number after the middle of the 1st millennium BC indicate that there were local south-Pannonian workshops, probably located somewhere in Syrmium (Vasić 2001, 25; 2005, 72; 2006, 120–121). Therefore, even the beads found in southeast Pannonia and attributed to the Early La Tène contexts had probably been produced in local workshops. The gold beads from the sites of Klasje and Osijek fit into this picture, and despite the fact that the exact context of their discovery is not known, they can probably be dated to the second half of the 4th century BC. Nonetheless, the links to the south were maintained during Hellenism, as attested to by the settlements of Hisar and Kale-Krševica in southern Serbia. These well-researched sites were important centres on the famous Vardar-Morava communication route, in which objects, ideas and knowledge were exchanged (Stojić 2007b; Popović 2006; 2012). The contacts of the second half of the 4th century BC are further confirmed by the bronze vessels

in the hoard of Rgajski Grad (Stojić 2007c), and the bronze vessels from the Early La Tène grave 22 in Karaburma are of the same origin. The latter should be considered as diplomatic gifts or traded objects, rather than war booty resulting from a campaign in the south, thus brought to the area surrounding the mouth of the River Sava on the Danube (Blečić Kavur and Kavur 2010, 75–76). At the beginning of the Late Iron Age, in the territory stretching from the Morava region to the Danube region, the cultural community of the Scordisci was formed, and it maintained communication with the south. In a certain way, this very communication caused the formation of this cultural group, because its members, like those before them, always yearned for precious items from the south. The fact that gold beads from the site in Klasje can be linked to the territory of the Scordisci on the basis of their direct analogies from Osijek indicates that both the eastern Sava region and the Požega Valley probably belonged to the same cultural circle.

Bibliography Blečić Kavur, M. and Kavur, B. 2010. Grob 22 iz beogradske nekropole Karaburma: retrospektiva i perspektiva, Starinar 60, 57–84. Božič, D. 1984. Srebro i bronca iz Čuruga (Bačka). In KEΛTOI, Ljubljana, 31–32. Brunšmid, J. 1902. Prethistorijski predmeti iz Srijemske županije, Vjesnik Hrvatskog arheoloskog drustva n. s. 6, 68–86. Brunšmid, J. 1909. Prethistorijski predmeti iz Srijemske županije, Vjesnik Hrvatskog arheoloskog drustva n. s. 10, 231–237. Despini, K. (ed.) 1985. Sindos (exhibition catalogue). Thessaloniki: Thessaloniki Archaeological Museum. Dizdar, M. and Potrebica, H. 2002. Latenska kultura na prostoru Požeške kotline, Opuscula Archaeologica 26, 111–31. Djuknić, M. and Jovanović, B. 1965. Illyrian Princely Necropolis at Atenica, Archaeologia Iugoslavica 6, 1–37. Fidanovski, S. 1998. Ulpijana. In Tasic, N. (ed.) Arheološko blago Kosova i Metohije od neolita do ranog srednjeg veka, 590–611. Belgrade: Serbische Akademie der Wissenschaften und Künste. Garašanin, D. 1961. Studije iz metalnog doba Srbije, VI. Jedna praistorijska livnica u zapadnoj Srbiji, VII. Srebrni ilirski nakit iz Umčara, Starinar 11 (1960), 75–92. Garašanin, M. 1973. Gvozdeno doba III (Sremska grupa zapadnobalkanskog kompleksa), Praistorija na tlu Srbije II, 511–15. Guštin, M. and Teržan, B. 1976. Malenškova gomila v Novem mestu, Prispevek k poznavanju povezav med jugovzhodnim alpskim svetom, severozahodnim Balkanom in južno Panonijo v starejši železni dobi, Arheološki Vestnik 26, 188–202. Guštin, M. and Teržan, B. 1977. Beiträge zu den vorgeschichtlichen Beziehungen zwischen dem Südostalpengebiet, dem nordwestlichen Balkan und dem südlichen Pannonien im 5. Jahrhundert. In V. Markotic (ed.) Ancient Europe and the Mediterranean, Festschrift H. Hencken, 77–89. Warminster: Aris & Phillips.

18.  Late Hallstatt and Early La Tène gold and silver beads in southeast Pannonia Jerem, E. 1968. The Late Iron Age Cemetery of Szentlörinc, Acta Archaeologica Academiae Scientiarum Hungaricae 20, 159–208. Jerem, E. 1973. Zur Geschichte des späten Eisenzeit in Transdanubia: späteisenzeitliche Grabfunde von Beremend, Acta Archaeologica Academiae Scientiarum Hungaricae 25/1–2, 65–86. Jerem, E. 1981. Zur Späthallstatt und Frühlatènezeit in Transdanubia. In C. Eibner (ed.) Die Hallstattkultur, Bericht über das Symposium in Steyr 1980 aus Anlaß der Internationalen Ausstellung des Landes Oberösterreich, 105–36. Linz: Oberösterreichischer Landesverlag. Jevtović, J. (ed.) 1990. Gospodari srebra. Gvozdeno doba na tlu Srbije. Belgrade: Narodni muzej Beograd. Jovanović, B. 1994. Horizont najstarijih keltskih grobova na severnom Balkanu. In N. Tasić (ed.) Kulture gvozdenog doba jugoslovenskog Podunavlja, 111–117. Belgrade: Balkanološki institut SANU, Posebna izdanja 55. Jovanović, B. 2007. Srebrne naušnice u nakitu ranog latena Srednjeg Podunvlja. In M. Blečić, M. Črešnar, B. Hänsel, A. Hellmuth, E. Kaiser and C. Metzner-Nebelsick (eds.) Scripta Praehistorica in honorem Biba Teržan, Situla 44, 821–7. Ljubljana: Narodni muzej Slovenije. Luci, K. 1998. Pećka Banja. In Tasic, N. (ed.) Arheološko blago Kosova i Metohije od neolita do ranog srednjeg veka, 577–86. Belgrade: Serbische Akademie der Wissenschaften und Künste. Ljuština, M. 2010. The Late Hallstatt Communities in the Serbian part of the Danube Basin. In S. Berecki (ed.) Iron Age Communities in the Carpathian Basin, Proceedings of the International Colloquiums from Târgu Mureş 9–11 October 2009, Bibliotheca Mvsei Marisiensis, Seria Archaeologica II, 59–78. Cluj-Napoca: Editura Mega. Majnarić-Pandžić, N. 1981. Vinkovci kod Silosa – kasnohalštatski grobovi, Arheološki Pregled 21, 39–40. Majnarić-Pandžić, N. 1995. Nekoliko napomena o uvođenju ranolatenskog stila u sjevernu Hrvatsku i Bosnu, Arheološki radovi i rasprave 12, 31–53. Medović P. and Hänsel B. 2006. Die Srem-Gruppe – Nekropolen bei den Siedlungen der Bosut-Gruppe. in: N. Tasić and C. Grozdanov (eds.) Homage to Milutin Garašanin, 489–512. Belgrade: Serbian Academy of Sciences and Arts – Macedonian Academy of Sciences and Arts. Mitrevski, D. 2011. The treasure from Tremnik and some traces of the Celts in the Vardar valley. In M. Guštin and M. Jevtić (eds.) The Eastern Celts, The Communities between the Alps and the Black Sea, 199–206. Koper-Belgrade: Univerza na Primorskem, Znanstveno-raziskovalno središče, Univerzitetna založba Annales. Palavestra, A. 1984. Kneževski grobovi starijeg gvozdenog doba na Centralnom Balkanu, Belgrade: Balkanološki Institut SANU, Posebna izdanja, Knjiga 19. Popović, Lj. B. 1994. Antička grčka zbirka. Antika VII. Belgrade: National Museum Belgrade. Popović, P. 1996. Early La Tène Between Pannonia and the Balkans, Starinar 47, 105–25. Popović, P. 2006. Central Balkans between the Greek and Celtic World: Case Study Kale-Krševica. In N. Tasić and C. Grozdanov (eds.) Homage to Milutin Garašanin, 523–536.

157

Belgrade: Serbian Academy of Sciences and Arts-Macedonian Academy of Sciences and Arts. Popović, P. 2012. Centralni Balkan između grčkog i keltskog sveta. In Centralni Balkan između grčkog i keltskog sveta. Kale-Krševica 2001–2011, 11–51. Narodni muzej (katalog izložbe): Beograd. Potrebica, H. and Balen, J. 1999. Prapovijesni nalazi iz Požeške kotline u Arheološkom muzeju u Zagrebu, Godišnjak Požeštine Zlatna dolina 5(1), 7–61. Rustoiu, A. 2012. The Celts and Indigenous Populations from the Southern Carpathian Basin. Intercommunity Communication Strategies. In S. Berecki (ed.) Iron Age Communities in the Carpathian Basin, Proceedings of the International Colloquiums from Târgu Mureş 9–11 October 2009, Bibliotheca Mvsei Marisiensis, Seria Archaeologica II, 357–390. Cluj-Napoca: Editura Mega. Sokolovska, V. and Pašić, R. 1975. Eden grob od Ždanec, Zbornik Arheološki muzej na Makedonija knj. 6–7 (1967–1974), 231–44. Spajić, E. 1954. Nalazišta mlađeg željeznog doba s terena Osijeka, Osječki Zbornik 6, 7–18. Srejović, D. and Vukadin, O. 1988. Blago iz Kruševice, Raška baština 3, 7–18. Stojić, M. 2007a. Zwei neue Gürtel aus Edelmetall vom Typ Mramorac aus Batinac in Serbien, Prähistorische Zeitschrift 82/1, 51–65. Stojić, M. 2007b. Hisar in Leskovac at the End of the Early Iron Age, Starinar 57, 175–89. Stojić, M. 2007c. Grupni nalaz bronzanih posuda IV veka pre n. e. na lokalitetu Rgajski grad u selu Rgaje kod Prokuplja. In M. Blečić, M. Črešnar, B. Hänsel, A. Hellmuth, E. Kaiser and C. Metzner-Nebelsick (eds.) Scripta Praehistorica in honorem Biba Teržan, Situla 44, 759–67. Ljubljana: Narodni muzej Slovenije. Stojić, M. 2008. New Finds from Rutevac and Deliberation on Purpose, Origin Place of Production and Ethnic Attribution of Mramorac Type Belts, Starinar 58, 87–94. Šimić, J. 1997. Kelti. In Kelti i Rimljani na području Osijeka, 3–49. Osijek: Muzej Slavonje Osijek. Teržan, B. 1977. O horizontu bojevniških grobov med Padom in Donavo v 5.in 4. stol. pr. n. št.. In M. Guštin (ed.) Keltske študije, 9–21. ed., Brežice: Posavski muzej Brežice knj. 4. Teržan, B. 1998. Auswirkungen des skytisch geprägten Kulturkreis auf die hallstattzeitlichen Kulturgruppen Pannoniens und des Ostalpenraumes. In B. Hänsel and J. Machnik (eds.) Das Karpatenbecken und die osteuropäische Steppe, Prähistorische Archäologie in Südosteuropa Band 12, 511–60. Rahden, Westf.: Leidorf. Vasić, R. 1982. A contribution to the study of ‘Illyrian’ helmets in north Yugoslavia, Archaeolgica Iugoslavica 22–23, 76–80. Vasić, R. 1987. Sremska grupa zapadnobalkanskog kompleksa. In A. Benac (ed.) Praistorija jugoslovenskih zemalja V, Željezno doba, 555–8. Sarajevo: Akademija Nauka i Umjetnosti Bosne i Hercegovine, Centar za Balkanološka Ispitivanja. Vasić, R. 1987. Kneževski grobovi iz Novog pazara i Atenice. In A. Benac (ed.) Praistorija jugoslovenskih zemalja V, Željezno doba, 644–50. Sarajevo: Akademija Nauka i Umjetnosti Bosne i Hercegovine, Centar za Balkanološka Ispitivanja. Vasić, R. 1988. Ein Beitrag zur Chronologie der Späthallstattzeit

158

Hrvoje Potrebica and Marko Dizdar

im Sremgebiet. In N. Tasić and J. Petrović (eds.) Gomolava 1, Cronologie und Stratigraphie der vorgeschichtlichen und antiken Kulturen der Donauniederung und Südosteuropas, 169–76. Novi Sad: Vojvoðanski Muzej. Vasić, R. 1989. Jedan prilog proučavanju Sremske grupe, Godišnjak Centra za balkanološka ispitivanja knj. 27/25, 103–13. Vasić, R. 1991, Boat-shaped earrings in the Central Balkan Area, Starinar 40–41 (1989–1990), 135–40. Vasić, R. 1995. Srebrni nakit IV veka pre n. e. na teritoriji srednjeg Podunavlja. In Radionice i kovnice srebra, 83–91. Belgrade: Narodni muzej u Beogradu. Vasić, R. 2001. Gold and Silver in Iron Age Serbia, Archaeologia Bulgarica 5/3, 23–8. Vasić, R. 2003. Die Nadeln im Zentralbalkan, PBF Abteilung 13, Band 11. Stuttgart: Steiner.

Vasić, R. 2004. Beleške o Atenici, ZRNM Čačak 34, 21–7. Vasić, R. 2005. Srebrninalaz iz Nikinaca, Starinar 55, 67–73. Vasić, R. 2006. Connections between Serbia and Macedonia in the 4th Century BC. In E. Maneva (ed.) In Honorem Verae Bitrakova Grozdanova, 119–24. Skopje: Folia Archaeologica Balkanica I. Vinski, Z. 1960. Povodom izložbe “Iliri i Grci”, Vijesti Mmuzelaca I konzervatora Hrvatske 9/2, 57–61. Vinski-Gasparini, K. 1973. Kultura polja sa žarama u sjevernoj Hrvatskoj. Zadar: Monografije Filozofskog fakulteta u Zadru 1. Zotović, M. 1991. Istraživanja praistorijskih humki u dolini Poblačnice i geneza ranog bronzanog doba u zapadnoj Srbiji, Starinar 40–41 (1989–1990), 77–88.

19 EAST MEETS WEST... THE STAMPED POTTERY FROM THE LA TÈNE CEMETERY AT FÂNTÂNELE-DEALUL POPII, (TRANSYLVANIA, ROMANIA) Aurel Rustoiu

Oh, East is East, and West is West, and never the twain shall meet, Till Earth and Sky stand presently at God’s great Judgment Seat; But there is neither East nor West, Border, nor Breed, nor Birth, When two strong men stand face to face, tho’ they come from the ends of the earth! (Rudyard Kipling, The Ballad of East and West, 1889).

Introduction In the first lines of a famous ballad, Rudyard Kipling was rightly pointing out that in spite of the apparent geographic and cultural opposition between the eastern and western worlds, some outstanding individuals belonging to these two spaces may discover that they shared similar values and ideas, thus being able to find a common language. On the same note, in a well-known synthesis regarding Celtic art published more than two decades ago, Ruth and Vincent Megaw use the line East meets West... to better describe the cultural encounter between the Celts and the local populations from the Carpathian Basin and the Balkans (Megaw and Megaw 2001, 144–51). During the last decades Vincent Megaw, to whom this volume is warmly dedicated, had the opportunity to directly investigate a series of important La Tène discoveries from Transylvania, including the funerary inventories from Fântânele. In September 1974, during a study visit in Romania, he had seen for the first time the finds from Fântânele-Dealul Popii while meeting and discussing with the late Dr. Ion Horaţiu Crişan, the site director. Our first meeting happened some decades later, at the beginning of the 1990s, when Vincent again visited Romania. Later, in October 2009 and also in Cluj-Napoca, we had another occasion to discuss the funerary inventories from Fântânele,

as well as many other aspects of the La Tène world (Fig. 19.1). One tangible result of these discussions is the publication of an article about a bracelet decorated in the Vegetal Style, coming from the above-mentioned cemetery, in a recent volume dedicated to Professor Mircea Babeş, another outstanding researcher of the La Tène period (Rustoiu and Megaw 2011). Over the last decades Vincent Megaw has expressed a continuous interest in the Celtic finds from Transylvania and their place in the culture and civilization of the eastern Celtic world, and so the present article is honouring him by discussing the ceramic vessels decorated with stamped motifs coming from the La Tène cemetery at Fântânele-Dealul Popii (Bistriţa-Năsăud County, Romania). The decorated pottery from this eastern Transylvanian cemetery also illustrates, amongst many archaeological artefacts and contexts, the aforementioned connection between the Orient and the Occident.

The finds The site was archaeologically investigated from 1969 until 1974 by I. H. Crişan from the Institute of History of the Romanian Academy from Cluj-Napoca. The investigations uncovered 84 burials, eight being inhumed and the remaining ones cremated (see some short notes in Crişan 1975; 1977, 75–7, 79–82). Aside from them, ten other graves were previously excavated (Dănilă 1978). Chronologically the cemetery evolves from the La Tène B2 to the La Tène C1 sub-phase, more precisely from the end of the 4th century to the beginning of the 2nd century BC. The site is very important for the Celtic horizon in Transylvania due to the large number of burials and its long lasting evolution. As a consequence, the related discoveries attracted the interest of many specialists shortly after the conclusion of field

160

Aurel Rustoiu

Fig. 19.1. Vincent Megaw (left) analysing the finds from Fântânele-Dealul Popii during a study visit in Cluj-Napoca in October 2009.

investigations. However, the cemetery remained largely unpublished (some preliminary studies have been published more recently: Rustoiu 2008a, 55, 76–8, 121–3, fig. 22, 35, 59; 2008b, 26–7, fig. 2; 2009, 10–11, fig. 2/4–5; 2011, 164–5; Rustoiu and Egri 2010, 25–7; Rustoiu and Megaw 2011). Amongst the artefacts from the funerary inventories discovered at Fântânele-Dealul Popii is a series of ceramic vessels decorated with stamped motifs. Their analysis begins with the list of vessels and of their contexts of discovery.

Grave no. 17 The rite is cremation, with the remains placed into a pit. The inventory consists of the panoply of weapons (sword, spear head and shield-boss, all made of iron), two iron knives and four ceramic vessels. Two of them have a stamped decoration. The grave is dated to the LT C1. 1. A wheel-made pitcher having a fine grey fabric. Dimensions: height – 320 mm; diameter of rim – 125 mm; maximum diameter – 220 mm; diameter of base – 115 mm. Decoration: six identical combinations of four concentric

circles (two horizontally and two vertically placed) were symmetrically stamped on the neck and shoulder. Initially the circles were inlaid with a white substance, but only some traces are now preserved. Diameter of circles: 9.5 mm (Fig. 19.2/1–2). 2. A wheel-made bi-truncated beaker having a ring base and being made of a fine black, burnished fabric. Dimensions: height – 90 mm; diameter of rim – 75 mm; maximum diameter – 95 mm; diameter of base –50 mm. Decoration: five identical groups of two concentric circles were stamped above the maximum diameter of the body. Like the first vessel, some traces of the white substance which once filled the circles can sometimes be noted. The diameter of the circles is identical with that of the circles stamped on the first vessel, indicating that the same stamping tool was used in both cases (Fig. 19.2/3–4).

Grave no. 28 The rite is cremation, with the remains placed into a pit. The inventory consists of the panoply of weapons (sword, spear head and shield-boss, all made of iron), a bronze brooch

19.  The stamped pottery from the La Tène cemetery at Fântânele-Dealul Popii

Fig. 19.2. Fântânele-Dealul Popii. Vessels from grave no. 17.

161

162

Aurel Rustoiu

of the middle La Tène scheme and three ceramic vessels. One of them is decorated with stamped motifs. The grave is dated to the LT C1. 3. A large bi-truncated vessel, wheel-made, having a fine brown-reddish fabric. Dimensions: height – 340 mm; diameter of rim – 200 mm; maximum diameter – 335 mm; diameter of base – 110 mm. Decoration: three groups of stamped elements were placed above the maximum diameter of the body. Each group consists of rows of four rosettes (resembling a small wheel with four spokes) and four concentric circles, grouped in parallel and horizontally, united by arches consisting of incised dots. In two groups the rosettes are on the upper line and the circles are on the lower line, whereas in the remaining group their position is reversed. Diameter of circles: 7 mm; diameter of rosettes: 11 mm (Fig. 19.3/1–2).

Grave no. 40 The rite is cremation, with the remains placed into a pit. The inventory consists of two bronze anklets with large hollow knobs, two iron bracelets, three bronze brooches of the middle La Tène scheme, one bronze chain and three ceramic vessels. One of them is decorated with stamped motifs. The grave is dated to the LT C1. 4. A wheel-made bi-truncated pitcher having a fine grey fabric. Dimensions: height – 260 mm; diameter of rim – 128 mm; maximum diameter – 220 mm; diameter of base – 100 mm. Decoration: a stripe in relief, decorated with oblique incisions, was placed above the maximum diameter of the body. Three identical groups of stamped elements were placed below it. Each group consists of five stamped concentric circles surrounded by other circles consisting of fine incisions. Diameter of stamped circles: 5 mm (Fig. 19.3/3–4).

a fine black, burnished fabric. Dimensions: height c. – 120 mm; diameter of rim – 160 mm; maximum diameter – 192 mm; diameter of base – 70 mm. Decoration: three groups consisting of four concentric circles, united by arches, were stamped above the maximum diameter of the vessel. The motif is similar to the one present on the previous vessel and was made with the same stamping tool (Fig. 19.4/2).

Grave no. 42 The rite is cremation, with the remains placed into a pit. The inventory consists of three iron bracelets (or maybe one bracelet and two anklets), three iron brooches of the middle La Tène scheme and three ceramic vessels. Two of them have a stamped decoration. The grave is dated to the LT C1. 7. A large bi-truncated vessel, wheel-made and having a fine brown fabric, burnished on the outside. Dimensions: height – 331 mm; diameter of rim – 162 mm; maximum diameter – 302 mm; diameter of base – 111 mm. Decoration: three groups of decorative elements were stamped above the maximum diameter of the vessel. Each group consists of four stamped concentric circles, surrounded by a circle of fine incised lines. The latter are also surrounded by a further 11 or 12 stamped circles. The two groups of motifs are united by two arches consisting of fine incisions, each having a stamped circle in the middle. Diameter of stamped circles: 7 mm (Fig. 19.4/3–4). 8. A wheel-made bi-truncated beaker having a fine grey fabric. Dimensions: height – 115 mm; diameter of rim – 85 mm; maximum diameter – 120 mm; diameter of base – 55 mm. Decoration: triangular groups consisting of three concentric circles were stamped above the maximum diameter of the vessel. Diameter of stamped circles: 8 mm (Fig. 19.4/5).

Grave no. 41

Discussion

The rite is cremation, with the remains placed into a pit. The inventory consists of an iron sword, three iron brooches of the middle La Tène scheme, one blue glass bracelet (Haevernick 1960, type 6b) and five ceramic vessels, of which two have a stamped decoration. The grave is dated to the LT C1. 5. A large bi-truncated vessel, wheel-made and having a fine brown-reddish fabric. Dimensions: height c. – 250 mm; diameter of rim – 122 mm; maximum diameter – 230 mm; diameter of base – 80 mm. Decoration: three groups consisting of four concentric circles were stamped above the maximum diameter of the body, between two parallel stripes in relief. The groups are united by arches consisting of incised lines. Diameter of stamped circles: 5.5 mm (Fig. 19.4/1). 6. A wheel-made carinated bowl, having a ring base and

The eight vessels presented above belong to a relatively unitary group characterised by a broadly similar style of decoration. They all belong to the ceramic category decorated with stamped concentric circles combined in various manners. This type of decoration was widespread in Europe (Schwappach 1977, 119, Fig. 19.1 – distribution map). Although the stamped decoration was considered to be specific to the early eastern La Tène art (see for example Schwappach 1973; contra Szabó 1992, 112), in reality it was encountered across a vast area, from the western extremity of France to Transylvania, sometimes reaching the regions eastward of the Carpathians, in Moldova (Babeş 1993, 71, pl. 39, 40), or even eastern Bulgaria, an area situated in the hinterland of the Celtic kingdom of Thrace, having its centre at Tylis or in its vicinity (Lazarov 2010, 105, Fig. 19.5/1, 3; Vagalinski 2007, 82, fig. 18; Anastassov 2011,

19.  The stamped pottery from the La Tène cemetery at Fântânele-Dealul Popii

Fig. 19.3. Fântânele-Dealul Popii. Vessels from graves no. 28 (1–2) and 40 (3–4).

163

164

Aurel Rustoiu

232, fig. 18). Chronologically such ornaments are present throughout the early and middle La Tène (LT A–C), but in certain areas they are already used in the middle Hallstatt period, being included in complex figurative compositions, for example on a series of funerary urns from the cemetery at Sopron-Burgstall (Megaw 1970, 46, no. 10; Eibner 1996, 108, Pl. 5 etc.). In spite of this vast distribution area and of a wide chronological span, several particular patterns have been identified in different regions or sites, regarding the frequency, decorative structure and motif combination on various ceramic forms. In general the number of vessels decorated with stamped motifs recovered from cemeteries is reduced by comparison with the total number of ceramic containers. Therefore, it can be presumed that the stamped motifs were not just a simple popular manner of ornamentation, but must have had a specific symbolic significance. Furthermore, it has to be noted that the use of stamped decoration had a different chronological evolution from one community to another. For example at Mannersdorf (Ramsl 2011), in the western part of the Carpathian Basin, the vessels with stamped decoration (consisting of concentric circles) are present in graves belonging to the early phases of the cemetery (LT A–B1). In this case the 12 vessels stamped with concentric circles represent 19.35% of the total number of funerary pottery belonging to the mentioned phases. On the other hand, in the subsequent phases only one vessel of this type has been placed in a grave, suggesting that the popularity of this decorative style gradually diminished within this community. The situation is totally different in other cemeteries from the Carpathian Basin. For example at Ludas – Varjú-dűlő (Szabó 2012) the stamped pottery (four vessels) represents 2.56% of the total number of vessels recovered from graves, but from the chronological point of view they are equally distributed in the first (LT B2a) and the last phase (LT C1) of the cemetery, while missing from the middle phase (LT B2b). In the cemetery from Pişcolt, in north-western Romania (Németi 1988b; 1989; 1992), the percentage of the stamped pottery belonging to the early horizons (LT B1/B2–B2a) is 2.32%, whereas the figure increases to 6.66% in later horizons (LT B2b–C1). Lastly, at Remetea Mare – Gomila lui Pituţ (Medeleţ ms.) in Banat (western Romania), a small cemetery dated to the LT B2 was excavated, in which a single vessel, decorated with stamped concentric circles associated with lyre motifs, was found (Megaw and Megaw 2001, 263, fig. 443; 2006, 371, fig. 21), representing 0.44% of the total number of vessels from the graves. These statistical data seem to suggest that the stamped concentric circles were more frequently used at the beginning of the La Tène period (LT A–B1), Mannersdorf being a good example, whereas their popularity diminished in the LT B2. A rather timid revival of this decorative style can be noted in the La Tène C1, at least in some communities

from the Carpathian Basin. For example, a series of preliminary results from the cemetery at Zvonimirovo, in Croatia, seems to indicate an intensification of the use of this decoration mainly in the middle LT period (Dizdar 2011, 104). However, these observations have to be further verified through a statistic analysis of the inventories from a larger number of cemeteries and settlements from this region. Moreover, the stamped vessels (representing 3.63% of the total number of ceramic vessels) are only present in the last phase of the cemetery from Fântânele-Dealul Popii, dated to the LT C1. This situation can be interpreted as a resurgence or rememoration of some earlier traditions from the community’s region of origin (probably located in Central Europe), from which some Celtic groups of ‘colonists’ migrated to the east, in Transylvania, a few generations before. Some other archaeological evidence of this eastward movement of the Celtic groups, and of the persistence of the collective memory of the first ‘colonists’, has been identified at Fântânele. One example is the iron bracelet enamelled and decorated in the Vegetal Style, recovered from grave no. 62. The object was produced at the end of the 4th or the beginning of the 3rd century BC (LT B2), somewhere in Western Europe. Afterwards it was brought over to the East by a migrating individual, being then inherited and further integrated into a set of jewellery belonging to a woman from the community from Fântânele, who belonged to a later generation and was buried in the second half of the 3rd century BC (Rustoiu and Megaw 2011). The stamped vessels from the cemetery at Fântânele appear both in graves containing weaponry (graves no. 17, 28 and 41) and in others containing feminine costume sets (graves no. 40 and 42). As a consequence, there is no connection between this type of decoration and a particular social or gender category. The same situation has been noted in the cemetery from Ludas–Varjú-dűlő, in which the stamped vessels are equally present in graves with or without weaponry. On the other hand at Mannersdorf this type of pottery appears in only two burials containing weaponry (from a total of 11 graves including vessels stamped with concentric circles in their inventory), whereas at Pişcolt only one grave with weaponry (from a total number of 10) contains such finds. These characteristics may suggest that the attitude towards these motifs and their meanings was different from one community to another, even if they are remarkably similar from the morphological point of view. Bringing into discussion the functionality of the vessels with stamped decoration, it has to be noted that mostly the typical tableware of the Carpathian Basin (Németi 1988a) is ornamented in this manner. The mentioned motifs are present on liquid storage and serving vessels, drinking vessels and open forms for eating or serving. It is important to note that different combinations of concentric circles were also stamped on vessels inspired by forms from the

19.  The stamped pottery from the La Tène cemetery at Fântânele-Dealul Popii

Fig. 19.4. Fântânele-Dealul Popii. Vessels from graves no. 41 (1–2) and 42 (3–5).

165

166

Aurel Rustoiu

eastern Mediterranean region. For example they are present on some Danubian kantharoi (Zvonimirovo, Balatongyörök, Kosd, Szombathely, Zalakomár – Fig. 19.5/1: MajnarićPandžić 2001, 91, Pl. 5; Dizdar 2011, 115, Fig. 19.4/1; Kruta and Szabó 1982, Pl. 2–3; Horváth 2008, 113, Fig. 19.2–7; Rustoiu and Egri 2011, Fig. 19.6/1, 15/6, 23/2) or on an oenochoe from the cemetery at Ludas–Varjú-dűlő (Fig. 19.5/2), which copies in a local manner a Greek ceramic or metal vessel (Szabó 2012, 82, no. 13, 135, Pl. 50/11). At the same time, similar decorative motifs are also present on a series of ceramic objects having different functions than those mentioned above. One example is the fragmentary zoomorphic figurine (Fig. 19.5/4) discovered in a pottery workshop from Biharea (Bihor County), in Romania (Dumitraşcu 1982, 165, Fig. 19.6/2). Lastly, some concentric circles were also used in combinations of motifs decorating some boot-shaped vessels which are present in a few cemeteries from the Carpathian Basin, for example at Mannersdorf (Ramsl 2011, 182–183, Pl. 36). The vessels from Fântânele-Dealul Popii belong to some types which are frequently found in cemeteries and settlements from the Great Hungarian Plain and Transylvania (Németi 1988a; Szabó 2007, 243, fig. 49; Szabó et al. 2008, 217). The large bi-truncated vessels (no. 3, 5 and 7), the pitchers (no. 1 and 4) and the carinated bowl (no. 6), are all typical La Tène forms. On the other hand the bi-truncated beakers (no. 2 and 8) are mainly specific to the eastern part of the Carpathian Basin. They were inspired by forms coming from the Greek Pontic region and from the northern Balkans, being morphologically adapted to the ‘Celtic’ tableware set from the inner Carpathians region (Zirra 1975, 50, 53; 1978, 132; see the distribution map of the single-handled vessels in the Carpathian Basin in Teleagă 2008, 120, fig. 18). These vessels became popular in this region due to the earlier tradition (of the Early Iron Age) of using single-handled drinking vessels, which commonly replaced the handleless beakers of La Tène type, frequent in the tableware of Central Europe. At the same time, their distribution area overlaps to the west, in the Great Hungarian Plain and the middle Danube region, the area in which the so-called Danubian kantharoi, inspired by Greek forms, were used. The incorporation of the latter forms into the La Tène tableware was also facilitated by the prior existence of an ‘Illyrian’ tradition of using two-handled drinking vessels in southern Transdanubia and the basins of the Sava and Drava rivers (Rustoiu and Egri 2010; 2011). The two types of drinking vessels are illustrating two different manners of manipulating, serving and consuming beverages, preferred by various communities from the aforementioned regions. In this context it has to be noted that some drinking vessels were important individualised items in certain situations. An example is the ceramic beaker recovered from the grave no. 107 from Pişcolt, which has a handle made of bronze sheet, replacing the broken original one; the vessel must have

remained in use up to the death of its owner, as a cherished belonging (Németi 1975, Pl. 3/11; 1992, 82, fig. 17/10). Concerning the style and repertoire of the decorative motifs, the vessels from Fântânele are decorated (with a single exception) with stamped concentric circles arranged in various combinations and sometimes united by incised arches. Such circles were impressed onto the vessels’ walls before firing, using a series of stamping tools made of animal bone or horn. A good example is the stamp made of horn (Fig. 19.5/3) discovered in the already mentioned pottery workshop from the settlement at Biharea (Dumitraşcu 1982, 157, Fig. 19.6/1), dated to the LT C1 and eventually up to the beginning of the LT C2 (Rustoiu and Egri 2011, 73; Megaw and Megaw 2006, 376–8, with a list of stamps for decorating pottery). In the case of the vessels no. 1 and 2 from grave no. 17 and of the vessels no. 5 and 6 from grave no. 41, only one stamping tool was used, indicating that the containers were made at the same time and in the same workshop. This hypothesis is also supported by the similar firing and composition of the fabric of all these vessels. Summarising these observations, it might be possible that the tableware placed into the graves was purposefully ordered for funerary rituals, an idea which was also suggested many years ago by I. Németi in his analysis of the cemetery from Pişcolt (Németi 1988a, 109). The remaining vessels were each decorated with different stamps, one type of tool being used for each vessel. Vessel no. 3 from grave no. 28 was decorated with concentric circles and rosettes having a wheel-like shape, with four spokes. Such rosettes, but with more spokes, are also encountered on the indigenous pottery from the Balkans during the 4th – 3rd centuries BC (Fig. 19.5/5–6) (Čičikova 1984, fig. 13, 32–3, Pl. 8, 10–11 etc.; Moscalu 1983, Pl. 67–9; Alexandrescu 1980, 35, 48, fig. 39/8), and reappeared later in Dacia, on vessels from the Kingdom period of the 1st century BC (Crişan 1969, 210–11, Pl. 111/1, 4). On the basis of this pattern, V. Zirra presumed that rosettes appeared on Celtic pottery from Transylvania due to contacts with the Thracian communities from the Balkans (Zirra 1975, 54), albeit later he abandoned this hypothesis (Zirra 1978, 131). The rosette-shaped stamps were rarely used on La Tène ceramic by comparison with the concentric circle motifs. For example, rosettes were stamped on a vessel from Dipşa (Fig. 19.6/3) (Zirra 1975, 54, 19, Fig. 19.3/1) and on another one from Orosfaia (Vaida 2000, 135, Fig. 19.2/3; 2006, 303, fig. 15/6). The latter cemetery was wrongly located in older literature in the nearby village at Comlod (Dănilă 1971; Zirra 1978, 129–30). The examples cited come from cemeteries located in the same region as the one from Fântânele. The rosette-shaped stamps (more or less closely resembling those from Fântânele) are also known from Hungary, on a vessel from Szob with another type of rosette of triskele type (Jerem 1974–1975, 52, Pl. 15/2), as

19.  The stamped pottery from the La Tène cemetery at Fântânele-Dealul Popii

167

Fig. 19.5. 1. Details of the stamped decoration on the kantharos from Zalakomár (after Horváth 2008). 2. Oenochoe from Ludas – Varjúdűlő (after Szabó 2012). 3–4. Stamp made of horn (3) and ceramic zoomorphic figurine (4) from the pottery workshop from Biharea (after Dumitraşcu 1982). 5–6. Pithoi from Seuthopolis (5) and Zimnicea (6) stamped with rosettes (after Čičikova 1984 and Moscalu 1983). Different scales.

168

Aurel Rustoiu

Fig. 19.6. Vessels stamped with rosettes from Schwebheim (1), Sopron-Krautacker (2), Dipşa (3) and Pottenbrunn (4) (after Schwappach 1975, Jerem 1984, Zirra 1975 and Ramsl 2002). 5–6. Bowls from Mannersdorf (after Ramsl 2011). Different scales.

19.  The stamped pottery from the La Tène cemetery at Fântânele-Dealul Popii

169

Fig. 19.7. Map with the main sites mentioned in text. 1. Fântânele – Dealul Popii; 2. Biharea; 3. Dipşa; 4. Dvory nad Žitavou; 5. KaleKrševica; 6. Ludas – Varjú-dűlő; 7. Malé Kosihy; 8. Mannersdorf; 9. Nitra-Šindolka; 10. Orosfaia; 11. Pişcolt; 12. Pottenbrunn; 13. Remetea Mare – Gomila lui Pituţ; 14. Schwebheim; 15. Seuthopolis; 16. Sopron; 17. Zimnicea; 18. Zvonimirovo.

well as from south-western Slovakia, at Dvory nad Žitavou (Benadik, Vlček and Ambros 1957, 78, 137, Pl. 27/7) and Malé Kosihy (Bujna 1995, Pl. 14/1, 55/7) or Lower Austria, at Pottenbrunn (Fig. 19.6/4) (Ramsl 2002, 107, Pl. 44/12). Similar rosettes also appear westward in Bavaria, at Schwebheim (Fig. 19.6/1) (Schwappach 1975, Pl. 2/3). However, the closest analogy for the decorative motif from Fântânele appears on a vessel from the pottery workshop at Sopron-Krautacker (Fig. 19.6/2), which functioned in the 4th – 3rd centuries BC (Jerem 1984, 60, Fig. 19.4/5). At the same time, the stamped rosettes on two vessels from the settlement at Nitra-Šindolka, in south-western Slovakia

(Březinová 2000, Pl. 43/5, 141/10; 2002, Fig. 19.2/2–3) are stylistically quite close. Lastly, some rather similar rosettes appear on a series of metal artefacts, for example those decorating the chape of the scabbards of some swords from Hatvan-Boldog, Kosd etc. (Szabó and Petres 1992, Pl. 18, 29, 31 etc.), or the countermarks on Celtic coins, like those from the hoard discovered at Chişineu-Criş (Arad County), in western Romania (Dembski 1994, Pl. 9/13, 19–21). In the same way in which a series of ornaments belonging to the Vegetal Style, usually represented on metal objects, were transferred onto pottery by craftsmen having a variable degree of technological and artistic ability (Megaw 1970,

170

Aurel Rustoiu

92–3, no. 115; Duval 1974; Szabó and Petres 1992, 53–4), some stamped decorative motifs specific to the metal art were also used to ornament the ceramic tableware, pointing to the possible existence of a common symbolic repertoire. The manner in which the groups of stamped motifs were composed is quite unitary. On beakers the concentric circles are displayed either in pairs (vessel no. 2) or in triangles (vessel no. 8). On larger vessels, either four concentric circles or rosettes were united with arches, or groups of four or five stamped elements were included into an incised circle: sometimes these motifs were combined, like on vessel no. 7. In all these cases the groups of stamped motifs were symmetrically placed on the vessels, three times. Vessel no. 1 is an exception in terms of the composition and the placing of ornamental groups on its surface. Thus the four stamped circles make a T-shaped motif, each group being symmetrically repeated six times. The same composition, as well as the same decorative rhythm, is encountered on a series of early LT bowls belonging to the so-called ‘Braubach’ type. Amongst the examples can be listed the vessels from Győr-Újszálás and Kosd (Hunyady 1942, Pl. 59/1, 4, 60/1), as well as the finds from graves no. 10 and 114 from Mannersdorf (Fig. 19.6/5–6), dated to the LT B1 (Ramsl 2011, Pl. 46/1, 125/6). Thus the vessel from Fântânele is again an expression of the rememoration of some earlier traditions from the ‘colonists’’ region of origin, who moved eastward at the end of the LT B1 and the beginning of the LT B2, through the transferring of some ancestral symbols on vessels used in funerary ceremonies which were performed a few generations later. In conclusion, the stamped pottery from the cemetery of Fântânele-Dealul Popii illustrates a series of particular aspects regarding the use of such ornaments in the eastern Celtic world. These aspects were determined by the cultural evolution of the La Tène communities which ‘colonised’ the Carpathian Basin in successive phases during the 4th and 3rd centuries BC (Rustoiu forthcoming). M. Szabó has already noted that in the Carpathian Basin, La Tène culture was characterised by a synthesis of Central and Western European elements and of many components belonging to the local communities. To this core part were added several cultural ‘borrowings’ from the eastern Mediterranean area and the northern Balkans, which were filtered and interpreted in a particular manner. The result of this ethnic and cultural synthesis was the formation of a veritable koiné of the eastern Celts (Szabó 1971; 1992; 2006 etc.). The stamped pottery from Fântânele, as well as many aspects from the same cemetery, illustrates the same cultural amalgamation. The decorative motifs, consisting mainly of stamped concentric circles and rarely of rosettes, were usually placed on typical vessels belonging to the Central European La Tène tableware set. The local dining sets also include

forms originating from the earlier local repertoire, for example beakers and mugs, but which were adapted to the consumption style of the newcomers. Chronologically, all of the stamped vessels from Fântânele come from graves dated to the LT C1. The preliminary statistic analysis of some cemeteries from the Carpathian Basin seems to suggest a resurgence of decorative motifs consisting of concentric circles precisely in this phase of the Late Iron Age, at least in some communities. More than that, in certain cases the organization of the motifs and the ornamentation rhythm illustrate the resurgence of some earlier Central European traditions belonging to the beginning of the La Tène period. This pattern might be interpreted as a rememoration of some cultural traditions from the region of origin of the communities which ‘colonized’ the eastern part of the Carpathian Basin. On the other hand, some decorative motifs – for example the rosettes – were inspired by products from the eastern Mediterranean region. The vessel stamped with rosettes and palmettes, discovered in grave no. 89 from Pottenbrunn (Fig. 19.6/4) and dated to the LT B1, is a good example (Ramsl 2002, 36, 107, 110, fig. 125, Pl. 44/12). In the western part of the Carpathian Basin the connections with Italy were established early along the famous Amber Route, facilitating such exchanges. At the same time, the communities from the central and south-eastern parts of the Carpathian Basin were connected to the eastern Mediterranean area soon after the colonization of these regions in the second half of the 4th century and the beginning of the 3rd century BC. Macedonia played a major role in the northward distribution of Greek products, using the route along the Vardar-Morava corridor (Rolley 2002, 49). A series of local commercial centres, like the one recently investigated at Kale-Krševica, in the upper basin of the Južna Morava, Serbia (Popović 2005; 2006 etc.), also played an important role in the distribution of such products. These contacts directly established or filtered through the Illyrian or Thracian populations from the northern Balkans, determined the appearance of some objects inspired from Greek forms in the La Tène environment from the Carpathian Basin, but adapted in a particular manner to the necessities of the local communities. For example, this was the case of the Danubian kantharoi (Rustoiu and Egri 2010; 2011). At least a part of the rosette-shaped stamped motifs which appear on pottery from the Carpathian Basin might have the same source of inspiration. Thus, in the Carpathian Basin East meets West in a wide variety of manners and means and the resulting synthesis displays several original characteristics. The stamped pottery from Fântânele illustrates in the local manner the amalgamation of the symbolic languages specific to a community living on the frontier between the East and the West.

19.  The stamped pottery from the La Tène cemetery at Fântânele-Dealul Popii

Bibliography Alexandrescu, A. D. 1980. La nécropole Gète de Zimnicea, Dacia N.S. 24, 19–126. Anastassov, J. 2011. The Celtic presence in Thrace during the 3rd century BC in light of new archaeological data. In M. Guštin and M. Jevtić (eds.) The Eastern Celts. The Communities Between the Alps and the Black Sea, 227–239. Koper: Analles Mediterranei. Babeş, M. 1993. Die Poieneşti-Lukaševka-Kultur. Ein Beitrag zur Kulturgeschichte im Raum östlich der Karpaten in den Jahrhunderten vor Christi Geburt, Bonn: Verlag R. Habelt. Benadik, B., Vlček, E. and Ambros, C. 1957. Keltische Gräberfelder der Südwestslowakei, Bratislava: Slovenská Akadémia Vied. Březinová, G. 2000. Nitra-Šindolka. Siedlung aus der Latènezeit. Katalog, Archeologica Slovaca Monographiae 8, Nitra: Slovenská Akadémia Vied. Březinová, G. 2002. Výzdobné motívy na keramike z laténskeho sídliska v Nitre-Šindolke, Študijné zvesti archeologického ústavu SAV 35, 203–212. Bujna, J. 1995. Malé Kosihy. Latènezeitliches Gräberfeld. Katalog, Archeologica Slovaca Monographiae 7, Nitra: Slovenská Akadémia Vied. Čičikova, M. 1984. Antična keramika. In D. P. Dimitrov, M. Čičikova, A. Balkanska, L. Ognenova-Marinova, Sevtopolis, vol. 1. Vit i kultura, 18–114. Sofia: Bulgarian Academy of Science. Crişan, I. H. 1969. Ceramica daco-getică. Cu specială privire la Transilvania, Bucureşti: Editura Ştiinţifică. Crişan, I. H. 1975. La nécropole de Fîntînele et son importance pour le problème des Celtes de l’Europe Centrale. In J. Fitz (ed.) The Celts in Central Europe, 185–186. Székesfehérvár: Az István Király Múzeum. Crişan, I. H. 1977. Începutul Latène-ului la daco-geţi, Marisia 7, 67–84. Dănilă, Ş. 1971. Noi descoperiri arheologice privind problema celţilor din zona Bistriţei, File de Istorie 1, 55–71. Dănilă, Ş. 1978. Primele săpături arheologice în necropola de epocă La Tène de la Fîntînele, Studii şi Cercetări de Istorie Veche şi Arheologie 29.2, 257–275. Dembski, G. 1994. Die keltischen Münzen aus dem Schatzfund von Chisineu-Cris, Römisches Österreich 19–20, 83–94. Dizdar, M. 2011. The La Tène culture in central Croatia. The problem of the eastern border of the Taurisci in the Podravina region. In M. Guštin and M. Jevtić (eds.) The Eastern Celts. The Communities Between the Alps and the Black Sea, 99–118. Koper: Analles Mediterranei. Dumitraşcu, S. 1982. Les fours de poterie découverts à Biharea. Dacia N. S. 26, 157–166. Duval, P.-M. 1974. Le décor du vase celtique de KálozNagyhörcsök, Acta Archaeologica Academiae Scientiarum Hungaricae 26.1–2, 105–112. Eibner, A. 1996. Die Bedeutung der Basarabi-Kultur in der Entwiklung des Osthallsattkreis. In M. Garašanin and P. Roman (eds.) Der Basarabi-Komplex in Mittel- und Südösteuropa, 105–118. Bukarest: Vavila Edinf. Horváth, L. 2008. Kelta pszeudo-kantharos Zalakomárból (Zala m.), Zalai Múzeum 17, 109–129. Hunyady, I. 1942. Die Kelten im Karpatenbecken, Dissertationes

171

Pannonicae II.18, Budapest: A Királyi Magyar Pázmány Péter Tudományetem Éremés Régiségtani Intézete. Jerem, E. 1974–1975. Stempelverziertes frühlatènezeitliches Gefäss aus Écs, Mitteilungen des Archäologischen Instituts der Ungarischen Akademie der Wissenschaften 5, 45–57. Jerem, E. 1984. An early Celtic pottery workshop in north western Hungary: some archaeological and technological evidence, Oxford Journal of Archaeology 3.1, 57–80. Kruta, V. and Szabó, M. 1982. Canthares danubiens du IIIe siècle avant notre ère. Une exemple d’influence hellénistique sur les Celtes orientaux, Études celtiques 19, 51–67. Lazarov, L. 2010. The Celtic Tylite state in the time of Cavarus. In L. F. Vagalinski (ed.) In search of Celtic Tylis in Thrace (III c BC), 97–113. Sofia: National Archaeological Institute and Museum. Majnarić-Pandžić, N. 2001. Grob ratnika LT 12 iz srednjolatenskog groblja u Zvonimirovu kod Suhopolja u Virovitičko-podravskoj županiji, Prilozi. Institut za arheologiju u Zagrebu 18, 83–101. Medeleţ, F. ms. Necropola La Tène de la Remetea Mare (jud. Timiş), manuscript. Megaw, J. V. S. 1970. Art of the European Iron Age. A study of the elusive image, Bath: Adams and Dart. Megaw, R and Megaw, V. 2001. Celtic art. From its beginnings to the book of Kells, 2nd edition, London: Thames and Hudson. Megaw, J. V. S. and Megaw, M. R. 2006. Strike the lyre: notes on an eastern Celtic motif, Acta Archaeologica Academiae Scientiarum Hungaricae 57, 367–393. Moscalu, E. 1983. Ceramica traco-getică, Bucureşti: Muzeul National de Istorie. Németi, I. 1975. Contributions concernant le facies latènien du Nord-Ouest de la Roumanie a la lumière de dècouvertes celtiques de Piscolt (dép. de Satu Mare). In J. Fitz (ed.) The Celts in Central Europe, 187–197. Székesfehérvár: Az István Király Múzeum. Németi, I. 1988a. Unele aspecte ale evoluţiei ceramicii din a doua epocă a fierului în nord-vestul României (Latène B-C), Studii şi Cercetări de Istorie Veche şi Arheologie 39.2, 87–111. Németi, I. 1988b. Necropola Latène de la Pişcolt, jud. Satu Mare. I, Thraco-Dacica 9, 49–74. Németi, I. 1989. Necropola Latène de la Pişcolt, jud. Satu Mare. II, Thraco-Dacica 10, 75–114. Németi, I. 1992. Necropola Latène de la Pişcolt, jud. Satu Mare. III, Thraco-Dacica 13, 59–112. Popović, P. 2005. Kale-Krševica: investigations 2001–2004. Interim report, Zbornik Narodnog muzeja 18.1, 141–174. Popović, P. 2006. Central Balkans between the Greek and Celtic world: case study Kale-Krševica. In N. Tasić and C. Grozdanov (eds.) Homage to Milutin Garašanin, 523–536. Belgrade: Serbian Academy of Sciences and Arts. Ramsl, P. C. 2002. Das eisenzeitliche Gräberfeld von Pottenbrunn, Wien: Verlag Ferdinand Berger und Söhne. Ramsl, P. C. 2011. Das latènezeitliche Gräberfeld von Mannersdorf am Leithagebirge, Flur Reinthal Süd, Niederösterreich, Mitteilungen der Prähistorischen Kommission 74, Wien: Verlag der Österreichischen Akademie der Wissenschaften. Rolley, C. 2002. Le travail du bronze à Delphes, Bulletin de correspondance hellénique 126.1, 41–54. Rustoiu, A. 2008a. Războinici şi societate în aria celtică

172

Aurel Rustoiu

transilvăneană. Studii pe marginea mormântului cu coif de la Ciumeşti, Interferenţe entice şi culturale în milenile I a. Chr.- I p. Chr. 13, Cluj-Napoca, Editura Mega. Rustoiu, A. 2008b. Celţii din Transilvania şi comunităţile indigene nord-balcanice. Schimburi culturale şi mobilitate individuală, Ephemeris Napocensis 18, 25–44. Rustoiu, A. 2009. Masters of metals in the Carpathian Basin (workshops, production centres and funerary manifestations in the early and middle La Tène), Ephemeris Napocensis 19, 7–21. Rustoiu, A. 2011. The Celts from Transylvania and eastern Banat and their southern neighbours. Cultural exchanges and individual mobility. In M. Guštin and M. Jevtić (eds.) The Eastern Celts. The Communities Between the Alps and the Black Sea, 163–170. Koper: Analles Mediterranei. Rustoiu, A. forthcoming. Indigenous and colonist communities in the Eastern Carpathian Basin at the beginning of the Late Iron Age. The genesis of an Eastern Celtic World. In C. N. Popa and S. Stoddart (eds.) Fingerprinting the Iron Age. Approaches to identity in the European Iron Age. Integrating South-Eastern Europe into the debate, Oxford, forthcoming. Rustoiu, A. and Egri, M. 2010. Danubian Kantharoi – Almost three decades later. In S. Berecki (ed.) Iron Age Communities in the Carpathian Basin, Proceedings of the International Colloquium from Târgu Mureş, 9–11 October 2009, 217–287. Cluj-Napoca: Editura Mega. Rustoiu, A. and Egri, M. 2011. The Celts from the Carpathian Basin between continental traditions and the fascination of the Mediterranean. A study of the Danubian kantharoi, ClujNapoca: Editura Mega. Rustoiu, A. and Megaw, J. V. S. 2011. A foreign flowering in Transylvania: the Vegetal style armring from Fântânele- Dealul Popii, jud. Bistriţa-Năsăud, grave 62. In D. Măgureanu, D. Măndescu and S. Matei (eds.) Archaeology: making of and practice. Studies in honor of Mircea Babeş at his 70th anniversary, 217–237. Piteşti: Editura Ordessos. Schwappach, F. 1973. Frühkeltisches Ornament zwischen Marne, Rhein und Moldau, Bonner Jahrbücher 173, 53–111. Schwappach, F. 1975. Zur Chronologie der östlichen FrühlatèneKeramik. In J. Fitz (ed.) The Celts in Central Europe, 109–136. Székesfehérvár: Az István Király Múzeum. Schwappach, F. 1977. Die stempelverzierte Latène-Keramik aus den Gräbern von Braubach, Bonner Jahrbücher 177, 119–183.

Szabó, M. 1971. Sur les traces des Celtes en Hongrie, Budapest: Éditions Corvina. Szabó, M. 1992. Les Celtes de l’Est. Le Second Age du Fèr dans la cuvette des Karpates, Paris: Éditions Errance. Szabó, M. 2006. Les Celtes de l’Est. In M. Szabó (ed.) Celtes et Gaulois. L’Archéologie face à L’Histoire. Les Civilisés et les Barbares du Ve au IIe siècle avant J.-C. Actes de la table ronde de Budapest 17–18 juin 2005, Collection Bibracte 12. 3, 97–117. Glux-en-Glenne: Centre archéologique européen du Mont Beuvray. Szabó, M. (ed.) 2007. L’habitat de l’époque de La Tène à Sajópetri Hosszú-dűlő, Budapest: L’Harmattan. Szabó, M. (ed.) 2012. La nécropole celtique à Ludas-Varjú-Dűlő, Budapest: L’Harmattan. Szabó, M. and Petres, É. F. 1992. Decorated weapons of the La Tène Iron Age in the Carpathian Basin, Budapest: Magyar Nemzeti Múzeum. Szabó, M. et al. 2008. Polgár 1: l’habitat du Second Âge du Fer (IIIe siècle av. J.-Chr.), Acta Archaeologica Academiae Scientiarum Hungaricae 59, 183–223. Teleagă, E. 2008. Die La-Tène-zeitlichen Nekropole von Curtuiuşeni/ Érkörtvélyes (Bihor, Rumänien). Der Forschungsstand, Dacia N.S. 52, 85–165. Vagalinski, L. F. 2007. Celtic pottery in northern Bulgaria. In L. F. Vagalinski (ed.) The Lower Danube in Antiquity (VI c. BC – VI c. AD), 73–82. Sofia: National Archaeological Institute and Museum. Vaida, L. 2000. The Celtic cemetery from Orosfaia (BistriţaNăsăud County). In C. Gaiu and A. Rustoiu (eds.) Les Celtes et les Thraco-Daces de l’Est du bassin des Carpates, 135–159. Cluj-Napoca: Editura Accent. Vaida, D. L. 2006. Celtic finds in North-East Transylvania (IVth – IInd Centuries B.C.). In V. Sîrbu and D. L. Vaida (eds.) Thracians and Celts. Proceedings of the International Colloquium from Bistriţa, 18–20 May 2006, 295–322. ClujNapoca: Editura Mega. Zirra, V. 1975. Influences des Gèto-Daces et de leurs voisins sur l’habitat celtique de Transylvanie. In J. Fitz (ed.) The Celts in Central Europe, 47–64. Székesfehérvár: Az István Király Múzeum. Zirra, V. 1978. The decorated Celtic pottery of Transylvania, Dacia N. S. 22, 125–141.

20 A VESSEL WITH STAMPED DECORATION FROM THE ŽELIEZOVCE COLLECTION Gertrúda Březinová1

Introduction During the preparation of my article on Celtic glass in 2012, and searching the catalogue of the collection in the Archaeological Institute of the Slovak National Museum in Bratislava, I found an entry regarding the reception of some finds from the archaeological collection in Želiezovce. Fragments of a vessel with registration number AP 2543, decorated by a stamp, were listed. The stated place of the find is an unknown site. In a reconstructed form, the vessel has already been published as part of the collection’s processing (Pichlerová and Tomčíková 1995, 125, tab. XV.1), in which, however, adequate attention was not paid to the decoration, and there were also differences in metrical data if compared with the original. The fact that it is an exceptional find with interesting decorative elements was a reason why I have decided to draw attention to it again. I would like to express my gratitude for a high quality drawing documentation to Radoslav Čambal.

Description of the vessel A vase-shaped vessel (Fig. 20.1) with low ‘S’-shaped profiled body of dark-grey colour. The neck is bent out, rim slightly strengthened. The lower part of the vessel and its base have not been preserved. The diameter of rim, measured from the original sherd (as a large part has been remodelled in plaster) is 14.4 cm, maximum diameter of the body being 18.6 cm. The maximum preserved height is 9.2 cm. The vessel was wheel thrown. The body is decorated by an oblong stamp, composed of a spirally coiled motif in a shape of the letter ‘S’. A larger stamp is placed along the neck’s girth, right over an engraved line which separates the vessel’s neck and body. Individual stamps are almost in a row, though the creator did not manage to place them

precisely and did not observe the rule of a unified distance either. Right under the sunken girth line one can find another stripe of oblong stamps of a spirally coiled motif in the shape of the letter ‘S’. The stamp is smaller and the matrix itself contains two subsequently placed horizontal letters ‘S’. The form and realisation of the motif are identical. This row of stamps almost exactly follows the girth line. In one case the oblong stamp is placed as a separating element perpendicularly to the girth line. On the opposite side of the vessel this vertical placement of stamp may only be assumed. The original of this part has not been preserved (Fig. 20.1.1a and 20.1.1b).

Stamp dimensions Stamp no. 1 placed above the girth line: width: 1.6 cm, height: 1.1 cm, width of spiral in length: 1.46 cm, height of spiral: 0.8 cm (Fig. 20.1.2a). Stamp no. 2 placed under the girth line: width: 1.8 cm, height: 0.47 cm, width of spirals in length: 1.74 cm, height of spiral: 0.42 cm (Fig. 20.1.2b).

Analysis of the decoration Generally, the decorative motif on the vessel belongs to the motif of a stamped simple ‘S’ letter, potentially lyre-shaped. It was made by a stamp and is characteristic for the La Tène Period. It occurs on pottery, artistic as well as utility objects. The regional occurrence and evaluation of the symbolism of stamps with S-motifs is summarised in the work of coauthors J. V. S. and M. R. Megaw (2006). The map of the spread of pottery finds, decorated by the mentioned motif, shows that they are concentrated in the western part of the

174

Gertrúda Březinová

Fig. 20.1. 1.1a and b: drawing reconstruction of the vessel with decoration – scale and 50% of the original’s size: drawing by R. Čambal. 1.2a and 1.2b: photographs of the decoration (gauge b – real size of stamp).

Carpathian Valley, with the dominance of the lyre motif. The authors consider the decoration, occurring on the pottery, to be typical for eastern Celts (Megaw and Megaw 2006, fig. 1). To the last list were added other finds from Hungary (Tankó 2010). As far as Slovakia is concerned, it is the Košice-Krásno site (Mirrošayová 2012). Here there is a vase with a high-placed maximum curve decorated by a stamp with the ‘S’-motif. Vertical imprints of the oblong stamp are arranged into two circular lines. The ends of the first one are superimposed, while the other one is continuous. The stamped decoration is not done precisely, probably because of the fact that it was mortuary pottery. The use of stamps with simple ‘S’-motifs seems not to

have been restricted by any rule. On pottery it occurs in various combinations and positions. The dating also falls within a more extensive time period ranging from the LTB2 stage through to LTC2/D1 (Megaw and Megaw 2006). As far as the decoration of the vessel from the Želiezovce collection is concerned, there are notable differences in the execution of the simple S motif, with the identified and very precisely made coiling into spirals. It is repeated on a large as well as small matrix. However, the motif is strikingly similar to the plastic decoration on the bracelet from grave no. 31 of the Brno-Maloměřice site (Čižmářová 2004, 151; 2005), the one on the brooch from the inhumation grave dated to LT B–C1 of Dobročkovice 2 site (Čižmářová 2012, tab. I.1),

20.  A vessel with stamped decoration from the Želiezovce collection

175

Fig. 20.2. Slovakia. The Ipeľ River Area. Probable places of the discovery of the stamped vessel.

Fig. 20.3. La Tène sites in the vicinity of Želiezovce. 1 – Hronovce-Domaša, 2 – Pohronský Ruskov, 3 – Želiezovce.

as well as on other small features belonging to the group of pseudo-filigree jewels. Its occurrence in Moravia was discussed by J. Čižmářová (2012). She dates the beginning of the jewel to LT B2 stage. Its numerous occurrences can

be detected in LT C1 as well. At the close of LT C2 and in the LT D stage it is rather exceptional. To what extent it is possible to claim that the maker of the observed vessel was influenced exactly by the motif from this kind of jewel is

176

Gertrúda Březinová

very questionable. So far there has been no other similar, or equal, application of this decoration on pottery.

Dating the vessel The vase-shaped vessel belongs to the pottery types occurring in La Tène settlements as well as gravesites in the LT B2 up to LT C stages.

Conclusion The vessel with stamped decoration most probably comes from one of the mentioned sites (Fig. 20.3). It is hypothetically possible, that it is from the Domaša gravesite itself. An exact identification of the find’s position remains open. However, other sites, detected and verified by newer activities, could be considered as well. The execution of the decoration into a spirally coiled letter ‘S’ opens possibilities for further interpretation of this decorative motif. Translated by Anton Pokrivčák

Discussion of the place of discovery At the beginning, one has to ask about the regional context that the features making up the Želiezovce collection come from – whether there are La Tène sites around Želiezovce, and which of them could eventually be considered (taking into account temporal aspects) as a place of discovery. The beginnings of this collection are associated with the acquisition of archaeological finds by Mrs. Kherndlova, a wife of the custodian of Želiezovce country estate, which was divided after 1903. One part of it was salvaged by Countess Coudenhove, who expanded it and made possible its scientific exploration. The collection was dominated by archaeological finds from the Neolithic,  Eneolithic and Bronze Age from over 20 sites in the vicinity of Želiezovce. Its detailed inventory was prepared and evaluated by Austrian archaeologists H. Mitscha-Märheim and R. Pittioni (1934, 147ff.). The collection was placed in the Želiezovce mansion, and after WW II – following the departure of its last owner, Countess Coudenhove, and upon its confiscation – it was transferred to the Archaeological Section of the Slovak National Museum in Bratislava, where it was processed in detail by M. Pichlerová and K. Tomčíková (1993–1997). The finds from the Domaša-Hronovce gravesite were re-evaluated and complemented by G. Brezňanová and M. Furman (2012). The collection itself includes mainly features from the closest vicinity of Želiezovce (Fig. 20.2). Some of them come from the neighbouring territory of Hungary, as well. For example, from the La Tenè Period comes a vase-shaped bowl and jug with handle from Békésgyula. As for the features dated to the La Tène Period, the authors list entire vessels. Three of them are from Domaša-Hronovce gravesite (Pichlerová and Tomčíková 1994, 93, tab. II.1–3). Other two vessels are from Pohronský Ruskov (Pichlerová and Tomčíková 1994, 93, tab.V.1 and 3). La Tène pottery has been identified in Želiezovce itself as well. This includes a bowl (Pichlerová and Tomčíková 1995, fig. 5.1, tab. XIV.3) from the Mikula area, a vessel fragment from the Kerekudvar area (Pichlerová and Tomčíková 1995, tab. XIII.5), and a thick-walled small situla-formed vessel (tab. XIV.4) from an unknown location.

Note 1

The research concerned was carried out within the project reg.-no. 2/0117/12 of the VEGA Grant agency.

Bibliography Brezňanová, G. and Furman, M. 2012. Laténske nálezy v zbierkach Slovenského banského múzea v Banskej Štiavnici. In G. Březinová and V. Varsik (eds.) Archeológia na prahu histórie: K životnému jubileu Karola Pietu, 59–83. Nitra: Archeologický ústav SAV. Čižmářová, J. 2004. Encyklopedie Keltů na Moravě a ve Slezsku. Praha: Nakl. Libri. Čižmářová, J. 2005. Keltské pohřebiště v Brně-Maloměřicích, Pravěk-Supplementum 14. Brno: Ústav archeologické památkové péče. Čižmářová, J. 2012. Filigránový šperk doby laténské na Moravě. In G. Březinová and V. Varsik (eds.) Archeológia na prahu histórie: K životnému jubileu Karola Pietu, 83–94. Nitra: Archeologický ústav SAV. Megaw, J. V. S and Megaw, M. R. 2006. Strike the lyre: Notes on an eastern Celtic motif. Acta Arch. Acad. Scien. Hungaricae 57, 367–93. Mirrošayová, E. 2012. Keltské žiarové hroby z Košíc-Krásnej. In G. Březinová and V. Varsik (eds.) Archeológia na prahu histórie: K životnému jubileu Karola Pietu, 349–62. Nitra: Archeologický ústav SAV. Mitcha-Märheim, H. and Pittioni, R. 1934. Zur Besiedlungsgeschichte des unteren Grantales, Mitteilungen der Anthropologischen Gesellschaft Wien 64, 147–73. Pichlerová, M. and Tomčíková, K. 1993. Archeologická zbierka zo Želiezoviec I, ZSNM 87 – Archeológia 3, 53–90. Pichlerová, M. and Tomčíková, K. 1994. Archeologická zbierka zo Želiezoviec II, ZSNM 88 – Archeológia 4, 85–128. Pichlerová, M. and Tomčíková, K. 1995. Archeologická zbierka zo Želiezoviec III, ZSNM 89 – Archeológia 5, 91–127. Pichlerová, M. and Tomčíková, K. 1996. Archeologická zbierka zo Želiezoviec IV, ZSNM 90 – Archeológia 6, 149–81. Pichlerová, M. and Tomčíková, K. 1997. Archeologické zbierka zo Želiezoviec V, ZSNM 91 – Archeológia 7, 113–31. Tankó, K. 2010. Late Iron Age Settlement in the Vicinity of Menfőcsanak (Road no. 83 and Bevasarlokozpont). Studia celtica classica et romana Nicolae Szabo Septuagesimo Dedicata 249–60. Budapest: Pytheas.

21 BALKAN KANTHAROI Petar Popović

One of the frequent finds in the Balkans and in the Danube basin are rather small pottery vessels with two handles mentioned in the literature as ‘two-eared beakers’, twohandled beakers, pseudo-kantharoi and in more recent time mostly as kantharoi. This characteristic shape is as a rule associated with serving and consuming alcoholic drinks, hence its identification as two-handled drinking vessel. They were identified as Danubian kantharoi in the beginning of the 1980s when a considerable quantity of vessels of this very type were encountered in the territories of the Eastern Celts, particularly the Scordisci. Their models or inspiration were the Hellenistic shapes, but it turns out that the ceramic tradition of the autochthonous Illyrians and Pannonians played a significant rôle in the process of manufacture of these objects (Kruta and Szabó 1982; Szabó 1992, 155–157; Rustoiu and Egri 2010). It is considered, on the basis of more recent results and analyses of earlier finds, that the La Tène kantharoi of the Taurisci and Scordisci had been preceded by similar handmade shapes dating from the end of the Early Iron Age, i.e. from the 6th–4th centuries BC, and because of long tradition they could have continued until the Roman times (Dizdar 2010, 299–300). Nevertheless, if we follow this distinct morphology we could conclude that two-handled vessels, disregarding all typological distinctions, belong to one large family, which will lead us gradually into the very distant past. In other words, everyone who has had the opportunity to leaf through the plates in volumes of Praistorija jugoslavenskih zemalja (Prehistory of the Yugoslav Lands) could easily be convinced that vessels with two handles are encountered from the Eneolithic through the Bronze Age until the very end of the Iron Age. Accordingly, it is quite obvious that these shapes are concentrated in the central Balkans and in the Danube basin (Benac 1979; 1983; 1987). This is confirmed in many publications, which, unfortunately, are not always

available outside our country, and archaeologists are often more interested in solving chronological, cultural and ethnic problems and speak only in passing about the function and rôle of these objects. My main intention in this work is to present a short review of the appearance and disappearance of twohandled vessels, which are identified in accordance with the explanation above, as Balkan kantharoi. We selected only finds from the Danube basin and the central Balkans that illustrate some basic shapes produced over a rather long period of time (Fig. 21.3). They are taken from published works or documentation in some museums and we quoted only the main literature because there are just a few syntheses, although kantharoi have been mentioned in the countless publications throughout the Balkans. This text is our tribute in honour of Professor J. V. S. Megaw, who is well acquainted with the Balkans. I do not believe that he would be sorry that there are no pieces of art but I think that every potter did his best to make kantharos the object of beauty. Of considerable importance for our topic was the discovery of one isolated object at the site of LivadeKalenić in north-western Serbia. Archaeological excavations confirmed that there were no traces of settlement in the vicinity and that it was just one building consisting of a few rooms which perished in a conflagration, while heterogeneous material was discovered under the ruins. To the great surprise of archaeologists there was Eneolithic pottery characteristic of Pannonian, Danubian and central Balkan cultures within the single structure with totally heterogeneous material. A special place belongs to rather small vessels with two handles that are typologically characteristic of a quite large Balkan cultural complex (Fig. 21.1.1–2). The structure with its complete inventory generally dated to the 5th/4th millennium BC looked like

178

Petar Popović

the unique example of conviviality of inhabitants of diverse origin. The great mobility of these groups is explained as a consequence of the intensive quest for ores and the first use of the new metal – copper (Blagojević 2005; Jovanović 2005). Vessels from Kalenić that we could consider as the earliest two-handled beakers can be traced subsequently within a rather wide area (Fig. 21.1.3). Almost as a rule we encounter them in culturally different milieus or sometimes even as single finds (Tasić 1995, 32). However, disregarding all regional distinctions and thanks to many finds, the impression is that kantharoi remain a permanent feature of populations in the central Balkans from the Eneolithic until the first centuries AD. They would experience numerous transformations during this long period of time but they would maintain their basic shape and function to the very end. The kantharoi in the Early Bronze Age, in contrast to the preceding period, had their handles rising high above the rim, and they were distributed from the Adriatic to the Black Sea (Korkuti 2006, T. 1–2; Benac 1983, 463–470, sl. 33). A distinct area is the Danube valley and the central Balkans as far as northern Greece, where two-handled vessels were one of the most frequent shapes (Benac 1983, 705–759; Mitrevski 2006, fig. 1; Heurtley 1939). A wide variation, including simple shapes (Fig. 21.1.4–5) as well as ‘baroque’ forms and use of diverse decorative techniques and motifs, are considered the essential element in the typology and chronology of Bronze Age pottery (Hänsel 1976; Hoddinott 1981). Certain stylistic traits depending on local affinities are ascribed to different cultures or cultural groups, and because of the large quantity of finds it is not without reason that a part of former Yugoslavia, first of all the central Balkans, was identified as the ‘two-handled bowl complex’ or ‘kantharos culture’ (Coles and Harding 1979, 144). If, within such a busy area, the first great changes are related to the appearance of the Indo-Europeans, and the close relations between the Balkans and Asia Minor are later considered the consequences of migration movements of certain populations, including Bryges, Moesi or Dardanians (Petrova 2006), then these hypotheses could explain, at least to a certain extent, why the Balkan kantharoi spread over such a vast area. From the end of the Bronze Age and during the Early Iron Age, two-handled vessels (Fig. 21.1.6–8) spread further towards the western Balkans and the Alps, and thanks to connections between both Adriatic coasts they are often encountered in neighbouring Italy (Benac 1987; Lollini 1976). A similar situation is also found at the necropoles between the Adriatic and the Aegean where kantharoi are common finds in grave associations from Albania, Serbia, FYR Macedonia or Greece (Prendi 1976, T. VII–IX; Benac 1987, 575–733; Mitrevski 2006, fig. 1; Andronikos 1969, 211–213; Vokotopoulou 1986, 235).

From the Proto-Geometric period, and with the appearance of related but more refined vessels with handles, specialists in classic archaeology raised the question of the origin of kantharoi, which in Attic workshops reached the height of quality and elegance from the 6th century BC (Courbin 1953; Sparkes and Talkott 1970, 113–124, fig. 7). They are often depicted on vases and coins, particularly in Dionysiac scenes, and with new aesthetic features they could have belonged to a distinct category of products intended for distinguished customers and for special occasions (Franke and Marathaki 1999, 34). In this case they do not have direct analogies, but their origin should certainly be looked for in the rich heritage of the earlier periods. Owing to circumstances, they had a special place as pieces of art, but after lasting for a few thousand years they could represent only exceptional but ephemeral phenomenon. In general, it could be said that all vessels identified as kantharoi originate from the same cradle. In the 6th–5th centuries BC on the periphery of the antique world and in the south of the Balkans, a generation of kantharoi with technically and visually more modern solutions had been created from old and new elements based on shapes from the Early Iron Age and classic period (Fig. 21.1.8–9). Their appearance was going to play a significant role in establishing connections between the central Balkans and the Danube valley, especially with the Celts who lived in these areas from the end of the 4th century BC. From the final horizon of the Early Iron Age that preceeds these events come the graves of the Pannonian population. Vessels with a short lower cone and high handles (Fig. 21.1.10) were found together with Certosa fibulae in female burials at the necropolis attributed to the settlement at Feudvar in the south Bačka region (Medović 2007, 10–19). Similar kantharoi have been encountered at many sites from Pannonia to the Alps, but in the Alpine area they were found together with the early La Tène fibulae (Dizdar 2010, pl. 1–3; Teržan 2009, 91–92). In the meantime, thanks to recent archaeological excav­ ations in south Serbia, we obtained valuable data about the first contacts of these two culturally different areas. Namely, significant remains of urban settlement from the 4th and first decades of the 3rd century BC, built after Greek models, have been discovered at the Kale site in Krševica (kalekrsevica.com). The settlement consisted of an acropolis and large suburbium and also had a complex water-supply system. In addition to buildings of various type and size and many various finds, most of the discovered material is pottery. Smaller in proportion are luxurious goods from the Attic workshops (from St. Valentin vases to the early Hellenistic kantharoi), while, with the exception of coarse kitchenware, most are wheel-made vessels from a local workshop. It turned out that they were almost as a rule made after Greek shapes, while numerous variants of kantharoi were of older as well as of modern shapes taken over from

21. Balkan Kantharoi

179

Fig. 21.1. Eneolithic: 1–2 Livade-Kalenić (Blagojević 2005, sl. 18, 21); 3 Krivelj (Benac 1979, T. X. 6). Bronze Age: 4 Ljuljaci (Bogdanović 1986, sl. 27); 5 Paraćin (Benac 1983, T. CI. 5). Early Iron Age: 6 Vajuga-Pesak (Popović and Vukmanović 1998, fig. 28. 7); 7 Romaja (Benac 1987, T. LXIII. 3); 8–9 Radolište (National Museum in Belgrade, Popović 1958); 10 Stubarlija (Medović 2007, fig. 9); 11–13 Kale-Krševica (Popović et al. 2012, 35. 6, 2, 5). La Tène: 14 Pećine (Cerović 1983); 15–16 Kupinovo (Zagreb Archaeological Museum, Hunyady 1942, Pl. LXVI. 4, 8); 17 Dumača (Sladić 2002, sl. 4); 18–19 Krajčinovići (National Museum Užice); 20 Kale-Krševica (Popović 2009, fig 5.19); 21 Sirmium (Museum of Srem, Sremska Mitrovica). Roman: 22–23 Komini (Regional Museum Pljevlja); 24 Davidovac (National Museum Vranje).

the south (Fig. 21.1.11–13; Fig. 21.2). A large number of kantharoi and skyphoi have been discovered in the course of archaeological excavations, so considering also many amphorae from the north Aegean workshops it could be

concluded that at one time inhabitants of Krševica preferred wine to other alcoholic drinks (Popović et al. 2012). Celtic existence in the Balkans is recorded in antique sources from the end of the 4th century BC and now we

180

Petar Popović

Fig. 21.2. Kantharoi from Kale-Krševica (Popović et al. 2012, no. 53–54, 56–58).

even have some archaeological evidence to probably support that (Kruta 2000, 240–241). Namely, it is quite conspicuous that there is no essential difference between certain kantharoi from Krševica and specimens discovered in the earliest Celtic burials from Pećine, Belgrade-Karaburma or Kupinovo (Jovanović 1984; Todorović 1972; Majnarić-Pandžić 1970, 33). Therefore, it is logical to assume that newcomers from the Danube basin established, at this or some other similar place, relations with the Balkan inhabitants to whom Greek civilization was not unfamiliar. Thanks to these circumstances, certain groups of Eastern Celts take advantage of an opportunity and include in their new ceramic repertoire characteristic vessels from the south associated with wine (Fig. 21.1.14–17). The consequence was not only the adoption of kantharoi, but to a certain extent also the introduction of some Mediterranean customs as well. After the Celtic episode in 279 BC and the founding of the Scordisci, the footed kantharoi inspired by Hellenistic shapes have a prominent place in the areas around the Sava and the Danube (Fig. 21.1.14–17). The repertoire of these shapes is certainly best illustrated at the necropolis at Karaburma in Belgrade (Todorović 1972), and on the whole it is obvious that in the 3rd century the Scordisci were inevitable intermediaries between populations in the central Balkans and the Carpathian basin. As members of the great Celtic koine they take part in the exchange of goods on a large scale, and one example of these activities is sporadic finds of kantharoi of this type (Rustoiu and Egri 2010, 218–21, 247–48). However, in the Carpathian basin, vessels with one handle prevail (Rustoiu and Egri 2010, 222) and kantharoi are represented by a special type of Early Iron Age provenance without pronounced Hellenistic influences. These are massive biconical vessels with a flat base and two handles, which are sometimes of anthropomorphic or zoomorphic shape (Hunyady 1942, pl. LXV; Rustoiu and Egri 2010, pl. 2.2, 223–27).

Footed Kantharoi are more and more infrequent from the 2nd century BC and are usually encountered in the areas closer to the Mediterranean. One example is the find from Krajčinovići – a group grave from the 2nd century BC in south-western Serbia (Fig. 21.1.18–19). Besides Hellenistic material originating from the Montenegro coast, a considerable number of hybrid forms with characteristics of Mediterranean and autochthonous shapes, and with conspicuous La Tène influence were discovered (Zotović 2007; Keltoi 1984, sl. 16). Another example concerns Krševica. In the mid 2nd century, Scordisci and other Balkan barbarians used the abandoned acropolis as a stronghold in their plundering raids towards Macedonia and Greece. It turned out that the discovered ceramic material resembles finds from the Danube basin as well as the specimens originating from the south (Popović 2009). An example of such merging is a kantharos from Krševica with typical hybrid characteristics. It had the bowl with polished ornament characteristic of late La Tène pottery from the territory of Scordisci and the high foot typical of the Hellenistic period (Fig. 21.1.20). In the Late La Tène period, many workshops, particularly in the Sava valley, produced various types of biconical vessels with a flat base, high handles and often polished ornament (Fig. 21.1.21). Material from the settlement at Gomolava offers the best selection, but kantharoi with similar traits have been recorded over the vast area occupied at that time by the Scordisci – from eastern Slavonia to the Iron Gates and from Bačka and part of the Banat region to the Južna (South) Morava valley (Jovanović and Jovanović 1988; Popović 2001). They are also not rare in the neighbouring regions especially in Pannonia (Bónis 1969, 181). For the appearance of the wide repertoire of these Late La Tène shapes we should thank, to a considerable extent, the autochthonous populations, which had been deeply rooted in this area (Majnarić-Pandžić 2009). Thus, combining

21. Balkan Kantharoi

181

of significant urban settlement (municipium S.) from the 2nd/3rd centuries AD in the village Komini near Pljevja in the north of Montenegro (Cermanović-Kuzmanović 1998). Local inhabitants had been buried in a traditional way and kantharoi with characteristics of the Early Iron Age or Late La Tène have been found in some of the graves indicating a slow process of Romanization (Fig. 21.1.22–23). In the Južna (South) Morava valley further to the east, graves from the 3rd/4th centuries AD have recently been discovered where perhaps the last specimens of small two-handled vessels have been encountered (Fig. 21.1.24) (Fidanovski and Cvjetićanin 2005, T. III 4, 7; IV 6; V 4–5). Kantharos as a distinct shape of drinking vessel, dis­ regarding all cultural and ethnic changes, remained in use in the Balkans from the early prehistoric cultures until the Greek and Roman civilization. From the 4th century AD and with the new epoch they disappeared together with all the achievements of the antique period and were substituted with goblets or other vessels without handles. If something remained in the modern times from this long tradition these are certainly decorative objects or voluminous two-handled cups made of silver or crystal that are awarded to the winners of sport competitions.

Acknowledgments Fig. 21.3. Map: 1 Livade-Kalenić; 2 Krivelj; 3 Ljuljaci; 4 Paraćin; 5 Vajuga-Pesak; 6 Romaja; 7 Radolište; 8 Stubarlija; 9 Kale-Krševica; 10 Pećine; 11 Kupinovo; 12 Dumača; 13 Krajčinovići; 14 Sremska Mitrovica-Sirmium; 15 Komini; 16 Davidovac.

earlier tradition and advanced La Tène production, the kantharoi amongst other items became a segment of the common cultural heritage of the Scordisci. This conclusion could be associated with the so-called ‘revival of IllyrianPannonian kantharoi’ that is related to the population the Celts encountered on their arrival (Megaw and Megaw 1990, 149; Rustoiu and Egri 2010, 236). Romans occupied the areas around the Sava and the Danube by the end of the 1st century BC, but La Tène and latenoid pottery remained in use disregarding all new circumstances also during the early Roman times. At the location of a small Late La Tène settlement by the Sava River, Sirmium, one of most important Pannonian cities was soon established, but La Tène shapes, gray pottery and polished ornament were important products of pottery workshops in the 1st century AD (Popović 2001, 94). More rapid processes of Romanization were connected with main communications, while for a long time the way of life did not change essentially in the peripheral regions. Rather interesting is the case of large necropoles

I wish to express my gratitude to my colleagues Mira Ružić from the Faculty of Philosophy in Belgrade (Fig. 21.1.22– 23), Sofija Petković from the Institute of Archaeology in Belgrade (Fig. 21.1.24) and Aca Djordjević from the National Museum in Belgrade, for providing me with illustrations (Fig. 21.1.8–9).

Bibliography Andronikos, M. 1969. Vergina I. Athens: Bibliothēkē tēs En Athēnais Archaiologikēs Hetaireias. Benac, A. (ed.) 1979. Praistorija jugoslavenskih zemalja III: eneolitsko doba. Sarajevo. Benac, A. (ed.) 1983. Praistorija jugoslavenskih zemalja IV: bronzano doba. Sarajevo. Benac, A. (ed.) 1987. Praistorija jugoslavenskih zemalja V: željezno doba. Sarajevo. Blagojević, M. 2005. Keramičke posude iz ranoeneolitske kuće sa lokaliteta Livade, Kalenić. Summary: Ceramic Vessels from Early Eneolithic House from the Locality of Livade in the Villige of Kalenić, Kolubara 4, 31–77. Bogdanović, M. 1986. Ljuljaci. Kragujevac: Narodni muzej. Bónis, É. 1969. Die spätkeltische Siedlung Gellérthegy-Tabán in Budapest, Achaeologia Hungarica XLVII. Budapest: Académiai Kiadó. Cermanović-Kuzmanović, A. 1998. Komini – Municipium S...: Nekropole/Komini – Municipium S..: Cemeteries, Beograd

182

Petar Popović

Cerović, M. 1983. Pećine: keltska nekropola. Unpublished thesis, Faculty of Philosophy, Beograd. Coles, J. M. and Harding, A. F. 1979. The Bronze Age in Europe. London: Methuen & Co Ltd. Courbin, P. 1953. Les origines de canthare attique archaique, Bulletin de Correspondance Hellénique 77.II, 322–45. Dizdar, M. 2010. Kantharoi of Autochthonous – ‘Pannonian’ Origin from the La Tène Culture Cemetery in Zvonimirovo, Croatia. In S. Barecki (ed.) Iron Age Communities in the Carpathian Basin, Proceedings of the International Colloquium from Târgu Mureş 9–11 October 2009, 297–307. Cluj-Napoca: Editura Mega. Fidanovski, S. and Cvjetićanin, T. 2005. Roman Villa and Necropolis at the site Kamenitica-Moravište in Mala Kopašnica near Grdelica, Arheološka istraživanja E-75 / Archaeological Research E-75, 49–121. Franke, R. P. and Marathaki, I. 1999. Wine and Coins in Ancient Greece. Athens: The Hatzimichalis Estate. Hänsel, B. 1976. Beiträge zur Regionalen und Chronologischen Gliederung der Älteren Hallstattzeit an der Unteren Donau. Bonn: Habelt. Heurtley, W. A. 1939. Prehistoric Macedonia. Cambridge: Cambridge University Press. Hoddinott, R. F. 1981. The Thracians. London: Thames and Hudson. Hunyady, I. 1942. Die Kelten im Karpatenbecken, Dissertationes Pannonicae II. 18, Budapest: A Királyi Magyar Pázmány Péter Tudományetem Éremés Régiségtani Intézete. Jovanović, B. 1984. Les sépultures de la nécropoles celtiques de Pécine près de Kostolac (Serbie du nord). Études celtiques XXI, 63–93. Jovanović, B. 2005. Poreklo i relativna hronologija ranog eneolita zapadne Srbije. Summary: The Origin and Relative Chronology of the Early Eneolithic Period in Western Serbia, Kolubara 4, 11–17. Jovanović, B. and Jovanović, M. 1988. Gomolava. Naselje mlađeg gvozdenog doba. Beograd: Novi Sad. Keltoi 1984. Keltoi: Kelti i njihovi savremenici na tlu Jugoslavije (Katalog izložbe). Ljubljana. Korkuti, M. 2006. Early Bronze Age: Milutin Garašanin’s important Work and the Research of Albanian Archaeology. In N. Tasić and C. Grozdanov (eds.) Homage to Milutin Garašanin, 31–57. Beograd: SASA. Kruta V. 2000. Les Celtes: histoire et dictionnaire. Paris: Robert Lafont. Kruta, V. and Szabó. M. 1982. Canthares danubiens de IIIe siècle avant notre ère. Une example d’influence hellénistique sur les Celtes orientaux, Études celtiques 19, 51–67. Lollini, D. G. 1976. Sintesi della civilita Picenà, Jadranska obala u protohistoriji, 117–53. Zagreb: Centar za Povijesne Znanosti Sveučilišta u Zagrebu, Arheološki Institut. Majnarić-Pandžić, N. 1970. Keltsko-latenska kultura u Slavoniji i Srijemu. Vinkovci. Majnarić-Pandžić, N. 2009. On the South Pannonian Population in the Late Iron Age. In G. Tiefengraber, B. Kavur, and A. Gaspari (eds.) Keltske studije II: Studies in Celtic Archaeology – Papers in honour of Mitja Guštin, 235–245. Montagnac: Editions Monique Mergoil. Medović, P. 2007. Stubarlija. Novi Sad: Muzej Vojvodine. Megaw, J. V. S. and Megaw, M. R. 1990. Celtic Art: From its beginnings to the Book of Kells. London: Thames and Hudson.

Mitrevski, D. 2006. Genesis of the Iron Age Culture in Macedonia, Folia Archaeologica I – In Honorem Verae Bitrakova Grozdanova, 85–96. Skopje. Petrova, E. 2006. Relations of the Mysians and Dardanians with the Paeonians and Brygians in the Balkans and Asia Minor. In N. Tasić and C. Grozdanov (eds.) Homage to Milutin Garašanin, 537–44, Beograd: SASA. Popović, Lj. 1958. Radolište, Zbornik radova Narodnog muzeja I 1956/57, 75–94. Popović, P. 2001. La ceramique de La Tène finale sur les territories des Scordisques, Starinar n.s. L/2000, 83–111. Popović, P. 2009. Scordisci on the Fringes of the Hellenistic World. In G. Tiefengraber, B. Kavur, and A. Gaspari (eds.) Keltske studije II: Studies in Celtic Archaeology – Papers in honour of Mitja Guštin, 247–258. Montagnac: Editions Monique Mergoil. Popović, P. and Vukmanović, M. 1998. Vajuga-Pesak: Early Iron Age Cemetery/Vajuga-Pesak: nekropola starijeg gvozdenog doba, Cahiers des Portes de Fer, Monographie 3. Beograd. Popović P., Radojčić, N., Krstić, V., and Djordjević, A. 2012. Central Balkans between Greek and Celtic World: KaleKrševica 2001–2011 (Catalogue of Exibition). Belgrade: National Museum. Prendi, F. 1976. Un aperçu sur la civilisation de la première période du fer en Albanie, Jadranska obala u protohistoriji, 155–75. Zagreb: Centar za Povijesne Znanosti Sveučilišta u Zagrebu, Arheološki Institut. Rustoiu, A. and Egri, M. 2010. Danubian Kantharoi – Almost Three Decades Later. In S. Barecki (ed.) Iron Age Communities in the Carpathian Basin, Proceedings of the International Colloquium from Târgu Mureş 9–11 October 2009, 217–87. Cluj-Napoca: Editura Mega. Sladić, M. 2002. Prilog proučavanju helenističkih uticaja na teritoriji srpskog Podunavlja. Summary: A Contribution to the Study of Hellenistic Influence on the Territory of Serbian Danube Basin. In M. Lazić (ed.) Vestigatio vetustatis Aleksandrini Cermanović-Kuzmanović od prijatelja, saradnika i učenika, 433–41. Beograd: Centar za Arheološka Istraživanja. Sparkes, B. A. and Talcott, L. 1970. Black and plain Pottery of the 6th, 5th and 4th Centuries BC, Athenian Agora XII. Princeton: American School of Classical Studies at Athens. Szabó, M. 1992. Les Celtes de l’est. Paris: Editions Errance. Tasić, N. 1995. Eneolithic Cultures of Central and West Balkans. Beograd: Institute of Balkan Studies. Teržan, B. 2009. Eine lateneartige Fremdform im hallstättischen Vače. In G. Tiefengraber, B. Kavur, and A. Gaspari (eds.) Keltske studije II: Studies in Celtic Archaeology – Papers in honour of Mitja Guštin, 85–99. Montagnac: Editions Monique Mergoil. Todorović, J. 1972. Praistorijska Karaburma I, Monografije 3. Beograd: Muzej grada Beograda. Vokotopoulou, I. 1986. Vitsa. Athens: Ekdosē tou Tameiou Archaiologikōn Porōn kai Apallotriōseōn. Zotović, R. 2007. Social and Cultural Aspects of the Burial ‘Krajčinovići-Slana voda’ (south-west of Serbia, middle of II c. B.C.). In V. Sîrbu and S. A. Luca (eds.) Funerary Practices in Europe, before and after the Roman Conquest (3rd century BC–3rd century AD), Acta terrae Septemcastrentis VI, 1, 199–205. Alba Iulia: Ed. Altip.

22 LA TÈNE AND PRZEWORSK STRAP SHIELD BOSSES FROM POLAND Tomasz Bochnak

Introduction The territory of present-day Poland was on the periphery of the La Tène culture area (in the present text, for stylistic reasons, the terms ‘La Tène culture’ and ‘Celtic culture’ are used interchangeably denoting an archaeological culture, without any ethnic connotations whatsoever). To the north of the Carpathian range Celtic settlement is recorded in isolated pockets; in Lower Silesia they are recorded in the region of Wrocław, between the Odra and the Oława rivers, and also, moving eastward, in the Głubczyce Upland and in the Upper San basin. On the Vistula River in the region of Kraków the La Tène culture horizon was succeeded by an archaeological unit known as the ‘Tyniec Group’, a conglomerate of different culture elements (Woźniak 1970; Cumberpatch 1995; Poleska 2006; Dulęba 2009). In his studies Professor Vincent Megaw has made reference on a number of occasions to La Tène material from Poland. He focused on dress accessories (Megaw 2005) and imported vessels (Megaw 1970, 162, 163; Megaw and Megaw 2001, 220; fig. 368), and also on weapons (Megaw and Megaw 1990; Megaw and Megaw 2012). Other La Tène military objects have been discovered in Poland. Some of them derive from assemblages attributed to Celtic culture but the majority are imports in the environment of cultures with aboriginal traditions, most notably the Przeworsk culture which took form in southern and central Poland in the late 3rd century BC and continued until the early 5th century AD. The present paper focuses on strap-type shield bosses, typical metal reinforcements of Celtic shields, which are encountered in Poland both in the earlier La Tène culture assemblages and later assemblages associated with the Przeworsk culture. The shield boss, or the umbo, from Latin, was mounted on the shield to reinforce and protect its central area, especially

at its weakest point – the hole cut in the board to make room for the hand holding the shield’s grip. Shield bosses were mostly made of metal but there are some rare examples of other material such as wood or wicker. Depending on the shape and the construction of the shield the protruding part (boss proper) of the shield boss may be open on both sides, or closed. The forms with an ‘open’ boss are referred to variously as strap-type, strip-type, band, linear, rectangular, butterfly shield bosses, those with the ‘closed’ boss proper (or bowl) are known as circular, although we know of a small group of shield bosses with a bowl and a rectangular flange. Strap shield bosses were mounted onto shields reinforced with a vertical midrib (spina) whereas in circular shield bosses the flange connected with the surface of the shield board the entire length of its perimeter showing that they were mounted onto flat shields without the midrib. The subjects of the present article are strap shield bosses: manufactured from a rectangular strip of metal sheet worked up at the centre to form the open-sided boss proper. Shields with the midrib tended to be complex composite structures whereas the board of the flat shield could be made from a single piece of wood. The form definitely dominant among Przeworsk culture shield bosses is the shield boss with a closed boss proper; strap forms account for merely c. 3% of all finds datable to the Later Pre-Roman period, that is, the time between the late 3rd century BC and the onset of the 1st century AD.

History of research The morphological variation of the strap shield bosses and the correct identification of their function is closely connected to the problem of the reconstruction of the Celtic shield. Thanks to the written evidence and to

184

Tomasz Bochnak

iconographic representations, the construction design and the general appearance of the shield used by Celtic warriors have been sufficiently well understood, but interpretation of archaeological evidence has presented a number of problems. Fundamental in this respect were artefacts from the site at La Tène where the shield reinforcements had survived complete with the organic parts of the shield (Vouga 1923, 59–62; pl. XVI.10a–c, XVII, XVIII). The Swiss site yielded mostly Middle La Tène material, with some artefacts from the onset of the Late La Tène period, and the construction of earlier shields continues to be the object of inquiry and guesswork. Questions related to the reconstruction of wooden shields fitted with a two-piece shield boss, also called the bivalve shield boss, were addressed in the early 20th century by J. L. Pič, but the reconstruction he proposed is flawed (Pič 1902, 18). The correct arrangement of the pair of symmetrical fittings reinforcing the midrib on the outer face of the shield board was presented by A. Procházka, who correctly identified some finds interpreted earlier as the metal border of a shield with thick edges (Procházka 1937, 38–39). The findings of A. Procházka were supplemented and expanded by I. Hunyady whose work has significant value even today, not the least thanks to its rich illustrations, because many of the artefacts documented in it have gone missing or have become seriously corroded (Hunyady 1942; 1944, 121–123). J. Filip addressing questions related to the distribution range and variation of La Tène culture artefacts in Central Europe discussed in brief the morphological transformation of the strap shield boss, paying attention to the gradual increase in the size of its two flanges reflected by the archaeological material from the Middle La Tène period (Filip 1956, 166–168). Based on the observation of materials from the oppidum at Manching, W. Krämer proposed to rectify some of J. Filip’s findings, including the dating of the two-piece shield bosses, and dated them to the 3rd rather than to the 2nd century BC (Krämer 1962, 306). However, these conclusions were not taken into account in the next work on the morphological and chronological evolution of shield boss finds from Lower Pannonia, where the appearance of two-piece shield bosses is dated to the 2nd century BC (Todorović 1966). Metal shield reinforcements from the territory of the Eastern Celts were investigated by M. Domaradzki, who, based on a typological-chronological analysis of 183 shield bosses, distinguished them into five main groups (including forms with a spherical bowl) and six types of shield grips (Domaradzki 1977). In studies published in subsequent years by D. Božič and M. Guštin, the relative chronology of shield fittings and other materials from the eastern and the western areas of the Balkan Peninsula was refined, and they were compared to the finds from the grave sites in Belgrade-Karaburma and northern Italy (Božič 1981; Guštin 1984; 1991, 56–58). Strap shield bosses from western

Europe were investigated by I. M. Stead. He proposed an evolutionary-typological scheme for the shield bosses from Yorkshire in correlation with the most widespread types of these shield fittings recorded in continental Europe (Stead 1969). The origin and evolution of shield bosses mounted onto shields with a midrib was investigated by P. F. Stary (Stary 1981). Drawing on archaeological and iconographic evidence from western and southern Europe, he proposed a scheme to illustrate the evolution of shields from the 8th century BC until the onset of the 1st century AD. In the view of A. Rapin, adopting such a broad source base had a negative impact on the precision and transparency of the said analysis (Brunaux and Rapin 1988, 24, fig. 14). Vital input to the typological study of Celtic shield bosses of the 2nd and 1st century BC came from analysis of the votive site at Gournay-sur-Aronde (dép. Oise, France), which yielded at least 250–300 examples of these shield fittings (Brunaux, Meniel and Poplin 1985, 71). Thanks to their fine preservation, 214 could be used in a typologicalchronological analysis (Brunaux and Rapin 1988, 46, 47). Drawing on this extensive database, nine technological groups of shield bosses were distinguished, marked with letters A – I, and seven types, marked with numbers I–VII. Other than in the source publication, the typology and dating of strap shield bosses from Gournay-sur-Aronde was addressed in subsequent studies by A. Rapin. Based on finds of strap shield bosses from Slovenia and northern, Italy M. Guštin distinguished types Skorba and Mokronog-Arqua (Guštin 1991, 56–58, fig. 29, 30; 2002, 15–16, fig. 2). These are large shield forms with a low raised central part dated to late LT C2 and especially to LT D1 (Guštin 1991, 57). Materials from the grave site at Oleggio (reg. Piemonte, Italy), were used by G. Spagnolo Garzoli to differentiate between Middle and Late La Tène shield bosses from Gallia Transpadana (Spagnolo Garzoli 1999, 351–354). The next contribution to the study of strap shield bosses dating from the onset of the Late La Tène period was made by M. Schönfelder, who analysed the materials from the rich grave at Verna (Isère, France) and distinguished three more types – type Nierstein, type Nîmes and type Lamadelaine (Perrin and Schönfelder 2002, 80–85). A synthesis of the above findings was proposed by L. Pernet, who identified four groups of Late La Tène shield bosses: type Mokronog, shield bosses with long, rectangular flanges, shield bosses with short flanges (here he included type Nierstein and type Nîmes), and circular shield bosses with a closed bowl, a separate typo-chronological branch (Pernet 2010, 109–112). The classification system proposed by L. Pernet takes into account the most frequent types of shield bosses and leaves out the less popular variants. In Poland, Celtic strap shield bosses are recorded within two chronological horizons and in two distinct archaeological cultures. The older horizon is that of inhumation graves of the La Tène culture, dated to the period LT B2 – LT C1a

22.  La Tène and Przeworsk strap shield bosses from Poland

185

Fig. 22.1. Finds of strap shield bosses in La Tène culture assemblages. A – La Tène culture during phases LT B2-C1; B – La Tène culture succeded by Tyniec group during phases LT B2-D2. 1 – Głownin grave 1/1930; 2 – Iwanowice grave 9; 3 – Iwanowice grave 34; 4 – Sobocisko grave 8/1940; 5 – Sobocisko grave 9/1940; 6 – Sobocisko grave 1/1952.

(Fig.  22.1), the more recent horizon is that of cremation burials of the Przeworsk culture dated to phases A1 and A2 of the Later Pre-Roman period, which correspond, grosso modo, to stages LT C2 and LT D1 (Fig. 22.2). In these graves, archaeologists have discovered strap shield bosses typical for the La Tène culture.

Strap shield bosses in La Tène culture Some of the finds from the Vistula Basin show that not all Celtic shields in our region were fitted with a metal shield boss. In La Tène graves 7/1940, 24/1940, 26/1940, 3a/1955, recorded at Sobocisko (German Zottwitz, Oława, Lower Silesia), and in graves 1699 and 1701 at Kietrz (Głubczyce) in the foreland of the Moravian Gate, the presence of shields was documented by finds of metal edge bindings and long u-shaped fittings covering the vertical midrib, but there were no metal shield bosses (Hoffman 1940, 12, 13, 17, 18, 20, 21; fig. 3.6; Czerska 1966, 96, fig. 16; Gedl 1978, 16, 17, 36, 37, 38–40). Shields lacking a metal shield boss and on occasion with a greatly reduced number of other

fittings are known from other regions of Celt territory. One parallel would be the inhumation burial at Chens-sur-Léman (Haute Savoie, France) phase LT B2, where the only element suggesting the presence of a shield was a small curved edge binding (Landry and Blaizot 2011, 152–154, 157, fig. 5.7). The oldest La Tène shield bosses from Poland are bivalve specimens. They surfaced in Lower Silesia, at Sobocisko (German Zottwitz), in inhumation graves 8/1940, 9/1940 and 1/1952 (Hoffmann 1940, 13–17, figs. 5, 6, 8.6; Czerska 1966, 90, 91, fig. 4:k). The shield boss from grave 8/1940 consists of two pieces, c. 11 by 7 cm, each of them attached using two nails, and a metal strip between them, originally running the length of the spina (Fig. 22.3.1). This ‘u’-shaped fitting of the midrib survived in six fragments, with a total length of 41 cm. The shield boss from grave 9/1940 from Sobocisko was presumably of a similar construction. This specimen was seriously fragmented and its shape cannot be reconstructed, but it appears to be another two-piece shield boss with a vertical midrib in between the two halves of the shield boss (Fig. 22.3.2a). The rest of the metal shield fittings published by W. Hoffmann, thick pieces of metal sheet, may be identified tentatively as fragments of a shield

186

Tomasz Bochnak

Fig. 22.2. Finds of strap shield bosses in Przeworsk culture assemblages. A – Przeworsk culture during the Younger Pre-Roman period; B – area where Przeworsk settlement disappears from the archaeological record during phase A2 of the Younger Pre-Roman period. 1 – Dzierżążnia Nowa grave 132; 2 – Jeszkowice stray find; 3 – Legionowo grave 47; 4 – Piotrków Kujawski stray find; 5 – Ziemnice grave 1; 6 – Żukowice grave 41; 7 – Żukowice grave 54; 8 – Żukowice stray find.

grip, Domaradzki type B (Fig. 22.3.2d and 22.3.2e), and circular appliqués fixed to the shield board (Fig 22.3.2b and 22.3.2c). Finally, from grave 1/1952 at Sobocisko comes one more iron shield fitting. It is very poorly preserved. Presumably, it represents the remains of another bivalve shield boss (Fig. 22.3.3). Shield bosses such as were discovered at Sobocisko belong to type A in the typology of M. Domaradzki, which covers bivalve forms. Bivalve shield bosses are known from e.g. Förk-Emmersdorf (Kärnten, Austria), Staňkovice (Žatec) grave 4, Mistřin (Hodonín) grave XXXIV, and Nový Bydžov (Hradec Králové, Czech Republic) (Pittioni 1954, 700, fig. 488.5; Filip 1956, 381, pl. XLV.15, LXXXVIII.19–21; Venclová 2008, 104, 106, fig. 53.2). Parallel examples are also encountered in western Europe, e.g. at Ecury-leCrayon (Marne, France) (Rapin 2001, 283, 290, fig. 6.2). The chronology of all the assemblages listed here fits within phase LT B2, this would also be the dating of the shield bosses from Sobocisko. A slightly later shield boss dated to the beginning of the Middle La Tène period comes from inhumation grave

1/1930 discovered at Głownin (German Glofenau, Strzelin) (Fig. 22.3.4), published by M. Jahn (Jahn 1934, 120–122, fig. 15). This is a c. 23 × 11 cm piece, with rectangular flanges and a highly domed c. 6.7 cm central part. The rivet holes are not marked on the drawing. The same assemblage from Głownin included late variants of iron fibulae of Early La Tène design and a lignite bracelet of a form encountered in LT B2 and LT C1. The strap shield boss from Głownin is a form classified by M. Domaradzki to his group II1A (Domaradzki 1977, 58). In the typology of A. Rapin this specimen belongs in group F, and among the strap shield bosses from Gournay-sur-Aronde it finds a parallel in type 1B examples (Brunaux and Rapin 1988, 44–46, 78, 79, fig. 24, fig. 39). Similar shield bosses are encountered in assemblages dated as early as late phase LT B2, but are frequent, especially during the Middle La Tène when they show a shift in their distribution range. Older examples are known mostly from Eastern Celt territory, e.g. from the Slovak cemetery site at Trnovec nad Váhom-Horný Jatov (Šaľa) grave 537, and Bajč-Vlkanovo (Komárno) grave 34 (Benadik, Vlček

22.  La Tène and Przeworsk strap shield bosses from Poland

187

Fig. 22.3. La Tène shield bosses in Poland. 1 – Sobocisko grave 8/1940 (after Hoffmann 1940); 2 – Sobocisko grave 9/1940 (after Hoffmann 1940); 3 – Sobocisko grave 1/1952 (after Czerska 1966); 4 – Głownin grave 1/1930 (after Jahn 1934); 5 – Iwanowice grave 9 (after Woźniak 1970); 6 – Iwanowice grave 34 (after Woźniak 1970).

188

Tomasz Bochnak

and Ambros 1957, 36, pl. VII.6; Benadik 1960, 401, pl. IX.12), where with time they disappear from the record, unlike in the Czech Republic and in the western zone of La Tène culture where this form of strap shield boss continues in evidence. Related shield bosses are known from burials discovered at Radovesice (Litoměřice) grave 13, Nová Ves (Mělník), Blučina (Brno-venkov) grave 16, Bučovice (Vyškov), Dražúvky (Hodonín) grave recorded in 1933 (Filip 1956, 167, 367, pl. LI.14; Waldhauser 1987, 117, pl. 20.12; Čižmářová 2004, 132, 158, 171, 172). Analogous forms were discovered at La Tène (Vouga 1923, 59–62, pl. XV.9 and XVII.1a). They were attached using two fastening rivets, which suggests that a similar solution was used in the shield boss from Głownin. Other types of Celtic strap shield bosses surfaced in closed assemblages discovered at Iwanowice (Kraków) cremation graves 9 and 34 (Woźniak 1970, 108, 120, 121, pl. XXIX.1, 1b, XXXI.1a; 1991). They have the form of a strip of sheet metal with tapering terminals (the flange), the specimen from grave 34 is somewhat less well preserved and its exact shape cannot be reconstructed (Fig. 22.3.6). In the shield boss from grave 9, the central raised part was lightly profiled (Fig. 22.3.5). Both shield bosses from Iwanowice were over 30 cm long and would have been attached using two fastening rivets. Similar specimens are encountered both in central and in western Europe. A similar shield boss comes from Predmostie (Veľký Krtíš, Slovakia), grave 4 from Makotřasy (Kladno, Czech Republic), from the grave site at Vác (Pest, Hungary) grave 15, and Ludas–Varjú-dűlő (Heves) graves 1056, 1057, 1140 (Filip 1956, 417, pl. CII.10; Domaradzki 1977, 60; Venclová 2008, 106, fig. 53.6; Szabó and Tankó 2012, 69–72, 124; pl. XLII.8, XLIV.2, XLVII.4). Similar specimens are known also from Western Celt territory, notably, from the votive site at Gournay-sur-Aronde. Shield boss no. 3972 has flanges with strongly attenuated terminals by which it resembles the specimen from grave 9 at Iwanowice (Brunaux and Rapin 1988, 178, 219, pl. XXXVII.3972). The shield bosses invoked here definitely belong to group II1Dbγ of M. Domaradzki, which are dated broadly to LT C (Domaradzki 1977, 60–61). In the typology of Rapin, the shield boss from grave 9 at Iwanowice would be type III B whereas the shape of its raised part and flanges could be traced back to group ‘a’ and ‘b’ shield bosses from Gournay-sur-Aronde (Brunaux and Rapin 1988, 41–43, 78, 80; fig. 23, fig. 39). Type III of A. Rapin is dated to the late 3rd century BC with a chronological range somewhat narrower than determinations made by M. Domaradzki with regard to the Eastern Celtic finds. Initially the graves from Iwanowice were placed in phase LT C2 but now their chronology was revised as stage LT C1b, and this is in agreement both with the conclusions of M. Domaradzki and of A. Rapin (Woźniak 1970, 108, 120, 121, pl. XXIX.1, 1b, XXXI.1a; 1991). The graves from Iwanowice are some of

the latest Celtic graves with weapons known from Poland. This is because beginning with phase LT C2 the people of La Tène culture in our region adopted a burial custom elusive to contemporary archaeological methods.

Strap shield bosses in Przeworsk culture This transition in the burial customs observed in La Tène culture may be treated as a terminus ante quem for the emergence of the Przeworsk culture, the local culture unit evolved by the aboriginal population under the impact of La Tène culture who adopted cremation and intentional destruction of grave goods (Dąbrowska 1988; Bochnak 2006a). This burial rite, typical for Celtic grave sites defined as type Ponětovice-Holiare (Meduna 1962, 88–96), was sustained with minor modifications throughout the entire duration of the Przeworsk culture, that is until the early 5th century AD. The Przeworsk culture is regarded as one of the most strongly ‘Latènized’ cultures of non-Celtic Europe (Dąbrowska 1988, 105; 2003, 153). Elements of La Tène culture are evident not only in the burial customs, but also in material culture as well. Przeworsk culture grave inventories feature some imports from the Celtic world. Among imported military objects the dominant form definitely is the sword. At present we know of some 150 of these weapons which have been discovered in sets with imported metal scabbards, and it is likely that of some 45 or so sword blades discovered without a scabbard a significant number also found their way to the Przeworsk environment from the Celtic sphere (Bochnak 2005, 17; since 2005 the number of imported swords has increased a little). In comparison to imported swords, Celtic defensive weaponry is much less well represented. We know of a single helmet, discovered at Siemiechów (Łask) grave 25 (Jażdżewska 1988; 1992; 1994a; 1994b; Kaczanowski 1992a, 172; 1992b, 53). Celtic shield fittings are equally scarce because the forms which definitely dominates in the archaeological record are shield bosses of local make (Bochnak 2006b). Currently from Przeworsk culture assemblages we know of just nine specimens of strap shield bosses, most of them preserved in fragmented form. Eight of them were discovered in Poland and one surfaced in a grave site on the left bank of the Odra River, at Oderberg (Barnim, Germany). All the better preserved specimens are shield bosses with a protruding central section and rectangular flanges which were discovered in cremation graves. Making finer morphological distinctions in this really modest series of strap shield bosses used by the people of Przeworsk culture is a real challenge, but we may assume that in this group the find from Piotrków Kujawski (Aleksandrów Kujawski) (Fig. 22.4.1) takes special place. This is a large strap shield boss unearthed by accident in 1954 (Zielonka 1955; 1956). At this time several Przeworsk

22.  La Tène and Przeworsk strap shield bosses from Poland

189

Fig. 22.4. Przeworsk culture shield bosses in Poland. 1 – Piotrków Kujawski stray find (after Bochnak 2005); 2 – Ziemnice grave 1 (after Łuczkiewicz 2000); 3 – Jeszkowice stray find (after Jahn 1931); 4 – Żukowice grave 41 (Courtesy of Krzysztof Demidziuk, Muzeum Archeologiczne, Wrocław. Drawing by Teresa Demidziuk); 5 – Żukowice stray find (Courtesy of Krzysztof Demidziuk, Muzeum Archeologiczne, Wrocław. Drawing by Teresa Demidziuk).

culture graves were destroyed and we cannot reconstruct individual assemblages or even their number. The shield boss analyzed here presumably belonged in a set with a sword and fragments of a scabbard also recovered at the time. The same inventory may have also included the iron point of a shafted weapon. The shield boss from Piotrków Kujawski has an oval-shaped raised central part decorated on the edges with a double engraved groove, and straight

flanges with two symmetrically placed fastening rivets. The edges of the flanges did not survive and it is possible that originally they were longer and were attached using two pairs of fastening rivets. The surviving dimensions of the shield boss are 23 × 12.5 cm, and as it is difficult to detach the narrow strip of metal it is possible that originally this specimen measured at least 30 cm. The strap shield boss from Piotrków Kujawski is the largest of its kind

190

Tomasz Bochnak

to be discovered in a Przeworsk grave assemblage. From outside the Przeworsk culture territory we have two large strap shield bosses attached with one, possibly two, pairs of fastening rivets. Strap shield bosses attached with a single pair of fastening rivets have their closest parallels in forms classified by M. Domaradzki as type IIIA (Domaradzki 1977, 61). Specimens with long flanges similar to the shield boss from Piotrków Kujawski were discovered e.g. at Hart (Steiermark, Austria), Szirmabesenyő (Borsod-AbaújZemplén, Hungary) and Perkáta (Fejér, Hungary), and Zvečka-Obrenovac (Serbia) (Babeş 1993, 116). A closely related shield boss with its raised central part decorated on the edges with four parallel lines, surfaced in grave 29 in the grave site of Poieneşti-Lukaševka culture at Boroseşti (Iaşi, Romania). The Romanian specimen is 35.7 × 13.5–14.2 cm (the flanges are slightly wider than the raised part), and these values do not differ significantly from the dimensions of the shield boss from Piotrków Kujawski (Babeş 1993, 116; pl. 5e). All the specimens invoked here belong in phase LT C2, and this suggests that if the shield boss from Piotrków Kujawski had a single pair of fastening rivets, then it ought to be dated to the later segment of phase A1, synchronized with the end of the Middle La Tène period. Similarly dated are large shield bosses with two pairs of fastening rivets known from western Europe, e.g. from the grave site at Giubiasco (Ticino, Switzerland) and the necropolis found at the confluence of the Yonne and the Seine between the communes Barbey and Misy-sur-Yonne (Île-de-France, France). The largest series of shield bosses with flanges that were attached using four fastening rivets comes from Picardy, from Gournay-sur-Aronde, and from the neighbouring sanctuaries at Saint-Maur (Oise) and Digeon (Somme), the earliest of them come on record in the first half of the 2nd century BC (Brunaux and Rapin 1988, 72, 78). In short, we have reason to date the shield boss from Piotrków Kujawski to phase LT C2, or in the relative chronology of Przeworsk culture, the end of phase A1 and the very onset of phase A2 of the Later Pre-Roman period. This chronological determination is not altered by the fact that we do not know whether the shield boss from Piotrków Kujawski originally had two or four fastening rivets. All other strap shield bosses from Polish finds are visibly smaller and form a quite typologically compact group. Most of them have a length of 20–25 cm and a width of c. 7–8 cm, and have rectangular flanges attached using four fastening rivets arranged in a single row, two fastening rivets to each flange. An incomplete shield boss from Żukowice, grave 41, has in its one surviving flange three rivet holes suggesting that it was attached more than once, presumably when the shield was damaged. The best preserved shield boss surfaced at Legionowo (Legionowo) in feature 47. Its dimensions are length c. 21.8 cm and width c. 7.7 cm, the rims of the raised central part are slightly thicker. This shield boss was found in an

inventory together with a sword resting in its scabbard, an iron belt-hook for the baldrick – a form with projections (‘wings’), three iron rings from the sword belt, a spearhead with a wavy incised blade, a spear ferrule, a knife, four small fastening rivets and some pottery (Orliński 2011, 9, 10, photo 1; personal communication and query made in regional museum in Legionowo. I am indebted to W. Orliński, Muzeum Historyczne w Legionowie, for his valuable assistance). This assemblage definitely belongs in phase A2 of the Later Pre-Roman period, which corresponds in general to stage LT D1. A shield boss from Ziemnice (Leszno) grave 1 had a length of c. 21 cm and a width of c. 12 cm (Fig. 22.4.2). It was discovered together with a spearhead and some pottery (Łuczkiewicz 2000, 381, 383, fig. 22). Although P. Łuczkiewicz dated this grave inventory to phase A1 of the Later Pre-Roman period, I believe it is safer to place it in phase A1 and A2, as there were no chronologically sensitive diagnostic forms in this assemblage. The inventory of grave 132 from Dzierżążnia Nowa (Płońsk) includes fragments of iron sheet, described on the museum catalogue card as the remains of a destroyed shield grip. The largest fragment, with a pair of holes, is c. 7.2 × 6.7 cm, and may be identified tentatively as the flange of a strap shield boss. Other fragments of iron sheet may be interpreted as fragments of the raised edge of central part of the shield boss and the other flange (Bochnak 2005, 109). The same assemblage included a fragment of a sword and a fibula, Kostrzewski type K, which date the inventory to phase A2 (Łuczkiewicz 2006, 302). The cemetery at Żukowice (Głogów) in Lower Silesia yielded three strap shield bosses (pers. comm. K. Demidziuk and data from archival records, Archaeological Museum in Wrocław corroborated by a query made in the Museum of Archaeology and History in Głogów, MAH. I am indebted to K. Demidziuk and to Z. Hendel, MAH, for their valuable assistance). The specimen from grave 41 had ornamented edges of the raised central part and measured c. 24 cm by c. 8 cm (Fig. 22.4.4). In the same assemblage, next to the pottery, were a sword and scabbard, four spearheads, two knives, shears, fragments of four fibulae, metal mounts from a casket and five or so undetermined tools, presumably surgical instruments, as well as a whetstone, possibly also a part of that set. The presence of shears dates this assemblage to phase A2, but we cannot dismiss an even earlier chronological position – in phase A1. The next, fragmented shield boss from Żukowice, grave 54, is too damaged to specify its dimensions. It was found together with a sword, two belt-hooks for the baldrick, two knives, a razor, fragments of three fibulae and ceramics. The grave inventory belongs to phase A2. The third strap shield boss from Żukowice was seriously fragmented (Fig. 22.4.5) and is a stray find from that grave site. We have only very patchy information about one more

22.  La Tène and Przeworsk strap shield bosses from Poland

191

Fig. 22.5. Type Nierstein shield bosses in Europe. A – certain type; B – uncertain type. 1 – Acy-Romance; 2 – Gournay-sur-Aronde; 3 – Bad-Nauheim; 4 – Dzierzążnia Nowa, grave 132; 5 – Gross-Gerau, grave 8; 6 – Hainburg/Klein-Krotzenburg; 7 – Heidelberg; 8 – Hochheim; 9 – Jeszkowice; 10 – Legionowo, grave 47; 11 – Manching; 12 – Montmartin; 13 – Nierstein; 14 – Nordheim/Bruchhöche; 15 – Obernau; 16 – Oderberg; 17 – Offenbach-Bürgel; 18 – Oleggio, grave 53; 19 – Plaidt/In der Keuzchen; 20 – Sainte Cécile; 21 – Schwangau; 22 – Thayngen; 23 – Verna; 24 – Ville-sur-Retourne; 25 – Wallau, 26 – Wallertheim, grave 3/1928; 27 – Wallertheim, grave 5/1928; 28 – Wallertheim, grave 34; 29 – Wederath, grave 1178; 30 – Wederath, grave 1216; 31 – Wederath, grave 1228; 32 – Ziemnice, grave 1; 33 – Żukowice, grave 41; 34 – Żukowice, grave 54; 35 – Żukowice, stray find.

strap shield boss from the Przeworsk culture territory. This specimen surfaced during the first half of the 19th century at Jeszkowice (German Jäschkowitz, Wrocław), and is known to us from a rough drawing (Fig. 22.4.3) and brief references in literature (Jahn 1931, 58, 60, fig. 60); it is often listed – incorrectly – in a group with archaeological artefacts also discovered at Jeszkowice and confused with a conical shield boss, type J.5 (Jahn 1924, 35). According to the report published in M. Jahn’s study on Celtic finds from the area of Silesia, this particular shield boss was associated with three fastening flat-headed rivets (Jahn 1931, 58, 60, fig. 60). As the number of rivets is even, we may conclude that this specimen too belongs in the group of shield bosses with four fastening rivets. If we analyse the distinctive features of 158 finds of shield bosses from the Eastern Celt territory listed by M. Domaradzki (Domaradzki 1977), we have to conclude that in the area to the south of Poland there are no exact counterparts of strap shield bosses from the Przeworsk culture territory. None of the strap shield bosses discussed by M. Domaradzki (Domaradzki 1977) was attached to its shield using four rivets arranged in a single row. All shield bosses found at a small distance from our study area, e.g. on the territory of the Púchov culture in northern

Slovakia, were circular forms (Pieta 2005, 51–53, 71–73; pl. XI.3, XII, XIII.1, 2, 6a–c). Let us note nevertheless that the forms with four fastening rivets have tangible, even if rather unexpected, connections with Celtic deposits from western Europe. Closely related specimens, described as type Nierstein (Fig. 22.5), were discovered in France, in the previously mentioned Late La Tène burial at Verna, at Ville-sur-Retourne (Ardennes) grave 3 and in Italy at Oleggio grave 53 (Spagnolo Garzoli 1999, 112, 114, fig. 95.3; Perrin and Schönfelder 2002, 80–85; Pernet 2010, 206, 207, 261, pl. 33.2, 241C.16). A poorly preserved shield boss with flanges characteristic for type Nierstein surfaced in a grave at Sainte Cécile (Vaucluse, France) and was dated to phase LT D1 (Pernet 2010, 218, 219; pl. 75A.3). The largest concentration of shield bosses type Nierstein is observed in Germany, in the basin of the Rhine and the Main (Perrin and Schönfelder 2002, 80–85). Small shield bosses with flanges attached using four fastenings are known from assemblages dated to phase LT C2 discovered at Gross-Gerau (GrossGerau) in grave 8, at Obernau (Altenkirchen) graves 3/1928, 5/1928, at Wallertheim (Alzey-Worms, perhaps, also grave 34 from the same cemetery), and possibly at Hochheim. A later chronological position is occupied by similar shield bosses discovered in a grave at Wallau (Main-

192

Tomasz Bochnak

Taunus), and in grave 1/1914 at Heidelberg, and possibly at Hainburg/Klein-Krotzenburg (Offenbach) (Polenz 1971, 48–50, 94, fig. 29.12, 58). From the cemetery at Wederath (Bernkastel-Wittlich) come further strap shield bosses with four fastening rivets analogous to the specimens discovered in Poland. Shield bosses from graves 1178 and 1216 at Wederath belong to phase LT D1 (Haffner 1978, pl. 297.1a, 1b, 309.5; 1989a, 65, fig. 42; 1989b, fig. h). This is also the case for the assemblage from grave 1228, which contained a shield boss of a shape reminiscent of the forms under discussion here but with curved flanges (Haffner 1978, pl. 314.3a and 3b). The latest assemblages with strap shield bosses of the described type come from assemblages dated to phase LT D1 at Nierstein (Mainz-Bingen) grave 1, from Offenbach-Bürgel (Offenbach am Main) (an assemblage with a fibula of Late La Tène design), and possibly from grave 9 at Bad-Nauheim (Polenz 1971, 48–50, 94, fig. 29.12, 58; Oesterwind 1989, 108, 109, fig. 28.1). A special case would be the shield boss from Schwangau (Ostallgäu) (Maier 1985, 234, 237, fig. 5.5). This specimen surfaced at the votive site dated to the Roman period but is certain to have an older chronology. Many shield bosses from the area between the Rhine and the Main have reinforced edges of the raised central part (Polenz 1971, 48) which would be an additional common feature shared by the shield bosses from Legionowo, feature 47, and from Dzierzążnia Nowa, grave 132, with type Nierstein bosses. The relatively late chronological position of shield bosses from western Germany corresponds in the main to the dating of the finds from Legionowo, grave 47, from Dzierzążnia Nowa, grave 132, Żukowice, graves 41 and 54 (perhaps, from Ziemnice as well?). In conclusion we can say that, thanks to the analysis of the chronology of assemblages with shield bosses attached using four fastening rivets, discovered in western Germany and eastern France, most of the strap shield bosses from Polish finds regarded previously as a form diagnostic for phase A1 of Przeworsk culture (Dąbrowska 1988, 55, 56, Łuczkiewicz 2000, 381) have lost their value of a chronologically diagnostic form now that we have evidence that they belong in phase LT D1. As mentioned earlier, strap shield bosses with rectangular flanges and four fastening rivets are recorded mainly in Germany and France. Significant similarity of the strap shield bosses from Polish finds gives us reason to argue that strap shield bosses from the Vistula Basin are imports. Those specimens which surfaced in closed assemblages were found together with swords typical for La Tène culture, and on occasion, also with other Celtic imports. Worth special note in this context is the assemblage from Żukowice, grave 41, which, next to a set of weapons and a casket with precision (surgeon’s?) tools, included shears, typical for the western area of Celtic settlement. These finds would testify to contacts between the people of the Przeworsk culture and communities where the Western Celtic model of military

outfit was in use. The dating of the La Tène and Przeworsk assemblages containing analogical shield bosses shows that this exchange would have been particularly lively in phase LT D1, the period we can synchronize with phase A2 of the Later Pre-Roman period, although there may have been sporadic contacts even back in phase LT C2. We can hardly expect that strap shield bosses were imported specifically to have them mounted onto locally made shields (making of a strap shield boss is not a particularly difficult task and, presumably, one that would not pose a problem to the local Przeworsk blacksmiths). We may conclude therefore that the shield bosses were brought to the region as an integral element of the shields. Celtic shields were a rarity in the territory of present-day Poland, where the dominant form were shields fitted with circular shield bosses. From the territory of the Przeworsk culture we have just eight finds of strap shield bosses as compared to more than 80 imported swords datable to the same period. This disproportion suggests that local shields with the spherical bosses answered their purpose sufficiently well and that there was no need to adopt Celtic models in this regard. We need to note at the same time that in Poland the distribution range of strap shield bosses overlaps in general with the distribution range of Celtic pottery. The latter has been recorded in Silesia, Lesser Poland, e.g. at Dzierzążnia Nowa, and in the region of Kujawy (Woźniak 1970, 162, 163; Makiewicz 1973, 119; 1986, 28, 29; Dąbrowska 1988, 127–130; map 15; 2008, 73–75). According to P. Łuczkiewicz Celtic painted wares were discovered in grave 132, whereas T. Dąbrowska claimed that this was grey pottery from grave 107 and 107a (Łuczkiewicz 2006, 302; Dąbrowska 1988, 129; 2008, 73, 74, 133). In the territory of Przeworsk culture the overwhelming majority of Celtic pottery comes from settlement sites, which suggests its utilitarian character. Occurrence of this type of material in settlement contexts is something characteristic for areas penetrated by arrivals from the oppidum zone and suggests the existence of a close relationship between the local population and the representatives of La Tène culture.

Conclusion To sum up the above discussion, we can say that shield bosses encountered in La Tène graves in Poland correspond to forms used by the people of this culture in the neighbouring areas, notably in the Czech Republic and Slovakia. This observation fully confirms observations made by Z. Woźniak and P. Poleska on the existence of ties between local islands of La Tène settlement with communities to the south of the Carpathian range (Woźniak 1970, passim; Poleska 2006, passim). The situation is evidently different in Przeworsk culture where in the group of strap shield bosses we observe a

22.  La Tène and Przeworsk strap shield bosses from Poland clear domination of forms classified as type Nierstein, characteristic for western Europe. We may assume that the majority of Celtic imports were introduced to the territory of present day Poland from the south, and their presence has to do with the operation of the Amber Road. At the same time, from the Vistula basin we have no record of shield bosses typical for the Adriatic region, or of ones similar to shield bosses documented in the Czech Republic and Slovakia. Neither are there finds of type Skorba and Mokronog-Arqua shield bosses. The distribution range of type Nierstein in Poland suggests they were brought to the area from the west. There is a striking concentration of these forms in Lower Silesia and in Lubusz Land, in the Odra Basin. How can this phenomenon be explained? Some materials attributed to Przeworsk culture were discovered in Thuringia (Saxony-Anhalt, Hesse and Wetterau) (Hachmann 1957, Peschel 1977; 1978; Dąbrowska 1988, 158–167; 2003, 545). Isolated fragments of Przeworsk pottery have surfaced in the oppidum at Manching (Stöckli 1979, 53). These finds document the existence of some relationship between the people of Przeworsk culture and Celtic tribal communities in Germany during phases A1 and A2 of the Later Pre-Roman period, that is, approximately during phases LT C2 and LT D1. T. Dąbrowska has argued that exchange was mainly through Lubusz Land and southern Brandenburg and the point of departure would have been the northern reaches of Lower Silesia and southwestern Greater Poland (Dąbrowska 1988, 166). During phase A2 these areas became seriously depopulated and it is likely that their inhabitants migrated west, which was connected with the movement of the Suebi and other allied Germanic peoples led by Ariovistus (Godłowski 1983, 311). The cemetery at Żukowice, with its three strap shield bosses, type Nierstein, went out of use. The region would be settled once again by the people of Przeworsk culture only during the Early Roman period. The presence in Lower Silesia of finds of strap shield bosses, and of other western Celtic imports too, shows that the communication between cultures was not one-sided, there was feedback and the contacts were bilateral. This is perfectly understandable given that migration to the west must have bypassed most of the more accessible communication routes, which in this part of Europe mostly run along the rivers which flow from the south to the north. As L. Moczulski has noted, the selection of a communication route other than the most obvious one, meaning one that is inconvenient, must be dictated by the awareness of one’s target destination (Moczulski 2007, 115). When migrating ‘from a place’ (for example, leaving one’s earlier abodes to escape famine) one tends to choose the most convenient migration route, but when migrating ‘to a new place’ the chosen destination dictates the route to be taken. Finds of western Celtic imports, most notably, of type Nierstein shield bosses, recorded in western Poland prove that the westward migration of the Przeworsk culture

193

people from Lower Silesia had a well-specified destination and was preceded by bilateral contacts. Translated by Anna Kinecka

Bibliography Babeş, M. 1993. Die Poieneşti-Lukaševka-Kultur. Ein Beitrag zur Kulturgeschichte im Raum östlich der Karpaten in den letzten Jahrhunderten vor Christi Geburt, Saarbrücken Beiträge zur Altertumskunde 30. Bonn: R. Habelt. Benadík, B. 1960. Besonders angelegte Gräber auf keltischen Gräberfeldern der Slowakei und ihre gesellschaftliche Bedeutung, Alba Regia 14, 97–106. Benadik, B., Vlček, E. and Ambros, C. 1957. Keltské pohrebiska na juhozápadnom Slovensku, Archeologica Slovaca, Fontes Instituti Archaelogici Nitriensis Academiae Scientiarum Slov­ acae 1. Bratislava: Vydavatel’stvo Slovenskej akadémie vied. Bochnak, T. 2005. Uzbrojenie ludności kultury przeworskiej. Rzeszów: Wydawnictwo Uniwersytetu Rzeszowskiego. Bochnak, T. 2006a. Les Celtes et ses voisins septentrionaux. In M. Szabó (ed.) Celtes et Gaulois. L’archéologie face à l’histoire, Les Civilisés et les Barbares, Actes de la table ronde tenu à Budapest, 17–18 juin 2005, Collection Bibracte 12/3, 159–83. Glux-en-Glenne: Bibracte. Bochnak, T. 2006b. Early Circular Umbones of the Przeworsk Culture. The Role of Local Tradition and Celtic Influences on the Diversity of Metal Parts of Shields at the Beginning of the Late Pre-Roman Period, Analecta Archaeologica Ressoviensia 1, 161–94. Božič, D. 1981. Relative Chronologie der jüngeren Eisenzeit im jugoslawischen Donauraum, Arheološki Vestnik 32, 315–36. Brunaux, J.-L., Meniel, P. and Poplin, F. 1985. Gournay I. Les fouilles sur le sanctuaire et l’oppidum (1975–1984), Amiens: Revue Archéologique de Picardie, Numéro spécial. Brunaux, J.-L. and Rapin, A. 1988. Gournay II. Boucliers et lances, dépots et trophées. Paris: Errance. Čižmářová, J. 2004. Encyklopedie Keltů na Moravě a ve Slezsku. Praha: Libri. Cumberpatch, C. G. 1995. Settlement and economy in late Iron Age. Slovakia, southern Poland and trans-danubian Hungary. In J. D. Hill and C. G. Cumberpatch (eds.) Different Iron Ages: studies on the Iron Age in temperate Europe, BAR International Series 602, 195–217. Oxford: British Archaeological Reports. Czerska B. 1966. Celtyckie cmentarzysko szkieletowe koło wsi Sobocisko, pow. Oława, Wiadomości Archeologiczne 32, 88–103. Dąbrowska, T. 1988. Wczesne fazy kultury przeworskiej. Chronologia-zasięg-powiązania. Warszawa: Państwowe Wydawnictwo Naukowe. Dąbrowska, T. 2003. Przeworsk-Kultur, § 1: Jüngere vorrömische Eisenzeit – a. Definition und Entstehung – b. Datierung – c. Bedeutung – § 2: Frühe Römische Kaiserzeit. In H. Beck, D. Geuenich, H. Steier (eds.) Reallexikon der Germanischen Alter­ tumskunde 23, 540–53. Berlin, New York: Walter de Gruyter. Dąbrowska, T. 2008. Młodszy okres przedrzymski na Mazowszu i zachodnim Podlasiu, Materiały Starożytne i Wczesnośredniowieczne 7.

194

Tomasz Bochnak

Domaradzki, M. 1977. Tarcze z okuciami metalowymi na terenie Celtyki Wschodniej, Przegląd Archeologiczny 25, 53–95. Dulęba, P. 2009. Przemiany kulturowe w zachodniej małopolscy w okresie od III do I wieku przed Chr. Przyczynek do kontaktów między Celtami a Germanami. In E. Droberjar and M. Karwowski (eds.) Archeologia Barbarzyńców 2008: powiązania i kontakty w świecie barbarzyńskim, Materiały i kontakty z IV Protohistorycznej Konferencji Sanok, 13– 17 października 2008, Archaeologie der Barbaren 2008: Beziehungen und Kontakte in der barbarischen Welt, Materialen aus der IV. Frühgeschichtlichen Konferenz in Sanok, 13.–17. Oktober 2008, 11–36. Rzeszów: Mitel. Filip, J. 1956. Keltové ve středni Evropě. Monumenta Archaeologica: Acta praehistorica, protohistorica et historica Instituti Archaeologici Academiae Scientiarum Bohemoslovenicae 5, Monumenta archaeologica 5. Praha: Nákladatelství Československé akademie věd. Gedl, M. 1978. Gräber der Latènekultur in Kietrz, Bezirk Opole. In M. Gedl (ed.) Beiträge zum Randbereich der Latènekultur, Prace Archeologiczne 28, 9–72. Kraków: Warszawa, Państwowe Wydawnictwo Naukowe. Godłowski, K. 1983. Przemiany osadnicze na Dolnym Śląsku i sąsiadujących z nim terenach w I w. p.n.e. In W. Hensel (ed.) Przemiany ludnościowe i kulturowe I tysiąclecia p.n.e. na ziemiach między Odrą i Dnieprem, Materiały z polskoradzieckiego sympozjum paleodemograficznego Warszawa, 6–9 grudnia 1977, 293–323. Wrocław-Warszawa-Kraków-GdańskŁódź: Zakład Narodowy imienia Ossolińskich, Wydawnictwo Polskiej Akademii Nauk. Guštin, M. 1984. Die Kelten in Jugoslavien. Übersicht über das archäologische Fundgut, Jahrbuch des Römisch-Germanischen Zentralmuseums Mainz 34, 305–65. Guštin, M. 1991. Posočje. Posočje in der jüngeren Eisenzeit, Catalogi et monographiae 27. Ljubljana: Narodni Muzej Ljubljana. Guštin, M. 2002. I Celti dell’Adriatico. Carni tra fonti storiche e archeologia. In Gli echi della Terra. Presenze celtiche in Friuli: dati materiali e momenti dell’immaginario, 11–20. Pisa: Giardini. Hachmann, R. 1957. Ostgermanische Funde der Spätlatènezeit in Mittel- und Westdeutschland. Ein Beitrag zum Problem des Nachweises von Bevölkerungsbewegungen auf Grund des urgeschichtlichen Grundstoffs, Archaeologia Geographica 6, 55–68. Haffner, A. 1978. Das keltisch-römische Gräberfeld von WederathBelginum. 3. Teil: Gräber 884–1260, Ausgegraben 1958–1960, 1971 u. 1974. Mainz: Philipp von Zabern. Haffner, A. 1989a. Das Gräberfeld von Wederath-Belginum vom 4. Jahrhundert vor bis zum 4. Jahrhundert nach Christi Geburt. In A. Haffner (ed.) Gräber – Spiegel des Lebens. Zum Totenbrauchtum der Kelten und Römer am Beispiel des Treverer-Gräberfeldes Wederath-Belginum, 37–128. Mainz: Philipp von Zabern. Haffner, A. 1989b. Das spätlatènezeitliche Kriegergrab 1178 mit Feinwaage. In A. Haffner (ed.) Gräber – Spiegel des Lebens. Zum Totenbrauchtum der Kelten und Römer am Beispiel des Treverer-Gräberfeldes Wederath-Belginum, 173–86. Mainz: Philipp von Zabern.

Hoffmann, W. 1940. Neue Keltenfunde aus Mittelschlesien, Altschlesien 9, 112–26. Hunyady, I. 1942. Die Kelten im Karpatenbecken. Tafelband, Kelták a Kárpátmedencében. Táblakötet, Dissertationes Pannonicae II.18. Budapest: Királyi Magyar Pázmány Péter Tudományegyetem Éremés Régiségtani Intézete. Hunyady, I. 1944. Die Kelten im Karpatenbecken. Textband, Kelták a Kárpátmedencében. Szövegkötet, Dissertationes Pannonicae II.18. Budapest: Királyi Magyar Pázmány Péter Tudományegyetem Éremés Régiségtani Intézete. Jahn, M. 1924. Die Gliederung der Wandalischen Kultur in Schlesien, Schlesiens Vorzeit in Bild und Schrift, Neue Folge 8, 20–36. Jahn, M. 1931. Die Kelten in Schlesien, Quellenschriften zur ostdeutschen Vor- und Frühgeschichte 1. Leipzig: Curt Kabitzsch. Jahn, M. 1934. Neue skythische und keltische Funde, Altschlesien 4 (1932), 112–26. Jażdżewska, M. 1988. Najciekawsze obiekty na stanowisku kultury przeworskiej w Siemiechowie nad górna Wartą, Prace i Materiały Muzeum Archeologicznego i Etnograficznego w Łodzi, seria archeologiczna 32 (1985), 109–42. Jażdżewska, M. 1992. Cimetière de La Tène III et de la période romaine à Siemiechów (suite), Inventaria Archaeologica 54, Pl. 387: 1–3. Jażdżewska, M. 1994a. Ciekawsze zespoły grobowe ze śladami wpływów celtyckich na cmentarzysku kultury przeworskiej w Siemiechowie, woj. Sieradzkie. In J. Gurba and A. Kokowski (eds.) Kultura przeworska I, 107–26. Lublin: Wydawnictwo Uniwersytetu Marii Curie-Skłodowskiej. Jażdżewska, M. 1994b. Nowe spojrzenie na hełm z Siemiechowa, Łódzkie Sprawozdania Archeologiczne 1/1994, 63–74. Kaczanowski, P. 1992a. Bemerkungen zur Chronologie des Zustroms römischer Waffenimporte in das europäische Barbaricum. In K. Godłowski and R. Madyda-Legutko (eds.) Probleme der relativen und absoluten Chronologie ab Latènezeit bis zum Frühmittelalter, Materialien des III. Internationalen Symposiums: Grundprobleme der frühgeschichtlichen Entwicklung im nördlichen Mitteldonaugebiet, KrakówKarniowice 3.–7. December 1990, 171–86. Kraków: Secesja. Kaczanowski, P. 1992b. Importy broni rzymskiej na obszarze europejskiego Barbaricum. Kraków: Uniwersytet Jagielloński. Krämer, W. 1962. Manching II. Zu den Ausgrabungen in den Jahre 1957 bis 1961, Germania 40, 293–317. Landry, Ch. and Blaizot, F. 2011. Une sépulture de guerrier celte à Chens-sur-Léman (Haute Savoie), Revue Archéologique de l’Est 60, 147–71. Łuczkiewicz, P. 2000. Zur späteisenzeitlichen Bewaffnung in Polen. Stand der Forschung, Jahrbuch des RömischGermanischen Zentralmuseums Mainz 47, 355–437. Łuczkiewicz, P. 2006. Uzbrojenie ludności zim Polski w młodszym okresie przedrzymskim. Lublin: Instytut Archeologii Uniwersytetu Marii Curie-Skłodowskiej, Standruk. Maier, R. A. 1985. Ein römerzeitlicher Brandopferplatz bei Schwangau und andere Zeugnisse einheimischer Religion in der Provinz Rätien. In J. Bellot, W. Czysz and G. Krane (eds.) Forschungen zur provinzionalrömische Archäologie in Bayerisch-Schwaben, Schwäbische Geschichtsquellen und

22.  La Tène and Przeworsk strap shield bosses from Poland Forschungen 14, 231–51. Augsburg: Historischer Verein für Schwaben. Makiewicz, T. 1973. W kwestii osadnictwa celtyckiego na Kujawach, Archeologia Polski 18, 213–18. Makiewicz, T. 1986. Znaleziska z kręgu kultury celtyckiej i trackiej na Kujawach. In Celtowie i ich mennictwo, Materiały z sesji naukowej, Warszawa, 26 kwietnia 1985, 25–33. Warszawa: Polskie Towarzystwo Archeologiczne i Numizmatyczne. Meduna, J. 1962. Laténské žarové hroby na Moravé, Památky Archeologické 53/1, 87–136. Megaw, J. V. S. 1970. Art of the European Iron Age: A study of the elusive image. New York, Evanston: Harper & Row. Megaw, J. V. S. 2005. Notes on Two Belt-Plates of Early La Tene Type from Northern Poland, Pomorania Antiqua 20, 257–76. Megaw, J. V. S. and Megaw, M. R. 2012. Poles Apart? Notes from the Fringes of the ‘Hungarian’ Sword Style, In P. Anreiter, E. Bánffy, L. Bartosiewicz, W. Meid and C. Metzner-Nebelsick (eds.) Archaeological, Cultural and Linguistic Heritage. Festschrift for Erzsébet Jerem In Honour of Her 70th Birthday, 401–414. Budapest: Archaeolingua Alapítvány. Megaw, R. and Megaw, J. V. S. 1990. “Semper aliquod novum...” Celtic dragon-pairs re-reviewed, Acta Archaeologica Academiae Scientiarum Hungaricae 12, 49–72. Megaw, R. and Megaw, J. V. S. 2001. Celtic Art. From its beginnings to the Book of Kells. (Revised and expanded edition). London: Thames & Hudson. Moczulski, L. 2007. Narodziny Międzymorza. Ukształtowanie ojczyzn, powstanie państw oraz układy geopolityczne wschodniej części Europy w późnej starożytności i we wczesnym średniowieczu. Warszawa: Bellona. Oesterwind B. C. 1989. Die Spätlatènezeit und die frühe römische Kaiserzeit im Neuwider Becken, Bonner Hefte zur Vorgeschichte 24. Bonn: Institut für Vor- und Frühgeschichte. Orliński, W. 2011. Cmentarzysko kultury przeworskiej w Legionowie – omówienie dotychczasowych wyników badań, Rocznik Legionowski 4, 7–18. Pernet, L. 2010. Armement et auxiliares gaulois (IIe et Ier siècle avant notre ère). Protohistoire européenne 12. Montagnac: Éditions Monique Mergoil. Perrin, F. and Schönfelder, M. 2002. La tombe à char à quatre roues de Verna (Isère). Contribution à l’étude des élites à l’époque de La Tène finale. Documents d’Archéologie en RhôneAlpes 24. Lyon: Association Lyonnaise pour la Promotion de l’Archéologie en Rhône-Alpes. Peschel, K. 1977. Frühe Waffengräber im Gebiet der südlichen Elbgermanen. In B. Chropovský (ed.) Ausklang der LatèneZivilisation und Anfänge der germanischen Besiedlung im Mittleren Donaugebiet, 261–81. Bratislava: VEDA, Vyd. Slovenskej Akad. Vied. Peschel, K. 1978. Anfänge germanischer Besiedlung im Mittelgebirgsraum. Sueben-Hermunduren-Markomanen. Arbeits- und Forschungsberichte zur sächsischen

195

Bodendenkmalpflege 12. Berlin: VEB Deutscher Verlag der Wissenschaften. Pič, J. L. 1902. Starožitnosti země české 2. Praha: Národní muzeum v Praze. Pieta, K. 2005. Spätlatènezeitliche Waffen und Ausrüstung im nördlichen Teil des Karpatenbeckens, Slovenská Archeológia 53–1, 35–84. Pittioni, R. 1954. Urgeschichte des österreichischen Raumes. Wien: F. Deuticke. Polenz, H. 1971. Mittel- und spätlatènezeitliche Brandgräber aus Dietzenbach, Landkreis Offenbach am Main, Studien und Forschungen, Neue Folge 4, 1–115. Poleska, P. 2006. Celtycki mikroregion osadniczy w rejonie podkrakowskim. Biblioteka Muzeum Archeologicznego w Krakowie 2. Kraków: Muzeum Archeologiczne w Krakowie. Prochàzka, A. 1937. Gallská kultura na Vyškovsku, (La Tène středomoravský). Slavkov u Brna. Rapin, A. 2001. Un bouclier celtique dans la colonie grecque de Camarina (Sicile), Germania 79, 273–96. Spagnolo Garzoli, G. (ed.) 1999. Conubia Gentum. La necropoli di Oleggio e la romanizzazione dei Vertamocori: Oleggio, Palazzo Bellini, 23 gennaio–30 aprile 1999. Torino: Omega edizioni. Stary, P. F. 1981. Ursprung und Ausbreitung der eisenzeitlichen Ovalschilde mit spindelförmigem Schildbuckel, Germania 59, 287–306. Stead, I. M. 1969. An Iron Age Hill-fort at Grimthorpe, Yorkshire, England, Proceedings of the Prehistoric Society 34 (1968), 148–90. Stöckli, W. E. 1979. Die Grob- und Importkeramik von Manching, Die Ausgrabungen in Manching 8. Stuttgart: Franz Steiner Verlag. Szabó, M. and Tankó, K. 2012. La nécropole laténienne à Ludas – Varjú-dűlő. In M. Szabó (ed.) La nécropole laténienne à Ludas – Varjú-dűlő, 9–152. Budapest: L’Harmattan. Todorović, J. 1966. Klasifikacija keltskih štitova otkrivenih u Donjoj Panoniji i na Balkanskom Poluostrvu, Vesnik Vojnog Muzeja JNA 11/12, 361–74. Venclová, N. (ed.) 2008. Archeologie pravěkých Čech 7, Doba laténská. Praha: Archeologický ústav AV ČR. Vouga, P. 1923. La Tène. Monographie de la station publiée au nom de la Commission des Fouilles de La Tène. Leipzig: Karl. W. Hiersemann. Waldhauser, J. 1987. Keltische Gräberfeldern in Böhmen, Bericht der Römisch-Germanischen Kommission 68, 25–179. Woźniak, Z. 1970. Osadnictwo celtyckie w Polsce. WrocławWarszawa-Kraków: Zakład Narodowy imienia Ossolińskich, Wydawnictwo Polskiej Akademii Nauk. Woźniak, Z. 1991. La nécropole d’Iwanowice et les Celtes en Pologne. In S. Moscati (ed.) Les Celtes, 378. Milan: Bompiani. Zielonka, B. 1955. Piotrków Kujawski, Dawna Kultura 2, 166. Zielonka, B. 1956. Cmentarzysko w Piotrkowie Kujawskim, Wiadomości Archeologiczne 23, 275–6.

23 DE L’ANNEAU EN BRONZE À TÊTES DE BÉLIERS DE CHERMIGNAC (CHARENTE-MARITIME) ET DE QUELQUES PIÈCES DE HARNAIS. LA TÈNE FINALE DE GAULE DE L’OUEST José Gomez de Soto

L’anneau en bronze par lequel nous initions notre propos a été recueilli en surface du sol lors d’une prospection pédestre au lieu-dit La Fenêtres, commune de Chermignac, en Charente-Maritime, France, par un chercheur bénévole qui le conserve actuellement. Le lieu de la découverte est un site de la période impériale romaine. Une ou plusieurs occupations antérieures sont attestées par la trouvaille de quelques tessons protohistoriques très érodés, sans élément de forme datable: on ne peut affirmer avec certitude de présence laténienne sur ce site, d’autant plus qu’aucune fouille ni sondage n’y ont été pratiqués, qui eussent permis d’apporter une réponse. La découverte du site a été déclarée au Service régional de l’Archéologie de Poitou-Charentes (n° DRACAR 10680). L’objet est affecté d’une forte usure, ce qui traduit un port prolongé, et probablement un abandon tardif. A défaut de pouvoir le rattacher à une incertaine occupation du site au cours de la période laténienne, l’hypothèse d’une longue utilisation jusqu’à la période romaine, qui justifierait cette usure, serait recevable. Une autre hypothèse tout aussi vraisemblable serait celle de la récupération d’une trouvaille, alors devenue objet de curiosité. Nous savons que les Anciens ne furent pas étrangers au goût de la collection (Schnapp 1993, 25 sq.).

L’anneau de Chermignac (Fig. 23.1 et 23.2) L’anneau, en bronze à patine verte, possède un jonc à section circulaire, maintenant très légèrement ovalisé. Son décor plastique comporte six reliefs, en alternance des têtes de béliers et des nodosités transversales oblongues. Ces

dernières conservent de discrètes traces d’un décor de stries. Sur les têtes de béliers, la trop forte usure a effacé tout détail incisé ou en léger relief qui eût pu exister. Le diamètre externe de l’anneau, mesuré en dehors des reliefs, varie de 2,5 à 2,6 cm, et, au niveau des reliefs céphaliques, de 3,1 à 3,3 cm; son diamètre interne passe de 1,65 à 1,9 cm. L’anneau de Chermignac est, à notre connaissance, le seul anneau à décor plastique portant des têtes de béliers actuellement connu en Gaule de l’Ouest et plus généralement, en Gaule. Il se rattache au type 1 du classement proposé par C. Tappert pour les anneaux laténiens portant des représentations zoomorphes: ceux à trois têtes de bélier reposant directement sur le corps de l’anneau (Tappert 1998, 178). De tels petits anneaux à décor plastique, ornés de simples globules, de nodosités ou de protomés d’animaux sont bien connus dans la Celtique d’Europe centrale pendant la période des oppida (Fig. 23.2, 2), en Bohême et Moravie (Píč 1903, pl. XI, XVI; Filip 1956, pl. CIV, CXXVII, CXXX; Čižmař 2011), en Bavière (van Endert 1991, taf. 5), etc. Mais on en trouve largement plus loin en Occident, et ce jusqu’en Gaule de l’Ouest. Pour ne citer que quelques exemples: en Palatinat, un anneau orné de globules et de figures animales du Donnersberg (Zeeb-Lanz 2008, 61); en Suisse, le remarquable anneau de Port, BE, orné de globules, d’un oiseau et de protomés de bovidés aux cornes bouletées, ou, de plus grands diamètres, ceux des tombes 62 et 248 de Giubiasco, TI (Wyss 1974, fig. 14, 19). En Gaule, outre des anneaux simplement ornés de globules comme celui de Forsfeld en forêt de Haguenau, Bas-Rhin (Schaeffer 1930, fig. 151), on ne mentionne, en

23.  De l’anneau en bronze à têtes de béliers de Chermignac sus de celui de Chermignac, que huit autres anneaux à motifs zoormorphes (Tappert 1998, 204), tels celui à têtes de bovidés aux cornes bouletées et globules tiré de la Saône à Chalon-Lux, Saône-et-Loire (Guillaumet 1983, 34) (Fig. 23.2, 1) ou les deux à décor ornithomorphe de Larina, Isère (Pelatan 1986, pl. XLVI) (Fig. 23.2, 3), etc. En Gaule de l’Ouest, avec une dizaine d’anneaux à décor de simples globules récoltés au cours de prospections de surface, le site de Lacoste à Mouliets-et-Villemartin, Gironde (Sireix et al. 1985, 17; Boudet 1987, 118), a peu à envier à bien des établissements de la Celtique centre-européenne: leur nombre paraît, par exemple, assez proche de celui récolté sur l’oppidum de Stradonitz, du moins au vu de l’ouvrage de J. L. Píč (1903). Un des anneaux de Lacoste, celui au plus grand diamètre, très usé, porte des globules et d’autres reliefs relativement volumineux très usés, peut-être eux aussi des motifs zoomorphes maintenant non identifiables (Fig. 23.4). Du même département que Lacoste, un autre anneau, lui aussi à simple décor de globules, est encore signalé, trouvé dans un contexte postérieur de l’époque impériale romaine, aux Murasses à Lugasson. En Europe centrale, les pièces à décor zoomorphe sont nettement minoritaires parmi les anneaux à décor plastique de la période des oppida: par exemple, pour la Moravie, M. Čižmař (2011) n’en recense que quatre à têtes de béliers, dont un, fragment d’un anneau de plus grand diamètre que les deux précédents, est peut-être plutôt un débris de bracelet. Encore en Moravie, le site de Nĕmčice-Vícemĕřice, daté il est vrai de LT C, pourtant riche en figurines zoomorphes, n’a pas livré de figurations de béliers, et ses anneaux en bronze ne s’ornent que de globules (Čižmař et al. 2008).

197

Fig. 23.1. Anneau en bronze de Chermignac (photo. J. Gomez de Soto)

Fig. 23.2. Anneau en bronze de Chermignac (dessin M. Coutureau, Inrap).

Fig. 23.3. Anneaux en bronze à figures zoomorphes. 1, Chalon-Lux, Saône-et-Loire (photo Lacoste, in Guillaumet 1983) ; 2, Malhostovice, Moravie (d’après Čižmař 2011) ; 3, Larina, Isère (d’après Pelatan, 1986).

José Gomez de Soto

198

Fig. 23 4. Anneaux en bronze de Lacoste à Mouliets-et-Villemartin, Gironde (photo. J. Gomez de Soto).

Béliers et monstres hybrides Le bélier est un animal par nature lié au monde chthonien. Il est inutile de discuter ici plus avant de sa symbolique, ni de la véritable nature du masque humain qui lui est parfois associé et que divers auteurs interprètent comme celui d’une grande divinité (Kruta 1992 et 2000, 460; Desenne et al. 2005, 254 sq.; Cluytens 2009; Charpy 2009). Nous nous bornerons à un rapide survol de la présence du motif dans l’art laténien. Les images du bélier, rarement complètes, parfois réduites aux cornes mais le plus souvent à la seule tête, naturalistes ou schématisées, sont souvent associées à d’autres images zoomorphes voire anthropomorphes pour créer des monstres. Ils peuvent aussi accoster l’arbre de vie dans certaines compositions. Ces images sont présentes dans la production plastique celtique dès la fin du premier âge du Fer, mais surtout à partir du début de la culture de La Tène (Verger 1991, tabl. IV), et ce jusque pendant la période impériale romaine (Cluytens 2009). Pour La Tène ancienne, par exemple, sur quelques rares fibules telles en France celles d’Aignay-le-Duc, Côte-d’Or, de Port-à-Binson et Suippes, Marne ou de Dompierre-les-Tilleuls, Doubs, en Autriche du cimetière de Hallein ou encore en Allemagne de Berlin-Niederschönhausen (Binding 1993). Sur la fibule de Droužkovice en Bohême, la tête de bélier se trouve tenue dans sa gueule par un fauve, mais sur une de Dompierreles-Tilleuls, par une sorte d’inversion des rôles, c’est le bélier qui tient dans sa bouche une tête humaine (ibid.). Cette image du bélier sur les fibules demeure rare: moins de 2% de celles à têtes animales, comme le fait remarquer

M. Cluytens (2009, 205), et les autres types de supports portant des figurations de bélier, comme le bracelet en or de Rodenbach (Jacobsthal 1944, no. 59; Megaw et Megaw 2001, no. 114), n’abondent guère. Pendant La Tène ancienne, la tête du bélier – ou uniquement ses cornes – peut encore participer de l’élaboration de monstres hybrides, tel le serpent à tête de carnassier cornu de la paragnathide du casque de la grotte des Perrats à Agris en Charente (Gomez de Soto et Verger 1999 et 2010), se substituer au corps humain sur l’anse de l’enochoé de la tombe de Reinheim, sur laquelle le masque humain se superpose directement à la tête de bélier (Keller 1965, pl. 20–24; Echt 1999, fig. 75–76, pl. 22) selon un modèle iconographique qu’on retrouvera bien plus tard à La Tène finale sur des garnitures de harnachement de Manerbio sul Mella en Lombardie (Jacobsthal 1944, no. 84), ou encore se combiner au masque humain à Erstfeld (Wyss 1975, fig. 11; Guggisberg 2000). Les sourcils démesurément prolongés avec terminaisons spiralées, si nombreux avec diverses variantes de La Tène A jusqu’à des dates tardives comme sur la tête bifrons de l’épée à poignée anthropoïde de Saint-André-de-Lidon en Charente-Maritime (Duval et al. 1986) pourraient aussi correspondre – ou correspondent effectivement? – à une fusion entre le masque humain et l’image du bélier. Sur la plaque ajourée réutilisée en arc de fibule de la tombe 3 de la Croyère à Orainville dans l’Aisne, le chagrinage du front des masques, lu comme la figuration de la toison du bélier, manifesterait de façon particulièrement significative cette fusion (Desenne et al. 2005, 254 sq.; Charpy 2009).

23.  De l’anneau en bronze à têtes de béliers de Chermignac

199

Fig. 23.6. Plaque d’applique. Lacoste à Mouliets-et-Villemartin, Gironde (dessin J. Hiernard, in Hiernard 1999, 116).

Fig. 23.5. Pendentifs de harnais de tête. Saintes, Charente-Maritime (dessin des archives de la Société des Antiquaires de l’Ouest à Poitiers, d’après Hiernard 1999, 106).

Fig. 23.7. Barre de suspension de disque de pendentif de harnais de tête et pièce cruciforme à bélières probablement pour harnais. Lacoste à Mouliets-et-Villemartin, Gironde (photo. J. Gomez de Soto).

M. Cluytens (2009, 205) associe l’image du bélier à celle de la paire de dragons, et souligne que la première se raréfie au moment où justement, la seconde se diffuse sur les fourreaux d’épées. Pendant la période de La Tène finale – celle de l’anneau de Chermignac – les images de têtes de béliers demeurent proportionnellement, parmi les figurations anthropomorphes et zoomorphes de toutes natures, en métal et autres, peu nombreuses voire dans certaines régions inexistantes (Megaw et Megaw 2001, 160): pour la Moravie, pour laquelle nous disposons d’un inventaire complet, elles n’apparaissent que dans cinq cas sur un total de vingt-deux (Čižmař 2011). Mais si, pour l’ensemble de l’Europe, on se

limite aux seuls anneaux portant des figurations zoomorphes, déjà minoritaires par rapport aux anneaux à décor bouleté, la proportion de ceux à têtes de béliers passe à près d’un sur deux (Tappert 1998). En Gaule, outre l’anneau de Chermignac, quelques rares autres figurations zoomorphes ornent d’autres objets. Le motif de la tête de bélier n’y est pas inconnu: par exemple, il termine les branches supérieures de la poignée anthropoïde de l’épée d’une tombe à incinération de la nécropole de Mouriès dans les Bouches-du-Rhône (Déchelette 1914, 1140; Marcadal et al. 2003; Guillaumet et Rapin 2000, 80). Pour citer quelques exemples parmi d’autres hors de

200

José Gomez de Soto

Gaule, mentionnons l’embout en bronze de corne à boire orné d’une tête humaine surmontée d’une tête de bélier du Donnersberg (Zeeb-Lanz 2008, 67 sq.), ou, la figurine d’animal d’identification incertaine attribuée par P. Vouga à la réoccupation tardive de La Tène (Vouga 1923, 121–122, pl. L 23). Comme pendant La Tène ancienne, pendant La Tène finale les cornes du bélier peuvent se combiner avec d’autres représentations d’animaux réels ou mythiques pour élaborer des monstres hybrides: sur les landiers en fer du Montéqueux à Beine dans la Marne, elles affublent des têtes de griffons, déjà inhabituelles sur de tels instruments du foyer (Charpy 2007); sur le torque de Frasne-les-Buissenal dans le Hainaut, elles se combinent à un protomé de bovidé (Jacobsthal 1944, no. 70; Hautenauve 2005, 194; Ginoux 2006, 144); on n’oubliera évidemment pas les serpents cornus du chaudron de Gundestrup.

Un anneau produit en Occident ou en Celtique d’Europe centrale ? A propos de quelques éléments de harnachement équestre de Gaule de l’Ouest. L’anneau de Chermignac est, nous l’avons vu ci-dessus, très comparable à divers autres d’Europe centrale. Pour cet exemplaire de Gaule de l’Ouest se pose la question de son origine: production centre-européenne, ou au contraire, occidentale? L’hypothèse de l’importation depuis la Celtique orientale a déjà été émise pour quelques objets de parure trouvés en Gaule de l’Ouest, par exemple pour divers objets en bronze de Lacoste, comparés à des homologues d’Europe centrale, de Bohême plus particulièrement (Boudet 1987, 117), auxquels pour le même site pourrait s’ajouter un bracelet à oves à décor plastique et masques anthropomorphes affrontés s’ouvrant par un fermoir amovible connu depuis peu (Sireix 2012, 38). On pourrait encore mentionner dans cette catégorie de parures peu courantes en Gaule de l’Ouest la paire d’anneaux de cheville à oves creux non décorés et fermoir mobile inédite présumée venir de la région de Niort dans les Deux-Sèvres (étude en cours par J. Gomez de Soto, A. Lehoërff et A. Masse). Semblable hypothèse d’une origine centre-européenne fut également émise par J. Hiernard et D. Simon-Hiernard à propos d’une paire de pendentifs de harnais équestre interprétés comme frontaux (mais en fait, des pendeloques latérales de tête, comme le montre la disposition d’une barre de suspension conservée en place sur une têtière métallique de La Tène: Vouga1923, pl. XXXVII 1), découverts associés à une broche en fer à Saintes en Charente-Maritime au XIXe siècle, maintenant malheureusement perdus mais connus par un très précis dessin de l’époque de leur découverte (Hiernard et Simon-Hiernard 1999, 107 sq. et 2000). L’un était complet de son disque ajouré de motifs à contours

curvilignes et de sa barre de suspension, l’autre réduit à sa barre de suspension (Fig. 23.5). La même hypothèse d’une origine centre-européenne est encore posée par les mêmes auteurs pour une plaque de bronze à décor ajouré de Lacoste (Hiernard et Simon-Hiernard 1999, 116) (Fig. 23.6), site dont proviennent encore une barre de suspension de pendeloque de harnais proche de celles de Saintes ou de la têtière de La Tène (Sireix et al. 1985, fig. 7) (Fig. 23.7, 1), ainsi qu’une pièce cruciforme à bélières probablement également destinée à un harnais (Fig. 23.7, 2), qui trouve un parallèle à Stradonitz (Píč 1903, pl. XXI 28). L’origine lointaine de ces différents objets est évidemment une hypothèse recevable, mais celle d’une production occidentale le paraît tout autant: la grande similitude des cultures matérielles d’Europe centrale et de Gaule de l’Est a de longue date été soulignée par J. Déchelette (1901 et 1914), qui confrontait les mobiliers de Stradonitz et de Bibracte, ce qui suppose non seulement des échanges à longue distance, mais aussi des centres de production largement répartis en Europe moyenne. Les observations de J. Déchelette, émises à une époque où la Gaule de l’Ouest était quasiment terra incognita, peuvent aussi s’appliquer, nous le savons maintenant, à l’Occident du monde celtique continental, pour lequel des productions très standardisées, entre autres dans l’armement et les manifestations artistiques, traduisent dès le Ve siècle la mise en place précoce de la koinè celtique. Pour en revenir aux pendentifs de Saintes, nous soulignerons que les différents exemplaires dont les disques sont connus offrent des différences notables dans leur ornementation, et que le caractère transeuropéen du style curviligne utilisé pour leurs ajours, comme d’ailleurs l’usage de banals rivets à tête striée, plutôt qu’inviter à rechercher un atelier unique ou une région de production limitée, suggèrent au contraire la probabilité de plusieurs: la répartition de ces objets de la Hongrie à l’Atlantique en passant probablement par la Champagne (Stead et Rigby 1999, pl. 186) n’autorise pas à privilégier une région davantage qu’une autre et ces ornements apparaissent de fait trans-celtiques. Le disque de Saintes, de qualité un peu maladroite sous sa flatteuse apparence, peut difficilement être attribué, par exemple, aux mêmes ateliers qui produisirent les beaux exemplaires presque parfaits de Skyjle en Tchéquie ou de Niedenstein en Hesse. Surtout, il porte en bordure un décor gravé ou estampé, un enchaînement d’arceaux et d’ocelles qui ne figure sur aucun des autres exemplaires connus. Ce motif est clairement inspiré des décors estampés communs sur les céramiques de Gaule du Nord-Ouest, d’Armorique en particulier (Daire 1992), décors dont on connaît aussi quelques exemples en Centre-Ouest dès le Ve siècle, et jusqu’au 1er siècle avant J.-C. (Gomez de Soto et al. 2007). Ce décor est d’ailleurs recoupé par un des ajours, celui maladroitement dépourvu de symétrique, ce qui prouverait qu’il fut tracé en premier et non rajouté postérieurement sur un objet de provenance lointaine qu’on eût voulu adapter

23.  De l’anneau en bronze à têtes de béliers de Chermignac au goût local: qu’il s’agisse d’une production occidentale est indiscutablement l’hypothèse la plus probable. De même pour les pièces de Lacoste mentionnées ci-dessus, une production par les ateliers de bronziers du site (Sireix 2012, 56) serait tout aussi plausible. Quant à la plaque ajourée de Lacoste, elle porte un décor périphérique d’ocelles, motif courant sur les céramiques de La Tène finale de Gaule du Nord-Ouest, indice pour elle aussi d’une fabrication occidentale. Bien qu’il reste encore isolé en Extrême-Occident, que l’anneau orné de têtes de béliers trouvé à Chermignac soit le produit d’un atelier occidental et non une pièce d’importation lointaine est l’hypothèse que nous retiendrons. La variété des types d’objets métalliques en tout genre livrés par un site comme celui de Lacoste (Sireix et al. 1985; Sireix 2012) et les apports de l’archéologie préventive, et ce malgré l’extrême rareté des contextes funéraires, rendent progressivement et radicalement obsolète le décalage entre la quantité et la qualité du matériel disponible entre la Celtique de l’Est et celle d’Occident, en affirmant régulièrement les similitudes. Désormais, plus n’est besoin de démontrer que la Gaule de l’Ouest participait activement de la koinè celtique, et ce dès la période de formation de la culture laténienne (Gomez de Soto 2005; Milcent 2006; Gomez de Soto et al. 2007).

Remerciements Nous remercions nos collègues Christophe Lassalle, inventeur et propriétaire de l’anneau de Chermignac, qui nous l’a confié et nous en a permis la présentation, et Christophe Sireix, qui nous a permis d’examiner les découvertes de son père à Lugasson et signalé l’anneau inédit des Murasses.

English translation

Introduction This bronze ring was recovered from the ground surface during fieldwalking in the area known as La Fenêtres, in the commune of Chermignac, in Charente-Maritime, France, by a volunteer who remains its current owner. The find location is on a site dating to the Roman Empire period. Evidence for one or more earlier occupation phases was determined by the discovery of several badly-worn sherds of prehistoric pottery, lacking in dateable diagnostic features: it is not possible to confirm the presence of La Tène activity at the site, especially since no surveys or excavations have been

201

carried out which might provide an answer. The discovery of the site was reported to the Poitou-Charentes Regional Archaeological Service (DRACAR no. 10680). The object was heavily abraded, which reflects wear over a prolonged period, and probably late discard. Without being able to relate it to the unproven occupation of the site during the La Tène period, it is still possible to hypothesize its long-term use into the Roman period to explain the level of wear on the ring. An equally plausible alternative hypothesis would be that it was a ‘found’ object which became a curiosity. We know that people in the past were no strangers to a penchant for collecting (Schnapp 1993, 25ff.).

The Chermignac Ring (Fig. 23.1 & 23.2) The ring, bronze with green patina, has a circular cross section, now slightly oval. Its decoration consists of six plastic reliefs alternating rams heads and transverse oblong nodules. The latter retain traces of barely-visible decorative lines. On the rams heads, the heavy wear has erased any details of potential incised or low relief. The outer diameter of the ring, measured between the relief decoration, ranges from 2.5 to 2.6 cm, and, on the animal head reliefs, from 3.1 to 3.3 cm; its inner diameter ranges from 1.65 to 1.9 cm. The Chermignac ring is, to our knowledge, the only ring decorated with ram heads in plastic style currently known from western Gaul in particular, and Gaul as a whole. It belongs to the type 1 classification proposed by C. Tappert for La Tène rings with zoomorphic decoration: rings with three ram heads placed directly on the ring (Tappert 1998, 178). Such small plastic-style decorated rings, decorated with simple beads, nodules or animal protomes, are well known in Celtic central Europe of the oppida period (Fig. 23.2, 2), in Bohemia and Moravia (Píč 1903 pl XI, XVI. Filip 1956 pl CIV, CXXVII, CXXX; Čižmár 2011), Bavaria (van Enderttal 1991, taf 5), etc. But they are distributed more widely in the west, and as far as western Gaul. To cite a few examples: in the Palatinate, a ring decorated with beads and animal figures from Donnersberg (Zeeb-Lanz 2008, 61); from Switzerland, the remarkable ring of Port, Berne, adorned with beads, a bird, and protomes of cattle with bouletee horns, or, with larger diameters, those from graves 62 and 248 from Giubiasco, Ticino canton (Wyss 1974, fig. 14, 19). In Gaul, in addition to rings plainly decorated with beads, like that from Forsfeld, Forêt de Haguenau, Bas-Rhin (Schaeffer 1930, fig. 151), only eight other zoomorphic rings, in addition to the Chermignac ring, have been recorded (Tappert 1998, 204), such as the ring with boulettee horned cattle heads and beads which comes from the Saône at Chalon-Lux, Saône-et-Loire (Guillaumet 1983, 34) (Fig.

202

José Gomez de Soto

23.2, 1), or the two with bird-shaped ornamentation from Larina, Isère (Pelatan 1986, pl. XLVI) (Fig. 23.2, 3). In western Gaul, the Lacoste site at Mouliets-et-Villemartin, Gironde (Sireix et al 1985, fig.17; Boudet 1987, 118), with ten rings decorated with simple spheres, retrieved during fieldwalking surveys, is equal to any site in Celtic central Europe: the number of rings seems to parallel, for example, retrieval rates from the oppidum at Stradonitz, according to the work of J. L. Píč at least (1903). One of the Lacoste rings, with the largest diameter, very worn, consists of beads and other relatively large reliefs, heavily abraded, and perhaps representing now unidentifiable zoomorphic motifs (Fig. 23.4). From the same department as Lacoste, another ring, also plainly decorated with beads, has been reported, found in a post-Roman context at Les Murasses at Lugasson. In central Europe, objects with zoomorphic decoration are in a distinct minority among plastic decorated rings of the oppida period: for example, for Moravia, M. Čižmár (2011) only identifies four with rams heads, including one, a fragment of a ring of larger diameter than the previous two, which is perhaps more likely to be part of a bracelet. Still in Moravia, the site of Nemčice Vícemĕřice, dated it is true to LT C, is rich in zoomorphic figurines, but has not yielded any representations of rams, and its bronze rings are only adorned with beads (Čižmár et al. 2008).

Rams and hybrid monsters The ram is an animal naturally related to the chthonic world. It is useless to discuss any further here its symbolism, or the true nature of the human face with which it is often associated, and which many scholars interpret as the mask of a powerful deity (Kruta 1992 and 2000, 460; Desenne et al 2005, 254 ff.; Cluytens 2009; Charpy 2009). We confine ourselves to a brief overview of the motif’s presence in La Tène art. Images of the ram, rarely complete, sometimes reduced to horns, but most often the head alone, naturalistic or schematically, are often associated with other anthro­ pomorphic or zoomorphic images to create monsters. They can also be represented next to the tree of life in some compositions. These images are present in Celtic plastic production at the end of the first Iron Age, but especially from the start of La Tène culture (Verger 1991, Tab. IV), and even during the Roman Empire period (Cluytens 2009). For Early La Tène for example, they can be found on a few rare brooches such as those from France – Aignay-le-Duc, Côte-d’Or, or Port-à-Binson and Suippes, Marne, or from Dompierre-les-Tilleuls, Doubs – from the Hallein cemetery in Austria, or again from Berlin Niederschonhausen in Germany (Binding 1993). On the fibula of Drouzkovice in Bohemia, the ram’s head is held in the jaws of a wild beast, but on one from Dompierre-les-Tilleuls, a sort of

role reversal, it is the ram which holds a human head in its mouth (ibid.). This image of the ram on fibulae is still rare: less than 2% of those with animal heads, as noted by M. Cluytens (2009, 205), and the other types of artefact bearing representations of rams, such as the gold bracelet from Rodenbach (Jacobsthal 1944, no. 59; Megaw 2001, 114), are scarce. During the Early La Tène, the ram’s head – or only its horns – is also present in the development of hybrid monsters, such as the serpent with a carnivore horned head on the cheekpieces of the helmet from the Perrats cave, Agris, Charente (Gomez Soto and Verger 1999 and 2010), or taking the place of the human body on the handle of the enochoe from the Reinheim burial, where the human face is superimposed directly onto the ram’s head (Keller 1965, pl. 20–24, Echt 1999, fig 75–76, pl 22), following an iconographic style which reemerges very much later in Late La Tène on harness fittings from Manerbio sul Mella in Lombardy (Jacobsthal 1944, no. 84), or again combined with a human face at Erstfeld (Wyss 1975, fig 11; Guggisberg 2000). Excessively extended eyebrows with spiral endings, so plentiful with a range of variations in La Tene A to later dates, as on the double-faced head of the anthropoid sword handle from Saint-André-de-Lidon in Charente-Maritime (Duval et al 1986), could also correspond to – or actually match? – a merger between the human mask and the image of the ram. On the perforated plate reused as a bow fibula from grave 3, Croyère Orainville in the Aisne, the punched decoration on the foreheads of the human masks reads as a representation of the ram’s fleece, marking this fusion in a particularly significant way (Desenne et al 2005, 254 ff.; Charpy 2009). M. Cluytens (2009, 205) associates the image of the ram with dragon pairs, and emphasized that the first becomes less common at the same time as the second begins to appear on sword scabbards. At the end of late La Tène – the date of the Chermignac ring – images of ram heads remain proportionally rare among anthropomorphic and zoomorphic representations of all kinds, on metal and other materials, and non-existent in some regions (Megaw 2001, 160); in Moravia, for which we have a complete inventory, they appear only in five cases out of a total of twenty-two (Čižmár 2011). But if, for the whole of Europe, the search is limited to rings with zoomorphic representations, already in the minority compared to rings with bead decoration, the proportion of those with ram heads rises to nearly one in two (Tappert 1998). In Gaul, besides the Chermignac ring, very few other zoomorphic representations decorate other objects. The ram head is not unknown there: for example, it ends the upper branches of the anthropoid sword hilt of a cremation grave in the necropolis of Mouriès in the Bouches-du-Rhône (Déchelette 1914, 1140; Marcadal et al 2003, Guillaumet

23.  De l’anneau en bronze à têtes de béliers de Chermignac and Rapin 2000, 80). To cite a few examples outside Gaul, we could mention the bronze tip of a drinking horn decorated with a human head surmounted by a ram’s head from Donnersberg (Zeeb-Lanz, 2008, 67 ff.), or the animal figurine of uncertain identification which P. Vouga attributed to Late La Tène reoccupation (Vouga 1923, 121–122, pl. L 23). As during the Early La Tène, during Late La Tène the horns of the ram can be combined with other representations of real or mythical animals to develop hybrid monsters: on iron andirons from Le Montéqueux à Beine in the Marne, they decorate griffin heads, already unusual in such hearth equipment (Charpy 2007); on the torc from Frasne-lesBuissenal in Hainaut, they are combined to a bovine protome (Jacobsthal 1944, no. 70; Hautenauve 2005, 194; Ginoux 2006, 144); let us not forget, of course, the horned snakes on the Gundestrup cauldron.

A ring manufactured in the west or in central Celtic Europe? On the subject of some elements of horse trappings in Western Gaul. The Chermignac ring is, as we have seen above, very similar to various others from central Europe. This raises the question of the origin of this example from western Gaul: central European manufacture, or conversely, western? The hypothesis of importation from the Celtic east has already been made for some ornaments found in western Gaul, for example for various bronze objects from Lacoste, by comparison with counterparts in central Europe, from Bohemia in particular (Boudet 1987, 117), to which, from the same site, could be added a recently discovered bracelet decorated with plastic ovals and opposing anthropomorphic masks, which opens by a removable clasp (Sireix 2012, 38). One could also mention in this category of ornaments – rare to Western Gaul – the still unpublished pair of anklets presumed to come from the area of Niort in the Deux-Sèvres, with hollow undecorated ovals and a unique removeable clasp (research in prep. J. Gomez de Soto, A. Lehoërff and A. Masse). A similar assumption of a central European origin was also put forward by J. Hiernard and D. Simon-Hiernard about a pair of horse harness pendants interpreted as frontal (actually pendants for the side of the head, as the disposition of a suspension bar kept in place on a metal La Tène head equipment shows: Vouga 1923, pl. XXXVII 1), found in association with an iron spit at Saintes in Charente-Maritime in the 19th century, now unfortunately lost, but known by a very precise drawing at the time of their discovery (Hiernard and Simon-Hiernard 1999 107 ff. and 2000). One was complete with its openwork disc decorated with curvilinear contours and its suspension bar, the other was reduced to its suspension bar (Fig. 23.5). The same assumption of a central European origin was also proposed by the same authors

203

for an openwork bronze plaque from Lacoste (Hiernard and Simon-Hiernard 1999, 116) (Fig. 23.6), from a site which also produced another suspension bar for a harness pendant similar to those from Saintes, or to the La Tène head equipment (Sireix et al. 1985 fig. 7.) (Fig. 23.7: 1), and a cruciform object with suspension rings, probably also for a harness (Fig. 23.7, 2), which finds a parallel at Stradonitz (Píč 1903, pl. XXI 28). Obviously, the distant origin of these objects is an admissible hypothesis, but the theory of production in the west is equally possible: the great similarity of the material culture of central Europe and eastern Gaul was emphasized long ago by J. Déchelette (1901 and 1914), who linked the artefacts from Stradonitz and Bibracte, which assumes not only long-distance trade, but also production centres widely distributed in middle Europe. The observations of J. Déchelette, written in an era when western Gaul was almost terra incognita, may also apply, we now know, to the west of the continental Celtic world, where highly standardized products, among others in weaponry and artistic production, reflect the early establishment of the Celtic Koine in the 5th century. Back to the Saintes pendants, we emphasize that, rather than encouraging the search for a single workshop or a region of limited production, the different examples of the known disks offering significant differences in their ornamentation and the trans-European character of the curvilinear style used for their openwork as well as the use of ordinary rivets with striated heads, suggests instead the probability of many workshops: the distribution of these objects from Hungary to the Atlantic, probably via the Champagne (Stead and Rigby 1999, pl. 186) does not allow us to favour one region more than another, and these ornaments appear to be trans-Celtic. The Saintes disk, its flattering appearance masking its poor quality, can hardly be attributed to, for example, the same workshop as that which produced the finest examples, almost perfect, from Skyjle in the Czech Republic, or Niedenstein in Hesse. Above all, its rim carries an engraved or stamped decoration, a series of arches and eyespots which do not figure on any of the other known examples. This pattern is clearly inspired by the stamped decorations common on the pottery of north-west Gaul, Brittany in particular (Daire 1992), decoration which is also known on a few examples from the centre-west from the 5th century to the 1st century BC (Gomez de Soto et al. 2007). This ornamentation is overlapped by an openwork, the one awkwardly lacking in symmetry, which proves that it was drawn first and not added later to an object of distant provenance which was adapted to local tastes: that it represents a western production is unquestionably the most likely hypothesis. The same applies to the aforementioned objects from Lacoste, for which manufacture by bronze workshops at the site (Sireix 2012, 56) is equally plausible. As for the openwork plaque from Lacoste, it has a peripheral decoration of eyespots, a

204

José Gomez de Soto

pattern common on La Tène pottery of northwestern Gaul, also indicative of western manufacture. Although it remains isolated in the far west, we maintain the hypothesis that the ring decorated with ram heads found at Chermignac is the product of a western workshop and not a long-distance import. The variety of types of metal objects of any kind yielded by sites like Lacoste (Sireix et al 1985. Sireix 2012), and the contributions of rescue archeology (and that despite the extreme rarity of funerary contexts), are making obsolete, progressively and radically, the discrepancy between the quantity and quality of material available from the Celtic east and west, while frequently confirming similarities. However, it is more than obvious now that western Gaul was actively engaged in the Celtic Koine, right from the formative period of La Tène Culture (Gomez de Soto 2005; Milcent 2006; Gomez de Soto et al 2007).

Acknowledgments Our gratitude goes to our colleague Christophe Lassalle, finder and owner of the Chermignac ring, who entrusted us the ring and permitted us to present it, and Christophe Sireix, who allowed us to examine the discoveries made by his father at Lugasson and mentioned to us the unpublished ring from Les Murasses. Translated by S. Crawford.

Bibliographie Binding, U. 1993. Studien zu den figürlichen Fibeln der Frühlatènezeit, Universitätsforschungen zur prähistorischen Archäologie aus dem Seminar für Ur – und Frühgeschichte der Universität Münster, 16, Bonn: Rudolf Habelt. Boudet, R. 1987. L’Age du fer récent dans la partie méridionale de l’estuaire girondin, Archéologies 2, Périgueux: Vesuna. Charpy, J.-J. 2007. La tombe aux landiers de Beine «Le Montéqueux» (Marne) et la représentation du bélier à bec de griffon. In V. Kruta and G. Leman-Delerive (eds.) Feux des morts, foyers des vivants. Les rites et symboles du feu dans les tombes de l’Âge du Fer et de l’époque romaine, 97–106. Lille: Revue du Nord (hors-série 11). Charpy, J. -J. 2009. Le bélier, l’oiseau rapace et leurs rapports avec la divinité principale, Études celtiques 38, 5–19. Čižmař, M. 2011. Die spätlatènezeitliche Tier- und Menschdarstellung aus Mähren, Études celtiques 37, 53–62. Čižmař, M., Kolnikova, E., Noeske, H.-C. and Noeske, H.-C. 2008. Nĕmĕice-Vicemĕřice – ein neues Handels- und Industriezentrum der Latènezeit in Mähren, Germania 86, 655–699. Cluytens, M. 2009. Réflexions sur la symbolique du bélier chez les Celtes protohistoriques à travers les représentations figurées, Lunula. Archaeologia protohistorica 17, 201–206.

Daire, M.-Y. 1992. Les céramiques armoricaines de la fin de l’Age du Fer. Travaux du Laboratoire d’Anthropologie de l’université de Rennes I. Déchelette J. 1901. Le Hradischt de Stradonic en Bohême et les fouilles de Bibracte, étude d’archéologie comparée. Mâcon, Protat (off-print from Congrès archéologique de Mâcon, juin 1899). Déchelette, J. 1914. Manuel d’Archéologie préhistorique, celtique et gallo-romaine, II, 3. Second Âge du Fer ou époque de la Tène. Picard: Paris. Desenne, S., Collart, J.-L., Auxiette, G., Martin, G., Rapin, A. and Duvette, L. 2005. La nécropole d’Orainville (Aisne). Un ensemble attribuable au Aisne-Marne IV. In G. Auxiette and F. Malrain (eds.) Hommage à Claudine Pommepuy, 233–270. Revue archéologique de Picardie, spécial 22: Lille. Duval, A., Gaillard, J. and Gomez de Soto, J. 1986. L’épée anthropoïde de Saint-André-de-Lidon (Charente-Maritime) In A. Duval and J. Gomez de Soto (éds.) Actes du VIIIème colloque sur les Ages du Fer en France non méditerranéenne (Angoulême, 1984), suppl.1, 233–238. Bordeaux: Aquitania. Echt, R. 1999. Das Fürstinnengrab von Reinheim, Blesa 2, Bliesbruck-Reinheim: Parc archéologique européen. Filip, I. 1956. Keltové ve stredni Europe, Praha: Nakladatelstvi ceskolovenské Akademie ved. Ginoux, N. 2006. Éléments d’iconographie celtique: le thème du taureau à cornes bouletées dans le répertoire du Nord de la Gaule, Archéologie de la Picardie et du Nord de la France 88, 129–150. Gomez de Soto, J. 2005. Actualités de l’art de la Tène ancienne en Gaule occidentale. Age du Fer en Europe – The Iron Age in Europe, 3–7. (Actes du congrès de l’UISPP, Liège, 2001), BAR International Series, 1378, Oxford, British Archaeological Reports. Gomez de Soto, J., [dir.] Lejars, T., Ducongé, S., Robin, K., Sireix, C. and Zélie, B. 2007. Du milieu du Ve au IIIe siècle avant notre ère en Centre-Ouest, Aquitaine septentrionale et ouest du Massif central. In C. Mennessier-Jouannet, A.-M. Adam and P.-Y. Milcent (éds.) La Gaule dans son contexte européen aux IVe et IIIe s. av. n. è. (actes du XXVIIe colloque international de l’AFEAF, Clermont-Ferrand, 29 mai–1er juin 2003), 69–89. Lattes: Monographies d’archéologie méditerranéenne. Gomez de Soto, J. and Verger, S. 1999. Le casque celtique de la grotte d’Agris. Angoulême: Groupe d’études et de recherche du musée. Gomez de Soto, J. and Verger, S. 2010. Le casque d’Agris, chefd’œuvre de l’art celtique occidental, L’Archéologue,106, 56–59. Guggisberg, M. A. 2000. Der Goldschatz von Erstfeld. Ein keltische Bilderzyklus zwischen Mitteleuropa und der Mittelmeerwelt, Basel: Schweizerische Gesellschaft für Ur- und Frühgeschichte. Guillaumet, J.-P. 1983. Chalon «Cabilonnum» – Marché et port fluvial des Eduens, La vallée de la Saône aux âges du Fer (VIIe – Ier siècle avant notre ère), 31–36. Rully: Revue archéologique de l’Est et du Centre-Est, 6ème supplément. Guillaumet, J. -P. and Rapin, A. 2000. L’art des Gaulois du Midi. In J. Chausserie-Laprée (ed.), Le temps des Gaulois en Provence, 79–83. Martigues: Musée Ziem. Hautenauve, H. 2005. Les torques d’or du second Âge du Fer

23.  De l’anneau en bronze à têtes de béliers de Chermignac en Europe. Techniques, typologies et symbolique, Rennes: Association des Travaux du laboratoire d’Anthropologie de l’université de Rennes 1, Rennes. Hiernard, J. and Simon-Hiernard, D. 1999. Les Santons, les Helvètes et la Celtique d’Europe centrale. Numismatique, archéologie et histoire, Aquitania 16, 93–125. Hiernard, J. and Simon-Hiernard, D. 2000. Le ‘frontal’ celtique de Saintes, Bulletin de la Société des Antiquaires de l’Ouest (5th série) 14, 1st–2nd trim., 3–32. Jacobsthal, P. 1944. Early Celtic Art, Oxford: The Clarendon Press. Keller, J. 1965. Das keltische Fürstengrab von Reinheim, Mainz: Römisch-Germanischen Zentralmuseum. Kruta, V. 1992. Brennos et l’image des Dieux: la représentation de la figure humaine chez les Celtes, Comptes-rendus de l’Académie des Inscriptions et Belles Lettres 4, 821–846. Kruta, V. 2000. Les Celtes. Histoire et dictionnaire des origines à la romanisation et au Christianisme, Paris : Laffont. Marcadal, N., Marcadal, Y. and Paillet, J., with contributions by Heijmans, M., Villemeur, I. and Columeau, P. 2003. La nécropole protohistorique et gallo-romaine de Servanes – Cagalou (Ier s. av. – IIIe s. ap. J.-C.) à Mouriès (Bouchesdu-Rhône): sépultures et monuments funéraires, Documents d’Archéologie méridionale 26, 251–348. Megaw, R. and Megaw, J. V. S. 2001. Celtic Art. From its beginnings to the Book of Kells, London: Thames & Hudson. Milcent, P.-Y. 2006. Premier âge du Fer médio-atlantique et genèse multipolaire des cultures matérielles laténiennes. In D. Vitali (éd.) Celtes et Gaulois, l’Archéologie face à l’Histoire, 2: la Préhistoire des Celtes, 81–105. (Bibracte, 12/2), Glux-enGlenne: Bibracte, Centre archéologique européen. Pelatan, J.-P. 1986. L’oppidum de Larina, Avignon: Revue archéologique Sites, hors-série 31. Píč, J. L. 1903. Hradištĕ u Stradonic jako historické Marobudum, Praha: Vlastním. Schaeffer, F. A. 1930. Les Tertres funéraires préhistoriques dans la

205

Forêt de Haguenau. II. Les Tumulus de l’Age du Fer, Haguenau: Imprimerie de la ville. Schnapp, A. 1993. La conquête du passé. Aux origines de l’archéologie, Paris: Carré. Sireix, C., Sireix, M. and Faravel, S. 1985. Le site gaulois de Lacoste: un exemple d’habitat du deuxième âge du Fer girondin, Bordeaux: Groupe de recherches archéologiques de Lacoste. Sireix, C. 2012. Lacoste, un village artisanal et commercial. In V. Mistro and C. Sireix (eds.) Au temps des Gaulois. L’Aquitaine avant César, 30–43. Bordeaux: Musée d’Aquitaine. Stead, I. M. and Rigby, V. 1999. Iron Age Antiquities from Champagne in the British Museum. The Morel Collection, London: British Museum Press. Tappert, C. 1998. Ein keltischer Widderkopfring aus Straubing und verwandte Tierprotomringe, Jahresberichte des historischen Vereins für Straubing und Umgebung 100, 1, 173–217. Van Endert, E. 1991. Die Bronzefunde aus dem Oppidum von Manching (Aus Grabungen in Manching, 13), Stuttgart: F. Steiner Verlag. Verger, S. 1991. L’utilisation du répertoire figuratif dans l’art celtique ancien, Histoire de l’art 16, 3–17. Vouga, P. 1923. La Tène. Monographie de la station publiée au nom de la commission des fouilles de La Tène, Leipzig: K. W. Hiersemann. Wyss, R. 1974. Grabriten, Opferplätze und weitere Belege zur geistigen Kultur der Latènezeit, Ur- und frühgeschichtliche Archäologie des Schweiz, IV, Die Eisenzeit, 167–196. Basel: Schweizerische Gesellschaft für Ur- und Frühgeschichte. Wyss, R. 1975. Der Schatzfund von Erstfeld. Frükeltischer Goldschmuck aus des Zentralalpen, Zürich: Gesellschaft für das Schweizerische Landesmuseum Zürich. Zeeb-Lanz, A. 2008. Der Donnersberg. Eine bedeutende spätkeltische Stadtanlage (Archäologische Denkmäler in der Pfalz, 2), Germersheim: Steimer Verlag.

24 A MOULD FOR CELTIC-TYPE RINGS FROM SANZENO IN THE VALLE DI NON, TRENTINO Franco Marzatico

In a stimulating article by Vincent and Ruth Megaw, entitled ‘From the Tyrol to Texas: lost and found in Middle La Tène’, dedicated to the analysis of some Celtic-type evidence originating from the Alpine area accredited to the Raetic population (Fig. 24.1), the controversial question of the connotations of the Celtic world’s contributions documented between the 5th – 1st century BC in this characteristic central-eastern Alpine area, is presented in an effective synthesis (Megaw and Megaw 1996). Calling to mind Anne Marie Adam’s considerations, the authors point out that: ‘To be taken seriously are claims for their native or Raetic production; the alternative to their representing a locally developed variant would see them as the result of actual Celtic settlement within the area…’, further observing that ‘…on one hand the “Alpine” helmets worn by the brochure finial heads and on the other the general La Tène brooch type and the assured nature of the low relief “vegetal” ornament, the assimilation of “Celtic” or trans-Alpine La Tène minority by the local non-“Celtic” Fritzens-Sanzeno group would seem to offer a middle way’ (Megaw and Megaw 1996). Though I have not had the privilege to personally meet Vincent, his intellect and humane qualities, irony notwithstanding, were appreciated through our correspondence, and to render homage to his scientific contribution the choice fell on a Celtic-type object that is part of the collection of the Castello del Buonconsiglio of Trento. It concerns a type of stone mould for casting brought to light in 1927 at Sanzeno in the Valle di Non, Trentino, recently taken into consideration on the occasion of the recognition and study of pre-Roman materials originating from the valleys crossed by the Noce River, in the framework of a ‘dig in the museum’, to use a definition that is now in vogue.

The stone mould numbered 7050 on the inventory sheet, measures 9.5 cm × 5.5 cm and is 2.1 cm thick. The stone, with two opposing worked surfaces, one for the production of solid rings in lenticular sections of 3.3 cm in diameter and the other of ribbed discs of 5 cm in diameter, appears to be full of holes on one side and, excluding this and the other fractured surfaces as well as the shorter side with only the edges polished, all the surfaces are carefully smoothed. (Fig. 24.2 a–d). The cavity for the production of rings is divided into four parts by etched curved lines that enclose meandering motifs on either side with two stylized ogival eyes (Fig. 24.2 a, c, e). At the level of these eyes is a funnel-shaped dip for the flow of the melted metal. In the centre of the ring is a cylinder (the diameter measuring 1.4 cm) with a deep central hole (0.5 cm in diameter) that with the use of a linchpin blocks the second mould within the bivalve mould. More in-depth examination results in a cylindrical intaglio of the second surface, with a base consisting of four sturdy concentric ribs and a central concavity of 1.5 cm in diameter (Fig. 24.2 b, d).

Considerations Directing one’s attention to the side of the mould that has the ring-shaped concavity (Fig. 24.2 a, c, e), due to its structure and the decorative motifs carved in the stone, it is very apparent that the intended use of the matrix under consideration was aimed at the production of solid Celtictype sectioned lenticular rings decorated with meandering motifs and stylized eyes in relief.

24.  A mould for Celtic-type rings from Sanzeno in the Valle di Non, Trentino

207

Fig. 24.1 Peoples and cultural aspects of northern Italy in the 5th century BC.

The model is in fact comparable, as a type in its own right, to the articulated Koppelringe group, according to the terminology of the German bibliography, documented as noted in the course of La Tène in much of Europe, with many versions from the constructive point of view, or else from the uses of bronze or iron and the presence or absence of decorative motifs (Raftery 1987; 1998, 68; Ramsl, 2002, 80–81, fig. 81). An example in this regard is the typological table of the rings brought to light in the necropolis of Pottenbrunn in Austria which includes more or less solid bronze and iron specimens, with very variable diameters and sections, asymmetrical and symmetrical, lenticular, sub-rhomboidal, flat and circular (Ramsl 2002, 80–81, fig. 81). In addition to these types, hollow rings have been noted

in a large part of Europe, composed of two laminates with C-shaped sections assembled in a similar manner to the documented half ring, always by way of example, in Tomb 6 of Müsingen-Rain in Switzerland, object of a careful analysis by Raftery (Raftery 1998). The author, distinguishing two constructive typologies, made respectively with the use of rivets and assembly of the laminates through folding the edges, signals the presence of the first grouping in tombs predominantly of La Tène A, persisting through La Tène B, and the second through the course of La Tène A and principally in La Tène B2/C, with a rarefaction in La Tène C 1 (Raftery 1987, 521–524; 1998, 64, 68). Always according to Raftery, the function of the hollow rings, principally attested in the male tombs, less so in those of females and only exceptionally even in those of children, does not appear

Franco Marzatico

208

a

b

c

e

g

d

f

h

Fig. 24.2 a–d) stone casting form from Sanzeno; e) graphical depiction of the engraved motif on the stone of the casting form; g–h) solid bronze ring from Lavis; f) drawing of solid bronze ring from Vadena/Pfatten (by Ghislanzoni 1940).

precisely definable given that their location in the burials is variable and therefore a ‘multiple use’ could be deduced (Raftery 1987, 524). The examples of the oldest hollow rings, documented from the beginning of the La Tène A, originated

predominantly from tombs of male individuals, frequently burials of warriors with rich grave goods including a sword. According to Raftery, for the majority of cases, the relationship between the weapon and the rings was not

24.  A mould for Celtic-type rings from Sanzeno in the Valle di Non, Trentino observed (Raftery 1987, 524; 1998, 61). The ornamental use of the rings is revealed in female tombs and probably also in the case of male tombs, as that of the renowned Weiskirchen burial mound in Germany where the richly decorated rings were meant to be hung from a belt as ornamental elements (Raftery 1987, 524; 1998, 65; Müller 2009, 188, fig. 248). In the tombs for females and children, the rings were either arranged in relation to the head or inserted in a neckband, or used as a decorative belt pendant, particularly in regards to the second typological group (Raftery 1987, 525). Regarding the rings blocked with rivets, those of the second group are connected mainly to male burials with a more evident connection to the sword (Raftery 1987, 524; 1998, 65). As indicated by the study of the sepulchral complex in the necropolis of Pottenbrunn, where as previously mentioned different typologies are represented, the rings are often in pairs or in groups of three, frequently placed in a functional relationship with the sword or the belt in the tombs of warriors (Raftery 1998, 65; Ramsl 2002, 80–81, fig. 81). Only in a limited group of tombs of those armed with a sword was there a single ring which, however, showed little or no connection with the sword (Raftery 1998, 65). By analyzing all this data and the attestation of the laminated half-ring positioned at the hip area of the female individual between the ages of 14 and 20, in Tomb 6 of Müsingen-Rain, Raftery raises the question whether the hollow rings had been used as ornaments, part of a belt or else as amulets, or as a religious object (Raftery 1998). Taking into account that the hollow rings in the female burials did not seem to always have the same function, the author gives these objects an important magical and religious significance, interpreting them as amulets. The symbolic value of the belts emerges from the rest in diverse chronological and cultural settings, as for example, the ‘survival’ of the large Villanoviano-type belts in the form of discoidal cut-outs placed in areas of worship or as elements of grave goods, or else the Samnite-types, as evidence of that historical age (Marzatico and Endrizzi 2009, 51). But getting back to the Sanzeno mould in question, it should be noted that the ring-shaped intaglio obtained by a careful carving of the stone (Fig. 24.2 a, c, e) fits almost perfectly in relation to the shape and dimensions of a solid bronze ring originating from Lavis (north of Trento) which is part of the Castello del Buonconsiglio collection, inventory number 5393 (Fig. 24.2 g–h). The Lavis ring (Fig. 24.2 g–h) represents the closest parallel, not only for the characteristic typologies, but also from a geographic point of view, given the decorative motifs in relief, however very worn down indicating a prolonged use of the object. Still in the Adige Valley, a few kilometres north of Lavis, in the necropolis of Vadena/Pfatten is additional comparison (Fig. 24.2 f) (Ghislanzoni 1940, 437–439, fig. 110; Roncador 2011, 172–173). As in the case of the Sanzeno mould (Fig. 24.2 a–d), the rings from Lavis (Fig. 24.2 g–h) and Vadena/Pfatten (Fig. 24.2 f) lack precise information in the context of the

209

findings that would facilitate the chronological framework. Keeping the characteristics of the wavy decorative motifs in relief and the stylized ‘lyre shape’ composition on the Lavis ring as dating parameters (Fig. 24.2 g–h), there is a possible placement of the production of the rings in question towards La Tène B, also compatible with the studies on the suspension fittings of sword scabbards conducted by Rapin, and as for the rings from the Adige Valley, for the rich decorations one can presume it had an ornamental function, perhaps similar to that of the Weiskirchen examples? (Rapin 1987; 1991; 2003; Müller 2009, 188, fig. 248; Vitali 2011). On the other hand, one cannot also rule out a further relevance in the same Raetic geographical area, of the recurring so-called mandolinshaped fibulae, with a meandering decoration similar to that presented on the Sanzeno mould (Adam 1996, 167–165, fig. 25; Gleirscher et al. 2002, 50–52 tav. 134 B; Neuner 2002; Gamper 2006, 159–160, k. 32), or the use of rings decorated with undulating motifs also at La Tène D found in Tomb 4 of Valeggio sul Mincio where the bronze ring differs, however, by a decoration of a line of tiny dots and circles, and a flat section (Salzani 1995, 59 table V n. 39). But omitting the aspects of chronological order which may be clarified by new research and discoveries, it should be stressed how the mould for the production of Celtic-type rings from Sanzeno, together with the bronze examples from Lavis and Vadena/Pfatten as mentioned (Fig. 24.2), are testimonies to a distinctive local production that assumes a specific interest from the historical-cultural dynamics point of view, again presenting particularly debatable questions. This is in reference, on the one hand, to the aspects established at the Sanzeno centre and, on the other hand, to the significance of the presence of Celtic-type artifacts in the central-eastern Alpine area attributed to the Raetians and therefore to the nature of the relations existing between Celts and Raetians (Lang 1999; Adam 2006; Marzatico 2001; Roncador 2011; Marzatico 2012). In this regard, and in light of the definition of Trento city of the Gauls, throughout all the studies made (not counting hypotheses of appropriations unsupported by substantial excavation data) confrontations of varying degrees have been made: a ‘Brenno model’, influenced by the idea of invasions, Celtic raids within Raetic areas and a ‘Helico model’ that led back to the Celtic contributions within the channel of contacts, exchanges, integration and acculturation (Marzatico 1992; Schönfelder 2002; Bats 2006; Marzatico 2013, 145). If we exclude the few pottery remains originating from sites in the more northern Raetic areas and ‘technological’ components (connected to metallurgic production such as bi-pyramidal ingots, cart parts, and long haymaking scythes), the Celtic-type elements consist essentially of ‘mobile’ objects to put on or to wield: ornaments, clothing, and offensive and defensive armament (Gleirscher 1987/1988; Egg 1992, 426–430; Adam 1996; Rapi 2000; Lang 2002; Roncador 2011; Marzatico in press). The solid rings in question are therefore included in this framework where,

210

Franco Marzatico

with reference to the fibulae, the imitations certainly prevailed over the proven imports, often reinterpreted locally in the flourishing regional workshops (Adam 1996; Piana Agostinetti 1998; Megaw and Megaw 2000). The Sanzeno site has been interpreted in contrasting terms. In the Italian bibliography it figures mainly as a place of settlement, metallurgical production and an empirical area, while in the German one, greater importance is given to shrines because of the consistent presence of objects related to the magical-religious sphere and to defunctionalized metal artifacts and ‘hoarding’ in ‘treasure-houses’, corresponding to the location of the buildings of aligned constructions in the Casalini area (Vidale 1992; Migliavacca and Ruta Serafini 1992; Nothdurfter 1992, 56–57; 2002, 1131–1136; Sölder 2002 46–55; Von Uslar 1997, 166–201, 167; Adam 2006; Marzatico 2012). The presence of the mould, belonging to a class of material generally absent from places of worship with burnt offerings (Brandopferplätze) of the Raetic territory (Gleirscher et al. 2002; Steiner 2010; Marzatico 2009), seems to confirm the development of metallurgical activities on the site that is also indicated by work tools, ingots and other forms of casting. On the other hand it seems quite plausible that the elite who controlled this metallurgical production also exercised its own supremacy over the religious aspects and therefore the Sanzeno site, in the Casalini locality displaying a proto-urban aspect, carried out more functions (Adam 2006; Marzatico 2012). In the light of the consistent presence of Celtic-type evidence at the site, particularly in the case of armament elements, it often does not lend itself to a precise distinction between imports and imitations, and with the lack of incontrovertible excavation data an open question remains whether and to what point these materials can be considered as products from local workshops that came into contact with artifacts, artisans, merchants, warriors and individuals from the Celtic world, or are due to achievements by a minority of ‘foreign’ workers who settled peacefully in Sanzeno, resuming the ‘middle way’ outlined years ago by Vincent and Ruth Megaw (Nothdurfter 1979; Marzatico 1992; Adam 1996; Roncador 1996; Megaw and Megaw 1996, 64; Bergonzi and Piana Agostinetti 1997, 368; Megaw and Megaw 1998, 398; Frey 1999; Marzatico 2001, 528–537; Roncador 2009; 2011; Marzatico 2011, 102–104).

Acknowledgements We thank Carmen Calovi and Michele Dalba; translation by Jill Kaufman; drawings by Dora Giovannini and Marco Pontalti.

Bibliography Adam, A. M. 1996. Le fibule di tipo celtico in Trentino, Patrimonio storico artistico del Trentino 19. Trento: Provincia Autonoma di Trento. Adam, A. M. 2006. Dépôts d’objets métalliques du second âge du Fer dans le nord-est de l’Italie et les Alpes orientales. In G. Bataille and J.-P. Guillaumet (eds.) Les dépôts métalliques au second âge du Fer en Europe tempérée, Actes de la table ronde organisée par l’UMR 5594, CNRS-Culture-université de Bourgogne. Archéologie, cultures et société en Bourgogne et en France orientale, Équipe 3: la société gauloise (Glux-enGlenne, 13–14 octobre 2004), Bibracte 11, 135–45. Glux-enGlenne: CNRS. Bats, M. 2006. L’acculturation et autres modèles de contacts en archèologie protohistorique europèenne. In M. Szabó (ed.) Celtes et Gaulois. L’archéologie face à l’Histoire. Les Civilisés et les Barbares du Ve au IIe siècle avant J.-C., Actes de la table ronde Budapest, 17–18 juin 2005, Bibracte 12/3, 29–41. Glux en Glenne: CNRS. Bergonzi, G. and Piana Agostinetti, P. 1997. La seconda età del Ferro nelle Alpi centrali. In Atti della XXXI riunione scientifica. La valle d’Aosta nel quadro della Preistoria e Protostoria dell’arco alpino centro-occidentale, Courmayeur, 2–5 giugno 1994, 361–91. Firenze. Egg, M. 1992. Spätbronze- und eisenzeitliche Bewaffnung im mittleren Alpenraum. In I. Metzger and P. Gleirscher (eds.) Die Räter / I Reti, Arbeitsgemeinschaft Alpenländer, 401–38. Bozen: Verlagsanstalt Athesia. Frey, O.-H. 1999. Beziehungen zwischen Rätern und Kelten. In G. Ciurletti and F. Marzatico (eds.) I Reti / Die Räter, Atti del Simposio Castello di Stenico, 23–25 settembre 1993, Archeologia delle Alpi 5, 671–701. Trento: Provincia Autonoma di Trento, Servizio Beni Culturali, Ufficio Beni Archeologici. Gamper, P. 2006. Die latènezeitliche Besiedlung am Ganglegg in Südtirol: Neue Forschungen zur Fritzens-Sanseno-Kultur, Internationale Archäologie 91. Rahden/Westf.: Leidorf. Ghislanzoni, E. 1940. Il sepolcreto di Vadena, Monumenti Antichi XXXVIII. Roma: Bretschneider. Gleirscher, P. 1987/1988. Zu latènzeitlichen Halsringen aus dem Bereich der Fritzens-Sanzeno-Kultur, Denkmalpflege in Südtirol/Tutela dei Beni Culturali in Alto Adige 1987/1988, 263–75. Lana: Tappeiner AG. Gleirscher, P., Nothdurfter, H., Schubert, E. 2002. Das Rungger Egg: Untersuchungen an einem eisenzeitlichen Brandopferplatz bei Seis am Schlern in Südtirol, Römisch-Germanische Forschungen 61. Mainz am Rhein: von Zabern. Lang A. 1999. Räter und Kelten. Archäologisches zu Nachbarn. In G., Ciurletti and F. Marzatico (eds.) I Reti / Die Räter, Atti del Simposio, Castello di Stenico, 23–25 settembre 1993 (Archeologia delle Alpi 5), 373–398. Trento. Lang, A. 2002. Der tordierte Bronzehalsreif. In G. Tomedi and J. Zeisler (eds.) Archaeo Tirol Kleine Schriften 4, 154–62. Wattens: Verein Archaeo Tirol. Marzatico, F. 1992. I Galli abitanti del Trentino preromano? Revisione critica della vecchia tesi alla luce delle attuali conoscenze archeologiche. In Per Aldo Gorfer: studi, contributi artistici, profili e bibliografia in occasione del settantesimo

24.  A mould for Celtic-type rings from Sanzeno in the Valle di Non, Trentino compleanno, 619–51. Trento: Provincia autonoma di Trento, Assessorato all’istruzione, attività e beni culturali. Marzatico, F. 2001. La seconda età del Ferro. In M. Lanzinger, F. Marzatico and A. Pedrotti (eds.) Storia del Trentino I, La preistoria e la protostoria, 479–573. Bologna: Società Ed. Il Mulino. Marzatico, F. 2009. Luoghi di culto nell’area retica. In L. Endrizzi, N. Degasperi and F. Marzatico, Luoghi di culto nell’area retica. Altnoi. Il santuario altinate: strutture del sacro a confronto e i luoghi di culto lungo la via Annia, Atti del Convegno, Venezia 4–6 dicembre 2006, a cura di G. Cresci Marrone e M. Tirelli, Studi e ricerche sulla Gallia Cisalpina 23. Altinum 5, 263–74. Roma: Quasar. Marzatico, F. 2011. I Reti, fra protostoria e storia. In F. Marzatico and E. Migliario (eds.) Il territorio trentino nella storia europea I, La storia antica, 77–120. Trento: Fondazione Bruno Kessler. Marzatico, F. 2012. Waffen in Sanzeno – Deutungshypothesen im Vergleich. In W. Meighörner (ed.) Waffen für die Götter: Krieger, Trophäen, Heiligtümer, Austellungskatalog Tiroler Landsmuseum, 152–5. Innsbruck: Tiroler Landesmuseum. Marzatico, F. 2013. Veneti e Reti. In Venetkens, Viaggio nella terra dei Veneti antichi, Catalogo della mostra, Padova, Palazzo della Ragione, 6–17 novembre, 145–55. Venezia: Marsilio. Marzatico, F. in press. Il mondo retico fra Etruschi e Celti, Colloque international de l’Association Française pour l’Etude de l’Age du Fer. Verona. Marzatico, F. and Endrizzi, L. 2009. Un nuovo cinturone villanoviano dai Campi Neri di Cles, Trentino, Ocnus 17, 45–54. Megaw, J. V. S. and Megaw, M. R. 1996. From the Tyrol to Texas: Lost and found in middle La Tène. In M. Lodewijckx (ed.) Archaeological and historical aspects of West-European Societies: album amicorum Andreé van Doorselaer, Acta Archaeolgica Lovaniensia Monographiae 8, 59–66. Leuven: University Press. Megaw, J. V. S. and Megaw, M. R. 1998. Cheshire Cats in the Tyrol: an essay on prehistoric multiculturalism. In A. MüllerKarpe et al. (eds.) Studien zur Archäologie der Kelten, Römer und Germanen in Mittel- und Westeuropa, 389–400. Rahden/ Westf.: Leidorf. Megaw, J. V. S. and Megaw, M. R. 2000. Le fibule di tipo celtico nel Trentino (A.M. Adam). Recensione, Germania 78.2, 502–7. Migliavacca, M. and Ruta Serafini, A. 1992. Casa retica’ o abitazione alpina dell’età del Ferro? In I. Metzger and P. Gleirscher (eds.) Die Räter / I Reti, Arbeitsgemeinschaft Alpenländer, 369–814. Bozen: Verlagsanstalt Athesia. Müller, F. 2009. L’art des Celtes. 700 av. J.-C. – 700 apr. J.-C. Berne-Bruxelles: Musée historique de Berne. Neuner, M. 2002. Die Mandolinenfibeln. In G. Tomedi and J. Zeisler (eds.) Archaeo Tirol Kleine Schriften 4, 163–7. Wattens: Verein Archaeo Tirol. Nothdurfter, J. 1979. Die Eisenfunde von Sanzeno im Nonsberg, Römisch-Germanische Forschungen 38. Mainz: von Zabern. Nothdurfter, H. 1992. Die Fritzens-Sanzeno-Kultur und ihre Beziehungen zur etruskischen Kultur. In L. Aigner-Foresti (ed.) Etrusker nördlich von Etrurien, Etruskische Präsenz in Norditalien und nördlich der Alpen sowie ihre Einflüsse auf die einheimischen Kulturen, Akten des Symposions von Wien

211

– Schloß Neuwaldegg, 2.–5. Oktober 1989, 45–62. Wien: Verlag der Österreichischen Akademie der Wissenschaften. Nothdurfter, H. 2002. Schlüssel – Schlüsselgriffe – Geweihvotive. In Kult der Vorzeit in den Alpen. Opfergaben – Opferplätze Opferbrauchtum / Culti nella Preistoria delle Alpi, Arbeitsgemeinschaft Alpenländer, 1127–1154. Bolzano: Athesia. Peyre, C. 1979. La Cisalpine gauloise du IIIe ai Ie siècle avant J.-C., Études d’Histoire et Archéologie I. Paris. Piana Agostinetti, P. 1998. Recensione di Anne Marie Adam. Le fibule di tipo celtico in Trentino, Archeologia Classica L, 447–50. Raftery, B. 1987. Anneaux creux formés de deux plaques de metal en Europe Latenienne. In D. Vitali (ed.) Celti ed etruschi nell’Italia centro-settentrionale dal V secolo a.C. alla romanizzazione, Atti del colloquio internazionale Bologna, 12–14 aprile 1985, 521–7. Bologna: University Press. Raftery, B. 1998. Der Hohlblechring aus Grab 6 von MünsingenRain: Amulett oder Schmuck? In F. Müller (ed.) MünsingenRain, ein Markstein der keltischen Archäologie. Funde, Befunde und Methoden im Vergleich, Akten Internationales Kolloquium, Das keltische Gräberfeld von Münsingen-Rain 1906–1996, Münsingen/Bern, 9–12 Oktober 1996, Schriften des Bernischen Historischen Museums 2, 61–70. Bern: Verlag Bernisches Historisches Museum. Ramsl, P. C. 2002. Die keltischen Gräberfelder von Pottenbrunn und Mannersdorf am Leithagebirge. Zwei Drehscheiben zwischen West und Ost, Archäologie Österreichs 13/2, fasc. 2, 6–23. Rapi, M. 2000. Le armille di vetro La Tène. In R. C. de Marinis and S. Biaggio Simona (eds.) I Leponti tra mito e realtà 2, 63–73. Locarno: Dado. Rapin, A. 1987. Le système de suspension des fourreaux d’Épèes lateniens aux IIIe siècle av. J.-C. innovations techniques et reconstitution des Élements perissables. In D. Vitali (ed.) Celti ed etruschi nell’Italia centro-settentrionale dal V secolo a.C. alla romanizzazione, Atti del colloquio internazionale Bologna, 12–14 aprile 1985, 529–39. Bologna: University Press. Rapin A. 1991. Le ceintouron métallique et l’évolution de la panoplie celtique au III siècle av. J.C., Études Celtiques 28, 349–368. Rapin, A. 2003. Les analyses sémiologiques de l’image: l’iconographie du deuxième âge du Fer. In Décors, images images et signes de l’âge du Fer européen, XXVI Colloque de l’Association Francçaise pour l’Etude de l’Age du Fer, Pairs et Saint-Denis 2002, Supplément à la Revue Archéologique du Centre de la France 24, 49–62. Tours: FERACF. Roncador, R. 2006. Armi tipo La Tène in Trentino Alto AdigeSüdtirol. Problematiche relative alla presenza celtica in ambito alpino centro-orientale, Tesi di Specializzazione in Antichità Celtiche, Università degli Studi di Bologna. Bologna: University Press. Roncador, R. 2009. La ‘riscoperta’ del carnyx di Sanzeno (Val di Non, Trentino, Italia). Storia degli studi ed inquadramento culturale. In S. Grunwald et al. (eds.) ArteFact. Festschrift für Sabine Rieckhoff zum 65. Geburststag, 547–55. Bonn: Habelt. Roncador, R. 2011. Celti e Reti tra V e I sec. a.C. Oggetti tipo LaTène all’interno della cerchia culturale Fritzens-Sanzeno

212

Franco Marzatico

(arco alpino centro-orientale), Tesi di dottorato di ricerca in Archeologia, Università degli studi di Bologna. Bologna: University Press. Salzani, L. (ed.) 1995. La necropoli di Valeggio sul Mincio, Documenti di Archeologia – Padus. Mantova: Ed. SAP. Schönfelder M. 2002, Das Spätkeltische Wagengrab von Boé (dép. Lot-et-Garonne). Studien zu Wagen und Wagengräbern der Jüngeren Latènezeit. Monographien des RömischGermanischen Zentralmuseums 54. Mainz: Habelt. Schönfelder, M. 2010. Keltische Wanderungen – welche Modelle bleiben bestehen. In M. Schönfelder (ed.) Kelten! Kelten? Keltische Spuren in Italien, Mosaiksteine 7, 46–8. Mainz: Römisch-Germanisches Zentralmuseum. Sölder W. 2002. Zur Urgeschichte und Römerzeit in Nordtiro. In W. Sölder (ed.), Zeugen der Vergangenheit. Archäologisches aus Tirol und Graubünden, Ausstellungskatalog (Chur, 26 April – 29 September 2002; Innsbruck, 23 Oktober 2002 – 16 Februar 2003), 19–75. Innsbruck.

Steiner H. (ed.) 2010. Alpine Brandopferplätze. Archäologische und naturwissenschaftliche Untersuchungen. Roghi votivi alpini. Archeologia e scienze naturali, Forschungen zur Denkmalpflege in Südtirol V. Trento: Editrice Temi. Vidale, M. 1992. Produzione artigianale protostorica. Etno­ archeologia e archeologia, Vol 4: Saltuarie dal laboratorio del Piovego, 267–83. Padova: Saltuarie del laboratorio del Piovego. Vitali, D. 2011. Arte lateniana e Celti d’Italia. In S. Casini (ed.) ‘Il filo del tempo’. Studi di preistoria e protostoria in onore di Raffaele Carlo de Marinis, Notizie Archeologiche Bergomensi 19, 427–45. Bergamo: Comune de Bergamo, Assessorato alla Cultura, Civico Museo Archeologico. von Uslar, R. 1996. Zu Räter und Kelten in den mittleren Alpen, Bericht der Römisch-Germanischen Kommission des Deutschen Archäologischen Institut 22, 156–212.

25 ‘LEOPOLD BLOOM I’ AND THE HUNGARIAN SWORD STYLE Paul Jacobsthal with an introduction by Katharina Ulmschneider and Sally Crawford

Introduction It is very nearly impossible to discuss Early Celtic art without reference to the scholarship of Paul Jacobsthal or Ruth and Vincent Megaw, as the papers in this volume indicate. For the authors of this introduction, the names of Megaw and Jacobsthal are inextricably interlinked. When we first began to look at the potential of the Paul Jacobsthal Archive housed at the Institute of Archaeology, Oxford, Vincent Megaw gave us tremendous encouragement, gave generously of his wisdom and knowledge, and gave us invaluable and effective support in our grant applications to develop a project on the Archive with Professor Chris Gosden. This volume arises out of the Jacobsthal Project, and represents our ‘thank you’ to an inspiring scholar (Fig. 25.1). Since Paul Jacobsthal provided our personal introduction to Vincent Megaw, we think it is appropriate that Vincent’s Festschrift is the place where a significant piece of unpublished Jacobsthalia can find a wider audience for the first time. ‘Leopold Bloom I and the Hungarian Sword Style’ was written as a birthday present for Jacobsthal’s great friend, Sir John Beazley, no later than 1943. The first part of Jacobsthal’s paper describes and details the ornament on four British and Irish sword scabbards, and compares these details with continental scabbards in the Hungarian Sword Style. The second part offers a joking skit based on James Joyce’s novel Ulysses (not available in Britain until the 1936 publication), telling the story of Leopold Bloom I, an Iron Age Hungarian swordsmith who set up a workshop in England. Jacobsthal began work on Early Celtic Art (hereafter ECA) in Europe in the 1920s (J/J Archive, Box 3, Letter 119) while he was still in charge of the Archaeological Seminar at

Fig. 25.1. Oxford, 2012. Vincent Megaw in front of Jacobsthal’s camel, on display (unattributed) in the ‘West meets East’ gallery at the Ashmolean Museum, Oxford. Photograph: Tony Randall.

the University of Marburg, Germany, and best known for his work on Greek vase painting. His friends, including Beazley, warned him off his odd idea of investigating ‘barbarian’ art (BA Archive Letter 13.5.1930). His decision to press on with the topic proved to be fortunate in unexpected ways. 1930s Oxford had an elegant sufficiency of classical

214

Paul Jacobsthal with Katharina Ulmschneider and Sally Crawford

archaeologists, but when Jacobsthal was deprived of his post at Marburg ‘aus rassischen Gründen’ in 1935, he was able to find a post at Oxford with Beazley’s help, by transforming himself into the Reader in Celtic Archaeology. ECA was eventually published in 1944 (Jacobsthal 1944). Long before this date, Jacobsthal had already turned the focus of his considerable energies and scholarship to researching a second book on Celtic art in Britain, the country that was now his home. In 1939 he was sufficiently confident in his knowledge of the subject to publish on the Witham Sword in the Burlington Magazine (Jacobsthal 1939), and to apply to the British Academy for funds to support this research (J/J Archive, Box 3, Letter 46). What happened next is a matter of history. War intervened. After the war, Jacobsthal, now in his mid 60s and in unreliable health, felt he needed an assistant to help him with a task that had become overwhelming. His first choice was Günther Haseloff, the art historian and archaeologist, who had spent the war at Kiel as an assistant at the Museum of Prehistoric Antiquities. Letters from the archives at the Institute of Archaeology, Oxford, show Jacobsthal negotiating with Oxford University Press for a salary to allow Haseloff to spend a few months a year in England to help him with the project (Jac. Archive, Box 53, Letter 19 – the volume was due for completion in 1958). He also planned an itinerary with Haseloff that would take in the key relevant museums in Britain and Ireland (Jac. Archive, Box 53, Letter 94). But Haseloff’s appointment as Chairman of the Institut für Archäologie und Vor- und Frühgeschichte at Würzburg in 1954 brought negotiations to an end. Jacobsthal’s second choice was Martyn Jope. When, in 1957, Jacobsthal died, Jope took on the task of bringing Early Celtic Art in the British Isles (hereafter ECABI) to completion. Ultimately, the volume was published posthumously in 2000, under Jope’s name alone, a mere 42 years overdue – Oxford University Press’s record for time lapsed between commissioning and publication. Given this history, it is hardly surprising that the broad consensus in the Celtic Art world is that Jacobsthal had not been able to undertake any serious work on the material culture of Britain and Ireland before his death. However, a recent campaign to sort and box the Martyn Jope archive at the Institute of Archaeology, Oxford (more than 75 cartons of material), which came to the Institute after Jope’s death, led to the discovery of a significant quantity of Jacobsthal’s notes, photographs and text relating to ECABI, including draft chapters Jacobsthal had typed out and corrected which covered British mirrors, brooches, basketry decoration, the ‘first wave’ of Celtic art, swords, daggers, dishes, buckets, and continental links. This material (catalogued as the Jope/Jacobsthal Archive) is still being assessed, but in the meantime the paper published below gives an insight into how far Jacobsthal had progressed in this research as early as 1943, as well as highlighting some of the difficulties

Jacobsthal faced in advancing his work – difficulties which ultimately proved insurmountable.

Part 1: The genesis of Leopold Bloom, and Early Celtic Art in the British Isles If the final version of ‘“Leopold Bloom I” and the Hungarian Sword Style’ presented to Beazley still exists, it is not in our Archive, nor in the Beazley Archive at the Classical Art Research Centre, Oxford. The Jope/Jacobsthal Archive has several typed versions of ‘Leopold Bloom’ which are almost identical. One is specifically dated to 1943. It is this dated version that we have presented for publication. Jacobsthal circulated the manuscript of the paper to a few select friends, but resisted pressure to publish it. The most concerted effort to push Jacobsthal to print came from his very good friend Hugh Hencken at the Peabody Museum of Archaeology and Ethnology, Harvard University. Letters in the Jacobsthal Archive, and corresponding letters in the Peabody Museum Archive. track the exchanges between the two. Sometime before July 1949, Hencken contacted Jotham Johnson, a classical archaeologist and first editor of Archaeology, about the paper. Archaeology, published by the American Institute of Archaeology (of which Hencken was then President), was a brand-new journal which only came into existence in 1948: Jotham had been careful to make sure the first volume was ‘safe’, but now he was on the look-out for innovative writing which would help to build a readership and establish the credentials of the journal. He was excited about ‘Leopold Bloom’, and had plans for the paper: ‘BY ALL MEANS get hold of the Jacobsthal ‘Story of Leopold Bloom, the First’ with or without illustrations – if illustrations come with it, so much the better, but this sort of thing could even be done without. If any foreword to smooth the reader’s path is necessary, I will get it written. This is by every means a must and if possible I would like to get it by September 10th, for publication in the December issue…’ (Jac. Archive, Box 49, Letter 182: 19.7.1949). Jacobsthal must have given the proposal some serious thought, because in August 1949 he writes to Hencken and mentions that ‘yesterday I was working on Leopold Bloom’ (Jac. Archive, Box 49). But at some point, reviewing the paper, Jacobsthal decided against. Hencken, disappointed, pleaded with Jacobsthal to reconsider: Dear Jacobsthal, I was afraid you would feel that way about ‘Leopold Bloom’. But you see, we believe in fun in archaeology. I should have though that anyone as distinguished as yourself would have no fear of being considered anything but a serious scholar. It is no doubt lèse majesté, but the picture of you and Beazley chuckling over this archaeological fantasy would have been marvellous. Please excuse our crazy, trans-Atlantic notions! Yours ever, Hugh Hencken (Jac. Archive, Box 49, Letter 121: 29.8.1949).

25.  ‘Leopold Bloom I’ and the Hungarian Sword Style Ultimately, Jacobsthal decided against publication, instead offering a ‘safe’ paper on Greek pins. Jacobsthal’s unease about his work on British scabbards may have underpinned his reluctance to see this paper in print, in spite of the efforts of his colleagues to persuade him. As he wrote to Hencken, Beazley was a great admirer of Joyce and ‘gets something more or less bogus every year’ … but he cautions this is ‘dangerous stuff – shown to the very few’. There may have been a number of reasons, personal, social or political, why the paper contained ‘dangerous stuff’ – but Jacobsthal was acutely aware of one particular ‘problem’ with a paper of this sort: I have from time to time shown it to a few especially well qualified friends … but for good reasons never thought of publication; those who enjoy the Joycean fun do not happen to know of the swords from Lisnacrogher, of the Witham shields, of the chamfrein from Torrs and the other works which I assign to my ‘Hungaryzing School’ in 3rd century Britain. On the other hand, those who would say ‘why does this silly Jacobsthal not publish his quite interesting stuff and ideas in a sensible form, well-documented, but mix it up with this nonsense: he seems to be one of those degenerate highbrows who take that Joyce for a poet…. (I see Sir Cyril [Fox]’s and Lady Eileen’s faces…..). I do not bother much about what people think of me and my papers, but I really feel that this paper is dangerous stuff, should be marked with a skull and two bones crossed and kept in the special cupboard in my shop and shown only to the very, very few… (Jac. Archive, Box 49, Letter 184: Christ Church 18.8.1949)

Jacobsthal may have claimed that he did not care about other people’s opinions of him, but his archive shows him to have been always very cautious of revealing too much about himself or his inner thoughts in public (Jacobsthal, Ulmschneider and Crawford 2011), and the experience of life as an enemy alien in Oxford during the war had done nothing to encourage openness. Nor was this the first paper Jacobsthal had written and shown to his friends but refused to publish: his account of internment on the Isle of Man was, in his view, too libellous to put into print (OUP Archive, Letter of 31.12.1940: the account was eventually published in 1992 (Jacobsthal 1992); see also Ulmschneider and Crawford 2012 for further discussion of the text). The problem underpinning Jacobsthal’s decision not to publish, however, rested in no small part on the issue of scholarship – though there are likely to have been political reasons as well: Jacobsthal had seen at first hand the dangerous outcome of claims of cultural origins based on archaeological artefacts (Crawford and Ulmschneider 2011). While Jacobsthal was confident that there was something useful in his paper – which is why he circulated it amongst friends – it was also unfinished, even unfinishable, research. In the opening section of ‘Leopold Bloom’, Jacobsthal explains that he has not been able to see all the artefacts for himself, a sentence which hides a long story of frustrated

215

attempts to carry out research in wartime conditions. The Jacobsthal Archive retains some examples of Jacobsthal’s attempts to commit research. Letters to museums asking for photographs of objects are returned with brusque replies, usually to the effect that objects were in storage and museum curators were away fighting – the responses to Jacobsthal’s requests for help are of the pursed-lip ‘don’t you know there is a war on?’ variety. As Jacobsthal lamented in 1940, ‘for years to come the objects will be buried in the basements of British museums, even the negatives in museums, for their greater part, are inaccessible, the materials required become scarcer and scarcer’ (J/J Archive, Box 3, Letter 157). Jacobsthal’s preference, of course, was to examine the objects himself, but from his internment to the end of the war he was under curfew as an enemy alien, and his movements were restricted. Records in Jacobsthal’s file at the National Archive (those which were released under the Freedom of Information Act: parts of this archive are redacted until 2031 for ‘reasons of national security’) show that Jacobsthal attempted to use the influence of his well-placed friends to help him travel. In February 1944, Robin Dundas, Master of Christ Church, Oxford, wrote a personal letter to his friend Sir John at the Home Office on Jacobsthal’s behalf. Jacobsthal wanted to be able to buy a map of Britain to plot out Celtic art find-spots on it, as well as travel to London to see Cyril Fox and on to Cambridge: he also wanted to be able to travel around Wales. Dundas made the plea that a harmless scholar like Jacobsthal should be able to work unhampered. The predictable response was that the idea of letting German scholars travel around wartime Britain armed with notebooks and maps was not going to be received sympathetically, not least, Dundas was reminded, because the Germans had given Celtic archaeology a bad name by using it as a cover for other activities, particularly in Ireland (TNA Archive, PRO HO 405/24418). Unable to complete his research in Britain to his own high standards, Jacobsthal was also frustrated by the lack of extant evidence to show how an art style originating in Hungary could be present in the British Isles. In fact, it would be many decades before archaeological evidence to support Jacobsthal’s hypothesis of long-distance transmission would be found (see Fitzpatrick and Schönfelder this volume: cf. Henry 1965, 27; Megaw 1970a, 29, 148; De Navarro 1972, 291ff.; Frey with Megaw 1976; Szabó 1977, 219). Jacobsthal’s vexation that he could not find the evidence to support what he felt to be a plausible hypothesis was not recent: he had already expressed the problem in his publication of the Witham sword: ‘This principle of scabbard decoration [as seen on the Witham scabbard] does not occur here for the first time: the Celtic sword from Hungary [Plate C], 200–300 years older, shows that this feature, like others of Celtic art in the British Isles, has its roots on the Continent, not in France, but in Hungary and Switzerland, a remarkable fact which I can only state here, but not try to explain’ (Jacobsthal 1939, 28).

216

Paul Jacobsthal with Katharina Ulmschneider and Sally Crawford

Jacobsthal’s research on British and Irish sword scabbards also formed the core of a chapter which Jacobsthal had written, at least in draft form, for ECABI. This chapter, typed on Jacobsthal’s typewriter and with corrections in his own distinctive handwriting, also bear the marks of Jope’s scrutiny, with his pencilled comments ranging from ‘use this’ to ‘tripe’. As in ‘Leopold Bloom I’, Jacobsthal used an arthistorical approach to the scabbards, noting comparisons in small decorative details to argue for links between Hungary and the British Isles: ‘So we have clear evidence of Swiss and Hungarian connexions of the workshop or school which produced the four scabbards’, he claimed. This received Jope’s marginal comment: ‘You’d need something better than this, chum’ – a more straightforward expression of the problem of evidence than Jope’s careful distancing from the theory in his 1954 publication on the River Bann sword (‘Dr. Jacobsthal’s demonstration that one of the main roots of the Early Celtic metalwork style of Britain is in Switzerland, Hungary and east-central Europe, rather than in north France, remains without an explanation except perhaps in terms of migrating master-craftsmen’ (Jope 1954, 84). Letters in the Jacobsthal Archive suggest that Jacobsthal, perhaps more than any other scholar in the British Isles at the time, was active in keeping abreast with developments in Hungarian archaeology, and his knowledge of Hungarian material for comparison with the material from the British Isles was unique, at least up to the outbreak of war. For example, Jacobsthal was sent photographs and information by Joseph Fleissig, a businessman from Budapest with a private collection, who had an important correspondence with Jacobsthal in February 1939, giving details of the Fleissig artefacts. Ernst Lauringer, Dr. Fr. Tempa and Lajos Marton all sent photographs and letters about Celtic metalwork: Marton (d. 1934) hoped to welcome Jacobsthal to Budapest soon (Jac. Archive, Box 14, Letter 7). Elemier von Kund (Jac. Archive, Box 14, Letter 27: 30.6.1939) sent photographs of his excavation of the Celtic cemetery at Kosd, and mentioned that he was already working on a monograph of the cemetery with Professor Ferenc von Tompa. Stephan Zichy, Director of the Hungarian Historical Museum, Budapest, sent information on a gold helmet, while Andrew Alfoeldi at Peter Pazmany University, Budapest, sent plates from an article on horse burials from Brigetio (Jac. Archive, Box 44, Letter 1: 18.7.1947) In ‘Leopold Bloom I’, Jacobsthal showed that he was well aware that his claims for a Hungarian influence based on a comparison of stylistic detail had a fragile foundation. He began the whole paper with a modified quote from Homer to warn the reader: This story is not true | nor did you travel in well-benched ships | nor did you come to the citadels of Ireland (trans.: Ewan Bowie). That the quotation comes before the ‘serious’ archaeology rather than before the ‘humorous’ section on Bloom reflects Jacobsthal’s intention to signal that the whole work was an idea incapable of proof.

This is not to say that Jacobsthal did not return to the ‘problem’ of explaining the Hungarian/Irish links, like an itch that must be scratched. From the Jope/Jacobsthal Archive again come fragments of a draft copy of a lecture given by Jacobsthal at Aberdeen in January 1949. Here, rather than the sardonic Joycean Bloom, Jacobsthal proposes another, more Tolkeinesque scenario: The Hungarian chieftain wore a sword which caught the eye of the bride’s father: the old man asked the future son-in-law to send him his armourer: so he did, and this was the origin of the Hungarizing [sic] style in Britain. This is a construction – if you like, fiction, but we know that the Celts, who in the third century formed a continuous belt stretching from this country over France, Germany, Bohemia, the Danunian lands across the Bosporus to Galatia in Asia minor, were in close contact with one another; and my story has good analogies in other periods of history (J/J Archive, Box 13/1).

Once again, Jacobsthal struggled with the limits of the archaeological evidence: it is a fiction if you like, but not for Jacobsthal. He worked on an instinct that this, or something like this, must be the explanation. Fascinatingly, in the context of this volume in honour of Vincent Megaw, Jacobsthal continued to touch on two other thorny topics which, he felt, had still to be properly understood in Celtic art. The first was what Megaw termed ‘the elusive image in La Tene art’ (Megaw 1970a; 1970b). Of the Witham shield, Jacobsthal pointed out that it was often illustrated, but always ‘horizontal or upside down’, so ‘people failed to see that there is a face at the upper end, with popping eyes, a nose and a long-drawn-out beard, passing over into the midrib’. The second is the Torrs chamfrain – Jacobsthal returned again to the proposal that it was found in the form it still has, with an intended use as a horse-mask. But as with what he calls the ‘Hungarizing’ style, archaeology in the 1950s could not support his case, and he resorted to classical, Asian and medieval models for comparisons. Jacobsthal’s ‘problem’ with the ‘Hungarizing influence’ was one which has persisted through the following decades of scholarship. Vincent Megaw identified basketry in-filling – something Jacobsthal also noted on British metalwork – as an indication that Newnham Croft was non-continental (Megaw 1983), but recanted in 2008 (Megaw and Megaw 2008, 45) because his recent detailed examination of Hungarian swords revealed basketry ‘is a feature of a substyle of Hungarian swords’. Finally, there is the question of how many Iron Age scholars, apart from Hencken and Jope, were familiar with Jacobsthal’s unpublished text? On the basis of Jacobsthal’s wide circle of academic contacts, there must be the suspicion that the gist of Jacobsthal’s theories on the Hungarizing influence in Early Celtic art in the British Isles must have been fairly well known, and it would be an interesting conference-dinner game for Iron Age specialists to try to spot it in publications after 1945 – De Navarro in The

25.  ‘Leopold Bloom I’ and the Hungarian Sword Style Celts in Britain and their Art, for example, on the possible continental sources of his Insular Style IV, arguing that it was: ‘Probably due to the bringing in of foreign artists… from Hungary and Switzerland… Its creators were obviously working for rich patrons, either princely or priestly’ – does this have a hint of ‘Leopold Bloom I’ (De Navarro 1952, 75)?

Part 2: Leopold Bloom – some notes The second part of Jacobsthal’s paper for Beazley follows the fortunes of ‘Leopold Bloom I’, Hungarian swordsmith of the Iron Age. It shows Jacobsthal’s clever, sardonic humour – the kind of entertaining cleverness which led Maurice Bowra to declare he was the ‘most lively and most fascinating’ of the German Jewish refugees who came to Oxford (Bowra 1966, 300), and Sir John Cecil Masterman (tutor of Modern History at Christ Church Oxford, and noted as the Chairman of the Twenty Committee, which ran the ‘XX System’ during WWII, controlling double agents in Britain) to write ‘true’ against the comment in his personal copy of Bowra’s Memoirs (Worcester College Library). We are grateful to Professor Ewen Bowie of Corpus Christi College for the following comments on Jacobsthal’s opening quotation: ‘The three lines of Greek are the lines of Stesichorus’ palinode (Davies 1991, fr. 192, with ‘Ireland’ replacing ‘Troy’) quoted by Plato’s Socrates in his Phaedrus before HIS palinode (and the only lines of the palinode then published: we have now a bit more on papyrus) – the ‘you’ is Helen (Stesichorus used them to reject standard views of Helen having gone to Troy and instead created or at least elaborated the story she went to Egypt. I don’t know if there is anything palinodic about the skit, or if Jacobsthal simply meant he was about to spin fiction’ (Bowie 2013, pers. comm.). There are conflicting reports on the standard of Jacobsthal’s English, but Jacobsthal’s comfortable familiarity with James Joyce’s Ulysses backs up the impression given by his letters in the Institute Archive, that his English was entirely competent, if occasionally idiosyncratic. Jacobsthal did not choose Ulysses as his model for its relevance to metalworking. Professor David Bradshaw noted that: ‘there is only one reference to a scabbard in Ulysses – in episode 14, ‘Oxen of the Sun’ – and that passing reference does not seem to fit with this document in any way. My guess is that Jacobsthal is casting himself in the mould of Leopold Bloom, whose father arrived in Ireland from Hungary (but not, as far as the text tells us, via Hook of Holland and Lincoln) and who is certainly Jewish. Bloom is subject to anti-Semitic comments in the novel, but only by plain bigots’ (Bradshaw 2013, pers. comm.). Joyce’s Bloom was born in Szombathely, apparently chosen as his fictional birthplace in honour of a man he met

217

in Trieste called Marino Szombathely, a Jewish Hungarian immigrant who was translating The Odyssey into Italian (McCourt 2000). Jacobsthal attributed an important scabbard to Szombathely (ECA 1994, no.127), but Jacobsthal’s Bloom begins his story in Bölcske, a Hungarian village on the western bank of the River Danube in the comitatus of Tolna. This location was chosen with care, as the Bölcske scabbard (Jacobsthal 1944, 95–6) represented a crucial stage in the development of the ‘Hungarian Sword Style’ (but see also Mansel Spratling: ‘Jacobsthal (1944, 95–6) argued that the style was developed from Waldalgesheim ornament and that it presupposed its existence. However, comparison of the Boelcske scabbard’s ornament… with such Hellenistic floral scroll work as is seen on the Hoby jug… suggests rather that a renewal of Hellenising influence may have been the prime source of the layouts and details of the ‘Hungarian Sword Style’ (Spratling 1972, 290)). In the same vein, Jacobsthal’s choice of Lincoln as the home of ‘Bloom and Co’ reflects the findspot of, amongst other artefacts, the Sutton Trent and Witham swords. Bloom’s Swiss partner, Heierli, references Jakob Heierli (1853–1912), prehistoric archaeologist at the University of Zurich, and author of Urgeschichte der Schweiz (1901). He excavated at Schaltenrail hill, Switzerland: an early La Tène sword from the 5th century was found here. Jacobsthal is also giving the nod to Heierli as a very early proponent of the links between Hungary and Switzerland. Writing about the sword from Martigny, Heierli noted that: ‘Die Form, noch mehr aber das Spiral-Ornament auf dem Griff dieses Schwertes beweist, dass es kein einheimisches Produkt ist, sondern aus dem Osten stammt. Eine fast identische Waffe, ebenfalls aus Bronze, stammt vom Semmering in Österreich und in Ungarn ist dieser Schwerttypus ganz besonders häufig’ (Heierli 1901, 237). Bloom and Heierli’s English apprentice, John Smith, may be intended as ‘everyman’, but there may be a reference here to John Alexander Smith (1863–1939), first lecturer of philosophy at Balliol College, Oxford, and from 1910 Oxford Waynflete professor of Moral and Metaphysical Philosophy, based at Magdalen College. An anecdote about Smith told by Harold Macmillan recounts that he began a lecture course in 1914 with the comment: ‘Nothing that you will learn in the course of your studies will be of the slightest possible use to you in after life – save only this – if you work hard and diligently you should be able to detect when a man is talking rot, and that, in my view, is the main, if not the sole, purpose of education’ (Horne 1988, 27). Is Jacobsthal enjoying a private joke here with Beazley, who had been reading Literae Humaniores, including Philosophy, at Balliol College just a few years before Harold Macmillan? If so, Jacobsthal may also be joking again that Beazley needs to work out which parts of this paper are ‘rot’. At the end of his life, Bloom reminds his wife that: ‘On revient toujours a son premier amour’. The quote is from one

Paul Jacobsthal with Katharina Ulmschneider and Sally Crawford

218

of the less well known works of D. H. Lawrence, Fantasia of the Unconscious (1922). The novelist is an apt choice, being closely linked with Nottingham, where the fictional Smith was said to have worked. The text of ‘Leopold Bloom I’ which survives in the Jacobsthal Archive lacks figures, and in some places, Jacobsthal left the figure numbers blank. We have followed Jotham Johnson’s suggestion that the paper can be read without the images: we have indicated Jacobsthal’s figure scheme with numbers in square brackets, but no corresponding images are offered.

NEW LIGHT ON LIFE AND WORK OF LEOPOLD BLOOM 1 By Paul Jacobsthal 13th September 1943 to J. D. Beazley. οὐκ ἔστ’ ἔτυμος λόγος οὗτος, οὐδ’ ἔβας ἐν νηυσὶν ἐυσσέλμοις οὐδ’ ἵκεο πέργαμα Iouerνhíā

The rise of Celtic art in the British Isles was the work of immigrant artists, who came from France, Switzerland and Hungary. Connexion with France is natural and accordance with facts which can be established by other than stylistic evidence: there are the Parisii in Yorkshire, and there is likeness of burial-customs between Britain and the region of the Marne. Immigration of artists from Switzerland is surprising, immigration from Hungary paradoxical but capable of exact proof. The Hungarian contribution was the most important: I interpret four scabbards of Hungarian Style. Scabbard (A). British Museum. From the Crannog at Lisnacrogher, near Brouchshane, Co. Antrim. [Fig. 1]. Scabbard (B). Belfast Museum. Same provenience as (A). [Fig. 2]. Scabbard (C). Hull Museum. From the Trent, near Sutton. [Fig. 3]. Scabbard (D). Belfast Museum. Same provenience as (A) and (B). [Fig. 4]. I have examined and photographed only (A): for the other scabbards I have to rely on the published drawings; the faithfulness of those of (B) and (D) can be checked by the comparison of photographs of (A) with the drawing made by the same draughtsman who drew (B) and (D); the drawing of (C) gives the impression of correctness. (A). There are six matrices, now empty: their inlay was more probably coral than enamel (Raftery (1940,

277) speaks of enamel studs, representing the eyes of stylized snakes: I have never seen a Celtic snake chape. In Marburger Studien (Raftery 1938, 203) he calls the chape ‘zoomorphic’). The engraved decoration consists of three complete pairs of lyres, foot to foot, and one single lyre below, above the top of the chape: its counterpart, owing to lack of space within the chape, is replaced by a chain of four tiny lyres, all upright: the foot of each of them is at the same time the head of the neighbouring lyre: [Fig. 5] (from ECA, pl. 272, pattern 341) gives an Early Celtic example, a device from the chariot-tomb at La Gorge-Meillet. The little lyres are poorer than the larger ones and lack trimmings. From the lowermost lyre leaves grow, similar to those which rise from the matrices near the scabbard-top. From the tip of each leaf a wave line runs towards the point of the scabbard. I describe the ornament in still greater detail. Along the arms of the lyres prickles stand, alternately inside and outside, according to the space available. Long sickle leaves swing from the foot and head of the lyres, the former within, the latter without: the sickle leaves effect a wave movement, intersecting with the rhythm of the lyres. The leaves do not branch off from spiral involutions – as would be the case in classical ornament – but from a member inserted, which might be compared with a Boeotian shield, a double pelta, or a double axe, [Fig. 6]. This motif, not without early forerunners [Fig. 7] (ECA pl. 265, patterns 101–4) was used as a clasp at the junction of tendrils in the Waldalgesheim Style ([Figs. 8a, 8b], from ECA pl. 277, patterns 447, 450b), and in the Hungarian Sword style Fig. 9. It appears in its clearest and classic form on our scabbard. (B). [Fig. 2]. Wave lines run along the edges of the scabbard down to the point where the chape, which is not preserved, began. The decoration consists of four S tendrils, the lowermost shorter, but with a tail of clever design appended. The tendrils bear prickles, and at long intervals, bipartite knots, alternately on the right and left side. Where two tendrils join, their axil-fillings form a double axe. At their free ends the tendrils swing out in an S shaped leaf swinging inwards, on their outer side they here coalesce with a half-flower (ECA 94) [Fig. 10a]. I illustrate in Figs. 10b and c, two examples of Waldalgesheim Style (from ECA pl. 277, patterns 453, 455), and show in [Fig. 11] an Hungarian sword (ECA no.115, pls. 67–8). (C). [Fig. 3]. The engraved decoration covers the scabbard for its whole length down to the point: the chape, which is now missing, hid part of the ornament below. The two halves into which the sheath is cut by the plastic midrib are decorated separately: but near the top the rightand left-hand patterns make together one ornament; in the other six sections only the right-hand or left-hand panel bears ornament, and the opposite side is cross-hatched: this contrast introduces a half-latent zigzag rhythm into the decoration. The pattern in the square top panel mentioned is a four-leaf star: the motif is found in the same place

25.  ‘Leopold Bloom I’ and the Hungarian Sword Style on British scabbards earlier and later: [Fig. 13] shows a scabbard from Minster Ditch, Oxfordshire (Salzman 1939, pl. 12d and 12e), which, with a certain reserve, I date to the fourth century B.C.; the later piece comes from Cambridge (Salzman 1938, 224, fig. 25, 9; Fox 1923, 107). The leafstar, [Fig. 14] is here translated into the peculiar dialect of this school of armourers: inside the lancet leaves, another leaf, filled with little circles, clinging to the outer contour. Notice that the leaf on the top right is bungled. The inner leaves are a development of a type frequent in Early Celtic art, [Fig. 15]. The pattern recurs identically in the following section. [Fig. 16c], illustrates a similar motive used for the adornment of the tendrils. The tendrils decrease in length towards the point of the scabbard; no two of them are alike. The uppermost runs through in two movements: sideways it ends in the pattern, in the middle the space between the midrib and the point where the tendril touches the edge is filled with patterns, [Fig. 14]. The second tendril is formed by two esses; each of these ends outside in pattern, [Fig. ]; in the middle they are clamped by pattern, [Fig. 16a]. The third tendril is similar but the treatment of the middle junction differs. In the fourth tendril, [Fig. 16b], the two parts do not join in the middle, but run dead against an horizontal line. The ornaments of the two following sections are tiny, slightly botched and need not be described. (D). The decoration consists of a tendril with seven large involutions; at the top it passes over into an upright split palmette, a motif frequent in the Waldalgesheim Style; the circumscription of the palmette grows upwards two sickle leaves, which reach the uppermost point of the scabbard. Where the chape begins, the tendril ends; below, a cup spiral with a tail, and lower down, within the chape-lozenge, two wave lines. The top contour of the scabbard is accompanied by a wave line. What I called a tendril is, in point of fact, a sequence of scrolls, alternatively right-hand upwards and left-hand downwards. Where two scrolls touch each other, the axil on either side is filled: the fillings form together a kind of double axe, similar to those on scabbard (B). To the concave lateral ends of the double axes little scrolls of peculiar form, [Fig. 17], are attached, those on the right side of the scabbard pointing upwards, those on the left side downwards (notice an irregularity at the second double axe from the top). So far I have taken the ornament as static: it is possible, though less easy, to see it in motion: you then discover that only double axe 1, 3 , 5 – counting from top – are the starting point of two tendrils, while 2, 4, 6 remain inactive. The split palmette, the double axe, the scroll, and the cup spiral are covered with basketry-work: that on the scrolls is interrupted by a spiral, that of the uppermost scroll, which alone is pelta-shaped in part, at two points. The four scabbards, in outline and form of chape, which is preserved in (A) and (D), are alike and correspond to Swiss and Hungarian swords of so-called La Tène 1, b, c – 11 periods, i.e. about the middle of the third century

219

B.C. (A) and (D) show wave lines along the upper edge, (B) sideways; (A) and (B) have barbed tendrils; (B) and (C) share the double axes; (D) in essential features stands apart. The following traits have analogies in West Celtic art and seem not to occur on Hungarian scabbards. (1) Tendril decoration: La Tène (Vouga 1923 pl.1, 3; pl. 2. 2); France, (Maubranche, Cher – Déchelette 1908–1914, fig. 463, 3), here not exactly tendrils but intersecting wave lines, running the whole length of the scabbard. The Hungarian scabbard, ECA no. 127, pl. 70, which covers the whole sheath with stamped patterns, is not comparable. (2) The wave lines along the scabbard edge have numerous analogies on Swiss swords (Vouga 1923). (3) For the sickle leaves on (B), Fig. 18, compare the Swiss scabbard Fig. 19 (from Déchelette 1908–1914, fig. 463, 6). (The knots on (B), Fig. 20a, have possibly an analogy in the bronze mountings from the chariottomb at Nanterre, Fig. 20b, (from ECA no. 174,c, pl. 110)). The following features are Hungarian. (1) Coalescence of half-flowers with tendrils ((B), [Fig. 10a]). The motif originates in the Waldalgesheim Style (see above and [Figs. 10b and c]), and passes over into the Sword Style: [Fig. 11] shows a detail of the scabbard from Kis Koszeg, ECA no. 115, pl. 67–8. (2) The double axe, [Fig.6]. This motif, like the foregoing, survives from the Waldalgesheim Style into the Sword Style: the scabbard from Simunovec, [Fig. 9] (from ECA no. 115, pl. 68) gives an example: I explain it in the drawing [Fig. 9b]. (3) The fillings on (A) [Fig. 20], three leaves pointing outwards, have an analogy on the Hungarian scabbard, [Fig. 9] (there bungled). (4) The spirals filling leaves on (B) recall the scroll on the uppermost tongue of the scabbard from Simunovec, [Fig. 9]. (5) The cup spiral on (D), below the top matrices of the chape, [Fig. 21], is a pattern with no analogy in Celtic ornament: it copies an Hungarian chape of this form like the piece from Kosd, tomb 34, in the Kund collection, Budapest, [Fig. 22]. The original type showed two birds: [Fig. 23], a scabbard from Hatvan Boldog (Márton 1933/4, 158, pl. 55; and Márton 1933, pl. 27,1). Other scabbards from the Kosd graves illustrate the gradual degeneration of the birds. (6) The leaves near the point of (B), [Fig. 18], are possibly connected with the plastic motif on the sword from Roje, [Fig. 24], (ECA no. 114, pl. 67), which they transpose into a flat, two-dimensional pattern. (7) Barbed tendrils. Though there is a Swiss example on a scabbard from La Tène (Gross 1886, pl. 1,7; not clear in Vouga 1923, 41, fig. 7), where the prickles stand on the contours of the body and neck of debased birds, the motif is preponderantly Hungarian: to the examples discussed and illustrated in last year’s Festschrift I add the scabbard, [Fig. 9], where the prickles accompany the tendril which runs from bottom left to top right into the tongue-shaped top finial of the sheath. (8) The motif in the fourth section of (C), [Fig. 16b], where the tendrils broadening run against a straight line, has an analogy in the scabbard from Boelske, [Fig. 25], (from ECA no. 116, pl. 67).

220

Paul Jacobsthal with Katharina Ulmschneider and Sally Crawford

Weightier than the comparison of details is the observation that the handwriting on our scabbards is more like that on Hungarian than on Swiss scabbards: see [Fig.25], which I have adduced for detail. So we have evidence of Swiss and Hungarian armourers collaborating in a British workshop in casting, hammering, engraving scabbards and forging swords. It was but natural that one called armourers from those countries where the best swords were manufactured. These were not artists who had to go abroad for lack of success at home: of our four scabbards (B) and (C) are at least on the level of average Hungarian and Swiss swords, and (A) and (D) are above it. (D), on the one hand, is inseparably connected with the other pieces by the following features: the form of the chape; the wave line along the upper edge; the little member appended sideways to the double axes, [Fig.17], which recurs in a related chape as filling of leaves on (B), [Fig.2]. On the other hand it has peculiarities which sever it from the rest and make it an object of importance for the history of Celtic art in Britain. I enumerate these features without discussing them in full. (1) The uppermost scroll, at its bottom, broadens into a kind of pelta, [Fig.26]: a motif which played a great role in Celtic art in a later age. (2) The ornamentation of (D) is less light and sober, more exuberant than that of the other pieces. The ‘basketry-work’ is one of the most characteristic elements of Celtic decoration in the British Isles. The majority of the objects decorated thus no doubt belong to a later period, the first century B.C. and the first century A.D., but to force a late date on our scabbard solely because of the appearance of these patterns, is putting the cart before the horse. (So did Dr. Mahr (1932, 14) and following him Raftery (1938, 279), who dates the scabbards about 100 B.C.). There is no external evidence for the date of any of the four scabbards: Lisnacrogher was not excavated properly – a sad chapter of Irish archaeology – and (C) was dredged from the Trent. I do not see how to avoid the conclusion that ‘basketry work’ and pelta tendrils were invented in third century Britain. It is unlikely that the master of scabbard (D) invented them: he probably got them from elsewhere and blended them with the Hungarian elements. Are the Lisnacrogher scabbards Irish? (Leeds (1933) takes them for made in Ireland but in British tradition. Raftery (1938, 277), claims them as Irish: ‘The decoration is purely insular and the chapes are peculiarly Irish. They represent the stylised heads of serpents with two enamel studs for the eyes. Obviously they have a long story behind them, and though no wooden or other organic prototypes are known, the high state of development would seem to suggest such’. I have pointed out that the decoration is not Irish, and that there are no snakes or enamel studs. Raftery overrates the role wood played in Celtic crafts. Francoise Henry (1933, 78) takes (B) for a continental import without

giving reasons). The piece from the Trent is inseparably connected with them: either the scabbard from the Trent is an Irish importation into Lincolnshire, which is unlikely, or the Lisnacrogher scabbards were made in Britain and exported to Ulster. That the four swords were made in Britain can be established by comparison with other objects. (1) The Chamfrein from Torrs, Kirkcudbrightshire, once in the possession of Sir Walter Scott, now in Edinburgh Museum. [Fig.27], shows the ornaments engraved on the horns in a drawing made eighty years ago: though it is very incorrect, enough motives of the Hungarian style are recogizable. The following patterns which we have encountered on the scabbards reappear on the chamfrein: [Figs.17, 20], and there is again a great likeness to the Hungarian scabbard, [Fig.25]. (2) The round shield-boss from the Thames at Wandsworth (Jacobsthal 1939): the Hungarian character of part of the engraved ornaments and their close connexion with those on the chamfrein are obvious. From the shield a way leads to the scabbard of the sword from the River Witham (Jacobsthal 1939): sword, shield and chamfrein share the half palmette with the scalloped outline, a motif not found in Hungary. The Hungarian element in the engraved decoration of the Witham scabbard is slight and confined to the spiral with adjoining fillings, top left; Hungarian, however, is the asymmetrical, diagonal run of the embossed ornament – its run only, for here, as in the shield and the chamfrein the modelling, vigorous and tender, is British, and has not its like in Celtic art on the Continent. All these weapons (and a few others which I have not mentioned) found widely distributed, in Ireland, in the Witham, the Trent, and the Thames, were made in England, in a few interconnected workshops: they illustrate the making of a new British style out of Swiss and Hungarian ingredients.

[Part II] I have to give out hero his place in history. In 251 B.C., the sixth year of the slump following the definite conquest of Hungary by the Celts, Leopold Bloom I, not heeding the paternal admonitions of Samuel Bloom, left Boelcske, Komitat Tolna, for England. He travelled on the route which today runs via Vienna, Prague, Nuremberg, Cologne, Hook of Holland; on his journey he declined tempting offers from Bohemian and Rhenish chieftains to enter their services: his prophetic mind foresaw trouble on the Continent. After his arrival he first tried his luck in other parts of Britain until he made up his mind to settle in Lincoln: in 249 B.C. he started the business L. S. Bloom and Co., Co being Mrs. Bloom. There was already established in Lincoln the Swiss armourer Heierli: in 247 B.C. Heierli preferred to merge with Bloom: the firm was now L. S. Bloom and J. Heierli.

25.  ‘Leopold Bloom I’ and the Hungarian Sword Style At the time there was much fighting in the country and a boom in the armourer trade: they could not satisfy the demand and had to increase their staff. It was Bloom and not his pigheaded Swiss partner, who saw that British labour was not only cheaper but that these British Celts were gifted for the job and had an inventive mind. The best of their apprentices was John Smith, who had worked before in a British business in Nottingham. Scabbard (D), which was afterwards sold through Bloom’s Stranraer agent to Lisnacrogher, was Smith’s work. Smith did so well that in 241 B.C. they made him a partner: the firm now became L. S. Bloom, J. Heierli and J. Smith, Ltd. In 235 B.C. Heierli died; Leopold Bloom retired, disinteresting himself in the arts and the production of tools satisfying the barbaric instincts of gentiles. He devoted the rest of his days to study the Talmud and debauches: ‘on revient toujours…’ he used to say to Mrs. Bloom: there was a tacit understanding that she referred the allusion to the Talmud only. He died in 225 B.C.

Acknowledgements SC and KU would like to thank Chris Gosden, Jas Elsner, Ewen Bowie, David Bradshaw, Martin Maw at Oxford University Press Archive, Peter Sawyer and Claudia Wagner at the Beazley Archive, Joanna Parker and Emma Goodrum at Worcester College Library and Archive, Marissa Kings for her work at the Peabody Archive, and all the Institute of Archaeology Archive volunteers, especially Roelie Reed and Pam England. We are grateful to the Reva and David Logan Foundation, the Heritage Lottery Fund, and the Fell Fund for their generous support.

Bibliography Archives BA = Beazley Archive, Classical Art Research Centre, Oxford Jac = Jacobsthal Archive, Institute of Archaeology, Oxford J/J = Jope/Jacobsthal: Jacobsthal material from the Jope Archive, Institute of Archaeology, Oxford Jope = Jope Archive, Institute of Archaeology, Oxford OUP = Oxford University Press Archive TNA = The National Archive, Kew. Bowra, M. 1966. Memories 1898–1939. London: Weidenfeld and Nicolson. Crawford, S. and Ulmschneider, K. 2011. Paul Jacobsthal’s Early Celtic Art, his anonymous co-author, and National Socialism: new evidence from the archives, Antiquity 85, 129–141. Davies, M. (ed.) 1991. Poetarum Melicorum Graecorum Frag­ menta, Vol. 1. Oxford: Oxford University Press. Déchelette, J. 1908–1914. Manuel d’archéologie préhistorique, celtique et gallo-romaine. Paris: A. Picard et fils. De Navarro, J. M. 1952 The Celts in Britain and their Art, in M. P. Charlesworth et al. Bell & Sons.

221

De Navarro, J. M. 1972. The finds from the site of La Tene, I: Scabbards and the swords found in them. London: Oxford University Press. Fox, C. F. 1923. The Archaeology of the Cambridge Region. Cambridge: Cambridge University Press. Frey, O.-H., with Megaw, J. V. S. 1976. Palmette and circle: early Celtic art in Britain and its continental background, Proceedings of the Prehistoric Society 42, 47–65. Gross, V. 1886. La Tène. Un oppidum helvète. Paris: Fetscherin and Chuit. Heierli, J. 1901. Urgeschichte der Schweiz. Zürich: A. Müller. Henry, F. 1933. Emailleurs d’Occident, Préhistoire 2.1, 64–146. Henry, F. 1965. Irish art in the Early Christian period (to A D 800). London: Methuen. Horne, A. 1988. Macmillan: the Official Biography. London: Macmillan. Jacobsthal, P. 1939. The Witham sword, Burlington Magazine 75 (no. 436), 28–31. Jacobsthal, P. 1944. Early Celtic Art. Oxford: The Clarendon Press. Jacobsthal, P. 1992. The long vac 1940. In S. Cooper (ed.) The Refugee scholars: conversations with Tess Simpson, 198–228. Leeds: Morland Books. Jacobsthal, P., introduction and translation by Ulmschneider, K. and Crawford, S. 2011. Childhood, art and education in late nineteenth century Berlin: an autobiographical account, Childhood in the Past 4, 11–30. Jope, E. M. 1954. An Iron Age Decorated Sword-Scabbard from the River Bann at Toome, Ulster Journal of Archaeology Third Series 17, 81–91. Jope, E. M. 2000. Early Celtic Art in the British Isles. Oxford: The Clarendon Press. Joyce, J. 1936. Ulysses. London: John Lane, The Bodley Head. Leeds, E. T. 1933. Celtic ornament in the British Isles down to A.D. 700. Oxford: Clarendon Press. McCourt, J. 2000. The Years of Bloom: James Joyce in Trieste, 1904–1920. Wisconsin: Wisconsin University Press. Mahr, A. 1932. Das irische Kunstgewerbe. In H. T. Bossert (ed.) Geschichte des Kunstgewerbes aller Zeiten und Voelker Bd 5, 1–45. Berlin: Wasmuth. Márton, L. 1933. Die Frühlatènezeit in Ungarn. Archaeologia Hungarica 11. Budapest. Márton, L. 1934. Das Fundinventar der Frühlatène-Gräber, Dolgozatok 9–10, 93–182. Megaw, J. V. S. 1970a. Art of the European Iron Age: a study of the elusive image. Bath: Adams and Dart. Megaw, J. V. S. 1970b. The elusive image in La Tène art. In E. Anati et al. (eds.) Valcamonica Symposium. Actes du Symposium International d’Art Préhistorique Capo di Ponte, 507–14. Brescia: Edizioni del Centro. Megaw, J. V. S. 1983. From Transdanubia to Torrs: further notes on a Gabion of the late Jonathan Oldbuck. In D. V. Clarke and A. O’Connor (eds.) From the Stone Age to the Forty-five: studies in Scottish material culture presented to R. B .K. Stevenson, 127–48. Edinburgh: John Donald. Megaw, J. V. S. and Megaw, M. R. 2008. A Celtic mystery: some thoughts on the genesis of insular Celtic art. In D. Garrow, C. Gosden and J. D. Hill (eds.) Rethinking Celtic art, 40–58. Oxford: Oxbow Books.

222

Paul Jacobsthal with Katharina Ulmschneider and Sally Crawford

Raftery, J. 1938. Zur Zeitbestimmung der irischen Eisenzeit. Marburger Studien. Raftery, J. 1940. A suggested chronology for the Irish iron age. In J. R. Ryan (ed.) Féil-sgríbhinn Eóin Mhic Néil: essays and studies presented to Professor Eoin MacNeill on the occasion of his seventieth birthday, 272–281. Dublin: Three Candles. Salzman, L. F. (ed.) 1938. The Victoria History of the County of Cambridgeshire and the Isle of Ely Volume 1. Oxford: Oxford University Press. Salzman, L. F. (ed.) 1939. The Victoria History of the County of Oxford Volume 1. Oxford: Oxford University Press. Spratling, M. 1972. Southern British Decorated Bronzes of the

Late Pre-Roman Iron Age. Unpublished Ph.D. thesis, University of London. Szabó, M. 1977. The origins of the Hungarian sword style, Antiquity 51, 211–20. Ulmschneider, K. and Crawford, S. 2012. Writing and experiencing internment: rethinking Paul Jacobsthal’s internment report in the light of new discoveries. In H. Mytum and G. Carr (eds.) Prisoner of War: Archaeology, Memory, and Heritage of 19thand 20th- Century Mass Internment, 223–238. New York: Springer Publications. Vouga, P. 1923. La Tène, Monographie de la station publiée au nom de la Commission de Fouilles de la Tène. Leipzig: Hiersemann.

26 THE CELTIC MERCENARY RECONSIDERED Jan Bouzek

My old friend Vincent with his wife Ruth are leading experts in Celtic art (esp. Megaw and Megaw 2001), to whose understanding they always brought something new (Megaw and Megaw 1990; Megaw 2005) more essentially than the present author, so here a small note should be devoted rather to one of the fields of relations between the Celts and the Mediterranean.

The context of Celtic mercenary service The bravery of Celtic warriors was admired already by Aristotle. The Celts, instructed by their druids, did not place a great value on life. Death in battle was one way to ‘get better’ in the next life (cf. Caesar, de Bello Gallico VI, 14; for surveys of Celtic religion for example Aldhouse Green 1986; 1995; 1998; Hatt 1989), but they also did not despise fighting for money. Polybius (Histories II 7.4–11 and II 5.4) writes about their wildness and unreliability in the contract agreements which were the seamy side of their courage. Another remark mentions that those who lost in some quarrel at home and were expelled from society became mercenaries. Probably also outcasts who were condemned by druids and were not allowed to take part in religious life were among the mercenaries, as mentioned by Caesar (de Bello Gallico VI,13). But already the story of the invasion of Rome shows that the Celts were willing to be paid in gold, which they held in high esteem – the means of acquiring it, as part of a booty, or a fee for ‘protection’, or as a salary, was of lesser importance. The warriors expected part of the booty or a fee as ‘salary’ from their kings or leaders, too. All these were good conditions for mercenary service even in the narrower sense, i.e. serving as paid warriors on a contract between employer and employee (cf. Dobesch 1995; Kruta 2000, 727–728).

Italy, Sicily and the Carthaginians It was supposed that some Celts were financially supported by some Etruscans, who used them against Rome, but genuine Celtic mercenaries are first known working with the Carthaginians in the 5th century BC; the Elysikoi took part in the battle of Himera in 480 BC (Herodotus Histories VII, 165), and their ties with the Carthaginians are also confirmed by Iberian fibulae found in their territory (Kruta 2000, 727). Shortly after conquering Rome, the Italian Celts concluded a treaty with the Syracusian tyrant Dionysius I, to whom some of them (probably mainly the neighbours of his Adriatic domain, Senones living south of the Boii, but probably also some Boii) served as mercenaries. According to other sources, Dionysius settled some Senones in the Otranto region. Celtic mercenaries took part in the actions of Dionysius’ army in southern Italy, and already in the year 369 BC they were also sent by Dionysius to Greece where they fought at Corinth on the side of Sparta against Epaminondas of Thebes (Xenophon, Hellenica VII, 1, 1; 2; 14). Dionysios the Younger and later Agathocles employed them in their wars against Carthage. In 307 BC their troops took part in the campaign of Agathocles and Syracusians into Africa against Carthage, but other Celts served in the Carthaginian army. Polybius mentions them as mercenaries of the Carthaginians in battles at Sicily and also as allies of Illyrians during their seafaring attacks in Epirus and in the Adriatic Sea (Polybius, Histories II 7.4–11 and II 5.4). In both cases he writes about their unreliability in contract agreements. Three thousands Celts were employed by the Carthaginians in 363 BC and brought to Sicily, where they plundered Acragas. In the first Punic war, Carthage employed a large group of Celts, but some of them under their leader Antarios revolted when not paid enough, and went over to the Romans. Gesati, who came to Italy from

224

Jan Bouzek

Gaul in 225 BC, were probably organized as troops of would-be mercenaries, even if they came to help North Italian Celts against the Romans (Polybius, Histories II, 22; cf. Kruta 2000, 727; Szabó 1991, 333–334). Many Celts also joined the army of Hannibal, and were apparently paid – as other Carthaginian troops; they fought bravely against their common enemy even at Zama in 202 BC. After the defeats of North Italic Celts in the early 2nd century, Celtic mercenaries ceased to play any role in Italy, though some of them apparently joined the auxiliary troops of Roman army. Mercenaries contributed to the development of Celtic coinage starting towards the end of the 4th century; in North Italy it was based on models from Massilia. The Ambiens on the Somme took their models from the golden staters of Tarent, but also those of Philip II of Macedon, while the staters of Alexander were the main models in Central Europe (see below). Objects received by mercenaries may have existed among early Greek, Etruscan and Carthaginian imports north of the Alps, notably among the Greek coins (early cf. Mielczarek 1989, 38–40, 4th–3rd century items are also registered by Miltký 2010), the Celtic horse may have been inspired by Carthaginian coins and the Early La Tène artisans took over their techniques from the Mediterranean (Eluère 1988). The Balkan invasions Celts from Bohemia and Moravia, notably the VolkoiTektosagoi, apparently participated in the Balkan campaign, which was not organized on the tribal level as that in Italy, but mainly by groups of warriors of different origin; the Tektosagoi preserved their name as the only one of the tetrarchies in Galatia. The first Celts were already in the Lower Danube area at the time of Philip II. In 335 BC a Celtic delegation met Alexander II, and on that occasion they told him that the only thing they feared was that the sky might fall on their heads (Arrian, Anabasis of Alexander 1,4, 6–8; Strabo Geographica VII 3,8). The torcs found at Gorni Cibar in NW Bulgaria (e.g. Venedikov and Gerasimov 1973, pls. 254–255) may well have arrived in Thrace on this occasion. Another delegation of Celts met Alexander in Babylon (Diodorus Bibliotheca historica 17, 113, 2; Arrian, Anabasis of Alexander 7, 15,4), and some Celts probably joined his army. In 310 the Celts ‘from Illyria’ attacked their southern neighbours, the Autariati; 20,000 of them settled Kassandros on Mt. Orbelos (now Pirin and Ograzden). In 298 BC the Celtic attacks were stopped by Kassandros at Heamus. The death of Lysimachus in 281 offered the Celts new possibilities (cf. the survey by Tomaschitz 2000, 92–141; Expansion 1983): in 280 three Celtic armies marched to the south. The first of them, led by Kerethrios, attacked Thrace. The second, led by Brennos and Akichorios, attacked Pannonia, Dardania and Macedonia, while in 279 BC the third, led by Bolgios, defeated Ptolemaios Keraunos, who fell in the battle and the

Celts carried his head stuck on a spear in their army. They destroyed Emporion Pistiros (cf. Bouzek 2005b), Seuthopolis (Dimitrov and Čičikova 1976) and other important cities in Thrace. In the ruins of the final destruction of Pistiros damaged weapons, and also one late Duchcov type fibula appeared, known also from Serbia and Hungary (Bouzek 2005b; 2007). The fibula and part of a Celtic chain pendant were found on the burnt layer, so somebody lost them when looting. Other Celts returned from the campaign against Delphi and formed in what is now Serbia – with their Illyrian allies – the tribal federations of Greater and Lesser Scordisci (Domaradzki 1984; Popović 1992; Todorović 1968; 1974). The Celtic invasion is marked by several hoards in Thrace and Macedonia. The destruction of Emporion Pistiros is dated by a large hoard of 549 silver and three gold coins, which contained (besides several items of Alexander and of other diadochs) tetradrachms and drachms of Lysimachus, including those struck after his death; the owner of the set ceased collecting soon after Lysimachus’ death (Ruseva and Bouzek 2011).

The impact of the Balkan invasion in Central Europe After his victory over Ptolemy Keraunos, Bolgios with his army returned to the north. Several objects from Slovakia and Hungary may well have come there as part of the booty from the Balkan campaigns. This includes the silver kantharos found at Szob and the bronze lekythos found at Hurbanovo, together with some amphoroid glass beads found in Moravia (Bouzek 2002a and 2002b) (Figs. 26.3 and 26.4), and apparently many coins (Mielczarek 1989, 45–69; Militký 2010). This Balkan experience with its booty, and stories of enrichments by mercenaries in the south, led Celtic leaders to minting their own coins (cf. Szabó 1993). Celtic coinage in Central Europe started – with very rare predecessors – briefly after the Balkan campaign. Among the models for the eastern coinage the tetradrachms of Philip II (Fig. 26.2a and b), Alexander III and Antigonos Gonatas (277–239 BC) were of basic importance, besides those of Thasos, Byzantion and Larisa. In central Europe the main models for the first coins (c.300 BC) were those of Alexander the Great with Pallas Athena on the obverse (Fig. 26.1) and Nike on the reverse; Athena Alkis on the obverse was later changed into a warrior. The weight system derived from the Macedonian standard; but during the 3rd century the stater diminished from 8.5 g to 8 g. Also the changing symbolism of the ‘Tree of Life’, which first was mistletoe and in the Manching tree was already ivy, may be connected with the Thracian milieu. The Němčice site in Moravia yielded coins from Italy, Sicily, Gaul, many Balkan centres, the Aegean, of Ptolemies and Carthage dating from 3rd to early

26.  The Celtic mercenary reconsidered

225

Fig. 26.1. Stater of Alexander the Great: head of Athena, Pistiros hoard.

Figs. 26.3 and 26.4. Amphoroid glass beads from Pistiros and similar in collier from Brno – Horní Heršpice (photo courtesy J. Čižmářová).

Fig. 26.2 a and b. Tetradrachm of Philip II, from Pistiros.

2nd century BC (Kolníková 2012); at least a proportion of the Mediterranean coins was probably brought there by mercenaries (Fig. 26.5).

Celts in the SE Balkans and the foundation of Galatia: fighting for profit Between 279 and 277 BC the Celtic leader Kommontorios founded a Celtic kingdom in south-western Thrace. Its capital was Tylis, a town not yet known. The Celts could not conquer any of the coastal Greek towns, but they menaced their chora and Greek towns paid taxes for ‘protection’: Byzantium paid the Celts 80 talents per year. The coins of the last king of the Tylis kingdom, Kavaros, are distributed over most of Black Sea Thrace up to the present frontier between Romania and Bulgaria (Domaradzki 1995; Emilov 2007; Tomaschitz 2007). Kavaros’ kingdom was destroyed in 218 BC. The remains of the Celtic military groups then joined their relatives in Galatia. In the second quarter of the 3rd century, graves equipped with Celtic armour and local or Greek pottery are a characteristic phenomenon in Bulgaria. One was found in Plovdiv and contained – besides a Celtic sword and fibula – also Attic pottery of the early West Slope style (Bospačieva 1995; Bouzek 2005a, 105–106; 2007; Emilov 2005). The most famous Celtic site in Bulgaria is the burial at Mezek placed into an older Thracian barrow. The cart with bronze

linchpins in Celtic Plastic Style has its closest parallels in the vicinity of Paris (Lejars 2005), but it was probably made in Central Europe. It shows Celtic mobility in the same way as the story according to which a part of the Celtic booty from Delphi could already be seen a year after the Celtic campaign in the region of Toulouse. On the top of the Mezek tumulus, covering the Thracian-type tomb (Filow 1937), a statue of a wild boar from the early Hellenistic period was put. Wild boar was a symbol of a mighty warrior with the Celts and as such it was depicted on coins and on the top of helmets on the Gundestrup cauldron, while boar hunt was also popular in Thrace and Macedonia. The explanation suggesting that is was a burial of a Thracian local ruler and that the cart in the grave was part of booty gained from the victory over the Celts (Stoyanov 2005; 2010) or of Adaios (Emilov and Megaw 2012), a Ptolemaic (?) general only known from coinage, are hypotheses not more probable at the time of Celtic victories than the ascription of the burial to a partly ‘thracised’ Celtic leader, perhaps the founder of the Tylis kingdom (Bouzek 2005a, 106–107). More modestly equipped graves with Celtic weapons, but with locally produced pottery, are known from all parts of Thrace between Byzantium and the Danube (Domaradzki 1984; Haralambieva 2004; Anastassov 2011; Szabó 1992). North of the Danube they are even more common, and these graves also contained La Tène or related pottery (Todorović 1974; Berecki 2012). It seems that the Celts conquered only the lowlands along the rivers; in the mountains the Thracian population persisted (Rustoiu 2012). It is interesting that foot rings found inside a well at Isthmia in Greece according to their style and manner of wear were probably made in Bohemia (Fig. 26.6) (Szabó 1968; Bouzek 2002a). The Isthmia rings probably belonged

226

Jan Bouzek

Fig. 26.5. The origins of coins found at Němčice, Moravia; part of them was probably brought there by mercenaries (After the list of Kolníková 2012, drawing A. Waldhauserová, after Bouzek 2011). 1–3 local (type Rosendorf), 4 Macedonia, 5 Danube area Celtic, 6 Ptomaic Egypt. – 1 Němčice, 2 southern Slovakia, type with lyre, 3 Hungary, type Audoleon, 4 Transylvania, ‘bird symbol’, 5 Moldova, type Husi-Vavriesti, 6 Adaios, 7 Alexander III, 8 Illyrian king Balleus, 9 Peithesa in Etruria, 10 Rome, 11 Neapolis, 12 Arpi 13 Tarent, 14 Bruttium, 15 Syracusae, Hieron II, 16 Alexandria, Ptolemaios VI, 17 Cyrenaica, Polemaios III, 18 Zeigitania Punica, Carthage, 19 Gallia, type with cross, 20–21 potins with big head, and ‘a mannequin’, 22 Massalia.

to a lady who came to Greece with the Celtic campaign. Kruta suggested that it fits well into the time of the Celtic mercenaries’ revolt in Megara in 265 BC (Kruta 2000. 687). Celtic sacrifices to the deities of springs were usual; in Bohemia e.g. the Podmokly coin hoard or fibulae placed in the Obří pramen (Giant’s source) near Duchcov.

Anatolia In 278 BC one Celtic army lead by Leonnorios was in Thrace and another came there under Kerethrios. Nicomedes of Bithynia invited Celtic troops under the leadership of Leonnorios and Lutarios to help him in his row over the throne, which he shared with his two brothers, and with

26.  The Celtic mercenary reconsidered

227

Fig. 26.6. Foot-rings. Above Isthmia, below Prague – Bubeneč (after Bouzek 2011, drawing A. Waldhauserová).

Fig. 26.7. Middle La Tène fibula from Skopje, Mus. Skopje (courtesy D. Mitrevski).

Celtic help he won the rule over Bithynia. Leonnorios at that time crossed the Bosporus with his troops while Lutharios crossed the Dardanelles. An expedition of 20,000 people, 10,000 of whom were men in arms, settled in Phrygia and founded the Galatian kingdom. The remaining part of the Celtic army was then defeated by Antigonos Gonatas, grandson of Alexander’s brother, and the rest of the Celtic troops joined his army (cf. the survey by Tomaschitz 2002, 92–134). After the death of Lysimachos, parts of his empire in Asia Minor were annexed by Seleukos Nikator of Syria, against whom Nicomedes of Bithynia used his paid Celtic allies. Seleukos’ son Antiochus Soter defeated Ephesus against the Celts, but without success. The Celtic leader Brennos

Fig. 26.8. Decorative plaques in the shape of Celtic shields, Messembria (after Bouzek 2004).

(probably a common name with the Celts) attacked the town and fell in love with a local girl, Demodike, who – in return for the promise of a big reward – helped him to enter the town. For her service she got many jewels collected

228

Jan Bouzek

Fig. 26.9. Celtic warriors, terracotta. Left: Kerch, right: from Myrina in the Louvre (after Winter 1903).

Fig. 26.10. Celtic warriors from Kerch, terracotta, Skt. Petersburg, Hermitage (after Winter 1903).

26.  The Celtic mercenary reconsidered from women in Ephesus, but in the end when the Celts left she is said to have been buried alive with the confiscated jewels like Tarpeia in Rome. But the story might have been changed later and in the original version she might have saved the city from plunder. One of the earliest Celtic coins was found in Ephesus and some connect it with the Celtic attack and occupation. In 218 BC Attalos I invited Celtic Aigosagi into Thrace (they were probably other Celts than those in the kingdom of Tylis) to help him against Antiochus III. He settled them by the Hellespont, but in the next year they were defeated by Prusias of Bithynia and then their traces disappeared. Many Greek towns in Asia Minor in the 3rd century had to pay Galatians for ‘protection’. Heraclea once paid 5000 staters to their squad, and 200 staters to their leaders. One Galatian army gained the throne for Antiochos Hieraktos in 238 BC, but Galatian warriors rose against him; he had to call the troops of Ptolemy against them and pay the Galatians a vast amount of silver. Even during the wars between Galatia and Pergamon some Galatians served in the Pergamon army, as they still did in the army of Mithridates in the first century BC (Strobel 1996; Dobesch 2001, 79–102; cf. also Maier 1973).

Other and later Balkan Celts The Scordisci ruled the territory of present-day Serbia and Northwest Bulgaria (Knez 1996). The 3rd century burials of warriors known from many parts of Thrace were probably of Celts and Thracians as well (Guštin and Jevtić 2011). Weapons in these graves are of Celtic type, but pottery is of local origin. The ethnic identity of the buried warriors is of course uncertain; Thracians and Celts often served together as mercenaries in Hellenistic armies. Some Celts either entered the service of Thracian rulers or fought on their own in smaller groups. They became the leading force in Thrace for 150 years, and their artistic style dominated in most parts of what is now Bulgaria and Romania (Theodossiev 2005; Thraker und Kelten 2000). The decisive impact can be seen in armour, jewels and other branches of the fine arts. Celtic fibulae served for fastening garments, the custom of flat inhumation graves with weapons and jewels became common in burial rites; pottery also accepted inspiration from La Tène style, mainly in Transylvania and western Walachia. Most of the inhabitants of the eastern Balkans remained, however, Thracian; the Celts were in a minority. Although some of them brought their wives and their children learned Celtic as their mother tongue, in the foreign environment their small groups were assimilated during several generations and mixed with Thracian surroundings, and they lost their ethnic character. The last clearly Celtic horizon in Transylvania comes from the transition period between La Tène C1 and C2, i.e. from the beginning of the 2nd century BC.

229

The group of warrior graves called ‘Padea-Panagjurski Kolonii’ known from Romania and Bulgaria from the middle of the 2nd century to the early 1st century BC (Wozniak 1974; Rustoiu 2005) document the beginning of new political structures in the Central Balkans. The swords in those graves are of Celtic type, but horse bridles are Thracian and the pottery is of local tradition; the Thracians probably prevailed among its members. Graves of warriors show local differences in their burial rite (in some areas flat graves prevail, while in other places barrows predominate), and also in their equipment. They seem to represent warriors from different tribes who joined forces for common adventures according to the prevailing situation. Such groups sometimes served in Hellenistic armies and in the armies of Greek cities at the Black Sea, but they also attacked their southern neighbours. Written reports mention invasions of Skordisci and Triballi or joined with other Thracian allies in the second half of the 2nd and early 1st century BC.

The North Pontic region At the beginning, the culture of the Bastarni in what is now Moldavia and in the western Ukraine was of Celtic character; later they were considered a Germanic tribe. Due to this fact it is probable that Bastarni came from the northern periphery of the Celtic world, bordering with Germanic territory. They used cavalry more than their western neighbours and accepted many habits from their eastern neighbours, the Sarmatians. In 179 BC, Bastarni got as far as Rila in the southwest Balkans, but later they never achieved similar military success (cf. also Eremenko 1997 for the role of Celtic fibulae in the Zarubineckoe culture; Sulimirski 1976, Babeş 2006 and Shchukin 1995, for general surveys). In the Bosporan kingdom, Celtic influence is shown first by the Celtic shields on coins minted by Leucon II, dated to the second to third quarter of the 3rd century, together with a stylized Celtic sword, and those on the Isis ship depicted at Nymphaeum on a wall painting of the same date (survey Treister 1993, 789–91), so some mercenaries may have served there or were heard of from the history of their activities in Thrace and Bithynia. The clay figurines of Celtic warriors date mainly from the 2nd–1st centuries BC (Figs. 26.9 and 26.10), the Celtic helmets found in the North Pontic and North Caucasian regions may have come there via Asia Minor (Simonenko 1987; Trejster 1993, 791). Several late Duchcov (Early La Tène) fibulae roughly contemporary with those along the Balkan invasions are only known from the western Ukraine (Ambroz 1966, 12 with pl. 1), while the Middle La Tène items are much better represented, notably in the area of the Zarubinetskoe culture (Ambroz 1966, 12–14, pls. 1–3, distribution pl. 18) and in the Upper Don region, but some also came from the area

230

Jan Bouzek

of the Greek cities and from North Caucasus (Berlizov and Eremenko 1998). Celtic swords are rare, but also known notably from Scythian Neapolis; perhaps they arrived there as booty (Rieth 1965; Trejster 1993, 793 fig. 4). La Tène fibulae are known from the northern Black Sea area still in Early Roman contexts.

with the Celts. Similar fibulae are kept and exhibited in several North Greek museums, notably at Dion, Maroneia, Komotini, Thessaloniki, Ioanina and Pella, in Skopje in Macedonia (Fig. 26.7), and also in Istanbul and Ҫanakkale museums in Turkish Thrace; at least some of them came from Hellenistic fortresses in the region.

Celtic mercenaries in the narrower sense in the Eastern Mediterranean

Celtic enemies and mercenaries and their weapons in the artistic imagination

It has been shown that the frontier between fighting for booty, and ransom and for salary is vague in many cases, but here we should consider the latter in its narrower sense (cf. Kruta 2000, 727–728; Szabó1991). Many Celts served as mercenaries in Hellenistic armies in the east. We mentioned above Nicomedes of Bithynia and Antigonos Gonatas as employers of Celtic soldiers; other Celts also fought in the armies of the first diadochs. When Pyrrhos of Epirus defeated Antigonus Gonatas he valued most the victory over his Celtic guard which was bravest in the battle; their shields he dedicated to the sanctuary of Athena Itonia (Pausanias Descriptio Graecia 13, 3). Ptolemaios II Philadelphos engaged 4,000 Celts to use them for the repression of an uprising that broke out against him in Egypt, but when they revolted he let them starve to death without food on one of the islands on the Nile. Their troops also revolted in other places, e.g. in Megara in 265 BC. Most of the stories of Celtic mercenaries date from the 3rd and early 2nd century BC, but in Asia Minor some Galatians and their relatives were still serving in the army of their enemy Pergamon, with the Seleucids and with Mithridates of Pontus (Strobel 1996). In Egypt a Celtic shield was made with the relief of a rare Upper Nile gazelle which testifies to the admiration of Celtic weapons even in the empire of the Ptolemies (Künzl 2004). Archaeological evidence for the presence of Celts in the eastern Mediterranean mainly comes from fibulae of Middle La Tène construction known from Turkey (Müller-Karpe 1988; 2006; Polenz 1978; Schaaf 1970), Syria and Lebanon (Courbin 1999); some of them might have belonged to Celtic mercenaries and their families who often accompanied them. An interesting feature in Asia Minor is the concentration of Middle La Tène fibulae with spiral decoration south of the Halys bend – fibulae with extremely high external cords appear in Centre-North Anatolia. This variety is also known from the Crimea, and was probably connected with the mercenaries serving in the army of Mithridates VI (Müller-Karpe 2006). Middle La Tène fibulae are known as gifts to sanctuaries in Dodona and on Delos (Szabó 1971). They might have been dedicated there by those Celts who valued the sanctuaries or on the contrary by their opponents who shared with the gods a part of their booty from battles

Pergamene art produced the most famous sculptures of Celtic warriors (Wenning 1978; Polito 1999) and their weapons, their weapons were dedicated to Delphi (Amandry 1978) and elsewhere. Hellenistic terracotta figurines known from western Asia Minor and the northern Black Sea area depicted Celtic warriors with their high shields (Winter 1903, 384; Dufková 2001), both naked and dressed and in slightly different attitudes (Figs. 26.9 and 26.10). Their shields were represented and imitated even in jewellery (Fig. 26.8). A well-known head of a Celt was found at Gizeh (Fig. 26.11) in Egypt and a Celtic shield decorated with a

Fig. 26.11. Head of a Celtic warrior from Gizeh in Kairo, after a cast.

26.  The Celtic mercenary reconsidered rare Central Africa gazelle in Stuttgart was made in Egypt, too (Künzl 2004).

The end of the Celts in the Balkans and of Celtic mercenaries: the phalerae and the Gundestrup cauldron The art of phalerae, bronze or silver discs with representations of a mounted warrior or a bust of a goddess, opens a new stylistic language characteristic at first of the eastern Balkans and the north-western part of the Black Sea, but it was also popular with the Sarmatians and Balkan Celts. In the East they were found even in Northwest India, in the West in the first Roman forts in the Rhineland (Haltern, Waldgirmes) and on the island of Sark off the coast of Normandy. One phalera with the name Mithridates is kept in the Bibliothèque Nationale in Paris; it may represent a dedication to the great king and passionate adversary of the Roman empire, in whose army many Celts served (Bouzek 2010; Mordvinceva 2001). Phalerae had their parallels also in Italy, notably in discs from Manerbio sul Mella. Celtic Skordisci living mainly in the area of modern Serbia and Vojvodina supported Mithridates even in the years 78–76 BC. They remained in this area till the Roman occupation and are still mentioned in connection with the Pannonian uprising at the end of the 1st century BC. They were neighbours of the Thracians, they had a lot in common with them and it is in their region that the workshop tradition from which the most famous object of late La Tène art – the cauldron found in Gundestrup in Denmark is often sought (Bergquist and Taylor 1987; Best, Flemming and Marazov 1991; Falkenstein 2004; Hachman 1990). Its bottom part is made from a phalera. From the 1st century BC on, some Celts from Italy, Gaul and from the Balkans served in the disciplined Roman army, but their position in it was very different from the earlier free-lanced mercenary.

Conclusion Celtic mercenary service was closely related to other military activities of the Celts; money gained by ransom and as booty was of equal value, as was that gained through mercenary service; Celtic coinage derived from Greek models gained by mercenaries and it was also used as payments by military leaders to their men.

231

Bibliography Aldhouse-Green, M. 1986. The Gods of the Celts. Gloucester: Alan Sutton. Aldhouse-Green, M. 1995. Celtic Goddesses, Warriors, Virgins and Mothers. London: British Museum. Aldhouse-Green, M. 1997. The World of The Druids. London: Thames and Hudson. Amandry, P. 1978. Consecration d’armes galates à Delphi, Bulletin de correspondance hellénique 102, 571–86 Ambroz, A. K. 1966. Fibuly juga jevropejskej časti SSSR. Moskva: Archeologia SSSR – SAI D1–30. Anastassov, J. 2011. The Celtic presence in Thrace during the 3rd century BC in the light of new archaeological data. In M. Guštin and M. Jevtić (eds.) The Eastern Celts. The communities between the Alps and the Black Sea, 227–39. Koper. Babeş, M. 2006. Les Celtes et la région de la Mer Noire. In M. Szabó (ed.) Celtes et Gaulois, Archéologie face à l’histoire. 3: les Civilisés et les Barbares du Ve au Ie siècle av. J.-C., Actes de la table ronde de Budapest, 17–18 juin 2005, 126–27. Gluxen-Glenne: Bibracte 12/3. Berecki, S. (ed.) 2012. Iron Age Rites and Rituals in the Carpathian Basin, Proceedings of the international colloquium 2011. Târgu Mureş: Ed. Mega. Bergquist, A. and Taylor, T. 1987. The origin of the Gundestrup cauldron, Antiquity 61, 10–24. Berlizov, N. E. and Eremenko, V. E. 1998. Latenskie importy v sarmatskich pamjatnikach Pričernomor’ja, Drevnosti Kubani 7, 25–33. Best, J., Flemming, K. and Marazov, I. 1991. Thracian Tales on the Gundestrup Cauldron. Amsterdam: Najade Pr. Bospačieva, M. 1995. Mogilno pogrebenie ot elenistechkata nekropol na Filippopol, Izvestia na muzeite ot Juzhna Bulgaria 21, 43–62. Bouzek, J. 2002a. Die keltische Kunst und der Osten. In A. Lang and V. Salač (eds.) Fernkontakte in der Eisenzeit. Praha: Archeologický ústav AV ČR, 98–105. Bouzek, J. 2002b. Die Bronzelekythos vom Typ Talcott aus Hurbanovo und andere vorrömische mediterrane Bronzegefässe aus der Slowakei. Anodos 2, In Honour of Mária Novotná, 53–57. Trnava: Trnavská Univerzita v Trnave. Bouzek, J. 2005a. Thracians and Their Neighbours, Studia Hercynia IX. Prague: Nakl. AMU. Bouzek, J. 2005b. The Duchcov fibula in Bulgaria and the destructions of Pistiros in 279/8 BC. In H. Dobrzańska, V. Megaw and P. Poleska (eds.) Celts on the Margin, Studies of European Cultural Interaction, 7th century BC – 1st century AD dedicated to Zenon Woźniak, 93–101. Kraków: Polish Academy of Sciences. Bouzek, J. 2007. La destruction de Pistiros, les Thraces et les Celtes en Bulgarie, Studia Hercynia 11, 76– 82. Prague: Nakl. AMU. Bouzek, J. 2010. The art of phalerae in broader context. In I. Cândea (ed.) Tracii şi vecinii lor în Antichitate (The Thracians and Their Neighbours in Antiquity), Studia in honorem Valerii Sîrbu, 95–110. Braila: Ed. Istros. Bouzek, J. 2011. Prehistory of Europe as seen from Its Centre, Studia Hercynia 15/1. Prague: Nakl. AMU.

232

Jan Bouzek

Courbin, P. 1999. Une fibule de La Tène à Bassit (Syrie), Préhistoire européene 3 (Mélanges a la mémoire de René Joffroy, 93–7. Dimitrov, D. P. and Čičikova, M. 1978. The Thracian City of Seuthopolis, British Archaeological Reports Supplementary Series 38. Oxford: British Archaeological Reports. Dobesch, G. 1995. Das Europäische Barbaricum und die Zone der Mediterrankultur: Ihre historische Wechselwirkung und das Geschichtsbild des Poseidonios, Tyche Suppl. vol. 2. Wien: Holzhausen. Dobesch, G. 2001. Ausgewählte Schriften. Vienna: Böhlau Verlag. Dobrzańska, H., Megaw, V. and Poleska, P. (eds.) 2005. Celts on the Margin, Studies of European Cultural Interaction, 7th century BC – 1st century AD dedicated to Zenon Woźniak. Kraków: Polish Academy of Sciences. Domaradzki, M. 1984. Keltite na balkanskija poluostrov. Sofia: Nauka i izkustvo. Domaradzki, M. 1995. La diffusion des monnaies de Cavaros au Nord-est de la Thrace, Eirene 31, 120–8. Dufková, M. 2001. Zur Problematik der sog. Galater-Krieger in griechischer Koroplastik, Anodos I, 81–4. Trnava: Univ. Eluère, Ch. 1988. Orfèvrerie des Celtes anciens et orfèvreries méditerranéennes. In Princes Celtes et la Méditerranée, Actes du colloque Paris, 199–220. Paris: Louvre. Emilov, J. 2005. Changing paradigms: Modern interpretation of Celtic raids in Thrace reconsidered. In H. Dobrzańska, V. Megaw and P. Poleska (eds.) Celts on the Margin, Studies of European Cultural Interaction, 7th century BC – 1st century AD dedicated to Zenon Woźniak, 103–108. Kraków: Polish Academy of Sciences. Emilov, J. 2007. La Tène finds and the indigenous communities in Thrace. Interrelations during the Hellenistic period, Studia Hercynia 11, 57–75. Prague: Nakl. AMU. Emilov, J. and Megaw, V. 2012. Celts in Thrace? A Re-Examination of the Tomb of Mal Tepe, Mezek with Particular Reference to the La Tène Chariot Fittings, Archaeologia Bulgarica 16.1, 1–32. Eremenko V. E. 1997. „Keltskaja rol“ i zarubineckaja kultura. Sankt-Petersburg. Expansion 1983. L’expasion des Celtes vers l’Orient, Dijon, Dossiers d’archéologie et histoire 77. Falkenstein, F. 2004. Anmerkungen zur Herkunftsfrage des Gundestrupkessels, Prähistorische Zeitschrift 79, 57–88. Filow, B. 1937. Kupolnite grobnici pri Mezek, IBAI Izvestia na bulgarski archeologicheski institut 11, 1–116. Guštin, M. and Jevtić, M. (eds.) 2011. The Eastern Celts. The communities between the Alps and the Black Sea. Koper. Hachmann, R. 1990. Gundestrup-Studien, Untersuchungen zu der keltischen Grundlage der frühgermanischen Kunst, Bericht der Römisch Germanischen Kommission 71, 565–903. Haralambieva, A. 2004. Eine Gussform aus Bronze aus der Mittellatènezeit, Izvestia Varna 36–7, 2000–1, 187–210. Hatt, J. J.1989. Mythes et dieux de la Gaule I, Les grandes divinités masculines. Paris: Picard. Kolníková E. 2012. Němčice. Ein Macht-, Industrie- und Handelszentrum in Mähren und Siedlungen an ihrem Rand. Fundkatalog: Münzen. Brno: Institut der Akademie Wissenschaften der Tschechischen Republik. Knez, T. 1996. Keltische Gräber von Novo Mesto. In E. Jerem, O. Urban and A. KrennLeeb (eds) Die Kelten in den Alpen und

an der Donau, Studien Zur Eisenzeit Im Ostalpenraum: Akten Des Internationalen Symposiums St. Polten, 14.–18., 441–446. Budapast: Archaeolingua Alapitvany. Kruta, V. 2000. Les Celtes, Histoire et dictionnaire. Paris: Robert Laffont. Künzl, E. 2004. Waffenfund und Zoologie in Alexandrien: Der hellenistische Prunkschild im Württembergischen Landesmuseum Stuttgart, Jahrbuch des Römisch-Germanischen Zentralmuseums Mainz 50, 2003 (2004), 279–303. Lejars, T. 2005. Le cimetière celtique de La Fosse Cotheret, à Roissy (Val- d’Oise) et les usage funéraires aristocratiques dans le nord du Bassin parisien à l’aube du IIIème siècle avant J.-C. In O. Buchsenschutz, A. Bulard and T. Lejars (eds.) L’âge du Fer en Ile-de-France 73–83. Tours-Paris: FERACF. Maier, F. 1973. Keltische Altertümer in Griechenland, Germania 51, 459–76. Megaw, J. V. S. 2005. Celts in Thrace? A reappraisal. In J. Bouzek and L. Domaradzka (eds.) The culture of Thracians and their neighbours, British Archaeological Reports Int. Ser. 1350, 209–14. Oxford: BAR. Megaw, J. V. S. und Megaw, M. R. 1990. Semper aliquid novum, Acta Arch. Hung. 42, 55–72. Megaw, J. V. S. und Megaw, M. R. 2001. Celtic art from its beginnings to the Book of Kells. 2 ed. London: Thames and Hudson. Mielczarek, M. 1989. Ancient Greek Coins Found in Central, Eastern and Northern Europe. Wroclaw: Ossolineum. Militký, J. 2010. Finds of Greek, Roman and Early Byzantine coins in the Territory of the Czech Republic. Vol. 1, 1–3. Bohemia, Wetteren, Warszaw. Mordvinceva, V. 2001. Sarmatische Phaleren, Archäologie in Eurasien 11. Rahden/Westf. Müller-Karpe, A. 1988. Neue galatische Funde aus Anatolien, Istanbuler Mitteilungen 38, 188–99. Müller-Karpe, A. 2006. Zur historischen Deutung von Funden keltischer Trachtelemente in Anatolien, in Szabó 2006, 3, 119–23. Polenz, H. 1978. Gedanken zu einer Fibel vom Mittellatèneschema aus Kayseri in Anatolien, Bonner Jahrbücher 178, 181–216. Polito E. 1999. I Galati vinti. Il trionfo sui barbari da Pergamo. Roma. Popović, D. 1992. Celtic Cemeteries at the Iron Gates Area – Scordisci and the Native Population in the Middle Danube Region. Belgrade. Rieth, A. 1965. Bemerkungen zu einem keltischen Langschwert von Neapolis auf der Krim, Germania 43, 159–63. Ruseva, B. and Bouzek, J. 2011. The journey of the coins from Vetren (Pistiros) hoard: A preliminary remark, Studia Hercynia XV/2, 9–13. Rustoiu, A. 2005. The Padea – Panagjurski Kolonii group in SW Transylvania (Romania). In Dobrzańska et al., 109–20. Rustoiu, A. 2012. The Celts and indigenous population from the southern Carpathian basin, Intercommunity communication strategies. In Berecki 2012, 357–90. Schaaf, U. 1970. Eine Mittellatènefibel aus Kleinasien, Jb. RGZM 17, 297–300. Shchukin, M. B. 1995. The Celts in Eastern Europe, Oxford Journal of Archaeology 14, 201–27. Simonenko, A. V. 1987. Kel’to-italiyskie shlemy na teritorii

26.  The Celtic mercenary reconsidered Vostochnoj Jevropy. In Pamjatniki bronzogo i ranego železnogo veka Podneprovja, 104–113. Dnetropetrovsk. Stoyanov, T. 2005. The Mal-Tepe complex at Mezek. In J. Bouzek and L. Domaradzka (eds.) The culture of Thracians and their neighbours, British Archaeological Reports Int. Ser. 1350, 123–8. Oxford: BAR. Stoyanov, T. 2010. The Mal-Tepe tomb at Mezek and the problem of the Celtic kingdom in south-eastern Thrace. In Vagalinski, L. F. (ed.) In search of Celtic Tylis in Thrace (IIIC BC), 115–19. Sofia. Strobel, K. 1996. Die Galater, Geschichte und Eigenart der keltischen Staatsbildung auf dem Boden des hellenistischen Kleinasien I. Berlin. Sulimirski, T. 1976. The Celts in Eastern Europe. In Megaw, J. V. S. (ed.) To Illustrate the Monuments, Essays presented to S. Piggott, 181–190. London: Thames and Hudson. Szabó, M. 1968. Zur Frage des keltischen Fundes von Isthmia, Acta Antiqua Acad. Scien. Hungricae 16, 173–7. Szabó, M. 1971. Une fibule celtique à Délos, Bull. Corr. Hellénique 95, 503–14. Szabó, M. 1991. Le mercenariat. In S. Moscati et al. (eds.) I Celti, 333–6. Milano. Szabó, M. 1992. Les Celtes de l’Est, Le seconde âge du fer dans la cuvette des Karpates. Paris: Centre archéologique européen du Mont Beuvray; ERRANCE edition. Szabó, M. 1993. Guerriers celtiques avant et après Delphes: contribution à une période critique de monde celtique. In

233

L’Europe Celtique du Ve au IIIe siècle av. J.-C., Contacts, échanges et mouvements de populations, Actès du 3ème symposium international d’Hautvillers, Bibracte 1992, 49–67. Szabó, M. (ed.) 2006. Celtes et Gaulois, Archéologie face à l’histoire. 3: les Civilisés et les Barbares du Ve au Ie siècle av. J.-C., Actes de la table ronde de Budapest, 17–18 juin 2005, Glux-en-Glenne, Bibracte 12/3. Theodossiev, N. 2005. Celtic settlement in NW Thrace during the late 4th and 3rd centuries B.C.: some historical and archaeological notes. In Dobrzańska et al. (eds.) 2005, 85–92 Thraker und Kelten 2000. Thraker und Kelten beiderseits der Karpaten, Hochdorf, Exh. cat. Todorović, J. 1968. Kelti u jugoistočnoj Evropi. Beograd. Todorović, J. 1974. Scordisci, Istoria i kultura, Novi Sad. Beograd. Tomaschitz, K. 2000. Die Wanderungen der Kelten in der antiken literarischen Überlieferung, 92–141. Wien. Tomaschitz, K. 2007. Die Kelten von Tylis nach den Schriftquellen, Studia Hercynia 11, 83–95. Treister, M. J. 1993. The Celts in the North Pontic area: A Reassessment, Antiquity 67, 789–804. Venedikov, I. and Gerasimov. T. 1973. Trakijsko izkustvo. Sofia: Izdatelstvo Bulgarski Hudoznik. Wenning R. 1978. Die Galateranatheme Attalos I, Pergamenische Forschungen 4, Berlin. Winter, F. 1903. Typen der figürlichen Terrakotten. Leipzig. Woźniak, Z. 1974. Wschodnie pogranicze kultury lateńskiej. Wroclaw – Warszawa.

27 THE DRAGON FROM OBERLEISERBERG Maciej Karwowski

Introduction Among archaeological finds from the Celtic hilltop settlement on Oberleiserberg held by the Museum of Prehistory of Lower Austria in Asparn an der Zaya is a small fragment of an object, presumably made of steel, with a dragon’s head terminal (inventory No. 22694.223; Karwowski 2009, 119, fig. 9.1). It was discovered on the surface of the site by Leopold Laab, private collector. The fragment has a length of 30 mm and the form of an S-shaped bar of oval section, 4 x 5 mm. One of its ends is evidently fractured, the other end is flattened, with a rather heavily stylized but nevertheless clearly legible dragon’s head. The beast has a wide open maw, a clearly marked eye and, on the top of its head and spine, a crest in the form of five bulges. The head passes fluidly to a fairly slender neck which has, on its inner and outer face, two more elongated bulges (Fig. 27.1). Oberleiserberg (‘Oberleis Hill’) near Ernstbrunn, where the dragon’s head fragment was recovered, is one of the key archaeological sites in the Middle Danube region. It occupies a high elevation (457 m above sea level) which is part of the Leiserberge chain at the very centre of the Weinviertel in Lower Austria. The summit of the elevation is an oval plateau with a surface area of about 6.5 ha (about 360 × 250 m), which is bordered to the west, north and east by steep slopes. On the south side, slightly below the plateau, is another flat ‘terrace’ (called Vorburg) with a surface area of about 1.5 ha. This is therefore a typical hilltop site with a very favourable position both topographically and in terms of communications. Over many years of archaeological research, rich traces of occupation spanning the late Neolithic period to modern times have been found.

An important part of the archaeological material coming from Oberleiserberg are finds dated to the late Iron Age and connected with Celtic La Tène Culture settlements (Karwowski 2009; 2012). Some of this material comes from regular archaeological research and some was collected from the surface of the site.

The dragon motif in Celtic art The dragon motif is known in Celtic art from stylistically quite diverse variants from the very beginning. Presumably it derives from the mythology and folklore of the ancient peoples of the Middle East and Greece where the dragon is similar to a giant serpent. The Greeks, and later the Romans, regarded the serpent as a guardian spirit and often represented it on their altars. Disgust and aversion for snakes came only later, presumably with the spread of Christianity as in its tradition the creature plays a definitely negative role as Satan, the Serpent in the garden of Eden (Kopaliński 1991, 1257–8; Simek 2004, 144). A motif particularly frequent in the Hellenistic tradition is that of ketos, a mythical serpentine sea monster, presumably born of the imagination of sailors, said to inhabit the deepest parts of the sea. Nevertheless, in Greek mythology we find a good many other beasts, mostly hybrids of two or more creatures (Kopaliński 1991, 917, 1080–1; Simmons 2006, 609; Szabó 2012, 440). Under the influence of the Greeks the symbolism of the serpentine monster was adopted in Etruscan art (e.g. Martelli 2006, 188–9, fig. on p. 187) from whence, presumably still during the Early La Tène period, it was introduced to the Celts. And so, in Early La Tène art

27.  The Dragon from Oberleiserberg

235

Fig. 27.2. An ornament in the form of a pair of dragons from an iron sword scabbard, Celtic grave no. 400 at Pottenbrunn, Lower Austria (photo: A. Rausch – BDA Vienna, permission of Dr P. Ramsl).

Fig. 27.1. The dragon’s head find from the Celtic hilltop settlement on Oberleiserberg, Lower Austria (photo: G. Gattinger – IUFG Vienna, drawing: E. Smagur).

the dragon usually is depicted as a creature with the body of a serpent arranged in the shape of the letter ‘S’ but with a head usually in the form of the head of a bird of prey or that of a sea horse (Szabó 2012, 439–40). In the Early La Tène period the S-shaped dragon usually is an element of a design consisting of a pair of serpentdragons arranged in a lyre-shaped motif. This motif, usually in the form of a considerably stylized variant, appears on iron and bronze belt buckles, both in north Italy and in the Alpine region (Lenerz-de Wilde 1980, 78–83; Bulard 1982, 151; Bill 2000; Megaw and Megaw 2001, 66–7). Soon it was adopted as a highly distinctive decorative element of iron sword scabbards widespread in much of the territory of Celtic settlement (Bulard 1982, 151–3; Petres 1982; Megaw and Megaw 1989; 2001, 126–8; Szabó and Petres 1992, 29–36; Ginoux 2007). It is notable at the same time that the paired dragons motif only rarely appears on other La Tène artefacts (e.g. de Navarro 1959, 132, 138–9, Pl. 20:1 and 21: 2–4; see also Olmsted 2001, 114, Pl. 141). But is known in the Scythian and the Middle Eastern environment, to which area we may have to go to seek its prototypes (de

Navarro 1959, 135, Pl. 20:4–5). One of the more popular variants of decoration on Celtic sword scabbards are two dragons with wide open maws depicted facing each other (Fig. 27.2) (Ginoux 2007). This motif is often encountered in the Eastern Celtic environment on scabbards dated to the middle La Tène period (Petres 1982, 162, 166–7; Ginoux 2007, 102–18, figs 60 and 61). The depictions of the ketos, serpentine sea monster inspired by Hellenistic art, appear as imitations only on a limited number of ornamental designs seen on Eastern Celtic objects. One fine example is a bronze decorative drinking horn mount discovered in a Middle La Tène grave in the burial complex at Jászberény on the Middle Tisa in Hungary (Kaposvári 1969, 180, figs 5 and 11). This piece has a terminal in the form of a monster, with a wide open maw and a crest running from its head down its spine. The lower part of the beast’s body is covered with scales (Fig. 27.3). A similar style of depiction of a mythical sea monster was identified on some of the Boii coin finds from the Bohemian Basin of the Oppida period. These are the relatively rare gold staters classified as a subsidiary series, described in numismatic literature as the ‘Rolltier’ type (Castelin 1965, 50–2, 167–88). The reverse of these coins depicts a coiled serpentine sea monster with a wide open maw and a series of bulges in imitation of a crest. The similarity of the style of the depiction seen on a coin find from Osov, central Bohemia

236

Maciej Karwowski

Fig. 27.4. Gold Boii coin type ‘Rolltier’, from a hoard of Celtic coins from Osov, Bohemia (photo: J. Vaněk – NM Prague, permission of Dr J. Militký).

Fig. 27.3. Partial reconstruction of a bronze drinking horn mount, Celtic grave no. 17 at Jászberény, Hungary (after: Kaposvári 1969).

(Fig. 27.4) (Militký 2008, 124, fig. 68:22, Pl. 15:3), to the dragon from Oberleiserberg is so evident that it does not seem to be fortuitous. The ‘Rolltier’ motif also appears on the obverse of a series of gold coins of the Vindelici known from south Bavaria classified by H.-J. Kellner as type I (Kellner 1990, 8; see also Castelin 1965, 179–80, fig. 20). However, the stylistic similarity applies only to the coiled body of the monster depicted on them, but not to its head, because the Vindelician ‘Rolltier’ has a closed maw and appears less aggressive (see Kellner 1990, Pl. 44–5, 49 and 54). The image of, one might say, a ‘dormant’ serpentine monster with a closed maw is seen also on the reverse of silver hexadrachms, type ‘Biatec’, with the inscription MACCIVS, minted around the middle of the 1st century BC, probably in the Boii oppidum at Bratislava in south-western Slovakia (Castelin 1965, 168; Kolníková 1978, 72, 82, figs 21–3). Similarly, as in the case of the paired dragons, the image of the coiled monster serpent depicted on the Boii coins apparently drew inspiration from oriental symbolism. The motif is quite widespread in Scythian and Thracian art (Castelin 1965, 168–71; Birkhan 1999, 35–6, 88, 275, figs 442–6). Very similar motifs are encountered also in

Hellenistic art, for example, on jewellery (Higgins 1961, 161–3, Pl. 47).

The Oberleiserberg dragon As stated in the opening paragraph, the artefact from Oberleiserberg with the depiction of the dragon was probably made of steel. The view that the find from Oberleiserberg was manufactured from high quality steel was expressed during conservation treatment made in the Jagiellonian University Archaeometallurgy and Conservation Unit of Institute of Archaeology by Dr M. Biborski. He stressed at the same time that a fully conclusive identification requires further specialist metallographic analyses. This special alloy of iron and carbon was a resource well known in antiquity. In the world of the ancient Celts greatest renown was won by Noric steel – ferrum noricum – which is mentioned in the written sources (Straube 1996, 115–21). The name described high quality steel produced in their kingdom by the Celtic Norici using eastern Alpine (mostly Carinthian) ores. When the Roman province of Noricum was established, both the production and distribution of Noric steel came under Roman control and the material itself was commonly used in the production of weapons (Sperl 2002). It is commonly accepted that the main centre of metallurgy and distribution of Noric steel during the 1st century BC was on Magdalensberg in Carinthia (Dolenz

27.  The Dragon from Oberleiserberg

237

Conclusion

Fig. 27.5. The find from Oberleiserberg reconstructed tentatively as a fragment of an armour fastening (drawing: E. Smagur).

1996; Straube 1996, 26–32; Sperl 2002, 64; Buchwald 2005, 124). Moreover there is evidence that the iron ore was obtained and steel of comparable quality produced in other centres in the kingdom of the Norici, e.g. in southern Burgenland (Kaus 1981, 76–86). Based on metallographic analyses V. F. Buchwald (2005, 118, 124–5, Table 5.1) has concluded that weapons made of steel corresponding in quality and technological parameters to Noric steel may have been manufactured already at the beginning of the Middle La Tène period. It is less easy to attribute the dragon head fragment from Oberleiserberg to a specific object. Its shape and material suggest that originally it belonged to some sort of a fastening (Fig. 27.5). Objects of similar form, with animal (often serpent) head terminals, are noted among items of Roman military gear, i.e. they are elements of fastenings of plate and mail armour. Similar fastenings, dated to the Augustan period or, more broadly, within the 1st century AD, are known mostly from Roman provinces on the Rhine and from Britannia (Deschler-Erb 1999, 37, Pl. 15:269–70; Hansen 2003, 73–5, figs 23–4, Map 6, List 9). Even though finds of similar fastenings from reliably dated pre-Roman contexts are rare (see Hansen 2003, 34–42, 209–10; Viand 2008, 40–1) we cannot rule out their concept of the dragon motif derives from the Celtic environment. This is indicated both by the early chronology of many of the provincial Roman finds and by their geographic distribution. A strong additional argument would be pieces of armour with decorative fastenings immortalized on the stone busts of Celtic warriors discovered in the oppidum at Entremont, Bouches-du-Rhône in southern France (Arcelin and Rapin 2002, 46–4, figs 7 and 13; Hansen 2003, 91–2, fig. 33; Viand 2008, 41, fig. 9).

The brief analysis presented here leads us to conclude that the representation of the dragon discovered on Oberleiserberg corresponds stylistically to the works of Celtic art which drew their prototypes from Greek or Scythian models. Even though similar motifs are known from much of the territory of the Celtic world, they appear to be more characteristic of its eastern region. One clue to assist interpretation may be the fact that in Celtic ornamentation the dragon tends to be encountered mostly on sword scabbards, i.e. in the context of military equipment. This may relate to the belief in the serpent, or the serpent-dragon, as a guardian spirit, adopted from the Mediterranean world. Drawing on analogies known from early Roman military gear we can venture to interpret the find from Oberleiserberg as an ornamental fragment of a fastening of a Celtic plate or mail armour. An additional argument is the material out of which this artefact is made. If we assume that this may be Noric steel, the material commonly used in manufacturing weapons, this would make the find from Oberleiserberg an import from the nearby kingdom of the Celtic Norici. Contacts between the settlement on Oberleiserberg and the Norici are documented so far only by a handful of finds. This is in evident contrast with the very intensive contacts with the area in the neighbourhood of the state of the Norici inhabited by the Celtic Taurisci which are well attested to by archaeological and numismatic materials (Karwowski and Militký 2011). Such a state of affairs seems to reflect the political situation then prevailing in the Middle Danube region including the ‘collaboration’ of the Norici with Rome. Parallels between the design of the Oberleiserberg dragon and representations seen on Boii ‘Rolltier’ coins suggests that our specimen was manufactured by the Norici on commission from the Boii. The chronology of the object decorated with the dragon’s head from Oberleiserberg also remains an open issue. There is probably no question that it dates from the time when the site was under Celtic occupation. Archaeological material recovered from Oberleiserberg proves that the settlement was in use from the second half of the 3rd to the beginning of the second half of the 1st century BC, after which came its quite abrupt decline. Thus, very likely, we have here an object dating from the Oppida period crafted in an environment where the Celtic tradition mingled with early Roman influences, somewhere in the eastern Alpine or the Middle Danube region.

Bibliography Arcelin, P. and Rapin, A. 2002. Images de l’aristocratie du second âge du Fer en Gaule méditerranéenne. À propos de la statuaire d’Entremont. In V. Guichard and F. Perrin (eds.) L’Aristocratie celte à la fin de l’âge du Fer (IIe s. avant J.-C. – Ier s. après

238

Maciej Karwowski

J.-C.), Collection Bibracte 5, 29–66. Glux-en-Glenne: Centre Archéologique Européen du Mont Beuvray. Bill, J. 2000. Der Drache am eisernen Gurt. In R. C. de Marinis and S. Biaggio Simona (eds.) I Leponti tra mito e realtà. Raccolta di saggi in occasione della mostra 2, 31–40. Locarno: Dadò. Birkhan, H. 1999. Kelten. Bilder ihrer Kultur. Wien: Österreichische Akademie der Wissenschaften. Buchwald, V. F. 2005. Iron and steel in ancient times, Historisk– filosofiske Skrifter 29, Copenhagen: Kongelige Danske Videnskabernes Selskab. Bulard, A. 1982. A propos des origines de la paire d’animaux fantastiques sur les fourreaux d’épée laténiens. In P.-M. Duval and V. Kruta (eds.) L’Art Celtique de la Période d’Expansion: IVe et IIIe siècle avant notre ère, Hautes Etudes du monde gréco-romain 13, 149–60. Genève: Librairie Droz. Castelin, K. 1965. Die Goldprägung der Kelten in den böhmischen Ländern. Graz: Akademische Druck- und Verlagsanstalt. Deschler-Erb, E. 1999. Ad arma! Römisches Militär des 1. Jahrhunderts n. Chr. in Augusta Raurica, Forschungen in Augst 28. Augst: Römerstadt Augusta Raurica. Dolenz, H. 1996. Eisenverarbeitung auf dem Magdalensberg. In H. Straube (ed.) Ferrum Noricum und die Stadt auf dem Magdalensberg, 140–67. Wien: Springer Verlag. Ginoux, N. 2007. Le thème symbolique de «la paire de dragons» sur les fourreaux celtiques (IVe– IIe siècles av. J.-C.). Etude iconographique et typologie, BAR International Series 1702, Oxford: Tempus Reparatum. Hansen, L. 2003. Die Panzerung der Kelten. Eine diachrone und interkulturelle Untersuchung eisenzeitlicher Rüstungen. Kiel: Eigenverlag des Autors. Higgins, R. A. 1961. Greek and Roman Jewellery. London: Methuen. Kaposvári, G. 1969. A Jászberény-Cserőhalmi kelta temető, Archaeologiai Értesítő 96, 178–98. Karwowski, M. 2009. A Celtic hilltop settlement on Oberleiserberg in Lower Austria, Światowit – Supplement Series B: Barbaricum 8, 115–31. Karwowski, M. 2012. Der keltische Oberleiserberg. In E. Lauermann and P. Trebsche (eds.) Beiträge zum Tag der Niederösterreichischen Landesarchäologie 2012, Katalog des Niederösterreichischen Landesmuseums N. F. 507, 40–8. Asparn/Zaya: Urgeschichtemuseum Niederösterreich. Karwowski, M. and Militký, J. 2011. The Oberleiserberg types in the context of Taurisci influences. In M. Guštin and M. Jevtić (eds.) The Eastern Celts. The Communities between the Alps and the Black Sea, 131–6. Koper-Beograd: Univerza na Primorskem. Kaus, K. 1981. Lagerstätten und Produktionszentren des Ferrum Noricum. In Bergmännischer Verband Österreichs (ed.) Eisengewinnung und -verarbeitung in der Frühzeit, Leobener Grüne Hefte NF 2, 74–92. Wien: Montan-Verlag. Kellner, H.-J. 1990. Die Münzfunde von Manching und die keltischen Fundmünzen aus Südbayern. Die Ausgrabungen in Manching 12, Stuttgart: Steiner Verlag.

Kolníková, E. 1978. Keltské mince na Sovensku, Dávnoveké Umenie Slovenska 2. Bratislava: Pallas. Kopaliński, W. 1991. Słownik mitów i tradycji kultury (4th Ed.). Kraków: Państwowy Instytut Wydawniczy. Lenerz-de Wilde, M. 1980. Die frühlatènezeitlichen Gürtelhaken mit figuraler Verzierung, Germania 58.1–2, 61–103. Martelli, M. 2006. Kunst und Kultur. In M. Cristofani, Die Etrusker. Geheimnisvolle Kultur im antiken Italien (Sonderausgabe), 168–209. Stuttgart-Zürich: Belser Verlag. Megaw, R. and Megaw, J. V. S. 1989. The Italian Job: some implications of recent finds of Celtic scabbards decorated with Dragon-pairs, Mediterranean Archaeology 2, 85–100. Megaw, R. and Megaw, J. V. S. 2001. Celtic Art. From its Beginnings to the Book of Kells. London: Thames & Hudson. Militký, J. 2008. Mincovnictví v době laténské. In N. Venclová (ed.) Archeologie pravěkých Čech 7. Doba laténská, 122–8. Praha: Archeologický ústav AV ČR. de Navarro, J. M. 1959. A bronze mount of the La Tène period from Kelheim, Lower Bavaria, Germania 37, 131–40. Olmsted, G. 2001. Celtic art in transition during the first century BC. An Examination of the Creations of Mint Masters and Metal Smiths, and an Analysis of Stylistic Development During the Phase between La Tène and Provincial Roman, Archaeolingua 12. Budapest: Archaeolingua Alapítvány. Petres, É. F. 1982. Notes on scabbards decorated with dragons and bird-pairs. In P.-M. Duval and V. Kruta (eds) L’Art Celtique de la Période d’Expansion: IVe et IIIe siècle avant notre ère, Hautes Etudes du monde gréco-romain 13, 161–74. Genève: Librairie Droz. Simek, R. 2004. Schlange. In J. Hoops (ed.) Reallexikon der Germanischen Altertumskunde, Vol. 27 (Schere-Secundus von Trient) (2nd Ed.), 144–6. Berlin-New York: Walter de Gruyter. Simmons, V. 2006. Dragons. In J. T. Koch (ed.) Celtic Culture. A Historical Encyclopedia, Vol. 2 (Celto-F), 609. Santa Barbara: ABC-CLIO. Sperl, G. 2002. Ferrum Noricum – Stand der Forschung über eine frühe Stahlqualität, Berg- und Hüttenmännische Monatshefte 147.4, 61–5. Straube, H. 1996. Ferrum Noricum und die Stadt auf dem Magdalensberg. Wien: Springer Verlag. Szabó, M. 2012. Drache. In S. Sievers, O. H. Urban, and P. C. Ramsl (eds.) Lexikon zur Keltischen Archäologie, Mitteilungen der Prähistorischen Kommission 73, 439–40. Wien: Österreichische Akademie der Wissenschaften. Szabó, M. and Petres, É. F. 1992. Decorated Weapons of the La Tène Iron Age in the Carpathian Basin, Inventaria Praehistorica Hungariae 5. Budapest: Magyar Nemzeti Múzeum. Viand, V. 2008. Les fragments de cotte de mailles de Vernon. Armure souple véliocasse ou présence romaine aux portes de l’oppidum? In M. Poux (ed.) Sur les traces de César. Militaria tardo-républicains en contexte gaulois, Colection Bibracte 14, 33–46. Glux-en-Glenne: Centre Archéologique Européen du Mont Beuvray.

28 A L’AUBE DU IIIE S. AV. J.-C.: LES FOURREAUX D’ÉPÉE À DÉCOR ESTAMPÉ SUR FER Thierry Lejars

Introduction Si nous ne disposions que des données funéraires pour connaître les Celtes de La Tène et leur production artistique, des territoires entiers demeureraient encore dans l’ombre. S’il ne fallait compter que sur les objets en parfait état, parce que le feu ou l’eau en aurait favorisé la conservation, quantité d’artefacts auraient sombré dans l’oubli. Or, nous savons d’expérience que les usages funéraires ont considérablement varié non seulement pendant les cinq siècles du second âge du Fer, mais aussi d’une région à l’autre, démontrant qu’il n’existe pas en ce domaine de rituel spécifiquement celtique. L’expérience nous a également appris que la rouille qui encombre la grande majorité des fers anciens, cache également bien des surprises. La procédure visant à débarrasser l’objet de son enveloppe corrodée est aussi ancienne que la discipline archéologique elle-même. Pour les savants du XIXe s. comme pour nous aujourd’hui la ruine des objets en fer a toujours constitué un problème. En 1865, A. Verchère de Reffye (1865a; 1865b) constatait amèrement «qu’au commencement de notre siècle, ce pauvre fer était tenu dans un tel mépris, que personne ne le ramassait, et qu’aucune collection ne daignait lui donner asile; si bien qu’il en était résulté ce préjugé qui demeure encore dans bien des esprits, que les anciens ne faisaient pas usage du fer». On peut se faire une idée assez précise des traitements mis en œuvre à cette époque, à Mayence et à Saint-Germain-en-Laye, à travers la note qu’il publie en 1865 dans la Revue Archéologique. L’article destiné à sensibiliser et informer les archéologues est repris dans les colonnes des Matériaux pour l’histoire positive et philosophique de l’homme. Il s’agit là d’un

véritable plaidoyer pour le fer, «l’agent le plus actif de la civilisation». Dans ce domaine comme dans bien d’autres, le nouveau Musée de Saint-Germain-en-Laye s’inspire de Mayence, référence obligée pour tout ce qui a trait à l’étude et à la conservation des antiquités. La méthode préconisée, appliquée au matériel d’Alésia, vise à retrouver «les formes primitives de la pièce, les gravures ou les inscriptions qu’elle portait». La mise au jour à Gournay-sur-Aronde, entre 1975 et 1984, d’un nombre considérable d’armes et fibules en fer, fortement corrodées, a conduit les archéologues à se doter des moyens nécessaires pour tirer le meilleur parti de cette documentation d’un genre inédit. Le recours de plus en plus systématique à la restauration à partir des années 80, a fini par porter ses fruits. La restauration complète ou partielle des fers de Gournay a rendu possible de grandes avancées quant à la connaissance de l’armement celtique (Brunaux and Rapin 1988; Lejars 1994) et a permis d’enrichir de manière considérable le corpus des armes ornées dont l’aire de distribution fut longtemps circonscrite aux domaines centre-européen (Hongrie, Autriche et Suisse) et insulaire (Jacobsthal 1944; Duval 1977; Megaw and Megaw 1989; Jope 2000; Stead 2006). En 1977, P.-M. Duval pouvait écrire – c’est le sentiment qui prévalait alors – que l’ornementation des fourreaux «s’est développée surtout en Hongrie; de là, a gagné la Suisse, puis à travers la Gaule, les îles atlantiques, et par la Roumanie, la Yougoslavie» (Duval 1977, 120). Cette vision, qui a longtemps prévalu, est liée à l’idée d’un monde celtique en expansion, mû par d’incessantes migrations, et une conception de l’artisanat calquée sur l’approche des historiens d’art du monde «classique». En France, dans ces mêmes années, se sont les recherches

240

Thierry Lejars

systématiques menées par A. Bulard qui permirent de constituer les bases d’un corpus des armes ornées (Bulard 1979; 1982). Là encore, se sont les trouvailles en milieu humide qui composent l’essentiel du corpus. L’inventaire publié une dizaine d’année plus tard par N. Ginoux permet de multiplier par cinq le corpus initial, engrangeant les premiers résultats d’une politique de restauration en plein développement (Ginoux 1994; 1995) (le catalogue fait état de 79 armes ornées pour la période Ve–IIe s. avant J.-C.). Dans le même temps, Otto Hermann Frey répertorie 144 représentations de paires de dragons sur l’ensemble de l’Europe celtique quand, deux décennies plus tôt, J. M. De Navarro n’en comptabilisait qu’une trentaine seulement (Frey 1995b; De Navarro 1972, 66). A Gournay-sur-Aronde, 65 des 111 entrées de fourreau nettoyées ont restitué les traces d’un décor mais il est fort probable que ce chiffre ait été à l’origine beaucoup plus important; trop souvent les décors ne subsistent plus qu’à l’état de traces et beaucoup ont dû disparaître. A La Tène, le taux de fourreaux ornés est d’environ 70%; parmi les 30% restant beaucoup sont fortement corrodés ou dégradés (Lejars 1994, 63). Dans ce panorama, le sanctuaire de Gournay-sur-Aronde n’est pas un cas isolé. Depuis sa découverte nombre de lieux de culte, certains au demeurant fort spectaculaires comme Ribemont-sur-Ancre ou Tintignac, ont été mis au jour et fouillés, livrant des centaines, voire des milliers, d’objets en fer. Si certains sont relativement bien conservés et complets, beaucoup sont réduits à l’état de fragments. Pour autant, ces résidus d’objets ne sont pas, loin s’en faut, dénués d’intérêt. Ainsi, les travaux menés entre 1994 et 2011 sur le grand sanctuaire romain de «La Tour aux Fées» à Allonnes, à quelques kilomètres au sud de la ville du Mans, ont révélé l’existence de niveaux d’occupation préromains et surtout un très abondant mobilier métallique laténien, avec en particulier de nombreux restes de fourreaux d’épée décorés (lyres zoomorphes, paire de dragons, décors de Style végétal continu, décors estampés, etc.) (présentation préliminaire dans Brouquier-Reddé and Gruel 2004. Nous voudrions nous arrêter ici sur le cas d’un modeste fragment orné qui vient compléter la petite série d’armes à décor estampé du début de La Tène moyenne disséminée à travers l’Europe celtique, et splendidement illustrée par le fourreau de Potypuszta en Hongrie et, plus près de nous, par deux pièces de Gournay-sur-Aronde sur lesquelles nous aurons à revenir.

La Tour aux Fées à Allonnes (Sarthe) Le sanctuaire gaulois et romain de La Tour aux Fées à Allonnes a livré au fil des campagnes de fouilles quantité d’armes, et quelques éléments de parure, qui attestent d’une fréquentation régulière du sanctuaire – bien que ces objets soient presque toujours trouvés en situation résiduelle et que les aménagements protohistoriques soient

difficilement datables en l’état – depuis le IVe siècle jusqu’à la romanisation et la monumentalisation des installations cultuelles. Si certaines pièces d’armement tardives datent du IIe et Ier siècle av. J.-C., voire de la période augustéenne, la majorité d’entre elles peut être assignée à La Tène ancienne et au début de La Tène moyenne (IVe et IIIe siècles av. J.-C.) (Brouquier-Reddé and Gruel 2004, 296–306, pour le matériel d’époque gauloise). Le fragment de fourreau inventorié Al 05 3008 a été découvert en juillet 2001 dans le secteur du Portique ouest, entre les murs 11 et 14. Il mesure 65 mm de longueur pour 47 de large (Fig. 28.1). La bordure, conservée sur un seul côté, est marquée par une forte gouttière arrondie large de 5 mm. Le fragment, comme la plupart des fers protohistoriques découverts sur le site, était recouvert d’une épaisse gangue sablo-argileuse qui masquait la surface. C’est la restauration qui a permis de reconnaître la forme de l’objet et de son décor. Le caractère résiduel de la pièce est également confirmé par le fait qu’aucun fragment complémentaire n’ait été identifié jusqu’à maintenant. Il semble que le décor estampé couvre la totalité de la plaque antérieure à l’exception de la gouttière. Le décor se présente sous la forme d’un bandeau oblique en relief (descendant) réalisé au repoussé, sur lequel vient se greffer perpendiculairement un second bandeau (ascendant). Le commencement d’un troisième (descendant) est perceptible dans le bas, sur la gauche. Les bandeaux, bombés, sont rehaussés d’une suite de cartouches quadrangulaires estampés, ornés chacun, autant que l’on puisse en juger, d’un motif asymétrique organisé autour d’une esse oblique (avec des gousses disposées tête bêche de part et d’autre de l’esse oblique). L’espace losangé axial, délimité par les bandeaux, est meublé au centre d’un cartouche en léger relief orné d’un rinceau (les bribes de décor conservées trahissent un motif semblable à celui des estampilles imprimées sur les bandeaux) tandis que des palmettes (incluses dans un segment de cercle dont les extrémités en volute tournées vers l’intérieur forment les feuilles basses d’une palmette à trois branches) coiffée d’un ocelle garnissent les écoinçons, réduisant considérablement les vides. Le champ triangulaire (ou demi-losangé) latéral est quant à lui uniquement orné d’une palmette adossée à la gouttière et coiffée, comme les précédentes, d’un ocelle.

Les armes à décor estampés Cette composition complexe, jouant sur les reliefs qui structurent géométriquement l’espace et les motifs imprimés qui l’animent (restitution du décor, cf. Fig. 28.2.3), présente d’évidentes similitudes avec les armes à décor estampés du début de La Tène moyenne, un sujet qui a retenu, il y a près de 35 ans, l’attention de quelques uns de nos illustres prédécesseurs, en particulier O. H. Frey qui proposait

28.  A l’aube du IIIe s. av. J.-C.: les fourreaux d’épée à décor estampé sur fer

241

Fig. 28.1. Allonnes 05 3008 (cliché, Patrick Ernaut, Inrap, dessins, Th. Lejars del./Cnrs).

en 1978 une brève, mais importante, étude consacrée au fourreau de Graz et John Vincent S. Megaw qui revenait sur cet argument dans une note publiée la même année à propos du décor du fourreau orné de la tombe 115 de Jenisuv Ujezd (Frey 1978/79; Megaw 1978). Cependant, avant même de rappeler ce que l’on doit à ces auteurs, il convient de nous arrêter quelques instants sur le fourreau de Potypuszta évoqué précédemment. Publié simultanément par Ilona Hunyady (Hunyady 1942/1944, 112, pl. XLII.7, XLV.3) et Paul Jacobsthal (Jacobsthal 1944, n 127, 177) et depuis souvent mentionné dans la littérature archéologique, l’arme a été plus récemment étudiée par Eva F. Petres et Miklos Szabo à l’occasion de l’édition du corpus des

armes ornées du bassin des Carpates en 1992 (Szabo and Petres 1992, n 50, 21–23, 96, pl. 54, ill. h.t. I.). Jacobsthal indique Velem-Szentvid et Szombathely (le nom du musée où est conservée la pièce) pour la provenance et parle d’un fourreau en bronze, une erreur induite par le très bon état de conservation du métal. Pour P.-M. Duval (Duval 1977, 255), il s’agit encore d’un fourreau en bronze, avec une excellente photographie de la partie médiane du fourreau, (ibid., fig. 299). Il est probable que cette arme, sans contexte connu, mais complète et d’une qualité de conservation tout à fait remarquable, provienne d’une incinération. Le fourreau mesure 69,5 cm de long (74,8 cm avec l’épée) pour une largeur de 5,4 cm. A la différence du fragment d’Allonnes, le

242

Thierry Lejars

Fig. 28.2. 1 – Potypuzsta (d’apr. Szabo and Petres 1992), 2 – Graz (d’apr. Frey 1978/79), 3 – Allonnes 05 3008, 4 – GSA 2351/2353 (d’apr. Lejars 1994), 5 – Manching «Steinbichel» t. 17 (d’apr. Jacobi 1982), 6 – GSA 1116/1122, 7 – Ensérune (cliché, P. Gardin, Irrap), (dessins, Th. Lejars del./Cnrs).

28.  A l’aube du IIIe s. av. J.-C.: les fourreaux d’épée à décor estampé sur fer fourreau de Potypuszta présente une forte nervure médiane. Les gouttières latérales sont larges et saillantes. Le pontet, arqué, est fixé au fourreau par deux pattes arrondies. La bouterolle, longue d’environ 15 cm, se termine par une extrémité arrondie, sans ajours visibles. Le décor, organisé en bandeaux obliques, ascendants et descendants, n’est pas côtelé comme à Allonnes, mais estampillé d’une suite de motifs losangés ornés chacun de deux demi-palmettes opposées (symétrie par rotation de motifs en amande coiffé d’une esse flanquée de hachures) (restitution du décor, cf. Fig. 28.2.1). Le décor est souligné près de l’embouchure par la juxtaposition de huit estampilles rangées en deux files horizontales. Les champs axiaux, également losangés, sont meublés au centre, comme à Allonnes, d’une estampille losangée bordée sur les côtés, de quatre palmettes (incluses dans un segment de cercle) alternant avec autant d’ocelles pointées. Les champs latéraux, triangulaires (ou demilosange), sont quant à eux ornés d’une palmette identique aux précédentes, adossée à la gouttière et flanquée de trois ocelles pointés. Les estampilles recouvrent la nervure et les gouttières, ce qui implique que le décor ait été réalisé avant la mise en forme de la plaque. L’organisation des estampilles en bandeau obliques croisés signale un petit groupe d’armes ornées qui retiendra ici notre attention pour montrer, en dépit d’un état de conservation très inégal, et au-delà des fortes similitudes de forme et de décor, la variété des solutions adoptées par les armuriers.

Graz, Manching, Gournay-sur-Aronde Il convient, en premier lieu, d’examiner le fourreau de Graz «Laubgasse», également sans contexte connu. L’épée et son fourreau ont été publiés par O. H. Frey en 1978/79 (Frey 1978/79, 67: l’épée aurait été trouvée avec une fibule de La Tène moyenne, non conservée, dans une ballastière), puis de nouveau par M. Szabo et E. F. Petres en 1992 (Szabo et Petres 1992, n 76, 102, pl. 79). En se fondant sur les travaux d’U. Osterhaus (1966) et de J. M. De Navarro (1972), Frey propose d’attribuer l’épée de Graz à la fin de La Tène ancienne ou au début de la période suivante. Il se réfère également aux trouvailles d’Europe centrale et d’Italie (Montefortino et Ceretolo). Moins d’une centaine de kilomètres séparent Graz et Szombathely (Potypuszta). L’arme, bien que complète, est fort dégradée. La partie supérieure du fourreau très altérée est lacunaire, tandis que le décor qui à l’origine recouvrait la totalité de la plaque frontale n’est plus visible que partiellement, dans la partie haute et près de la bouterolle. L’épée mesure (avec le fourreau) 74,8 cm pour 4,4 cm de large et 110 mm pour la soie. La longueur du fourreau est d’environ 62/63 cm pour une largeur de 5,4 cm. Ce dernier se signale en outre par des gouttières latérales et une nervure médiane larges. Le

243

pontet, trapu, présente des bords concaves et deux larges pattes de fixation arrondies. La bouterolle, longue d’environ 15 cm, se signale quant à elle par une extrémité arrondie (la corrosion ne permet pas de vérifier l’existence d’ajours). Le décor estampé couvre cette fois encore la nervure médiane (restitution du décor, Fig. 28.2.2). Il se compose de motifs losangés juxtaposés, organisés en rangées obliques perpendiculaires. Comme sur les exemplaires précédents, le losange renferme un motif composé de deux demi-palmettes opposées avec hachures. En dépit du mauvais état de conservation de la pièce on discerne deux autres estampilles, la palmette (demi-cercle avec enroulement terminaux) et l’ocelle, utilisés pour meubler les grands losanges axiaux et les champs triangulaires latéraux (voir la reconstitution que nous proposons à partir du relevé publié par O. H. Frey). Il convient enfin d’insister sur la similitude des estampilles, à quelques détails près (les motifs à l’intérieur des losanges sont figurés en miroir) ornant les fourreaux de Graz et de Potypuszta ) (Frey 1978/79, 68). Le second exemple nous est donné par le fourreau de la tombe 27 de la nécropole de Steinbichel à Manching, en Bavière. Le contexte de la découverte est important dans la mesure où il permet de préciser, pour la première fois, le cadre typo-chronologique de cette forme de fourreau (cf. infra). L’épée, longue de 755 mm (longueur fourreau restituée environ 66 cm pour 5,5 cm de large) est gainée dans un fourreau avec bouterolle circulaire ajourée (la longueur de la bouterolle n’est pas déterminable) (Kramer 1985, 83, pl. 16). Le pontet, trapu, est fixé par deux larges attaches arrondies. Le fourreau, non nervuré, est intégralement couvert d’estampilles losangées à bords concaves (le décor a d’abord été publié par Jacobi en 1982, 567, fig. 2; voir aussi Szabo and Petres 1992, 22–23, fig. 6). Contrairement aux exemplaires précédents, la répartition est uniforme et sans relief. Le motif interne diffère également. Il s’agit d’un rinceau curviligne avec des bossettes placées dans les boucles et sur les bords de l’estampille (Fig. 28.2.5). Le troisième exemple provient du sanctuaire de Gournaysur-Aronde dans l’Oise qui a livré plusieurs fourreaux, complets ou fragmentaires, avec décor estampé. Ces armes, au nombre de quatre, sont datées, par analogie avec les trouvailles en contexte funéraire du début de La Tène C1. Pour le moment, seuls les fourreaux GSA 2351/2353 et GSA 1116/1122 retiendrons notre attention. Les deux pièces sont incomplètes : il manque au premier la bouterolle, tandis que seule la plaque antérieure du second est conservée. Comme les exemplaires précédents, ils sont pourvus de larges gouttières latérales, mais sans nervure médiane. Le décor est constitué dans les deux cas d’un maillage oblique régulier en léger relief. Il déborde sur les gouttières, ce qui laisse penser qu’il a été réalisé avant le formage des bords. L’embouchure de GSA 2351/2353 est soulignée par deux bandes horizontales réalisées au repoussé. Les estampilles losangées de ce dernier, d’environ 1 cm de côté, sont

244

Thierry Lejars

enrichies d’un motif composé de deux parties symétriques (symétrie par rotation): une feuille horizontale avec sur le côté un rinceau vertical se terminant par un enroulement en crosse (Fig. 28.2.4 et 3). Le décor du second fourreau, GSA 1116/1122 est tout aussi complexe. Les observations effectuées au cours du nettoyage consentent de restituer, en dépit du mauvais état des surfaces, un motif interne en forme de rinceau curviligne analogue à celui mis en évidence sur le fourreau de Manching (Fig. 28.2.5 et 3).

La chronologie Si le fragment d’Allonnes est trop lacunaire pour offrir un quelconque point d’appui à la chronologie, les armes examinées présentent d’indéniables similitudes quant à leur taille et leur morphologie. Elles sont perceptibles dans les dimensions avec des longueurs comprises entre 62 et 67 cm et entre 5,2 et 5,5 cm pour la largeur (Fig. 28.4). La pièce de suspension est constituée d’un pontet trapu encadré de deux larges pattes de fixation arrondies. La bouterolle, longue de 15 à 16 cm se signale par une extrémité massive circulaire ou arrondie; les ajours sont quasiment inexistants. Les gouttières sont larges et saillantes ainsi que la nervure médiane lorsqu’elle est présente. Malheureusement, leur contexte de découverte est loin d’être aussi clair. Seul le fourreau de Manching bénéficie de cet avantage. L’assemblage se compose, outre l’épée, d’un fer de lance à flamme biconvexe, d’une chaîne de suspension à double torsade et anneaux hypertrophiés et d’un brassard en sapropélite. Si la bouterolle circulaire ajourée du fourreau renvoie à la tradition des armes de La Tène ancienne, la morphologie du pontet et le système de suspension de l’épée permettent de situer l’ensemble dans un horizon tardif de La Tène B2 ou du début de La Tène C1. C’est la datation que nous avions déjà proposé pour les fourreaux d’épée avec bouterolle à extrémité massive arrondie présents dans le sanctuaire de Gournay-sur-Aronde (Lejars 1994, 55–56). La confrontation de ces critères morphologiques avec les associations en contexte funéraire nous avait également permis d’associer chronologiquement cette forme de fourreaux aux exemplaires équipés de bouterolle circulaire ajourée les plus tardifs. Pour compléter notre propos et appuyer notre démonstration, nous pouvons mentionner le mobilier de la tombe 1 de Moulin-sur-Yèvre à Maubranches (Cher). Le fourreau, long de 61 cm et large de 5,5 cm, était associée à une chaîne de ceinturon avec maillons à double torsade, du type de celle qui accompagne le fourreau de Cernon-sur-Coole, un fer de lance et un umbo de bouclier à ailettes rectangulaires (Ginoux 1994, 74, n 44 avec bibliographie, pl. XI; Kruta Poppi 1984, 58–59, fig. 6, avec photographies du décor). L’épée et le mobilier associé sont attribués à La Tène moyenne. Dans le catalogue de l’exposition L’art celtique en Gaule (1983/84, 177, n 214),

une datation excessivement basse est proposée pour l’épée de Maubranches (La Tène finale), alors que Déchelette (1914, 1119, fig. 463.3) rangeait déjà cette épée parmi les armes de La Tène moyenne. Il arbore sur la plaque frontale un décor, en relief, réalisé au repoussé qui se présente sous la forme d’un entrelacs (Fig. 28.5.2). Il semble que des cercles poinçonnés, disposés dans les vides, complétaient le décor. Des traces d’ornementation sont également perceptibles sur la bouterolle, massive et arrondie, longue de 15 cm. Comme pour les armes précédentes, les gouttières sont larges et repliées sur l’arrière, et le pontet relativement court, fixé par deux attaches arrondies. Le décor de cette arme n’est pas sans analogie avec celui d’un autre fourreau d’Italie du nord, découvert en 1883 à Saliceta San Giuliano, près de Modène (Kruta Poppi 1978, 427; 1983, 22). Les premières trouvailles remontent à 1876, en particulier le mobilier d’une sépulture féminine. Les vestiges recueillis en 1883 proviennent de plusieurs ensembles datant pour certains de la phase étrusque antérieure (fibule de type Certosa), les autres de la période gauloise et romaine. Outre l’équipement d’un guerrier, on dénombre les vestiges d’un ou plusieurs assemblages féminins avec pas moins de six bracelets en bronze. L’équipement du guerrier connu par les photographies d’A. Crespellani se compose d’une épée avec son ceinturon métallique composé de maillons à triple torsades (analogue à la chaîne de ceinturon de Cernon-sur-Coole) et un fer de lance (Kruta Poppi 1978, 429, fig. 4: dessins d’après les photographies de Crespellani 1887). En 1984, L. Kruta Poppi (1984, 53) indique la présence possible d’un anneau double muni de pointe (en faite une pièce de harnais de cheval, mais que l’on ne rencontre jamais dans la tombes celtiques de Cisalpine) et de deux fragments de lame en fer d’un couteau ou d’un rasoir (Bergonzi 1988, 157, fig. 108). Elle propose de voir dans la partie de l’épée conservée, longue d’une quarantaine de centimètre, les deux tiers inférieurs de l’arme (Kruta Poppi 1984, 53). Du fourreau, il ne subsiste que trois fragments disjoints de la plaque frontale (ibid., 52–53, l’auteur indique qu’un seul des trois fragments a été restauré. Le décor n’est pas signalé dans la publication de 1978). De la bouterolle, n’est conservée que la pointe triangulaire. Le décor, en relief, est constitué par un entrelacs angulaire qui court entre les gouttières latérales et délimite une série de rhombes au centre et des triangles sur les côtés. S’il y eut d’autres motifs, plus discrets, il n’en subsiste aucune trace du fait de la corrosion. Les deux larges cannelures perpendiculaires qui ferment la partie du fourreau conservée marqueraient d’après L. Kruta Poppi l’emplacement de la barrette transversale de la bouterolle (ibid., 53: l’auteur note un léger rétrécissement de cette partie). On découvre dans le catalogue de l’exposition consacrée à l’histoire de Modène, publié en 1988 (Bergonzi 1988, 157, fig. 108: photographie de l’épée avec la base de la soie par Kruta 1988, fig. 231) un dessin de l’arme sensiblement différent,

28.  A l’aube du IIIe s. av. J.-C.: les fourreaux d’épée à décor estampé sur fer

245

Fig. 28.3. 1 – GSA 2351/2353, 2 – GSA 1116/1122 (d’apr. Lejars 1994 ; clichés, Patrick Ernaut, Inrap et Renaud Bernadet, dessins, Th. Lejars del./Cnrs).

246

Thierry Lejars

Allonnes 05 3008 Potypuzsta Graz « Laubgasse » Manching, « Steinbichel » 27 Gournay 2351/2353 Gournay 1116/1122 Maubranches 1 Saliceta San Giuliano Malé Kosihy 31

Longueur 695 620/630 660 610

largeur > 47 54 54 55 52

Long. bouterolle

55 55 58

150

150 150

Fig. 28.4. Tableau synthétique des mesures connues des fourreaux d’épée à décor estampé sur fer.

avec un fragment correspondant au départ de la soie placé juste au-dessus des bandeaux horizontaux. Cette dernière présentation nous paraît plus conforme à la réalité dans la mesure où les bandeaux – deux larges bandes en relief et une plus étroite au centre – ne se rencontrent jamais dans la partie du fourreau signalant l’emplacement de l’entrée de la bouterolle. On peut également douter de l’appartenance du fragment terminal de bouterolle à cette arme… une forme arrondie aurait été plus appropriée. Pour ces raisons, et au regard des armes décrites précédemment, je crois préférable de voir dans les bandeaux horizontaux un motif signalant l’embouchure du fourreau (vue restituée, Fig. 28.5.1). G. Bergonzi indique également la présence de bossettes dans les triangles latéraux de l’entrelacs. Pour clore cet excursus chronologique, on peut signaler le mobilier de deux tombes de guerriers équipés de fourreaux supportant un décor en relief, bien que de nature différente des exemples examinés jusqu’ici. Le premier est illustré par la tombe de guerrier de Rungis qui a livré avec l’épée une chaîne de ceinturon de type «échelle», un fer de lance, un umbo de bouclier avec coque étroite nervurée et ailettes circulaires, et une fibule en fer à pied libre et arc torsadé (Kruta and Rapin 1987; Ginoux 1994, 84, n 74, pl. XIX). Du fourreau, il ne subsiste que les deux tiers supérieurs. Le pontet est fixé par deux larges attaches arrondies. Le décor se développe sur l’ensemble de la surface sur deux plans successifs: on distingue un premier décor de 16 bossettes rehaussées de fines gravures qui divise la plaque en champs obliques et un second, gravé (peu lisible), figurant une lyre hachurée sous l’entrée (pour le décor, voir Ginoux 1994). Le second exemple nous est offert par le mobilier de l’incinération 31 de Malé Kosihy en Slovaquie (Bujna 1995b, 24–25, pl. 6, le mobilier est très fragmenté). L’assemblage est constitué de l’épée, d’une chaîne de suspension de type «échelle», de garnitures diverses se rapportant pour partie à un bouclier et des restes de deux fibules en fer à pied libre. Le fourreau, non nervuré, est partiellement conservé. Il manque la bouterolle mais le pontet est à pattes de fixation larges (Fig. 28.5.4). Le décor du fourreau, restitué, se compose de paires de côtes parallèles réalisées au repoussé. Elles sont disposées, les premières, horizontalement sous l’entrée et, les secondes, suivant un axe oblique descendant sur le reste

du fourreau. Une trame de motifs losangés concentriques estampés meuble les vides entre les côtes (Bujna 1995a, 260: largeur fourreau 58 mm ). Si ce décor estampé sur la face interne semble attester une réutilisation de la plaque, comme le suggère J. Bujna, on s’étonne de voir les losanges parfaitement insérés entre les côtes qui auraient été réalisées (également sur l’arrière) dans un second temps. Plutôt que de voir là le résultat de deux états successifs, on peut se demander si les petites estampilles losangées (environ 5 mm de côté contre 1 cm pour les estampes réalisées sur les fourreaux précédents) n’ont pas été imprimées à dessein sur l’arrière de la plaque comme on peut le voir avec les esses estampés sur le fourreau plus ancien, également en fer, de Sogny-aux-Moulins (Marne) (Fig. 28.6). L’association récurrente de ces fourreaux avec des formes anciennes de ceinturons métalliques (maillons à double ou triple torsade, maillons «échelle») et d’umbos de bouclier (umbos monocoques à ailettes rectangulaires hautes ou demi-circulaires et coque nervurée), mais aussi les dernières fibules à pied libre, conforte la datation de ces armes à la transition La Tène ancienne – moyenne avancée il y a maintenant un peu plus d’une trentaine d’années par O.H. Frey (Frey 1978/79, 68–69).

L’organisation du décor: fer et bronze Les décors examinés ont en commun la technique de l’estampage sur fer avec l’usage de poinçons élaborés, combinée ou non avec des parties exécutées au repoussé, et une très forte propension à couvrir la totalité de la plaque frontale. Bien entendu, on connaît de nombreux exemples de fourreaux d’épée entièrement ornés, avec des décors qui parfois couvrent non seulement les deux faces du fourreau mais également le pontet et la bouterolle comme celui de Cernon-sur-Coole (Duval and Kruta 1986). Pour nous en tenir au registre des décors estampés ou réalisés au repoussé, on rappellera pour La Tène ancienne les exemples fameux de Filottrano, de Moscano di Fabriano, d’Epiais Rhus, de Saint-Germainmont et de Jenisuv Ujezd qui ont fait l’objet de nombreuses études (Frey 1971; Megaw 1978; Kruta et al. 1984; Rapin 2006a; 2008). Il s’agit de fourreaux mixtes,

28.  A l’aube du IIIe s. av. J.-C.: les fourreaux d’épée à décor estampé sur fer

247

Fig. 28.5. 1 – Saliceta san Giuliano (d’apr. Kruta Poppi 1984 et Bergonzi 1988), 2 – Maubranches (d’apr. Ginoux 1994), 3 – GSA 2186 (d’apr. Lejars 1994), 4 – Malé Kosihy t. 31 (d’apr. Bujna 1995a), 5 – GSA 2676 (d’apr. Lejars 1994) (dessins 1–3, 5, Th. Lejars del./Cnrs).

combinant fer et bronze. Les estampilles souvent élaborées (rinceaux de Style végétal continu, peltes, esses et ocelles) sont imprimées sur les plaques antérieures en bronze plus malléables. Les exemples de décors estampés sur tôle en fer

demeurent relativement exceptionnels pendant toute cette période, probablement en raison de la difficulté de réaliser des tôles aussi fines que celles en bronze (Rapin 1999, 42). A la différence des estampes sur bronze, celles pour le fer

248

Thierry Lejars

Fig. 28.6. Sogny-aux-Moulins, Marne (cliché, P. Gardin, Irrap).

demeurent pendant tout La Tène ancienne (Ve et IVe s. av. J.-C.) beaucoup plus élémentaires: des bossettes simples dessinant des figures curvilignes ou géométriques comme à Prosnes (Bataille-Melkon and Rapin 1997) et à Cortrat (Rapin and Zurfluh 1998); des ocelles et des esses – paires d’esses organisées en quadriscèles – sur l’exemplaire de Sogny-auxMoulins (Fig. 28.6). A La Tène ancienne, les estampilles ne couvrent pas la totalité de la plaque frontale mais s’organisent en bandeaux horizontaux ou verticaux alternant avec des espaces entièrement vides (gravé au burin balancé, le décor du fourreau de Méroux présente une organisation très semblable (Megaw 1968; Ginoux 1994, 70, pl. VI.1)). Si la technique du repoussé est fréquemment employée sur les tôles de bronze

Fig. 28.7. Entrelacs. 1 – grand entrelacs asymétrique, 2 – petit entrelacs, 3 – entrelacs anguleux régulier, 4 – entrelacs curviligne régulier, 5 – entrelacs anguleux fibule de type Duchov, 6 – entrelacs curviligne, fibule à décor plastique de Juvigné (dessins, Th. Lejars del./Cnrs).

28.  A l’aube du IIIe s. av. J.-C.: les fourreaux d’épée à décor estampé sur fer

249

Fig. 28.8. Palmettes. 1 – Potypuzsta (d’apr. Szabo and Petres 1992), 2 – Allonnes (d’apr. Fig. 1), 3 – Canosa (d’apr. Jacobsthal 1944), 4 – Comacchio (d’apr. Jacobsthal 1944), 5 – Voivodine (d’apr. Jacobsthal 1944), (dessins, Th. Lejars del./Cnrs).

pour donner du relief, c’est par le ciselage que l’on cherche à animer les surfaces en fer – par exemple, Monte Bibele, tombe 126 (Rapin, Szabo and Vitali 1992), Flavigny (Rapin 2006a, 201; 2006b), bouterolle et pontet de Cernon-sur-Coole (Duval and Kruta 1986) – ou par l’ajout d’éléments rapportés, comme à Gournay-sur-Aronde (Gournay 1826, 1860/2651, 1863, 2676, Lejars 1994, 64–73), Drna (Megaw 1973; Szabo and Petres 1992, 103–104 n 82, pl. 84–85), etc. La structuration du décor à partir d’axes obliques croisés (ou entrelacs) n’est pas à proprement parler une nouveauté. Si les méandres couchés encadrant les côtés de la dernière scène du fourreau de Hallstatt (Jacobsthal 1944, pl. 60, n 96; Egg and Schönfelder 2009) n’ont pas un rôle structurant, il en va tout autrement de l’exemplaire de Vert-la-Gravelle (Jacobsthal 1944, 92, pl. 56, n 90, pl. 268.216; Ginoux 1994, 71, n 36, pl. VI ). Là, un entrelacs anguleux se déploie sur la totalité de la plaque, avec dans les rhombes une svastika et dans les triangles latéraux des cercles concentriques. Ce schéma est également attesté sur des torques et certaines fibules de type Duchcov, sous forme de bandes obliques croisées, parallèles ou en zig zag (Kruta 1979, 109). A souligner en particulier le décor d’une des fibules de Repin, en Bohême, avec ses bandes hachurées croisées enfermant à l’intérieur de chaque rhombe un triple ocelle (Fig. 28.7.5) (ibid., 110). Sur les fourreaux du début de La Tène moyenne, la trame peut se développer suivant un schéma relativement dense et parfaitement uniforme (Manching, Gournay 1116/1122, 2351/2353, Malé Kosihy), ou lâche à la manière d’un entrelacs (Potypuszta, Graz, Allonnes, ainsi que Saliceta San Giuliano et Maubranches). Des réseaux de petites estampilles losangées distribuées de façon très régulières caractérisent de la même manière un second fragment de fourreau découvert à Allonnes (Lejars 2003, 26, fig. 15.3) ainsi qu’un autre venant d’Ensérune (M.006, inédit). Le motif se présente comme un ocelle inscrit dans un losange.

La trame peut, suivant les cas, être matérialisée unique­ ment par l’alignement des estampilles losangées comme à Potypuszta et à Graz, ou par un traitement en relief avec la formation de véritables cordons (Allonnes, Gournay 2351/2353, Saliceta San Giuliano et Maubranche). Enfin, l’entrée est signalée par deux bandeaux transversaux (Potypuszta, Gournay 2351/2353, Saliceta San Giuliano, Malé Kosihy). L’entrelacs curviligne de Maubranches trouve quant à lui une intéressante comparaison dans le décor plastique (pseudo-filigrane) de la fibule en bronze de Juvigné, en Mayenne (Fig. 28.7.6) (Lejars dans Santrot, Santrot and Meuret 1999, 121: fibule à arc arrondi, pied triangulaire à grosse perle globulaire et ressort à six spires. L’appendice terminal est solidaire de l’arc même si sa morphologie est encore celle des fibules à pied libre de La Tène ancienne). Alors que l’entrelacs se déploie sur l’arc, des globules en meublent les vides internes et externes.

L’ocelle, la palmette et le losange Le nombre d’estampilles employées est limité. Le corpus se compose d’ocelles, de palmettes inscrites dans un segment de cercle et de losanges. Les motifs emblématiques du Style végétal continu en usage pendant la période précédente ne sont plus utilisés (rinceaux à triscèles de type A1 (terminologie élaborée par St. Verger 1987) des fourreaux de Moscano di Fabriano, d’Epiais Rhus et de Saint-Germainmont). De tous les motifs, l’ocelle (cercle simple ou double, pointé ou non) est le plus simple. Son emploi est documenté durant toute La Tène ancienne. On le trouve dès le début de la période sur les étuis de poignards, à Saint-Jean-sur-Tourbe (Ginoux 1994, 65, n 24, pl. II.5), et plus tard sur les fourreaux d’épée, par exemple Vert-laGravelle (Megaw 1978, 107, cf. note 48), Bussy-le-Château (Ginoux 1994, 68, n 30, fig. 4, pl. VII.2), Filottrano,

250

Thierry Lejars

Moscano di Fabriano, Epiais-Rhus, Jenisuv Uzejd (Kruta et al. 1984), Drna (Szabo and Petres 1992, n 81, pl. 83; Gournay 2186, Lejars 1994, 67) (Fig. 28.5.3). Les ocelles sont le plus souvent utilisés pour souligner les contours, délimiter l’entrée et la bouterolle (Megaw 1978, 107, avec une présentation relativement complète de l’utilisation de ce motif à La Tène ancienne sur des supports aussi bien métalliques que céramics). Leur emploi fait écho au travail du cuir comme on peut le voir sur l’étui de couteau de L’Eglise, en Belgique, avec ses motifs géométriques et floraux exécutés au repoussé (Cahen-Delhaye 1981). Des ocelles sont parfois utilisés en compléments sur des représentations de paires d’animaux fantastiques. Dans d’autres cas, le motif réduit à une simple bossette est organisé en compositions géométriques (fourreau en fer de Cortrat, Rapin and Zurfluh 1998, tombe 17, 57–59) ou couvre uniformément toute la surface (Allonnes et AubignéRacan Lejars 2003, 26, fig. 15.4–5; 2007, 277). L’ocelle, sans jamais vraiment disparaître, devient plus rare par la suite. Plus complexe, la palmette s’inscrit dans un demicercle terminé par des volutes tournées vers l’intérieur. Un motif foliacé, en forme de goutte, se déploie dans l’espace interne, entre les volutes, pour former une palmette à trois branches. Ce motif, s’il évoque d’une certaine manière les estampilles de palmettes imprimées sur les vases à vernis noir et certains casques en fer avec appliques décoratives en tôle de bronze (Smarjeta/Sanct Margaraten, Slovénie), trouve également des analogies dans le répertoire décoratif laténien comme l’avait déjà noté O.H. Frey (Frey 1978/79, 69). On peut mentionner, en particulier, des éléments de frises de la cruche de Besançon, du casque de Canosa, du torque en or de la tombe de Waldalgesheim, de l’applique de Brunn ou encore de celle de «Comacchio» (Fig. 28.8.3–4) (Jacobsthal 1944, n 143, pl. 83; Frey 1995a ). Sur le fer de lance de Joressant, au Haut-Vully (canton de Fribourg, Suisse), le décor, dans la tradition du Style des épées, combine, dans des champs triangulaires disposés de part et d’autre d’un losange rehaussé d’une svastika, un tourbillon de rinceaux curvilignes associant pelte et lyre zoomorphe (Megaw 1978, 108; Lejars 2012). Encore présent à l’aube de La Tène moyenne (Voivodine, Szabo and Petres 1992, n 127, pl. 119; le motif est placé entre les appendices supérieurs des deux lyres affrontées) (Fig. 28.8.5) le motif se transforme ensuite progressivement en lyre (goutte encadrée de deux cercles et disparition du contour circulaire) ou en boucle, e.g. Bölcske-Madocsahegy, Bodrogahalom (ibid., Bolcske-Madocsahegy 1, n 5, pl. 7 et Bodrogahalom, n 3, pl. 4; le motif coiffe l’ensemble du décor). En dépit de ces modifications, le motif n’en conserve pas moins sa centralité dans l’organisation générale du décor (comme élément de symétrie par rabattement pour les frises du IVe s., ou comme élément de liaison axial pour les exemplaires plus tardifs). A Bölcske-Madocsahegy, le motif issu de la fusion d’une

paire d’oiseaux à bec crochu, prend la forme, lorsque l’on renverse la figure, d’un masque coiffé de la double feuille de gui (Szabo 1993; 2003). Le dernier motif est assurément le plus complexe et celui qui pose a priori le plus de problèmes. Il se présente sous la forme d’un rinceau inclus dans un cartouche quadrangulaire posé sur la pointe (losange ou rhombe). Le motif interne se compose de deux feuilles en amande coiffées d’une esse à volute, flanquée de hachures, ou en d’autres termes deux demi-palmettes opposées. La figure est obtenue par le procédé de symétrie par rotation, une technique souvent employée pour les décors de Style hongrois (BölcskeMadocsahegy) mais aussi dans de nombreuses réalisations de Style végétal continu; c’est même un des ressorts de la dynamique des rinceaux à triscèles (Filottrano, Moscano di Fabriano – cf. Frey 1974, 142 note 37 : Kruta et al. 1984). Comme le notait O.H. Frey, l’estampille de Graz est le miroir quasi exact de celle de Potypuszta. O. H. Frey faisait dériver le motif avec ses hachures (parfaitement visibles sur l’exemplaire hongrois) de la «palmette hachurée» que l’on retrouve dans l’ornementation des vases de Basse-Yutz ou des appliques de Comacchio, ainsi que dans certaines réalisations plus récentes, en particulier du Style des épées (Frey 1978/79, 69, avec référence à Jacobsthal 1944, n 381, pour Basse-Yutz, n 113, pour Cernon-sur-Coole et n 126, pour de Mitricova, Slovénie). Les estampilles de Gournay et d’Allonnes sont à l’évidence – autant que l’on puisse en juger d’après leur état de conservation – distinctes des précédentes. A Gournay, les volutes terminales des esses font place à de larges boucles et les hachures à un semis de points disposés en périphérie, comme sur le fourreau de Manching. A Allonnes, le motif contraint par le cartouche s’inscrit dans un quadrilatère légèrement oblique. Les esses latérales ne sont plus perpendiculaires mais plaquées sur l’esse centrale, et les hachures supprimées. L’origine du motif est probablement à rechercher dans les rinceaux en esses du Style végétal continu (esses de la fibule de Schosshalde ou du casque de Canosa) (Fig. 28.9.6–8). L’estampage sur fer a conduit à une nécessaire simplification du motif. A la même époque, l’esse prolongé à chacune des extrémités d’une double volute, avec dans certains cas une évocation manifeste de la tête d’oiseau fantastique, figure en médaillon sur quelques armes comme le pontet du fourreau de Vojvodine ou encore sur les bossettes de la pièce de renfort du fourreau d’une épée de type Hatvan Boldog, provenant de Szob (Fig. 28.9.9) (Szabo and Petres, 1992, 100, n 68, pl. 71). A l’évidence, la finesse du ciselage permet des raffinements de détails que n’autorise pas la technique de l’estampage. Inscrits dans un cartouche losangé, les méandres curvilignes de Manching et de Gournay 1116/1122 se déploient de façon continue et symétrique de part et d’autre de l’esse centrale. A bien y regarder, ces motifs ne sont pas si différents des précédents dans la mesure où ils s’organisent,

28.  A l’aube du IIIe s. av. J.-C.: les fourreaux d’épée à décor estampé sur fer

251

Fig. 28.9. Rinceaux des losanges. 1 et 16 – Potypuzsta (d’apr. Szabo and Petres 1992), 2 – Graz (d’apr. Frey 1978/79), 3 – Allonnes 05 3008, 4 – Gournay 2351/2353 (d’apr. Lejars 1994), 5 – Manching (d’apr. Jacobi 1982), 6 – Waldalgesheim (frette, d’apr. Frey 1995a), 7 – Bern-Schosshalde (fibule, d’apr. Frey 1995a), 8 – Canosa (rinceau casque, d’apr. Jacobsthal 1944), 9 – Szob (détail frette, d’apr. Szabo and Petres 1992), 10 – Sopron-Bécsidomb (céramique, d’apr. Szabo and Petres 1992), 11 – Gourgançon, La Corbillère t. 7 (détail torque, d’après Kruta Poppi 1999), 12 – Steinenbronn (détail stèle, d’apr. Jacobsthal 1944), 13 – Voivodine (détail fourreau, d’apr. Szabo and Petres 1992), 14 – Stichill (détail troque, d’apr. Megaw and Megaw 1989), 15 – rinceau à triscèle de type A1/A2 (d’apr. Verger 1987), 17 – Drna (détail fourreau, d’apr. Megaw 1973), 18 – bronzes de «Paris» (détail garniture de char, d’apr. Jacobsthal 1944), 19 – Brno (applique, d’apr. Duval 1977), 20 – Brno (symétrie par rotation), (dessins, Th. Lejars del./Cnrs).

les uns comme les autres, autour d’une esse centrale, anguleuse pour les uns, nettement curviligne pour les autres. L’estampille de Manching a été rapprochée du dessin d’un poinçon imprimé sur une céramique de Sopron-Bécsidomb, en Hongrie (Fig. 28.9.10) (Megaw and Megaw 1989, 119; Szabo and Petres 1992, 23). Le tracé curviligne du rinceau figurant sur ce dernier se signale en outre par la présence, aux points de jonction entre l’esse et ses appendices latéraux, de deux excroissances triangulaires qui conservent le souvenir de rinceaux complémentaires et évoquent le schéma de construction des rinceaux à triscèle de type A1/ A2 (Fig. 28.9.15) (Verger 1987, 288). C’est une estampille semblable, avec méandre curviligne, que l’on retrouve sur les bandeaux obliques du fourreau de Gournay 2676. Sur cet exemplaire, la palmette fait place à une pelte (il manque la branche centrale de la palmette) qui n’est pas estampée mais ciselée sur les bossettes de la plaque rapportée placée sous l’entrée, tandis que des ocelles simples occupent les vides (Fig. 28.5.5) (Lejars 1994, 74, 222 – dessin in extenso de l’épée et de son fourreau. Groupées par trois, les peltes dessinent deux triscèles imbriqués de direction contraire). Les demi-palmettes opposées des estampilles de Potypuszta et Graz, avec leurs motifs en amande, les volutes terminales et les hachures, ont été rapprochées par J. V. S. Megaw de certaines figures animalières caractéristiques du Style plastique, tels les bronzes de «Paris» et la fibule de

Conflans, ou encore celles ciselées sur plusieurs fourreaux d’épée contemporains, comme Drna (Fig. 28.9.17–18) et Cernon-sur-Coole (Megaw 1978, 108). On peut ajouter à ces quelques exemples les visages grotesques des appliques de la cruche de Brno (Fig. 28.9.19) et de l’agrafe de ceinture de Loisy-sur-Marne, avec leurs yeux en amande et leur nez piriforme proéminent (Brno, Duval 1977, 131; Kruta and Bertuzzi 2007. Loisy-sur-Marne, Roualet and Charpy 1991, 265). Que l’image très stylisée des estampilles évoque un masque ne saurait surprendre, eut égard à l’importance du visage (souvent figuré de face, parfois très expressif, parfois tout juste esquissé) et des yeux dans la tradition iconographique laténienne (Fig. 28.9.16–20). L’association d’un masque et d’une palmette à trois branches incluses dans un segment de cercle (avec des hachures dans les écoinçons séparant les pétales) n’est pas rare comme en témoigne les garnitures en bronze ajourées de Cizkovice datées de La Tène B1 (Kruta 1976, 121, fig. 6). L’effet de distorsion induite de la symétrie par rotation n’est pas non plus sans évoquer les contorsions et grimaces qui déforment le visage de figures légendaires, tel Cuchulainn, un des héros de la poésie épique irlandaise, Táin Bó Cúailnge (Guyonvarc’h 1994). Avec ces fourreaux le motif n’est pas unique, mais soigneusement dupliqué de manière à couvrir uniformément la totalité de la surface. On connaît les vertus apotropaïques de l’œil toujours en éveil et du regard qui

252

Thierry Lejars

méduse et terrifie. Probablement, la répétition de ces «paires d’yeux» ne vise qu’à en affirmer et augmenter l’efficacité: à titre de comparaison, on peut rappeler dans la mythologie grecque, la figure du géant Argos, le fidèle gardien d’Héra, qui veillaient en permanence. Héra rendit hommage à sa fidélité en plaçant ses cent yeux sur la queue de son oiseau préféré, le paon (Ovide Metamorphoses I, 625).

Une production multipolaire Cette série d’armes, que l’on peut situer dans cette phase transitoire marquant le passage de La Tène ancienne à La Tène moyenne, autrement dit dans le courant de la première moitié du IIIe s. av. J.-C., se signale par d’évidentes similitudes tant morphologiques qu’ornementales. Pour autant, ces fourreaux d’épée sont loin d’être parfaitement identiques comme l’indiquent les techniques mises en œuvre pour le décor, le choix des estampilles et leur agencement. La composition générale peut prendre la forme d’un entrelacs anguleux, plus rarement curviligne (Maubranches) ou d’un maillage oblique dense et uniforme. Les estampilles losangées sont agencées de manière à matérialiser l’entrelacs. A Allonnes, les estampilles sont directement appliquées sur l’entrelacs réalisé au repoussé. Les ocelles et palmettes sont utilisés pour meubler les plages circonscrites par le réseau. Lorsqu’il s’agit d’un maillage uniforme, une estampille unique, de forme losangée, est employée, tandis que les bandes obliques, avec ou sans relief, demeurent parfaitement lisse. Lorsque des objets présentent de fortes ressemblances comme les fourreaux de Graz et Potypuzsta, le dessin des estampilles losangées ne sont pas identiques mais le miroir l’un de l’autre. Ces quelques exemples distribués sur un vaste espace géographique (depuis la Slovaquie jusque dans le Sud et l’Ouest de la France) suffisent pour montrer, à travers les choix opérés, la diversité des solutions adoptées, et ce faisant nous amène à voir là le résultat d’une production extrêmement diversifiée (Fig. 28.10). On conçoit une image tout à la fois unitaire et multiforme de la genèse de ces décors dans la mesure où ils puisent dans un répertoire ancien, adapté et actualisé simultanément, en différents lieux (conclusions similaires pour la genèse des rinceaux à triscèles, Verger 1987, 333). Cette diversité rend assez improbable l’attribution de ces armes à un unique centre de production, qu’il s’agisse d’un atelier ou même d’une région (sauf peut-être dans le cas des armes de Graz et Potypuzsta). Considérant la brièveté de cette mode ornementale, il faut croire que ces armuriers spécialisés ont cherché à reproduire librement des modèles qui circulaient rapidement. Cette mode, de courte durée, ne saurait être comparée à celle du chagrinage qui se développe plus tard, à partir de la fin du IIIe siècle (De Navarro 1972; Lejars 2013). La diversité des centres de productions est une hypothèse qui s’accorde bien, il nous

semble, avec le contexte culturel de cette période marquée par le foisonnement des expériences et le renouveau des panoplies militaires – avec l’introduction des systèmes de suspension d’épée semi-rigides et des premiers umbos de bouclier monocoques – et de leur décoration. Du point de vue ornemental, cette période qui voit le plein épanouissement du Styles des épées, avec certaines réalisations des plus remarquables (les fourreaux de Cernon-sur-Coole, de Flavigny et d’Halimba) mais aussi les débuts du Style plastique, est également marquée par le développement de thèmes comme la paire d’animaux fantastiques (lyres zoomorphes et dragons) et de techniques peu employées jusque là comme l’adjonction d’éléments rapportés (Lejars 2003, 24–29; 2012). Cette situation n’a probablement rien d’original, même si les études portant sur l’art des Celtes anciens sont encore fortement marquées par l’idée de productions centralisées, liées à des ateliers ou centres régionaux, diffusant sur de vastes territoires. Probablement, ce schéma interprétatif doit-il son succès aux travaux réalisés sur quelques productions quasi-industrielles du monde antique comme les vases grecs peints, les amphores vinaires romaines ou même certains biens artistiques, renseignées par une très riche documentation archéologique et des sources textuelles. Dans le cas des productions artistiques laténiennes, nous n’avons rien de tel. Pour chaque catégorie de décor, nous disposons le plus souvent de séries limitées, parfois quelques unités seulement, disséminés sur une bonne partie du monde celtique. La mise en évidence d’ateliers de fabrication d’armes au sein d’habitat ne garantit pas l’unicité de la production. A signaler l’atelier de fabrication d’armes identifié par J.-P. Guillaumet à Sajopetri en Hongrie (Szabo and Czajlik 2007). L’hypothèse d’une estampille commune pour les fourreaux d’épée d’Epiais-Rhus (lequel compte six estampes distinctes) et Moscano di Fabriano (pour lequel on dénombre trois estampes seulement) a été avancée par A. Rapin (2006a, 194), qui conclut à l’emploi d’un même outil (peut-être un même artisan), mais renonce à préciser le lieu et la date de la conception. Pour la transition HallstattLa Tène, on peut également mentionner les différents lieux de production de fibules à timbales mis en évidence ces dernières années sur plusieurs gisements du Centre-Est de la France (Bourges, Lyon, Plombières-les-Dijon; Carrara et al. 2013). La primauté d’une découverte et la bonne conservation des vestiges ne sont certainement pas des arguments suffisants pour indiquer l’origine d’une forme, d’une mode ou d’un style. La concentration d’une série en une région donnée ne fournit pas davantage la preuve d’une origine locale en particulier lorsqu’il s’agit de contextes funéraires dans la mesure où nous savons que les coutumes diffèrent fortement d’un groupe à l’autre, d’une région à l’autre, ainsi que d’une époque à l’autre pour une même région. L’exemple des fourreaux d’épée à décor estampés est à cet égard très éclairant; en extrême occident, ces objets,

28.  A l’aube du IIIe s. av. J.-C.: les fourreaux d’épée à décor estampé sur fer

253

Fig. 28.10. Distribution des fourreaux d’épée à décor estampé sur fer. 1 – Allonnes, 2 – Aubigné-Racan, 3 – Gournay-sur-Aronde, 4 – Rungis, 5 – Maubranches, 6 – Ensérune, 7 – Manching «Stainbichel», 8 – Graz, 9 – Potypuzsta, 10 – Malé Kosihy, 11 – Saliceta san Giuliano.

absents des contextes funéraires, nous sont, pour le moment, uniquement connus à travers les débris d’armes conservés dans les lieux de culte. Si certaines armes proviennent de butins de guerre, d’autres sont des offrandes des membres de la communauté. Là encore, il n’est pas simple de distinguer le local de «l’étranger». Cette difficulté est aussi la conséquence de la forte homogénéité qui caractérise l’armement celtique pendant la période laténienne et de son adoption par les populations des aires culturelles limitrophes. Le décor du fourreau d’Allonnes démontre l’importance de ces fragments, soigneusement collectés mais également souvent délaissés du fait de la corrosion. Seule la nécessité de procéder à un examen technique détaillé, comme l’appelait de ses vœux O. H. Frey en conclusion de la note consacrée à l’épée de Graz, permet de dépasser l’handicap de la corrosion, d’approfondir notre connaissance de l’artisanat du fer, d’enrichir un corpus ornemental encore extrêmement réduit et d’ouvrir sur de nouvelles perspectives quant à l’art celtique.

At the Dawn of the 3rd Century BC: stampdecorated iron sword scabbards If the only evidence to understand the Celts of the La Tène and their artistic output came from grave goods,

whole regions would still be obscure to us. If we had to rely on items in perfect condition, because fire or water bias preservation, the number of useful artifacts would be negligible. We know from research that funeral customs varied widely, not only during the five centuries of the second Iron Age, but also between regions, demonstrating that there was no specific Celtic mortuary ritual. Research has also taught us that the rust which encrusts the great majority of ancient iron, also hides many surprises. The procedure to remove the corroded casing from the object is as old as the discipline of archaeology itself. For scholars of the 19th century, as for us today, iron corrosion has always been a problem. In 1865, A. Verchère de Reffye (1865a; 1865b) noted bitterly: ‘at the beginning of this century, this poor iron was so little valued that no one picked it up, and no collection deigned to give it space, so much so that it resulted in a belief, still held by many, that people in the past did not use iron’. We can get a fairly accurate idea of the methods used at this time in Mainz and Saint-Germain-en-Laye, through the note he published in the Revue Archéologique in 1865. The article, intended to educate and inform archaeologists, is reprinted in the columns of Matériaux pour l’histoire positive et philosophique de l’homme. It constitutes a clear plea for iron, ‘civilization’s most significant agent’. In this area as in many others, the new Museum of Saint-Germain-enLaye is inspired by Mainz, the obligatory reference for all

254

Thierry Lejars

that relates to the study and conservation of antiquities. The recommended method, applied to the artefacts from Alesia, aims to recover: ‘the original forms of the artefact, the markings and inscriptions they carry’. Between 1975 and 1984, the discovery of a considerable number of heavily corroded iron weapons and fibulae in Gournay-sur-Aronde, led archaeologists to develop the means to make the most of this new kind of archaeological record. Use of more systematic restoration from the 1980s eventually bore fruit. Major advances in terms of our understanding of Celtic weaponry have been made possible by the complete or partial restoration of ironwork from Gournay (Brunaux and Rapin 1988; Lejars 1994), and has enabled, to a considerable extent, a significant increase in the corpus of decorated weapons, whose distribution has long been confined to central European areas (Hungary, Austria and Switzerland) and the British Isles (Jacobsthal 1944; Duval 1977; Megaw and Megaw 1989; Jope 2000; Stead 2006). In 1977, P.-M. Duval was able to write – this was the widespread view – that ornamented scabbards ‘developed particularly in Hungary and from there reached Switzerland, then moved through Gaul, the Atlantic islands, and via Romania to Yugoslavia’ (Duval 1977, 120). This perception, which has long prevailed, is linked to the idea of an expanding Celtic world, driven by constant migrations, and a concept of craftsmanship modelled on the ‘classical’ art historians’ approach. In France, in the same period, the systematic research conducted by A. Bulard formed the basis of a corpus of decorated weaponry (Bulard 1979; 1982). Here again, finds from waterlogged conditions make up the bulk of the corpus. The inventory published ten years later by N. Ginoux multiplied the initial corpus by five, using the first results of a fully-developed policy of restoration (Ginoux 1994; 1995) (the catalogue mentions 79 decorated weapons for the 5th – 2nd century BC). At the same time, Otto Herman Frey listed 144 representations of dragons pairs across the whole of Celtic Europe when, two decades earlier, J. De Navarro could only identify about 30 (Frey 1995b; De Navarro 1972, 66). At Gournay-sur-Aronde, 65 of the 111 entries of cleaned scabbards had retained traces of decoration, but it is very likely that this figure was originally much higher: often decoration survived only as traces, and some must have disappeared altogether. At La Tène, about 70% of scabbards were decorated; amongst the remaining 30%, many are badly corroded or damaged (Lejars 1994, 63). In this context, the sanctuary of Gournay-sur-Aronde is not an isolated case. Since its discovery, numbers of cult sites, some even very spectacular as at Ribemont-sur-Ancre or Tintignac, have been identified and excavated, yielding hundreds, even thousands, of iron objects. While some are relatively well preserved and complete, many are reduced to a fragmentary state. However, these fragmentary artefacts are far from uninteresting. The work between 1994 and

2011 on ‘La Tour aux Fées’, the great Roman sanctuary at Allonnes, a few kilometres south of the town of Le Mans, revealed the existence of pre-Roman occupation levels and above all a significant assemblage of La Tène metalwork, with, in particular, numerous fragments of decorated sword scabbards – zoomorphic lyres, pairs of dragons, ornamentation in continuous Vegetal Style, stamped decorations, etc. (preliminary publication in Brouquier-Reddé and Gruel 2004: I am pleased to thank K. Gruel and V. Brouquier-Reddé for entrusting me with the study of the Gallic metal excavated on this remarkable site). In this paper, we will focus on the case of a small decorated fragment which complements the limited catalogue of stamp-decorated weapons spread across Celtic Europe from the beginning of the middle La Tène, and splendidly illustrated by the Potypuszta scabbard from Hungary and, more recently, by two pieces from Gournay-sur-Aronde, to which we shall return.

La Tour aux Fées at Allonnes (Sarthe) During the course of excavations at the Gallic and Roman sanctuary of La Tour aux Fées, Allonnes, a quantity of weaponry and some dress accessories were discovered, which indicate regular activity at the shrine – although these objects were almost always residual deposits and the proto-historic development of the site is difficult to date – between the 4th century to the Romanization and monumentalization of cultic structures. If certain items of late weaponry date from the late 2nd and 1st centuries BC – that is, the Augustan period – the majority of them can be assigned to the Early La Tène and to the beginning of the Middle La Tène (4th and 3rd centuries BC) (ibid., 296–306 for the Gallic era material). The scabbard fragment, inventory number Al 05 3008, was discovered in July 2001 in the area of the west portico, between walls 11 and 14. It is 65 mm long by 47 mm wide (Fig. 28.1). The edge, preserved only on one side, is marked by a deep rounded convex overlap 5 mm wide. The fragment, like most of the protohistoric ironwork found on the site, was covered with a thick sandy clay matrix which masked the surface. It is only through restoration that the shape and decorative detail of the object are recognizable. The residual nature of the piece is also confirmed by the fact that no conjoining fragment has been identified so far. It seems that the stamped decoration covers the entire front plate with the exception of the overlap. The decoration appears as an oblique repoussé band in relief (descending) to which a second perpendicular strip (ascending) has been applied. The beginning of a third (descending) is visible at the bottom, on the left. The curved bands are enhanced by a series of quadrangular stamped cartouches, each decorated, as far as we can judge, with an

28.  A l’aube du IIIe s. av. J.-C.: les fourreaux d’épée à décor estampé sur fer asymmetric motif organized around an oblique ‘S’ (with comma shapes arranged head to tail on either side of the oblique ‘S’). The axial diamond-shaped space defined by the bands is decorated by a central cartouche in low relief (the fragments of the surviving decoration indicate a motif similar to that incised on the bands) while palmettes (situated in a segment of a circle whose ends spiral inwardly to form the lower leaves of a palmette with three leaves) adorned with a dot are placed on the sides of the cartouche, greatly reducing the undecorated space. The lateral triangular (or half-lozenge) shaped field is itself decorated with a single palmette abutting the overlap and decorated with a dot, like the others. This complex composition, playing on the reliefs which structure the space geometrically, and which animate it (restored decoration, cf. Fig. 18.2.3), has obvious similarities to weapons with stamped decoration from the beginning of the Middle La Tène, a subject that has received, over the last 35 years now, the attention of some of our illustrious predecessors, especially O. H. Frey who published in 1978 a short but important study on the Graz scabbard, and John Vincent S. Megaw, who returned to this argument in a note published in the same year about the decoration on the scabbard from grave 115 at Jenisuv Ujezd (Frey 1978/79; Megaw 1978). However, having remembered what we owe to these authors, it is worth pausing for a moment on the Potypuszta scabbard mentioned above. Published simultaneously by Ilona Hunyady (Hunyady 1942/1944, 112, pl. XLII.7, XLV.3) and Paul Jacobsthal (Jacobsthal 1944, no. 127, 177), and often mentioned subsequently in the archaeological literature, the weapon was studied more recently by Eva F. Petres and Miklos Szabo on the occasion of the 1992 publication of the corpus of decorated weaponry from the Carpathian basin (Szabo and Petres, 1992, no. 50, 21–23, 96, pl. 54, fig. I). Jacobsthal attributed the find to Velem-Szentvid and Szombathely (the name of the museum holding the artefact) and speaks of a scabbard made of bronze, an error caused by the very good state of preservation of the metal. For P.-M. Duval (1977, 255), it is still referred to as a bronze scabbard, with an excellent photograph of the middle portion of the scabbard (ibid., fig. 299). It is likely that this weapon, without known context, but complete and in a quite remarkable state of preservation, comes from a cremation. The scabbard measures 69.5 cm in length (74.8 cm with the sword) and is 5.4 cm wide. Unlike the Allonnes fragment, the scabbard has a high midrib like the Potypuszta example. The overlaps are large and prominent. The suspension loop, arched, is attached to the scabbard by two rounded loop plates. The chape, about 15 cm long, ends in a rounded extremity, without any visible openwork. The decoration, organized in oblique bands, ascending and descending, is not ribbed like the Allonnes scabbard, but stamped with a series of diamond-shaped patterns, each decorated with two opposing half-palmettes

255

(symmetrical by rotation of the almond motifs dressed with an ‘S’ flanked by hatching) (reconstructed decoration, fig. 28.2.1 cf.). The decoration is highlighted near the mouth of the juxtaposition of eight stamps arranged in two horizontal rows. The axial fields, also lozenge-shaped, are furnished in the centre, like Allonnes, with a lozenge stamp bordered along the sides with four palmettes (included within a segment of a circle) alternating with as many circle-anddot motifs. As for the triangular (or half-diamond) lateral fields, they are decorated with a palmette identical to the previous ones, abutting the overlap and flanked by three circle-and-dot motifs. Stamps cover the overlaps and the raised midrib, which implies that the decoration was done before the sheet of metal was shaped. Organization of stamps in oblique cross-bands is indicative of a small group of decorated weapons which we focus on here to demonstrate, in spite of a very uneven state of preservation, and beyond the strong similarities in shape and decoration, the variety of solutions adopted by armourers.

Graz, Manching, Gournay-sur-Aronde It is necessary, first, to examine the ‘Laubgasse’ scabbard from Graz, also without a known provenance. The sword and its scabbard were published by O. H. Frey in 1978/79 (Frey 1978 to 1979, 67: the sword was found with a Middle La Tène brooch, not kept, in a gravel pit), and again by M. Szabo and E. F. Petres in 1992 (Szabo and Petres 1992, no. 76, 102, pl. 79). Based on the work of U. Osterhaus (1966) and J. M. De Navarro (1972), Frey proposes allocating the Graz sword to the end of the Early La Tène or to the beginning of the following period. He also refers to finds from Central Europe and Italy (Montefortino and Ceretolo). Less than a hundred kilometres separate Graz and Szombathely (Potypuszta). The weapon, although complete, is badly corroded. The upper part of the very distorted scabbard is incomplete, while the decoration which originally covered the entire front panel is only partially visible, in the upper part and near the chape. The sword (with scabbard) measures 74.8 cm long and 4.4 cm wide and 110 mm with the tang. The length of the scabbard is about 62/63 cm and a width of 5.4 cm. The latter is further distinguished by lateral overlaps and a broad central ridge. The squat suspension loop has concave edges and two large rounded fastening plates. The chape, about 15 cm long, is distinguished for its part by a rounded end (corrosion prevents the identification of openwork). The stamped decoration once again covers the midrib (reconstruction of the decoration, Fig. 28.2.2). It consists of juxtaposed diamond-shaped patterns organized in oblique perpendicular rows. As in the previous examples, the diamond contains a pattern composed of two opposed half-palmettes with hatching. Despite the piece’s poor state of preservation, two other stamps are identifiable, a palmette (half-circle with volute terminals), and a circle-and-

256

Thierry Lejars

dot, used to decorate the large axial diamond and the lateral triangular fields (see our proposed reconstruction from the summary published by O. H. Frey). Finally, we should stress the similarity, in some details identical (the patterns within the diamond are mirrored), of the stamps adorning the Graz and Potypuszta scabbards (Frey 1978/79, 68). The second example is provided by the scabbard from grave 27 from the Steinbichel cemetery in Manching, Bavaria. The context of discovery is important because of the extent to which it allows us to specify, for the first time, the typo-chronological context of this style of scabbard (see below). The sword, 75.5 cm long (the reconstructed length of the scabbard is about 66 cm by 5.5 cm wide) is sheathed in a scabbard with an openwork circular chape (the length of the chape is not determinable) (Kramer 1985, 83, pl. 16). The squat suspension fitting is secured by two large rounded loop plates. The scabbard, not ribbed, is entirely covered with diamond-shaped stamps with concave edges (the pattern was first published by Jacobi in 1982, 567, fig 2. See Szabo and Petres 1992, 22–23, fig 6.). Unlike the previous examples, the distribution is uniform and without emphasis. The internal pattern also differs. It consists of a curvilinear foliage with dots placed in the loops and on the edges of the stamp (Fig. 28.2.5). The third example comes from the sanctuary of Gournaysur-Oise in Aronde, which produced several scabbards, complete and fragmentary, with stamped decoration. These weapons, four in number, are dated, by analogy – morphological and technical – with the findings in a funerary context from the beginning of La Tène C1. For the moment, only scabbards GSA 2351/2353 and GSA 1116/1122 are our focus. The two objects are incomplete: the first is missing its chape, while only the front plate of the second survives. Like the previous examples, they are fitted with wide lateral overlaps, but lack the midrib. In both cases the ornamentation consists of a regular oblique lattice in low relief. It overflows the overlaps, which suggests that it was made before the edges were formed. The mouth of GSA 2351/2353 is highlighted by two horizontal bands made in repoussé. The latter’s rhomboid stamps, about 1 cm square, are enriched with a pattern composed of two symmetrical portions (rotational symmetry): a horizontal leaf with a vertical tendril terminating in a returned curve (Fig. 28.2.4 and 28.3). The decoration of the second scabbard, GSA 1116/1122, is equally complex. The observations made during cleaning allow us to reconstruct, despite the poor state of the surfaces, an internal pattern in the form of curvilinear foliage analogous with that exhibited on the Manching scabbard (Fig. 28.2.5 and 28.3).

The chronology If the Allonnes fragment is too incomplete to offer a

specific chronological point, the weapons we have examined present undeniable similarities in terms of their size and morphology. This is most apparent in their dimensions, with lengths between 62 and 67 cm and widths between 5.2 and 5.5 cm (Fig. 28.4). Suspension fittings consist of a squat loop flanked by two large rounded loop plates. Chapes, 15 to 16 cm long, are marked by a solid circular or rounded end, openwork is virtually nonexistent. Overlaps are large and prominent, like the midrib where it survives. Unfortunately, the context of their discovery is far from being as obvious. Only the Manching scabbard has this advantage. That assemblage consists of, in addition to the sword, a spearhead with biconvex blade, a suspension chain with double twisted and enlarged rings, and a lignite armring. If the rounded, openwork chape of the scabbard reflects Early La Tène traditions in weaponry, the morphology of the suspension loop and the suspension system of the sword allows the assemblage to be placed in a late La Tène B2 horizon or at the beginning of La Tène C1. This is the dating we have already proposed for sword scabbards with solid rounded chape ends present in the Gournay-sur-Aronde sanctuary (Lejars 1994, 55–56). The comparison of these morphological criteria with their parallels in funerary contexts has equally allowed us to make the chronological link between this form of scabbard to the latest examples equipped with openwork rounded chapes. To complete our discussion and support our demonstration, we can mention the furnishings of grave 1 from Moulinsur-Yèvre at Maubranches (Cher). The scabbard, 61 cm long and 5.5 cm wide, was associated with a belt chain with double twisted links, of the kind that accompanies the scabbard from Cernon-sur-Coole, a spearhead and shield boss with rectangular flanges (Ginoux 1994, 74, no. 44 with bibliography, pl. XI; Kruta Poppi 1984, 58–59, fig. 6, with photographs of the decoration). The sword and the associated grave goods are attributed to the Middle La Tène. In the catalogue of the L’Art celtique en Gaule exhibition (1983/84 177, no. 214), an excessively late dating is proposed for the Maubranches sword (Final La Tène), even though Déchelette (1914, 1119, fig. 463.3) had already placed this sword alongside Middle La Tène weaponry. On the front panel it features relief decoration, made in repoussé, which takes the form of an interlace pattern (Fig. 28.5.2). It seems that punched circles, arranged in the gaps, completed the decoration. Traces of ornamentation are also visible on the chape, solid and rounded, 15 cm long. As in the previous weapons, the overlaps are wide and folded over the back, and the suspension loop is relatively short, fixed with two round loop plates. The ornamentation of this weapon is not unlike that of another scabbard from northern Italy, discovered in 1883 at Saliceta San Giuliano, near Modena (Kruta Poppi 1978, 427; 1983, 22). The first finds date back to 1876, especially grave goods from a female burial. The remains collected in 1883 derive from several

28.  A l’aube du IIIe s. av. J.-C.: les fourreaux d’épée à décor estampé sur fer assemblages which certainly date to the earlier Etruscan phase (Certosa type fibula), others to the Gallic and Roman period. In addition to the equipment of a warrior, we can list the remains of one or more female assemblages with not less than six bronze bracelets. Based on the photographs of A. Crespellani, the warrior equipment consisted of a sword with a metal belt made of triple-twisted links (analogous to the chain belt from Cernon-sur-Coole) and a spear (Kruta Poppi 1978, 429, fig. 4. Drawings after photographs by Crespellani 1887). In 1984, L. Kruta Poppi (1984, 53) suggested the possible presence of a double ring equipped with point (apparently a piece of horse harness, except that they are never found in Cisalpine Celtic burials) and two fragments of an iron knife blade or razor (Bergonzi 1988 157, fig. 108). She proposes that the preserved part of the sword – about 40 cms in length – represents the lower two-thirds of the weapon (Kruta Poppi 1984, 53). As for the scabbard, only unconnected fragments of the front plate survive (ibid., 52–53, the author indicates that only one of the three fragments was restored. The decoration is not recorded in the 1978 publication). Only the triangular tip of the chape remains. The decoration, in relief, is formed by an angular interlace which runs between lateral overlaps and marks out a series of lozenges in the centre and triangles at the sides. If there were other, more delicate, motifs, no trace survives because of the corrosion. The two wide perpendicular corrugations which terminate the surviving portion of the scabbard marked, according to L. Kruta Poppi, the location of the transverse bridge of the chape (ibid., 53: the author notes a slight narrowing of this part). In the catalogue of the exhibition devoted to the history of Modena, published in 1988 (Bergonzi 1988, 157, fig. 108: photograph of the sword with the base of the tang in Kruta 1988, fig. 231), there is a substantially different drawing of the weapon, with a fragment corresponding to the base of the tang placed just above the horizontal bands. This last representation seems more consistent with reality to the extent that the bands – two broad bands in relief and one much narrower at the centre – are never found in the section of the scabbard marking the location of the opening of the chape. One can also question the terminal fragment belonging to this chape ... a rounded shape would have been more appropriate. For these reasons, and given the weapons described above, I think it best to see the horizontal bands as indicating the mouth of the chape (reconstruction, Fig. 28.5.1). G. Bergonzi also indicates the presence of dots in the lateral triangles of the interlace. To close this chronological excursus, we can report on the grave goods from the burials of two warriors equipped with scabbards with relief decoration, although of a different nature to the examples discussed so far. The first is illustrated by the tomb of the warrior from Rungis whose sword was furnished with a ‘ladder’ type chain belt, a spearhead, a shield plate with narrow ribbed boss and circular flanges,

257

and an iron fibula with a open foot and twisted bow (Kruta and Rapin 1987; Ginoux 1994, 84, no. 74, pl. XIX). Of the scabbard, only the upper two thirds survive. The suspension loop is fixed by two large rounded loop plates. The ornamentation covers the whole surface following two successive schemes: the first is distinguishable as 16 repoussé dots accented with fine engraving which divides the plate into oblique fields, and a second, engraved (barely visible), featuring a hatched lyre beneath the opening (for the decoration, see Ginoux 1994). The second example is provided by the grave goods from cremation 31 from Male Kosihy, Slovakia (Bujna 1995b, 24–25, pl. 6. – the grave goods are very fragmented). The assemblage consists of the sword, a ‘ladder’ type suspension chain, various fittings related in part to a shield and the rest to two iron open foot fibulae. The scabbard, without a midrib, is partially preserved. The chape is missing, but the suspension loop has large loop plates (Fig. 28.5.4). The decoration of the scabbard, reconstructed, consists of pairs of parallel lines made in repoussé. They are arranged, first, horizontally beneath the mouth, and second, following an oblique axis running down the rest of the scabbard. A decorative grid of a concentric diamond-shaped stamped pattern fills the spaces between the parallel lines (Bujna 1995a, 260: scabbard width 58 mm). If this grid stamped on the internal spaces seems to attest to a reuse of the plate, as J. Bujna suggests, it is surprising to see the lozenges perfectly inserted between the parallel lines which must have been made (as on the reverse) at a later date. Rather than seeing the result of two successive stages, one may wonder if the small diamond-shaped stamps (about 5 mm square compared to 1 cm square for the stamps on the preceding scabbards) were not deliberately printed on the back of the plate as can be seen with the ‘S’ stamps on the earliest scabbard, also iron, from Sogny-aux-Moulins (Marne) (Fig. 28.6). The recurrent links between these scabbards with early forms of metal belts (links with double or triple twists, ‘ladder’ links) and shield bosses (single shell bosses with long or semi-circular flanges with ribbed ‘umbos’), but also the late brooches with open feet, confirm the dating of these weapons to the early/middle La Tène transition – put forward by O. H. Frey a little over thirty years ago now (Frey 1978/79, 68–69).

The organization of the ornamentation: iron and bronze The common features of the decoration in this study are the technique of iron stamping with the use of elaborate punched dots, whether or not combined with sections executed in repoussé, and a strong propensity to cover the entire front plate. Of course, there are many known examples of sword scabbards decorated in their entirety, with decorations that

258

Thierry Lejars

may cover not only both sides of the scabbard, but also the suspension loop and plates and the chape, such as Cernonsur-Coole (Duval and Kruta 1986). Staying with the range of stamped or repoussé decorations, we note the famous early La Tène examples from Filottrano, Moscano di Fabriano, Epiais Rhus, St. Germainmont, and Jenisuv Ujezd, which have been the subject of numerous of studies (Frey 1971; Megaw 1978; Kruta et al. 1984; Rapin 2006a; 2008). It is about combination scabbards, combining iron and bronze. The stamps, often developed (foliage in continuous Vegetal Style, peltas, ‘S’s, and circles) are printed on the front plates in more malleable bronze. Examples of stamped decoration on iron plates remain relatively exceptional throughout this period, probably due to the difficulty of making sheets as thin as those in bronze (Rapin 1999, 42). Unlike stamps on bronze, those for iron remain much more basic throughout the early La Tène (5th and 4th centuries BC): simple punched dots forming curvilinear or geometric figures as at Prosnes (Battaille-Melkon and Rapin 1997) and Cortrat (Rapin and Zurfluh 1998); circles and ‘S’s – pairs of ‘S’s arranged in four-part whirligigs – as on the example from Sogny-auxMoulins (Fig. 28.6). In the Early La Tène, stamps do not cover the entire front plate, but are organized in horizontal or vertical bands alternating with completely empty spaces (engraved with a chisel, the decoration of the Méroux scabbard is very similar (Megaw 1968; Ginoux 1994, 70, pl. VI.1)). If the repoussé technique is frequently used on bronze sheets to create relief patterns, iron surfaces were animated by incised designs – for example, Monte Bibele, grave 126 (Rapin, Szabo and Vitali 1992), Flavigny (Rapin 2006a, 201; 2006b), the chape and suspension loop from Cernon-sur-Coole (Duval and Kruta 1986) – or by adding inserts, as Gournay-sur-Aronde (Gournay 1826 1860/2651, 1863, 2676; Lejars 1994, 64–73) Drna (Megaw 1973; Szabo and Petres 1992, 103–104 no. 82, pl. 84–85), etc. To be accurate, creating a pattern from intersecting oblique (or interlaced) lines is not itself a novelty. Even if the meander-hooks framing the borders of the final scene on the Hallstatt scabbard (Jacobsthal 1944, pl. 60, no. 96; Egg and Schoenfelder 2009) do not have a structural role, it is a different case with the Vert-la-Gravelle example (Jacobsthal 1944, 92, pl. 56, no. 90, pl. 268 (216); Ginoux 1994, 71, no. 36, pl. VI). There, an angular interlace covers the entire plate, with a swastika inside each central lozenge and concentric circles within the edging triangles. This pattern is also attested on torcs and some Duchcovtype fibulae, in the form of crossed oblique parallel or zigzag stripes (Kruta 1979, 109). Particularly important is the decoration on one of the brooches from Repin, Bohemia, with its criss-cross, hatched bands enclosing triple concentric-circles inside each rhombus (ibid., 110) (Fig. 28.7.5). On Middle La Tène scabbards, the decorative scheme sometimes develops following a relatively dense and completely uniform pattern (Manching amd Gournay

1116/1122, 2351/2353, Malé Kosihy) or more loosely as interlace (Potypuszta Graz, Allonnes, as well as Saliceta San Giuliano and Maubranches). In the same way, a second fragment of scabbard discovered at Allonnes is characterised by networks of small diamond-shaped stamps distributed in a very regular fashion (Lejars 2003, 26, fig. 15.3), as well as another from Ensérune (M.006, unpublished). The motif appears as a concentric circle inscribed in a lozenge. On the basis of these examples, the pattern scheme can be expressed uniquely by the alignment of lozenge-shaped stamps, as at Potypuszta and Graz, or by relief decoration to create proper ribbons (Allonnes, Gournay 2351/2353, Saliceta San Giuliano and Maubranche). Finally, the mouth is highlighted by two transverse bands (Potypuszta, Gournay 2351/2353, Saliceta San Giuliano, Malé Kosihy). As for the curvilinear interlace from Maubranches, it forms an interesting comparison to the Plastic decoration (pseudofiligree) on the bronze fibula from Juvigné, in Mayenne (Fig. 28.7.6) (Lejars in Santrot, Santrot and Meuret 1999, 121: the brooch has a rounded bow and triangular foot, with large globular beads, and has a six-spiral twisted spring. The raised foot turns back and is attached to the bow even though its morphology is still Early La Tène open foot). Where the interlace covers the bow, beads decorate the internal and external spaces.

Circles, palmettes and lozenges The number of stamps used is limited. The corpus consists of circles, palmettes inscribed in circle-segments, and lozenges. The motifs emblematic of continuous Vegetal Style from the previous period are no longer used (type A1 scrolled triskeles (terminology developed by St. Verger 1987) on scabbards from Moscano di Fabriano, Epiais Rhus, and St. Germainmont). Of all the motifs, the circle (single or concentric, tear-shaped or not) is the simplest. Its use is documented throughout the Early La Tène. It is found from the beginning of the period on dagger scabbards, at SaintJean-sur-Tourbe (Ginoux 1994, 65, no. 24, pl. II.5), and later on sword scabbards, for example Vert-la-Gravelle (Megaw 1978 107cf. footnote 48), Bussy-le-Château (Ginoux 1994, 68, no. 30, fig. 4, pl. VII.2), Filottrano, Moscano di Fabriano, Epiais-Rhus, Jenisuv Uzejd (Kruta et al. 1984), Drna (Szabo and Petres 1992, no. 81, pl. 83; Gournay 2186, Lejars 1994, 67) (Fig. 28.5.3). Circles are most often used to highlight contours, to mark the scabbard mouth and suspension loop (Megaw 1978, 107, with a relatively comprehensive overview of the use of this motif in Early La Tène metal and ceramic). Their use echoes leatherworking, as can be seen on the knife sheath from L’Eglise, Belgium, with geometric and floral designs executed in repoussé (CahenDelhaye 1981). Circles are sometimes used as supplements on representations of fantastic paired animals. In other

28.  A l’aube du IIIe s. av. J.-C.: les fourreaux d’épée à décor estampé sur fer cases, the motif is reduced to a simple dot organized into geometric compositions (iron scabbard from Cortrat, Rapin and Zurfluh 1998, grave 17, 57–59) or uniformly covering the entire surface (Allonnes and Aubigné-Racan, Lejars 2003, 26, fig. 15.4–5; 2007, 277). The circle, without ever completely disappearing, becomes rarer in as time goes on. More complex, the palmette is drawn in a semi-circle terminated by inward-turning volutes. A foliated pattern, teardrop-shaped, unfolds in the internal space, between the scrolls, to form a palmette with three branches. If this motif evokes, to a certain extent, the palmette stamps decorating black burnished vases and certain iron helmets with decorative bronze sheet appliques (Smarjeta/Sanct Margaraten, Slovenia), there are also analogies in the La Tène decorative repertoire, as already noted by O. H. Frey (Frey 1978/79, 69). Particular mention should be made of elements from the Besançon jug friezes, the Canosa helment, the gold torc of the Waldalgesheim tomb, the appliqué from Brunn, or of that from ‘Comacchio’ (Fig. 28.8.3–4) (Jacobsthal 1944, no. 143, pl. 83; Frey 1995a). On the Joressant spearhead from Haut-Vully (Fribourg, Switzerland), the decoration, in the Sword Style tradition, combines, in triangular fields arranged on either side of a diamond enhanced with a swastika, a flurry of curvilinear foliage combining curvilinear pelta and zoomorphic lyre (Megaw 1978, 108; Lejars 2012). Still present at the dawn of the Middle La Tène (Voivodine – Szabo and Petres 1992, n 127, pl. 119 – the pattern is placed between the two upper arms of the opposing lyres) (Fig. 28.8.5) the pattern gradually transforms into a lyre (droplet enclosed by two circles and the disappearance of the circular outline) or a loop, eg Bölcske-Madocsahegy, Bodrogahalom (ibid., Bölcske-Madocsahegy 1, no. 5, pl. 7 and Bodrogahalom, no. 3, pl. 4 – the motif embellishes the overall design). Despite these changes, the pattern nevertheless retains its centrality in the overall organization of the ornamentation (as an element of diagonal symmetry for compositions of the 4th century, or as part of an axial link for later examples). At BölcskeMadocsahegy, the pattern resulting from the merger of a pair of birds’ heads with curled beaks takes the form, when the figure is reversed, of a mask adorned with a double mistletoe leaf headdress (Szabo 1993; 2003). The last motif is undoubtedly the most complex and one that poses, a priori, the most problems. It comes in the form of a leaf enclosed in a quadrilangular cartouche turned onto its corner (as a diamond or rhombus). The internal design consists of two almond-shaped leaves finished with a volute ‘S’, flanked by hatching, or in other words two opposing halfpalmettes. The figure is obtained by rotational symmetry, a technique often used for Hungarian Style ornamentation (Bölcske-Madocsahegy) but also in numerous continuous Vegetal Style works; it is even one of the dynamic springs turning tendrils into triskeles (Filottrano, Moscano di Fabriano – cf. Frey 1974, 142 note 37: Kruta et al. 1984). As O. H. Frey noted, the Graz stamp is the almost exact mirror

259

of that from Potypuszta. O. H. Frey argued a derivation of the hatched motif (clearly visible on the Hungarian example) from the ‘hatched palmette’ found in the decoration of the Basse-Yutz flagon or the Comacchio appliqués, as well as in specific more recent works, especially the Sword Style (Frey 1978/79, 69, with reference to Jacobsthal 1944 no. 381 for Basse-Yutz, no. 113, for Cernon-sur-Coole and no. 126, for Mitricova, Slovenia). The Gournay and Allonnes stamps are obviously – as far as we can judge from their state of preservation – distinct from their predecessors. At Gournay, the terminal scrolls of the ‘S’s give way to large curls, and the hatching to half dots arranged on the periphery, as in the Manching scabbard. At Allonnes, the motif constrained by the cartouche inscribes a slightly oblique quadrilateral. The lateral ‘S’s are no longer perpendicular but attached to the central ‘S’, and the hatching has disappeared. The origin of the pattern is likely to be found in the in the Continuous Vegetal Style ‘S’ tendrils (‘S’s of the Schosshalde fibula or the Canosa helmet) (Fig. 28.9.6–8). Stamping the iron led to a need to simplify the design. At the same time, the ‘S’ extended at each end into a double volute, with, in some cases, a clear evocation of a fantastic bird’s head, appeared as medallions on some weapons, like the suspension loop of the Vojvodina scabbard, or again on the bosses on the reinforcing piece of a scabbard of Hatvan Boldog type, from Szob (Fig. 28.9.9) (Szabo and Petres 1992, 100, no. 68, pl. 71). Clearly, the fineness of the carving allowed refinements of detail that were not possible with the stamp technique. Inscribed in a diamond-shaped cartouche, the curvilinear meanders of Manching and Gournay 1116/1122 unfold in continuous symmetrical fashion on either side of the central ‘S’. Looking more closely, these motifs are not so different to the previous ones to the extent that they are all organized around a central ‘S’, angular for some, for others clearly curvilinear. The Manching stamp was close to the drawing of a pattern on a pot from Bécsidomb-Sopron, Hungary (Fig. 28.9.10) (Megaw and Megaw 1989, 119; Szabo and Petres 1992, 23). The curved line of the leaf on this latter is also interesting for the presence, at the junction points between the ‘S’ and the lateral arms, of two triangular extrusions which are the relic of additional leaves and which evoke the scheme of leaves constructed as triskeles, type A1/A2 (Fig. 28.9.15) (Verger 1987, 288). There is a similar stamp, with curvilinear meanders, found on the oblique bands on the Gournay 2676 scabbard. On this example, the palmette is replaced by a pelta (it is missing the central leaf of the palmette) which is not stamped but incised on the roundels of the insert plate beneath the mouth, while simple dots occupy the empty spaces (Fig. 28.5.5) (Lejars 1994, 74, 222 – drawing in extenso of the sword and scabbard). Grouped in threes, the peltae describe two opposing triskeles. J. V. S. Megaw linked the opposing half-palmettes on the Potypuszta and Graz stamps, with their almond-shaped

260

Thierry Lejars

motifs, their terminal scrolls, and hatching, to certain animal motifs characteristic of the Plastic Style, such as the ‘Paris’ bronzes and Conflans fibula, as well as those incised on several contemporary sword scabbards, such as Drna (Fig. 28.9.17–18) and Cernon-sur-Coole (Megaw 1978, 108). To these we can add some examples of grotesque faces appliquéd to the Brno flagon (Fig. 28.9.19) and the Loisysur-Marne belt buckle, with their almond eyes and prominent pear-shaped noses (Duval 1977, 131; Kruta and Bertuzzi 2007; Charpy and Roualet 1991, 265). That the highly stylized stamped motifs evoke a mask is not surprising, given the importance of the face (often shown full-face, sometimes very expressive, other times just sketched) and eyes in the iconographic La Tène tradition (Fig. 28.9.16–20). The combination of a mask and a trilobed palm leaf included in a segment of a circle (with hatching in the spandrels between the petals) is not uncommon, as evidenced by the bronze openwork fittings of Cizkovice dated to La Tène B1 (Kruta 1976, 121, fig. 6). The effect of the distortion induced by the rotational symmetry is not without reminiscences of the contortions and grimaces which deform the face of legendary figures such as Cuchulainn, one of the heroes of the Irish epic, Táin Bó Cúailnge (Guyonvarc’h 1994). With these scabbards the motif is not alone, but carefully duplicated to cover the entire surface evenly. The apotropaic virtues of the unblinking eye and the gaze that petrifies and terrifies are well known. Probably, the repetition of these ‘pairs of eyes’ serves only to affirm and increase the effect: for comparison, one might recall the figure of the giant Argos, the faithful guardian of Hera in Greek mythology, whose eyes were always open. Hera paid tribute to his loyalty by placing his hundred eyes on the tail of her favourite bird, the peacock (Ovid Metamorphoses I, 625).

A mulitpolar production This series of weapons, which can be placed in the transitional phase from Early to Middle La Tène, ie sometime in the first half of the 3rd century. BC, is notable for its obvious similarities, as much morphological as decorative. However, these sword scabbards are far from being completely identical, as shown in the techniques used for the decoration, the choice of stamps and their arrangement. The general composition may take the form of an angular interlace, more rarely curvilinear (Maubranches), or a dense and uniform oblique mesh. The diamond-shaped stamps are arranged to create an interlace effect. At Allonnes, stamps are applied directly onto the repoussé interlace. Circles and palmettes are used to decorate the areas described by the stamped pattern. When it comes to the uniform stamped lattice, a single stamp of lozenge form is used, while the oblique bands, with or without relief, remain perfectly smooth. When objects have strong similarities, like the Potypuzsta and Graz scabbards,

the design of the lozenge-shaped stamps is not identical but mirror images of each other. These few examples distributed over a wide geographic area (from Slovakia as far as the South and West of France) are sufficient to show, through the choices made, the diversity of the adopted solutions, and thereby leads us to see the result of a highly diversified production (Fig. 28.10). An image of the genesis of this ornamentation is conceived which is at once unified and multifaceted, in that they draw in an old repertoire, simultaneously adapted and updated, in different places (similar to conclusions about the genesis of tendrils into triskeles: Verger 1987, 333). This diversity makes it highly unlikely that these weapons can be attributed to a single centre of production, whether a workshop or even a region (except perhaps in the case of the Graz and Potypuzsta weapons). Considering the brevity of this ornamental fashion, we must believe that these specialized dealers reproduced freehand models which were rapidly distributed. This method, of short duration, cannot be compared to that of chagrinage which develops later, from the late 3rd century (De Navarro 1972; Lejars forthcoming). The diversity of the production centres is a hypothesis that fits well, it seems, with the cultural context of this period marked by the proliferation of developments and the renewal of military outfits – with the introduction of semi-rigid sword suspension systems and the first monoshell shield umbos – and their decoration. From the ornamentation point of view, this period saw the full development of Sword Style with some of the most remarkable achievements (the Cernon-sur-Coole, Flavigny and Halimba scabbards), but also the beginnings of Plastic Style are equally marked by the development of themes such as pairs of fantastic animals (zoomorphic lyres and dragons) and techniques hitherto little used, such as the addition of linked elements (Lejars 2003, 24–29; 2012). There is probably nothing original in this situation, even if research on the art of the ancient Celts is still strongly influenced by the idea of centralized production, linked to regional workshops or centres, spreading over large areas. Probably, this interpretive scheme owes its success to research on some quasi-industrial production from the ancient world, such as Greek painted vases, Roman wine amphorae, or even some artistic goods, supported by a very rich archaeological documentation and textual sources. In the case of La Tène artistic production, we have nothing like it. For each category of decoration, we often have a very limited series to work from, sometimes only a few objects scattered over a large part of the Celtic world. The argument for weapon workshops within settlements does not guarantee unique production. This is demonstrated by the weaponry workshop identified by J.-P. Guillaumet in Sajopetri, Hungary (Szabo and Czajlik 2007). The hypothesis of a common stamp for the Epiais-Rhus sword scabbards (which have six different patterns) and Moscano di Fabriano (for which there are only three motifs) was advanced by A. Rapin (2006a, 194),

28.  A l’aube du IIIe s. av. J.-C.: les fourreaux d’épée à décor estampé sur fer who concluded that there was a single tool (maybe even a single craftsman), but declined to specify the place and date of production. For the Hallstatt- La Tène transition, we can also mention different production locations for ‘timbale’ brooches which have come to light over the last few recent years from many sites in east-central France (Bourges, Lyon, Plombières-les-Dijon; Carrara et al. 2013). The primacy of a discovery and good preservation of the remains are certainly not sufficient arguments to indicate the origin of a form, a fashion or style. The concentration of a series in a given area does not provide further evidence of local origin, in particular when it comes to funerary contexts, since we know that customs differ greatly from one group to another, from one region to another and from one era to another for the same region. The example of stamp decorated sword scabbards is very revealing in this respect: in the extreme west, these objects, missing from funerary contexts, are, at the moment, only known to us from the debris of weaponry preserved in cult sites. If certain weapons come from the spoils of war, others were votive offerings from members of the community. There again, it is not easy to distinguish local from ‘foreign’. This difficulty is also the consequence of the high homogeneity that characterizes the Celtic weapons during the La Tène period and its adoption by the people of neighbouring cultural areas. The ornamentation of the Allonnes scabbard demonstrates the importance of these fragments, carefully collected but often neglected due to corrosion. Only detailed technical review, such as O. H. Frey called for in his conclusion to the note devoted to the Graz sword, can overcome the handicap of corrosion, to deepen our understanding of iron working, to enrich the still extremely limited corpus of ornament, and to open new perspectives on Celtic art. Translated by S. Crawford.

Remerciements Il m’est agréable d’exprimer ma plus vive reconnaissance à K. Gruel et V. Brouquier-Reddé qui m’ont confié l’étude du matériel métallique gaulois mis au jour à Allonnes, un site remarquable. Il faut louer ici la qualité du travail réalisé par l’équipe d’Arc’Antique à Nantes, et en particulier Stéphane Lemoine qui suit le dossier Allonnes depuis le début de l’opération.

Bibliographie Bataille-Melkon, A. and Rapin, A. 1997. Un fourreau celtique en fer orné au repoussé du Ve siècle avant J.-C, au Musée Saint-Rémi de Reims: Prosnes, Les Vins-de-Bruyère (Marne), Bulletin de la Société archéologique champenoise 90–2, 3–14. Bergonzi, G. 1988. I Celti nel Modenese. In A. Cardarelli, I. Pulini

261

and Chr. Zanasi (eds.) Modena dalle origini all’anno Mille. Studi di archeologia e storia I, 153–67. Brouquier-Redde, V. and Gruel, K. (dir.) 2004. Le sanctuaire de Mars Mullo chez les Aulerques Cénomans (Allonnes, Sarthe) Ve s. av. J.-C. – IVe s. apr. J.-C. Etat des recherches actuelles, Gallia 61, 291–386. Brunaux, J.-L. and Rapin, A. 1988. Gournay II, boucliers et lances, dépôts et trophées. Paris: Errance. Bujna, J. 1995a. Les contacts entre l’Europe centrale et la Gaule septentrionale au IIIe siècle av. J.-C.: la sépulture n 31 de Malé Kosihy, Slovaquie du Sud-Ouest. In J.-J. Charpy (éd.) L’Europe celtique du Ve au IIIe siècle avant J.-C., contacts, échanges et mouvements de populations, Actes du deuxième symposium international d’Hautvillers, octobre 1992, Société Archéologique Champenoise, Mémoire 9, 259–268. Sceaux: Kronos B.Y. éds. Bujna, J. 1995b. Malé Kosihy, latènezeitliches Gräberfeld, Katalog, Archaeologica Slovaca Monographiae 7. Nitra: Archeologický ústav slovenskej akadémie vied. Bulard, A. 1979. Fourreaux ornés d’animaux fantastiques affrontés découverts en France, Etudes Celtiques XVI, 27–52. Bulard, A. 1982. A propos des origines de la paire d’animaux fantastiques sur les fourreaux d’épée laténiens. In P.-M. Duval and V. Kruta (eds.) L’art celtique de la période d’expansion, IVe–IIIe siècle avant J.-C., colloque Collège de France, Paris, sept. 1978, 156–163. Genève: Droz. Cahen-Delhaye, A. 1981. Tombelles de La Tène I à Léglise, I. Inventaire, Archéologia Belgica 245, 5–40. Carrara, S., Dubreucq, E., Pescher, B. and Filippini, A. 2013. La fabrication des fibules à timbale comme marqueur des contacts et des transferts technologiques au cours du Ha D-LTA1. In A. Colin and F. Verdin (eds.) L’âge du Fer en Aquitaine et sur ses marges. Mobilité des hommes, diffusion des idées, circulation des biens dans l’espace européen à l’âge du Fer, actes du 35e Colloque international de l’AFEAF (Bordeaux, 2–5 juin 2011), Aquitania suppl. 30, Bordeaux, 595–608. Dechelette, J. 1914. Manuel d’archéologie préhistorique, celtique et gallo-romaine, t. II: Archéologie celtique ou protohistorique, Troisième partie, Second Age du Fer ou époque de La Tène. Paris: Picard. De Navarro, J. M. 1972. The Finds from the site of La Tène Vol. I: Scabbards and the Swords found in them, 2 vol. London: Oxford University Press. Duval, P.-M. 1977. Les Celtes, coll. L’Univers des Formes. Paris: Gallimard. Duval, P.-M. and Kruta, V. 1986. Le fourreau celtique de Cernonsur-Coole (Marne), Gallia 44–1, 1–27. Egg, M. and Schönfelder, M. 2009. Zur Interpretation der Schwertscheide aus Grab 994 von Hallstatt, Beiträge zur Hallstatt- und Latènezeit in Nordostbayern und Thüringen, Beiträge zur Vorgeschichte Nordostbayerns 7, 27–44. Nürnberg: Naturhistorische Gesellschaft. Frey, O. H. 1971. Das Keltische Schwert von Moscano di Fabriano, Hamburger Beiträge zur Archäologie I–1, 173–9. Frey, O. H. 1974. Akanthusornamentik in der keltischen Kunst, Hamburger Beiträge zur Archäologie IV, 141–57. Frey, O. H. 1978/1979. Ein Verziertes La Tène Schwert aus Graz, Schild von Steier, Beiträge zur Steirischen Vor- und Frühgeschichte und Münzkund 15/16, 67–73.

262

Thierry Lejars

Frey, O. H. 1995a. Das Grab von Waldalgesheim. Eine Stilphase des keltischen Kunsthandwerks. In H.-E. Joachim (éd.) Waldalgesheim: Das Grab einer keltischen Fürstin, 159–206. Köln: Rheinland Verlag / Bonn: R. Habelt. Frey, O. H. 1995b. Some comments on swords with dragon-pairs. In B. Raftery (éd.) Sites and Sights of the Iron Age, Essays on Fieldwork and Museum Research presented to Ian Mathieson Stead, Oxbow Monograph 56, 163–172. Oxford: Oxbow. Ginoux, N. 1994. Les fourreaux ornés de France du Ve au IIe siècle avant J.-C., Etudes Celtiques XXX, 7–86. Ginoux, N. 1995. Lyres et dragons. Nouvelles données pour l’analyse d’un des principaux thèmes ornementaux des fourreaux laténiens. In J.-J. Charpy (éd.) L’Europe celtique du Ve au IIIe siècle avant J.-C., contacts, échanges et mouvements de populations, Actes du deuxième symposium international d’Hautvillers, octobre 1992, Société Archéologique Champenoise, Mémoire 9, 405–412. Sceaux: Kronos B.Y. éds. Guyonvarc’h, Chr.-J. (trad.) 1994. La razzia des vaches de Cooley. Paris: Gallimard. Hunyady, I. 1942/1944. Keltak a Karpatmedenceben (Die Kelten im Karpatenbecken), Dissertationes Pannonicae II. 18. Budapest: A Királyi Magyar Pázmány Péter Tudományetem Éremés. Jacobsthal, P. 1944. Early Celtic Art. Oxford, Clarendon Press. Jacobi, G. 1982. Verzierte Schwertscheiden vom Frühlatèneschema aus den Flachgräbern von Manching, Germania 60, 565–8. Jope, E. M. 2000. Early Celtic Art in the British Isles. Oxford: Oxford University Press. Kramer, W. 1985. Die Grabfunde von Manching und die Latènezeitlichen Flachgräber in Südbayern, Die Ausgrabungen in Manching 9. Stuttgart: Steiner. Kruta, V. 1976. Le premier style laténien en Bohême. In C. Hawkes and P.-M. Duval (éds.) Celtic Art in Ancient Europe Five Protohistoric Centuries (L’art Celtique en Europe protohistorique), 111–40. London, New-York, San Francisco, Ca: Seminar Press. Kruta, V. 1979. Duchcov-Münsingen: nature et diffusion d’une phase laténienne. In P.-M. Duval and V. Kruta, V. (éds.) Les mouvements celtiques du Ve au Ier siècle avant notre ère, 81–116. Paris: éditions du Cnrs. Kruta, V. 1988. I Celti. In G. Pugliese Carratelli (ed.) Italia omnium terrarum alumna, La civiltà dei Veneti, Reti, Liguri, Celti, Piceni, Umbri, Latini, Campani e Iapigi, 263–311. Milan: Libri Scheiwiller. Kruta, V. and Bertuzzi, D. 2007. La cruche celte de Brno. Chefd’œuvre de l’art, miroir de l’univers. Dijon: Ed. Faton. Kruta, V. and Rapin, A. 1987. Une Sépulture de guerrier gaulois du IIIe siècle avant Jésus Christ découverte à Rungis (Val-deMarne), Cahiers de la Rotonde 10, 5–35. Kruta, V., Lambot, B., Lardy, J.-M. and Rapin, A. 1984. Les fourreaux d’Epiais-Rhus (Val-d’Oise) et de Saint-Germainmont (Ardennes) et l’art celtique du IVè siècle av. J.C., Gallia 42–1, 1–20. Kruta Poppi, L. 1978. Les vestiges laténiens de la région de Modène, Etudes Celtiques XV, 425–39. Kruta Poppi, L. 1983. Testimonianze celtiche nel territorio modenese, Miscellanea di Studi archeologici e di antichità, Deputazione di Storia Patria per le Antiche Provincie Modenesi, 21–31. Modena. Kruta Poppi, L. 1984. Contacts transalpins des Celtes cispadans au

IIIe s. av. J.-C.: le fourreau de Saliceta San Giuliano (province de Modène), Etudes Celtiques XXI, 51–61. L’art celtique en Gaule, collection des Musées de province, 1983/1984, Marseille, Paris, Bordeaux, Dijon. Paris: Ministère de la Culture, Direction des musées de France. Lejars, T. 1994. Gournay III: Les fourreaux d’épée, le sanctuaire de Gournay-sur-Aronde et l’armement des Celtes de La Tène moyenne, Archéologie Aujourd’hui. Paris: Errance. Lejars, T. 2003. Les fourreaux d’épée laténiens, supports et ornementation. In D. Vitali (éd.) L’immagine tra mondo celtico e mondo etrusco-italico – aspetti della cultura figurativa nell’antichità, Studi e scavi 20, 9–70. Bologna: Gedit Edizioni. Lejars, T. 2007. Lieux de culte et pratiques votives en Gaule à La Tène ancienne. In Mennessier-Jouannet, C., Adam, A.-M. and Milcent, P-Y. (eds). La Gaule dans son contexte européen aux IVe et IIIe siècles avant notre ère , actes du 27e colloque international de l’AFEAF, Clermont-Ferrand, 2003, Monographies d’archéologie méditerranéenne, hors série, Lattes 2007, 265–282. Lejars, T. 2012. Der Schwertstil. In Die Welt der Kelten. Zentren der Macht – Kostbarkeiten der Kunst, Eine Ausstellung des Archäologischen Landesmuseum Baden-Württemberg, des Landesmuseum Württemberg in Zusammenarbeit mit dem Landesamt für Denkmalpflege im Regierungspräsidium Stuttgart und dem Historischen Museum Bern, 15 septembre 2012 – 17 février 2013, 318–325. Ostfildern: Jan Thorbecke Verlag. Lejars, T. 2013. La Tène, un site, un mythe. La collection Schwab à Bienne (Suisse), Lausanne: Cahiers d’Archéologie Romandes. Megaw, J. V. S. 1968. Une épée de La Tène I, avec fourreau décoré, Revue archéologique est centre-est 19, 129–44. Megaw, J. V. S. 1973. The decorated Sword-scabbards of Iron from Cernon-sur-Coole (Marne) and Drna, Rimavska Sobota (Slovakia), Hamburger Beiträge zur Archäologie III–2, 119–37. Megaw, J. V. S. 1978. The decoration on the sword-scabbard from Grave 115. In Waldhauser, J. (éd.) Das Keltische Gräberfeld bei Jenisuv Ujezd in Böhmen, 106–113. Teplice. Megaw, R. and Megaw, J. V. S. 1989. Celtic art, from the beginnings to the Book of Kells. London: Thames and Hudson (rééd. augmentée 2001). Osterhaus, U. 1966. Die Bewaffnung der Kelten zur Früh-La-TèneZeit in der Zone nördlich der Alpen, unpublished dissertation, Univ. of Marburg. Rapin, A. 2006a. L’abstraction narrative dans l’imagerie des Celtes. In F.-H. Massa-Pairault (éd.) L’image antique et son interprétation, Collection de l’Ecole française de Rome 371, 185–207. Rome: Ecole française de Rome. Rapin, A. 2006b. Les élites guerrières celtes: données récentes de l’archéologie. In X. Delestre, M. Kazanski and P. Perin (éds.) De l’âge du fer au haut Moyen Age: archéologie funéraire, princes et élites guerrières, actes des tables rondes Longroy I (1er et 2 septembre 1998) et Longroy II (24 et 25 août 1999), Mémoires de l’association française d’archéologie mérovingienne 15, 43–57. Saint-Germain-en-Laye: AFAM. Rapin, A. 2008. Les Celtes d’Italie et leurs voisins septentrionaux: nouveaux outils d’analyses pour l’armement laténien du sud de l’Europe aux Ve et IVe s. av. J.-C. In St. Verger and D. Vitali (éds.) Tra mondo celtico e mondo italico. La necropoli di Monte Bibele, Atti della tavola rotonda, Rome, EFR, 1997,

28.  A l’aube du IIIe s. av. J.-C.: les fourreaux d’épée à décor estampé sur fer Dipartimento di Archeologia, 237–68. Bologna: Università di Bologna. Rapin, A., Szabo, M. and Vitali, D. 1992. Monte Bibele, Litér, Rezi, Piscolt, contributions à l’origine du Style des épées hongroises, Communicationes Archaeologicae Hungariae, 23–54. Rapin, A. and Zurfluh H. 1998. Le cimetière celtique de Cortrat (Loiret), Revue archéologique du Centre 37, 33–79. Roualet, P. and Charpy, J.-J. 1991. Les Celtes en Champagne – Cinq siècles d’histoire, catalogue d’exposition, Musée d’Epernay, 22 juin – 3 novembre 1991. Epernay: Musée d’Epernay. Santrot, H., Santrot, J. and Meuret, J.-Cl. 1999. Nos ancêtres les Gaulois, aux marges de l’Armorique. Nantes: Musée Dobrée. Stead, I. M. 2006. British Iron Age Swords and Scabbards. London: British Museum Press. Szabo, M. 1993. Eléments anthropomorphes dans le décor des fourreaux danubiens. In J. Briard and A. Duval (dir.) Les représentations humaines du néolithique à l’âge du Fer, actes du 115e Congrès national des sociétés historiques et scientifiques, Avignon, 271–86. Paris: éds CTHS.

263

Szabo, M. 2003. Eléments anthropomorphes dans le décor des fourreaux danubiens. In D. Vitali, D. (éd.) L’immagine tra mondo celtico e mondo etrusco-italico: aspetti della cultura figurativa nell’antichità, Studi e scavi 20, 71–4. Bologna: Gedit. Szabo, M. and Petres, E. F. 1992. Decorated Weapons of the La Tène Iron Age in the Carpathian Basin, Inventaria Praehistorica Hungariae V. Budapest: Magyar Nemzeti muzeum. Szabo, M. and Czajlik, Z. (eds.) 2007. L’habitat de l’époque de la Tène à Sajópetri Hosszú-Dűlő. Budapest: L’Harmattan. Verchère de Reffye, A. 1865a. Procédés pour le nettoyage et la conservation des objets en fer, Revue archéologique, mars, 392–7. Verchère de Reffye, A. 1865b. Procédés pour le nettoyage et la conservation des objets en fer, Matériaux pour l’histoire positive et philosophique de l’homme, Paris, première année, 418–22. Verger, St. 1987. La genèse celtique des rinceaux à triscèles, Jahrbuch des römisch-germanischen Zentralmuseums Mainz 34–1, 287–339.

29 ‘…TO BOLDLY GO WHERE NO MAN HAS GONE BEFORE.’ DEDICATED TO RUTH AND VINCENT… Boris Kavur and Martina Blečić Kavur

What do we have? The bed of the river Cetina, running through Sinjsko polje, in central Dalmatia (Croatia) was a well known archaeological site from the 19th century on. At the end of the 1980s, the quantity of archaeological finds deriving from the river increased – finds, mostly weapons and armour, ranged chronologically from the Neolithic to the medieval period. In the early ’90s the area in focus was between the confluence with the stream Ruda and the bridge in Trilj. The main objective was to determine the origin of the artifacts, but the archaeological survey in the river did not yield any finds or structures and consequently it was considered that the finds discovered in the river Cetina, might have actually derived from the river Ruda (Milošević 1992a, 46; 1992b, 88). Amongst other prehistoric finds kept in the private property of Milan Gilić, was also an Early La Tène style sword. It was briefly mentioned by Ante Milošević as a ‘curiosum’ on the site, and a photograph of the chape end was published in a preliminary report in 1992 (Milošević 1992b, 87, fig. 2). A few years later it was mentioned again in the Archeological topography of the Cetina (Milošević 1998, 291, fig. 482). In another preliminary presentation in 1999 it was mentioned that the sword was discovered in 1990 at the site of Mali Drinić in the river Cetina below the confluence with the stream Ruda (Milošević 1999, 208, fig. 6). The same photograph of the find was reproduced again in the exhibition catalogue Numini Hippi Fluvii, but for the first time it was also mentioned that the prehistoric finds might have been ritually deposited in the river (Milošević 2003, 27) (Fig. 29.1). Since its discovery, although the sword has been

mentioned and its picture reproduced at least four times in Slovenia, Croatia and Germany, it has not drawn any attention. Despite the fact the same photograph was reproduced every time and that its quality was rather bad, the sword’s general form – and most importantly, the form of the chape end and its decoration – were clearly visible. More than 20 years after its discovery, and taking into consideration that it was a river find which was not conserved, the sword is still in a rather good state of preservation – although its surface is completely covered with corrosion and several fragments of the scabbard are missing. The whole find is 70 cm long while the scabbard is 62 cm long. The preserved hilt is short – only 10.5 cm long and strangely formed – pointed towards the end, allowing the assumption that this form is a consequence of its weathering, and that the original grip was at least a little bit longer. The scabbard is almost completely preserved, but completely embedded in corrosion – it tapers only slightly but throughout. It consists of two chambered plates where the front plate overlaps the reverse one with edges – the overlaps extend to the point of the sheath. Its mouth is campanulate with concave sides and almost 6 cm wide. The upper edge of the mouth is decorated by a barelyvisible hatched band running between two incised lines. The front plate is decorated, but the decoration is only barely visible due to the corrosion. There are two incised lines running parallel to the edge of the sword all along the edges. On the top of the scabbard, several concave, convex and circular incised lines have been preserved which are visible, and which are the remains of a depicted dragon pair. Unfortunately we can identify only a few elements of

29.  ‘…to boldly go where no man has gone before.’ Dedicated to Ruth and Vincent…

265

Fig. 29.1. Map with major sites mentioned in the text.

the dragons – on both sides we can recognize the pointed ear and the concave upper line of the head. On both sides also several lines forming the beak can be recognized – it seems as if the upper jaw ends with an inward-curved end while the lower jaw can be identified only as a line. Below the ears there is a point of rupture, and the dorsal lines of the back can be further followed. They run transversely in a convex form to the central part of the sword where they end in two upwards rolled spirals. In several spots, where

the corrosion was broken off, we can see that the surface between the dragons was covered with hammered points. In some places, mostly below and between the dragons, we can see traces of fan-shaped hatched ornaments. The chape end is open and heart-shaped. It is short (some 11 cm long) with two circular clamps on the upper, and a slightly concave bridge on the lower side. The clamps are of the round type and decorated with two circular incisions with a triangular cross section, creating the visual impression

266

Boris Kavur and Martina Blečić Kavur

of an elevated circular button in the middle encircled with a thin line. On the point of rupture, where the chape end widens, there are two circular, slightly flattened buttons tapped into the inner side. The concave rods of the chape end between the circular buttons and the chape clamps are decorated with incised transverse lines. The larger chape clamps are circular – on the reverse side they are plain and decorated only with a thin circular line running along the edges, while on the front side they are decorated with a motive of a triskele ending with open-beaked dragon heads. The rods forming the end of the chape are a little bit convex and thickened towards the curved tip (Fig. 29.2).

What is it? Sword scabbards with an open reinforced chape end were, accepting the suggestion of Ian Stead, named as the Kosd type by Éva Petres and Miklos Szabó. They further subdivided them into four groups designated ‘A’ to ‘D’ according to the form of the chape end – oval shapes were included in the designations type A and B, conical ones as type C, and deltoid shaped ones as type D (Petres and Szabó 1985, 91). The last ones formed the smallest group, and mapping them, Peter Ramsl observed that their occurrence is concentrated in the Carpathian basin (Ramsl 2003, 256, fig. 12). In the revision of grave number 22 from the necropolis of Belgrade-Karaburma we have further supplemented the list of these swords, but also pointed to their occurrence outside of this region – in Northern Italy and in Dalmatia, in the river Cetina (Blečić Kavur and Kavur 2010, 71). Contrary to other forms of chape ends, the ones included in the Kosd D type tend to be rhomboidal shaped, shorter, with circular buttons on the clamp positioned above the middle of the chape end, with a wider opening of the chape end, with a more concave upper and convex lower part. But when observed on a larger scale, they are, outside of the Carpathian basin and within as well, much more variable, and widening the perspective we could propose a further division of the later type, observing them in the Celtic world – in the territory between France and Romania. Generally we would agree that it would be better to describe the chape ends as ‘heart-shaped’ as already proposed by P. Ramsl (2002, 79). All of the swords are approximately 70 cm long with scabbards long up to 62 cm – commonly they tend to be less than 4 cm wide with scabbards up to 5 cm wide. The chape ends tend to be short and the rods between the clamp plates and the circular buttons are as short – as short as the clamp plates. Observing their general form we can divide them into two groups (Fig. 29.3). In the first group we can include the swords with a more curved and shorter chape end (for example Pottenbrunn grave 520 and Cetina near Trilj) while in the

second group we can include the more pointed and elongated chape ends (for example Monte Bibele, Monte Tamburino grave 118 and Kosd grave 15) (Fig.29.3).

The first group In the first group – the curved heart-shaped chape ends with clamps positioned on the middle of the chape end – we should include scabbards from the swords discovered in grave 21 in Gourgançon (Charpy 1987, Pl. IX, 14) and the stray find discovered in the vicinity of Chalon-sur-Seine in France (Charpy 1987, Pl. VII, 17); graves number 121 and 94 from the necropolis of Monte Tamburino at Monte Bibele (Vitali 1987, 359, fig. 35b; Lejars 2008, 207, 215) and the sword from the old Rilli collection from Numana in Italy (Lollini 1979, 67, T. VI, B); the old grave number 2 from 1930 (Ramsl 2002, T. 22) and grave number 520 from Pottenbrunn in Austria (Ramsl 2002, T. 58, 15a); as well as grave number 3 from Novajidráný-Sárvár in Hungary (Hellebrandt 1997, 77, fig. 6). Finally we have to add to this group the sword discovered in the river Cetina near Trilj in Croatia. Perhaps we could add also the heavily corroded scabbard from grave number 8 from Trivio di Serra S. Quirico in the Marche region in Italy (Lollini 1979, 78; T. XVII, B) and the isolated find from Chens-sur-Léman in Switzerland (Bocquet 1991, 99, fig. 3, 13). The latter differs from all the others due to the form of its buttons above the chape clamps which are not circular but pointed and turned up on the outer side. Unfortunately none of the preserved scabbards was decorated – only the chape clamps of the scabbards from Pottenbrunn grave 1930/2, grave number 94 from Monte Bibele and grave number 3 from Novajidráný-Sárvár were decorated with plastic convex buttons with triskeles positioned in the middle of a circle with lateral incisions, while the sword from Cetina had buttons decorated with triskeles ending in the form of dragons heads with open beaks. It was only the sword from Cetina which had a dragon pair depicted on the mouth of the scabbard.

The second group In the second group – the pointed heart-shaped chape ends with clamps positioned slightly above the middle of the chape end – we should include the scabbards from graves discovered mostly in the Carpathian basin; from grave number 8 from Getzersdorf in Austria (Baumgartner 1906, 291, fig. 111); grave 15 from Kosd (Szabó and Petres 1992, 147, Pl. 29; Szabó 2008, 234, fig. 18, 2b); the accidental find from Szob (Szabó and Petres 1992, 186, Pl. 68); grave number 4 from Rezi-Rezicseri in Hungary (Horváth 1987, 156, Pl. XVII, 2; Szabó and Petres 1992, 175, Pl. 57); grave number 137 from Pisçolt in Romania (Németi 1988, 89, fig. 10, 1a); accidental find from Ruma-Borkovac (Todorović

29.  ‘…to boldly go where no man has gone before.’ Dedicated to Ruth and Vincent…

267

should also add two poorly preserved accidental finds – the scabbard of the sword from Szob in Hungary (Szabó and Petres 1992, 186, Pl. 68) and the fragment of the scabbard from Ruma-Borkovac in Serbia (Todorović 1968, T. XLIII, 10). The latter is questionable since its chape clamps do not appear circular in form – making them look more knob-like in shape and lacking the clearly pronounced buttons on the very slightly thickened rod above them. The mouths of scabbards from Kosd, Szob, ReziRezicseri, Monte Bibele and most probably Pisçolt were decorated with a Type I dragon pair, according to José Maria De Navarro. Several of the chape clamps were also decorated – the scabbard from Kosd had an incised rosette while the scabbards from Szob and Rezi-Rezicseri had an almost identical decoration of a point inside a circle with lateral incisions. Only the scabbard from Ruma-Borkovac, decorated with a Hungarian style ornament, had incised vegetable ornaments on the chape clamps and transverse incisions on the rod above them.

Chronology

Fig. 29.2. The sword from the river Cetina near Trilj (Croatia). General view and details.

1968, T. XLI, 3); and grave number 22 from BelgradKaraburma in Serbia (Blečić Kavur and Kavur 2010, 70, fig. 66, 8). But we should also add the scabbards from graves number 70 (Vitali 1987, 356, 358, fig. 35a; Ginoux 2007, Pl. 43; Lejars 2008, 196), number 118 (Lejars 2008, 213), number 127 (Lejars 2008, 218) and number 135 (Lejars 2008, 221) from Monte Tamburino, and from grave number 32 from Monterenzio Vecchia in Monte Bibele from Italy (Bernadet et al. 2007, 244, fig. 15). Most probably we

In an analysis of scabbards with open chape ends of a noncircular form, Thierry Lejars demonstrated that they should be dated to the last horizon of the Early La Tène period (LT B2) while early forms already make their appearance in the LT B1 (Lejars 1994, 44–47). In terms of the division of swords as proposed by É. Petres and M. Szabó (1992), we can conclude that the majority of swords included in group Kosd C could be dated to the LT C while the swords from group Kosd D could be still included in to the period LT B2. This is also reflected in their size – the majority of the swords from the first group are longer than 70 cm while the majority of the swords from the second groups are shorter (Blečić Kavur and Kavur 2010, 69), but when observing them in detail, we can conclude that the swords with more curved chape ends tend to be shorter and narrower than the examples with the more pointed ones. When presenting the preliminary results of the chronological analysis of the graves from Monte Bibele, Daniele Vitali concluded that the swords included in group Kosd D should be dated at the end of LT B2 and the beginning of C1, which was in chronological terms between 320 and 280 BC (Vitali 1987, 356, 368, fig. 39). A slightly earlier dating was proposed by P. Ramsl who, on the basis of the analysis of the swords from the cemetery of Pottenbrunn, suggested that the examples with a heartshaped chape end could be dated to the later part of the LT B2 (Ramsl 2002, fig. 148). One of the most important and interesting details, enabling a more precise dating of the sword from Cetina near Trilj, are the decorated chape clamps. They have deep plastic relief decoration in the form of triskele with open-

268

Boris Kavur and Martina Blečić Kavur

Fig. 29.3. Division of scabbards with open reinforced heart-shaped chape ends. 1. Pottenbrunn, grave 520 (Ramsl 2002, T. 58, 15a); 2. Cetina near Trilj; 3. Monte Bibele, Monte Tamburino, grave 118 (Lejars 2008, 213); 4. Kosd, grave 15 (Szabó and Petres 1992, 147, Pl. 29).

Fig. 29.4. Evolution of a motive – from dragons to geometric motives. 1. Cetina – triskele ending in dragon heads with open beaks; 2. Szob – two dragon heads facing one another (Szabó 1989b, 20; Pl. 2, 1); 3. Srednica near Ptuj – the deconstructed dragon’s head (Lubšina Tušek and Kavur 2009, 137, fig. 9); 4. Apahida – triskele with vegetable offshoots (Rusu 1969, 292, fig. 10, 3); 5. Ciumeşti – triskele with vegetable offshoots (Rusu 1969, Taf. 146).

beaked dragon-head terminals. The eyes of the dragons are clearly depicted and the ends of the widely open beaks are curled (Fig. 29.2, Fig. 29.3, Fig. 29.4). They form a rare motive present mostly in a developed but reduced form in the Carpathian Basin. Among the older swords the best analogy would be the decorated plastically modelled buttons on the scabbard reinforcement of the

sword from Szob, where on each button two dragons were depicted facing one another. The eyes are clearly visible and the ends of the beaks are also curled (Szabó 1989b, 20, Pl. 2, 1) (Fig. 29.4.2). Perhaps another such example is the unfortunately badly-preserved lower loop plate from the Hatvan-Boldog type sword from Gáva-Katóhalom in Hungary. The loop plate is circular, decorated with a band

29.  ‘…to boldly go where no man has gone before.’ Dedicated to Ruth and Vincent… of hatched ornament along its edge while in the centre, inside a circle, from the damaged remains we can recognize in the lower right angle the head of a dragon with an eye clearly depicted and a curled beak (Szabó and Petres 1992, 131, Pl. 13, 255, Ill. III, 1). Perhaps we could assume a similar decoration on the chape clamps on the reverse of the eponymous sword from Hatvan-Boldog which is unfortunately also badly preserved (Szabó and Petres 1992, 136, Pl. 18). In an attempt to demonstrate the development of the decoration of the Hatvan-Boldog type swords in the Carpathian basin we have added, along with the two Hungarian swords from Hatvan-Boldog and Gáva-Katóhalom, the find from Srednica near Ptuj in Slovenia (Lubšina Tušek and Kavur 2009, 136, fig. 7). The latter had its loop plates decorated with a further developed form of the same decoration – the dragon’s head was deconstructed – the eye was omitted, the curved lower jaw of the beak was transformed into a circle, while the upper jaw of the beak was reduced to a vegetable off-shoot (Lubšina Tušek and Kavur 2009, 137, fig. 9; Megaw and Megaw 2012, fig. 4) (Fig. 29.4.3). Generally in LT C1 the motive becomes reduced to a triskele where the ends only just still resemble open beaks as, for example, on the lower loop plate of the sword from Szob (Szabó and Petres 1992, 179, Pl. 61) or on the decorative buttons on the upper side of the scabbard reinforcement of the sword from the vicinity of Balassagyarmat (Szabó 1989b, 20–1, Pl. 2, 2a; Szabó and Petres 1992, 121, Pl. 3). Further, the motive can be observed on other elements of weaponry, such as the decorative button attached to the triangular reinforcement of the helmet from Apahida in Romania (Rusu 1969, 292, Abb. 10, 3; Szabó and Petres 1992, 57, fig. 33). Here, the endings of the triskele still look similar to the open beaks of dragons, although their morphology is reduced (Fig. 29.4.4). Unquestionably the most important find of such a motive is the decoration of the inner circle of the two small roundels on the clamp of the chain mail from the grave from Ciumeşti in Romania (Rusu 1969, T. 146; Szabó 1989a, 18; Pl. 1, 2a). The endings of the triskele demonstrate the further development of the reduction started in the decoration of the loop plate from Srednica. The dragon’s heads are completely reduced – the lower jaw of the beak forms a circle ending with a point while the upper jaw becomes detached and reduced only to a line. The ornament was discussed by Aurel Rustoiu and Vincent Megaw (2011, 233) from a different perspective – they observed it in connection with the distribution of the Vegetal Style. Further they have noted in their discussion the matching or actually overlapping distribution of swords decorated with dragon pairs and the style végetal continu as defined by Stéphane Verger (1987). Looking at the known finds decorated with the motive of triskeles – the ones ending with dragons heads and the reduced and transformed example – we can conclude that the plastic decoration of the chape clamps of the sword

269

from Cetina, with its triskele finishing in the form of the dragon’s head is, according to the form’s elaboration and semantic meaning, closely comparable to the decoration of the older Kosd type A swords from the Carpathian basin. It seems that triskeles made in plastic style and composed from dragon heads may be dated to LT B2, while later in phases LT B2/C1 and C1, the stylistic development of the dragon’s heads becomes reduced and deconstructed – they are transformed into a geometric pattern composed of a triskele containing three circles (Fig. 29.4).

What does it mean? The first finds demonstrating contacts between the communities of the Alpine area and the cultures of the central and eastern Mediterranean world can be dated back to the Early Iron Age. Not only artifacts from the western Mediterranean discovered in central Europe, but also finds demonstrating the opposite movement of contacts and cultural flow – mostly female attire – were discovered in Greek and Sicilian sanctuaries (Verger 2003; 2011). Coming from the most important Pan-Hellenic sanctuaries, such as the Hearion in Perachora where anklet rings, bracelets and decorative discs were discovered, they could be dated to the period Ha D1 according to central European chronology. That this was a tradition obviously sustained in the following centuries is demonstrated by the presence of walnut-shaped anklet rings in Isthmia (Caskey 1960; Krämer 1961), as well as fibulas of the Middle La Tène schema from Dodona (Megaw 1968, 187–8; 2004, 100) and the sanctuary of Apollo in Delos. Discussing the latter fibula, M. Szabó formulated three hypotheses regarding its origin – it might have been a present from the ‘Celtic world’, or the property of a slave originating from the same environment, or booty captured from the Celts in Asia Minor. In his opinion the first hypothesis was most likely, since he noted that the majority of isolated artifacts originating from Balkan territory were discovered in sanctuaries of the Greek world (Szabó 1971, 514). Later, Ferdinand Maier observed that the majority of these finds could be included in a consistent chronological horizon encompassing the end of the Early and the beginning of the Middle La Tène. This horizon could be linked to the events of 279 and 278 BC, since according to his opinion, the presence of finds in sanctuaries indicated that they were parts of trophaia (Maier 1973, 477). He included the Early La Tène style sword from Dodona and the anklet rings from Isthmia in the same horizon, although M. Szabó demonstrated that the context of their discovery predated the event (Szabó 1971, 503). The sword from Dodona is an old discovery kept in the National Museum in Athens and originating from the collection of Constantin Carpanos (Szabó 1971, 504). The only information about its origins states that it was

270

Boris Kavur and Martina Blečić Kavur

discovered in Dodona in the Epirus (Meier 1973, 464). Discussing it some 45 years ago, J. V. S. Megaw assumed that it could be connected with the armament of Celtic mercenaries, whose military service diffused their artifacts all across the eastern and central Mediterranean (Megaw 1968, 191). Decades later, when returning to this topic, he was more precise and noted that the finds from Isthmia could be connected to the mercenaries present during the actions of Timoleon of Corinth on Sicily, and the finds from Dodona to the troops housed in the garrison of Phoenice not far from Dodona (Megaw 2004, 97–99). In his conclusion, he drew attention again to the fact that these artifacts were discovered in ritual or sacred places: ‘recalling the Celtic propensity for depositions particularly in watery contexts’, and raising the question of whether they were deposited there by their original owners (Megaw 2004, 104). But to shed light on the problem we should look across the sea and further back in time. In the years 369–368 BC Dionysius the Elder from Sicily had already sent Celtic mercenaries to help Sparta in their battles against the Thebans (Occhipinti 2006, 79; Sordi 2002, 496; Treister 1996, 245). The short description of this action in the Hellenica of Xenophon is the first incontestable description of Celtic people in ancient sources (Freeman 1996, 20). He must have established contact with the Celts at least some years before in the framework of his plans to build up a maritime empire in the Adriatic to act as a counterpart to the Carthaginian empire in the Western Mediterranean. These mercenaries were battle-hardened Italic Celts who, as Timothy Bridgman suggested, were collaborating with Syracuse even before Dionysius the Elder presumably coordinated them in the military action that finally led to the sack of Rome (Zuffa 1978, 139; Bridgman 2003, 49). In the next decades they were active as mercenaries of Syracuse in their campaigns against Kroton and Locri (Bouzek 2002, 55; Szabó 1991, 333) and crucially during the campaign against Corsica, when Sardinia and Pyrgy were also taken (Sordi 2002, 494), consequently making plausible the assumed origin of the finger rings discussed long ago by J. V. S. Megaw (1965/1966, 97; Lejars 2006, 79). It was in Sicily, when Dionysius the Elder was accused of being a barbarian-lover, that the propaganda machinery from Syracuse coined the myth explaining that Poliphemus and the nymph Galatea were the parents of Keltos, Galata and Illyros, consequently bringing their favorite mercenaries into the Mediterranean cultural circle with their constructed genealogy (Bridgman 2003, 49–50). It was a period when their detachments – armed with their own equipment and fighting in their own tradition, according to Dionysius’ program, as noted by Mikhail Yu. Treister – had entered the Mediterranean to be employed in the distant parts of the oikumene (Treister 1996, 248). When concluding his presentation of the Celts in Italy, Thierry Lejars referred back to this decision of Dionysius the Elder to supply his

mercenaries with their traditional weapons, to demonstrate how they were integrated into the Mediterranean communities but at the same time were able to preserve their identity (Lejars 2006, 91). According to many classical authors, their identity and reputation for toughness in battle ensured that, in the centuries to come, no ruler would wage war without Celtic mercenaries (Freeman 1996, 21). Consequently, the Celtic sword represented not only a weapon, but also a symbolic object designating the status of the individual and his affiliation with a group of warriors identifying themselves as being ‘Celts’ – a group of mercenary warriors keeping their privileges in the Mediterranean and sustaining their contacts with prehistoric communities in the north: a weapon, decorated with unique symbols, which was removed from public display and violently destroyed when its owner passed over into the community of the dead.

Conclusion Taking into consideration the proposed revision of absolute chronology for the central Balkan Celtic presence (Blečić Kavur and Kavur 2010, 73–76; Kavur and Guštin 2011, 129–130), and in accordance with the chronological observations of D. Vitali (1987), the sword from Cetina near Trilj predates the catastrophic events in the Balkans in 279 BC – due to its size and form the same could be assumed also for the sword from Dodona. The earliest deposition of Celtic weapons in Greek sanctuaries, as confirmed by the epigraphic evidence from an inventory list from the treasury of the old temple of Athena on the Acropolis in Athens, took place even before the middle of the 4th century BC (Freeman 1996, 23). Clearly the ritual deposition of these weapons, loaded with symbolism, should be observed as a part of the process taking place in the 4th century BC when the Celts from Northern Italy entered the Mediterranean as mercenaries engaged in the political ambitions of Dionysius the Elder from Syracuse, thus becoming the ‘Adriatic Celts’ (Guštin 2005a, 112–113; 2005b, 50–52). The general form of the sword (as well as other finds of Celtic provenience discovered in the river Cetina) subtly suggest that we should look for their origin in Northern Italy, while the plastic decoration of the chape clamps of the sword from Cetina, with its triskele finishing in the form of the dragon’s head are, according to its form elaboration and semantic meaning, closely comparable to the decoration of the younger group of Kosd A type swords from the Carpathian basin dated to Lt B2. Still, in general traits the distribution of the swords we have included in our first group – the swords with the curved heart-shaped chape ends and with clamps positioned on the middle of the chape end – demonstrate a wide distribution similar to the general distribution of the Hatvan-Boldog (or Munsingen or Kosd A) type swords as demonstrated by Thomas Stöllner (Stöllner 1998, Beilage 3).

29.  ‘…to boldly go where no man has gone before.’ Dedicated to Ruth and Vincent… This brings us again to the observations of M. Szabó, who compared the finds from Italy and stressed their founding role in relation to the Carpathian basin to demonstrate the evolution of the dragon pair motive (Szabó 1989a). Furthermore, J. V. S. and M. R. Megaw were the ones who pointed out that the oldest swords decorated with dragon pairs discovered in Italy, were discovered on the territory of the Boii, who were assumed to have been of Central European or Danubian origin (Megaw and Megaw 1990, 71). The oldest examples of the second group just look as if they were a form further developed from the first one in the region of Northern Italy. In many elements, the sword from Cetina resembles the Hatvan-Boldog type swords from the Carpathian basin – above all the Plastic Style decorative knobs on the chape plates decorated with a triskele with dragon heads. Understanding the structure of the image, isolating the individual forms and observing the relations among them, we are able to identify individual elements used in the artistic vocabulary to reproduce a realistic image. The decoration of the chape clamps stands at the beginning of the process of simplification and abstraction of the form: a process so typical for Celtic art which is just a reflection of the basic deconstruction of different elements used, deconstruction enabled by the profound understanding of the mythological schemes behind the images. But most problematic of all is the barely-visible incised decoration of the dragon pair on the top of the scabbard – the wide open beaks with the upper jaws curved into the mouth, the angle of the back lines and the curled tail suggest that we might be looking at a dragon pair of the older, II type according to J. M. De Navarro. On the other hand, the majority of the swords from this group, which represent the ‘classical’ swords with heart-shaped open chape ends from the Carpathian basin, were decorated with dragon pairs of the II type and some of them already exhibit decorative elements of the Hungarian sword style.

Conclusion To review the story – the sword was discovered in the river Cetina in Dalmatia in Croatia far away from any other Celtic sword discovered in the region (Guštin 1984, 343, fig. 28). Originating from the area of Northern Italy, it was fabricated in a workshop in the region around today’s Bologna, most probably controlled by the Boii. It was manufactured as a local product reflecting the traditions of the universal Early La Tène forms such as the shape of the scabbard, but it featured several decorative elements linking it to the new stylistic features developing in the workshops of the Carpathian basin. It was discovered in an area where these specific weapons were not used in the Late Iron Age – it was ritually deposited together with several other Celtic

271

artifacts in a river at a place where similar sacrifices had already been performed over several centuries (Milošević 1999, 206–210; Blečić Kavur and Pravidur 2012, 77), being one of the few swords decorated with dragon pairs ritually deposited in a sanctuary or in water (Stöllner 1998, Beilage 2). We cannot be sure whether it was deposited as a war trophy or by its owner, but we can be pretty sure that it was brought across the Adriatic by Celtic mercenaries in the 4th century BC who were determined to ‘...boldly go where no man has gone before’.

Acknowledgments We have to thank Milan Gilić (Sinj) who was kind enough to allow us to see and publish the sword from Mali Drinić. Also we have to thank Angela Tabak (Muzej triljskog kraja, Trilj) and Ivana Ožanić Roguljić (Institut za arheologiju u Zagrebu, Zagreb) for their help and support, and professor Mitja Guštin (Institut za dediščino Sredozemlja, Univerza na Primorskem, Koper) for his valuable comments.

Bibliography Baumgartner, G. 1906. Prähistorische Funde nächst Getzersdorf (N.-Ö.), Mitteilungen der k.k. Zentral-Kommission 5, 287–296. Bernadet, R., Lejars, T., Naldi, V., Verger, S. and Vitali, D. 2007. Monterenzio, (prov. De Bologne). La nécropole Celto-étrusque de Monterenzio Vecchia, Mélanges de l’école française de Rome 119.1, 238–245. Blečić Kavur, M. and Kavur, B. 2010. Grob 22 iz beogradske nekropole Karaburma: retrospektiva i perspektiva. Starinar 60, 57–84. Blečić Kavur, M. and Pravidur, A. 2012. Ilirske kacige s područja Bosne i Hercegovine / Illyrian helmets from Bosnia and Herzegovina. Glasnik Zemaljskog muzeja Bosne i Hercegovine u Sarajevu 53, 35–136. Bocquet, A. 1991. L’archéologie de l’âge du Fer dans les Alpes Occidentales Françaises. In A. Duval (ed.) Les Alpes a l’age du Fer. Revue Archéologique de Narbonnaise, Supplément 22, 91–155, Editions CNRS, Paris. Bouzek, J. 2002. Die Bronzelekithos des Typus Talcott aus Hurbanovo und andere vorrömische mediterrane Bronzegefäße in der Slowakei. Anodos. Studies of the Ancient World 2, 53–57. Bridgman, T. P. 2003. The ‘Gallic disaster’; Did Dionysius I of Syracuse order it? Proceedings of the Harvard Celtic Colloquium 23, 40–51. Caskey, J. L. 1960. Objects from a well at Isthmia. Hesperia 29.2, 168–176. Charpy, J.-J. 1987. Les épées laténiennes àbouterolle circulaire et ajourée des IVe et III siècles avant J.C en Champagne. Études celtiques XXIV, 43–80. Freeman, P. M. 1996. The earliest Greek sources on the Celts. Études celtiques XXXII, 11–48. Ginoux, N. 2007. Le thème symbolique de „la paire de dragons“ sur les fourreaux celtiques (IVe – IIe siècles avant J.-C.). Etude

272

Boris Kavur and Martina Blečić Kavur

iconographique et typologique, BAR International Series 1702, Oxford, British Archaeological Reports. Guštin, M. 1984. Die Kelten in Jugoslawien. Jahrbuch des Römisch-Germanisches Zentralmuseums 31, 305–363. Guštin, M. 2005a. The Adriatic Celts and their Neighbours. In W. Gillies and D. W. Harding (eds.), Celtic connections, Volume 2. Papers from the 10. International Congress of Celtic Studies, 111–124. Edinburgh, Edinburgh University Press. Guštin, M. 2005b. Celts on the margin of the northern Adriatic. In H. Dobrzańska, J. V. S. Megaw and P. Poleska, P. (eds.) Celts on the Margin. Studies in European Cultural Interaction 7th Century BC – 1st Century AD. Dedicated to Zenon Woźniak, 49–55. Kraków, Institute of Archaeology and Ethnology of the Polish Academy of Sciences. Hellebrandt, M. B. 1997. Keltische Eroberung und Ansiedlung in Nordungarn. Zalai múzeum 8, 69–77. Horváth, L. 1987. The surroundings of Keszthely. In L. Horváth, M. Kelemen, A. Uszoki and É. Vadász, Transdanubia I. Corpus of Celitc finds in Hungary, Volume I, 63–17, Budapest, Akadémiai kiadó. Kavur, B. and Guštin, M. 2011. Contribution to chronology. In M. Guštin and M. Jevtić (eds.) The Eastern Celts. The Communities between the Alps and the Black Sea. 129–130, Koper, Beograd, Annales Mediterranei. Krämer, W. 1961. Keltischer Hohlbuckelringe vom Isthmus von Korinth. Germania 39, 32–42. Lejars, T. 1994. Gournay III, Les fourreau d’épée, Le sanctuaire de Gournay-sur-Aronde et l’armement dec Celtes de La Tène moyenne, Paris, Édition Errance. Lejars, T. 2006. Les Celtes d’Italie. In M. Szabó (ed.) Celtes et Gaulois, l’archéologie face à l’histoire, 3 : les civilisés et les Barbares (du Ve au IIe siècle avant J.-C.). Actes du colloque de Budapest, 17–18 juin 2005. 77–96, Glux-en-Glenne, Centre archéologique européen. Lejars, T. 2008. Les guerriers et l’armement celto-italique de la nécropole de Monte Bibele. In D. Vitali and S. Verger (eds.) Tra mondo celtico e mondo italico. La necropoli di Monte Bibele. Atti della Tavola Ronda, Roma 1997, 127–222, Bologna, Università di Bologna. Lollini, D. G. 1979. I Sennoni nell’Adriatico alla luce delle recenti scoperte. In P.-M. Duval and V. Kruta (eds.) Les mouvements celtiques du Ve au Ier siècle avant notre ère, 55–79. Paris, Éditions CNRS. Lubšina Tušek, M. and Kavur, B. 2009. A sword between: the Celtic warriors grave from Srednica in North-Eastern Slovenia. In G. Tiefengraber, B. Kavur and A. Gaspari. Keltske študije II, Studies in Celtic Archaeology, Papers in honour of Mitja Guštin. Protohistoire européenne 11, 125–139. Maier, F. 1973. Keltische Altertümer in Griechenland, Germania 51, 459–477. Megaw, J. V. S. 1965/1966. Two La Tène finger rings in the Victoria and Albert Museum, London: an essay on the human face in the Early Celtic Art. Praehistorische Zeitschrift XLIII/ XLIV, 96–166. Megaw, J. V. S. 1968. Two finds of the Celtic Iron Age from Dodona. In K. Jażdżewski (ed.) Liber Iosepho Kostrzewski octogenario a venetoribusdicatus, 185–193. Wrocław, Warszawa, Kraków, Zakład narodowy im ossolińskich – wydawnictwo. Megaw, J. V. S. 2004. In the footsteps of Brennos? Further

archaeological evidence for Celts in the Balkans. In B. Hänsel and E. Studeniková (eds.) Zwischen Karpaten und Ägäis. Neolithikum und Ältere Bronzezeit. Gedenkschrift für Viery Němejcová-Pavúkova. Internationale Archäologie, Studia honoria, Band 21, 93–107. Rahden/Westfahlen, Verlag Marie Leidorf. Megaw, J. V. S. and Megaw, M. R. 1990. ‘Semper aliquid novum…’ Celtic dragon-pairs re-reviewed. Acta Archaeologica Academiae Scientiarum Hungaricae 42, 55–72. Megaw, J. V. S. and Megaw, M. R. 2012. Poles apart? Notes from the fringes of the ‘Hungarian’ sword style. In Anreiter, P., Bánffy, E., Bartosiewicz, L., Meid, W. and Metzner-Nebelsick, C. (eds.) Archaeological, Cultural and Linguistic Heritage. Festschrift for Erzsébet Jerem in Honour of her 70th Birthday. 401–414, Archeolingua, Budapest. Milošević, A. 1992a. Arheološki nalazi u koritu rijeke Cetine u Sinjskom polju, Obavijesti Hrvatskog arheološkog društva XXIV.2, 45–48. Milošević, A. 1992b. Arheološki nalazi u koritu rijeke Cetine u Sinjskom polju, Arheo 15, 86–88. Milošević, A. 1998. Arheološka topografija Cetine. Katalozi i monografije 3, Split, Muzej hrvatskih arheoloških spomenika. Milošević, A. 1999. Archäologische Probenuntersuchungen im Flussbett der Cetina (Kroatien) zwischen 1990 und 1994. Archäologisches Korrespondenzblatt 29.2, 203–210. Milošević, A. 2003. Numini Hippi Fluvii. Split. Ministarstvo Kulture RH, Muzej hrvatskih arheoloških spomenika. Németi, I. 1988. Necropola Latène de la Pişcolt, jud. Satu Mare. I. Thraco-Dacica IX.1–2, 49–73. Occhipinti, E. 2006. Dionisio il Vecchio e il consenso delle popolazioni anelleniche della Sicilia. Hormos 8, 65–82. Petres, É. F. and Szabó, M. 1985. Bemerkungen zum sogenannten ‘Hatvan-Boldog’ – Schwerttyp, Alba Regia XXII, 87–96. Ramsl, P. 2002. Das Eisenzeitliche Gräberfeld von Pottenbrunn. FÖMat A11, Wien. Ramsl, P. C. 2003. La nécropole laténienne de Pottenbrunn (BasseAutriche), miroir des relations Est-Ouest. In O. Buchsenschutz, A. A. Bulard, M.-B. Chardenoux and N. Ginoux (eds.) Décors, images et signes de l’âge du Fer européen. Actes du XXVIe Colloque de l’Association Française pour l’Etude de l’Age du Fer, 247–258. FERACF, Tours. Rustoiu, A. and Megaw J. V. S. 2011. A foreign flowering in Transylvania: The vegetal style armring from Fântânele – Dealul Popii, Jud. Bistriţa-Nǎsǎud, grave 62. In D. Măgureanu, D. Măndescu and S. Matei (eds.) Archaeology: making of and practice. Studies in honor of Mircea Babeş at his 70th anniversary, 217–237, Piteşti, Institutul de Arheologie ‘Vasile Pârvan’ Bucureşti, Editura Ordessos a Muzeului Judeţean Argeş. Rusu, M. 1969. Das keltische Fürstengrab von Ciumeşti in Rumänien. Bericht der Römisch-germanischen Kommission 50, 267–300. Sordi, M. 2002. Dionigi e il Tirreno. In N. Bonacasa, L. Braccesi and E. De Miro (eds.) La Sicilia dei due Dionisî. 493–499, Roma, « L’erma » di Bretschneider. Stöllner, T. 1998 Grab 102 vom Dürrnberg bei Hallein. Germania 76.1, 67–176. Szabó, M. 1971 Une fibule celtique a Délos. Bulletin de la correspondence hellénique 95, 503–514.

29.  ‘…to boldly go where no man has gone before.’ Dedicated to Ruth and Vincent… Szabó, M. 1989a. Beiträge zur Geschichte des keltischen Drachenpaarmotivs. Communicationes Archaeologicae Hungariae, 119–128. Szabó, M. 1989b. Contribution au problème du style plastique laténien dans la cuvette des Karpates. Acta Archaeologica Academiae Scientiarum Hungarica 41, 2–32. Szabó, M. 1991 Il mercenario. In S. Moscati (ed.) I Celti. 333–336. Milano, Bompiani. Szabó, M. 2008. Styles celtiques à Monte Bibele: les fourreaux décorés. In D. Vitali, and S. Verger (eds.) Tra mondo celtico e mondo italico. La necropoli di Monte Bibele. Atti della Tavola Ronda, Roma 1997, 223–236. Bologna, Università di Bologna. Szabó, M. and Petres, É. F. 1992. Decorated weapons of the La Tène Iron Age in the Carpathian basin, Inventaria praehistorica Hvngariae V, Budapest, Magyar Nemzeti Muzeum. Todorović, J. 1968. Kelti u jugoistočnoj Evropi. Dissertationes Tom. VII, Beograd, Muzej grad Beograda. Treister, M. Y. 1996. The Role of Metals in Ancient Greek History. Leiden, Brill.

273

Verger, S. 1987. La genese celtique des rinceaus a trisceles. Jahrbuch des Römisch-Germanisches Zentralmuseums 34, 287–339. Verger, S. 2003. Des objets gaulois dans les sanctuaires archaïques de Grèce, de Sicile et d’Italie. Comptes-rendus des séances de l’Académie des Inscriptions et Belles-lettres 147.1, 525–573. Verger, S. 2011. Les objets métalliques du sanctuarire de Pérachora et la dynamique des échanges entre mers Ionienne, Adriatique et Tyrerhénienne à l’époque archaïque. In G. De Sensi Sestio and M. Intrieri, (eds.) Sulla ritta per la Sicilia: L’Epiro, Corcire e l’occidente. 19–59. Firenze, Edizioni ETS. Vitali, D. 1987. Monte Bibele tra Etruschi e Celti: Dati archeologiche e interpretazione storica. In D. Vitali (ed.) Celti et Etruschi nell’Italia centro-settentrionale dal V secolo a.C. alla romanizzazione. Atti del colloquio internazionale Bologna 12–14 aprile 1985, 309–380. Bologna, University Press Bologna. Zuffa, M. 1978. I Galli sull’Adriatico. In I Galli e l’Italia, 138–162. Soprintendenza archeologica di Roma, Roma.

30 ART AND CRAFTSMANSHIP IN ELITE-WARRIOR GRAVES: ‘FROM BOII TO PARISII AND BACK AGAIN…’ Nathalie C. Ginoux and Peter C. Ramsl1

Dear Vincent, We all know about your thirst for knowledge and your curiosity, which wasn’t easy for us to deal with sometimes. Your ‘freedom of science’ was not always ours in old Central Europe. But something would have been missing in our field, if there wouldn’t have been your wonderfully sharp remarks. But I should not forget to mention all of your support – as a referee and reviewer for applications and projects, either as a co-writer with articles or as a supporter, in order to put them into major journals. Thank you very much for all of this!!!

(PCR)

Mag auch die Spieglung im Teich oft uns verschwimmen: Wisse das Bild.

To Vincent who knows…

R. M. Rilke (NG)

Celtic warrior elites and mercenaries were the most important actors of this mobility, in particular those who carried a sword and spears who were omnipresent on the European field of operations to the borders of modern Turkey during the entire 3rd century BC, but also in Greece, Sicily and Asia Minor, on the Anatolian plateau. Their omnipresence is attested by funerary archaeological sources, the rapid transmission of Latenian material over long distances, some figurative representations of Gauls, and ancient written sources. Swords and decorated metal scabbards, bronze fittings adorning the chariots and their fittings, fibula and other clothing accessories have formed, since the foundation of research into ancient Celts, the elements on which theories, and, as a result, many of the indicators for retracing the path of these movements, are based. In the following paper, we will be focusing on two comparable cases to shed light on the Iron Age exchange network: weaponry from burials in the Paris basin, and comparable metalwork from two burials at Mannersdorf, Austria.

Introduction Among the main markers of what is called the ‘expansion period’ which progressively extended the sphere of influence of the Latenian culture, first to the Southern Alps, towards Italy from the beginning of the 4th century BC, then towards south-eastern Europe and the Balkans in the course of the following century, are undoubtedly the successive artistic innovations and the apparent diversity of some local variations that have been regrouped and labelled as several ‘styles’ since Paul Jacobsthal’s fundamental book Early Celtic Art was published in 1944 (Jacobsthal 1944).

The Paris Basin Over more than a decade, a sudden increase in archaeo­ logical finds – some of them are really exceptional – in the north-east quarter of France (in particular the Paris basin) deeply changed our vision of the Latenian populations and their integration into the vast exchange and circulation network in inland and Mediterranean Europe. The iconographic repertoire on the weapons, sword scabbards in particular, provides us with a numerically poor

30.  Art and Craftsmanship in elite-warrior graves: ‘from Boii to Parisii and back again…’ but still fundamental record for understanding the Latenian elites’ mobility, from the use of iconographic symbols. In particular, the inter-Celtic motif of the so-called ‘Confronted Dragon Pair’ (De Navarro 1972; Megaw and Megaw 1989; Szabó and Petres 1992; Ginoux 2007) which is omnipresent in its two forms (one-legged small griffins and zoomorphic lyre) among the grave goods dating from the first half of the 3rd century BC in every Parisii cemetery (Ginoux 2009). Thanks to these recent discoveries, the Parisian region is reaching an important position in the history of Celtic art, a position it already holds for the later period, with the production of coinage images on Parisii gold staters, and the small series of chariot fittings, with bronze moulded plastic adornment. Among these is a linchpin with a decorated bronze head and a guide ring, aquired by the Musee des Antiquites nationales in 1907, with the unreliable provenance of Paris (‘Rue Tournefort, 5th district’) but perhaps from Champagne. For a long time this was an isolated find in the Parisian region, leading many authors to question the conditions and location of its discovery (Collection Triantaphillos. Jacobsthal 1944, n. 163, 175; Duval 1977, 115; Megaw and Megaw 2005, 139). In fact, except for the remains unearthed in 1899 in Nanterre (Hubert 1900; Olivier and Schönfelder 2002), nothing had allowed such decorated objects to be placed in the local archaeological context until now, with the discoveries made in Le Plessis-Gassot and Roissy-en-France (Ginoux 2003; 2009; Lejars 2005) complementing other finds published in the late 1970s: a well-made bronze artifact found on the ancient settlement of Le Tillay, from the place known as ‘La Vieille Baune’ opposite the Bouqueval and Le PlessisGassot cemeteries (French department of Val-d’Oise) (see recently Ginoux 2009, 104ff., 135), as well as the pottery bowl (from grave 1) and bronze bracelet (from Chariot grave no. 3) (Guadagnin 1984; Megaw and Megaw 2005, 137). The most important assemblage consists of 25 decorated bronze pieces unearthed from one of the two chariot graves in the ‘La Fosse Cotheret’ cemetery of Roissy-en-France, now on display at the Saint-Germain-en-Laye Museum (Lejars 2005; Emilov and Megaw 2012, 15ff.). This discovery, whose publication is awaited, gives new major arguments in favour of a connection with central Europe, notably with the bronze fitting of the flagon from BrnoMaloměřice (Moravia) (Poulík 1942; Meduna and Peškař 1992 and the most recently Kruta and Bertuzzi 2007) and the famous Celtic bronze chariot fittings found in the Hellenistic tomb of Mal-Tepe in Mezek (Haskovo district, Bulgaria), recently re-examined by Julij Emilov and Vincent Megaw (Emilov and Megaw 2012). What is exceptional about the context of the grave found in Roissy-en-France is the round bronze openwork armature interpreted as the lid of a wooden container; and the ornamental work on the fittings of the ceremonial chariot and particularly the decoration on its yoke. All these pieces of unusual plastic quality form a

275

very coherent iconographic program in which elements are technically and artistically equivalent to the rare artifacts found in the Parisian region and the surrounding area. Some similarities are found in the volumetric conception of certain pieces or parts of pieces from Roissy-en-France, the shield plates from Le Plessis-Gassot and the ‘Parisian’ chariot linchpins. The three-part division of the ornament is based on the volumetric expression of three wrapped S-curves dilated at the midpoint. The spiral volute extremities develop into six large medalllions. The same kind of concept is displayed on top of a ‘Parisian’ guide-ring (Jacobsthal 1944, n. 175; Duval 1977, 107–108ff.; Emilov and Megaw 2012, 14ff.; Ginoux 2009, 130). A cleverly-placed long and almond-shaped eye allows the fleeting evocation of a face. The same process is found on the bronze lattice of the Roissy container, where the principal gyratory motif develops around the central element. The principle of the composition based on S-motifs linking into continuous sequences – one S-motif ending in volute spirals and the other with the head of a fantastic animal – recalls the also spiral composition of the decoration on the Bouqueval bowl. Other similarities can be found between all the pieces, particularly the application of the Métamorphose Plastique process (Kruta 1992, 835, 15ff.) to the decoration of the ‘Paris’ and Roissy linchpin heads, on which a central mask developing from the fusion of a simplified Celtic palmette and foliage compounds anthropomorphic features with vegetal contours (Ginoux 2003, 6ff.; 2009, 130). These pieces have to be compared with several ceremonial objects discovered in the Nogentais-Sénonais area, among which are two iron fibulae from Conflans (Aube) and the bronze and iron fibula found in Villeneuve-la-Guyard (Yonne, in Burgundy) (Rapin and Baray 1999). As well as a debate on their place of manufacture, these objects obviously pose the question of the conditions for the rapid diffusion of this ornamental fashion which appeared without any precedent in this local context. The chronological framework given by the iron brooches, which are the big globular foot type and are the first examples of La Tène II design, is consistent in all points with the homogeneous character of all the ornamented pieces, which belong conceptually to the second phase of the ‘Plastic Style’. In the same period, the production of decorated metallic openwork made from iron and bronze lattices may be consistent with this dynamic of mobility in the formation process of the continental Parisii, in which an influence from the north-western part of the Carpathian basin needs to be acknowledged. The new character and the chronological, geographic and stylistic coherence demonstrated by the archaeological materials, iconography, geographic distribution, the structuring of the cemeteries and the heterogeneity of funeral practices, are only explicable by a long-distance population movement, on an east-west axis, which led to a partial repopulation of the current Île-de-

276

Nathalie C. Ginoux and Peter C. Ramsl

France region in the second quarter of the 3rd century BC. The movement over a long distance of some populations from the Danubian regions and along the river Seine is also attested in the Val de Seine area by the discovery of ankle rings dating from the second quarter of the 3rd century BC (Leman-Delerive 1987). We can also mention, without yet establishing a direct link, the Orval ‘Les Pleines’ chariot grave found on the west coast of the Cotentin peninsula (French department of La Manche, in Lower Normandy), and particularly the plastic ornamentation of the head of a bronze linchpin which represents a mask flanked by two symmetrical figures, like the linchpins from Mezek, ‘Paris’ and Roissy (Lepaumier et al. 2007, 68). To contribute to the still heated discussion about ‘questions of style’, it seems to be an appropriate time to present an interesting, though fragmentary, piece which is situated amongst the series of La Tène scabbard plates with openwork ornamentation, dating from the 3rd century BC. The artefact was found in a small inhumation cemetery, excavated in 1990 in Barbey-sur-Seine ‘Le Chemin de Montereau’ (French Department of Seine-et-Marne), which consisted of 20 graves, of which eight were studied and dated to the middle La Tène Period (Gouge and Griffisch 2008, 269ff.; Ginoux 1993. The material was conserved at the Institute of Conservation and Archaeological and Paleometallurgical Research (now Conservare), Compiègne, France. We cordially thank our colleague, Patrick Gouge (archéologue départemental de Seine-et-Marne) for having given us access to the material and the excavation reports). Unfortunately, we do not have much information about the context of most of the graves – except warrior Grave 8, already published (Ginoux 2002, 294–296, 100ff.) – because of the unexpected circumstances of discovery. Nevertheless, we know that the longitudinal direction of the graves was exclusively northwest. The morphology of Grave 6 seems to be consistent with the general pattern observed here: rectangular in shape with rounded corners. Its dimensions are estimated at about 1.60 m long and 0.50 m wide. Like the other weapon burials in the cemetery, it seems that the individual in Grave 6 was deposited in supine position with his head northward. Only the tibia and fibula bones were preserved (bone preservation across the site was generally poor due to the acid substratum and the burial context). In accordance with local burial customs, the deposits did not contain any ceramics, only metallic ornaments and iron weapons. In Grave 6, each of 11 metallic fragments attest to the deposition of an iron sword in its iron scabbard, the latter decorated with bronze openwork; a shield, of which only a fragment of its external overlap survives, and an iron brooch of unidentifiable typology – only the truncated cone bead on its foot was preserved.

Fig. 30.1. Detail of the chape – fragment no. 1 from Barbey 6 (photo P. Gouge).

Fig. 30.2. Fragments no. 2 and 3 of the scabbard from Barbey 6 (photo P. Gouge).

The decorated fragments The most identifiable part of the weapon belongs to the distal extremity of a sword blade, which is slightly ribbed and still attached to a piece of the front plate. Still in place on the lower edge, a part of the openwork which was either developed or segmented certifies that the decoration originally covered the whole scabbard. Fragment 1 (Fig. 30.1), that is part of the open-ring chape on the obverse, comprises a conical copper-alloy rivet set on the iron front plate of the scabbard. A fragment of the back plate is still visible. Once reassembled during conservation, fragments 2 and 3 (Fig. 30.2) can be attributed without hesitation to the distal part of the weapon, due to the convergence of the sheet and the motif (length: 60 mm). Fragment 4, the final fragment (Fig. 30.3), was on the midrib or near the mouth of the scabbard, which is shown by the width conserved from the remains of the two lateral overlaps, and the midrib’s position (maximum surviving length: 91 mm; surviving width: 28 mm; reconstructed width: 55 mm).

Shaping the metallic ornament and technique All the different motifs identified were created by cutting or casting an openwork copper-alloy plate with a constant

30.  Art and Craftsmanship in elite-warrior graves: ‘from Boii to Parisii and back again…’

Fig. 30.3. Fragments no. 4 of the scabbard from Barbey 6 (photo P. Gouge).

277

to the model of broad weapons highlighted for the latter openwork dragon-pairs (Ginoux 2007, 61 sq). The best comparisons have to be found in France, in Saint-Benoîtsur-Seine ‘La Perrière’ (grave 6), Le Plessis-Gassot (grave 1002), Chens-sur-Léman (Haute-Savoie) dated LT B2 and, eastward, in Sanislau (grave 1, Romania). In conclusion, this fragmented decorated scabbard fits within the production series of metallic ornamented openwork made from iron and bronze lattices dated from the first half of the 3rd century BC, in particular the ones including bird-heads and interlaces (Ginoux 2003, 8ff.) (Fig. 30.4).

Comparable finds from Mannersdorf plano-convex section. This kind of technique differs from that of the Brno-Maloměřice flagon, (Moravia) shaped with a thin ridge (Kruta and Bertuzzi 2007, 69–70ff.). From what can be understood of the iconographic scheme, on the basis of the different fragments, it seems to be organised in the following manner: a composition structured by the (continuous or non-continuous) line of the interlace motifs that join and shape the figured motifs, organise the voids, and thus the light and duotone effects of bronze upon iron. The heads of several imaginary birds appear in a series of patterns based on a ‘Sword Style’ variation. The most conspicuous is characterised by an almond eye underlined by a fine double incised line; the rivet is at the centre of the eye. Most curious is another one, a profile of small bird with a curved beak which recalls bronze bird head motifs from the British Isles, especially Ireland (Raftery 1994, 177, 107ff.). Nails (or rivets) are incorporated in the ornament in a more or less visible manner. The most visible mode places them at the centre of some small slightly convex discs on the extremities of the S-scrolls that shape the interlace segments. Some other rivets, only visible by radiography (X-rays), are cleverly hidden inside the eyes of the imaginary birds. The first method of fixing is not typical of the ornamental technique on sword scabbards dating from the 3rd century BC. Indeed, it is found on other artefacts and most of all on the metallic lattices on the Brno-Maloměřice flagon and, maybe most surprisingly, transposed on an embossed ornament which was not attached to the cauldron found in Brå (East Jutland, Denmark) (Emilov and Megaw 2012, 13c/d ff.). They are precise disc-shaped scrolls closing the foliage shoots which extend the griffin’s double crest moulded on the rim of the container. One can easily recognize on fragment no. 4 (near the mouth of scabbard) the uncluttered shape of a fantastic animal (left part of a zoomorphic lyre) based on S-motifs. The ornament seems to have been set up on the front plate before the completion of the scabbard. The width restored to about 6 cm corresponds

Mannersdorf, grave 117 Grave 117: wide-rectangular shaft with sharp corners and a grave of an armed individual. At a depth of 0.45 m is a 2.50 × 2.00 m wide, SW-NE-oriented grave pit, with a layer of limestone as a grave cover. This grave was within object 14 (surrounding ditch). The burial is placed at a depth of 2.10 m, in a 2.45 × 1.60 m wide grave pit, with teeth and bone fragments suggesting a body buried in an extended position (Fig. 30.5). On the left side of the body ‘was an approximately 1.15 × 0.40 m big shield of organic material’, with only iron fragments of the shield rim surviving. On the right side of the body was an approximately 80 cm long iron sword. On the left side of the skull, a 45 cm long iron spearhead, at the the neck was an iron fibula, at the pelvis four iron rings, as well as a bronze bracelet and a bronze finger ring. In the eastern part of the grave pit, from south to north, were a 24 cm long pair of iron scissors, a 44 cm-long iron knife, a grinding stone, animal bones, a 7.3 cm-high clay pot with three slanting grip-handles, a bowl, a bottle-shaped vessel made of grey clay, another bowl and another bottle-shaped vessel (with markings at the bottom) (Fig. 30.6). The iron sword is of particular importance and is detailed as follows: iron sword with scabbard: tang with oval to deltoid section, rivet on handle tang; Scabbard: roof-like mouth of the scabbard with slightly fluted sides, at the end of the central ridge is a V-shaped chape, crosspiece is 8 cm above the chape; on the rear is a belt fixing; on the front side are two open-worked tendril-ornaments, heavily restored, L. 82.5 cm, L. (tang) 12.2 cm, max.B. 5.6 cm, max.D. (scabbard) 1.9 cm (Fig. 30.7).

Dating The iron fibula nr. 12 unfortunately does not contribute much for dating (Ramsl 2011, pl. 138). But the combination of the hollow rings with the bludgeon-shaped pendants (FNr.

278

Nathalie C. Ginoux and Peter C. Ramsl

1

7

2

3

4

5

6

8

Fig. 30.4. Repertoire: 1; Le Plessis-Gassot (scabbard grave no. 1002), 2: Brunn am Steinfeld (after Jacobsthal 1944); 3, 6–8: Le PlessisGassot (scabbard grave no. 1004); 4: Gournay-sur-Aronde (after Lejars 1994); 5: openwork motif on the spearhead from Le Plessis-Gassot (graphics: N. Ginoux and I. Pasquier, after Ginoux 2003).

9), as well as the spearhead with a broad blade (FNr. 8) generally suggests La Tène B. In this context, the sword Fnr. 6 seems more interesting (Ramsl 2011, pl. 138). The sword has some attributes, such as the tang with deltoidshaped cross-section and the tight-fitting, short ending of the scabbard, which suggests a younger Period (LT C1?).

The open-worked ornaments, which are part of the scabbard, have their counterparts at the necropolis of Plessis-Gassot in grave 1002 and 1004 (Ginoux 2003, 53), which are dated to the 3rd century BC. Thus, one possible suggestion for dating grave 117 could be the final phase of Sub-period LT B2 (B2c?).

30.  Art and Craftsmanship in elite-warrior graves: ‘from Boii to Parisii and back again…’

279

Fig. 30.5. Plan of the cemetery of Mannersdorf (graphics: G. Melzer and P. C. Ramsl) Concerning the openwork decoration on scabbards, especially those with an iron latticework pattern, one can observe the same areas of focus as for the map of openworked spearheads: one in the north-eastern part of Gaul and one in Lower Austria. The example from Barbey highlights this recurrence. In this case, as for the two PlessisGassot adorned scabbards (graves no. 1002 and 1004) and Mannersdorf (grave no. 117) the iconography is strictly related and based on a genuine program for each piece which associated the themes favoured by the warlike ideology from the 5th century BC: the Tree of Life materialised by the simplified Latenian palmette and its two guardians (dragons), the ‘pelta’ (crescent-shaped motif), the ‘S’ motif (Kruta 2010), the mask and the double leaf (Fig. 30.8–30.9). The same program is found on some decorated spears – they are completely standard in most cases – which suggests discoveries in Gaul, specifically in the Paris Basin, could be linked to a military particularity of the Parisii referring to their name’s origin, in the same was as has been suggested for the Gesates (suggested by Pierre-Yves Lambert, Kruta 2002, 12–13; Ginoux 2009, 124).

Mannersdorf, grave 180 Approx. rectangular shaft with the grave of a male (19–30 y.) At a depth of 1.30 m is a 2.60 × 1.25 m wide, NNE-SSWoriented grave pit, with several layers of limestone slabs at a depth of 0.90 to 1.30 m. At a depth of 1.51 m is a 2.40 × 1.10 m wide grave pit, oriented to the south and north and, along the walls, there are some limestone slabs. Burial: SSW-NNE-oriented inhumation in an extended supine position. The skeleton was in the western part and the grave goods in the eastern part of the grave pit. To the left of the skull was a 50 cm long, leaf-shaped iron spearhead with various open-worked ornaments. On the left femoral head was a bronze bracelet, on the right side of the pectoral were two iron fibulae, and between the right side of the body and right arm bones was an approx. 85 cm long iron sword with a scabbard and an iron ring. The grave goods located in the eastern part of the grave pit from S to N are as follows: bones of an animal, an iron knife with scissors, iron and organic fragments, a reddishbrown, broken bowl, a dark-grey clay pot, a dark-grey, tall clay pot, a red-brown clay pot and several iron rolls (shield rim) (Fig. 30.10).

280

Nathalie C. Ginoux and Peter C. Ramsl

Fig. 30.6. Plan of grave 117 from Mannersdorf (graphics: G. Melzer and P. C. Ramsl).

An artifact of particular importance An iron spear-head consisted of a short socket with rivet, broad leaf-shaped blade with reduced tip, a strong central ridge, and on the blade are complex open-worked ornaments of floral motifs and rosettes; L. 46 cm, L. (socket) 5 cm, max.B. 14.9 cm, D. 0.25 cm, diameter (socket) 1.9 cm (Fig. 30.11).

Fig. 30.7. Scabbard with open worked ornament from grave 117 – Mannersdorf (graphics: M. Imam).

Dating Dating primarily concerns the two iron fibulae with large spherical ornamentation at their feet (Ramsl 2011, pl. 184/8a and 8b). These are most likely dated to LT B2a. Also, the two-part shield boss with rivets with big heads suggests the same. Furthermore, there is an open-worked iron standard.

This type of spearhead is of a shape consistent with Moravia, however, we know the ornaments from France. A good comparison is known from Plessis-Gassot 1004, dated to the first decades of the 3rd century BC, which is LT B2c in Middle Europe. The situla-shaped pottery points to a

30.  Art and Craftsmanship in elite-warrior graves: ‘from Boii to Parisii and back again…’

281

Fig. 30.8. Detail of the first metallic ornament on Le Plessis-Gassot’s scabbard 1004 (photo P. Gardin).

Fig. 30.9. Detail of the second metallic ornament on Le Plessis-Gassot’s scabbard 1004 (photo P. Gardin).

Fig. 30.11. Open worked spearhead from grave 180 – Mannersdorf (graphics: M. Imam)

However, the more pronounced end of the scabbard suggests a placement at the beginning of LT B2. In summary, dating to the end of LT B2a is proposed.

Iron open worked ornaments (ajourés)

Fig. 30.10. Plan of grave 180 from Mannersdorf (graphics: G. Melzer and P. C. Ramsl).

late phase of LT. B2. The approx. 45 cm high bottle-shaped vessel (Plate 186/11) is the largest of its kind in the cemetery and shows the importance of the person in grave 180.

The first point it is to see the different kind of techniques 1 Incised or punched ornaments 2 Forged ornaments Because of the varied kind of preservation of iron artifacts, it is usually not possible to decide if the ornaments are incised or forged. Thus, flat pieces with a wide rim around the cutout are seen as incised, those with small ridges are seen as forged. Of course, it is to be assumed that tools such as files and saws were also used.

282

Nathalie C. Ginoux and Peter C. Ramsl

Fig. 30.12. Distribution map: Scabbard with open worked ornaments and shield bosses with open worked wings (graphics: P. C. Ramsl).

Fig. 30.13. Distribution map: open worked spearheads and spearheads with reduced tips (graphics: P. C. Ramsl).

Openwork ornaments in iron have –as well as those of bronze – a long tradition in the Iron Age. It begins with the ‘openworked belthooks’ of LT A (e.g. Roseldorf in Sievers, Urban and Ramsl 2011, 1610), continues with the openwork ornaments on spearheads and shield buckles in LT B2/C (e.g. Szob in Szabó and Petres 1992, pl.62) and ends with the opus interrasile of late La Tène scabbards (e.g. Lamadelaine in Łuczkiewicz 2006, 189ff.). If we have a look at the distribution map of scabbards with openwork ornament, we can see a main area in the West

(Champagne, Picardie,…) and in the East (Lower Austria, NW-Romania). In between there are single find points in Switzerland and Bavaria (Fig. 30.12.) A similar, but slightly different situation is shown by the distribution of shield bosses with openwork wings. Here, there is a main area in the east – from Lower Austria to Slovakia and NE-Hungary with some single find spots in the west. At this point it should be mentioned, that in this case, maybe a lot of the openwork ornament is possibly missing because of corrosion.

30.  Art and Craftsmanship in elite-warrior graves: ‘from Boii to Parisii and back again…’

283

Table 30.1. List of scabbards with open worked ornaments. No. 29 30 28 26

France

13 . . . . 25 23 22 . 27 24 20

Middle Danube / Swiss Carpathian Basin

21 19 16 14 17 15 18

Finds

Iron

Barbey-sur-Seine T. 8 Chens-sur-Léman Sp 88 Chouilly « Les Jogasses » T. 1 Ensérune T. 163 Gournay-sur-Aronde 2666 Gournay-sur-Aronde 1860/ 2651 Gournay-sur-Aronde 2624 Gournay-sur-Aronde 1863–4 Gournay-sur-Aronde 1826 Meaux Nantes Plessis-Gassot T. 1002 Plessis-Gassot T. 1004 Saint-Benoît-sur-Seine T. 8 Villeperrot T. W25 Ering T. 1 Gumefens/Sus Fey T. 2 Manching Steinbichel T. 26 Guntramsdorf T. 2 Mannersdorf T. 117 Pișcolt T. 40 Pottenbrunn T. 562 Sanislău T.1

Co. All. X

11 12 13

Site Potzneusiedl Neunkirchen Halmajugra Bodroghalom Kosd (T. 25) Szob (T. 1) Magyarszerdahely (T. 63) Piliny Ižkovce (T. 10) Ludas (T. 1057) Ribemont-sur-Ancre Gourgançon “Saint Mard” (T. 9) Gournay 1215

Dating

Literature

LT B2/C1

Unpublished.

X

LT B2

Landry, Blaizot 2011.

X

LT A

X

LT C1

Hatt, Roualet 1976 ; Les Celtes en Champagne 1991, n° 83, p. 86 ; Ginoux 1994 Pl. I/3; cat 13. Rapin, Schwaller 1987 ; Ginoux 2007 cat. 27.

X X

LT C1 LT C1

Lejars 1994 p. 221. Lejars 1994 p. 72, 199.

LT C1

Lejars 1994 p. 66. 219.

X

LT C1

Lejars 1994 p. 65.

X

LT C1

Ginoux 1994 Pl X/4 ; cat . 40 ; Lejars 1994 p. 193.

LT B2 LT B2/C1 Late LT B2

Ginoux 1994 Pl X/3 ; cat . 45. Ginoux 1994 Pl X/1 ; cat . 47 ; Lejars 1994 p. 65 n° 8. Ginoux 2003 ; 2009.

Late LT B2

Ginoux 2003 ; 2009.

LT C1 LT B2

Bienaimé 1989, p. 62–63 ; Ginoux 1994 Pl. XVI. Bulard 1979 p. 38, fig. 1/1, 5/3, 6/2.

X

LT B2

Krämer 1985 p. 147, fig. 20, pl. 78, 112.

X

Late LT C1

Schwab 1984 ; Jud 2009, Abb. 2.

X

Late LT B2

Krämer 1985, Abb. 12/2.

LT B2/C1

De Navarro 1972 pl. CXXVIII ; Urban/Teschler/Schultz 1985, Abb. 17/4–6. Ramsl 2011, Taf. 138/6.

X X

X X

X X

X

X X

LT B2

X

LT B2

X X

Table 30.2. List of shield bosses with open worked wings. 1 2 3 4 5 6 7 8 9 10

gold

Literature Sauer 2007, 39. Caspart 1929, Abb. V/3. Hellebrandt 1999, Pl. LVIII/10. Hellebrandt 1999, Pl. LXII/1. Rapin 1988, Pl. 38/1. Szabó/Petres 1992, Pl.62. Rapin 1988, Pl. 38/3. Hunyady 1942, Taf. XLIX/5. Vizdal 1976, 166. Szabó (ed.) 2012, Pl. XLIV/2, XLV/4). Lejars 1998, fig.3. Brisson/ Loppin 1938. Rapin 1988, Pl. XII.

X

LT B2/C1 LT C1

Nemeti 1975 p. 96, pl. VI-VII ; Rustoiu 2012,1487. Ramsl 2002, Taf. 65/9. Zirra 1971, p. 190, fig. 8.20, p. 221, fig. 22 ; Szabó, Petres 1992 n° 98, pl. 95.1.

But we can once more see again the area of the ‘Middle European Corridor’ in the Late Iron Age as a road of connection across Europe. Looking at the map with the openwork spearheads we can see a main area of lances with reduced tip in the east (Lower Austria, Moravia, SW-Slovakia, W-Hungary), but spear heads with openwork ornamentation more in the West (Fig. 30.13) Anyway, at the moment it is senseless to ask about the origin of this technique or which workshop starts with it – East or West. Chronology is too weak to answer these questions and the question should be asked in another way. All these phenomena show an Iron Age exchange network of art styles and technique. It is the mobility of ideas – and that of people, whether they are craftsmen, warriors or others (Table 30.1, Table 30.2, Table 30.3).

Nathalie C. Ginoux and Peter C. Ramsl

284

Table 30.3. List of open worked spearheads and spearheads with reduced tips. Site 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19

Velké Číčovice Brno-Chrlice (9) Kobylnice (9) Křenovice (11) Horny Játov (111) Hurbanovo Bacherov-Majer (4) Mannersdorf/Lgb. (180) Halmajurga Káloz-Felső Törzsök Langengeisling (2) Bauernbauch-Bretten Gumefens (2) La Tène Marolles-sur-Seine (2012) Plessis-Gassot (1004) Plessis-Gassot (1004) Barbuise-Courtavant Fère-Champenoise (7) Fère-Champenoise (8) Nogent-sur-Seine Villevenard

Reduced Tip X X X? X? X X X X? X?

Openworked ornament (simple)

X X? X X

X X X X X

Acknowledgements We would like to thank Jennifer Douétil (NG contribution) and Micheline Welte (PCR contribution) for translating our words into English.

Note 1

Openworked ornament (complex)

The author is funded by the Austrian Science Fund (FWFProject P23517-G19)

Bibliography De Navarro, J.-M. 1972. The finds from the site of La Tène. Volume I: scabbards and the swords found in them, Part I: text, XXVI + 353, p. 37 fig; Part II: catalogue and plates, XVI, p. 357–456: CXLVIII. London: Oxford University Press for the British Academy. Duval, P.-M. 1977. Les Celtes. Paris: Gallimard. Emilov, J. and Megaw, V. 2012. Celts in Thrace? A Re-Examination of the Tomb of Mal Tepe, Mezek with Particular Reference to the La Tène Chariot Fittings, Archaeologia Bulgarica, XVI.1, 1–32. Guadagnin, R. 1984. La nécropole Celtique de Bouqueval, Jeunesse Préhistorique et Géologique de France 8, 12–65. Ginoux, N. 1993. Parure et ornementation des mobiliers métalliques des nécropoles de La Tène ancienne de Marolles-sur-Seine «le Carreau Franc», «le Parc Saint Donain», Barbey «le Chemin de Montereau, étude et rapport. Dammarie-les-Lys: Conseil Général de Seine-et-Marne. Ginoux, N. 2002. Ajuar de la sepultura no 8 de Barbey-surSeine. In: Torques belleza y poder, Museo Arqueologico

X X? X X X X X X

Lotus

X

Dating

Lt B2 Lt C Lt B2 ? LT C (?) Lt B2(b) Lt B2a Lt C(2)

X

Lt B2 Lt B2 (~ 300) Lt B2(?), C

X X

Lt B2 Lt B2

Lit. Sankot 1991 Čižmářová 2011 Čižmářová 2004 Čižmářová 2009 Benadík 1957, Taf. I/11 Benadík 1957, Obr. 19/2 Ramsl 2011 Hellebrandt 1972 Szabó/Petres 1992 Krämer 1985 Ludwig 2012 Jud 2009 Vouga 1923 Ginoux 2007 Ginoux 2003 Ginoux 2003 Chevallier 1956 Celtes en Champagne Celtes en Champagne Joffroy-Thénot 1976 Charpy 1996

Nacional, septiembre–Diciembre 2002, 294–296. Madrid: Museo Arqueológico Nacional. Ginoux, N. 2003. L’excellence guerrière et l’ornementation des armes aux IVe et IIIe siècles av. J.-C: découvertes récentes, Etudes Celtiques 35, 33–67. Ginoux, N. 2007. Le thème symbolique de «la paire de dragons» sur les fourreaux celtiques (IVe– IIe siècles av. J.-C.). Etude iconographique et typologie, BAR International Series 1702, Oxford, British Archaeological Reports. Ginoux, N. 2009. Elites guerrières au Nord de la Seine au début du IIIe siècle av. J.-C. La nécropole celtique du Plessis-Gassot (Vald’Oise). Lille: Revue du Nord, Hors-série Art et Archéologie 15. Gouge, P. and Griffisch, J.-N. 2008. Barbey. In J.-N. Griffisch, D. Magnan, and D. Mordant (eds.) La Seine-et-Marne, Carte archéologique de la Gaule, 269–273. Paris: Académie des Inscriptions et Belles-Lettres. Hubert, H. 1900. Sépulture à char de Nanterre. In Comptes-rendus du Congrès international d’anthropologie et d’archéologie préhistorique, XIIe session, 410–17, fig. 1–11. Paris: Masson et Cie. Jacobsthal, P. 1944. Early Celtic Art. Oxford: Clarendon Press. (Reprinted with corrections, 1969) Kruta, V. 1992. Brennos et l’image des dieux: la représentation de la figure humaine chez les Celtes, Académie des inscriptions et belles-lettres. Comptes rendus des séances de l’année 1992, Novembre–Décembre, (CRAI), 821–43. Paris: De Bocard (paru en 1994). Kruta, V. 2002. Les origines des Parisii, Commission du Vieux Paris, Procès – verbal de la séance du 8 janvier 2002, encart au Bulletin municipal officiel de la Ville de Paris no 29 du vendredi 12 Avril 2002, 7–14. Kruta, V. 2010. La question de l’art géométrique des Celtes, Ktèma 35, 243–53.

30.  Art and Craftsmanship in elite-warrior graves: ‘from Boii to Parisii and back again…’ Kruta, V. and Bertuzzi, D. 2007. La cruche celte de Brno. Dijon: Faton. Leman-Delerive, G. 1987. Anneaux de cheville laténiens en HauteNormandie, Etudes Celtiques 24, 33–41. Lepaumier, H., Chanson, K. and Giazzon, D. 2007. La tombe à char d’Orval, présentation préliminaire, Association Française pour l’étude de l’âge du Fer, Bulletin no 25, 69–71. Lejars, T. 1994. Gournay III, les fourreaux d’épée. Le sanctuaire de Gournay-sur-Aronde et l’armement des Celtes de La Tène moyenne. Paris: Errance. Lejars, T. 2005. Le cimetière celtique de La Fosse Cotheret, à Roissy (Val-d’Oise) et les usages funéraires aristocratiques dans le nord du Bassin parisien à l’aube du IIIe siècle avant J.-C. In O. Buschenschutz, A. Bulard, and T. Lejars (eds.) L’âge du Fer en Île-de-France, Actes du XXVI) colloque de l’Association Française pour l’Etude de l’Âge du Fer, Paris et Saint-Denis 9–12 mai 2002, 26e, 73–85. Tours: Supplément à la Revue Archéologique du Centre de la France. Łuczkiewicz, P. 2006. Uzbrojenie ludności ziem Polski w młodszym okresie przedrzymskim, Archaeologia Militaria II. Lublin: Inst. Archeologii. Meduna, J. and Peškař, I. 1992. Ein latènezeitlicher Fund mit Bronzebeschlägen von Brno-Malomĕřice (Kr. Brno-Stadt), Bericht der Römisch-Germanischen Kommission 73, 182–267, pl. 32–40. Megaw, J. V. S. and Megaw, R. 1989. The Italian Job: Some implications of Recent Finds of Celtic Scabbards decorated with Dragon-pairs, Mediterranean Archaeology, 1989.2, 85–100.

285

Megaw, J. V. S. and Megaw, R. 2005. L’art de la Celtique. VIIe siècle av. J.-C. – VIIIe siècle apr. J.-C. Paris: Errance. Olivier, L. and Schönfelder, M. 2002. Nanterre (Hauts-de-Seine): un char de parade de La Tène Moyenne. In V. Guichard and F. Perrin (eds.) L’aristocratie celte à la fin de l’âge du Fer (IIe s. av. J.-C., Ier s. ap. J.-C.). Actes de la table ronde des 10 et 11 Juin 1999, (Glux-en-Glenne – F58), 113-18. Glux-en-Glenne: Centre archéologique européen du Mont-Beuvray, Collection Bibracte 5. Poulík, J. 1942. Das keltischeGräberfeld von Brünn-Malmeritz, Zeitschrift des mährischen Landesmuseum, N.F. II, 49–86. Ramsl, P. C. 2011. Das latènezeitliche Gräberfeld von Mannersdorf am Leithagebirge, Flur Reinthal Süd, NÖ, Mitteilungen der Prähistorischen Kommission 74. Wien: Verlag der Österreichischen Akademie der Wissenschaften. Raftery, B. 1994. Pagan Celtic Ireland. The Enigma of the Irish Iron Age. London: Thames & Hudson. Rapin, A. and Baray, L. 1999. Une fibule ornée dans le „Style Plastique“ à Villeneuve-la-Guyard (Yonne), Gallia 56, 415–26. Sievers, S., Urban, O. H. and Ramsl, P. C. (eds.) 2012. Lexikon zur keltischen Archäologie, Mitteilungen der Prähistorischen Kommission 73. Wien: Verlag der Österreichischen Akademie der Wissenschaften. Szabó, M. and Petres, E. 1992. Decorated Weapons of the La Tene Iron Age in the Carpathian Basin. Budapest: Magyar Nemzeti Múzeum.

31 ASCOT HATS: AN IRON AGE LEAF CROWN HELMET FROM FISKERTON, LINCOLNSHIRE? Andrew Fitzpatrick and Martin Schönfelder

Introduction The River Witham in eastern England has yielded some of the finest and earliest pieces of Iron Age metalwork in these islands. The best known of these finds are the bronze facing of the Witham shield and the relief-decorated scabbard mounts of the ‘Witham sword’, finds that have been considered many times not only by Vincent and Ruth Megaw (e.g. 2001), but also by Vincent’s teacher, Stuart Piggott (1950; 1959; Atkinson and Piggott 1955). In recent years these finds from the River Witham have been widely interpreted as votive offerings, gifts to the gods of the water. In 1980 one of the platforms from which these offerings were made was discovered. Around it were some of the earliest examples of insular Celtic art yet found.

The timber jetty or causeway at Fiskerton In June 1980, whilst metal detecting, Vernon Stuffins noticed two lines of timber posts weathering out of the bank of the North Delph drain that flanks the northern side of the River Witham at Fiskerton, 7 km east of Lincoln (Fig. 31.1). He found, amongst other things, in the vicinity of the posts a La Tène I sword in its iron scabbard; the relief-decorated and coral-inlaid anthropomorphic bronze handle of a second La Tène I sword (possibly a short sword), and part of the shoulder of third sword. The subsequent excavation by Naomi Field in 1981 showed that the timber causeway or jetty at Fiskerton projected across the marshes on the north bank of the Witham and out into the modern main course of the river (Field and Parker Pearson 2003). No traces of the structure have been

found on the opposite (south) side of the river, but work on the north side has shown that the structure was at least 160 m long. Only a relatively small length of the structure (some 20 m) was excavated in 1981 (Field and Parker Pearson 2003) (Fig. 31.2). Another short length was examined in 2001 immediately to the south of the 1981 excavations, though these excavations are not yet fully published (Palmer-Brown and Rylatt 2002; Field et al. 2003, 22–6, fig. 4). On the basis of the samples from the 1981 excavations, the trees used in the structure were felled over a period of at least 135 years. The first trees were felled in 457/6 BC, and the last timber was added after 321 but before 319 BC. Most of the felling dates are between 406–388 BC with very few timbers being added after 350 BC. The great majority of the felling dates obtained for the timbers found in 2001 match those found in 1981 (Field et al. 2003, 28). A series of objects were found in the sediments around the structure in 1981. Many of the finds came from a layer (32) that sealed three holes in the sediments next to three upright timbers that were felled between 359–317 BC. The holes (‘eddy holes’) had been created by water erosion of the sediments that had formed around the timbers. The three timbers therefore offer a terminus post quem of 359–317 BC for the formation of layer 32. In the excavation report the wooden structure at Fiskerton is interpreted as a trackway that was laid directly onto the marsh peats rather than as a walkway or jetty that was raised above the marsh. This was because of the small number of worked logs and planks that could be from a superstructure, and the lack of angled bracing posts of the sort found at the site of La Tène, Canton Neuchâtel. It was suggested that

31.  Ascot hats: an Iron Age leaf crown helmet from Fiskerton, Lincolnshire?

287

Fig. 31.1. Location and excavation plan of the 1981 excavations of the Fiskerton causeway, after Field and Parker Pearson 2003. Reproduced by permission of Lindsay Archaeological Services.

288

Andrew Fitzpatrick and Martin Schönfelder

the trackway was held in place by some of the wooden pegs that were found. It is, however, perhaps more likely that the structure was a raised walkway. Prehistoric trackways and walkways across European marshes are usually of a quite different construction. They were either of wicker hurdles laid on heaped brushwood or, less frequently, corduroy roads laid on raised rails, as at Corlea, Co. Longford, in Ireland (Raftery 1996). No significant quantities of brushwood, and certainly no rails, were found at Fiskerton, which suggest that the structure was a raised walkway. The rarity of decking planks found at Fiskerton is probably to be explained by them having either having rotted away or simply floated away. The absence of angled bracing posts is because the structure at Fiskerton stood in the still waters of a marsh or a river backwater, and not in a main river channel or a lake that had to withstand the high energy environments created by Alpine melt and flood waters. Colin Palmer-Brown and Jim Rylatt also suggest that the structure was a causeway, noting that the wooden pegs are smaller than the holes in the planks that they are suggested to have held in place (2002, 38). The discovery of two, possibly three, log boats also suggests the presence of flowing water nearby.

Votive Offerings The weapons found at Fiskerton in 1981 were reported on with typical thoroughness by Ian Stead (2003a; 2006, 23, 27). In addition to the three swords found in metaldetecting, a further three swords were found; another La Tène I example (Small Find 222), a short sword (Small Find 149), and a fragmentary blade and handle (Small Find 429). There were 11 iron spear heads, a shield mount decorated with red glass (Small Find 3), and a decorative circular fitting (Find 409) that may be compared with the terminal mounts on the Witham shield. A few bronze rings could be from sword suspension systems, and some pieces of binding could be from shields. Fifty-five bone points, made mainly from sheep’s tibia, are interpreted as spear tips, largely on the basis of examples still with their wooden shafts from the Hjortspring and Krogsbølle pre-Roman Iron Age (4th to 2nd century BC) deposits in Denmark (Kaul 1988). At Fiskerton almost all of the weaponry, including the majority of the bone points, was found to the west of the jetty. In contrast, an important group of ironworking and woodworking tools was found to the east of the jetty. The tools find includes some of the earliest examples of La Tène decoration in Britain and Ireland. The antler handle of an iron file has a simple Waldalgesheim scroll at its base, while a fragmentary saw has traces of curvilinear decoration on both sides of the blade (Stead 2003b, fig. 4.15, 1 and 3; fig. 4.16–17). The importance of these finds for the chronology of British Celtic art was quickly seen

(e.g. Megaw and Megaw 1991, 291–4, fig. 6–6b; 2008, 48–9, fig. 3.8). Other finds included a timber that might be from a boat, and two complete pots, which were found amongst the timbers of the jetty. A small number of animal bones, possibly, though not certainly, of Iron Age date, are suggested to represent the deposition of the remnants of meals. A small number of bones from bird’s wings were also found (Field and Parker Pearson 2003). Three fragments of human bone were found: the left parietal bone with a sword wound which may have been inflicted peri-mortem or post-mortem; a fragment of a left tibia; and part of a right femur. All three bones were from adults, with the skull and femur possibly from males. Another sword was found in the 2001 excavations, making seven in all (Stead 2006, 203, no. A282), along with what is described as a dagger that is missing its handle; two iron spearheads – one complete with its ash shaft; two bone points; an iron axe with its handle; two bronze roundels; some u-shaped bronze bindings and some decorated bronze strips; an iron file with its antler handle; and a spindle whorl with its wooden spindle. Two dugout canoes were also found, one of which may have been almost unused and was positioned across the causeway and pinned in place. The skeleton of young pig was found next to the iron axe and it is suggested that it may be a foundation deposit, though the axe need not be as early as the 8th or 7th century date suggested (Palmer-Brown and Rylatt 2002, 37, fig. 3; Field et al. 2003, 25, fig. 7–9). Most of the objects are described as being found amongst the timbers of the structure, but the sword and dagger are described as lying to the west of it and the spindle whorl to the east, which is consistent with the distribution recorded in 1981. On the basis of the 1981 excavations, Fiskerton has been compared to the lake at Lynn Cerrig Bach, the bog at Lisnacrogher, Co. Antrim, and a course of the Nene at Orton Meadows in these islands, and to La Tène and Port in Switzerland, all sites at which Iron Age weaponry was placed in watery contexts (Field and Parker Pearson 2003, 179–88; Macdonald 2007, 174–89). However, as the typological dates suggested for the objects at Fiskerton (especially the weaponry) were later than the felling dates of the causeway, it was suggested that votive deposition only started there when the causeway was up to a century old.

An Iron Age helmet? While most of the Iron Age objects found in the 1981 excavation were of well-known types, a few bronze objects remained unidentified. Amongst them from Iron Age layer 32 was a hollow, almost tubular, S-shaped object made from two pieces of bronze riveted together (Small Find 208) (Fig. 31.3). The ‘tube’ defines two open lobes, one of

31.  Ascot hats: an Iron Age leaf crown helmet from Fiskerton, Lincolnshire?

289

Fig. 31.2. View of the excavations in 1981 looking east. The North Delph ditch is to the right of the excavation and the River Witham is behind the high bank. Reproduced by permission of Lindsay Archaeological Services

which is larger than the other. The object is 241 mm long and 96 mm wide. It has a lentoid or diagonal profile, some 10–14 mm deep, apart from at the lobe at the base of the ‘S’. Here the object widens to form a small, teardrop-shaped recessed area. There are holes in the centre of both faces of the teardrop, through which a rivet may have passed helping to fix the object to another one (though Ian Stead thought this unlikely (2003, 59). The rivet may also have secured coral or red glass mounts to the object. There is a small (c. 2 mm wide) flange along the outside edge of the object and along the inside of the smaller open lobe. There are rivets at regular intervals along both these flanges. On the inside of the larger open lobe, the two pieces of metal abut each other: there is no flange and there are no rivets. In the absence of evidence of use wear and of any parallels, Ian Stead concluded of the object that ‘its purpose is unknown’ (2003a, 59). Although it was published in 2003, Ian Stead’s report was written in the early 1980s. This was before the publication of Barry Raftery’s work on the technically similar hollow rings that were often used as parts of baldrics to suspend swords (Raftery 1988). And it was before the publication of the gold and coral decorated iron helmet found in 1981 at La Grotte des Perrats at Agris, Charente.

Since then, the stunning La Tène A/B1 early to mid 4th century parade helmet has become one of the most famous pieces of Celtic art (Gomez de Soto 1986; Eluère et al. 1987; Ducongé and Gomez de Soto 2007; Gomez de Soto and Verger 2010), omnipresent in exhibitions from Venice to Bern (Megaw 2007) and in standard works on Celtic art (e.g. Megaw and Megaw 2001) (Fig. 31.4, a). But while the artistic aspects of the Agris helmet have been well studied, the definitive technical publication is yet to appear. That publication will include the ‘loose’ elements of the cheek pieces, the knob on the top of the helmet, and pieces that may be grouped, rather loosely, as belonging to ‘the helmet crest’. Directly above both cheek pieces, a line of four rectangular holes runs up the side of the helmet. These holes indicate that the helmet had a crest of some sort, and their size suggests that it needed substantial support. Some elements that must belong to this crest, and which were called ‘external pieces’ by José Gomez de Soto (1986) in one the first publications of the find, include at least two fragments of hollow iron tube (Fig. 31.4, b). Like the rest of the helmet and the cheek pieces with their serpent-like ‘S’ motifs, they are covered with gold leaf. The larger of the two pieces that have been briefly published so far is 77 mm long and 68 mm wide,

290

Andrew Fitzpatrick and Martin Schönfelder a

b Fig. 31.3. Small Find 208 from Fiskerton, after Field and Parker Pearson 2003. Reproduced by permission of Lindsay Archaeological Services.

and it encloses an open S-shaped lobe in the form of ‘un enroulé en ying-yang ouvert’ (Gomez de Soto 1986, 180, colour plate VI, 1; Eluère et al. 1987, 15, fig. 7, 8–9). This hollow lobe is very similar in shape, size, and in form (i.e. hollow) to the Fiskerton object. The outline of these S-shapes recalls the famous leaf crowns or Blattkrone that are known from many representations, both figural and sculptural, across Europe (cf. Gomez de Soto and Verger 2010). The mounts from Agris, and by analogy that from Fiskerton, may therefore be identified decorative mounts or crests from helmets. By comparison with the Agris helmet, the Fiskerton mount is likely to date to the earlymid 4th century BC. The Fiskerton mount was presumably supported in a frame of some sort, though whether the helmet was of leather or metal is not known. It is not yet clear how the Agris mounts were attached to the iron helmet. One of the most famous of all Blattkronen is the one on

Fig. 31.4. The Agris helmet (a) and the external fragments (b). Photographs courtesy of the Römisch-Germanischen Zentralmuseum Mainz; Photo V. Iserhardt, RGZM.

the statue that surmounted the Glauberg tumulus in Hesse in the second half of the 5th century BC (Megaw 2003; 2007, 439) (Fig. 31.5). A key aspect of the Glauberg site is the apparently precise correspondence between the objects found in the grave and those represented on the sandstone statue. This is best seen on the gold neck ring, bracelet and finger ring where the actual objects and the sculpted representations are all but identical (Frey and Hermann 1997; Frey 2002). Where organic objects did not survive

31.  Ascot hats: an Iron Age leaf crown helmet from Fiskerton, Lincolnshire?

291

a

b

Fig. 31.6. The iron rods from the grave at Glauberg, Hesse (a) and their reconstruction (b). After Fröhlich 2006, Abb. 6 and 8. Photograph courtesy of hessenARCHÄOLOGIE; Drawing P. Rispa, hessenARCHÄOLOGIE, photo R. Fröhlich, hessenARCHÄOLOGIE. Fig. 31.5. The statue from Glauberg, Hesse. Height 1.86 m. Photograph courtesy of hessenARCHÄOLOGIE; Photo U. SeitzGray.

in the grave, the statue can be taken as reliable evidence that they were originally present. Such objects include the cuirass of leather and linen, and the helmet. As the burial and grave goods were lifted en bloc in the field and then micro-excavated in the laboratory, it was only after the initial publications that some curved iron rods were found at the right shoulder of the dead man and next to the sword handle. The iron rods have been interpreted as the frame

supporting the lobe or tear-shaped ‘leaves’ that were attached to the (presumably) leather helmet (Fröhlich 2006) (Fig. 31.6). This arrangement is simpler than that seen at Agris. The origins of the Blattkrone clearly lie in those most common of motifs in early Celtic art, the palmette and the lyre (Jacobsthal 1944). These motifs and their derivatives occur on many types of objects, including the decorative bronze bands applied to the bases of iron helmets in France and Italy. They also occur on the bronze plaques found on the sides of a series of slightly later La Tène B1 iron helmets found in Italy. Such plaques are best known on the helmet said to be from

292

Andrew Fitzpatrick and Martin Schönfelder

‘Umbria’ in the Lipperheide Collection in Berlin (Schaaff 1974, 182, no. 15, Abb. 28, Taf. 74–5; 1988, no. 103, 514, Abb. K103, a, f) (Fig. 31.7), but they are now represented by several recent finds from north Italy, including four examples from Monte Bibele – graves 14, 85, 116 and 120 (Vitali 2010, 36–7, Abb. 5). As well as the lyre motifs on the bronze cheek pieces, the principal decorative motif on these plaques is the lyre, which sometimes harbours elusive faces. Lyres are also present on several of the cheekpieces, though on the Gottolengo helmet (prov. Brescia) they are transformed into birds or dragons with long beaks or snouts (Rampinelli 1968). These decorative helmet mounts may reflect the transferral of a free-standing decorative element on helmets, such as a helmet crest, to an applied plaque. The transferral of decorative motifs between different elements of the weapon panoply has been suggested on a number of occasions (e.g. Verger 2003; Fitzpatrick 2007a). The lyres on the helmet plaques may reflect the transferral of motifs on the same object, in this case from the crest to the cheek piece. The tubular sockets on the broadly contemporary (La Tène B1) Canosa di Puglia helmet which was found in Apulia, well to the south of the main areas of Celtic settlement in Italy, are usually suggested to have held feathers (Schaaff 1988, no. 104, 516–18, Abb. K 104; Mazzoli 2010) in imitation of contemporary Apulian-Corinthian helmets.

From the waters of the Witham If the identification of the S-shape object from Fiskerton as part of an Iron Age helmet is correct (while mindful that the iron ‘spike’ which long surmounted the reconstruction of the bronze helmet from grave 12 at San Paolino di Filottrano has recently been shown to be part of a candelabrum... (Schönfelder 2010, 21)), it should not occasion too much surprise.1 While not as numerous as the Iron Age finds from the River Thames, the quality of the individual Iron Age objects from the Witham and their early date in British Celtic art is noteworthy (Fitzpatrick 1984, 179–81). The parallel and contemporary development of British Celtic art with that of Continental Europe is increasingly assured (e.g. Rapin 2002, 56; Garrow et al. 2009). With the parallel development of La Tène type brooches in Britain well established (Hull and Hawkes 1987; Megaw and Megaw 2008, 47) there is little reason to maintain a depressed chronology for Celtic art in these islands. At Fiskerton, very few timbers were added to the excavated length of the causeway after 350 BC. If the purpose of the walkway was not to cross the water but to get over or onto it, then the votive offerings may be argued to be broadly contemporary with it: a slightly earlier date than Vincent and Ruth would suggest for the objects with Waldalgesheim-related decoration (2008, 53).

Fig. 31.7. Copy of the ‘Umbria’ helmet in the Antikensammlung Staatliche Museen zu Berlin. Photograph courtesy of the RömischGermanischen Zentralmuseum Mainz; Photo R. Müller, RGZM.

In addition to the three La Tène I (B2/Ic) swords or short swords found at Fiskerton recently, at least four of the other nine or 11 Iron Age swords known from the River Witham are likely to be of similar date and to be some of the earliest La Tène swords in these islands (Stead 2006, no. 1, 36 and 38–9). Many of these items were found during the scouring and embanking of the Witham below Lincoln in the 18th and 19th centuries (Field and Parker Pearson 2003, 148, 162–4). It has been suggested that some of the swords attributed only to the ‘River Witham’ may have been found at Fiskerton, which is only 7 km east of Lincoln, or Washingborough, which lies between them (Fox 1958, 31, n. 14; Field and Parker Pearson 2003, 164; Field et al. 2003, 16), though it should be remembered that Joseph Banks advertised for artefacts (Stead 2006, 26), so some of the provenances may not be reliable. A currency bar found 30 m to the west of the Fiskerton causeway hints that there may also be other, later, structures with votive deposits in the vicinity (Field et al. 2003, 26, 30). Other weapons certainly or probably of Iron Age date from the Witham include the so-called ‘Witham imp’ anthropomorphic short sword found in 1787, now lost but

31.  Ascot hats: an Iron Age leaf crown helmet from Fiskerton, Lincolnshire? preserved through its colour engraving published in Horae Ferales (Kemble et al. 1863; Clarke and Hawkes 1955, 215–16, 227 fig. 6, 8, pl. xxvi, 2; Field and Parker Pearson 2003 pl. 15b; Jope 2000, 244, pl. 53d). Rarer still is the ‘dagger’ from Barlings Eau, a tributary of the Witham (Field and Parker Pearson 2003, 162–4). The dagger from Barlings Eau was described by Sir Joseph Banks as being: ‘of a singular structure and finished with a guard for the hand. The crescent of gold on one side of the blade is also very remarkable’ (1893, 233). Ian Stead’s identification of this weapon, now also lost, as an anthropomorphic handled short sword with astral signs is plausible (Stead 2006, 49, 199, no. 232; cf. Chamberlain and Parker Pearson 2003, 144; cf. Fitzpatrick 1996). The ‘Witham sword’ found in 1826 which had an organic scabbard with bronze fittings (Jacobsthal 1939; Jope 2000, 28–30, 243, pl. 50–1; Field and Parker Pearson 2003, pl. 15, a) is likely to be no later than the earlier third century BC. The diagonal shape of the mount at the hilt and its incised decoration recalls the Hungarian scabbard style sword (Jacobsthal 1939, 28: Stead 2006, no. 36, 21–2, 26, 160–1, fig. 51; Jope 2000, 29). A similar diagonal arrangement is seen in the hilt decoration of the anthropomorphic handled weapon from Fiskerton. Field et al. suggest that both the Witham sword and shield were found in the immediate vicinity of Washingborough or Fiskerton (2003, 16). The Witham shield was also made no later than the 3rd century (Brailsford 1975, 21; Jope 1971; 2000, 54–65, pl. 60–9; Fitzpatrick 2007a, 346). Jope emphasised the wide-ranging connections of the motifs and suggested that the shield was made by a craftsman who had worked in continental Europe (2000, 61). Similarities, perhaps fortuitous, between the elements that support the roundels on the shield and a scabbard decorated with a dragon pair from Kröshegy (Kom. Somogy) in Hungary have been noted (Fitzpatrick 2007a, 348–50). Fewer finds from the Witham certainly date to the later Iron Age, but they include at least one of the Bardney swords (Stead 2006, no. 102, 175–6, fig. 76, 102; Field and Parker Pearson 2003, pl. 16), and probably the carnyx from Tattershall Thorpe (Piggott 1959, 19–21, fig. 1, A, pl. vi; Hunter 2001, 86–7; 2008, 81). The actual date of many of the iron spear heads from the river, while plausibly of Iron Age date as the finds from Fiskerton show, is still uncertain (cf. Fitzpatrick 1984, 178). The continental connections of the motifs on the scabbard mount of ‘the Witham sword’ have already been mentioned and the fact that in the Iron Age the River Witham would have been tidal almost to Fiskerton is very relevant. At Fiskerton, Ian Stead showed that the simple symmetrical decorative motifs on the finials on the anthropomorphic handled sword find their parallels in the second half of the 4th century in continental Europe (2003a, 52; 2006, 27, 162–3, no. 52, fig. 54). The coral inlays on the handle

293

are ultimately of Mediterranean origin, and the method by which the handle was made recalls that of other early anthropomorphic hilted swords and short swords (Stead 2003a, 51–2; Clarke and Hawkes 1955, 203–6, fig. 2, 1–2; Rapin 2002, 165–8). There are also far-flung parallels for the decorated iron saw from Fiskerton. While the decorated saw is apparently unique, sword blades and spear heads were occasionally decorated (e.g. Duval 1982; Vitali 1996, 596–600, fig. 5–8) as were a few knife blades found in the Czech Republic (Sankot 1994). All these objects, which often have Waldalgesheim-related decoration, are earlier than the late 3rd century BC date tentatively ascribed to the Fiskerton file by Ian Stead on the basis of British parallels (2003b, 71; 2006, 127), and suggest that an earlier date is possible. The decoration on the antler handle of the iron file from Fiskerton is also a rare example of Waldalgesheim style in these islands. The tendrils are of Verger’s type A2, which are dated elsewhere to La Tène B1 or the beginning of B2 (Verger 1987), broadly speaking the first half of the 3rd century BC. Turning away from the Witham, other early pieces of Celtic art from eastern England include the bronze scabbard plate from Wisbech acquired by ‘Philosopher’ Smith in 1847 (Piggott 1950, 4–5, fig. 1; Jope 2000, 22–23, pl. 28–9, a) and the bronze shield spine from Ratcliffe-on-Soar. Found in 1895 but only recognised as Iron Age almost a century later, the Ratcliffe-on-Soar shield is another find from an east-flowing river in eastern England (Watkins et al. 1996). The Waldalgesheim-derived decoration on the spine places it early in the genesis of insular Celtic art with the tendrils of Verger’s type A1 dating to the later 4th or early 3rd centuries BC (Verger 2003, 352–3, fig. 8; Fitzpatrick 2007a, 334–8, fig. 7). Although distinctively insular in style, some of the objects reviewed above are amongst the finest pieces of Celtic art found in Europe. In view of the location of these objects close to the east coast or on a major river, their early date, quality and their wide-ranging stylistic connections, the presence at the important religious site of Fiskerton of a rare helmet adorned with a leaf crown or Blattkrone is perhaps less surprising.

Conclusion The Fiskerton helmet is not the only extravagant Iron Age helmet from a river in Britain. One of the other major sources of early Celtic art in these islands is the River Thames, and one of the most famous finds from it is the horned helmet found at Waterloo in 1868 (Brailsford 1975, 32–9). Where the idea of having horns on the helmet came from has puzzled scholars for generations, but in the context of the swan helmet and the three-ringed helmet from Tintignac, dép. Corrèze,

294

Andrew Fitzpatrick and Martin Schönfelder

France (Maniquet et al. 2011, 108–39); the iron helmet surmounted by a raven whose wings flap from Ciumeşti, Satu Mare, Romania (Rusu 1969; Rustiou 2006); and the bronze helmets with sheet metal horns found in northern Italy (e.g. Mazzoli and Paribeni 2010), the horned helmet from Waterloo seems at ease as part of this mobile bestiary. The Battersea helmet and the much later heads on the escutcheons of 1st century BC bronze-bound buckets such as that from Aylesford (Brailsford 1975, 83–9; Stead 1971, 261–2, fig. 4, pl. xc; Jope 1983; 151, 156, Fitzpatrick 2007b, 303–4, fig. 16.6) have long meant that Britain has been able to hold its head high in the remarkable parade of ‘Ascot hats’ from the Iron Age. Now it seems that, to borrow one of Vincent Megaw’s favourite analogies in explaining and interpreting Celtic art (e.g. 1970; Megaw and Megaw 1993), the find from Fiskerton shows that there were real ‘Mickey Mouse’ hats too.

Note 1

In this light, it is possible that one of the other unidentified finds from Fiskerton, a curved U-shaped bronze object (Small Find 237; Stead 2003a, 59, fig. 4.8, 20; 4.9, lower; pl. 7) might also be from a piece of military equipment. It comes from the west side of the causeway and was found in the same context (31) as the S-shaped object. If the object is indeed Iron Age rather than Roman in date (a few Roman objects were found in the layer) as almost all of the finds from the west side of the causeway are from military objects, it seems likely that this piece will also be from a piece of military equipment.    What it might be is another matter. It has been suggested that the object is an epaulette but without any comparanda being adduced (Field et al. 2003, 22, fig. 5, 2; Parker Pearson et al. 2007, 440, fig. 6, 1). On the basis of the evidence currently available, the piece seems too large to have been the ear from a boar-shaped standard of the type known at Soulacsur-Mer (dép. Gironde) and elsewhere (Moreau et al. 1995; Vial and Kaufmann-Heinimann 2007) and too early to be from a carnyx (cf. Hunter 2009). But the recent identification of Blattkrone helmets shows that there is still much to learn about Iron Age military parade equipment.

Appendix It has been argued that the felling dates of the timber uprights in the Fiskerton causeway excavated in 1981 have a strong statistical correlation with a series of midwinter lunar eclipses in two Saros cycles of lunar eclipses (Saros cycles 46 and 47). Six out of 16 felling dates fall exactly on Saros 46/7 eclipse years (Chamberlain 2003; Chamberlain and Parker Pearson 2003; Field et al. 2003, 28–9, fig. 8; Parker Pearson et al. 2007, 444–6). It was argued that the correlation between the felling dates and the eclipses in these Saros cycles was not a result of the fortuitous observation of eclipses but instead demonstrated the ability to predict eclipses, and that the causeway was partially rebuilt or

repaired in advance of, or at the time of, the eclipses. This calculation is based on several assumptions: 1. That the felling dates from the c. 20 m length of the causeway excavated in 1981 are representative of the whole length of the causeway, which is estimated to be at least 160 m long (i.e. a c. 13% sample). 2. That only two out of the c. 20 Saros cycles that generate total lunar eclipses for any given period were identified and used in the 118 years during which trees were felled to build or repair the causeway (457/6–339/8 BC). The eclipses in Saros cycles 47 and 46 alternate, with each cycle having 8 and 7 eclipses respectively in the 118 year period (the eclipse in 320 BC should be omitted, as the last felling event cannot be dated more closely than 321–91 BC). Four of the 16 felling dates coincide with the eight eclipses in Saros cycle 47 during the 118 year period. Two of the 16 felling dates coincide with the seven eclipses in Saros cycle 46 in the 118 year period (see also Pitts 2002). 3. That the other 18 Saros cycles of the 20 charted by Chamberlain and Parker Pearson can be discounted. While the eclipses in these other cycles are not as frequent as in cycles 47 and 46, this excludes 46 other lunar eclipses, including two eclipses in different Saros cycles with which some felling dates coincide (cycles 50 and 54). The case for lunar eclipses having been predicted and marked at Fiskerton is regarded here as not proven, though it is possible that, as Chamberlain and Parker Pearson consider (2003, 145–6), the eclipses were observed and then marked by the rebuilding or repair of the causeway.

Bibliography Atkinson, R. J. C. and Piggott, S. 1955. The Torrs chamfrein, Archaeologia 96, 197–235. Banks, J. 1893. An account of some ancient arms and utensils found in Lincolnshire, chiefly in the bed of the River Witham between Kirksted and Lincoln when it was scoured out in 1787 and 1788 (from Sir Joseph Banks’ MS at Revesby Abbey), Lincolnshire Notes and Queries 3, 196–201, item 116; 232–6, item 134. Brailsford, J. W. 1975. Early Masterpieces from Britain in the British Museum. London: British Museum Press. Chamberlain, A. T. 2003. Lunar eclipses, Saros cycles and the construction of the causeway. In N. Field and M. Parker Pearson (eds.) Fiskerton. An Iron Age timber causeway with Iron Age and Roman votive offerings, 136–43. Oxford: Oxbow. Chamberlain, A. T. and Parker Pearson, M. 2003. Social implications of lunar measurement for calendrical knowledge and archaeoastronomy. In N. Field and M. Parker Pearson (eds.) Fiskerton. An Iron Age timber causeway with Iron Age and Roman votive offerings, 144–8. Oxford: Oxbow. Clarke, R. R. and Hawkes, C. F. C. 1955. An iron anthropoid sword from Shouldham, Norfolk, with related Continental and British weapons, Proceedings of the Prehistoric Society 21, 198–227. Ducongé, S. and Gomez de Soto, J. 2007. Les dépôts à caractère

31.  Ascot hats: an Iron Age leaf crown helmet from Fiskerton, Lincolnshire? culturel en milieu humides et dans les cavités naturelles du centre-ouest de la France à l’âge du Fer. In P. Barral, A. Daubigney, C. Dunning, G. Kaenel and M.-J. RoulièreLambert (eds.) L’âge du Fer dans l’arc jurassien et ses marges. Dépôts, lieux sacrés et territorialité à l’âge du Fer, Actes du XXIXe colloque international de l’AFEAF; Bienne, 5–8 mai 2005, 477–92. Besançon: Presses universitaires de FrancheComté (Annales Littéraires, Série ‘Environnement, sociétés et archéologie’). Duval, A. 1982. Note sur une point de lance provenant de Rebourseaux (Yonne). In P.-M. Duval and V. Kruta (eds.) L’Art celtique de la période de l’expansion IVe et IIIe siècles avant notre ère: actes du colloque au Collège de France à Paris 1978, du 26 au 28 septembre 1978, 137–47. Paris and Geneva: Libraire Droz. Eluère, C., Gomez de Soto, J. and Duval, A.-R. 1987. Un chef d’œuvre de l’orfèvrerie celtique: le casque d’Agris (Charente), Bulletin de la société préhistorique française 84, 8–22. Field, N. and Parker Pearson, M. 2003. Fiskerton. An Iron Age timber causeway with Iron Age and Roman votive offerings. Oxford: Oxbow. Field, N., Parker Pearson, M. and Rylatt. J. 2003. The Fiskerton causeway: research – past, present and future. In S. Catney and D. Start (eds.) Time and Tide: the archaeology of the Witham Valley, 16–32. Lincoln: The Witham Valley Research Committee and Heritage Trust for Lincolnshire. Fitzpatrick, A. P. 1984. The deposition of La Tène Iron Age metalwork in watery contexts in southern England. In B. W. Cunliffe and D. Miles (eds.) Aspects of the Iron Age in Central Southern England, Committee for Archaeology Monograph 2, 178–90. Oxford: Oxford University Committee for Archaeology. Fitzpatrick, A. P. 1996. Night and Day: the symbolism of astral signs on later Iron Age anthropomorphic hilted short swords, Proceedings of the Prehistoric Society 62, 1996, 373–98. Fitzpatrick, A. P. 2007a. Dancing with Dragons. Fantastic animals in the earlier Celtic art of Iron Age Britain. In C. Haselgrove and T. Moore (eds.) The Later Iron Age in Britain and Beyond: recent work, 339–57. Oxford: Oxbow. Fitzpatrick, A. P. 2007b. Druids: towards an archaeology. In P. de Jersey, C. Gosden, H. Hamerow and G. Lock (eds.) Communities and Connections: essays in honour of Barry Cunliffe, 287–315. Oxford: Oxford University Press. Fox, C. 1958. Pattern and Purpose: a survey of Early Celtic Art in Britain. Cardiff: National Museum of Wales. Fröhlich, R. 2006. Experiment Glauberg – zur Blattkrone des Keltenfürsten, Denkmalpflege & Kulturgeschichte 2006/3, 34–6. Frey, O.-H. 2002. Menschen oder Heroen? Die Statuen vom Glauberg und die Frühe Keltische Großplastik. In H. Baitinger and B. Pinsker (eds.) Das Rätsel der Kelten vom Glauberg. Glaube – Mythos – Wirklichkeit, 208–18. Stuttgart: Theiss. Frey, O.-H. and Hermann, F.-R. 1997. Ein frühkeltischer Fürsten­ grabhügel am Glauberg in Wetteraukreis, Hessen. Berichte über die Forschungen 1994–1996, Germania 75, 459–550. Garrow, D., Gosden. C., Hill, J. D. and Bronk Ramsey, C. 2009. Dating Celtic art: a major radiocarbon dating programme of Iron Age and early Roman metalwork in Britain, Archaeological Journal 166, 79–123.

295

Gomoz de Soto, J. 1986. Le casque du IVème siècle avant notre ère de la grotte des Perrats à Agris, France, Archäologisches Korrespondenzblatt 16, 179–83. Gomoz de Soto, J. and Verger, S. 2010. Le casque d’Agris, chefd’œuvre de l’art celtique occidental, L’archéologue 106, 56–9. Hull, M. R. and Hawkes, C. F. C. 1987. Corpus of Ancient Brooches in Britain: pre-Roman bow brooches, BAR British Series 168, Oxford: British Archaeological Reports. Hunter, F. 2001. The Carnyx in Iron Age Europe, Antiquaries Journal 81, 77–108. Hunter, F. 2009. Un oreillle de carnyx découverte à La Tene. In M. Honegger, D. Ramseyer, G. Kaenel and B. Arnold (eds.) Le site de La Tène: bilan des conniassances – état de la question. Actes de la table ronde de Neuchâtel, 1–3 novembre 2007, 75–85. Archéologie neuchâteloise 43, Neuchâtel: Office et musée cantonal d’archéologie. Jacobsthal, P. 1939. The Witham sword, Burlington Magazine 75 (no. 436, July 1939), 28–31. Jacobsthal, P, 1944. Early Celtic Art. Oxford: Clarendon Press. Jope, E. M. 1971. The Witham shield. In G. de G. Sieveking (ed.) Prehistoric and Roman Studies, 61–90. London, British Museum. Jope, E. M. 1983. Torrs, Aylesford and the Padstow hobby-horse. In A. O’Connor and D.V. Clarke (eds.) From the Stone Age to the ‘Forty-five. Studies presented to R.B.K. Stevenson, former Keeper, National Museum of Antiquities of Scotland, 149–59. Edinburgh: John Donald. Jope, E. M. 2000. Early Celtic Art in the British Isles. Oxford: Clarendon. Kaul, F. 1988. Da Våbnene tav. Hjortspringfundes og dets baggrund. Copenhagen: Nationalmuseet. Kemble, J. M., Latham, R. G. and Franks, A. W. (eds.) 1863. Horae Ferales: Studies in the Archaeology of the Northern Nations. London: Lovell Reeve. Macdonald, P. 2007. Lynn Cerrig Bach: a study of the copper alloy artefacts from the insular La Tène assemblage. Cardiff: University of Wales Press. Maniquet, C., Lejars, T., Armbruster, B., Pernot, M., DrieuxDaguerre, M., Mora, P. and Espinasse, L. 2011. Le carnyx et le casque-oiseau celtiques de Tintignac (Naves-Corrèze). Description et étude technologique, Aquitania 27, 63–150. Mazzoli, M. 2010. Was macht ein keltischer Prunkhelm in Apulien? Der Helm von Canosa. In M. Schönfelder (ed.) Kelten! Kelten? Keltische Spuren in Italien, Mosaiksteine 7, 30–3. Mainz: Römisch-Germanischen Zentralmuseum. Mazzoli, M. and Paribeni, E. 2010. Die ligurisch-apuanische Nekropole von Pulica: die Bewaffnung aus den Gräbern 1 und 5. In M. Schönfelder (ed.) Kelten! Kelten? Keltische Spuren in Italien, Mosaiksteine 7, 24–7. Mainz: Römisch-Germanischen Zentralmuseum. Megaw, J. V. S. 1970. Cheshire Cat and Mickey Mouse: analysis, interpretation and the art of the La Tène Iron Age, Proceedings of the Prehistoric Society 36, 261–70. Megaw, J. V. S. 2003. Celtic Foot(less) soldiers? An iconographic note, Gladius 23, 61–70. Megaw, J. V. S. 2007. Imag(in)ing the Celts, Antiquity 81, 438–45. Megaw, J. V. S. and Megaw, M. R. 1991. The earliest insular Celtic Art: some unanswered questions, Études Celtiques 28, 283–307. Megaw, J. V. S. and Megaw, M. R. 1993. Cheshire Cats, Mickey

296

Andrew Fitzpatrick and Martin Schönfelder

Mice, the new Europe and ancient Celtic art. In C. Scarre and F. Healy (eds.) Trade and Exchange in Prehistoric Europe, Oxbow Monograph 33, 219–32. Oxford: Oxbow. Megaw, J. V. S. and Megaw, M. R. 2008. A Celtic mystery: some thoughts on the genesis of insular Celtic art. In D. Garrow, C. Gosden, and J. D. Hill (eds.) Rethinking Celtic Art, 40–58. Oxford: Oxbow. Megaw, R. and Megaw, J. V. S. 2001. Celtic Art. From its beginnings to the Book of Kells, 2nd edition. London: Thames and Hudson. Moreau, J., Ankner, D., Boudet, R., Dhénin, M., Fecht, M. 1995. Le Sanglier-enseigne gaulois de Soulac-sur-Mer (Gironde). Étude de l’emblématique du sanglier dans le monde celtique. Soulac-sur-Mer: Musée Archéologique Soulac-sur-Mer and Association Médullienne Archéologie et Histoire du Médoc. Palmer-Brown, C. and Rylatt, J. 2002. Gifts to the gods at Iron Age Fiskerton, Minerva 13 (5), 37–8. Parker Pearson, M., Chamberlain, A., Field, N. and Rylatt, J. 2007. Fiskerton: déposition votive et éclipses lunaires en Angleterre et sur le continent. In P. Barral, A. Daubigney, C. Dunning, G. Kaenel and M.-J. Roulière-Lambert (eds.) L’âge du Fer dans l’arc jurassien et ses marges. Dépôts, lieux sacrés et territorialité à l’âge du Fer, Actes du XXIXe colloque international de l’AFEAF; Bienne, 5–8 mai 2005, 439–57. Besançon: Presses universitaires de Franche-Comté (Annales Littéraires, Série ‘Environnement, sociétés et archéologie’). Piggott, S. 1950. Swords and scabbards of the British Iron Age, Proceedings of the Prehistoric Society 16, 1–28. Piggott, S. 1959. The carnyx in early Iron Age Britain, Antiquaries Journal 39, 19–32. Pitts, M. 2002. Altar of the Druids. New Scientist 2330 (16 February 2002), 40–2. Raftery, B. 1988. Hollow metal two-piece rings in La Tène Europe, Marburger Studien zur Vor-und Frühgeschichte 11. Marburg: Dr. Wolfram Hitzeroth Verlag. Raftery, B. 1996. Excavations in the Mountdillon Bog, Co. Longford 1985–1991. Dublin: Transactions of the Irish Archaeological Wetland Unit 3. Rampinelli, A. 1968. Elmo gallico di Gottolengo (Brescia), Sibrium 8, 1964–66 (1968), 61–3. Rapin, A. 2002. Une épée celtique demasquinée d’or du Ve s. av. J.-C. au Musée des Antiquites Nationales, Antiquites Nationales 34, 155–71. Rustiou, A. 2006. A journey to the Mediterranean. Peregrinations of a Celtic warrior from Transylvania, Studia Universitatis ‘Babes-Bolyai’, Historia, Special Issue: focusing on Iron Age élites, 51/1, 42–85. Rusu, M. 1969. Das keltische Fürstengrab von Ciumeşti in Rumanien, Bericht der Römisch-Germanischen Kommission 50, 267–300. Sankot, P. 1994. Decorated La Tène single edged knives in Bohemia. New aspects of Early La Tène art, Památky Archeologické 85, 35–58.

Schaaff, U. 1974. Keltische Eisenhelme aus vorrömischer Zeit, Jahrbuch Römisch-Germanisches Zentralmuseum 21 (Festschrift Hans-Jürgen Hundt zum 65. Geburtstag), 149–204. Schaaff, U. 1988. Keltische Helme. In Antike Helme. Sammlung Lipperheide und andere Bestände des Antikenmuseums Berlin, Monographien 14, 293–317. Mainz: Römisch-Germanisches Zentralmuseum. Schönfelder, M. 2010. Kelten im Picenum: die Senonen. In M. Schönfelder (ed.) Kelten! Kelten? Keltische Spuren in Italien, Mosaiksteine 7, 20–3. Mainz: Römisch-Germanischen Zentralmuseum. Stead, I. M. 1971. The reconstruction of Iron Age buckets from Aylesford and Baldock. In G. de. G. Sieveking (ed.) Prehistoric and Roman Studies, 250–82. London: British Museum. Stead, I. M. 2003a. The Iron Age military finds. In N. Field and M. Parker Pearson, Fiskerton. An Iron Age timber causeway with Iron Age and Roman votive offerings, 49–63. Oxford: Oxbow. Stead, I. M. 2003b. Comments [On the decoration on the woodworking tools]. In N. Field and M. Parker Pearson, Fiskerton. An Iron Age timber causeway with Iron Age and Roman votive offerings, 70–1. Oxford: Oxbow. Stead, I.M. 2006. British Iron Age Swords and Scabbards. London: British Museum Press. Verger, S. 1987. La genèse celtique des rinceaux à triscèles, Jahrbuch des Römisch-Germanischen Zentralmuseums Mainz 34, 287–339. Verger, S. 2003. Le bouclier de Diviac. A propos de Liv. V, 34. In D. Vitali (ed.) L’immagine tra mondo celtic e mondo etruscoitalico. Aspetti della cultura figurative nell’antichità, 333–69. Bologna: Gedit. Vial, E. and Kaufmann-Heinimann, A. 2007. Les figures animales. In C. Gorget and J.-P. Guillaumet (eds.) Le cheval et la danseuse. À la redécouverte du trésor de Neuvy-en-Sullias, 48–87. Paris, Orléans and Bavay: Somogy éditions, Musée des Beaux Arts, Musée historique et archéologique de l’Orléanais, Musée/site archéologique départemental à Bavay, Département du Nord. Vitali, D. 1996. Manufatti in ferro di tipo La Tène in area Italiana: le potenzialità non sfruttate. In S. Verger (ed.) L’armement celtique en fer. Résultats et perspectives d’une politique scientifique de restoration, Mélanges de l’École française de Rome. Antiquité 108, 1996 (1997), 575–605. Vitali, D. 2003. La necropli di monte Tamburio a Monte Bibele. Bologna: Gedit. Vitali, D. 2010. Die keltische Kunst und die Boier in Italien. In M. Schönfelder (ed.) Kelten! Kelten? Keltische Spuren in Italien, Mosaiksteine 7, 34–7. Mainz: Römisch-Germanischen Zentralmuseum. Watkins, J., Stead, I. M., Hook, D. and Palmer, S. 1996. A decorated shield-boss from the River Trent, near Ratcliffe-on-Soar, Antiquaries Journal 76, 17–30.

32 SNETTISHAM SWANSONG I. M. Stead

I offer these further thoughts on the Snettisham Treasure in honour of Vincent Megaw, a distinguished archaeologist and good friend. They supplement a report on the discovery of hoards in 1990 (Stead 1991), an account of subsequent excavations including the identification of a large polygonal enclosure (Stead 1995) and a brief discussion of the art styles represented at Snettisham (Stead 2009).

The Great Torque The Great Torque (Hoard E) was discovered by a ploughman in November 1950. It was taken to Ken Hill, the home of the landowner, Sir Stephen Lycett Green, a barrister who had a real interest in antiquities and local history. Sir Stephen took the torque across to Sandringham, where King George VI was in residence. The story goes that the King asked to whom this fine piece belonged and Sir Stephen responded: ‘To you, sire’. As Treasure Trove it belonged to the Duchy of Lancaster and the King was Duke of Lancaster. I had heard that the King then took the torque and placed it on his neck, so when the opportunity arose I questioned Sir Stephen. He denied it and said that as far as he was aware only one person had worn it in modern times, a French lady who was dining at Ken Hill, became bored with the conversation and wandered over to the mantelpiece where the torque was on display. When Sir Stephen took the torque to Sandringham there were three other people in the room, the King, the Queen and Princess Margaret. Years afterwards, when Hoards F to L had been acquired by the British Museum, Princess Margaret paid a private visit and I was able to ask

her for her recollection of that event. She had no doubts. The King had certainly worn the Great Torque. So much for witness statements. Another Great Torque anecdote is more significant because it has chronological implications. A Gallo-Belgic DC quarter stater was discovered within the terminal of the torque and a story circulated that it had been placed there at the time that the torque was manufactured (e.g. Clarke 1956, 31) and revealed when a senior conservation officer, Henry Batten, removed one of the terminals. Rupert BruceMitford told this story in 1967 in the presence of Sylvia Schweppe, secretary to Tony Werner, Keeper of the Research Laboratory. She was convinced that the Research Laboratory would never have removed the terminal without written permission from the Department, so she questioned Harold Plenderleith, the former Head of the Research Laboratory and sent his response to Bruce-Mitford. Plenderleith wrote: No one would want to take the ends off, and had anyone wished this, Batten would have been the first to object! The twisting of the torc (the original, not the electrotype!) has the effect of opening the wire part connecting the terminals which normally clicks back again when one stops twisting. Batten discovered this when cleaning and the coin fell out. He promptly put it back again and summoned me to the workshop where he did this little conjuring trick again before my eyes. Can’t you imagine the scene: the sand spray and coin falling out and Batten, as was usual, demonstrating his prowess with glee and waiting for a word of appreciation!

The coin could have entered the body of the torque at any time up to the occasion of its burial.

298

I. M. Stead

Fig. 32.1. The excavation team, Snettisham 1990: Standing, Tony Pacitto, Ian Stead, Dave Miller (driver of the bull-dozer), Pete Boyce, Ian Blomeley, Charles Hodder, Tony Spence, Lori-Ann Foley. Crouching, Dave Webb and Peter Makey. Photograph by Dave Webb.

Metal detecting Ken Hill, Snettisham, has been the subject of much metal detecting, good, bad and indifferent. As far as is known, early attempts at metal detecting produced few results until 1990 when a determined local enthusiast, Cecil ‘Charles’ Hodder, discovered Hoard F. It was too deep to have been detected by his machine, but he dug a hole to recover a small fragment of metal and from the bottom of that hole he detected the hoard. He tried to contact an archaeologist but nobody was available because it was a Bank Holiday weekend. He did not want to leave the hoard exposed for several days so he decided to excavate it. Masses of solid artefacts that had been buried as fragments were not unduly damaged by his excavation but there were also fragments of sheet bronze, much more delicate, and they did suffer from his work. There was a suggestion that the fragments were from a vessel in which the hoard had been buried, but re-excavation of the hoard pit found no supporting evidence. Eventually painstaking work by Fleur Shearman, in the Conservation Laboratories at the British Museum, showed that many if not all of the fragments came from an object resembling a face-mask or helmet. The British Museum followed up the discovery of Hoard F with an immediate excavation based on metal detecting. Hodder’s experience suggested that hoards would not be detected from the surface so top-soil was removed by machine. First an area of almost 1,000 square metres was stripped and then thoroughly excavated to reveal the bases of Neolithic pits, but no Iron Age or Roman features other

than hoards. In that area four hoards (B–E) had been found in the past and three new hoards (G–J) were discovered in 1990. Charles Hodder had been keen to help with the British Museum excavations and we had been very happy to make use of his expertise. He was a key member of the team, though an unpaid volunteer, and like everyone else on the site he signed a document disclaiming his rights to any reward under Treasure Trove regulations (Fig. 32.1). Hodder located three of the hoards (G, H and J) found on the British Museum excavations. The initial 1,000 square metre area was extended in all directions, with a view to stripping top-soil without a following excavation. One more hoard (K) was found and it was decided to end the operation. But Tony Pacitto, who was operating a second metal detector, was not convinced and persevered at the southern edge of the area where he found the biggest of the hoards, Hoard L. But for his determination that hoard would have fallen to the night hawks. The area was then extended and 1.2 hectares were excavated in 1990, but even that was not enough. Hoard P, the Bowl Hoard, was missed because it was underneath the large spoil heap on the north side of the excavation. The night hawks found it and distributed it to archaeology’s loss. The following year the British Museum excavation was resumed and eventually 2¼ hectares were stripped. The project had been a great success, allowing archaeologists to excavate hoards under controlled conditions: the field plans showing the arrangement of the torques within the hoards and the draft plans that I prepared for publication have been deposited at the British Museum

32.  Snettisham swansong

299

Fig. 32.2. Plotting of the gradiometer survey by Tony Pacitto.

and I hope that in the future an illustrator will be able to interpret them.

The polygonal enclosure About the time of the discovery of Hoard P, the Bowl Hoard, in the winter of 1990–91, unauthorized metal detectorists located a large quantity of iron slag. When the British Museum excavations were resumed, at the end of 1991, the distribution of slag was investigated and it was found to be in the filling of a ditch. The ditch was 1.8 m deep below the present ground level and 2.8 m wide at the base of the plough-soil and the iron slag had been used to fill it over a length of 20 m in order to create a causeway. The slag and furnace debris had been dumped there no earlier than the end of the 4th century because it sealed a sherd of Oxford ware dating AD 375 or later (information from Valery Rigby). Tony Pacitto used a gradiometer to follow the course of the ditch and identified a polygonal enclosure. The north and east sides of the enclosure were surveyed in the open fields, the

west side was tracked by traverses in the adjoining wood but both gradiometer and trenching failed to locate a south side down a slope towards what would have been an inlet from the Wash (Fig. 32.2). Perhaps a ditch on the south side was not needed because of marshy land. Assuming that missing south side, the enclosure was shaped as a six-sided polygon defining an area of about eight hectares. The only entrance identified was excavated in 1992 when lengths of the ditch were fully excavated and an area 24 m by 14 m was stripped, including land on both sides of the ditch (Fig. 32.3). But there was no indication of an internal or external bank, no hint of a gate and no metalling, hollow-way or wheel-ruts in the 6 m causeway between the terminals of the ditch. There was Roman pottery in the filling of the ditch, including Flavian and Antonine pottery in the lower levels, but Valery Rigby emphasises that all the pottery was weathered and abraded. It seems likely that the ditch was open in Roman times but precisely when it was dug is obscure. The area enclosed by the ditch is similar in size to a small Roman town: Irchester and Worcester are comparable in area, and some other small Roman towns are smaller (e.g.

300

I. M. Stead

The Roman Building

Fig. 32.3. The 1992 excavation of the entrance on the west side of the polygonal enclosure. Photograph by Dave Webb.

Millett 1990, 152–3). But this Snettisham site is certainly not a Roman town. Perhaps it was a temenos defining a sacred site. The temenos of a Roman temple is usually square or rectangular in plan, usually walled rather than ditched and usually very much smaller in area. Polygonal temenos walls for temples at Colchester and Farley Heath enclosed just under one hectare and about half a hectare respectively (Lewis 1966, figs 116, left, and 117). By comparison Snettisham’s eight hectares is enormous. It is tempting to relate this unusual enclosure to the unique collection of hoards within its boundaries and to suggest that it might have defined a sacred grove (Fig. 32.4). Celtic sacred groves are mentioned by classical writers and one was presumably in East Anglia: according to Dio Cassius (xlii, 7) following Boudicca’s attacks on Colchester and London her forces celebrated ‘not only in all their other sacred places, but particularly in the grove of Andate’, Andate being a goddess he correlated with Victory. Essendon is another candidate for a sacred grove, its gold torque and coins alongside a collection of weapons are quite appropriate for a Victory goddess (Stead 2006, 51). A sacred grove would presumably provide ideal security for buried treasure.

Metal detecting, some of it authorized, continued at Snettisham after the completion of the British Museum excavations and an extensive disturbance in the wood to the west of the ‘Gold Field’ prompted another organised archaeological excavation (approximately 70 m to the west of Hoards L and N, Fig. 32.4). Roman pottery had been found there since the middle of last century (Clarke 1954, 34) and in 1990 Mike Cowell (Department of Scientific Research, British Museum) had recorded phosphate levels higher than those in the ‘Gold Field’. In 2004 an area 8 m square was cleared to reveal remains interpreted as a 6 m square Romano-Celtic temple (Hutcheson 2011). The nearest approach to a wall is a band of stones interpreted as the east wall. There were no dressed stones but perhaps it was the remains of a robbed wall. At right angles to it is the southern edge of a similar band of stones interpreted as the north wall, whose northern edge is ill-defined and extends to the limits of the excavation. The south and west walls are more enigmatic, but there was a square stoneless area in the centre of the excavation. There was no floor and no stratification but finds included tesserae, painted wall-plaster and tile fragments. A late Roman ‘fallen horseman’ coin found amongst the stones was regarded as a terminus post quem for the destruction of the building. A more extensive excavation, that might have revealed more of this building, was impossible because of the trees. It seems that there had been a Roman building of some sort but its remains had been hopelessly disturbed. No RomanoBritish religious artefacts were found and coin evidence does not suggest that this was the site of a Roman temple. Marsden, who has examined the coins from the area found by metal detectorists comments: ‘the amount of Roman coinage is so small that, were it not for the small, possible Roman temple enclosure identified by Hutcheson, one would probably not even speculate that religious activity was taking place at Snettisham at this date’ (Marsden 2011, 49). Interpreting the Roman building as a temple rests entirely on identifying the polygonal enclosure as a temple temenos. That in turn depends on suggesting that the hoards were buried on a sacred site. But there is no hoard dated after the Claudian conquest and no evidence for religious activity thereafter.

Summary of the hoards Table 32.1 shows the numbers of artefacts found in the various hoards, with the torques classified by types. Apart from torques of type Va and triangular ingots, all types are represented in Hoard F, associated with Gallo-Belgic A–C staters, and all sub-types of torques are represented in Hoard F apart from two, one in Hoard E (associated with a coin contemporary with the A–C staters) and the other in Hoards H and L. So there is little scope for arranging a seriation

32.  Snettisham swansong

301

Fig. 32.4. Plan of the polygonal enclosure and the distribution of the hoards.

of torque types. Art styles are of no help for chronology because both styles represented occur in Hoards F and L and in no other hoard (Stead 2009). It may be that all the hoards of torques were buried about the same time. Hoards A, D, G, H, J, K and L had torques but no coins; B, C, E and F had torques and coins and seem to belong to the first half of the 1st century BC. ‘Hoards’ D and E were found by ploughmen and are artefacts that might well have been derived from one or two of the other hoards. Hoard N is a scattered hoard of coins of the same types as those in B, E and F. There is a suggestion of another coin hoard buried slightly later, about the middle of the first century BC, a widely scattered collection of ‘Norfolk wolf’ staters. About a century after that comes Hoard P with more than 6,500 coins buried in a silver bowl, excavated and distributed by night hawks. Amanda Chadburn has made a determined effort to track down and record these coins and it is apparent that they were Icenian silver and gold buried no later than about AD 40. Coins took the place of torques in the treasure buried at Snettisham but there is insufficient evidence to suggest a sequence starting with torques alone,

then hoards of torques and coins. Hoard M is very different from the others, a collection of five metal lumps (75% silver, 25% copper) weighing 6.6 kg. They were found by Charles Hodder on the surface in the wood, apparently re-deposited, and were dated by C14 to the 1st century BC or the first century AD. They are consistent with the broken and partly melted artefacts found especially in Hoard F. Other hoards may have been recovered in antiquity. Even in the area that was thoroughly excavated by hand it would have been impossible to recognise a shallow pit emptied of artefacts. Indeed, we were able to identify only one corner of one of the trenches excavated in 1948 even though we knew their approximate positions (Clarke 1954, fig. 2). Analyses of the few gold fragments in the Roman jeweller’s hoard (mid-2nd century AD) found at Snettisham in 1985 are typical of Iron Age gold and very different from Roman gold. Catherine Johns has speculated that they may have been derived from accidental discoveries of hoards at Ken Hill or even from deliberate excavation of hoards by descendants of those who buried them (Johns 1997, 73, also 46 and 58–9).

I. M. Stead

302

A

B

Torques Ia 2 Ib 1 Ic II III IV Va 4 Vb–e Total torques 4 3 Ingots Ingot rings 11 Straight ingots Triangular ingots Total ingots 11 Bracelets Hollow Loop terminals Solid Sheet metal artefacts

B/C

C

13 6

D

Hoards E F

1 1 1

3

H

J

K

L

4 3 2

2 1

2 1 2

3 1 1

8

4

2

2

2 4 2 6 6 1

17

7

7

7

31 16 7 5 15 8 36 118

19

3

1

16 1 1 18

5

1

85 3

4

5

1

88

4

Metal strands, ribbons, lumps, etc. X X X Coins Gallo-Belgic A–C 9 Gallo-Belgic D/C 1 Potins 145+ 30 B.C. to A.D. 40

2

G

1

M

N

1 3 3 15

21

3 1 1

2 1

X 9

loose finds 8

3

1

1

1

P

1 X

X 7

many loose coins 6,500

Table 32.1. Showing the numbers of artefacts found in the hoards. Hoards B and C have not been examined so the information is taken from Clarke 1954. ‘Loose finds’ include three complete torques found in 1964, 1968 and 1973 and fragments found on the 1990–92 excavations but not in hoards (other fragments have been found on the site before and since). Numbers of torques include fragments and are counted from surviving terminals. Their typology, simplified by merging sub-types, is: I, loop terminals (a, twin strands; b, multi-strand with multi-loop terminals; c, multi-strand with cast on loop terminals); II, multi-strand with cast on ring terminals; III, multi-strand with buffer terminals; IV, multi-strand with cage terminals; V, hollow tubular torques.

The intended publication The academic world expects a comprehensive and fully illustrated account of the Snettisham hoards: ‘their met­ iculous publication is assured’ wrote Fitzpatrick (1992, 395) referring to their acquisition by the British Museum. And the Keeper of Prehistoric and Romano-British Antiquities had promised the landowners such a complete publication. But redundancies, staff changes and different priorities at the British Museum have rendered it impossible. The following account is now a matter of history: a record of the facts. When I retired from my post at the British Museum in 1996 I had several commitments, unpublished excavations and unfinished catalogues. My plan was to complete four monographs: three were duly published and the fourth was intended to deal with Snettisham and Essendon. Work started on the Snettisham publication in 2003 and it was agreed that I should produce an excavation report and a fully illustrated catalogue of all the artefacts found in the ‘Gold Field’ at Ken

Hill. Norwich Castle Museum agreed to lend the artefacts in their custody so that everything could be studied and recorded consistently. There were two restrictions: new photographs would be limited because staff redundancies had put huge pressures on the Photographic Department; and I was asked to omit Essendon which could be tackled as a separate paper after Snettisham had been published. Within a year I had finished the excavation report and the first draft of the catalogue and six months later drafts of the plans and sections were completed. But the illustrators had also suffered redundancies though I was promised the services of one, and then two, in the near future. Eventually I was assigned one illustrator to work part-time on the project and we made a start with artefacts that were temporarily off display at Norwich. Thus Hoard A was studied and illustrated but further progress was slow. The annual report anticipating completion in 2007 listed: 2003, 2004 – no illustrations; 2005 – illustrations of 14 artefacts; 2006 – illustrations of

32.  Snettisham swansong 2 artefacts. Work on 22 artefacts was completed before my illustrator took early retirement in 2008. Then the goal-posts were moved. No more line drawings could be authorized: photographs would have to be used instead, the reverse of the position in 2003. Throughout my career accurate measured line drawings had been essential but I was now told that they were not necessary. Ian Longworth and Barry Cunliffe (then a British Museum trustee) took up the cudgels on my behalf and a decade earlier they would have carried the day. But we were told that we would have to accept a ‘less than ideal’ publication because the Museum had limited finance and the priority now was for visitor-orientated projects. The suggestion that finance could be raised elsewhere was considered and rejected because the Museum had other priorities. Indeed they were applying for a substantial grant to finance another British Iron Age project, a project that ‘is important, but nowhere near in the same league (as Snettisham)’ according to its external referee. But the Department of Prehistory and Europe did not agree. Elsewhere in the Museum research scientists and conservation officers continued to regard Snettisham as a very high priority. When I agreed reluctantly to use photographs instead of drawings more obstacles appeared. At first I was not allowed to supervise the photography, then it was agreed that I could attend on days specified by the Museum: if they were inconvenient for me photography would go ahead regardless, which it did on at least two occasions with unfortunate results. I needed a specific view of a Snettisham artefact for a paper for Barry Raftery’s Festschrift: I was not allowed to supervise it, had to describe and sketch the view, given a quite different photograph and told that I would have to accept it. That really hurt: we did not treat aged scholars like that in my day. Fortunately I still had a friend in the Photographic Department so that photograph was achieved. And another friend, Ralph Jackson, intervened to give me more control of the photographic programme, but even he could not prevent the subsequent decline. In January 2009 it was decreed that photography would have to end. I would be allowed 50 more photographs and ‘any items that cannot be photographed by then will not be included in the report as there is no more photographic time’. What a ludicrous situation! Meanwhile new British Museum loan regulations had rendered it either impossible or too difficult to borrow artefacts from Norwich: a photographer would be sent to record them on specific days to be named by the Museum, I was given no choice. Draft page arrangements were submitted to British Museum Press and I was not consulted. I had started with the intention of producing the com­ prehensive and fully illustrated report that artefacts of international importance deserve and for which the British Museum is renowned. Take a look at Catherine Johns’ The

303

Snettisham Roman Jeweller’s Hoard (1997) and see what could be achieved only a decade earlier. But the British Museum, a centre of excellence proud of its scholarship with David Wilson and Ian Longworth in control, had become a rigid dictatorial bureaucracy with no interest in past commitments. Standards had been lowered and it was obvious that my involvement was no longer welcome. Reluctantly I withdrew from the publication project. Snettisham had taken up a lot of my time, in the field, in the Museum and at home. I had written 60,000 words and since retirement had been working in a voluntary capacity. The British Museum has a draft version of my text and I will allow them to publish the excavation report but I made it clear that the catalogue and discussion were uncorrected and incomplete and must not be published.

Bibliography Clarke, R. R. 1954. The Early Iron Age treasure from Snettisham, Norfolk, Proceedings of the Prehistoric Society 20, 27–86. Clarke, R. R. 1956. The Snettisham Treasure. In R. L. S. BruceMitford (ed.) Recent archaeological excavations in Britain, 21–42. London: Routledge & K.Paul. Davies, J. A. (ed.) 2011. The Iron Age in northern East Anglia: new work in the land of the Iceni, BAR British Series 549. Oxford: British Archaeological Reports. Fitzpatrick, A. P. 1992. The Snettisham, Norfolk, hoards of Iron Age torques: sacred or profane?, Antiquity 66, 395–8. Hutcheson, N. 2011. Excavations at Snettisham, Norfolk, 2004: re-investigating the past. In J. A. Davies (ed.) The Iron Age in northern East Anglia: new work in the land of the Iceni, BAR British Series 549, 41–8. Oxford: British Archaeological Reports. Johns, C. 1997. The Snettisham Roman jeweller’s hoard. London: British Museum. Lewis, M. J. T. 1966. Temples in Roman Britain. Cambridge: Cambridge University Press. Marsden, A. 2011. Iron Age coins from Snettisham. In J. A. Davies (ed.) The Iron Age in northern East Anglia: new work in the land of the Iceni, BAR British Series 549, 49–58. Oxford: British Archaeological Reports. Millett, M. 1990. The Romanization of Britain. Cambridge: Cambridge University Press. Stead, I. M. 1991. The Snettisham treasure: excavations in 1990, Antiquity 65, 447–65. Stead, I. M. 1995. Die Schatzfunde von Snettisham. In A. Haffner (ed.) Heiligtümer und Opferkulte der Kelten, 100–110. Stuttgart: Theiss. Stead, I. M. 2006. British Iron Age swords and scabbards. London: British Museum. Stead, I. M. 2009. The chronology of La Tène art in Britain. In G. Cooney, K. Becker, J. Coles, M. Ryan and S. Sievers (eds) Relics of old decency: archaeological studies in later prehistory. Festschrift for Barry Raftery, 323–32. Dublin: Wordwell.

33 THE IRON AGE OPEN-AIR RITUAL SITE AT HALLATON, LEICESTERSHIRE: SOME WIDER IMPLICATIONS Colin Haselgrove and Vicki Score

Introduction (CH) I first met Vincent in October 1975 at a conference in Oxford on Iron Age oppida organised by Barry Cunliffe and Trevor Rowley. At the time, Vincent was Professor of Archaeology at Leicester, whilst I was a second year PhD student at Cambridge. On the Saturday evening, at a party given by Barry, I had an enjoyable and wide-ranging conversation with Vincent, of which I remember three aspects very well. The first, given our great difference in status, was just how friendly and informal Vincent was in an era when Professors were often distant figures. Second, I was struck and cowed by his encyclopaedic knowledge of the European Iron Age, with a scholarly command of its art and decorated metalwork (but by no means restricted to these), which I knew I could never match. Third, Vincent did little to conceal his scepticism for the theoretical ideas and approaches, which were then starting to gain ground in the discipline, and have retrospectively come to be known as processual archaeology. Next morning, I was to present my paper at the conference. Under the influence of David Clarke, Colin Renfrew and Ian Hodder among others, this agenda had found a particularly receptive audience at Cambridge. I still remember that sinking feeling as I realised that the content was not going to go down at all well with Vincent! Not only did my approach take its inspiration from (economic) anthropology, but the text made free with the word ‘model’ (a term to which Vincent indicated a particular aversion) and was peppered with fulsome references to the leading proponents of the then ‘new’ archaeology (Haselgrove 1976). In the event, whatever he thought privately and no doubt recognising that

I had much to learn, Vincent chose afterwards not to dwell in detail on the perceived shortcomings of my presentation, for which I have always been grateful! In retrospect (as I am sure Vincent will agree, whatever our differences at the time), the 1975 conference did prove a turning point in the study of Iron Age oppida, albeit thanks to the contributions by John Collis, Barry Cunliffe, Daphne Nash, Warwick Rodwell and Mansel Spratling among others, rather than my own. In the ensuing publication (Cunliffe and Rowley 1976), oppida were at last recognised for what they are: a complex and – in some respects – unique phenomenon, that is probably explained by some combination of local circumstances and various wider processes that affected large regions of temperate Europe during La Tène C–D. No one paradigm yet holds sway, but nearly 40 years on, there is no doubt that we are far better informed about the sites empirically and are conceptually better able to understand the changes of which oppida were probably at once symptoms and causes (e.g. Collis 1984; Guichard et al. 2000; Fichtl 2005; Haselgrove 2006). Vincent was at Leicester from 1972–82. It was during his tenure of the chair that I now hold that a single honours BA in Archaeological Studies was established, soon followed by a BSc (McWhirr et al. 2008, 9, 13). The wider horizons that Vincent brought to the School also remain firmly in place and many undergraduates choose to develop these by spending a year at one of our many partner universities overseas, of which Flinders naturally is one. Given his formative contributions to the development of what is now one of the largest archaeology departments in Britain, it is a pleasure to be able to offer a paper to Vincent – and in

33.  The Iron Age open-air ritual site at Hallaton, Leicestershire: some wider implications

305

Fig. 33.1. The location of the site at Hallaton.

memory of Ruth, who has been a partner in so much of his research – drawing out some implications of the remarkable finds from the late Iron Age ritual site at Hallaton near Leicester. As we shall see, these require us to reconsider the nature of relations between Britain and Rome in the invasion period. As it happens – and I hope Vincent will forgive me for harping on about this – Roman impact on insular Iron Age peoples was a key theme of my 1975 Oxford paper and underpinned the core–periphery model that I went on to formulate in the heyday of processual archaeology (Haselgrove 1982; 1984). Despite its many shortcomings (e.g. Millett 1990, 39–9; Hill 2007), this model still influences many aspects of our thinking about late Iron Age Britain (e.g. Cunliffe 2005; Darvill 2010, 294–332). Vicki will begin by briefly describing the Hallaton site, for those who are not familiar with the finds. We will then turn to the wider implications for our understanding of Iron Age ritual sites and the nature of the Iron Age to Roman transition in Britain.

The Hallaton site (VS) The open-air ritual site at Hallaton lies on a hilltop overlooking the Welland valley, 21 km south-east of Leicester (Fig. 33.1). Fieldwork there was undertaken between 2001–2009 by the Hallaton Fieldwork Group and University of Leicester Archaeological Services, under the direction of Vicki Score. A monograph has been published (Score 2011), along with an account of further research since

the report was completed (Score 2012) and a study of the site for a wider audience (Score 2013). The finds are now on permanent display in the Market Harborough Museum. The principal archaeological feature was a north–south boundary ditch located on the eastern side of the hill just below the brow, broken in the middle for an entrance (Fig. 33.2). The ditch probably originally supported a palisade; it curves inwards at both ends before fading out, though it is unclear whether the boundary continued in another form (a hedge or fence?) to create an enclosure of some kind, or whether the ditch was essentially a symbolic threshold and the space was effectively open. East of the entrance were several pits associated with a dense spread of animal bone, apparently the remains of feasting or sacrifice. Almost all the bones were of young pig (97%) and some bear butchery marks. Some of the animal parts seem to be placed offerings and there is clear evidence of selection, with right fore limbs markedly under-represented (Browning 2011). The spread was only partly excavated; in all over 300 animals are probably represented. Around 25 uninscribed gold coins from this area are probably the scattered remains of earlier deposits dating to the mid-late 1st century BC. The entrance was the focus for a series of deposits, including two dog burials in the oblique slot that regulated access. In the ditch south of the entrance were 139 coins, mostly local (East Midlands) uninscribed Iron Age silver issues, a silver bowl and two ingots (one of silver, made from melted coins, the other of tin bronze) and other objects. The entire group was probably deposited in a single episode some time in the period c.AD 20–40. West of the entrance,

306

Colin Haselgrove and Vicki Score

Fig. 33.2. Plan of the excavated area showing the principal features in the deposition area.

33.  The Iron Age open-air ritual site at Hallaton, Leicestershire: some wider implications

307

Fig. 33.4. Tankard handle from Hallaton.

Fig. 33.3. Artist’s impression of the Hallaton helmet (image courtesy of Bob Whale; copyright Leicestershire County Council). The helmet is on display at Harborough Museum, Market Harborough.

14 coin hoards had been buried, tightly packed – possibly in fabric bags – in small pits, containing a total of 2027 coins, again mostly local Iron Age silver types, but this time mostly inscribed and including some gold and 26 Roman denarii, the latest an unworn coin of Claudius struck in AD 41/42. Cutting the eastern side of the ditch was a pit in which over 1350 coins (including four late denarii of Tiberius) had been buried, along with a highly decorated Roman cavalry helmet of 1st century AD date accompanied by a minimum of six cheek pieces (Score 2012, 104–109). The helmet had a scalloped brow guard, dominated by a female bust, perhaps of the goddess Cybele, flanked by two lions, whilst the bowl is ornamented by a laurel wreath (Fig. 33.3). The coin hoards are of very similar composition and were probably assembled and buried within a short time of one another sometime between c.AD 43–50 (Leins 2011, 40–43), although a date a few years to either side remains possible. A further 1945 coins were recovered, mostly from superficial contexts. Many of these had clearly been ploughed out of the hoards, but there were enough later issues to indicate that deposition continued for up to a

century after the main hoard horizon and again in late Roman times. Other mostly ploughsoil finds included 20 brooches of 1st to 2nd century AD date, some of them deliberately damaged, fragmentary copper alloy sceptre mounts like those from the Roman-British temples at Farley Heath and Wanborough in Surrey (O’Connell and Bird 1994), various Roman objects including a gold necklace clasp, and two middle Bronze Age spearheads, which were presumably curated. Regrettably, there is little Iron Age metalwork to offer to Vincent. The most significant items are a tankard handle and rim fragments (Fig. 33.4), a ‘U’-shaped chape from a sword scabbard, a possible shield clip, and a circular silver mount with relief decoration (Fig. 33.5), perhaps from a horse harness of Roman or native origin, which was found in the ditch (Cooper and Score 2011, nos 28–29, 31–32, 35). Also from the ditch were two blue and white glass ‘eyes’ with silver collars that might have adorned a gilded wooden figure (Score 2012, 111–12). Despite this paucity of decorated metalwork and the overwhelming predominance of coins, had Hallaton been found during Vincent’s time in Leicester, we are sure that he would have taken the same keen interest in the project as his successor-but-one! We will not rehearse the arguments for interpreting Hallaton as an Iron Age open-air ritual site, which are set out at length elsewhere (Score 2011). It is however worth making the point that an initial presumption, from its elevated position, that this was an ‘isolated’ shrine,

308

Colin Haselgrove and Vicki Score

Fig. 33.5. Silver mount from Hallaton.

was quickly dispelled. A combination of fieldwalking, geophysical survey and trial trenching revealed that the shrine lay within a highly organised landscape (Fig. 33.6), the evolution of which we can now trace through to the late Roman period and beyond. Only c.100 m to the northwest of the ritual area lay a large sub-rectangular ditched enclosure. This was certainly inhabited from the mid 1st century AD onward and might already have been occupied when the hoards were deposited. Interestingly, the extensive, continuing Roman activity noticeably avoided the earlier ritual focus, implying that the later inhabitants were aware of its existence and deliberated avoided building over it. In 2012, another small Iron Age coin hoard was found inside a short-lived enclosure on a prominent hilltop at Weston by Welland across the valley from Hallaton (Browning 2012), but apart from the setting, there is nothing to suggest that the sites are comparable.

Hallaton and other multiple coin hoard sites Over the centuries, it has not been uncommon for Iron Age coin hoards to be found separately at the same site, both in Britain and on the continent, as at Mark’s Tey (Essex) in 1807 and 1843, Clacton (Essex) in 1898 and 1905, or Westerham (Kent) in 1927 and 2000–1, to name but three of many examples. Place names provide clues to possible earlier instances like the Ryarsh (Kent) hoard, which was found in ‘Golden Piece’ Field (Evans 1864), or the Chapelle-

à-Oie hoard (Belgium) from the slopes of ‘Le Mont d’Or’ (Van Heesch 2005, 258). Until recently, the received wisdom among specialists was that most such finds were the remains of a single larger deposit that had either been accidentally disturbed (e.g. by ploughing), or deliberately divided into separate lots, whether by the depositors or by the finder(s). Indeed, in a study of the important late Iron Age finds from Le Câtillon (Jersey), it was shown that what were thought to be three separate deposits were probably in fact all part of a single discovery (Fitzpatrick and Megaw 1987). As an alternative – noting the chronological differences that existed between groups of coins recovered at different times on the same site – one of us proposed that many of these ‘multiple’ finds were in fact discrete hoards buried over a period of time, and moreover that such hoards could have been religious offerings, rather than wealth concealed for safe-keeping but never recovered (Haselgrove 1987, 119–20; 133–4). As was also noted, this could in principle be explored by examining findspots when the opportunity arose, with the aim of clarifying the depositional context. Not that this will necessarily result in a cut and dried answer! The 1990–91 finds at Snettisham, for example, sparked off vigorous debate about the reasons for these multiple deposits (Stead 1991; Fitzpatrick 1992), although opinion now leans towards interpreting Snettisham as a sacred locale and the hoards as an expression of a distinctive East Anglian tradition of torc deposition in the 1st century BC (Hutcheson 2004). In Hallaton, we now possess a site with multiple coin hoards and strong evidence of ritual activity. Armed with it as our ‘type site’, we can readily identify many other examples of multiple Iron Age coin hoards, both in southern Britain (Haselgrove 2008) and on the opposite side of the Channel in northern France and the Low Countries (Van Heesch 2005; Roymans et al. 2012), which could well be further examples of similar practices. Like Hallaton, many of these finds are from significant natural locations such as hills, springs and bogs, and display a similarly compressed chronology (e.g. Westerham, Kent; Chute, Hants; Thuin, Belgium) and/or mix Iron Age and Roman coins (Chatteris, Cambs). Others, whilst sharing the same locational traits, display greater time-depth (e.g. Essendon, Herts; Mark’s Tey, Essex; Snettisham, Norfolk). The 2012 discovery – widely reported in the media – of a new hoard of over 50,000 Iron Age coins at Le Câtillon, implies that, after all, it too falls into this category of ‘multiple hoard site’, as do the older finds at Rozel, also on Jersey. Nor was the phenomenon was confined to dry land, since multiple hoards apparently of different date are known from many coastal sites in southern England (e.g. Selsey, Sussex; Haselgrove 1987) and northern France (Wellington 2005). So far, only a minority of the sites in question have been investigated in the field. In Britain, the find that most closely resembles Hallaton to date is Wanborough (O’Connell and

33.  The Iron Age open-air ritual site at Hallaton, Leicestershire: some wider implications

309

Fig. 33.6. The location of the excavation trenches and features in the vicinity of the open-air ritual focus at Hallaton as revealed by geophysical survey.

Bird 1994), but this site was unfortunately only investigated after systematic looting by treasure hunters destroyed any evidence that could have shown how the coins were deposited (Haselgrove 2005, 401–9). How many coins were stolen remains a matter of debate – Bean (2000, 275) suggests 3000–4000 – but the overall profile of the 950 identified Iron Age coins recalls that of Hallaton: most are local (Southern) silver with a few gold types, whilst the 66 Roman denarii end with a single coin of Claudius of AD 50–51. At the time of discovery in the 1980s, the Wanborough coins were believed to be a single enormous hoard buried shortly after the Conquest, but the pitted nature of the looted site would be consistent with multiple hoards buried within a short period of time, just as at Hallaton (Haselgrove 2005, 404–8).

Like Hallaton, Wanborough occupies a locally prominent setting and it too, probably started as an open-air focus of some kind. In the later 1st century AD, however, a sub-circular timber structure was built and then rebuilt in stone, sealing an earlier cut feature that might have been contemporary with the main coin deposits (D. Williams 2007). This Roman building then partially collapsed and was replaced by a double-square stone temple of the more usual type. A small hoard of Flavian bronzes within the circular structure and a scatter of other finds implies that, as at Hallaton, some coin offerings were made here into the 2nd century, but then fell away, only to resume in the late Roman period. Where Hallaton differs from Wanborough and other coinrich Iron Age temples like Harlow and Hayling Island which

310

Colin Haselgrove and Vicki Score

originated as open-air sites (Haselgrove 2005) and their more numerous continental counterparts (e.g. Delestrée 1996), is in its lack of Iron Age or Roman religious structures. Clearly only a limited area was excavated at Hallaton and the ridge and furrow cultivation could have obscured minor anomalies on the geophysical survey, but there is certainly no sign of any significant constructed features other than the boundary ditch itself. The overall layout also seems different from these more ‘formal’ temples, where the greatest intensity of deposits tended to be in front of and to the left of the cult focus (Haselgrove 2005, 417). An argument might be made that the early deposits in the Hallaton boundary ditch were following a similar rule, but the location of the main hoard cluster to the north-west of the ‘turnstile’-like entrance and the burying of the feasting debris beyond the ditch, imply a different logic hinting at social distinctions between those participating in activities performed inside and those outside (Score 2011). It is possible that a more substantial boundary was erected only on the eastern side of the hilltop because both participants and those excluded from the rituals approached the site from this direction. At many hillforts and enclosed settlements, the earthworks beside the entrances are noticeably more elaborate than the rest of the circuit, presumably to impress visitors arriving at the site. It is also common for enclosure and building entrances to face due east (Hill 1995), as do many temple and sanctuary entrances, suggesting that similar ideological considerations were at work. As well as warning against drawing too rigid a distinction between sacred and profane, this is a timely reminder that in the earlier Iron Age, many hillforts and smaller settlements probably performed a role not dissimilar to Hallaton, as symbolic foci and places of periodic assembly, where communal rituals and ceremonies were enacted and/or political, social or economic business transacted at certain times of year. It is against this tradition of ritual and sacrifice on hillforts and other settlements that the late Iron Age emergence of specialised ritual sites like Hallaton is to be viewed (Hill 1995). What drove the choice of location is unclear, but the existence of a possible ploughed-out barrow 100 m south-east of the shrine (Score 2011, 17) echoes a pattern at other Iron Age ritual and mortuary sites (Fitzpatrick 1997; Haselgrove 2005). Alternatively, the choice might have been down to something as simple as a now vanished tree or a waterhole. Elevation was presumably also a factor. Philip de Jersey (pers. comm.) has noted that Iron Age coin hoards are often found just off the tops of hills, raising the possibility that, as with the false-cresting of barrows and hillfort earthworks, visibility from below was important. We might think in terms of acts of conspicuous consumption, with locations chosen to be highly visible to those excluded from direct participation in the ceremony. Selective representation of animal species or body parts

is a regular feature of ritual sites in both Britain and Gaul. At Harlow, for example, juvenile sheep killed in autumn predominate (Legge and Dorrington 1985), although the overwhelming dominance of baby pigs at Hallaton is more unusual. In his analysis of Iron Age cemeteries in East Yorkshire, Parker Pearson (1999) suggested that pig was a marker of high status, since it occurs in graves with carts, weapons and decorated artefacts, whilst sheep was found in simpler burials. Interestingly, the pig bones from East Yorkshire graves are always frontal parts – head, ribs and front legs (ibid, 53). The missing right forelegs at Hallaton might well reflect a similar practice. There was no in situ evidence of metalworking at Hallaton, but the presence of gold and silver droplets and sheet fragments as well as the ingot formed from melted coins suggest that it could have taken place nearby. There is an extensive anthropological literature about the beliefs surrounding the smelting and working of metals. The transformation by fire from ore to finished product is widely seen as a dangerous and magical process, requiring the intervention and protection of the gods (Haselgrove and Wigg-Wolf 2005, 11–12). If similar beliefs existed in Iron Age Britain, it would make sense for metalworking to be undertaken at religious sites and other locations thought to be under the control and protection of the gods. In ethnographic contexts, there is also often a tradition of sacrifice, whereby part of the finished product is offered to the gods, who owned the ore, protected metalworkers from accident and oversaw the transformation of ore into object. This would work well for Hallaton. So far we have focused entirely on sites with multiple coin deposits, but it is worth broadening out the discussion to include other kinds of metalwork. As is well known, the character of the material found at continental Iron Age sanctuaries changed significantly during La Tène D1–D2, with a wholesale shift away from weaponry and human remains and an exponential increase in the deposition of coins and objects like brooches and pottery, accompanied by abundant evidence for animal sacrifice and feasting, although there are of course regional variations (Haselgrove and Wigg-Wolf 2005, 17–21). Whilst the British evidence is different in character, the overall trend is not dissimilar. Brooches apart, most types of metal object are rare on coinrich sites, with a few exceptions like Snettisham (torcs) and Essendon (swords). At Hayling Island, for example, other metal finds were essentially restricted to two iron currency bars and small quantities of horse and vehicle equipment and weaponry (King and Soffe 2001). In this respect, the somewhat eclectic array of metal finds from Hallaton is not untypical. The large quantities of late Iron Age horse and vehicle equipment, typically single finds, but including some hoards, found in East Anglia, Lincolnshire and other areas like the West Midlands, imply that – like coins –decorated metalwork

33.  The Iron Age open-air ritual site at Hallaton, Leicestershire: some wider implications was also the object of purposeful deposition (Hutcheson 2004; Worrell 2007). Although these hoards, which are mostly of 1st century AD date, have been explained in other ways – for safe-keeping, from burials, or connected with metalworking – many of them come from significant natural locations both high and wet (cf. Garrow and Gosden 2012), prompting a suspicion that they too were deposited at openair ritual sites, albeit within a different cultural tradition to the coin-rich sites. Hoards like Ringstead, Santon Downham and Seven Sisters spring to mind. In the case of the 1843 Stanwick-Melsonby hoard, fieldwork failed to locate the findspot (Fitts et al. 1999), but revealed that, like Hallaton, it lay within a complex archaeological landscape. The nature of the Melsonby assemblage is still far from clear, but it seems to be a multiple deposit and there could be a ritual dimension (Haselgrove forthcoming). Coincidentally, the closest parallel for the Hallaton silver bowl is a small bronze bowl among the Melsonby finds.

Core-periphery revisited: early Roman contacts with the east Midlands? (CH) Core–periphery thinking became widely established in Iron Age studies during the 1980s, particularly the idea that commercial trade to supply the Roman world was a factor driving many of the archaeological changes seen in southern Britain between Julius Caesar’s invasions in 55–54 BC and the Claudian conquest (Haselgrove 1982; 1984; Cunliffe 1988). As Vincent would undoubtedly interject, in many respects the core-periphery model was merely an updating of the emphasis on externally-induced change embedded in Hawkes’ classic ABC framework, only then it was Belgic settlers (Iron C), who by introducing innovations like coinage and in ‘unifying the tribal groups over ever larger areas, took the place in British history of the Hellenistic monarchs, and like them prepared the way for Rome’ (Hawkes 1931, 93). Among a new generation brought up in a post-colonial era, the focus has shifted from generalising models to the agency of individual British rulers like Tincomaros and Tascoivanus, recognised as Roman allies and friends, and their role in the formation of new, larger political entities in late Iron Age Britain (e.g. Creighton 2000, 2006; Hill 2007). Migration from Gaul has also made a comeback. In many respects, we have come full circle, except that Romaneducated client kings favoured by Augustus and Tiberius have now replaced Hawkes’ ‘Hellenistic monarchs’. As in core-periphery thinking, political and economic contacts between British groups and the Roman world are still seen as focused on south-east England, where the consequences are most evident. Finds that could indicate otherwise – like the Arretine and Gallo-Belgic wares (Clay and Mellor 1985; Clay and Pollard 1994) from the late Iron Age riverside

311

settlement beneath the subsequent civitas capital at Leicester (Ratae) – tend to be played down, interpreted as imports passed on from south-eastern England via internal social and economic networks. To date there has been minimal discussion of the extent of Roman intervention before AD 43 in the affairs of peoples living in the rest of Britain, let alone debate about its possible effects. Thanks to the finds from Hallaton, this is set to change. As we have seen, the exact date of the main hoard horizon is delicate, with Leins (2011, 40–3) proposing a date of c.AD 43–50, but there is a degree of leeway either way and post-conquest deposition need not in any case rule out pre-conquest arrival. From copying on Iron Age coins, it has long been inferred that Roman denarii reached Britain before AD 43, but definite pre-Conquest imports are hard to come by. Among the best candidates are some of the stratified finds from Hayling Island (Haselgrove 2005), a Republican denarius reused on a necklace from an Iron Age mirror burial found near Portesham in Dorset (2010 T415), and two denarii from the late Iron Age site at Humberstone on the edge of Leicester (Thomas 2011). Interestingly, one of the Humberstone coins was a legionary denarius of Mark Antony, of which two examples were found at Hallaton in a pit cutting the boundary ditch. Since the pit contained Iron Age coins clearly disturbed from the underlying ditch, it is certainly possible that this was also the case with the legionary denarii and they too were pre-conquest arrivals. In broad terms, the profile of early denarii from Hallaton resembles other mid 1st century AD hoards in Britain (Leins 2011, 42–3; Walton 2012), but this is not to say that none of them arrived before AD 43. Imports just before the invasion would not stand out and the slight wear on some of the Pontif Maxim denarii in the helmet deposit would not preclude their arrival in the six years following the death of Tiberius in AD 37. Even from our depleted sources, it is clear that this was a period of heightened Roman involvement in Britain (Creighton 2006). In the event, Gaius abandoned his planned conquest, but this engagement would have provided a context for renegotiating alliances with British rulers and forming new ones, in both cases potentially oiled with diplomatic gifts and subsidies. Last but not least, analytical work by Farley (2011, 98) indicates that the silver used to produce Iron Age coins circulating in East Midlands was sourced (at least in part) from the Roman world. The Roman cavalry helmet is the earliest known find of its kind in Britain. Helmets were not a normal feature of military kit in late Iron Age Britain (Hockey and James 2011, 65). The helmets of Romano-Gaulish type found in burials at North Bersted (Taylor 2008) and near Canterbury (Farley 2013) probably reached Britain in a Caesarian context, either brought by refugees or as booty. Nevertheless, helmets are depicted on a few British coin types and one on a bronze of Cunobelin (BMC 1956) is clearly of Roman style, implying that some examples were imported before AD

312

Colin Haselgrove and Vicki Score

43. They could have arrived as Roman diplomatic gifts to British leaders, or, following Creighton’s (2006) ideas, with hostages returning from a period at the Roman court. Roman troops stationed at the courts of British client rulers or in their service would be another possible source. Even before AD 43, young men of high birth may have left Britain to seek a career as auxiliaries in the Roman army. That said, we cannot yet rule out the possibility that the helmet (and the denarii) were gifts, passed on to their Midlands counterparts by rulers to the south, who were in direct contact with the Roman world. The depositors of the Hallaton hoards clearly had strong ties with the North Thames kingdom ruled up to c.AD 40 by Cunobelin (Haselgrove 1984; 1987; Creighton 2000; 2006), of which Camulodunum and Verulamium were the principal centres. Nearly three-quarters of non-local Iron Age coins from Hallaton (73.4%) came from this area (Leins 2011), several of them in the ditch. Among the finds from Leicester is a Gaulish coin inscribed Germanus Indutilli L (Clay and Pollard 1994, 134). The type is unusual in Britain, but at least three are known from Camulodunum. Nevertheless, it seems improbable that the imperial court did not exchange any embassies and or broker alliances before AD 43 with groups outside south-east England. Roman imports were reaching the fortified regional centre at Stanwick (North Yorkshire) by the Augustan period (Haselgrove forthcoming). The number of rulers (eleven) who submitted to Claudius also points at ties with groups in many different parts of Britain, particularly if one of these leaders really was king of the Orkneys (Fitzpatrick 1989). Moreover, even if Tiberius did little more than renew alliances established by Augustus, the planned expedition by Caius and the actual invasion under Claudius are both likely to have sparked off fresh waves of treaties. The date of the Roman coins accompanying the helmet would not rule out their having arrived together as gifts or subsidies at this period. A key point to arise from detailed study of the 5000+ East Midlands silver and gold issues found at Hallaton is that this coinage does not form a single sequence (Leins 2011). Instead, several individuals were actively issuing coinages at same time in different parts of what became the Roman civitas of the Corieltavi. As Leins shows, this tradition originated in north Lincolnshire and did not achieve its full territorial extent, including the Leicester area, until the 1st century AD. This has significant implications for the more traditional picture of a region inhabited by a unified tribal grouping prior to the conquest (e.g. Cunliffe 2005). Whilst the minting of similar coin types implies a degree of shared identity, which was later recognised by the Roman authorities, we would probably be better to think of a region inhabited by a number of distinct groups, linked by crosscutting economic and social networks, but competing with one another as well as cooperating (Hill 2011).

Thanks to the Hallaton finds and other recent excavations, there are good grounds for believing that developments in the Leicester area followed a broadly similar pattern to a number of major late Iron Age centres in south-east England. The site at Leicester (Ratae) was a new foundation, perhaps superseding the regionally important settlement at Humberstone, which went into decline in the late 1st century BC, around the time Leicester was founded. The new settlement enjoyed important long-distance contacts from the outset, as shown by the early arrival of Roman imports. Initially Leicester was peripheral to the main East Midlands coin distribution, but the discovery of a large quantity of baked clay flan trays at the Merlin Works near the river indicates that coins were being minted there by the mid-1st century AD. The best candidate is the Tatisom series, which displays a southerly bias (Leins 2011). In this, Ratae shares many attributes – as has long been recognised – with some of the other late Iron Age sites known collectively as oppida, but better termed ‘royal’ sites, due to their close association with the political development of kingship (Creighton 2000; Hill 2007). These sites, which include Camulodunum, Verulamium and Calleva, appear to be new foundations. Most of them lay near of the boundaries of previously distinct groups that they presumably served to unite; often they are beside water. Some may have begun as neutral meeting places, where groups from a wider region periodically assembled to transact business and effect exchanges of various kinds, accompanied by ceremonies and feasting (Haselgrove 2000). The one difference is that Verulamium – which in other respects resembles Leicester – and the other sites minted coins bearing the Latin place name and/or that of their ruler (and sometimes title) from the outset. At Ratae, this did not occur until much later and even then only partially, since the Tatisom legends may never have been fully intelligible thanks to blundered copying (Leins 2011). If, as Jonathan Williams (2007) has proposed, coin legends did represent the adoption of a ‘Roman cultural package’ by leading groups in south-east England or a later cultural alignment on the part of neighbouring peoples, this evidently went only so far in the Leicester area. Hill’s (2007, 25–6) suggestion that Iron Age coins helped to articulate developing ties of clientage and dependency, and to create and sustain ties and power relationships between members of these networks, that were less socially encumbered than existing forms of exchange, may be pertinent to the East Midlands. Ratae was not fully integrated into these networks until a relatively late date, and we might speculate that other focal sites existed in the territory that was eventually to form the civitas of the Corieltavi, an obvious possibility being beside the Brayford Pool at Lincoln. Not only does Lincoln lie at a key transit point where the River Witham cuts through the Lincolnshire Edge, but it also sits at the junction between the primary area of coin use in northern Lincolnshire and its secondary

33.  The Iron Age open-air ritual site at Hallaton, Leicestershire: some wider implications extension into eastern Lincolnshire, which later became united in the stylistically linked Avn, Vep and Iisvprasv series (Leins 2011). Old Sleaford with its mass of minting debris appears to have been another major centre. The eventual civitas of the Corieltavi, then, was partly a Roman creation (cf. Moore 2011). When the region was incorporated into the province, Leicester was selected as the civitas capital, whereas Lincoln had a fortress imposed and then a colonia. This may just have been for reasons of strategy, but it could reflect the way that different local groups reacted to the invasion. The finds from Hallaton strongly suggest that the inhabitants of the Leicester area welcomed the Romans. They may already have had a formal treaty of friendship; at very least, they had strong contacts with one of the client states in the south-east. By contrast, the people living in the Lincoln area may have opposed the Romans. Pushing the evidence to its limits, we might even see the helmet and denarii as gifts to a leader who actively facilitated the Roman invasion, perhaps even one of those who surrendered to Claudius. The reason for the choice of Hallaton for the deposition of such a symbolic object could well be its location overlooking the important road from Colchester to Leicester, close to its junction with two other long-established routes (Score 2011).

Conclusion It would be nice to think that we might eventually be able establish whether the Hallaton helmet and denarii came directly from a Roman source and whether they arrived before or after AD 43, but this is of secondary importance in any case. The primary significance of Hallaton is threefold: first, as a model for the investigation and interpretation of other coin-rich sites; and second, in drawing our attention to the less visible changes that were under way in late Iron Age societies outside south-east England, but driven largely by endogenous factors (Hill 2007). Third, the finds are a timely reminder that Roman interests in Britain after Caesar were not confined to the ‘known’ client kingdoms of south-east England, but touched peoples and institutions in many other areas of Britain, with differing consequences.

Bibliography Bean, S. C. 2000. The Coinage of the Atrebates and Regni. Oxford: Oxford University Committee for Archaeology Monograph 50. Browning, J. 2011. The animal bones. In V. Score, Hoards, hounds and helmets: a conquest-period ritual site at Hallaton, Leicestershire, 103–35. Leicester: Leicester Archaeology Monograph 21. Browning, J. 2012. An archaeological strip, map and sample excavation at Barn Farm, Weston by Welland, Northamptonshire. Leicester: Unpublished ULAS Report 2012–157.

313

Clay, P. and Mellor, J. 1985. Excavations in Bath Lane, Leicester. Leicester: Leicestershire Museums Archaeological Report 10. Clay, P. and Pollard, R. 1994. Iron Age and Roman excavation in the West Bridge area, Leicester. Leicester: Leicestershire County Council, Museums Arts and Records Services. Collis, J. R. 1984. Oppida: earliest towns north of the Alps. Sheffield: Department of Archaeology, University of Sheffield. Cooper, N. and Score, V. 2011. The small finds. In V. Score, Hoards, hounds and helmets: a conquest-period ritual site at Hallaton, Leicestershire, 67–87. Leicester: Leicester Archaeology Monograph 21. Creighton, J. 2000. Coins and Power in Late Iron Age Britain. Cambridge: Cambridge University Press. Creighton, J. 2006. Britannia. The Creation of a Roman Province. London: Routledge. Cunliffe, B. 1988. Greeks, Roman and Barbarians. Spheres of interaction. London: Batsford. Cunliffe, B. 2005. Iron Age communities in Britain (4th edn). London: Routledge Cunliffe B. and Rowley, T. (eds.) 1976. Oppida: the beginnings of urbanization in temperate Europe, BAR Supplementary Series 11, 25–49. Oxford: British Archaeological Reports. Darvill, T. 2010. Prehistoric Britain (2nd edn). London: Routledge. Delestrée, L.-P. 1996. Monnayages et Peuples Gaulois du NordOuest. Paris: Editions Errance. Evans, J. 1864. The coins of the Ancient Britons. London: J. Russell Smith. Farley, J. 2011. Analysis of silver and other metal objects. In V. Score, Hoards, hounds and helmets: a conquest-period ritual site at Hallaton, Leicestershire, 88–99. Leicester: Leicester Archaeology Monograph 21. Farley, J. 2013. An Iron Age Canterbury tale, Current Archaeology 275, 9. Fichtl, S. 2005. La ville Celtique. Les oppida de 150 av J -C à 15 ap J. –C. (2nd edn). Paris: Editions Errance. Fitts R. L., Haselgrove, C., Lowther P. and Willis S. 1999. Melsonby revisited: excavations in 1994–95 on the site of the discovery of the 1843 ‘Stanwick’ hoard, Durham Archaeological Journal 14, 1–52. Fitzpatrick, A. 1989. The submission of the Orkney Islands to Claudius: new evidence? Scottish Archaeological Review 6, 24–33. Fitzpatrick, A. 1992. The Snettisham, Norfolk, hoards of Iron Age torques: sacred or profane? Antiquity 66, 395–8. Fitzpatrick, A. 1997. Archaeological excavations on the route of the A27 Westhampnett Bypass, West Sussex: Vol. 2: the cemeteries. Salisbury: Wessex Archaeological Report 12. Fitzpatrick, A. and Megaw, J. V. S. 1987. Further finds from the Le Câtillon hoard, Proceedings of the Prehistoric Society 53, 433–44 Garrow, D. and Gosden, C. 2012. Technologies of Enchantment. Exploring Celtic Art: 400 BC to AD100. Oxford: Oxford University Press. Guichard, V., Sievers, S. and Urban, O-H. (eds.) 2000. Les processus d’urbanisation à l’âge du Fer. Glux-en-Glenne: Collection Bibracte 4. Haselgrove, C. 1976. External trade as a stimulus to urbanization. In Cunliffe and Rowley 1976, 25–49. Haselgrove, C. 1982. Wealth, prestige and power: the dynamics

314

Colin Haselgrove and Vicki Score

of late Iron Age political centralization in south-east England. In C. Renfrew and S. Shennan (eds.), Ranking, Resource and Exchange, 79–88. Cambridge: Cambridge University Press. Haselgrove, C. 1984. Romanization before the Conquest. In T. Blagg and A. King (eds.), Military and civilian in Roman Britain, 5–63. BAR British Series 136. Oxford: British Archaeological Reports. Haselgrove, C. 1987. Iron Age Coinage in South-East England: The Archaeological Context. BAR British Series 174. Oxford: British Archaeological Reports. Haselgrove, C. 2000. The character of oppida in Iron Age Britain. In V.Guichard, S. Sievers and O-H. Urban (eds.) Les processus d’urbanisation à l’âge du Fer, 103–10. Glux-en-Glenne: Collection Bibracte 4. Haselgrove, C. 2005. A trio of temples: a reassessment of Iron Age coin deposition at Hayling Island, Harlow and Wanborough. In Haselgrove and Wigg-Wolf 2005, 381–418. Haselgrove, C. (ed.) 2006. Les mutations de la fin de l’âge du Fer (Celtes et gaulois: l’Archéologie face à l’Histoire). Glux-enGlenne: Collection Bibracte 12/4. Haselgrove, C. 2008. Iron Age coin-finds from religious sites in N Gaul. In R. Haeussler and A. King (eds.), Continuity and Innovation in Religion in the Roman West, Vol. 2, 7–23. Portsmouth (RI): Journal of Roman Archaeology Supplementary Series 67. Haselgrove, C. forthcoming. Cartimandua’s capital? The late Iron Age royal site at Stanwick, North Yorkshire: fieldwork and analysis 1981–2009. York: CBA Research Reports. Haselgrove, C. and Moore, T. (eds.), 2007. The later Iron Age in Britain and beyond. Oxford: Oxbow Books. Haselgrove, C. and Wigg-Wolf, D. (eds.) 2005. Iron Age Coinage and Ritual Practices, Studien zu Fundmünzen der Antike 20. Mainz: Philipp von Zabern. Hawkes, C. F. C. 1931. Hillforts, Antiquity 5, 60–97. Hill, J. D. 1995. Ritual and rubbish in the Iron Age of Wessex, BAR British Series 242. Oxford: Tempus Reparatum. Hill, J. D. 2007. The dynamics of social change in Later Iron Age eastern and south-eastern England c. 300 BC–AD 43. In C. Haselgrove and T. Moore (eds.), The later Iron Age in Britain and beyond, 16–40. Oxford: Oxbow Books. Hill, J. D. 2011. How did British middle and late pre-Roman Iron Age societies work (if they did)? In T. Moore and L. Armada (eds.), Atlantic Europe in the first millennium BC. Crossing the divide, 242–63. Oxford: Oxford University Press. Hockey, M. and James, S. 2011. The Roman helmet components. In V. Score, Hoards, hounds and helmets: a conquest-period ritual site at Hallaton, Leicestershire, 61–66. Leicester: Leicester Archaeology Monograph 21. Hutcheson, N. 2004. Later Iron Age Norfolk. Metalwork, Landscape and Society, BAR British Series 361. Oxford: Archaeopress. King, A. and Soffe, G. 2001. Internal organisation and deposition at the Iron Age temple on Hayling Island, Hampshire. In J. R. Collis (ed.) Society and settlement in Iron Age Europe. Actes du XVIIIe colloque de l’AFEAF, Winchester 1994, 111–24. Sheffield: J. R. Collis Publications. Legge, A. J. and Dorrington, E. 1985. The animal bones. In N. France and B. Gobel, The Romano-British temple at Harlow, 122–33. Harlow: West Essex Archaeological Group.

Leins, I. 2011. The coins. In V. Score, Hoards, hounds and helmets: a conquest-period ritual site at Hallaton, Leicestershire, 39–60. Leicester: Leicester Archaeology Monograph 21. Millett, M. 1990. The Romanization of Britain. Cambridge: Cambridge University Press; McWhirr, A., Christie, N. and Palmer, M. 2008. Digging up our past. Fifty years of Archaeology and forty years of Ancient History at the University of Leicester, 1957/8–2007/8. Leicester: School of Archaeology and Ancient History. Moore, T. 2011. Detribalizing the later prehistoric past: Concepts of tribes in Iron Age and Roman studies, Journal of Social Archaeology 11, 334–60. O’Connell, M. G. and Bird, J. 1994. The Roman temple at Wanborough, excavation 1985–1986, Surrey Archaeological Collections 82, 1–168. Parker Pearson, M. 1999. Food, sex and death: cosmologies in the British Iron Age with particular reference to East Yorkshire, Cambridge Archaeological Journal 9, 43–69. Roymans, N., Creemers, G. and Scheers, S. 2012. Late Iron Age gold hoards from the low countries and the Caesarian conquest of Northern Gaul. Amsterdam: Amsterdam Archaeological Studies 18. Score, V. 2011. Hoards, hounds and helmets: a conquest-period ritual site at Hallaton, Leicestershire. Leicester: Leicester Archaeology Monograph 21. Score, V. 2012. Helmets, ingots and idols: an update on the Hallaton finds, Transactions of the Leicestershire Archaeological and Historical Society 86, 103–15. Score, V. 2013. Hoards, Hounds and Helmets: The Story of the Hallaton Treasure. Leicester: University of Leicester Archaeological Services. Stead, I. M. 1991. The Snettisham treasure: excavations in 1990, Antiquity 65, 447–64. Taylor, M. 2008. Warrior grave found in Bognor, Current Archaeology 223, 4. Thomas, J. 2011. Two Iron Age ‘Aggregated’ Settlements in the Environs of Leicester: Excavations at Beaumont Leys and Humberstone. Leicester: Leicester Archaeology Monograph 19. Van Heesch, J. 2005. Celtic coins and religious deposits in Belgium. In C. Haselgrove and D. Wigg-Wolf, (eds.), Iron Age Coinage and Ritual Practices, Studien zu Fundmünzen der Antike 20, 247–63. Mainz: Philipp von Zabern. Walton, P. J. 2012. Rethinking Roman Britain: coinage and archaeology. Wetteren: Collection Moneta 137. Wellington, I. 2005. Placing coinage and ritual sites in their archaeological context. In C. Haselgrove and D. Wigg-Wolf, (eds.), Iron Age Coinage and Ritual Practices, Studien zu Fundmünzen der Antike 20, 227–45. Mainz: Philipp von Zabern. Williams, D. 2007. Green Lane, Wanborough: excavations at the Roman religious site 1999, Surrey Archaeological Collections 93, 149–265. Williams, J. 2007. New light on Latin in pre-conquest Britain, Britannia 38, 1–11. Worrell, S. 2007. Detecting the Later Iron Age: a view from the Portable Antiquities Scheme. In C. Haselgrove and T. Moore (eds.), The later Iron Age in Britain and beyond, 371–88. Oxford: Oxbow Books.

34 BRIT-ART: CELTIC ART IN ROMAN BRITAIN AND ON ITS FRONTIERS Jody Joy

As is amply demonstrated in the Megaw’s extremely useful book Celtic Art, first published in 1989 and reprinted in 2001, designs recognisable as Iron Age or Celtic in character persist in Roman Britain, particularly in the north and the west where flourishing new regional art styles develop. Objects include horse-gear, as well as new varieties of other well-known Iron Age object types such as torcs. Recent surveys reveal that these objects are in fact more numerous than art made before the conquest (Gosden and Hill 2008, 2; Garrow et al. 2009). The influence of Rome can be seen especially in the use of enamels of multiple colours arranged in geometric patterns and brass, a Roman metal. However, although this paper is situated within a wider discourse of Romanisation, it is from the perspective not of how pre-Roman peoples became Roman but rather the role of art in the construction and renegotiation of identity. Building on recent research by Fraser Hunter (2006a; 2006b; 2008a; 2008b; 2010; 2012) and Mary Davis and Adam Gwilt (2008), which highlight regionality and diversity, it is argued that this art is not an historical fossil. Rather, through the making and wearing of these objects, people were actively working out how to live in the Roman world, or on its frontiers.

Artistic Background: art in the two centuries before the Roman conquest and beyond Art from Britain in the last two centuries before the Roman conquest differs markedly from art on the Continent and there has been much discussion of this insular art style (e.g. De Navarro 1952). It adorns a number of object

categories, such as items of dress and horse-harness gear, but not every object of a particular type is decorated (see Joy 2011, 212). The art is characterised by a repertoire of curvilinear motifs used in combination to produce complex, interlocking designs. Voids created in designs are often just as distinctive as the defined motifs themselves (Fox 1946; 1958; Joy 2008; 2010, Ch. 4), adding further complexity. Of particular importance is the trumpet void, delineated by two S-shaped lines of equal length and a third, concave line (Fox 1946, 48–49). Decorative techniques include chasing and engraving, as well as repoussé and lost-wax casting to create motifs in relief. By the mid-1st century BC this style of art all but stopped being deposited in southern England and the increasing influence of Gaul, at the time under Roman control, can be detected in the styles of material culture recovered, particularly those found in cremation graves, where ‘Roman’ and ‘Roman inspired’ objects are found alongside indigenous objects such as zoomorphic iron firedogs and stave-built wooden buckets with decorative mounts (Hunter 2006a, 100–103; Stead 1996, 35). Artefacts associated with feasting and personal appearance dominate grave assemblages. Alterations in the style of art such as the increasing use of straight lines and, less commonly, the adoption of berried rosettes are further indications of Roman influence, as is the production of die-stamped sheet metalwork, sometimes called casket ornament, comprising repeated raised motifs, some in the classical style, which probably adorned wooden boxes or caskets (Megaw and Megaw 2001, 228). One of the most familiar aspects of the material culture of the Roman Empire is the huge variety of artefacts. Exposure

316

Jody Joy

to these objects and the technologies used to make them is likely to have had a profound impact on people used to a very different material world with a more restricted range and quantity of artefacts (Gardner 2007, 32–33). The availability of these objects would have created new aspirations, new visual environments, and new possibilities for display and the accumulation of material (Garrow and Gosden 2012, 309). Prior to the 1st century AD, art was primarily manu­ factured from bronze (an alloy of copper and tin) and more rarely gold and silver. Red, and sometimes yellow, glass was also employed as decoration. Glass was heated and pressed into cells cast or cut into the metal (Rigby in Stead 2005, 120). These cells are in the form of curvilinear motifs characteristic of art in the later Iron Age. During the 1st century AD, new technologies are adopted and new materials are worked, such as enamelling and the use of brass. Brass is an alloy of copper and zinc and is generally regarded as a Roman metal. Although there is evidence for its usage much earlier (Craddock et al. 2004), brass is not used frequently in Britain until at least the beginning of the 1st century AD and continues in usage after the conquest (Dungworth 1996, 410–11). It has been thought that the zinc used was obtained by recycling metal from the Roman world (ibid.) although analysis of the material from the Severn Sisters hoard implies the acquisition of relatively newly manufactured brass and/or local smelting of zinc ores (Davis and Gwilt 2008, 161). Polychrome enamel was also used as a decorative technique. Although the end result can often look quite similar to the traditional Iron Age use of glass, albeit with more colours, the actual technique of manufacture is very different. Rather than pressing heated glass into cells, enamelling involves filling cells with ground glass and then heating the object to fuse the enamel in place, creating solid blocks of colour. A number of different coloured enamels were used including red, yellow, white and blue. Compositional analysis shows that the enamels used are closer to Roman enamels than to Iron Age glass (Davis and Gwilt 2008, 156). Polychrome enamels are often set into square, rectangular and triangular cells, arranged in geometric patterns. During the 1st and 2nd centuries AD, art adorns a wider range of objects but with some preservation of pre-Roman decorative conventions. Neck-ornaments, bracelets and fasteners, including brooches, are known from both preRoman and Roman cultural traditions, however, with the exception of simple spiral rings and a small number of ‘massive’ finger rings from NE Scotland (see Hunter, this volume), familiar forms of adornment in the Roman world such as earrings and finger rings were not commonly worn by Iron Age peoples (Johns 1996, 30–31). As Johns (1996, 184) identifies, it seems that object categories introduced by the Romans were adopted without alteration, whereas hybrids of familiar object types, such as brooches, retained

some features even though their form and decoration were also influenced by Roman technology and fashions.

History of Research Until recently, art dating to the very late Iron Age into the 1st and 2nd centuries AD has received relatively little attention, as the main focus of previous studies has often been on earlier metalworking traditions or unusual singular items, with continental parallels and comparisons sometimes sought. Where it is examined, it is most often argued that continuity of design is an indicator of the maintenance of traditional practices and customs in response to the threat of social pressures posed by the influence of Rome (e.g. Macgregor 1976, 177–178). By creating and wearing this highly conspicuous art, which encompasses traditional motifs and the use of new technologies and decorative techniques, societies are seen to reinforce their own cultural traditions in the face of external cultural pressures (Hunter 2006a, 107). Amongst many Roman scholars, it is most often viewed as provincial art, adorning small objects often referred to as trinkets (Henig 1995, 13; see Hunter 2008a, 130). Its preponderance in the early Roman period, particularly horse-trappings, has sometimes also been interpreted as a result of artefacts being commissioned and distributed by the Roman legions as they moved through Britain (e.g. discussion of Severn Sisters hoard in Dudley and Webster 1965, 194). Against a backdrop of the explosion in finds made by metal detector users and the consequent data made available via the Portable Antiquities Scheme (http://finds.org.uk/), as well as a re-emergence of interest in the conquest period (e.g. Creighton 2000; 2006), interest in the continuance of preRoman art styles in the conquest and post-conquest periods has been reignited (e.g. Hunter 2008a; Davis and Gwilt 2008). As stated in the introduction, these papers highlight the diversity and regionality of the material, which also makes the art hard to summarise (although see Hunter 2008a, table 8.1). For example, Fraser Hunter (Hunter 2006b) advocates the influence of the threat of Roman military incursions on the production of massive metalwork, which encompass pre-Roman decorative techniques and Roman technology (see below) in northeast Scotland in the 1st and 2nd centuries AD, but also demonstrates that in southeast England in the 1st century BC there was a very different reaction to Rome, as local styles were rejected in favour of Roman and Roman-inspired artefacts (Hunter 2006a, 107).

‘Becoming’ Roman Despite this re-ignition of interest, recent theoretical discussions on identity categories in the 1st and 2nd centuries in Roman archaeology (e.g. James 2001a, Gardner 2007),

34. Brit-art: Celtic Art in Roman Britain and on its Frontiers particularly the critical attention directed at the categories of Roman and native (e.g. Hill 2001, 12; Revell 2009, xi), have yet to fully permeate into discussions of art in the 1st and 2nd centuries, hence the continuation of the terms ‘Celtic’ or ‘native’ to describe art or manufacturing techniques, which are then set against ‘Roman’ art styles and/or technology. Whilst these are useful general categories, they are too value-laden, static and homogenous and can be used to create a simple continuum in which people are more, or less, ‘Roman’ (Hill 2001, 12). By setting ‘Celtic’ or ‘native’ against ‘Roman’, a dichotomy is created which acts as a focus of discussion and consequently, commentators tend to concentrate on identifying ‘Roman’ and ‘native’ constituents of the art. As John Creighton (2000; 2006) makes clear, many of the changes traditionally linked to Romanisation have roots in the preceding centuries. By providing a longer chronology for social change Creighton lessens the significance of the Roman conquest, helping to break down this dichotomy. It is also possible, for example, that during the conquest period for some groups or individuals ‘Rome’ or ‘Roman-ness’ was not that relevant, rather Roman objects, or new opportunities such as trade with the Roman world, were appropriated to achieve locally situated agendas. In this instance, ideas of resistance to Rome or becoming Roman are insignificant (James 2001a, 203). Although the form of most artefacts is concretised at the time of production, their meaning is mutable and contextdependent. In reality, as Gardner neatly expresses: Individuals [and artefacts] are typically able to change their affiliations, and more importantly to ‘have’ multiple ethnicities (or none at all), expressing different group identities in different contexts of interaction. An individual [or artefact] in Britain in the Roman period, for example, is thus likely to have had a number of nested affiliations that might be termed ‘ethnic’ at varying socio-spatial scales, up to and including ‘Roman’ (Gardner 2007, 198–199).

Rather than viewing the meaning of art as static or set, its role in renegotiations of, and performing, identity will be the focus of study throughout the rest of this paper.

Describing and categorising the art To counteract some of these issues, the terminology used throughout this paper broadly follows that set out by Louise Revell in her book Roman Imperialism and Local Identities (2009, xi). ‘Roman’ is used to refer to people or material culture from within the Roman Empire, including parts of Britain following Roman incursions into southern England in AD 43. ‘Native’ is avoided as it is too value-laden. ‘Iron Age’, ‘pre-Roman’ or ‘non-Roman’ is preferred instead. Although ‘Celtic art’ can be useful as a term to describe a broad and diverse group of material (Garrow and Gosden 2012, 47), its use is avoided here as it is also a value-laden term bound up with the maintenance of some form of native

317

identity. Where the term ‘Celtic’ is used, it refers to a style of art which spans pre-Roman and Roman Britain and encompasses swirly motifs and curvilinear designs.Fraser Hunter (2010, 91) usefully divides this art into three broad categories: 1. Continuing production of pre-Roman objects in the Roman period, particularly horse-gear. 2. Roman objects decorated in Celtic styles. This includes some enamelled vessels and seal boxes, as well as more singular examples discussed below. 3. Hybrids combining Roman and pre-Roman influences to create new types of object. These include beaded torcs, Wraxall collars and Dragonesque brooches.

The following summary broadly follows Hunter’s categor­ isation and is intended to illuminate the variety and complexity of the objects. It is followed by a revised interpretation of how they may have operated in the socially dynamic and unstable environment of Britain in the first two centuries AD.

Campaigning art From AD 50–75 objects, particularly horse-gear and vessels, are decorated in a new style which is found in western and northern Britain, as well as in East Anglia. As these objects are made and used at the same time as the period of confrontation between pre-Roman peoples and Roman armies they have been labelled ‘campaigning art’ (Davis and Gwilt 2008, 147). At this time a number of metalwork hoards are buried across Britain, such as the hoard from the Polden Hills in southwest England (Brailsford 1975), the Seven Sisters hoard from South Wales (Davies and Spratling 1976; Davis and Gwilt 2008) and the Stanwick hoard, Yorkshire (Fitts et al. 1999; Macgregor 1962). These hoards vary in content but most contain decorated metalwork, most commonly items of horse-harness, as well as other military equipment and feasting and drinking items, particularly tankards. The horse-gear stands out with items such as terrets and mounts often being beautifully decorated (Fig. 34.1). Recent analysis indicates that this art falls into two distinct categories: ‘curvilinear’ and ‘geometric’ (Davis and Gwilt 2008, 158). ‘Curvilinear’, or Celtic-style, art comprises bronze objects decorated with curvilinear motifs and often voids filled with red or occasionally yellow glass. In contrast, ‘geometric’, or Roman-style, art is most often made from brass decorated with multi-coloured enamel. Occasionally techniques are blurred, with brass being used in combination with curvilinear decoration. The differences between the two styles may not appear particularly significant to us, but the objects would have looked extremely different to people at the time. Brass is a different colour to bronze, and its sheen is similar in effect to gold. During the Iron Age, colour may

Jody Joy

318

Fig. 34.1. Enamelled terrets from the Polden Hill Hoard. P&E 1846,0322.94,96,101. © Trustees of the British Museum.

also have had links to gender. Blue and white beads are found in the graves of women in East Yorkshire, whereas objects decorated with red glass, such as weaponry, are most often associated with a more masculine identity (see Fitzpatrick 2007, 344–345; Giles 2008, 72–74). The use, for example, of Roman enamel of multiple colours would therefore have mixed up these pre-Roman associations. Campaigning art is most often seen as a selective use of Roman technology and materials by pre-Roman peoples. However, we do not know who made these objects, and they are not slavishly adopting Roman techniques. Objects continue to be made using pre-Roman methods. Other objects are made using a combination of pre-Roman and Roman techniques. Finally, when they are deposited, some hoards, like Polden Hills, contain almost exclusively curvilinear objects. Other hoards, such as Seven Sisters, contain both curvilinear and geometric objects. At all stages, therefore, different messages are arguably being communicated through this art by the selective use of pre-Roman and Roman techniques in manufacture and the selection and incorporation into hoards of artefacts made using these different techniques.

New versions of old favourites During the 1st century AD, new versions of pre-Roman

artefacts also appear. For example, several neckrings dating to the 1st and 2nd centuries AD have been discovered predominantly in south-west Britain. These are known as ‘Wraxall’ collars after one of the finest known examples (Megaw 1971). This is quite a varied group of objects, although all have a hinge at the back, to aid in getting the collar on and off, and some kind of joint at the front to secure it in place when worn. Illustrated (Fig. 34.2) is an unprovenanced example, probably from southern England. The collar is decorated in a distinctive style of art found on many objects dating to the latest Iron Age and early Roman period in Britain. The pattern of the decoration was cast in the mould when the two parts of the collar were made. It was then completed by use of a fine engraving tool to fill some background areas by stippling. The pattern creates a mirror image of itself on each arm. 26 holes drilled into the metal would have originally held multi-coloured glass, which is now lost. Alternating red and white enamel inlay does survive in a series of triangular cells situated close to the terminals. The collar is made of cast brass. It was therefore probably made after the Roman conquest using Roman metal, but the decoration borrows from pre-Roman artistic traditions, as does the object type itself, as torcs and collars were used throughout the Iron Age period (Hautenauve 2005). A second type of neckring, this time a torc, is also seen. Torcs are often viewed as objects emblematic of Iron

34. Brit-art: Celtic Art in Roman Britain and on its Frontiers

Fig. 34.2. Collar from southern England. P&E 1963,0407.1. © Trustees of the British Museum.

Age peoples: many descriptions of Iron Age peoples by classical authors specifically mention torcs. Beaded torcs are a peculiarly British variant and distinctly different from pre-Roman torcs. They are made in two parts, a hoop and a front section, which typically forms about a third of the circumference and is decorated with beads (Hunter 2010, 91). Two types of beaded torc have been identified (Hunter 2006c, 370; 2010, 91). In the first type the beads are cast individually and threaded onto a rod. The beaded section of the second is cast as one block. Around 40 examples are now known distributed between the English midlands and the River Forth in central Scotland (Hunter 2008a, 133; 2010, 91). The torcs date to the 1st and 2nd centuries AD (Hunter 2006c, 369; 2010, 93) and although the two types overlap chronologically, it is thought that the type with the beads cast as one block are later than the torcs with individual beads. Whilst beaded torcs clearly have pre-Roman origins, the most intensive period of production and usage was in the early Roman period (Hunter 2008a, 133). An exceptional example of a beaded torc comes from Lochar Moss, Dumfriesshire (Fig. 34.3). It was found inside a bronze bowl, and the torc and bowl were originally contained within a bag and deposited in a bog, which explains the beautiful preservation of the torc. The torc is made of brass. 13 separate beads and 11 spacers survive; six beads and five spacers on one side, seven beads and six spacers on the other. It is thought that originally the torc would have had at least one or two more beads and an equivalent number of spacers. The hoop is richly decorated with a flowing scroll pattern. This has been cut out from a sheet of brass and is attached to the hoop with rivets. The Lochar Moss torc is an unusual example, with the majority of beaded torcs decorated with simple geometric patterns. Combined with the available dating evidence, this indicates that beaded torcs cannot simply be viewed as pre-Roman

319

Fig. 34.3. Beaded Torc from Lochar moss, Scotland. P&E 1853,1105.2. © Trustees of the British Museum.

relics. Rather they should be seen as new interpretations of a familiar pre-Roman object type. As we have seen, Wraxall collars and beaded torcs are broadly contemporary, with beaded torcs found in northern Britain and Wraxall collars in southwest England, although there is some overlap in south Wales (Hunter 2006c, 371). Objects of both types are often prominent in books on Celtic art, yet for both types the main period of usage is actually the Roman period (Hunter 2010, 94). These are both good examples of the borrowing of something quintessentially pre-Roman – the tradition of wearing neckrings – and its reinterpretation within the context of Roman Britain to make something new. The Dinnington torc, from near Dinnington, North Yorkshire, is a unique object which combines the bead end-decorations of beaded torcs with the central hinge of Wraxall collars (Beswick et al. 1990). It is a complex object and various components are made of different alloys, specifically gunmetal and brass, both of which are Roman metals. Dinnington is also located towards the southern boundary of the distribution of beaded torcs making the Dinnington torc a transitional object in multiple respects. As Garrow and Gosden (2012) explain, like so much art dating to the 1st and 2nd centuries AD, the Dinnington torc ‘…draws on material elements from the distant past [the form of a torc], and accumulates a variety of different object ‘styles’ within its form. But it also looks forward; it is in some ways a new and very different kind of object’ (Garrow and Gosden 2012, 326).

Massive metalwork In northeast Scotland in the 1st and 2nd centuries AD, between the firths of Moray and Forth, a distinctive form

320

Jody Joy

Fig. 34.4. Massive Armlets from Castle Newe, Scotland. P&EE 1946,0402.1-2. © Trustees of the British Museum.

of massive, cast metalwork is manufactured (Hunter 2006b). The most striking objects include so-called massive armlets, as well as large, zoomorphic spiral bracelets. The famous Deskford carnyx is the only related sheet-metal object known from this tradition, linked to the cast objects through its decoration (Hunter 2006b, 135). Around 30 massive armlets are now known (ibid.). These comprise a large pennanular hoop with expanded terminals and are sometimes decorated with multi-coloured enamels. Two particularly fine examples were discovered at Castle Newe in Aberdeenshire (see Macgregor 1976, nos 239 and 240) (Fig. 34.4). They were found at a settlement site, lying at the entrance of a souterrain, a kind of stone-lined cellar. Each armlet weighs over 1.5 kg and fitted a bicep with a circumference of around 37 cm. The armlets were made with some skill. They are made of cast metal and an inset surviving in one of the terminals of one armlet is decorated with red and yellow enamel arranged into a geometric pattern. The size of the armlets indicates that they were probably worn by men. Zoomorphic spiral bracelets are far smaller and were probably worn by women (Hunter 2006b, 151).

Fig. 34.5. Dragonesque Brooch, no provenance. P&E POA201. © Trustees of the British Museum.

Other types bearing pre-Roman style ornament include disc brooches with triskele patterns, and bow and fantail brooches (Hunter 2008a, fig. 8.3; Johns 1996, chapter 7).

Dragonesque brooches Brooches Some of the best examples of the art which forms the focus of this study are small items of personal adornment, particularly brooches. Many of these are singular examples of a particular type of brooch. Others show affinities to preRoman traditions as a group. Aesica brooches are named after the famous type-object from Great Chesters (Latin name Aesica), Northumberland. This very fine gilded brooch is ornamented with raised, swirling motifs on its head, bow and foot (Megaw 2001, fig. 388). Other examples of this brooch type bear relief decoration with affinities to pre-Roman designs (Johns 1996, 159). Trumpet brooches also include examples which are decorated with Celtic motifs, most notably the brooch from Carmarthen. There are however, many trumpet brooches which are quite plain.

Of the brooches decorated with Celtic art, the dragonesque particularly stands out. Indeed the Megaws selected an image of a dragonesque brooch for the front cover of Celtic Art (Megaw and Megaw 2001) (Fig. 34.5). The dragonesque brooch is formed of an S-shape with dragonheads at either end. Many are also decorated with polychrome glass enamel inlays, including blue, red and yellow. Enamelled motifs vary in form, from recognisable Celtic style circular and swirly motifs to Roman style squares and lattices arranged in geometric patterns. Nearly 300 dragonesque brooches are known, and they date from AD 50–175. They are found across much of England and southern Scotland, with a distinct concentration in northeast England and southeast Scotland, particularly in the vicinity of Hadrian’s Wall, and there is much regional variation (Hunter 2010, 95). A small number of finds have been found on the continent,

34. Brit-art: Celtic Art in Roman Britain and on its Frontiers

321

Fig. 34.6. The Illam Pan, Staffordshire. P&E 2005,1204.1. © Trustees of the British Museum.

particularly along the Roman frontier, perhaps connecting these brooches with the movement of Roman soldiers. What is striking about dragoneseque brooches is the lack of preRoman parallels. A very few plain, S-shaped brooches are known, but essentially this supposedly exemplary example of Celtic art is actually an object-type with origins firmly rooted in the Roman period (Hunter 2010, 96).

Roman objects decorated with Celtic motifs Celtic ornament is sometimes found on Roman objects, for example, on trullae, used for the service of food and libations, as well as seal-boxes used to contain wax impressions of signet rings (Henig 1995, 13). One of the most beautiful examples of a Roman object decorated with Celtic motifs is the Illam Pan, discovered by a metal detectorist at Ilam in Staffordshire in 2003 (Jackson 2012) (Fig. 34.6). The pan is made of bronze inlaid with polychrome enamel arranged into a curvilinear pattern. The red, blue and turquoise enamel inlays would have contrasted wonderfully with the original golden-colour of the bronze bowl. The pan belongs to a small group of colourful, enamelled bronze pans dating from the late 1st to 3rd centuries AD, with the decoration on a number of these vessels also being influenced by Celtic art traditions (Hunter 2012). The Meyrick helmet has been in the collections of the British Museum since 1872 (Jackson 1995, 67) (Fig. 34.7). It is a Roman helmet made sometime in the second half of the 1st century AD, and is thought to come from the north of England, but no other details of its origin are known. It is of interest here because it is also decorated with Celtic style motifs. The neck-guard of the helmet is the main area of decoration. This broad, flat area, designed primarily for protection, is decorated with a swirling repoussé design. This is arranged symmetrically in an open-lyre pattern. At either side of the centre of the design are two scored bosses, which would originally have been decorated with glass enamel. Ornamental bosses and a repoussé plate would also have

Fig. 34.7. The Meyrick Helmet. P&E 1872,1213.2. © Trustees of the British Museum.

adorned the side-plate on both sides of the helmet, although one is now missing and the other very badly damaged. These would have originally supported hinged cheek-pieces which are now missing. The helmet has been variously interpreted as belonging to a ‘Celtic noble in Roman service’ (Fox 1958, 119) and a ‘Roman auxiliary’ (Megaw 1970, 173). Roman military equipment was not as standardised as is sometimes imagined, and it is thought that there was a certain amount of latitude allowed in the diversity of equipment of the Roman army in the 1st century AD (Jackson 1995, 72). Recent studies of the Roman military (e.g. James 2001b; Gardner 2007) also show that it was not necessarily the ‘imperial war machine’ of popular literature. In reality the boundaries between ‘military’ and ‘civilian’ were far more fluid, complex and regionally varied. Whoever it belonged to, the Meyrick helmet is not simply a Roman object showing pre-Roman influences. It was made for a soldier serving in the Roman army and adopted and adapted some of the Celtic art styles of northern Britain to create a complex object with potentially highly nuanced and complex associations depending on the context in which it was viewed and the viewer. The Meyrick helmet is a hybrid, neither straightforwardly Roman nor pre-Roman (see also discussion of material from Newstead in Garrow and Gosden 2012, 300–304). Perhaps that was the intention. This has implications for the interpretation of other decorative metalwork in north Britain (including dragonesque brooches and beaded torcs), which, because of its location of use and deposition, has been referred to as ‘frontier art’, and has most often been discussed in relation to its association with the Roman military (e.g. Hunter 2008a). As Hunter (2008a, 136) argues, in this context ideas of pre-Roman

322

Jody Joy

or Roman could have become blurred and side-lined at a time of flux on the frontiers. Similarly, so could notions of ‘military’ and ‘civilian’. Frontier art was at the forefront of negotiations of identity in frontier zones, which perhaps accounts for its preponderance in regions where art was not previously made and used in the pre-Roman period.

Conclusions: What does this art do? It is hoped that, through the examples presented here, we have illustrated that through social interactions, and by borrowing from pre-Roman and Roman decorative and technological traditions, distinctive new art styles were developed (Hunter 2008a). These art styles are characterised by their variability and regionality, and are also prevalent beyond the frontiers. Clearly the art which is the subject of this paper can only be understood when set against the wider background of Roman expansion and incursion (Hunter 2006a, 106–107). However, it is argued that the production of regionally distinctive art styles, sometimes in areas where art was not previously manufactured, is not wholly due to the influence of Rome. Art draws on pre-Roman practices, often fusing pre-Roman and Roman techniques to create something new. This art also does not form a coherent single style or serve a single purpose. Rather, pre-Roman and Roman decorative traditions and technologies are drawn upon in manufacture, throughout the lives of the objects, as well as in deposition, in social negotiations which are regionally and temporally specific. This means that no single explanation for this art will suffice. Neither will explanations framed solely around ‘resistance to Rome’ or ‘becoming Roman’. Through detailed case studies, Hunter and Davis and Gwilt have already offered interpretations of particular regions or types of object and it is hoped that these will continue to be expanded. The rest of this paper seeks therefore to question what it is that makes this art such a potent social artefact with such great transformative potential. This is tied up not only with the properties of the objects but also the social milieus which they helped form and create. In her recent PhD thesis, Julia Farley (2012) draws upon Post-Colonial Theory (see for example Gosden 2004 for its use in archaeological contexts) and the literature examining North American colonial societies (e.g. White 1991) to examine the social role of gold in Late Iron Age Britain. It is worth quoting her at length: In both Iron Age Britain and North America, objects played a key role in mediating and facilitating colonial encounters. From gifts to trade goods to new technologies of exchange and diplomacy, objects and the social relationships which underpinned their circulation were a driving force for colonial

expansion, and a means through which both sides sought to understand, incorporate and ultimately (at least in America) dominate the other (Farley 2012, 135).

Farley (2012, 137) also draws on the work of Star and Griesemer (1989) who identify a category of objects they call ‘boundary objects’, which are ‘…both adaptable to different viewpoints and robust enough to maintain identity across them’ (Star and Griesemer 1989, 387). Boundaries are foci of active maintenance and interaction, and their definition and maintenance is a performative process dependent on people doing things in specific ways (Gardner 2007, 199). Identity is often performed between boundaries, be they social or ethnic, or based on gender or age. Identity is also a two-way process as categories are imposed on people, but through their actions people can reproduce or transform those categories (ibid.). The art objects which form the focus of this study make excellent ‘boundary objects’, providing multiple pre-Roman and Roman facets which could be drawn upon in social endeavours. For example, much of what we remember is not stored in our brains. What we remember is where to find the information, for example data stored in a phone book or an atlas. In pre-literate or semi-literate societies objects can carry information and act as aide-mémoires. While it is possible to find similarities between the decoration – for example on the Illam Pan, or the Meyrick helmet – and pre-Roman objects, it is important to remember that we are not sure how these similarities were viewed by an ancient audience. It is most likely that different viewers would pick out different elements from the art and this was no doubt the intention of the makers. Importantly, art was not just a product of these colonial encounters. It was active in creating them in specific contexts or ‘middle grounds’, created by the external influences and pressures of the Roman Empire, as well as local, socially significant factors (see Gosden 2004, 107). This agency is not secondary, where the real agency or initiative is seen to lie with humans or groups of humans. Rather, people at this time were constituted through their relations with other people and non-human actors, including art (see González-Ruibal et al. 2011). Rather than being viewed simply as a product of native resistance to Rome or a sign of becoming Roman, here it is argued that art was at the centre of many of the social transformations which took place in early Roman Britain and on its frontiers, which were structured around the particular non-human properties of the art. It was these properties built into this art at the time of manufacture with its references to the pre-Roman and Roman worlds which created its social potency. By manipulating art and technology in various ways, difficult transactions of ideas and renegotiations of identity could be accomplished. This was not a static process; rather it was fluid and changing. By subtly making something in a different way, or presenting an object in

34. Brit-art: Celtic Art in Roman Britain and on its Frontiers a different context, or depositing various collections of artefacts with different combinations of pre-Roman and Roman, meanings could be radically altered.

Acknowledgements I first met Vincent in Oxford in 2006 at a workshop dedicated to the study of Celtic Art in Britain. We shared a very enjoyable evening of lively discussion about all manner of subjects. Our contact has continued ever since and I will always appreciate his support, particularly in the very early part of my career. This paper has its origins in a brief summary chapter I was asked to contribute for the catalogue of the Die Welt der Kelten exhibition held in Stuttgart from 15 September 2012 – 17 February 2013 (Joy 2012) which is revised and expanded here. I thank Marion Uckelmann and Duncan Garrow for their comments on that paper, as well as the editors of the catalogue, particularly Prof. Dr. Ralph Röber, for their help and advice in writing that chapter, which they have kindly allowed me to re-produce here. Thank you also to Fraser Hunter and Ralph Jackson who allowed me to see their papers on enamelled vessels prior to publication and to Julia Farley, Duncan Garrow and Fraser Hunter for their useful comments on earlier drafts. All errors remain my own.

Bibliography Beswick, P., Megaw, M., Megaw, J. V. S., and Northover, P. 1990. A Decorated Late Iron Age Torc from Dinnington, South Yorkshire, The Antiquaries Journal 70, 16–33. Brailsford, J. W. 1975. The Polden Hill Hoard, Somerset, Proceedings of the Prehistoric Society 41, 222–34. Craddock, P., Cowell, M., and Stead, I. M. 2004. Britain’s first brass, Antiquaries Journal 84, 339–46. Creighton, J. 2000. Coins and Power in Late Iron Age Britain. Cambridge: Cambridge University Press. Creighton, J. 2006. Britannia: The Creation of a Roman Province. London: Routledge. Davies, J. L. and Spratling M. G. 1976. The Severn Sisters Hoard: a centenary study. In G. C. Boon and J. M. Lewis (eds.) Welsh Antiquity; Essays mainly on Prehistoric Topics Presented to H. N. Savory upon his Retirement as Keeper of Archaeology, 121–147. Cardiff: National Museum of Wales. Davis, M. and Gwilt, A. 2008. Material, style and identity in first century AD metalwork, with particular reference to the Seven Sisters Hoard. In D. Garrow, C. Gosden and Hill, J. D. (eds.) Rethinking Celtic Art, 146–184. Oxford: Oxbow Books. De Navarro, J. M. 1952. The Celts in Britain and their art. In M. D. Knowles (ed.) The Heritage of Early Britain, 56–82. London: G. Bell & Sons. Dudley, D. R. and Webster, G. A. 1965. The Roman Conquest of Britain A.D. 43–57. London: B.T. Batsford Ltd. Dungworth, D. B. 1996. The Production of Copper Alloys in

323

Iron Age Britain, Proceedings of the Prehistoric Society 62, 399–421. Farley, J. 2012. At the Edge of Empire: Iron Age and early Roman metalwork in the East Midlands. Unpublished PhD thesis, School of Archaeology and Ancient History, University of Leicester (http://hdl.handle.net/2381/10840). Fitts, R. L., Haselgrove, C. C., Lowther, P. C. and Willis, S. H. 1999. Melsonby Revisited: Survey and Excavation 1992–95 at the Site of Discovery of the ‘Stanwick’, North Yorkshire, Hoard of 1843, Durham Archaeological Journal 14–15, 1–52. Fitzpatrick, A. P. 2007. Dancing with Dragons: fantastic animals in the earlier Celtic art of Iron Age Britain. In C. Haselgrove and T. Moore (eds.) The Later Iron Age in Britain and beyond, 339–357. Oxford: Oxbow Books. Fox, C. 1946. A Find of the Early Iron Age from Llyn Cerrig Bach, Anglesey. Cardiff: National Museum of Wales. Fox, C. 1958. Pattern and Purpose: a Survey of Early Celtic Art in Britain. Cardiff: National Museum of Wales. Gardner, A. 2007. An Archaeology of Identity: soldiers and society in late Roman Britain. Walnut Creek: Left Coast Press. Garrow, D. and Gosden, C. 2012. Technologies of Enchantment? Exploring Celtic Art: 400 BC to AD 100. Oxford: Oxford University Press. Garrow, D., Gosden, C., Hill, J. D., and Bronk Ramsey, C. 2009. Dating Celtic Art: a Major Radiocarbon Dating Programme of Iron Age and Early Roman Metalwork in Britain, The Archaeological Journal 166, 79–123. Giles, M. 2008. Seeing red: the aesthetics of martial objects in the British and Irish Iron Age. In D. Garrow, C. Gosden and J. D. Hill (eds.) Rethinking Celtic Art, 59–77. Oxford: Oxbow Books. González-Ruibal, A., Hernando, A. and Politis, G. 2011. Ontology of the self and material culture: Arrow-making among the Awá hunter-gatherers (Brazil), Journal of Anthropological Archaeology 30.1, 1–16. Gosden, C. 2004. Archaeology and Colonialism: Cultural Contact from 5000BC to the present. Cambridge: Cambridge University Press. Gosden, C. and Hill, J. D. 2008. Introduction: re-integrating ‘Celtic’ Art. In D. Garrow, C. Gosden and J. D. Hill (eds.) Rethinking Celtic Art, 1–14. Oxford: Oxbow Books. Hautenauve, H. 2005. Les Torcs D’Or du Second Âge du Fer en Europe: techniques, typologie et symbolique. Rennes: Association du Travaux du Laboratoire d’Anthropologie de l’Université de Rennes 1. Henig, M. 1995. The Art of Roman Britain. Ann Arbor: University of Michigan Press. Hill, J. D. 2001. Romanisation, gender and class: recent approaches to identity in Britain and their possible consequences. In S. James and M. Millett (eds.) Britons and Romans: advancing an archaeological agenda, 12–18. London: CBA Research Report 125. Hunter, F. 2006a. Art in Later Iron Age Society. In C. Haselgrove (ed.) Celtes et Gaulois, l’Archéologie face à l’Histoire, 4: les mutations de la fin de l’âge du Fer. Actes de la table ronde de Cambridge, 7–8 juillet 2005, 93–115. Glux-en-Glenne: Bibracte, Centre archéologique européen. Hunter, F. 2006b. New Light on Iron Age Massive Armlets, Proceedings of the Society of Antiquaries of Scotland 136, 135–160.

324

Jody Joy

Hunter, F. 2006c. A Romano-British beaded torc fragment. In A. P. Fitzpatrick, A. B. Powell, P. Booth and Crockett, A. D. (eds.) The Archaeology of the M6 Toll 2000–2003, 369–71. Oxford: Wessex Archaeology. Hunter, F. 2008a. Celtic art in Roman Britain. In D. Garrow, C. Gosden and Hill, J. D. (eds.) Rethinking Celtic Art, 129–145. Oxford: Oxbow Books. Hunter, F. 2008b. A Romano-British Beaded Torc Fragment. In A. Fitzpatrick, A. B. Powell, P. Booth and A.D. Crockett (eds.) The Archaeology of the M6 Toll 2000–2003, 369–71. Oxford: Wessex Archaeology. Hunter, F. 2010. Changing Objects in Changing Worlds: dragonesque brooches and beaded torcs. In S. Worrell, G. Egan and J. Naylor (eds.) A decade of discovery: Proceedings of the portable antiquities scheme conference 2007, 91–107. Oxford: BAR British Series 520. Hunter, F. 2012. Frontier Finds, Frontier Art – Views of Enamelled Vessels. In D. J. Breeze (ed.) The First Souvenirs: Enamelled Vessels from Hadrian’s Wall, 85–106. Kendal: Cumberland and Westmorland Antiquarian and Archaeological Society. Jackson, R. 1995. The Meyrick Helmet: A New Interpretation of its Decoration. In B. Raftery, V. Megaw and V. Rigby (eds.) Sites and Sights of the Iron Age, 67–73. Oxford: Oxbow Monograph 56. Jackson, R. 2012. The Ilam Pan. In D. J. Breeze (ed.) The First Souvenirs: Enamelled Vessels from Hadrian’s Wall, 41–60. Kendal: Cumberland and Westmorland Antiquarian and Archaeological Society. James, S. 2001a. ‘Romanization’ and the peoples of Britain. In S. Keay and N. Torrenato (eds.) Italy and the West: Comparative issues in Romanization, 187–209. Oxford: Oxbow Books. James, S. 2001b. Soldiers and civilians: identity and interaction in Roman Britain. In S. James and M. Millett (eds.) Britons and Romans: advancing an archaeological agenda, 77–89. London: CBA Research Report 125. Johns, C. 1996. The Jewellery of Roman Britain: Celtic and Classical Traditions. London: UCL Press. Joy, J. 2008. Reflections on Celtic Art: a re-examination of mirror

decoration. In D. Garrow, C. Gosden and J. D. Hill (eds.) Rethinking Celtic Art, 78–99. Oxford: Oxbow Books. Joy, J. 2010. Iron Age Mirrors: a biographical approach, Oxford, BAR British Series 518. Joy, J. 2011. ‘Fancy Objects’ in the British Iron Age: Why Decorate? Proceedings of the Prehistoric Society 77, 205–29. Joy, J. 2012. Kunst in der Provinz Britannia. In R. Röber, M. Jansen, S. Rau and C. von Nicolai (eds.) Die Welt der Kelten: Zentren der Macht – Kostbarkeiten der Kunst, 489–97. Ostfildern: Jan Thorbecke Verlag. Macgregor, M. 1962. The Early Iron Age Metalwork Hoard from Stanwick, N.R. Yorks, Proceedings of the Prehistoric Society 28, 17–57. Macgregor, M. 1976. Early Celtic Art in North Britain: a study of decorative metalwork from the third century BC to the third century AD. Leicester: Leicester University Press. Megaw, J. V. S. 1970. Art of the European Iron Age. Bath: Adams and Dart. Megaw, J. V. S. 1971. A Group of Later Iron Age Collars or Neck-Rings from Western Britain, British Museum Quarterly 35, 145–56. Megaw, R. and Megaw, J. V. S. 1989 [2001]. Celtic Art: from its beginnings to the Book of Kells (new ed. 2001). London: Thames & Hudson. Revell, L. 2009. Roman Imperialism and Local Identities. Cambridge: Cambridge University Press. Star, S. L. and Griesemer, J. R. 1989. Institutional Ecology, ‘Translations’, and Boundary Objects: Amateurs and Professionals in Berkeley’s Museum of Vertebrate Zoology 1907–39, Social Studies of Science 19.3, 387–420. Stead, I. M. 1996. Celtic Art in Britain before the Roman Conquest (2nd ed.). London: British Museum Press. Stead, I. M. 2005. Iron Age Swords and Scabbards. London: British Museum Press. White, R. 1991. The Middle Ground: Indians, empires, and republics in the Great Lakes region, 1650–1815. Cambridge: Cambridge University Press.

35 ART IN CONTEXT: THE MASSIVE METALWORKING TRADITION OF NORTH-EAST SCOTLAND Fraser Hunter

One of the few Scottish finds which regularly graces syntheses of Celtic art is the Deskford carnyx head (e.g. Megaw 1970, no. 272; Megaw and Megaw 2001, 232). My first encounter with Vincent was over this beast when, as a callow youth freshly ensconced in the National Museum in Edinburgh, I embarked on a study of it. I sought the master’s advice – and indeed dared to question some of his views. Having been roundly slapped into place for sins of language and intellect (an ongoing facet of our relationship), Vincent offered some gems, and he has been a regular source of assistance, new perspectives, titbits of gossip, and goading to think more, think further, think better ever since. The returns for him have been slender, I fear – a few titbits of Caledonian Celtic art (though new discoveries have been too sparse for both our tastes), and occasional despatches from one of his adopted cities, where he spent his formative undergraduate years under Stuart Piggott. This paper offers some more formal payback for assistances willingly offered. It is not about the carnyx (though that would appeal to his musical tastes), but about its context – the massive metalwork tradition of north-east Scotland, its objects, origins, uses and setting. This paper picks the flowers from a more detailed reassessment (Hunter in prep.) and seeks a broader understanding of this remarkable metalworking tradition.

‘Speak weel o’ the Hielans but live in the Laich’: the landscapes of north-east Scotland The Scottish tourist industry would place our study area in the Highlands, but traditional Moray folk-wisdom says the

opposite: ‘speak weel o’ the Hielans but live in the Laich’ (Keillar 1993, 25). The Laich is the narrow strip of fertile low-lying arable land on the southern edge of the Moray Firth, between the uplands and the sea (Fig. 35.1). It is the land of malts and Macbeth. Today its produce goes into Speyside whiskies; in the past it supported rich farming societies, able to raise a man who would be king in the 11th century, and to create Celtic art in the 1st. The dour Grampian hills created an isolation from more southern areas, but between mountain and water a strip of good land runs down the east coast to the firth of Forth. This area was defined by Stuart Piggott as the north-east province in his four-fold division of the Scottish Iron Age (1966); the definition has seen little subsequent challenge, although questions remain over the relation between the lowlands and the upland glens or upper Strathspey, where passes lead over to the west coast (e.g. Harding 2006). The area is almost split in two where the mountains come down to the sea at the Mounth, south of Aberdeen; this natural division is often reflected in the area’s archaeology (e.g. Maxwell 1990, 45). In our period of concern, the Iron Age, we can play the distribution-map game to find both linking and dividing factors. A distinctive class of oblong forts with massive timber-laced stone walls neatly covers the province (Feachem 1966, 67–8; RCAHMS 2007, 100). Rarely do hillfort types provide such clear, bounded distributions; these ones compensate for this spatial compliance by rampant chronological bad behaviour, with a long-running debate over their date (see Alexander 2003), but recent evidence puts them in the middle Iron Age, c.300–100 BC (Cook 2010). Their distribution is close to that of our massive metalwork some centuries later; the

326

Fraser Hunter

Fig. 35.1. Distribution of massive-style metalwork, with key geographical features indicated. The inset shows the detailed north-east Scottish distribution, with regional groups marked.

two indicate some long-running version of a north-east identity. Other distributions split south from north around the Mounth; glass bead styles, characteristic of the northern area (Guido 1978, figs 34 and 36), or a short-lived Roman Iron Age floruit of architecturally-exotic brochs, an Atlantic

Scottish form adopted for display purposes in the southern area and parts of southern Scotland (Macinnes 1984). A two-fold split at the Mounth hides further subtleties. We will explore these shortly through the metalwork, but it is worth turning once more to the land. The open landscapes

35.  Art in context: the massive metalworking tradition of north-east Scotland and low ridges of the Laich of Moray are very different from the valley-worlds of Aberdeenshire and Banffshire. South of the Mounth the country opens up once more, with an expanding wedge of good arable land down to the Tay. This is constrained by hills around Perth, with valleys again dominant. To the south the wide valley of the Forth and the peninsula of Fife form ‘a beggar’s mantle fringed with gold’ (James VI’s term for Fife), open landscapes dissected by hills and bogs, with rich edges. We are among Piggott’s regions here, the level below his provinces which is rarely considered today; and while I would query details of his scheme, this pen-picture broadly correlates with his. By the core period of massive metalwork, hillforts were ancient monuments; there is no evidence of fort construction at this time (Alcock 1987, fig. 4; Ralston 1996, 135–7; Cook 2010, 87–8) and only very limited activity at a few small promontory forts (e.g. Wilson 1980). The area is characterised by settlements of timber-built roundhouses, often spectacularly large (up to 20m in diameter) and sometimes with semi-subterranean cellars (souterrains or earth-houses). Most settlements were unenclosed, but small enclosures (often for single houses) also occur (RCAHMS 2007, 82–96; Davies 2007). A combination of increasing excavation and metal-detecting has harvested fresh information and finds which let us contextualise the metalwork.

327

Fig 35.2: oval-style massive armlet, Auchenbadie, Banffshire (height at terminals 135 mm). © National Museums Scotland.

Massive metalwork The basis of the corpus was established in the late 19th century (Wilson 1851, 446–8; Smith 1868; 1881; Anderson 1883, 140–61; 1904), and was integrated into the wider British picture by Leeds (1933, 126–36), although curiously avoided by Fox (1958). Piggott’s identification of the Deskford carnyx (1959) was contextualised by Stevenson’s (1966, 31–5) broader consideration of Scottish Iron Age metalwork. The fundamental studies remain those of Morna MacGregor/Simpson (Simpson 1968, 1970; MacGregor 1976); subsequent papers have considered individual items and new discoveries (Hunter 2001; 2006a). Distribution and decoration give the tradition a regional coherence which is exceedingly rare in studies of insular Celtic art. This covers an underlying diversity, as we shall see, but the core area, from Forth to Moray Firth, is clear. The term ‘massive metalwork’ comes both from the sheer scale of many castings and the preference for prominent three-dimensional forms in the decoration. Designs are not complex but they are striking, with liberal use of slender trumpets, keeled diagonals and crescents. Personal ornaments predominate: armlets, bracelets and finger rings. The massive cast penannular armlets weigh up to 2kg, their flat broad hoops typically in three integrallylinked decorated strands: the expanded perforated circular

Fig 35.3. Folded-style massive armlet, Seafield Tower, Fife (height at terminals 68 mm). © National Museums Scotland.

terminals are sometimes filled by enamelled insets (Simpson 1968; MacGregor 1976, nos 231–50; Hunter 2006a). Their distribution spans the region, but with different types in the north and south reflecting different workshop traditions (Figs 35.1–35.3): oval examples (which have fold-over symmetry) are northern, folded ones (with rotational symmetry) southern. Such regional variety occurs also with the zoomorphic

328

Fraser Hunter

spiral bracelets, which are found overwhelmingly on Tayside in the southern area (Figs 35.1 and 35.4) (MacGregor 1976, nos 213–9). Finger rings (discussed below) show no clear typological division, but the updated distribution has two very marked clusters in Moray and Perthshire/ Stirlingshire with Aberdeenshire, a heartland of armlets, being a conspicuous gap. The situation is thus more complex than a north/south division: there are localised variations of expression within the broader picture (Fig. 35.1), reflecting some of the landscape units noted above. Only in the south do the distributions of armlets, bangles and rings overlap, but

only immediately north of the Tay. To the north the current distributions of armlets and finger rings are essentially complementary, the former concentrating in Aberdeenshire, the latter along the southern Moray Firth. The single northern bracelet may be an import from Tayside. Thus the distribution is coherent but not united: the art styles are shared across this zone, but no object type was common to all areas. Different groups within the area chose to make and use different object types within overall linked decorative schemes. At one level the decorative styles show a regional identity; at another, the types show more local affiliations. These ornaments were made to be used: almost all show use-wear. Consideration of internal diameters indicates that upper arms which fitted the armlets (interior diameters in the range 95–115 mm) would struggle to squeeze into the bracelets (typically 50–60 mm), and these are best seen as male and female ornaments respectively. For the bracelets, this is supported by comparison with Iron Age ‘jet’ bangles; they correlate with the smaller peak in the diameters of such bangles, whose bimodal distribution suggests different sizes for females and males (Hunter 2008a, 107). Rare smaller version of both armlets and bracelets are most likely for children or youths (e.g. Fig. 35.5) (Hunter 2006a, 151). The finger rings are consistently large, almost all having an internal diameter of 19–20 mm; in modern jewellery, these are men’s sizes. Thus, in the southern area there is evidence for a division of jewellery by gender; in the north, the focus is male, with armlets or rings, and only a single (imported?) bracelet.

Finger rings Fig 35.4. Zoomorphic spiral bracelet, Culbin Sands, Moray (height 89 mm). © National Museums Scotland.

Many of the new finds are finger rings: Simpson (1970) listed four, but the total has now grown to 17 (see Appendix

Fig 35.5. Large and small armlets, from Braes of Doune (adult) and Lismore (youth). © National Museums Scotland.

35.  Art in context: the massive metalworking tradition of north-east Scotland

329

Fig 35.6. Massive-style tankard handle, Castle Craig, Perthshire (length 85 mm). By courtesy of University of Glasgow SERF project. © National Museums Scotland.

and Figs 35.13–35.14). All bar one have lentoid or trumpethead mouldings where bezel and hoop meet, a typical motif of the massive tradition. Bezels show both openwork and enamelled decoration in considerable variety. Triskeles are most common, but several have enamelling similar to the massive armlet discs, with chequer or quatrefoil patterns. The decorative bezels point to Roman influence, as earlier indigenous finger rings were typically bronze spirals or plain bands in bronze or jet-like materials (Clarke 1971; Callander 1916, 222–8). They must have been developed very quickly after the initial design impetus, as one was found in the Roman fort of Strageath, Perthshire in a Flavian context (AD 83–7; Frere and Wilkes 1989, 154, fig. 79 no. 98; Hunter 2006b, 84–5). In developing a prototype from the classical world, these Scottish rings fall into a long tradition: from the early La Tène period onwards, Greek and Etruscan originals had inspired rings with bezels decorated in Celtic art styles (Déchelette 1914, 1264–70, esp. 1269–70; Jacobsthal 1944, 124–6, pl. 52; Megaw 1970, nos 58, 88, 138–9, 169; see also Schönfelder 2003; Polenz 2007).

Other Personal ornaments are the dominant product, but the tradition was broader. A few widely-scattered strap junctions were probably for horse harness (MacGregor 1976, nos 36–8). Other singletons are linked to the tradition by decorative style. A recent addition is a tankard handle from the lowland broch of Castle Craig, Auchterarder, Perthshire (Fig. 35.6), identical in form to one from Oxtrow, Orkney (MacGregor 1976, no. 291) but with massive-style slender trumpet decoration. Sheet-work is very poorly attested, the Deskford carnyx head being the best example, but a pair of (physically) massive tweezers from beyond the main distribution at Kettleburn (Caithness) may be linked, not only on account of their size but from the feathered decoration which can be paralleled on massive armlets (Fig. 35.7). So-called massive terrets have traditionally been

Fig 35.7. Tweezers, Kettleburn (Caithness), perhaps in the massive tradition (length 111 mm). © National Museums Scotland.

linked to the style as well, but their dating is later (3rd–7th century AD) and their distribution markedly broader (Hunter 2010a, 100–3). An unusual zoomorphic bow brooch from Bin Hill, Garmouth, Moray, could be linked to the massive tradition in its use of animal-head terminals, yellow enamel and zig-zag decoration, although its form seems closer to a series of unusual La Tène II brooches (Hunter 2009, 146–7).

When and why? Detailed evidence for dating will be reviewed elsewhere (Hunter, in prep.), but a combination of stylistic analysis, contextual associations and technological information indicates a floruit of the later 1st–2nd centuries AD – the earlier Roman Iron Age. Examples from excavated sites are consistently of this date, as are associations from hoards, while the reused Roman metal in many of the alloys indicates a post-invasion date (in this area, post c.AD 80), as there is minimal artefactual evidence for contact before conquest (Hunter 2007a, 22). The finger rings, inspired by Roman prototypes, also indicate a post-contact date. Yet this art should not be seen as a purely post-invasion

330

Fraser Hunter

phenomenon; our dating, as too often, is straitjacketed by these Roman connections. Some of the alloys are traditional Iron Age tin-bronzes, and while this can only be suggestive, better evidence comes from further south. A burial from Snailwell (Cambridgeshire) produced a zoomorphic spiral armlet (Lethbridge 1953) which Fitzpatrick (1989) has convincingly seen as evidence of long-distance contacts to Tayside. The dating of the ceramics in the burial (AD 50–65; V Rigby, pers. comm.) is pre-conquest in northern terms. A decade or two may seem small beer, but it confirms that massive metalwork started before direct contact with the Roman army. There may be few pre-Flavian finds in the area, but through networks of contacts they were feeling the ripples in the pond.

Production Cast products dominate the tradition, often large castings which would be tricky to carry out; indeed, many armlets show very skilful post-casting repairs, barely visible to the uninitiated eye. Non-ferrous metalworking was a rare skill in the area: casting evidence of this general date is known from under 10% of 235 excavated later prehistoric sites, and only ten of these are likely to be Roman Iron Age in date. Thus production was restricted, although the products were widespread. The skills required were those not just of the bronzesmith but the enameller as well. The enamelling technology is both varied and innovative. Finger-rings provide the best evidence. Normal champlevé is attested, with cast-in fields, but the chequerboard patterns show no such field divisions, with blocks of glass being set in solid form and fused. A recent find from Castle Stuart, near Inverness, shows an even more complex picture (Fig. 35.13); although only partly preserved, the design shows a red substrate with inlays sunk into it. The technique is more familiar from glass-working, and indeed one of the inlays appears to be a reused multicoloured bead fragment. Such technical skills are seen also in the surviving enamelled armlets, with Castle Newe using blocks without cell divisions while Pitkelloney has cloisonné field divisions (V Rigby, pers. comm.). Red and yellow are the dominant colours, with blue attested only in the eyes of the Culbin bracelet. The northern area was known for its glass-working (see below), and this link between technologies is attested at Culduthel (Inverness-shire). Here, excavation of a workshop area of 2nd century BC–2nd century AD date provided evidence of a wide range of technologies, including bronzecasting, glass-working and enamelling in the same area (Murray forthcoming; Hunter forthcoming a and b). There was also evidence of sheet-working, which is otherwise difficult to identify. Although there is no evidence of massive metalwork from the site, Culduthel provides a model of

central production sites consistent with the wider evidence that non-ferrous metalworking was a restricted craft. Analysis provides some further insights into technological process. Of sheet products, only the carnyx has been analysed: this showed careful control of different alloys, selected to create striking visual contrasts between bronze and brass (Hunter 2001). The cast alloys show much more variety; even armlet pairs are not consistently of the same alloy. There are some technological patterns: analysed finger rings were heavily leaded to facilitate casting, while the larger items have low lead levels, probably as a compromise between ease of casting and the need for extensive postcasting working and bending to shape (which was easier with a low-lead alloy). The variety of alloys is likely to reflect varying supply sources. The data can be categorised into three groups, which may represent a chronological progression: no or minimal zinc (typical Iron Age alloys), high zinc (primary reuse of Roman metal), and lower zinc, representing volatilisation of the zinc through continued recycling. Davis and Gwilt (2008, 149–50) have recognised similar patterns in Welsh late Iron Age metalwork. This is, of course, an argument lacking independent chronological control, but it hints at some time-depth to the tradition (discussed more fully in Hunter, in prep.).

Patterns and biases of deposition In any study of decorative metalwork, processes of deposition and recovery can cast a critical bias on interpretation. With massive metalwork, a remarkable number of finds in northeast Scotland come from certain or probable settlement sites, suggesting a tradition of deliberate on-site deposition of valued material (Hunter 1997). There are clear patterns – (male) armlets and (female) bracelets were not mixed, and typically one or two armlets or a single bracelet were deposited, suggesting these were offerings linked to specific individuals in a domestic setting. We have little evidence for how they were deposited; for instance, was it a private offering by or for the individual, or in a public ceremony linked to them on their death or another rite of passage? Contextual details are frustratingly sparse, but some appear to be linked to the abandonment of sites: the Castle Newe armlets came from the entrance to a souterrain, while the Hurly Hawkin bracelet was in the shell of an abandoned broch (Proceedings of the Society of Antiquaries Scotland 6 (1864–6), 13–14; Hunter 1997, 115–6). Such deposits on settlement sites were clearly infrequent, as armlets and bracelets are rare finds today. In the writer’s 20+ years at the National Museum, eleven finger rings have been found (four excavated, seven metal-detected) but, sadly, only one massive armlet. While some rings (five of the total of 17) are intact and may well have been deliberate deposits, the remainder were broken; they seem to show use-wear rather

35.  Art in context: the massive metalworking tradition of north-east Scotland than purposeful damage, and are likely to be accidental losses from on- and off-site contexts. Deposition on settlement sites may have been the local norm, but when massive metalwork left its home area the rules changed – where the context is known, exported examples are from wet or other remote settings, fitting local practices in these areas. Off-site deposition was the exception in the north-east, and the rare examples are revealing. One exception was the Deskford carnyx. This was deposited in a peat bog: excavations showed it was the most spectacular of a long-running tradition of ritual offerings (pot, butchered animal bone and quartz pebbles). An adjacent promontory had been cut off from the settled landscape by a palisade and was kept free of normal domestic use, acting like a temenos for this sacred site (Hunter 2001). Here both setting and find are non-domestic. It seems carnyces were not linked to individuals – they are never found in burials, for instance – but were symbols of a group’s identity, and their deposition thus took place in special areas away from the domestic sphere. Such objects may have been seen as too powerful, even dangerous, to be buried near the settlement. Likewise, vessels such as cauldrons, while rare in the area, come from off-site contexts (Hunter 1997, table 12.6), and these too are best linked to groups rather than individuals. The specific location of Deskford is interesting – in a subsidiary valley which links one of the major rivers to the coastal zone, and between the distributions of groups using armlets and those using finger rings, suggesting it may have functioned as a site where various communities came together. The other major off-site find, the hoard from Bunrannoch, breaks all the ‘rules’ – remote from settlements, and mixing male and female ornaments (both a bracelet and an armlet were found in a vessel along with lost and tantalising ‘smaller articles’ (Anderson 1904)). Its unusual nature is matched by its unusual location, away from the core distribution in the uplands at the foot of Schiehallion – one of the most striking hills of the southern Grampians, its triangular profile catching the eye for miles around. The quantity of finds implies a connection to more than one individual, making it more of a communal or combined act, and the location is perhaps significant in this regard. The name means ‘fairy hill of the Caledonians’ (Watson 1926, 21). Was this where groups came together and offerings were made to cement alliances? Fiction perhaps, but both context and contents indicate this was different from the normal pattern of deposition. We thus have four different processes forming our record: the deliberate burial of personal ornaments on settlement sites, linked to individuals; the deliberate burial of communal items in significant natural locations; locallyvaried deposition of metalwork as it passed out of its core zone; and accidental losses of smaller items through usewear.

331

Precursors and stimulus Massive metalwork has little recognisable parentage in its homeland, although occasional metalworking debris of early-middle Iron Age date shows that copper alloys were being used, while gold ribbon torcs can now be recognised as a tradition of the area. Long assumed to be Bronze Age, Richard Warner’s arguments to subdivide them into a Bronze Age and an Iron Age group based on analysis, associations and typological variety (1993, 111–2; 2003; 2004) have been emphatically confirmed by the gold hoard found at Blair Drummond (Perthshire) in 2009, with two ribbon torcs in association with a French-style tubular torc and an unusual ring-terminal torc (Hunter 2010b). The Scottish distribution of ribbon torcs focuses in the northeast, and typological differences in the terminals of Irish and Scottish finds suggest regional manufacturing traditions (Coles 1968, 170–1). Dating evidence needs reconsideration in the light of recent finds, but is best seen as around the 4th–1st century BC, predating the massive tradition. It indicates a sheet gold-working tradition in the area, but provides little background for massive metalwork in terms of technology, form or decoration. The latter was highly innovative. Previous writers have recognised its style links to southern Britain and Ireland (MacGregor 1976, 184–5; Piggott 1959, 31–2), and the metalworkers of the area were clearly open to external influences, as seen in the creation of the carnyx (a local version of a Europe-wide type) and the influence of Roman finger rings. Yet these influences were absorbed and reinvented into new, distinctively local forms: the finger rings are transformed into the massive tradition, the carnyx is technically unlike any other examples known, while massive armlets have no convincing parallels. We see here the products of a highly creative society. The stimulus for this is likely to be two-fold. A growth in ornamental material in the British late Iron Age (Hill 1995, 85; Hunter 2007b, 289) points to societies and individuals increasingly concerned with marking individual variation and affiliation. This was certainly the case in the north-east, where a vibrant tradition of glass jewellery developed in the later 2nd/1st centuries BC (see below). Yet a strong catalyst is likely to have been the perceived and actual threat of Roman invasion, with the opportunities and uncertainties it presented; this is seen as a key stimulus for the flourishing of British Celtic art at this time (e.g. Davis and Gwilt 2008, 147, 165, 177–8). Dating evidence tends to be biased by Roman associations, although the exported spiral bracelet from Snailwell shows that the habit began before conquest, by c.AD 50–65. How much earlier, if at all, is uncertain, but this puts it into a context before any significant evidence of direct contact, although societies in the area were clearly becoming aware of Rome. An existing local trend to increasing use of ornament as a key part of social interaction was heightened in this fast-changing place on the margins of Roman influence.

332

Fraser Hunter

Fig 35.8. The movement of massive metalwork. (The source of the York fastener is not clear as the type cannot be allocated to a specific sub-region.)

It is worth tracking what happened to massive metalwork in comparison to the art of central Britain. This area also saw a flourishing of Celtic art around the time of conquest, and this was key to the negotiation of new identities in the complicated world of the Roman frontier (Hunter 2008b). Objects of typically Iron Age styles commonly come from

Roman contexts, and the decorative traditions continued to develop on Romano-British objects. This was not the case with massive metalwork. While it drew on Roman influences in aspects of its form (for finger rings) and raw material, it did not get drawn into these frontier identities. There are virtually no examples from Roman sites – a

35.  Art in context: the massive metalworking tradition of north-east Scotland finger ring from the fort of Strageath and a strap junction from York are the only exceptions. This may, of course, be an accident of history, since north-east Scotland stayed essentially beyond the Roman frontier. Yet it is noteworthy that the area was otherwise heavily drawn into Roman politics and material culture, but kept its art separate. Roman finds are commonplace from Iron Age sites in the area; the inhabitants were intimately involved in Roman frontier politics (Hunter 2007a). Yet it seems there were things it was considered appropriate to mix, and things which it was not. Celtic art was never hoarded with Roman finds in the area, in contrast to the south of Scotland. Certain spheres of life were separated from others, even when the objects were made from recycled Roman objects and, in some cases, influenced by Roman designs (Hunter 2013, 23–4).

More than massive: contemporary styles Massive metalwork was mobile; it is found in Sutherland, Shetland, Skye, Lismore, south-east Scotland, northern Ireland and eastern England. The contacts are often counterintuitive: the armlets from northern Scotland are of the southern folded type, not the nearby oval one, while the reverse is true of the two armlets in southern Scotland, suggesting a preference for more exotic longer-range contacts over neighbouring ones (Fig. 35.8). This is seen also in the links to East Anglia, with spiral bracelets from Snailwell, Cambridgeshire (Lethbridge 1953) and an unpublished fragment from Charsfield, Suffolk (J. D. Hill, pers. comm.). The Snailwell bracelet is worthy of note on other grounds. It is the exception to the gender associations noted earlier, being linked to a male grave (an assumption based on a shield boss; the cremated remains were not sexed); this is consistent with its larger diameter. Yet there are oddities to it – the shape is uneven, the inset eyes invisible as worn (Lethbridge 1953, 30). It seems the bracelet was transformed and reworked in its new homeland, from a female to a male ornament. Temptations to see this as a double-edged gift, with the northern tribes giving a proud male warrior a woman’s ornament fit for effete southerners, would simply be an imposition of modern prejudices. These connections are hard to spot within the north-east because the hoarding tradition prioritised local metalwork over anything else, and settlement finds are rare. Recent metal-detecting finds appeared to offer a way around this: a range of fragmentary finds in non-massive-tradition styles were interpreted as settlement finds and stray losses which reflected contacts with southern Scotland and further afield (Hunter 2006a, 151–7). This proved too simple. Excavations at the Culduthel (Inverness) workshop provided the complication. Moulds were too fragmentary to identify, but the one clearly unfinished item was a cruciform strap junction decorated with small-cell enamelling (Fig. 35.9)

333

which owed nothing to the massive tradition. It looks like a product of central Britain, but was clearly being made on site as the fastening loop had not been cleaned out after casting. More circumstantial evidence comes from other items of horse harness: there is now a cluster of unusual button-and-loop strap fasteners of openwork teardrop form in the area, a type which is otherwise rare (to the examples in Hunter 2006a, ill. 16c, 19c, add two unpublished finds from Clarkly Hill, Moray). This evidence allows more complicated stories to be told. It seems societies in the north-east shared in more widespread traditions of decorated horse gear and thus marked connections to central Britain (in contrast to Atlantic Scotland, where such finds are unknown); but for personal ornaments they developed their own tradition, the massive style. In other areas of social display they made affiliations to neighbouring areas, but personal display provoked a clear expression of regional (and sub-regional) identity; whether this made long-lived patterns more visible or was a new creation of the moment remains for debate.

A decorated world? Massive metalwork in context It is over-easy to focus on decorated metalwork at the expense of other ornament (Sharples 2008). What was the decorative world of the people who saw and used massive metalwork? As ever, our picture is biased by the non-survival of colour and of organic materials – in this area, even bonework is rarely preserved. But decoration is otherwise rare. Pottery was almost never decorated, and stone only rarely: there are a few decorated quernstones, with a circle around the hopper or with radial decoration (McLaren and Hunter 2008, fig. 4), while only a single painted pebble is known, with a simple pattern of dark dots (unpublished, from writer’s excavations at Birnie). Ornament is markedly more common on the stone cups or lamps common in the area, about a fifth of which carry decoration (Loudon 2000, 29; contra Steer 1956, 244). This is mostly simple linear geometric ornament, typically a band (sometimes raised) of herringbone or zig-zags under the rim (Fig. 35.10a; cf. Ralston and Inglis 1984, 35; Close-Brooks 1972). The herringbone patterns are also found on massive metalwork, while an example from Knockargity, Tarland has a bossed band suggesting a skeuomorph of a bronze vessel (Fig. 35.10b; cf. MacGregor 1976, nos 292, 305, 308). On one from Newton of Auchingoul, Banffshire, the banded herringbone pattern covering the surface and zoomorphic terminal are reminiscent of spiral bracelets (Fig. 35.10c) MacGregor 1976, no. 334). These links to metalworking decoration are confirmed in rare examples with crude curvilinear decoration from Howbury (St Andrews, Fife; Proceedings of the Society of Antiquaries Scotland 45

334

Fraser Hunter a

b

Fig 35.9. Cruciform strap junction, Culduthel (Inverness-shire). Height 61 mm. © National Museums Scotland.

(1910–11), 222–3) and Hill of Syde, Aberdeenshire (Ralston and Inglis 1984, 37). There was no absolute separation of decorative styles by material, although the overlaps are small and rare. Other metalwork of the period is quite plain: brooches are virtually unknown, pins are simple projecting ringheaded types, and other ornaments such as spiral rings or belt hooks bear only occasional simple linear geometric decoration. The trumpet and crescent decoration of massive metalwork was a very restricted phenomenon, found only on very selected material. Significantly, the other group of material which was regularly decorated is glass jewellery (Fig. 35.11) – which had a technical connection to metalworking in the use of enamelling. Not only do beads provide splashes of colour in our rather monochrome view of the Iron Age, but their curvilinear and spiral decoration resonates with Celtic art styles. There are three relevant categories of finds: triangular glass beads of Guido (1978) class 13, with interlinked spiral decoration; class 14, with a range of loops, swirls and rays; and small glass balls, perhaps playing pieces, with inlaid spiral-decorated eyes. The beads concentrate predominantly north of the Mounth, while the much rarer glass balls are more widespread (Ralston and Inglis 1984, 41). Dating evidence focuses on the Roman Iron Age, but examples from Culduthel (Inverness) and Dun Bharabhat (Lewis) are linked to radiocarbon dates of the later 2nd century BC–early 1st century AD (Hunter forthcoming c). Given the technological connection, the conceptual

c

Fig 35.10. Decorated stone cups or lamps. a, with herringbone and zig-zag patterns, Knockargity, Aberdeenshire (length 122 mm). b, with skeuomorphic bossed band, Knockargity, Aberdeenshire (length 125 mm). c, with snake-like decoration, Newton of Auchingoul, Banffshire (length 133 mm). © National Museums Scotland.

link of curvilinear decoration is perhaps not surprising; it is noteworthy that glass beads from other areas similarly show curvilinear patterning which is otherwise rare outwith metalwork (e.g. Guido 1978, plate I–II). Thus, the technically complex materials of glass and copper alloy were decorated

35.  Art in context: the massive metalworking tradition of north-east Scotland a

b

335

rather differently from other materials which, if decorated at all, tended to be linear geometric. The other decorative strand was a novelty: Roman imports. These opened people up to new imaginative worlds and materials: brooches decorated with silvering or tinning, pottery with shiny slips and decoration, coins and pottery with naturalistic images. Roman brooches are widespread in the area, with a preference for Romano-British styles which reflected indigenous tastes; functionally these complemented rather than supplanting massive-style ornaments. The frequency of brooches compared to massive metalwork confirms the latter were more restricted and prestigious. The impact of other materials, like samian or silver coins, is as yet harder to gauge, though in the long term such naturalistic decoration may have influenced the development of Pictish art (Hunter 2007a, 38–42).

Changes

c

Fig 35.11. Other ornamental material – glass jewellery. a, class 13 beads, Culduthel (Inverness-shire); diameter of largest, 19 mm. b, class 14 beads; clockwise from left, Aberdeenshire, Cawdor and Wigtownshire (the latter an import); diameter 12 mm. c, playing piece with spiral inlays, Birnie (Moray); diameter 11 mm. © National Museums Scotland

Defining the lifespan of the massive metalwork tradition is tricky. The earliest good evidence is mid-1st century AD, and no other associations are any firmer than 1st–2nd century. It could all have been made in less than a generation, or over 150 years. There is no evidence for its manufacture or use later than the 2nd century: this fits the pattern of central British Celtic art, which is found regularly in 2nd-century contexts but rarely thereafter (Garrow and Gosden 2012, 70–9). Other sources of evidence point to major changes in the north-east in the 3rd and 4th centuries, with the end of the long-lived roundhouse tradition, often a shift in settlement location, and a general difficulty in finding where people were living and what they were doing (Hunter 2007a, 42–50). This can plausibly be linked to major (and chaotic) social and political readjustments caused by the proximity and policies of the Roman world; it is the archaeological correlate of the emergence of the Picts. The processes are much debated (e.g. Mann 1974; Hunter 2007a; Fraser 2009, 30–61), but it is worthwhile examining any material traces. The massive metalwork tradition may have ceased, but other styles developed and new ones emerged. The best evidence for continuity comes from minor items, notably pins: the development of projecting ring-headed pins into more decorative forms, apparently in the 3rd and 4th centuries AD, has long been a typological classic (Stevenson 1955, 288–92). Along with other artefact types which mostly lack these clear links to earlier traditions, they define a new suite of material culture marking new connections stretching beyond the north-east to cover northern Britain, Ireland, and into Britannia (Laing and Laing 1986; Hunter 2010a). The regionality of massive metalwork had been replaced by much more outward-looking styles, reflecting the range of contacts prioritised by these groups. There may be clear dislocation in aspects of the archaeological record, notably

336

Fraser Hunter

Fig 35.12. Vessel handle, Culbin (Moray). Length 48 mm. © National Museums Scotland

settlements, but material culture provides signs of continuing development. This is true also in the stylistic development of Celtic art. The significance of a small find from Culbin Sands (Moray) has been widely overlooked (MacGregor 1976, no. 289). It is most plausibly a vessel handle, but its decoration is an interesting hybrid (Fig. 35.12). The handle itself has a lentoid motif flanked by slender trumpets, familiar from the massive tradition, but the zoomorphic terminals are those of late Roman Iron Age and early Mediaeval pins and brooches (e.g. Fowler 1963, figs 1–2). In this rare survival we see the development of later styles of Celtic art from the art of the Roman Iron Age.

Discussion The massive metalwork tradition is woven from different strands, and the variety in its fabric reflects social variety in north-east Scotland in the late and Roman Iron Age. At one level, shared styles and widespread artefact types reflect a shared identity over a large area from the Forth to the Moray Firth, but teasing typology and distribution apart shows that this is made up of smaller distributions, overlapping or complementary, which reflect different uses of the metalwork in smaller areas. North of the Mounth, along the Moray plain, (male) finger rings were the dominant type, while in Aberdeenshire it was (male) armlets. In Angus, the wider spectrum of material included female ornaments (the spiral bracelets) as well as male. Regional differences are seen in other contemporary developments, such as the building of brochs in Angus and adjacent areas, distinctive and exotic buildings often rich in indigenous and imported

material culture. The link of massive metalwork to such sites is striking, though not absolute – four brochs have produced such finds, but only one souterrain in the same area. It suggests increasingly hierarchical societies in the Angus area, or some groups making a claim for control and using metalwork as a medium for this. The chronology in relation to the expanding Roman world provides a plausible context for such very visible symbols of identity, though to see them as acts of resistance is to underplay a complicated relationship with the Empire where Roman material could be accepted and used, reused, absorbed as influences and, in some contexts, rejected. Studies of such visually rich ornaments elsewhere have rightly questioned any uncritical assumption that they necessarily represent an elite, and discussed other possibilities; for instance, they could reflect adopted roles required by the community at particular moments, such as community leader, war champion or religious guide (Garrow and Gosden 2012). I would see the massive metalwork as more directly linked to individuals, for a variety of reasons. They show extensive wear and thus extensive use, not just on special occasions, while the wear is typically more pronounced on one side, suggesting habitual (individual?) preferences. The frequent link to settlements connects them to the people who inhabited these sites, while their deposition, typically singly or in pairs, suggests a link to individuals. The rare occurrence of child- or youth-sized versions of bangles and armlets suggests a concern with inheritance and genealogy. Such concern with individual status fits wider evidence for a growing emphasis on personal ornament and other indicators of inter-personal or inter-group differentiation, from the architectural (such as brochs and enclosed roundhouses) to the material (access to restricted craft skills and other imported goods). For me, the evidence indicates that jewellery in the massive tradition was used as an indicator of personal position within a defined regional identity, a process accelerated by proximity to the Roman world. It also opens windows onto a world where, at least in some areas, men, women and children were all appropriate wearers of prestigious metalwork, a rare situation for the period. Yet this is not to see these as absolutely rare, elite ornaments. Within their core area they are relatively common – some 18 armlets, six bracelets and 15 finger rings, with the numbers of the latter in particular growing dramatically. This is consistent with the settlement pattern of numerous small-scale house-clusters, lacking major central places such as hillforts. In the south (closer to the Roman world), the brochs developed as part of an increasingly hierarchical society, but this was not the case further north. Any ‘elites’ were small scale and potentially short-lived, as the limited lifespans of most southern brochs suggest (Macinnes 1984). The unusual hoards – Bunrannoch and Deskford – have

35.  Art in context: the massive metalworking tradition of north-east Scotland

337

a

b

Fig 35.13. a, massive-style finger ring, Castle Stuart (drawn by Alan Braby). b, detail of bezel. © National Museums Scotland

highlighted other aspects of the tradition, with the deposition of material away from settlements reflecting, perhaps, some of the places and rituals which bound these groups together. This gives these decorated pieces considerable social impact, in signalling the power and position of particular people and groups, the marking of large-scale affiliations as well as local identities, and the building of links between groups in the area. Above all, artistically this was an innovative style, drawing in influences and changing them to create striking new artefacts. In this the Deskford carnyx, international in concept but north-eastern in style, is emblematic – but it fits into this wider, innovative and versatile tradition which sparked into life beyond the edges of the Roman world.

Appendix: recent finds of massive-style finger rings Massive-style finger rings were first synthesised by Simpson (1970) and updated by Hunter (1998, 344–5; see also Hunter 2006a, 152 (Middleton, Moray)). Excavated finds from Castle Craig (Perthshire) and Leckie (Stirlingshire) will feature in the excavation reports, as will one from Clarkly Hill which awaits conservation. Recent metal-detecting finds are reported here.

Castle Stuart, Inverness-shire (Fig. 35.13) Enamelled finger ring which has seen heavy wear: the D-sectioned hoop broke in antiquity, and the doubletrumpet mouldings marking the hoop-bezel junction are worn (one was slightly deformed in the casting). The single large recessed cell on the oval bezel contained a complex enamelled design, only half of which survives. Into an opaque discoloured red, slightly bubbly substrate were set multi-coloured inlays of glass. Parts of three survive, none

complete. The best-preserved is a crescent or pelta, the ‘stalk’ leading into the centre lost. Uniquely, these inlays are multi-coloured and must represent the reuse of remelted pieces of glass, probably broken beads. The largest has an area of opaque white interleaved with translucent pale blue in one part, and opaque red interleaved with translucent near-clear glass in the other, suggesting the reuse of two different glass fragments. The shape of the next inlay anticlockwise is uncertain; it may have been similar to the first, but runs to the edge of the field. It primarily comprises opaque yellow interleaved with translucent clear glass; an opaque red/clear border along one edge may have been created when it was inserted (a number of the inlays show similar border effects). Only an interleaved red/clear corner of the third inlay survives. The overall design is unclear, but if all the attested inlays were the size of the best-preserved one then there would be room only for three; the crescentic one could be seen as an arm of a slightly irregular anti-clockwise triskele. The inlays are likely to be reused fragments of beads – such broken decorated beads would be difficult to remelt owing to the use of multiple colours, so this would be a practical way to reuse them. It is vivid testament to the technical skills of the area. The findspot (NGR: NH 74 50) lies in the same field as a later prehistoric settlement (NMRS NH75SW 35), and it is likely to derive from this site. L 25, W 15.5, H 19 mm; enamelled area 15 × 13 mm; hoop 3.5 × 3.5 mm, internal W 18.3 mm. Treasure Trove case TT 181/12; allocated to Inverness Museum.

Clarkly Hill, Moray Four rings are known from the site, three from metaldetecting (reported here) and one from excavation (still undergoing conservation). All will go to Elgin Museum in due course. (NGR: NJ 131 675.)

Fraser Hunter

338

Fig 35.14. Massive-style finger rings. a–b, Clarkly Hill. c, Coldstream (drawn by Marion O’Neil).

1: small circular bezel with raised rim; the flat interior has a channel round the edge to assist in holding the lost enamel. Slight nicks flanking the hoop where it expands to meet the bezel create a trumpet-head effect, protruding above the bezel; they are worn and polished from use. The oval-sectioned hoop broke in antiquity. Some corrosion pitting; margins of bezel damaged. External D 23; bezel 11.5 × 14 (field D 11); H 10; internal D at least 21; hoop section 2.5 × 1.5 mm (Fig. 35.14a). 2: very pronounced double-trumpets at junction with bezel, which has openwork peltae flanking a central figureof-eight-shaped stalk with a central shallow dot in each lobe, perhaps once enamelled. About half of oval-sectioned hoop lost. Surface heavily corrosion-pitted. Bezel L 28, W 20; H 20; internal diameter 18.9 mm (Fig. 35.14b). 3: enamelled oval bezel; edge and much of design lost. Double-trumpet mouldings at junction with lost hoop. Ring-and-dot top centre, in reserved metal; remains of a tapering field curving around the top contain translucent pale blue enamel. A further blue fragment survives on the left edge, in remains of a marginal channel, and there may be a peltate field on the right hand side, but the overall design is unclear. 22 × 13 × 5 mm; internal diameter at least 18.5 mm (not illus.).

Coldstream, Berwickshire (Fig. 35.14c) Oblong openwork bezel carrying a variant yin-yang design; opposed peltaes, joined at the stalk, spiral clockwise into the bezel rim. Two conical perforations flank the stalk. Doubletrumpets mark the junction of the bezel and the D-sectioned

hoop. Some wear and edge-damage, but intact. External D 23 mm, H 22.5 mm, bezel; W 12.5 mm; internal D 17 × 15 mm. NGR: NT 83 39. Found by metal-detecting c.250 m ENE of a bivallate cropmark enclosure (NMRS NT83NW 18). NMS FA 131.

Acknowledgements I am grateful to David Clarke, Martin Goldberg, Jody Joy and Mhairi Maxwell for very helpful comments on an earlier draft of this paper, to Marion O’Neil and Alan Braby for illustrations, and to Neil McLean for taking some excellent new photos at very short notice.

Abbreviations NMRS: National Monuments Record of Scotland

Bibliography Alcock, L. 1987. Pictish studies: present and future. In A. Small (ed.) The Picts: a new look at old problems, 80–92. Dundee. Alexander, D. 2003. The oblong fort at Finavon, Angus: an example of the over-reliance on the appliance of science? In B. Ballin Smith and I. Banks (eds.) In the Shadow of the Brochs, 44–54. Stroud: Tempus. Anderson, J. 1883. Scotland in Pagan Times: The Iron Age. Edinburgh: David Douglas. Anderson, J. 1904. Note on a late Celtic armlet of bronze

35.  Art in context: the massive metalworking tradition of north-east Scotland now presented to the National Museum; with notes on the identification of two other late Celtic armlets in the museum, and on a massive bronze armlet recently found in Sutherlandshire, Proceedings of the Society of Antiquaries Scotland 38 (1903–4), 460–6. Callander, J. G. 1916. Notice of a jet necklace found in a cist in a Bronze Age cemetery, discovered on Burgie Lodge Farm, Morayshire, with notes on Scottish prehistoric jet ornaments, Proceedings of the Society of Antiquaries Scotland 50 (1915– 16), 201–40. Clarke, D. V. 1971. Small finds in the Atlantic province: problems of approach, Scottish Archaeological Forum 3, 22–54. Close-Brooks, J. 1972. Two steatite lamps, Proceedings of the Society of Antiquaries Scotland 104 (1971–2), 295–7. Coles, J. M. 1968. The 1857 Law Farm hoard, Antiquaries Journal 48, 163–74. Cook, M. 2010. New light on oblong forts: excavations at Dunnideer, Aberdeenshire, Proceedings of the Society of Antiquaries Scotland 140, 79–91. Davies, M. H. 2007. Dominated by unenclosed settlement? The Later Iron Age in eastern Scotland north of the Forth. In C. C. Haselgrove and T. Moore (eds.), The later Iron Age in Britain and beyond, 266–85. Oxford: Oxbow. Davis, M. and Gwilt, A. 2008. Material, style and identity in first century AD metalwork, with particular reference to the Seven Sisters hoard. In D. Garrow, C. Gosden, and J. D. Hill (eds.) Rethinking Celtic art, 146–84. Oxford: Oxbow. Déchelette, J. 1914. Manuel d’Archéologie Préhistorique, Céltique et Gallo-Romaine II: Archéologie Celtique ou Protohistorique – troisième partie, Second Age du Fer ou Époque de La Tène. Paris: Auguste Picard. Feachem, R. W. 1966. The hill-forts of northern Britain. In A. L. F. Rivet, The Iron Age in Northern Britain, 59–87. Edinburgh: EUP. Fitzpatrick, A. P. 1989. The submission of the Orkney Islands to Claudius: new evidence?, Scottish Archaeological Review 6, 24–33. Fowler, E. 1963. Celtic metalwork of the fifth and sixth centuries A.D. – a reappraisal, Archaeological Journal 120, 98–160. Fox, C. 1958. Pattern and Purpose. Cardiff: National Museum of Wales. Fraser, J. E. 2009. From Caledonia to Pictland: Scotland to 795. Edinburgh: EUP. Frere, S. S. and Wilkes, J. J. 1989 Strageath: Excavations within the Roman fort 1973–86. London: Society for the Promotion of Roman Studies. Garrow, D. and Gosden, C. 2012. Technologies of Enchantment? Exploring Celtic art: 400 BC to AD 100. Oxford: OUP. Garrow, D., Gosden, C. and Hill, J. D. (eds.) 2008. Rethinking Celtic art. Oxford: Oxbow. Guido, M. 1978. The glass beads of the prehistoric and Roman periods in Britain and Ireland. London: Society of Antiquaries of London, Research Report 35. Harding, D. W. 2006. Redefining the northern British Iron Age, Oxford Journal of Archaeology 25(1), 61–82. Haselgrove, C. C. and Moore, T. (eds.) 2007. The later Iron Age in Britain and beyond. Oxford: Oxbow. Hill, J. D. 1995. Ritual and Rubbish in the Iron Age of Wessex. Oxford: British Archaeological Reports (British Series 242).

339

Hunter, F. 1997. Iron Age hoarding in Scotland and northern England. In A. Gwilt and C. Haselgrove (eds.) Reconstructing Iron Age Societies, 108–33. Oxford: Oxbow. Hunter, F. 1998. Copper alloy. In L. Main, Excavation of a timber round-house and broch at the Fairy Knowe, Buchlyvie, Stirlingshire, 1975–8, Proceedings of the Society of Antiquaries Scotland 128, 293–417 (338–46). Hunter, F. 2001. The carnyx in Iron Age Europe, Antiquaries Journal 81, 77–108. Hunter, F. 2006a. New light on massive armlets, Proceedings of the Society of Antiquaries Scotland 136, 135–60. Hunter, F. 2006b. Recent finds from Strageath Roman fort, Tayside and Fife Archaeological Journal 12, 81–8. Hunter, F. 2007a. Beyond the edge of the Empire: Caledonians, Picts and Romans. Rosemarkie: Groam House Museum. Hunter, F. 2007b. Artefacts, regions and identities in the northern British Iron Age. In C. C. Haselgrove and T. Moore, The later Iron Age in Britain and beyond, 286–96. Oxford: Oxbow. Hunter, F. 2008a. Jet and related materials in Viking Scotland, Medieval Archaeology 52, 103–18. Hunter, F. 2008b. Celtic art in Roman Britain. In D. Garrow, C. Gosden and J. D. Hill (eds) Rethinking Celtic art, 129–45. Oxford: Oxbow. Hunter, F. 2009. Miniature masterpieces: unusual Iron Age brooches from Scotland. In G. Cooney, K. Becker, J. Coles, M. Ryan and S. Sievers (eds.) Relics of Old Decency: archaeological studies in later prehistory. A Festschrift for Barry Raftery, 143–55. Dublin: Wordwell. Hunter, F. 2010a. Beyond the frontier: interpreting late Roman Iron Age indigenous and imported material culture. In R. Collins and L. Allason-Jones (eds.) Finds from the frontier: material culture in the 4th–5th centuries, 96–109. York: CBA. Hunter, F. 2010b. Buried treasure: a major Iron Age gold hoard from the Stirling area, The Forth Naturalist and Historian 33, 61–4. Hunter, F. 2013. The lives of Roman objects beyond the frontier. In P. S. Wells (ed.) Rome beyond its frontiers: imports, attitudes and practices, 15–28. Portsmouth, Rhode Island: Journal of Roman Archaeology. Hunter, F. forthcoming a. Crucibles and moulds. In Murray forthcoming. Hunter, F. forthcoming b. The copper-alloy finds. In Murray forthcoming. Hunter, F. forthcoming c. The glass artefacts and glass-working debris: catalogue, typology and discussion. In Murray forthcoming. Hunter, F. in prep. The carnyx in the European Iron Age: the Deskford carnyx in context. Jacobsthal, P. 1944. Early Celtic Art. Oxford: Clarendon Press. Keillar, I. 1993. Man in Moray: 5000 years of history. In W. D. H. Sellar (ed.) Moray: province and people, 24–46. Edinburgh: Scottish Society for Northern Studies. Laing, L. and Laing, J. 1986. Scottish and Irish metalwork and the ‘conspiratio barbarica’, Proceedings of the Society of Antiquaries Scotland 116, 211–21. Leeds, E. T. 1933. Celtic Ornament in the British Isles down to AD 700. Oxford: Clarendon Press. Lethbridge, T. C. 1953. Burial of an Iron Age warrior at Snailwell, Proceedings of the Cambridge Antiquarian Society 47, 25–37.

340

Fraser Hunter

Loudon, M. E. 2000. Stone artefacts of cup-like form in Scotland: a preliminary overview. Unpublished MA dissertation, University of Edinburgh. MacGregor, M. 1976. Early Celtic Art in North Britain. Leicester: Leicester University Press. Macinnes, L. 1984. Brochs and the Roman occupation of lowland Scotland, Proceedings of the Society of Antiquaries Scotland 114, 235–49. McLaren, D. and Hunter, F. 2008. New aspects of rotary querns in Scotland, Proceedings of the Society of Antiquaries Scotland 138, 105–28. Mann, J. C. 1974. The northern frontier after AD 369, Glasgow Archaeological Journal 3, 34–42. Maxwell, G. 1990. A Battle Lost: Romans and Caledonians at Mons Graupius. Edinburgh: EUP. Megaw, J. V. S. 1970. Art of the European Iron Age. Bath: Adams and Dart. Megaw, R. and Megaw, V. 2001. Celtic Art: From its beginnings to the Book of Kells (revised and expanded edition). London: Thames & Hudson. Murray, R. forthcoming. Excavations at Culduthel, Inverness. Piggott, S. 1959. The Carnyx in Early Iron Age Britain, Antiquaries Journal 39, 19–32. Piggott, S. 1966. A scheme for the Scottish Iron Age, in Rivet (ed.), 1–15. Polenz, H. 2007. Ein frühlatènezeitlicher Goldfingerring aus Trebur. In H. Kelzenberg, P. Kießling and S. Weber (eds.) Forschungen zur Vorgeschichte und Römerzeit im Rheinland: Hans-Eckart Joachim zum 70. Geburtstag gewidmet, Beihefte der Bonner Jahrbücher 57, 65–9. Mainz: von Zabern. Ralston, I. 1996. Recent work on the Iron Age settlement record in Scotland. In T. C. Champion and J. R. Collis (eds.) The Iron Age in Britain and Ireland: recent trends, 133–53. Sheffield: J. R. Collis Publications. Ralston, I. and Inglis, J. 1984. Foul hordes: the Picts in the north-east and their background. Aberdeen: Anthropological Museum, University of Aberdeen. RCAHMS 2007. In the Shadow of Bennachie: a field archaeology of Donside, Aberdeenshire. Edinburgh: RCAHMS. Rivet, A. L. F. (ed.) 1966. The Iron Age in Northern Britain. Edinburgh: EUP. Schönfelder, M. 2003. Ein goldener Fingerring der frühlatènezeit aus Veringenstadt, Kr. Sigmaringen, Archäologisches Korrespondenzblatt 33, 363–74.

Sharples, N. 2008. Comment I. Contextualising Iron Age art. In Garrow et al. (eds.), 203–13. Simpson, M. 1968. Massive armlets in the North British Iron Age. In J. M. Coles and D. D. A. Simpson (eds.) Studies in Ancient Europe, 233–54. Leicester: Leicester University Press. Simpson, M. 1970. Some Roman-Iron Age finger rings, Proceedings of the Society of Antiquaries Scotland 102 (1969–70), 105–8. Smith, J. A. 1868. Notice of a remarkable bronze ornament with horns, found in Galloway, now at Abbotsford. Also of a bronze ornament like a ‘swine’s head’, found in Banffshire, Proceedings of the Society of Antiquaries Scotland 7 (1867–8), 334–57. Smith, J. A. 1881. Notice of a massive bronze ‘late Celtic’ armlet and two small objects of bronze (horse-trappings), found with a Roman bronze patella, at Stanhope, Peeblesshire, in 1876; with an account of other bronze or brass armlets found in Scotland, Proceedings of the Society of Antiquaries Scotland 15 (1880–1), 316–61. Steer, K. A. 1956. An early Iron Age homestead at West Plean, Stirlingshire, Proceedings of the Society of Antiquaries Scotland 89 (1955–6), 227–51. Stevenson, R. B. K. 1955. Pins and the chronology of brochs, Proceedings of the Prehistoric Society 21, 282–94. Stevenson, R. B. K. 1966. Metal-work and some other objects in Scotland and their cultural affinities. In Rivet (ed.) 17–44. Warner, R. B. 1993. Irish prehistoric goldwork: a provisional analysis, Archeomaterials 7, 101–13. Warner, R. 2003. Old letters and new technology – the Ballyrashane gold hoard. In J. Fenwick (ed.) Lost and found: discovering Ireland’s past, 151–64. Dublin: Wordwell. Warner, R. 2004. Irish gold artefacts: observations from Hartmann’s analytical data. In H. Roche, E. Grogan, J. Bradley, J. Coles and B. Raftery (eds.) From megaliths to metal: essays in honour of George Eogan, 72–82. Oxford: Oxbow. Watson, W. J. 1926. The history of the Celtic place-names of Scotland. Edinburgh: Blackwood. Wilson, D. 1851. The Archaeology and Prehistoric Annals of Scotland. Edinburgh: Sutherland & Knox. Wilson, E. M. 1980. Excavations at West Mains of Ethie, Angus, Proceedings of the Society of Antiquaries Scotland 110 (1978–80), 114–21.

36 THE TORRS CHAMFREIN OR HEAD-PIECE: RESTORING ‘A VERY CURIOUS RELIC OF ANTIQUITY’ C. Stephen Briggs

Introduction Discovery and earlier history of study The Torrs Chamfrein, ‘pony-’ or ‘horned-cap’ is a two-piece sheet bronze plaque (or head-plate) shaped to sit comfortably on the front of a modest-sized horse’s face and forehead, or atop a small pony’s head and mane (Fig. 36.1). The plaque is almost covered in repoussée ornament. It is surmounted by two chased bronze horns attached transversely to the plaque by flanges and collars. All the decoration is executed in an early La Tène style currently defined as Torrs-WithamWandsworth which has links to the continental Plastic and Sword Styles (Jope 2000 passim). (The terms head-piece or composite are used here to describe the head-plate mounted with horns.) Today, probably the best known piece of artwork from Iron Age Scotland, it was given to Sir Walter Scott by his friend (or ‘antiquarian agent’) Joseph Train, in 1829 (Atkinson and Piggott 1955 [henceforth Torrs] 197–8; Train u.d.). Train’s biographer explains how it was sent to the baronet along with ‘a Roman javelin’ from Auchengibbert (Cheape et al 2003, 76, 79) and a transcript on old Manx traditions copied for him by Train (Patterson 1857, 134). Probably following Train, Patterson described the headpiece as ‘a visor of brass, found in a morass at Torrs’ .. ‘an antique mask’ ..with ..‘horns projecting from the place where eye-holes should have been, and which were turned back like the horns of a goat’ (id.). Apparently Train thought it most likely a medieval mummer’s mask (id; cf. Megaw 1983, 146–7). Until the present time, it was believed to have been found c.1820 and to have come from ‘the earth at the bottom of

a peat moss’ or ‘a morass’ (probably Torrs Loch) at Torrs Farm, parish of Kelton, near Castle Douglas, Dumfries and Galloway (OS NGR NX 780 618; Canmore ID 64630; originally Kirkcudbrightshire). In Scott’s day and later, it became a centrepiece of the Abbotsford cabinet, and since 1921, when bought for the nation from the novelist’s descendants, its iconic craftsmanship has been studied and universally admired at the National Museum of Antiquities of Scotland (now the National Museum of Scotland: NMS). The head-piece was first dated to the Iron Age by Dr John Alexander Smith in 1867, when, like Train (in Paterson 1857, 134) he also compared its function to that of a chamfrein

Fig. 36.1. Photo of Torrs head-piece from Jope (2000, Plate 100) (Reproduced by permission of Oxford University Press).

342

C. Stephen Briggs

– the kind of medieval equine face-mask worn in knightly combat. Although there is still uncertainty among historians of Celtic craftsmanship about its components’ origins, they could well be local to southwest Scotland. Given that dating by art style alone can offer relatively broad parameters of periodisation, its components may date from the late 3rd to early 2nd centuries BC (Jope 2000, 58–9, 122–6).

Interpreting the headpiece It was for long believed that the horns were mounted on the head-plate at the time of their discovery; that they were indubitably integral components of the original head-piece, and that they had survived to look largely as intended by the prehistoric craftsperson(s) who designed them. Then, in 1955, after their detailed examination of the composite artefact, Richard Atkinson and Stuart Piggott argued that this belief was untenable because they had found evidence to refute it. Their study drew four main conclusions. The first was that the ‘head-piece, horns and collars and flanges [are] the product of a single school of prehistoric metal-workers’. Second, they argued that although its components were all ‘found together … [they] ... were not then or at any time in antiquity structurally associated [my italics]. (In this, his theory of separation, Piggott was particularly keen to treat the horns apart from their base-plate because at that time he believed they had originally functioned independently as drinking-horn mounts). Third, it was said that ‘the present assembly of the parts dates only from after their discovery, though before 1829’, while ‘the attachment of the horns was partially reconstructed … at some subsequent date, probably during the latter half of the nineteenth century’. This implied that in its present form the headpiece with horns upon it amounted to an antiquarian hoax. Finally, it was proposed that the mask’s holes were more likely to have accommodated a pony’s ears than its eyes. Rather than a chamfrein, it was therefore a pony cap which (Piggott believed) was originally intended to have ‘supported a central plume or other ornamental feature’ (Torrs, 215). The object of this essay is to ascertain how far the conclusions promulgated by Atkinson and Piggott in 1955 still today carry credible conviction. To address that objective, Atkinson and Piggott’s arguments will be reexamined and the implications of a recently-discovered newspaper account detailing the Torrs discovery in 1812 will be discussed. But before proceeding, it seems important first to reflect upon the impact of the interpretations of 1955 on later appraisals of the find.

The headpiece in scholarship after 1955 Atkinson and Piggott’s conclusions were soon to influence the study of British Iron Age metalwork through their students and contemporaries. But whereas in having the

Torrs artefact repaired and restored its past investigators had left the ‘constituent parts mounted [enabling them] to be taken apart without difficulty’ (Torrs 198), dismantling it had been a remarkable event that few would ever dare emulate – not, at least, in the lifetimes of two such powerful professors. It is therefore no indictment of their intellects or the rigour of their methodologies to say that most later scholars of Torrs naturally relied upon the exemplary detailed observations made, the original deductions drawn, and the novel conclusions reached in 1955. Moreover, even the more experienced investigators were unlikely to re-examine such convincing arguments as had been presented. Unsurprisingly, therefore, in his study of Celtic art written soon after, Sir Cyril Fox (1958, 21) rightly noted only how the Torrs horns had been ‘incorrectly fixed’. And in his magnum opus of European Iron Age Art published in 1970 (as in the host of later authoritative articles and books he has written – often together with Ruth), Vincent Megaw – whose undergraduate career at Edinburgh witnessed the latter part of Piggott’s investigation – listed the horns independent of their bronze headpiece (Megaw 1970, 145–7). Ian Stead’s instructive British Museum handbooks on Iron Age Art similarly accepted the horns’ conceptual separation (Stead 1985, 62–5; 1996, 82–3) and Piggott, of course, continued to promote his own interpretations without caveat (e.g. Anon. 1970). Since 1955, these and many others have discussed and documented Early Celtic Art, those more specifically focusing on the Torrs art style most recently including Martyn Jope (2000), Dennis Harding (2002) and Ruth and Vincent Megaw (2001; 2005). In spite of the great depth of scholarship represented by these researches, the separation of the Torrs horns from their base plate (the separation theory) has never been seriously questioned. In fact it took almost 20 years before that uneasy topic attracted adverse criticism. Then, writing in 1974, Martin Henig (who is perhaps best known as a Romanist writing on coins, sculpture and gems) drew attention to an apparently unique British silver coin struck by Tasciovanus, king of the Catavellauni after c.20 BC (Henig 1974). Its obverse bears an image of Pegasus, a relatively common subject, one regularly copied in Britain from Roman prototypes. What makes this coin unique, however, is that its Pegasus bears a ‘sheath or cap that covers the animal’s head’ which has ‘two knobbed horns rising from the crown’ (ibid., 374). Henig added ‘this chamfrein extends as far as the animal’s ear and terminates sharply in a right-angle at the bottom’. Strangely, Henig’s informed observations appear to have been by-passed in virtually all subsequent writings about the Torrs head-piece, except Vincent Megaw’s, who in 1983 p. 133) accepted ‘that whatever their original use, the horns prior to the abandoning of the find were in fact one with the chamfrein in the Iron Age’.

36.  The Torrs Chamfrein or head-piece: restoring ‘A very curious relic of antiquity’ Regarding Piggott’s proposal of an earlier life for the horns as drinking horn mounts, noted briefly above, Atkinson and he informed Henig that by 1974 their ground had shifted and they then preferred ‘to think of the horns as yoke-terminals, especially as such a reconstruction would allow use to be made of certain minor components of the ‘original’ chamfrein, namely the collars and flanges’ (Henig ibid.; cf. Munby and Henig 1971). Additionally, Atkinson and Piggott implied that this newer notion was supported by a recent article of Piggott’s (1969), reference to which gave the impression that the Torrs horns had been temporarily confused with British Iron Age chariot cap fittings, an impression confirmed nine years later by Piggott and Atkinson in their personal communication to Vincent Megaw (1983, 130). Atkinson and Piggott would not have found Henig’s critical points easy to dismiss or overlook, for additionally he cited important published works describing European Later Bronze Age helmets. These invited useful comparison with the Torrs horns (Henig 1974; e.g. Hencken 1971, 169–174; Megaw 1970, 170, no. 294). Of particular note were: two horned helmets from a votive deposit at Vixø, Northern Zealand [which] make use of all the ‘Torrs’ components in a way remarkably like that envisaged when these were thought to compose a chamfrein. The horns are hollow castings riveted into cast bronze mountings which in turn are riveted to the helmets. This in itself is something of a coincidence if the ‘creator’ of the Torrs chamfrein (in modern times) was merely ‘a successful perpetrator of an archaeological hoax’. A second coincidence is that the chamfrein was always believed to have the horns curving backwards with the straight edge of the sheath lying towards the horse’s nose. The coin suggests that although the placing of the horns relative to the sheath would have been an inspired guess by the ‘hoaxer,’ the chamfrein was envisaged as being worn back to front.

Henig concluded in rational frame: ‘surely, we may be allowed to surmise that the Torrs metalwork was discovered already assembled as a chamfrein?’ (Henig 1974, 375). Few others were ever to confront any of these problems. In his essay for R. B. K. Stevenson’s Festschrift Martyn Jope did briefly challenge the practical utility of the ‘drinking-horn mount interpretation’ as advanced by Piggott (Jope 1983). Curiously, however, in that essay he neither mentioned Henig nor appeared aware of any revised view by Piggott himself, which seems odd, since by 1983 Jope was intending a comprehensive monograph on Early Celtic Art, having then already spent almost 30 years revising the British section of Jacobsthal’s eponymous masterwork (Jacobsthal 1944; Jope 1986, 17; Stead Preface to Jope 2000, v). Because the descriptive entries of his later metalwork corpus were written piecemeal during such a long gestation, when eventually the book did appear (posthumously), it was obvious that some artefact descriptions had been up-dated inconsistently. This explains how Tasciovanus’s

343

illuminating coin is illustrated (Jope 2000, vol. 2, plate 100–101d), and Henig’s 1974 paper is referenced in the accompanying commentary to that plate (Jope 2000, vol. 1, 251, and bibliography 366), whereas there is nowhere acknowledgment of its potential significance to the Torrs find either in Jope’s authoritative description of the horns or of the head plate (id. 35–7 and 72–4). Furthermore, the horned Pegasus is absent even from his detailed discussion of Celtic art on British coins (id. 164–74)! Jope’s scepticism of Piggott’s conjectural ‘drinkinghorn’ pedigree was, it seems, a relatively minor diversion and his dissent on that point did not affect his overall acceptance of the investigators’ conclusions. Consequently, his descriptions of the horns and head-piece as they were later to appear – separately in his corpus – shows that in the main he was entirely convinced by their arguments (Jope 2000 passim). Morna MacGregor also appeared concerned about Atkinson and Piggott’s theories. In the text accompanying her comprehensive metalwork catalogue, she admitted with a trace of irony that her separate listing of horns and headplate (MacGregor 1976, 2, no. 1 and 285–6, respectively) was ‘evidence of the persuasive powers of Atkinson and Piggott’ (1955; MacGregor 1976, 1, 176). Finally, in his day Keeper of Antiquities at the National Museum of Antiquities of Scotland, Dr R. B. K. Stevenson is believed to have privately rejected Atkinson and Piggott’s ‘separation theory’ and its implication of an antiquarian hoax. However, as currently exhibited at the museum (June 2013), the Torrs head-piece caption reads: ‘The horns were found with the cap in a bog. They didn’t originally have horns. When they were attached is uncertain’. It is thus clear that the conclusions reached in 1955 still carry authority and conviction.

Towards authentic restoration? The Caledonian Mercury account In recent years a handful of historic newspaper databases have become available for searching digitally. Since 2005 the present author has combed them seeking information on a variety of artefact and monument types and has enjoyed varying degrees of success in making useful discoveries. The following account of the Torrs find which appeared in the Caledonian Mercury on December 17th, 1812 is set in a particularly small typeface which did not lend itself to easy digital recognition. Its text runs: DUMFRIES, Dec. 8. – A very curious relic of antiquity was lately found covered about four feet deep in a moss in the neighbourhood of Castle Douglas, which, till about 20 years ago, formed part of the bottom of a lake. It is formed out of a thin plate of brass, and in shape resembles a mask for the

344

C. Stephen Briggs

Fig. 36.2. Train’s drawing of 1829, National Library of Scotland MS 912, fol. 80 (Reproduced by courtesy of the National Library of Scotland). human face, being about 10 inches long, and four inches deep. There are two crooked horns, which project from between the circular openings apparently Intended for the eyes, and which are each about 12 inches long. The external surface is fantastically ornamented, and has a strange and grotesque appearance. Some antiquarians who have seen it, suppose it to have formed the face of one of the rude divinities which our forefathers worshipped; while others, whose fancy is less enthusiastic, are disposed to think that it may have constituted a most humble ornament on the prow of a pleasure boat. In the same place, eight rings, about three inches in diameter, have been found, composed of similar materials, and joined in pairs by vincula, about two inches long, accompanied by five small cups or hollow hemispheres, also of brass, and an ink bottle of earthenware (Anon. 1812).

This account clearly removes any doubt that when first discovered, the Torrs headpiece was assembled in at least something approaching its present composite form. Before discussing how that news impacts upon the conclusions drawn in 1955, it is first necessary to discuss the deposit’s accompanying artefacts. The most obvious question here is: ‘how far can they be identified or may parallels be suggested explaining their functions?’ The four sets of bronze ring three inches in diameter [c.7.6 cm] paired together by vincula 2 inches [c.5 cm] wide make a useful starting point. Both their sizes and the articulation described suggest they could have served as harness bits. If these rings actually make up four horse bits (cf. Haworth 1971; MacGregor 1976, vol. 1, 24–30, vol. 2, 2–11; Raftery 1983, vol. 1, 7–82, vol. 2, figs 1–46; 1984, 15–44), their discovery here might suggest the presence of two pairs of draught animals inferentially pulling a

wheeled vehicle. Curiously, however, when compared to conventional Iron Age harness, as bits they seem to have been remarkably small for horses, even for Shetland ponies, so perhaps they may have taken goat or even deer, particularly as, with its headpiece, the assemblage seems to have ceremonial leanings. The earliest known drawing of the chamfrein of 1829 by Joseph Train, includes a bronze ring (Torrs pl. LXXI) (Fig. 36.2 and Appendix 36.1). In a thorough discussion of this ring’s various potential associations (Cheape et al 2003, 64–5 & 80), it has been compared to one now at the NMS (DO49) from Dungyle (or Dunguile), a fort also in the parish of Kelton (OS NX 773 571; Canmore ID 64470l; Torrs id.). As this was also at one time apparently in Train’s collection, Atkinson and Piggott wondered if it, too, could have come from the Torrs deposit (see Torrs caption beneath pl. LXXI opp. p.198). But whereas that drawing suggests the slightest hint of differential wear – which might be expected in a harness-ring (cf. MacGregor 1976, vol 2, nos 13–14) – scaling it up against the headpiece computes a diameter no greater than c.6.5 cm (if his depiction is accurate), and anyway the museum’s catalogue gives a diameter of 2.95 inches (7.5 cm, Fraser Hunter pers comm.). The ‘Dungyle’ ring is very close in size to those found at Torrs, and if not one of them, it could have served the same purpose. The second artefact group here was made up of ‘five small cups or hollow hemispheres’. Absence of greater detail inhibits better understanding of their shapes or sizes. Fraser Hunter (pers. comm.) suggests they may have been phalerae. Alternatively, they were possibly ‘horn caps’ or ‘trumpet ends’, an artefact group constituting ‘a distinctively British type, not found on the Continent’ (Stead 1996, 25–6). These latter were first discussed in detail by Sir Cyril Fox in the context of the Llyn Cerrig Bach hoard and chariot in 1946 (Fox 1946, 14–17, fig. 12). These ‘horn caps’ were known to be distributed over England and Wales, but are not recorded from Ireland or Scotland hitherto (if indeed that is what they were, Macdonald 2007, 135–46; Spratling 1972, 430–436, nos 119–135). Although ‘none has been found in a context exclusively linked with harness or vehicle fittings’ (Stead 1996, 83), but because some, at least, have been recorded associated with harness bits, it is tempting to think of the Torrs cups as having belonged to a cultural milieu which included ceremonial or even utilitarian chariots or wagons (cf. Piggott 1983). With its limited tantalising detail, this neglected account now obviously assumes a significance beyond that of the deposit it describes: it entirely negates Atkinson and Piggott’s 1955 ascendant view that the horns were found as separate items from the repoussée bronze decorated headplate they are still mounted upon. In retrospect, it must seem strange that Atkinson and Piggott were never deterred from their unknowable claim that the Torrs’ components were all ‘found together …

36.  The Torrs Chamfrein or head-piece: restoring ‘A very curious relic of antiquity’ [though] ... were not then or at any time in antiquity structurally associated’ (Torrs, 215). But even demonstrating the composite nature of the head-piece when discovered need not absolutely negate the possibility that ‘the design of the head-piece pre-suppose[s] the existence of some form of central feature in the nature of a light panache or crest, the seating of which was destroyed by the later botched attachment of the pair of horns’ (id.). As in this re-examination concerns must lean towards the probable, rather than the possible, the origin and substance of that statement need careful examination.

Stuart Piggott’s theories: Chamfrein or Pony Cap? Stuart Piggott became curious about the Torrs headpiece soon after succeeding Vere Gordon Childe as Abercromby Professor of Prehistoric Archaeology at Edinburgh University in 1946. His burgeoning interest is evident from this response to Paul Jacobsthal’s invitation to Oxford in March 1947 (Jac. Archive, Box 23, Letter 2): If I don’t see you, I should like you to consider whether the Torrs champfrein [sic] isn’t of two periods (not two art styles though). I feel sure that the horns were not in the first place intended for the champfrein – they are too heavy, and their attachment is very odd and makeshift. They have also broken the main body of the thing twice – the bits of decorated bronze are to cover repairs caused by the weight of the horns – there is also evidence that they have been re-set at least once and they are awkwardly squeezed in between the relief pattern. I think the same craftsman made the horns and the other part – or at least they are both products of the same school (Parisian). The repairs look rather less good and might be later. (? made in Scotland). What were the horns originally ? I should like to think they were a pair of drinking-horn mounts, or (less probably) chariot-horns. Can you bring yourself to agree ? And another thing – all the illustrations of the Torrs object seem to me upside-down – i.e. the thing is conceived as back-to-front. Surely the holes are for ears, not eyes, and the narrow part goes over the neck and not the nose. I can’t make it fit the usual way round – there would be no room for the ears, and the eye part would be too narrow even for a small pony. It seems to me a sort of cap, with originally not horns, but some central mount (? a panache, or plume or what-not), and with the horns later added to make it into a sort of Pazyrik head-piece (Talbot-Rice and I had independently thought of this as a parallel) (Piggott 1947).

This demonstrates how Piggott had begun interrogating the headpiece in 1947. Possibly the first time he had written seriously proposing it as a pony-cap, this narrative nevertheless betrays that he was having difficulty coming to terms with one important outcome of his new theory. As long as the piece conceptually remained a chamfrein, its horns pointed backwards like a modern British billy goat’s (Harris and Yaldon 2008, 556, pl. 10C; 628–633; cf. Torrs, 227, where the horns are favourably compared to those of a

345

bovid). Turned around as a pony-cap, its horns would need to face the opposite way with their tips falling forward to look more like a nanny goat or even a fawn (Harris and Yaldon op cit.). As already acknowledged, Henig had grasped that point in 1974. Oddly, Piggott seems to have championed only backwardlooking head-gear, perhaps having felt that to offer the best aesthetic dynamic. Such preferences notwithstanding, he drew a sketch for Jacobsthal showing forward-pointing horns to show how the pony’s ears might have projected through the ‘eye holes’ in his then nascent pony-cap (Fig. 36.3a). Later, he seems to have avoided repeating similar imagery altogether, so that when comparing the relative looks of a chamfrein as against those of a pony-cap, the former (which he placed on a pony’s smaller head) continued to carry backwards-facing horns, while by contrast, the latter (a ‘pony-cap’ placed on the rather larger head of a horse sporting much larger ears) displayed a backwards-facing plume (Torrs, pl. LXXIX) (Fig. 36.3b)!

Panaches, feathers and drinking horns At the time he wrote to Jacobsthal, Piggott was intensively researching Early Iron Age swords. His was an interest offering critical insights into insular Celtic metalworking traditions (Piggott 1950) which also seems to have lent him confident authority in related matters. Possibly then still somewhat uncertain, in 1951 he cautiously described the chamfrein as a ‘Parade-Mask for a Horse’ (Piggott and Daniel 1951, 18). But increasingly convinced about his theory of separation, he went on to declare in print how it was ‘probable that the heavy horns, though from the same atelier as the mask, [were] a later addition replacing a lighter feature such as a panache of feathers, and they may themselves have originally served as drinking-horn mounts’ (Piggott and Daniel 1951, 18 and pl. 38–40). It should here be recalled that whilst Piggott’s speculations were gradually gathering momentum, at this point the pony-cap idea with his separation theory and its attendant deductions all remained unsupported by any evidence-based argument. But boldly to question the horns’ original use and propose them as replacements for a single plume surely needed proof! Could dismantling such an iconic relic be justified in its pursuit?

Dismantling the Chamfrein?1950–1955 Whereas the 1812 account ought to remove any doubt that the Torrs head-piece was a horned composite when found, so strong was the case made against its integrity in 1955, that concerns may still linger about the intentions of its original maker(s). This is because, following their thorough investigation, Atkinson and Piggott made clear that their observations quite independently provided evidence of the

346

C. Stephen Briggs Fig. 36.3a 1947 sketch by Stuart Piggott of the Torrs head-piece showing horns pointing forward (reproduced by permission of Oxford Institute of Archaeology Jac. Archive Box 23, letter 2).

Fig. 36.3b Sketch showing chamfrein compared to plume (Torrs, pl. LXXIX; reproduced by permission of the Society of Antiquaries of London).

separation theory, and therefore also justified the existence of an earlier plumed head-piece. Two vital questions therefore now remain. The first: why were exhaustive reparations or restorations necessary on the head-piece post-discovery? And secondly: was Piggott’s belief in plumed phase ever justifiable from the constructional evidence observed and interpreted in 1955? Rather than repeat the entire descriptive apparatus of 1955, the briefest of summaries taken mainly from that account is offered here. It should be noted that whereas the original observations were in many ways exemplary, Atkinson and Piggott’s text is occasionally punctuated by reminders of Piggott’s theories. In re-examining the

original text it is therefore important to distinguish between indisputable evidence-based argument and theoretical claims or assertions.

The head-plate The two-piece repoussée-decorated base-plate (now lacking a non-essential, decorated segment, probably broken off at the time of discovery) measures 28 cm long by 10–15.5 cm wide and from 8.7 to 11 cm deep (Fig. 36.4). Its component halves are attached transversely, with the overlapping joint crossing all four of the inserted apertures (horn mount holes and eye- or ear-holes). There are several notable repairs.

36.  The Torrs Chamfrein or head-piece: restoring ‘A very curious relic of antiquity’

347

Fig. 36.4. Developed scheme of repoussée ornament on the Torrs head-piece (Torrs, fig.1, 201).

Three are certainly ancient and cover cracked (probably fatigued) metal with bronze strips decorated in a style difficult to date precisely. Atkinson and Piggott believed these repairs resulted from ‘the botched job’ of mounting the horns secondarily to a (conjecturally) plumed, primary phase (Torrs, 200). But a more likely origin could have been the severe stress caused when, during its original construction, the sheet metal was bent over into a channelled U-shape, little more than c.5 cm (or 2 ins) wide to accommodate an equine nose or mane. (That narrowness is best appreciated in MacGregor’s unique section 1976, vol. 2, no.1).

as the base-plate, they are also fashioned in forged sheet bronze, though the pattern is chased and the terminals are of cast-on metal. Looked at as a chamfrein, the dexter horn (known in Torrs as B) terminates in a cast-on duck’s head, the eyes of which were until 1955 filled with tinsmith’s solder (Torrs, 202). Without supporting evidence, there is no reason to favour an original infill of coral over coloured enamel (contra Torrs, 207). Horn A now lacks its matching terminal. For discussion of the flanged collars which attached them to the head-plate see below.

The Horns

Summary of Repairs and Restorations (Appendix 2)

Slightly twisted, the surmounted horns (their precise lengths, not given in Torrs are: horn A 210 mm long, horn B 215 mm [Fraser Hunter pers.comm]), with diameters basally of 3.3 cm and 0.7 cm at the tips. Decorated in the same style

Atkinson and Piggott importantly established how the entire artefact – most particularly its horn-mounting apparatus – is now riddled with evidence of attempted repairs or

348

C. Stephen Briggs

restorations. This revelation led them to conclude it was ‘unquestionable that in its present form the attachment of the horns is modern’ (Torrs, 210), a conclusion that on the face of it appears to be unassailable. Precise interpretation of that deduction does, however, rest on what is meant by ‘its present form’. Indeed, given so many substantial interventions to the actual horn-holding apparatus, perhaps a re-examination might bring to light some clue that was missed or misunderstood in 1955? The 1955 investigation showed which ancient repairs could be distinguished from the recent (Appendix 2 and 356 above). The oldest patching – probably to the head-plate – has already been noted briefly (id. above). To this should be added a piece of skilfully re-decorated bronze sheet grafted on to horn A (Torrs, 206–7). Virtually all other repairs were convincingly ascribed to the post-discovery period (Torrs, 206–9). It is thus difficult to disagree with the first part of the first conclusion reached by the investigators, which runs: Though there can thus be no question that in its present form the attachment of the horns to the head-piece is wholly modern, it may be objected that this does not preclude the possibility that the horns were so mounted in antiquity, the present attachments being merely an unskilful copy, made soon after the discovery, of the work of the original prehistoric craftsman.

The argument continues, going on to claim that: [how] this hypothesis is wholly untenable is shown by the nature and position of the ‘rivets’ which held the collars, horns, and copper tubes together (fig. 2, m). On first examination these ‘rivets’ resembled other flush-headed rivets on the horns and head-piece, but scratching the surface showed that they were of brass, not bronze. By permission, several of these rivets were extracted, and found to be brass screws of non-standard diameter and pitch. They had been inserted into holes drilled and tapped for the purpose, the heads afterwards being cut off close to the collar and the projecting part carefully hammered down to counterfeit the appearance of a rivet-head. (Torrs, 210)

The conclusion reads: It thus follows that the collars (and therefore the flanges) cannot have been fastened to the horns in antiquity, since the only means of attaching them have been shown to be modern; and without their aid it is impossible to see how the horns can have been fastened to the head-piece at all, since there is no other visible means of attachment. (Torrs, 211)

Though this underlines that most surviving repairs are of recently-manufactured materials, the certainty of the investigators’ argument here employs a different logic from that applied elsewhere to justify the existence of the hypothetical plume mount (Torrs, 209, 215). So whereas it was argued that an absent plume could have been attached to the head-plate through smaller base-plate holes than those later accommodating the horns, it was quite out of the question to consider the possibility of enlarging and tapping the collar’s existing rivet holes for stabilising screws

or stitches. (Stitches are here suggested because of the possibility the collars were intended to be fastened without rivets (below 359–60)). Whether or not a botched cosmetic job or skilfullyexecuted repair, it is perhaps worth pausing at this point to consider who had actually undertaken the complex metalworking interventions that were to be so carefully chronicled in 1955. Here, the best clues once more come from digitised newspapers, notably The Berwickshire News and General Advertiser of 14th January 1902, supplemented by Mr Aaron Forrest’s obituary of 10th November 1903 [added in brackets here]. The former account carries a heading in undistinguished font, which runs: A Link with Sir Walter Scott. – At Jedburgh on Wednesday, Miss Forrest died at the residence of her brother, Mr Aaron Forrest, Abbey-place, Jedburgh, at the age of 84 years. She was a woman of vigorous intellect and retentive memory, and she possessed a great store of local history and knowledge of archaeological subjects. One of her cherished memories was the fact that Sir Walter Scott said a few kindly words to her when she was at play as a girl in Friars, Jedburgh. Her father was a gunsmith in High-street, and Sir Walter often visited his shop, both on business and friendly pursuits. Many of the antiquities that Sir Walter collected, and that are now preserved at Abbotsford, were repaired in Mr Forrest’s shop [on Sir Walter’s order].

Both accounts go on further to cement a connection between the ballad singing grandmother of the deceased Forrest siblings and Scott’s early writings. Although a jeweller or tinsmith may have possessed some of the requisite skills, a gunsmith is likely to have been better acquainted with tapping techniques for producing small [grub] screws like those recognised by Atkinson and Piggott. Also, gunsmiths may be able to draw upon wider ranges of metal-working equipment than jewellers, and in Forrest’s case – if he it was – he would have needed tools equal to stabilising the chamfrein. It would be interesting to know whether or not other Abbotsford pieces still carried any comparable hallmarks of Forrest’s work, or indeed if any archival evidence survives to cement the connection between his craftsmanship and Scott’s antiquarian collection.

Reconstructing the original head-piece As it has been shown that a great deal of effort was spent on stabilising the head-piece throughout its recent history, it has now to be considered what might have been lost from its original make-up that it was left needing so much remedial attention.

The broken flanged collars As already noted, although their bases are held into the head-plate by thin bronze strips the horns are currently

36.  The Torrs Chamfrein or head-piece: restoring ‘A very curious relic of antiquity’ stabilised by skilfully-devised beaten-bronze flanges and collars. These clamp the horns firmly onto the sheet bronze, their internal diameters (3.55 cm – horn A; 3.45 cm – horn B). The collars still take the horns’ open terminals and affix them through conjoined circular holes cut into the sheet base equidistant between the circular ornamental ‘eye’ (or ear) holes sited on either side. Whereas these ‘eye-holes’ were integral to the original repoussée patterning, the ‘horn holes’ do not obviously relate to its designed surface. There is some ambivalence about the way these rings were discussed in 1955. Although they are described as flanged collars, strictly-speaking, that term can no longer apply, for at an uncertain stage in their history each broke in half. This fracturing probably now holds an important clue to solving the dilemma between the greater certainty of the newsprint on the one hand, and Atkinson and Piggott’s conviction for outright rejection of the antiquity of the horned head-piece concept on the other. Fracturing occurs along the horizontal angle between the upper collars and their flanged basal attachments. That could well have resulted from the horns’ weight stressing the sheet base plate. Less likely, though still a possibility, breakage could have been due to over-diligent polishing or even clumsy handling, post-discovery. The outcome was anyway to create separate, independent upper collars and lower flanges. Whereas the fact of this breakage was noted (Torrs, 198, fn 7), given their central role in the complex history of the headpiece, it is puzzling that no attempt was made to explain the cause of the break (Torrs, passim). Until they fractured, however, the complete flanged collars probably held the horns firmly in place with minimal need of auxiliary riveting. Breaking them apart would have caused the horns’ support to fail entirely through removal of any stability previously intended by firmly-bedded base flanges. Consequently, once broken, a new solution would have been needed to maintain the horns’ firm positioning. In the course of the present enquiry, all known graphic and written descriptive material about the collars has been vainly scrutinised to try and establish when they broke. One of Joseph Train’s significant insights was to recognise how the horns were ‘fixed in [i.e. mounted] by two small rings’. Unfortunately, this does not exactly clarify whether or not he meant a total of two rings, or that the collars had already sheered apart by 1829. His drawing (Fig. 36.2) stylises only one collar and shows its upper section inclining very slightly towards the flanged lower. Unless this inclination resulted from perspective drawing, that detail could have been what led Piggott in 1947 to mention backward pressure from the horns upon the plate when he also pertinently deduced that its creator probably underestimated the strength of mechanism needed to carry the horns’ weight on their relatively thin bronze sheet base. While still entire, the collars probably needed auxiliary

349

support, or at least a mechanism to prevent them from moving axially. In their useful sketch of the horns, Atkinson and Piggott show rectangular holes 0.4 cm by 0.8 cm (horn A) and 0.4 cm by 0.6 cm (horn B), pierced at right angles to the vertical decorated surface of each horn with their longer axes parallel to the horn shafts (Fig. 36.5a; from Torrs, fig. 2, 205). As it is unclear how far the horns projected beneath the head-plate before restorations began, at this point it seems worth suggesting that if pushed only slightly beneath the sheet base towards the horse’s mane, transverse pegs inserted through these rectangular holes above could have prevented the horns’ upward movement. If such a mechanism ever existed, the complete flanged collars would have played a complementary role as tight-fitting horn supports maintaining rigidity perhaps even without rivets. . (Rivets were the stabilising solution originally proposed by Munby (Munby and Henig 1971)). Whether fashioned from iron or wood, any holding pegs or pins involved in such a mechanism could have easily disappeared in the bog, thus loosening the horns. Certain other inexplicable features were noted during the 1955 examination. For example, the investigators were puzzled that the collar bases appeared never to have located precisely onto the metal sheet base (Torrs, 211). Ever since 1829 the horn bases have been set so close together that one flange always slightly overrode the other (Train 1829) (Fig. 36.2). This basal crowding, better seen on Smith’s later illustration (1867, pl. XLIV) (Fig. 36.6) and currently best appreciated on Jope’s photograph (2000, vol. 2, 101e) (Fig. 36.7), results in a slight compensatory lift and even throws part of one flange out of contact with the base plate (Torrs, idem). It was cogently argued that skilled Iron Age craftsmen were unlikely to have countenanced such imprecision (Torrs, 212). That alleged imprecision was to become a key factor in arguing an antiquarian, rather than an Iron Age origin for the piece as examined. An attempt will now be made to explain this apparent deficiency in craftsmanship.

Leather: the loss of an essential production medium? (Fig. 36.5b) Up to this point, in common with previous enquiries, the headpiece has been examined in the exclusive materials context of metal-working. There is no doubt, however, that Iron Age smiths owed much to traditions of woodand leather- working. Embossing and chasing must have developed from designing and decorating organic media. Furthermore, practices of combining organic with inorganic materials in artefact construction demonstrably go back to the Later Bronze Age, for example in bronze buckets and cauldrons (Briggs 1987; Gerloff 2010), if not much earlier. Here, leather would have seemed an appropriate medium for cutting washers to cushion the horns and their flanged collars from the sheet metal base. Indeed, when originally

350

C. Stephen Briggs

Fig. 36.5. Key to rivets and rivet-holes in the horns. Adapted from Torrs (Atkinson and Piggott 1955), 205, fig. 2, 205.

assembled, the flanged collars could have ideally bedded onto a single piece of hide intended to cover the entire undecorated central area of the bronze sheet ‘cap’. In fact decorated leather could have been used to complement the plate’s repoussée ornament. Leather offers a practical solution to stabilising the horns by its cushioning and deflecting some of their downwardpressure. Its presence would have minimised damage to the sheet metal base by raising the horns’ collars upon washers of hide. Such a mechanism probably also kept the horns separated. The loss of these washers could explain how ‘the traces of two circular indentations, a little less than 5 cm in diameter [were observed] on the upper surface of the head-piece’. These were considered to have resulted from the recent fastening of horns and flanges to the headpiece (Torrs, 202 and 209). But an alternative explanation could be that they represent the sites of the two hard-pressed flanged collar bases, with the indentation resulting from progressive shrinkage, tightening, dissolution and loss of their protective leather washers. Mindful that it may have been necessary to use more than one medium to stabilise the horns, it becomes easier to appreciate their likely instability at the time of discovery. All the organic components would have deteriorated in the bog, and any iron clamping pins would have corroded or even dissolved. Coming from the bog with loosened horns, its new owners would have been obliged to set about restoring them as best they may. And as, after being cleaned up, the metal may well have lost any evidence of its complementary organic media, Atkinson and Piggott were probably correct in attributing its (relatively) clumsy workmanship to the shortcomings of 19th-century restorers, rather than to unskilled Iron Age craftsmen.

Summary and Conclusions Towards a history of the head-piece A recently-discovered newspaper account is presented which explains how, in 1812, an assemblage of Iron Age metalwork was found in a bog forming the basin of a former lake at Torrs Farm, Castle Douglas, Galloway, Scotland. The assemblage included ‘a thin plate of brass … [with] … two crooked horns … project[ing]’, together with four pairs of horse-bits and five brass ‘cups, or hollow hemispheres’ – interpretable as decorative ‘chariot horns’. Though these horse-bits and ‘cups’ are believed lost, the brass plate and horns were given to Sir Walter Scott in 1829. Displayed at Abbotsford (Scott’s home) throughout the 19th century, the artefact later became known as The Torrs Chamfrein (or head-piece), and having been bought for the nation in 1921, it is now exhibited at the National Museum of Scotland. After an investigation undertaken sometime between 1947 and 1955 involving its complete dismantling, Professor Stuart Piggott and (the then) Mr R. J. C. Atkinson, strenuously argued that the horns and base-plate could never have been part of the same head-piece in antiquity: the horned chamfrein, head-piece or pony-cap, was therefore a counterfeit artefact which could only have been brought together by a hoaxer after its recovery from the bog. It was further asserted that the head-piece was originally designed to carry a plume or panache, but certainly not horns. Though there were no parallels for such a plume, the horns were discounted as part of a head-piece and re-ascribed the function of wine-horns. These conclusions were absorbed by and largely accepted into the scholarship of most later writers, whilst dissent was either ignored or suppressed. The arguments underlying the conclusions of 1955 have

36.  The Torrs Chamfrein or head-piece: restoring ‘A very curious relic of antiquity’

351

Fig. 36.6. The relationship of the horns to Repairs 1 and 2 (from J. A. Smith 1867–8, pl. XLV. opp. p. 238).

been re-examined and it is demonstrated that aspects of the original investigational methodology were based to varying degrees upon preconception and inconsistency of argument. The finding here is that such certainty of assertion by the investigators was unwarranted. Such a finding does accommodate the original investigators’ view (based upon thorough inspection and meticulous description) that in its present state, the headpiece incorporates many repairs and restorations effected post-discovery which could be misinterpreted as ancient. Critical retrospection does, however suggest that the recent interventions recognised in 1955 may mask the clues testifying to the combined Iron Age craft skills of both metal-and leather-workers which may have obviated the use of screws, or even of rivets.

The headpiece was almost certainly conceived and constructed as a composite. And just as was believed in the 19th century, it was made up by mounting two horns upon their repoussée bronze sheet head-plate base. There is no evidence for any earlier separate function or use for the horns. Neither is there reason to suppose that the horn mountings were preceded in the same place by an earlier fixture carrying a plume or panache. For that, again there is neither real nor circumstantial evidence. Upon the basis of the 1812 news report and the detailed retrospection of the present study, it must be concluded that the horned headpiece cannot be ‘a modern fabrication from separate parts, found together and stylistically related, but never structurally united in antiquity’. Furthermore, it seems probable that an organic medium lost during immersion in

352

C. Stephen Briggs

the bog – most likely leather – was an important component of the head-piece, one originally fundamental to the horns’ stability. The Torrs metalwork assemblage could be one of the first manifestations of a fugitive Scottish Iron Age chariot or wagon culture, a culture until recently suspected only from single finds of horse-bit and terret-ring, though complemented by a 5th century BC chariot burial from Newbridge, near Edinburgh (Carter et al. 2010; Carter and Hunter 2003). Not known to have been accessioned or obviously displayed in any museum, the missing Torrs rings, cups and ‘ink bottle’ have, for the moment at least, disappeared without trace. And while nowadays prehistoric finds from watery places almost invariably attract interpretations of votive offering or ritual burial (Hunter 1997), the present writer’s preference for the potentials of accidental loss are elsewhere discussed in relation to bogbodies (Briggs 1995) and more pertinently to beaten bronze metalwork (Briggs 1987). Though certainty about the original connection between horns and base-plate has now been re-established, of necessity, this investigation has considered only limited aspects of the construction and discovery of the Torrs find. Like those given in 1955, the interpretations offered here will remain speculative until (if ever) more is known of the newly-described enlarged Torrs deposit, or until fresh finds produce comparable artefact groupings elsewhere. All the deductions drawn and conclusion reached here about the head-piece and its context will usefully repay further inquiry through model-building and experiment. These should enable clearer insights into the technological limitations and supposed functions of all its components. The artefact’s history aside, over the past 60 years the form and ornamentation of the Torrs headpiece has gained in importance as a document focal to interpreting regional La Tène workshops in Britain and Ireland. Through his own thorough descriptions of the material evidence, Vincent Megaw has not only played a vital role in stimulating discussion and debate upon the interpretation of that decoration; he can claim responsibility for promoting a fuller appreciation of Iron Age art internationally through its thorough aesthetic and technical interrogation.

Acknowledgments I am indebted most particularly to Sally Crawford, Archivist to the Institute of Archaeology at Oxford, for help well beyond the call of duty, for without her discovery of Stuart Piggott’s letter to Paul Jacobsthal an important part of the argument here would be wanting critical evidence; to Mansel Spratling, whose interests in Celtic Art have impressed me since our first meeting in the late 1960s and who kindly

Fig. 36.7. Close-up sketch of horns showing the condition of the broken flanged collars in 1867 (taken from the engraving in J. A. Smith’s 1867–8, pl. XLIV, opp. p. 236).

made available an extract from his doctoral thesis (Spratling 1972) updated 1983–4; to Richard Haworth, who through his consistent friendship has infused parts of Martyn Jope’s teaching that I failed to absorb as an undergraduate; also to Rev Martin Henig and his co-author Julian Munby in their unpublished 1971 paper, for sharing their scholarship with a complete stranger and testifying to Dr R. B. K. Stevenson’s doubts about the conclusions reached in 1955. Its conclusions share several fundamentals reached in the present text, but as it was received as that was on its way to the press, it has not been possible to absorb fully their detailed arguments which clearly included significant doubts about the 1955 conclusions. In addition, Trevor Cowie, Fraser Hunter and Ian Stead have saved me from elementary pitfalls by offering much advice and encouragement, while for advice on animals I thank Gareth Jones and Arthur MacGregor. Miss Mariamne Briggs kindly transcribed the Torrs’ exhibit caption at the National Museum of Scotland. The author alone is responsible for the views expressed here and for any errors of fact. For the graphics, grateful acknowledgements are due to Oxford University Press for Fig. 36.1; to Alison Metcalfe at The National Library of Scotland for the reproduction of Fig. 36.2; to The Society of Antiquaries of London for Figs 36.3b, 36.4 and 36.5a, and to the Oxford Institute of Archaeology Archive for Fig. 36.3a.

36.  The Torrs Chamfrein or head-piece: restoring ‘A very curious relic of antiquity’

Envoi Having been acquainted with Vincent for almost 40 years, it has given me considerable pleasure to offer him this essay. I am particularly grateful of his bonhomie and for generously sharing his circle of friends at recent meetings of the European Association of Archaeologists. Vincent will no doubt appreciate how this piece comes from a complete novice in Iron Age studies who is deeply sensible of its deficiencies.

Appendix 1: Joseph Train’s 1829 drawing and MS text Joseph Train was the exciseman and antiquarian collector who gave Sir Walter Scott the first known manuscript documentation connected with Torrs (Cheape et al. 2003). This focuses on a competent drawing of the chamfrein, showing it complete with horns and apparently mounted firmly in 1829 (Train 1829a; see Torrs, pl. LXXI opp. p. 198). Below the drawing the key runs: AAA The [? Shell] or mask made of metal resembling Jeweller Gold. B a [---illeg.] made seemingly to [----illeg.] from the edges. E the dark shade represents a piece broken off since found. The shaded ornaments are [----illeg.] [?beaten] by a hammer and Punch from the inside leaving a kind of oval on the outsides. D This small cast ornament is merely a patch covering a crack there are several other patches put on in like manner to cover cracks in the opposite side but each patch is of a different pattern. F a hole on each side evidently for the eyes. The horn at the root are about an inch and a half apart. They are made of the same kind of metal as the rest of the mask. They are fixed in by two small rings as represented at G. The whole mask is pinned together, there is no solder about it except a little white metal resembling Block Tin to fill up the eye at the point of the longest horn. The ornament on the horn appears to be punched in not engraven.

One problem of this drawing is that Train’s key does not mention the large ring shown at bottom right (see above p. 354).

Appendix 2: Evidence for Repairs, a summary 2.i Evidence for repairs in antiquity There are three repairs to the head-plate made of strip bronze and covered in simple decorative motifs. They were noted by Train (Fig. 36.2); are later itemised by Smith in 1867, and are then described in detail in 1955 (Torrs, 205–6). They appear to have been necessitated by metal fatigue (see above 356–7).

353

At some time during its early use, Horn A appears to have lost a substantial piece of chased sheet, as it narrows towards the point (Torrs, 206–7; fig. LXXVIIC). This casting-on was apparently effected when it was still possible to replicate the chased decoration using the original tools and know-how of the original art-piece.

2.ii Evidence for Modern (i.e. post-discovery) Repairs a Drill holes and screws Between the indentations and the edges of the holes which they surround, are a number of small holes. Of varying sizes and irregularly spaced, they are presumably rivet-holes (Torrs, pl. LXXIId). A number of them have clearly been drilled or punched from the inside. ’It is probable that most of them are of modern origin’(Torrs, 202). b Modifications to Flanges and Collars Although the authors expressed surprise that the flanged collars were of hammered sheet bronze rather than of cast, the structure of the former is generally tougher and less prone to breakage. Though it is accepted that the flanges and collars were originally of a piece, in Torrs they are treated separately, it being noticed that they came to be separated by a ‘complete fracture … along the weakest and thinnest line, where the flange splayed outwards from the base of the collar’. One collar bears a recent repair in sheet copper. Respecting the details, each collar is pierced by three radial ‘rivet holes’ set approximately equidistant round its periphery (Torrs, 204), These ‘are undoubtedly modern’ (ibid. 211). There is a notch on the outside edge of one flange but not on the other (ibid. 205) which seemed ‘meaningless in relation to the present use and position of the flanges’ (ibid. 211). The flanges are pierced by three older, countersunk holes together with three obviously executed with a twist-drill. The status of the former is unclear, whereas the latter are obviously modern. Several puzzling holes and rivets appear above the collars. However, the investigators all but dismissed the possibility that the bar-rivets currently holding together Horn A (one each of bronze, brass and copper) might have replaced earlier, thinner ones (Torrs, 208). They also held that ‘though it is conceivable that these bars replace ancient bar-rivets of smaller size in the same position, it is in the highest degree unlikely’ (id.). At this point the question arises: why were such holes made if they had no purpose? The corresponding screws of Horn B (one each of brass, copper and the third of uncorroded iron, probably part of an iron nail) could not be examined in the same detail but these were also pronounced modern. Below the collars and inserted into the horns for

354

C. Stephen Briggs

attachment to the base-plate copper tube 3 cm in diameter have been inserted. In each case, they are left to project 1.5 cm, and then (quite ingeniously) was cut into 8 narrow strips with hammered bases bent to right angles onto the underside of the head-plate surface. Two phases were detected in this apparatus, as evidence was detected demonstrating how one earlier strip had been soldered on to the surface (Torrs, 209). This is all that now testifies to the existence of two predecessor tubes. At some point, melted resin had been used to hold the collars in place (ibid. 210). Repairs on flange A were probably executed at the same time as modern repairs in copper to the fore-end of the head-piece (ibid. 209). c The use of modern materials Copper, solder, paper, pitch, plaster and putty were all observed during the 1955 examination. All the tinsmiths’ solder has to be post-discovery. Indeed, Train mentioned its existence in his day, so the first phase of restoration must have happened between 1812 and 1829. Whether or not plaster was also a medium used at that time or later is unclear. Evidence for it was still detectable in 1955, when ‘the broken parts [were] apparently [found to have been] held in place by adherence to an internal mass of hardened plaster or putty’ (Torrs, 208). Evidence for molten pitch was found plugged in to Horn A by a piece of tissue paper containing mechanical wood-pulp which was not introduced until 1869.

Bibliography Anon. 1812. Account of a ‘very curious relic of antiquity’, Caledonian Mercury, 17th December 1812. Anon. 1970. Early Celtic Art. Edinburgh: University Press for the Arts Council of Great Britain. Atkinson, R. J. C. and Piggott, S. 1955. The Torrs Chamfrein, Archaeologia 96, 197–235. Briggs, C. S. 1987. Buckets and Cauldrons in the Late Bronze Age of North-west Europe: a review (with appendix on buckets by M. Holland). In J. C. Blanchet et al. (eds.) Les Rélations entre le Continent et les Iles Britanniques a L’âge du Bronze, Actes du colloque de Lille dans le cadre du 22ième Congrès Préhistorique de France, 2–7 Septèmbre, 1984, supplément de la Revue Archéologique de Picardie, Socièté Préhistorique Française, 161–86. Amiens: Revue Archéologique de Picardie. Briggs, C. S. 1995. Did he fall or was he pushed? Some Unresolved questions about Lindow Man. In R. C. Turner and R. Scaife (eds.) Bog Bodies: New Discoveries and New Perspectives, 223–251. London: British Museum Press. Carter, S. and Hunter, F. 2003. An Iron Age chariot burial from Scotland, Antiquity 77, 531–5. Carter, S., Hunter, F. and Smith, A. 2010. A 5th century BC Chariot Burial from Newbridge, Edinburgh, Proceedings of the Prehistoric Society 76, 31–74. Cheape, H., Cowie, T. and Wallace, C. 2003. Sir Walter Scott, the Abbotsford collection and the National Museums of Scotland.

In I. G. Brown (ed.), Abbotsford and Sir Walter Scott: The Image and The Influence, 49–89. Edinburgh: Society of Antiquaries of Scotland. Fox, C. 1946. A Find of the Early Iron Age from Llyn Cerrig Bach, Anglesey. Cardiff: National Museum of Wales. Fox, C. 1958. Pattern and Purpose: A Survey of Early Celtic Art in Britain. Cardiff: National Museum of Wales. Gerloff, S. 2010. Atlantic Cauldrons and Buckets of the Late Bronze and Early Iron Ages in Western Europe, Prähistorische Bronzefunde. Stuttgart: Steiner. Harding, D. W. 2002. Torrs and the early La Tène Ornamental Style in Britain and Ireland. In B. Ballin Smith and I. Banks (eds.) In the shadow of the brochs: the Iron Age in Scotland, 191–204. Stroud: Tempus. Harris, S. and Yalden, D. W. (eds.) 2008. Mammals of the British Isles, Handbook, 4th ed. Southampton: The Mammal Society. Haworth, R. G. 1971.The horse harness of the Irish Early Iron Age, Ulster Journal of Archaeology 34, 26–49. Hencken, H. 1971. The Earliest European Helmets. Cambridge, Mass: Peabody Museum of Archaeology and Ethnology, Harvard University. Henig, M. 1974. A Coin of Tasciovanus, Britannia 5, 275–6. Hunter, F. 1997. Iron Age hoarding in Scotland and northern England. In A. Gwilt and C. Haselgrove (eds.) Reconstructing Iron Age societies: New approaches to the British Iron Age, 108–133. Oxford: Oxbow. Jac. Archive, The Institute of Archaeology, Oxford. Jacobsthal, P. 1944. Early Celtic Art. Oxford: The Clarendon Press. Jope, E. M. 1983. Torrs, Aylesford, and the Padstow Hobby Horse. In O’Connor and Clarke 1983, 149–59. Jope, E. M. 1986. Paul Ferdinand Jacobsthal. In D. Ellis Evans, J. G. Griffith and E. M. Jope (eds.) Proceedings of the Seventh International Congress of Celtic Studies in Oxford, 10th to 15th July, 1983, 15–18. Oxford: D. E. Evans. Jope, E. M. 2000. Early Celtic Art in the British Isles. Oxford: Oxford University Press. Macdonald, P. 2007. Llyn Cerrig Bach, a study of the copper alloy artefacts from the Insular La Tène assemblage. Cardiff: National Museum of Wales and University of Wales Press. Macgregor, M. 1976. Early Celtic Art in North Britain. Leicester: University Press. Megaw, J. V. S. 1970. Art of the European Iron Age: a study of the elusive image. Bath: Adams & Dart. Megaw, J. V. S. 1983. ‘From Transdanubia to Torrs: further notes on a Gabion of the late Jonathan Oldbuck’, in D. V. Clarke and A. O’Connor, (eds), From the Stone Age to the FortyFive: studies in Scottish material culture presented to R. B-K. Stevenson, John Donald: Edinburgh, 127–48. Megaw, R. and Megaw, J. V. S. 2005. Early Celtic Art in Britain and Ireland, 2nd Edition. Shire: Tring. Megaw, R. and Megaw, J. V. S. 2001. Celtic Art from its beginnings to the Book of Kells, 2nd Edition. London: Thames and Hudson. Munby, J. and Henig, M. 1971. New Light on the Torrs Chamfrein? Unpubl. typescript submitted to the journal Antiquity on 13th October, 1971. O’Connor, A. and Clarke, D. V. (eds.) 1983. From the Stone Age to the ‘Forty Five: Studies Presented to R.B.K Stevenson, Former Keeper, National Museum of Antiquities of Scotland. Edinburgh: John Donald.

36.  The Torrs Chamfrein or head-piece: restoring ‘A very curious relic of antiquity’ Patterson, J. 1857. Memoir of Joseph Train, F.S.A.Scot., Glasgow: Thomas Murray and Sons. Edinburgh: John Menzies. Piggott, S. 1950. Swords and Scabbards of the British Early Iron Age, Proceedings of the Prehistoric Society 16, 1–28. Piggott, S. 1969. Early Iron Age ‘horn-caps’ and yokes, Antiquaries Journal 49, 378–81. Piggott, S. 1983. The Earliest Wheeled Transport from the Atlantic Coast to the Caspian Sea. London and New York: Thames and Hudson. Piggott, S. and Daniel, G. E. 1951. A Picture Book of Ancient British Art. Cambridge: University Press. Raftery, B. 1983. A Catalogue of Irish Iron Age Antiquities, 2 vols, Veroffentlichung des Vorgeschichlichen Seminars Marburg, Sonderband I. Marburg. Raftery, B. 1984. La Tène in Ireland: Problems of Origin and Chronology, Veroffentlichung des Vorgeschichlichen Seminars Marburg Sonderband 2. Marburg.

355

Spratling, M. G. 1972. Southern British Decorated Bronzes of the Late Pre-Roman Iron Age, Thesis submitted for the degree of Doctor of Philosophy, University of London. Smith, J. A. 1867–8. Notes of a Remarkable Bronze Ornament with Horns found in Galloway, now at Abbotsford, Proceedings of the Society of Antiquaries Scotland 7, 334–57. Stead, I. M. 1985. Celtic art in Britain before the Roman conquest, 1st ed. British Museum: London. Stead, I. M. 1996. Celtic art in Britain before the Roman conquest, 2nd ed. British Museum: London. Train, J. 1829. National Library of Scotland MS 912, fo.80. Drawing by Joseph Train of Torrs ornament. Train, W. u.d. Biographical Memoir, National Library of Scotland MS 3277.

37 VINCENT, IN APPRECIATION Mansel Spratling

Vincent, I wish you Happy Birthday and many more of them, and good health. I write about what art is, about the Celts and Early Celtic art, as well as the worth of you. The bare essentials.

Art …can be most briefly summarised in the form of an equation of my own devising. Where a stands for art, c for culture and e for excellence,   Let: a = ce, c and e multiplied rather than added to effect the transformation. Should it be referred to, I call it ‘special’ theory in the hope one day of ‘general’, on which I have got nowhere in ten years. I have made no systematic testing of societies worldwide, but it often seems to work very well. The concept of art wholly depends upon excellence and of the person who made it; art and artist indivisibly go together and originated in Renaissance Italy about 500 years ago. If you want to see almost continuous art, visit St Peter’s and the Vatican in Rome. Patrons gave in to artists’ demands for better working conditions and pay and higher status in society. What artists claimed was that their work was a cut above the others’ (compare Martindale 1972). The definition and practice of art today still follow the original definition, underwriting such things as the Turner and Stirling Prizes, the Venice Biennale, the Cannes Film Festival, Oscars and, yes, the Chelsea Flower Show.

The Celts …are considered to have settled in a large part of Europe, but who exactly were they? We read often without distinction of Celts, Gauls and Galatians; surely different terms mean different things, however similar sounding, if not so much as megaw, macaw and meerkat. I examine here what Caesar, Livy and Herodotus had to say. Caesar and Livy say more or less the same thing, but of times widely separated, over 500 years, and of France and Italy in turn. Caesar opens his war in Gaul with this: Gaul comprises three areas, inhabited respectively by the Belgae, the Aquitani and a people who call themselves Celts, though we call them Gauls. All of these have different languages, customs and laws. The Celts are separated from the Aquitani by the river Garonne, from the Belgae by the Marne and the Seine (Caesar I.1, Handford and Gardner 1982, 28).

Clear enough, and clear that the main body of Early Celtic Art as defined by Jacobsthal (1944) was located to the east of the Caesarian Celts. Next Livy reporting a Roman account of the time of the Sack of Rome as relayed by Celts of Italy. I cite only the beginning of a long passage in his History of Rome, V. 34: ‘The following account has come down to us of the Gallic migration. During the reign of Tarquinius Priscus in Rome [600 BC], the Celts, one of the three Gallic peoples [that is, in Gaul] were dominated by the Bituriges…’ (de Sélincourt and Ogilvie 1971, 379). Odd that, three peoples 550 years apart, though the Bituriges are just one of many in Caesar (I.18 etc). If you are feeling uncomfortable, let us take a look at Herodotus who provides the kiss of death in two passages in his Histories (II.33;

37.  Vincent, in appreciation IV.50). Herodotus is writing in La Tène A, about 1 or 2, that is, about 430 BC just after the Parthenon was completed during the ascendancy of Pericles: …the Danube, which has its source amongst the Celts near Pyrene and flows right through the middle of Europe to reach the Black Sea at the Milesian colony of Istria. The Celts live beyond the Pillars of Hercules, next to the Cynesians who are the most westerly people of Europe.

And …the Danube which, rising among the Celts, the most westerly after the Cynetes of all European nations, traverses the whole length of the continent before it enters Scythia (de Sélincourt and Marincola 1996, 98 and 232).

The main problem here is not so much that the Danube does not run through all of Europe as that we think he means the river rose in Württemberg or more particularly Schwaben for which he provides no support. Herodotus could have mentioned the transalpine trade from Etruria, but he nowhere refers either to the alps or to the Etruscans let alone to the older trade up the Rhone from Marseilles. This is a part of Europe of which he knew nothing. So how do we resolve the problem of the Danube’s length? One possibility is that another river was tagged on (compare Avons and Ouses in England). If we examine what else Herodotus says, we arrive in the most westerly parts of the continent. First, one goes out to the Celts via the Straits of Gibraltar (to go on round Iberia, across the Bay of Biscay, up the English Channel and up the Rhine seems a very odd way of going about things). Then there are the Cynesians/Cynetes who are the only people living to the west of the Celts. Finally, there is Pyrene. Why this cannot be the Pyrenees is one of those mysteries in which Celticists freely indulge. In conclusion, taken together, Caesar, Livy and Herodotus are essentially telling the same story, which is that the Celts lived in the most westerly parts of continental Europe. Despite the fine accompanying book, the 2012–13 Stuttgart exhibition was thus misconceived. And no trace of a Celt in prehistoric Britain and Ireland. The linguistic issue is like this. Just because English is the dominant language in the Republic, it does not follow that the Irish are English (compare and contrast Collis 2003).

Early Celtic Art …is a concept which largely derives from the two-volume work written by Paul Jacobsthal published by the Clarendon Press in 1944. The problem for us now is that the great majority of finds studied by him come from lands which cannot historically be identified as having been inhabited by Celts in antiquity. The situation may be improved when Vincent completes and publishes his heralded Supplement

357

(to 1995) to Jacobsthal’s corpus. I am well aware of how distributions of archaeological finds can profoundly change. Still, what are we to do? There is a simple way out. That is to detach Early Celtic Art from historically known Celts and to preface the sequence of European styles with Early Celtic. Thus, Early Celtic, Romanesque, Gothic, Renaissance, Baroque, Neo-Classical and International Modern, for example. No-one suggests that Goths were responsible for Gothic, so the same can obtain with Early Celtic. I am not dodging any issues here, for we are still up against the same problem in interpretation. What did Early Celtic Art stand for? Fürsten and their followers? Not at all likely, because we only find Fürsten in the Medieval and Modern periods (along with Kurfürsten). What there appear to have been in abundance in the Iron Age were kings and queens. Kingship is an ancient Indo-European/-Germanic institution of which Britain’s Queen is the oldest living exemplar (her lineage reaches back in Scotland well before AD 1000). By the mid-first millennium BC some of these societies were overthrowing their kings for good, for example, Athens, Corinth and Rome. There are hints that late in the Iron Age this phenomenon was taking hold in (transalpine) Gaul. Please note here, if you haven’t worked it out, that a Celt is a kind of Gaul and not the other way around; in Athenian studies a Gaul is a kind of statue in Pergamene times (Smith 1991, 99–104, Pls 118–32).

The worth of Vincent …in the study of Early Celtic Art is what I shall concentrate upon, aware that there are other things of which I know little. Details of his life and work can be found in his 2011 memoir in Antiquity, much of which was news to me. But let me begin by lightening my sombre mood a little with a couple of limericks, the peculiarly English art-form with rhyming pattern aabba developed by Edward Lear (1812–88; cf. Jackson 1947, much reprinted) and written here for the occasion. The first was polished by Julia Farley, research fellow at Leicester. The second refers to a Suffolk village at the head of the Alde estuary, home of Julia’s boyfriend’s parents, a nice place for a day out and home to the international Aldeburgh Music Festival whose opening in 2013 happened to coincide with a visit. Music plays a large part in Vincent’s life, though I think the Salzburg Festival is more to his taste, particularly when he is digging at the nearby famed Iron Age site of Hallein. I There was a young man from down under Old upside-down faces he did ponder He found them in traces From all the known phases, That man he’s a real Celtic wonder.

358

Mansel Spratling

II There was an old man from Snape, Who showed his scabbard and chape When they asked ‘Dredged from the fen?’ He said ‘They come from La Tène’, That sensible old man from Snape.

Vincent’s artistic family heritage detailed in the 2011 memoir helps us understand why he has always since student days been interested in ancient art. That the painter John Constable also had a Hampstead village home may be significant too. As Vincent began fully to come into his own in the Sixties, he was to encounter a world, the world of younger prehistorians, which was barely interested in art as work, a corner of the discipline which regarded art as for airy-fairies, as something more than a little suspect. This new generation laid the greatest store in everyday culture, as if there were noblesse in the proletariat of artefacts. Previous generations had not shared this view, concerned as they were with the generality of past experience. New-style homes mixed posters of Botticelli and Picasso on plain living-room walls, Che Guevara and Salvador Dali in the bathroom and a Modigliani nude in the bedroom. The cultural referents of the young prehistorian were very confusing; I hesitate to say confused, for who could doubt their polymathic status. Images of the (Australian) feminist Germaine Greer burning her bra were in short supply, Brigitte Bardot without hers everywhere. It was into this rather complacent world, one which pretended that it had never watched Bambi or its girlfriends’ never having imbibed a babycham, that Vincent (1970) blasted his way with his seminal paper, ‘Cheshire Cat and Mickey Mouse: analysis, interpretation and the Art of the La Tène Iron Age’, in a journal, Proceedings of the Prehistoric Society, usually fed a diet of axehead petrology and wooden tracks. The paper (his best in my view) hurtled from down under into the apparently radical milieu of D. L. Clarke and Lewis Binford, gurus of the age. The dominant message was ‘Are you sure what you are looking at?’ Of course, few could say that they were. Archaeological (indeed art historical and scientific) procedure turns in a simple cycle: look, think, look again and see. Vincent was telling us something very fundamental: if we are not sure (or even if we are), pause and consider the possibilities and not just in visual image and pattern. Few in archaeology there have been who have pushed quite so hard at the distinction between what we see and what is actually ‘there’. Earlier in 1970 Vincent’s first book, Art of the European Iron Age: a study of the elusive image had been published, a work I had glimpsed in 1968–9 in typescript on our first meeting in London, he over on sabbatical leave from Sydney. His coverage was broader than Jacobsthal’s, including Britain and Ireland as well down to the end of the pagan period in Ireland, C2 in Britain. He was eager for corrigenda and addenda; I had a surfeit of data. Others all over had been or were being consulted, typical of the thoroughness

which makes so much of his published work invaluable. Art of the European Iron Age is an unusual treatment in art of the human form (mainly the face) and of further interest for its rare deployment in prehistoric studies of the catalogue raisonné; this is no easy medium. The book was well received by reviewers. The paper and the book had two consequences. First, a search high and low by other students for unrecognised faces within patterns both on the continent and in Britain and Ireland. There is not quite an inexhaustible supply, but it often feels like it. Further ones are identified in this volume by Jennifer Foster. It has to be said that the preoccupation with faces and heads in pattern has not been matched by a search for other parts of their bodies. The second consequence, and an immediate one, was being offered and taking the Chair of Archaeology at Leicester vacated by Charles Thomas. That was 1971. At Sydney he had also become interested in Australian aboriginal art which he hoped to develop at Leicester. However, his time back in England proved somewhat disappointing, for he was unable to broaden the British framework of the teaching and coursework. On top of that he was to learn that the Australian Institute of Aboriginal Studies headed by Peter Ucko could not allow him to do the fieldwork he wanted to do, because apparently that could then only be done by aboriginal researchers. He thus eventually took the decision to return to Australia, where following greatly increased resources allocated to archaeological and anthropological research by Whitlam’s Labour government he was able greatly to extend his operations in the Early Celtic Art domain on which he has concentrated his attention on for the past thirty years. Finding himself at Flinders University in South Adelaide he was able to develop coursework in both the visual arts and archaeology, becoming the only place in the English-speaking world where you could take a course in Early Celtic Art. For a long time Vincent has been a great encourager, especially of the younger and much younger than him. Early ones were Morna MacGregor and Graham Ritchie, Edinburgh students too. The former’s thesis was published at Leicester during Vincent’s time there. Encouragement is important but often overlooked even when we seem most confident. Vincent is one of the few who has travelled all over our Kulturgebiet, as if he were preparing his Habilitationsschrift, examining and photographing, discussing and conferring. His hands-on knowledge of our subject matter is thus really very impressive, even if at times he is unsure and seeks advice from those who perhaps will know. And so at what was supposed to be lunch chez Ian Stead about a decade ago illustrations were produced of a bronze piece from Bohemia he thought might be part of a Basse Yutz/Dürrnberg style flagon. It could not really be anything else, an identification later confirmed after consulting the dimensions table in his and Ruth’s very useful Basse Yutz monograph of 1990.

37.  Vincent, in appreciation Way back in 1976 I gained an insight into how Vincent operated on the international scene, when I found myself inadvertently booked into the seat next to him on an Air France flight from Heathrow to Nice en route to the International Congress of Pre- and Protohistoric Sciences. When the meal was served, we each found that we had been given a not ungenerous little bottle of a rather nice red, to which Vincent tutted that we should not have been given it on a short-haul flight. Not that that affected his consumption; or mine. After that he turned to such matters as his attempts to get his Basse Yutz project moving, which I at the British Museum was supposed to be expediting. As soon as we touched down, he bounded across the arrivals lounge to the car hire and spent the rest of the day (some hours) ferrying delegates (me first) from the airport to the Congress venue in the university high above the town. I did not always see as much of him that week as I expected, but I knew he had Pacific interests too. I need to bring this appreciation to an imminent close, since writing references is not my scene. If it has not yet been filled, Vincent, would you like the Abercromby? Chairs can be held in plural. All you have to do is to mark a pile of slips for the Scotland vote and, if you cannot stand bagpipes, the Tattoo is not obligatory. I regret I have little clout in Oxford, so cannot like Beazley for Jacobsthal fix you up with something, poor impecunious Paul Ferdinand with only a wife, a typewriter, a large van full of furniture, a MS of a book in the wrong language, no photographic negatives (confiscated), and a huge pile of papers (now carefully sorted by Sally Crawford and Katharina Ulmschneider) including school reports needed to matriculate in an English university. How he managed to steal all this as a refugee across the Grenze is not clear. How he managed to get as far as Berlin as late as 1937 without papers with ‘Jude’ stamped on them is also not clear. And back.

In conclusion …let me say that there are those who want me to say, Vincent, how nice you are. I cannot dissent from that. And helpful. Not all are so complimentary as one opining: ‘I wouldn’t buy a used car from him’. I cannot dissent from that either, as there’s bound to be a dead ’roo under the bonnet. The worst thing, Vincent, is that I have been asked to assess your likely legacy, to which I (of course) replied: ‘He’s not dead yet’, but I do know what is meant. Apart from the myriad papers which document and assess, papers which set an example in how to look closely and which encourage us to take both a broad and a detailed look simultaneously – and the very singular Cheshire Cat/Mickey Mouse one of 1970 – I would like to focus attention upon the three books, Art of the European Iron Age (1970), Celtic Art from its beginnings to the Book of

359

Kells written with Ruth his wife (1989, enlarged edition with an extra chapter, 2001) and the still-to-be completed massive supplement to Jacobsthal’s Early Celtic Art. These three books, all without competitors, will form the core of his legacy. Few, I imagine, will manage to read and digest all that he has written upon Early Celtic art continuously for more than 50 years at a rate of about one item every other month. This prodigious quantity bears witness to a mind well able to concentrate despite the usual distractions, a mind full of creative scholastic energy. For it all he deserves some kind of medal and that, modestly, he has kept from us, Membership of the Australian Order of Merit in 2004. Prima facie the Supplement will bring the subject area the most value, reinvigorating inquiry into how Early Celtic Art may be divided by style and by phase and indeed by area and into its meaning in relation to La Tène culture in general. But such things are hard to predict. Given Vincent’s well known interaction with fellow investigators all over Europe, the text and illustrations of the 1989/2001 Celtic Art, much improved in clarity through collaboration with his wife Ruth, are likely to sustain a long life. A quarter of a century already, long years these days for a renewed print run. And half a century since I began to read his work and made contact, my first letter further than the Ruhrgebiet. I leave it there, Vincent, and all you others, for though I could write more I think that enough is probably enough. So keep writing Vincent in your oldest house on Eden Hills (1838, Victoria’s Coronation Year), and Happy Birthday again.

Acknowledgements I take full responsibility for all I have written; what Caesar, Herodotus and Livy wrote is their affair. I thank Vincent for a lot of offprints (not a complete set) and for copies of his Art of the European Iron Age and the enlarged edition of his and Ruth’s Celtic Art. I thank Professor Mike Rowlands and Dr Ian Stead for advice on major matters; the lesser comment but no less interesting Drs Paul Craddock, Julia Farley, Jennifer Foster and Jody Joy, (Miss) Val Rigby and Dr Paul Sealey. I decided to make no contact with Vincent, lest he wonder what was going on. Last but first I thank Drs Sally Crawford and Katharina Ulmschneider for asking and ever so nicely pressing me to write here at any length (halved), to Sally for offering to convert my typescript into electronic language and then doing so, and to them both for a three-hour presentation in an Oxford basement of the Jacobsthal Project to Jody Joy accompanying me in late spring 2013. Their booklet Persecution and survival: the Paul Jacobsthal Story (Institute of Archaeology 2012) is now essential reading for a proper understanding of the genesis of Early Celtic Art.

Mansel Spratling

360

Art of the European Iron Age Academic reviews of Art of the European Iron Age (1970) listed in my copy, collected at the time: Driehaus, J. 1972. Bonner Jahrbücher 172, 613– 16. 2 Spratling, M. G. 1972. University of London Institute of Archaeology Bulletin 10/1971, 148–52. 3 Jope, E. M. 1971. Ulster Journal of Archaeology ser. III, 34, 116–19. 4 Sandars, N. K. 1971. Antiquity 45, 227–8. 5 Powell, T. G. E. 1972. Archaeological Journal 128/1971, 269–70. 6 Savory, H. N. 1972. Archaeologia Cambrensis 120/1971, 116–18 7 Schwappach, F. 1975. Germania 53, 226–36. 8 Alexander, J. 1975. Proceedings of the Prehistoric Society 41, 253–4. 1

Bibliography Collis, J. R. 2003. The Celts. Origins, Myths and Inventions.

Stroud: Tempus. De Sélincourt, A. and Marincola, J. (trans.) 1996. Herodotus. The Histories. New Edition. Harmondsworth: Penguin Books. De Sélincourt, A. and Ogilvie, R. M. (trans.) 1971. Livy. The Early History of Rome. Books I–V of the History of Rome from its Foundation. Harmondsworth: Penguin Books. Handford, S. A. and Gardner, J. F. (trans.) 1982. Caesar. The Conquest of Gaul. Harmondsworth: Penguin Books. Jackson, H. (ed.) 1947. The Complete Nonsense of Edward Lear. London: Faber and Faber. Jacobsthal, P. F. 1944. Early Celtic Art. 2 volumes. Oxford: Clarendon Press. Martindale, A. 1972. The Rise of the Artist in the Middle Ages and Early Renaissance. London: Thames and Hudson. Megaw, J. V. S. 1970. Art of the European Iron Age: a Study of the Elusive Image. Bath: Adams and Dart. Megaw, J. V. S. 1970. Cheshire Cat and Mickey Mouse. Analysis, interpretation and the art of the La Tène Iron Age, Proceedings of the Prehistoric Society 36, 261–79. Megaw, J. V. S. and Megaw, R. M. 2001. Celtic Art from its Beginnings to the Book of Kells. Revised and enlarged edition (original 1989). London: Thames and Hudson. Smith, R. R. R. 1991. Hellenistic Sculpture, a Handbook. London: Thames and Hudson.

Bibliography EMERITUS PROFESSOR JOHN VINCENT STANLEY MEGAW, MA DLitt (Edinburgh) FAHA FSA HonFSAScot Korresp Mitglied des Deutschen Archäol Inst MiFA

Membership of Editorial Boards i Editor of the Elek Archaeology and Anthropology (Paul Elek Ltd. 1974–9; Granada Publishing Ltd. 1979–82). ii Archaeology Area Editor for the Repertoire Internationale de Litterature Musicale (= RILM Abstracts 1965–75). iii Consultant Editor for the Kelta Corpus (= Corpus of Celtic Material in Hungary) being prepared under the auspices of the Hungarian Academy of Sciences (MTA). iv Member, Editorial Advisory Board, Mediterranean Archaeology. v Advisory Editor, Antiquity, Cambridge (1987–93). NB: In the following list of c.300 publications, reviews are not listed save for major or review articles; equally, curated exhibitions are not separately listed except where they have been marked by published catalogues or related publications.

Bibliography 1957 Folklore and tradition on North Uist, Folklore LXVIII, 483–8. 1957–58 Neolithic Cornwall, Proc. W. Cornwall Field Club. II:1, 13–25. A bronze bull’s head in Glasgow and its affiliations, Proc. Soc. Antiq. Scot. XCI, 179–182.

1959–61 Contribution to Introduction and Gazetteer sections of the following volumes of the ‘Buildings of England’ series (ed.) (Sir) Nikolaus Pevsner (Penguin Books Ltd., Harmondsworth); Buckinghamshire, Suffolk, Northamptonshire, Norfolk (North), Norfolk (South) and Surrey. 1960 Penny whistles and prehistory, Antiquity XXXIV, 6–13. ‘Hallstatt’, ‘Leubingen’, ‘La Tène’: articles in (ed.) Leonard Cottrell, Concise Encyclopaedia of Archaeology. London: Hutchinson, revised ed. 1970. With A. C. Thomas and Bernard Wailes, The Bronze Age settlement at Gwithian, Cornwall: preliminary report, Proceedings of the West Cornwall Field Club II:5, 200–15. 1961 Across the North Sea: a review, Antiquity XXXV, 45–52. Penny whistles and prehistory: further notes, Antiquity XXXV, 55–57. The Neerharen silver vase, Helinium I.3, 233–41. An end-blown flute or flageolet from White Castle, Medieval Archaeology V, 176–8. 1959–61 With D. A. Simpson. A short cist burial on North Uist and some notes on the prehistory of the Outer Isles in the second millenium BC, Proceedings of the Society of Antiquaries of Scotland XCIV, 61–78.

362

Emeritus Professor J. V. S. Megaw Bibliography

1962 Translation and adaptation (from the Dutch) with Thérèse Megaw. M. J. Vermaseren, Mithras, the Unknown God. London: Chatto and Windus. A bronze mount from Mâcon: a miniature masterpiece of the Celtic Iron Age reappraised, Antiquaries Journal XLIII, 24–29. 1963 A British bronze bowl of the Belgic Iron Age from Poland, Antiq. J. XLIV, 27–37. A medieval bone pipe from White Castle, Monmouthshire, Galpin Soc. J. XVI, 85–94. The Neolithic in the South-West of England: a reply and some further comments, Cornish Arch. 2, 4–8. ‘An outline archaeology of Europe’ and ‘Hand list of exhibits: Foundations of Europe, 6000 BC–AD 600’ in (ed.) Noel Barnard, Jack Golson and Helmut Loofs, Patterns of Culture: a Handbook, 1–12. Canberra: ANU. The Australian Institute of Aboriginal Studies and the 1963 Conference on Nomenclature of Implements and Culture, Antiquity 36, 298–301. With James Barber. A decorated Iron Age bridle-bit in the London Museum, Proceedings of the Prehistoric Society 29, 206–13. 1964 Notes and queries: a medieval bone pipe from White Castle, Galpin Society Journal 17, 116–7. An Irish gold neck-ring in the Nicholson Museum, Sydney, Journal of the Cork History and Archaeology Society 69, 94–100. 1965 Excavations in the Royal National Park, NSW: a first series of radiocarbon dates from the Sydney district, Oceania 35:3, 212–7. Iona and Celtic Britain with an interim account of the Russell Trust excavations, 1958–1963, Journal of Religious History 3, 212–37. A bronze spear-head from the Heathery Burn Cave, Co. Durham, Antiquaries Journal 46, 112–4. New light on the most ancient West, Cornish Archaeology 4, 46–50. Museum without walls: the history of ‘the Nicholson’, Teaching History 17, 7–15. 1966 Report on excavations in the South Sydney district, 1964/65, Australian Institute of Aboriginal Studies: Newsletter 2:3, 4–15. The Vix burial reconsidered, Antiquity 40, 38–44. A ‘bull-cult’ in Late Bronze Age Europe: a musical footnote, Antiquity 40, 56–58.

With a contribution from R. V. S. Wright. The excavation of an Aboriginal rock shelter in Gymea Bay, Port Hacking, NSW, Arch. and Phys. in Oceania 1:1, 23–50. Lidská tvár a Umení ranych Keltu, Kulturni; Tvorba 35:IV (1 Sept.), 4–5. With John Burke. British decorated axes: a footnote on fakes, Proc. Prehist. Soc. 32, 34–6, 3pls. A British bronze bowl of the Belgic Iron Age from Poland and its implications, Atti del VI Congresso Int. Sci. Pre& Protostoriche (Rome) 3, 144–147. Australian Archaeology – how far have we progressed? Mankind 6:7, 306–12. ‘Evening in a Scots Cottage’ – a note on the stock-and-horn, Scottish Studies 10, 119–211. A Celtic cult head from Port Talbot, Glamorganishire, Arch. Cambrensis CXV – published 1967, 94–98. 1965–66 Two La Tène finger rings in the Victoria and Albert Museum, London: an essay on the human face and Early Celtic Art, Praehist. Zeitschr. 43/44, 96–166. A carved cult figure from Maastricht, Ber. v.d. Rijksdienst voor het Oudheidkundig Bodemonderzoek 14–15, 109– 12. 1967 Radiocarbon dates from Curracurrang Cove, NSW, Austr. Inst. Aboriginal Studies: Newsletter 2:5, 26–30. With Anne Nippard. A horse hoof core from Wattamolla, NSW, Mankind 6:8, 357–62. Tribute to Sir Cyril Fox: the Trenoweth collar, Cornish Arch. 6, 6–8. Ein verzierter Frühlatène-Halsring im Metropolitan Museum of Art, New York, Germania 45, 59–69. Art styles and analysis, Mankind 6:9, 393–401. Archaeology, art and Aborigines: a survey of historical sources and later Australian prehistory, Journal of the Royal Australian Historical Society 53, 277–94. Notes on Iron Age and Neolithic material from Sidbury Camp, North Tidworth, Wilts. Arch. Mag. 62, 115–117. The animal-headed torc from Vieille-Toulouse, Antiq. J. XLVIII, 209–13. A further note on Celtic cult heads in Wales, Arch. Cambrensis CXVI, 192–4. 1968 Problems and non-problems in palaeo-organology: a musical miscellany. In (ed.) J. M. Coles and D. D. A. Simpson, Studies in Ancient Europe: essays presented to Stuart Piggott, 333–58. Leicester: Leicester University Press. Two finds of the Celtic Iron Age from Dodona. In (ed.) K. Jazædzæewski, Liber Iosepho Kostrzewski octogenario, 186–93. Warsaw. Trial excavations in the Captain Cook Landing Place

Emeritus Professor J. V. S. Megaw Bibliography Reserve, Kurnell, N.S.W., Aust. Inst. Aboriginal Studies: Newsletter 2:9, 17–20. A dated culture sequence for the South Sydney region of New South Wales, Current Anthropology 9–4, 325–9. With Julia Cox and Leslie Maynard. The excavation of a rock shelter at Audley, Royal National Park, N.S.W., Arch. and Phys. Anthrop. in Oceania III, 94–104. The earliest musical instruments in Europe, Archaeology 231, 124–32. Les fragments de feuille de bronze décorés de Levroux (Indre), Gallia, XXVI, 33–41. Une epée de La Tène I avec fourreau décoré, Rev. Arch. de l’Est et du Centre-Est XIX, 129–44. 1969 Captain Cook and bone barbs at Botany Bay, Antiquity XLIII, 213–6. With D. F. Branagan. The lithology of a coastal Aboriginal settlement at Curracurrang, N.S.W., Arch. and Phys. Anthrop. in Oceania IV, 1–17. Two native bronzes of the Roman period from London in the Nicholson Museum, Sydney, Trans. London and Middlesex Arch. Soc. 22, 45–47. An end-blown flute from medieval Canterbury, Medieval Archaeology 12, 149–50. Analyses of British Early Bronze Age axes in the Nicholson Museum, Sydney, Proc. Prehist. Soc. 34, 354–64. Captain Cook and the Australian Aborigine, Australian Natural History 16:8, 255–60. Doppelsinnigkeit in der keltischen Kunst, dargestellt an einem Beispiel aus dem Fürstengrab von Bad Dürkheim, Pfälzer Heimat 3, 85–6. 1970 With Ralph Merrifield. The Dowgate plaque, Arch. J. 126, 154–59. The elusive image in La Tène art. In (ed.) E. Anati et al. Valcamonica Symposium. Actes du Symposium International d’Art Préhistorique, 507–14. Capo di Ponte, Brescia: Edizioni del Centro. Cheshire Cat and Mickey Mouse: analysis, interpretation and the art of the European Iron Age, Proc. Prehist. Soc. 36, 261–79. The art of the European Iron Age: a study of the elusive image. Bath: Adams and Dart/ New York: Harper and Row. European precedents in the protection of archaeological sites, Newsletter Austr. Inst. Abor. Studies 2:12, 22–31. Towards an absolute chronology of the Australian Stone Age, Actes du VIIe Congr. Int. des Sci. Pre. et Protohist. 213–5. Prague. Further Early La Tène rings and other material of the ‘Horchheim’ and ‘Andernach’ classes, Germania 48, 126–30.

363

1971 Employ’d as a Discoverer: proceedings of the Captain Cook Bi-Centenary Symposium, Sutherland (Sydney and Wellington: A. H. & A. W. Read) – edited by J. V. S. Megaw and including a paper on ‘Cook and the Aborigines’, 55–63. The Maskenfibel fragment from Ksice, Stri;bro, okr. Tachov, Arch. roz. XXIII, 212–4. A Group of later Iron Age collars or neck-rings from Western Britain, British Museum Quarterly XXXV, 145–56. The Human face in Early Celtic Art – some problems of analysis, Actes du VIIe Congr. Int. des Sci. Pre. et Protohist. II, 817–20. Prague. The Felmersham fish-head spout, Antiq. J. LI, 2199–300. An unpublished Early La Tène Tierfibel from Hallstatt, Oberösterreich, Arch. Austriaca 50, 176–84. Two axes of the fake ‘Glencar’ class in the County Museum, Armagh, Ulster J. Arch. 34, 17–8. 1970–71 The possible wrest-plank from Dùn an Fheurain (Appendix I to J. H. Graham Ritchie, ‘Iron Age finds from Dùn an Fheurain, Argyll’), Proc. Soc. Antiq. Scot. 103, 106–7. 1972 With Rhys Jones. The Dawn of Man. London: Wayland Publishers. More eighteenth century trophies from Botany Bay?, Mankind 8, 225–6. Marks of natives everywhere: a review article, Mankind 8:3, 230–2. Style and style analysis in continental Early La Tène art, World Arch. 3:3, 276–92. With D. F. Branagan. Geography and Geology: Aboriginal Archaeology Sutherland. Shire Studies 3, 9–12. 1973 Archaeology from Down Under: a personal view (an inaugural lecture 24 February 1973). Leicester: Leicester University Press. A group of later Iron Age collars from Western Britain: an analytical footnote, British Mus. Quarterly, 37, 70–71. With Peter Wade-Martins. A tenth-century bone flute from North Elmham, Norfolk, Galpin Soc. J. 26, 142–3. An Irish Middle Bronze Age spearhead in the Queensland Museum, Brisbane, Mem. Queensland Mus. 16:3, 485–7. Prehistoric economics and archaeology on the coast of New South Wales, Actes du VIIe Congr. Int. des Sci. Pre. et Protohist. 3, 420–9. ‘The Finds from the site of La Tène . . .’ By J. M. de Navarro [review article], Archaeological Journal 130, 299–303 1974 Gussage All Saints: discussion, Antiquity 48, 306–8.

364

Emeritus Professor J. V. S. Megaw Bibliography

An Early La Tène decorated bronze bracelet from Aix-surCloie (Halanzy, Lux.), Helinium 14, 52–56. The recent archaeology of the Sydney district: excavations 1964–67. Aust. Inst. of Abor. Studies Research Publication 1 (Canberra) – edited and with contributions by J. V. S. Megaw.

Miklós Szabó: The origins of the Hungarian sword style, Antiquity 51, 211–20 – translated and edited by Ruth and Vincent Megaw. Editor: Hunters, gatherers and first farmers beyond Europe: an archaeological survey. Leicester: Leicester University Press, revised edition 1980.

1974–75 Deux paires de bracelets du La Tène ancien de la Suisse occidental, Jahrb. d. Schweiz, Ges. f. Urgeschichte 60, 68–70.

1978 The shape-changers: art of the Iron Age Celts, Archaeology 31:3 (May/June), 30–43. Chapter fourteen: The bronze spout. In R. Canham et al., 2000 years of Brentford: London Museum Archaeological Report, 131. London: HMSO. Article ‘Prehistoric art’, in Encyclopedia Americana (ed.), 542–5. With H. S. Green. A possible bone flute from Bacon Hole Cave, Gower, Gower XXLX, 52–53. Past research in archaeology and Department of Archaeology, University of Leicester BA theses, Current Research in Archaeology 5 (Nov.), 3 and 8–10. Editor and contributor with Michael Greenhalgh. Art in society: studies in style, culture and aesthetics. London: Gerald Duckworth. Another Tierfibel from Hallstatt, Oberösterreich. Prace i Materialy Muz. Arch. i Etnograf. w Òodzi (Festschr. Jazædzæewski) ser. arch. 25, 263–72.

1975 Notes on Bronze Age antiquities in the South Australian Museum, Records in the South Australian Museum 17:3, 22–29. The bone pipe. In C. Platt and R. Coleman-Smith (eds.), Excavations in Medieval Southampton 1953–1969, 252–3. Leicester: Leicester University Press. 1976 The science of rubbish: aspects of archaeology (review article), Encounter 47: 1, 71–77. (Ed.). To illustrate the monuments: essays on archaeology presented to Stuart Piggott on the occasion of his sixtyfifth birthday. London: Thames & Hudson. Joint compiler. Signposts for archaeological publication: a guide to good practice in the presentation and printing of archaeological periodicals and monographs. Council for British Archaeology, 2nd ed. 1979. The orientalizing theme in early Celtic art, Alba Regia 14 (Proc. Conference ‘The Celts in Central Europe’, Székesfehérvár 1974), 15–33. Gwithian, Cornwall: some notes on the evidence for Neolithic and Bronze Age settlement. In (eds.) C. Burgess and R. Miket, Settlement and economy in the third and second millennium BC., 51–79. British Archaeol. Reports 33. With O-H. Frey. Palmette and circle: Early Celtic art in Britain and its continental background, Proc. Prehist. Soc. 42 (Memorial Volume for T. G. E. Powell), 47–68. An Iron Age sword with decorated scabbard mounts of Piggott’s Group V from Aldwinckle, Northants, Northants Archaeology 2, 165–70. 1975–76 The decorated sword-scabbards of iron from Cernon-surCoole, Marne and Drna, Rimavská Sobota, Slovakia, Hamburger Beiträge zur Arch. III:2, 119–37. 1977 Tibor Kovács: The Bronze Age in Hungary (Corvina, Budapest) – translation revised by J. V. S. Megaw.

1979 Celtic art – product of travelling craftsmen or chieftainly vassals? In (eds.) P.-M. Duval and V. Kruta, Les Mouvements Celtiques à partir du Ve siècle avant notre ère, 49–54. Paris: CNRS. With R. J. Lampert. Unearthing Sydney’s prehistory. In (eds.) P. Stanbury and J. Birmingham, 10,000 years of Sydney Life 64–71. Macleay Museum; University of Sydney. Archaeology. In (eds.) K. Boehm and N. Wellings, The Student Book 1979–80, 57–59. Basingstoke: Macmillan, and annual revisions to 1984. Principal author and editor with D. D. A. Simpson, Introduction to British Prehistory: from the arrival of Homo Sapiens to the Claudian invasion. Leicester: Leicester University Press, 5th impression. Leicester UP/ Barnes & Noble Savage, MD 1992). 1980 The decoration on the sword-scabbard from Jenisuv Ujezd, gr. 115. In (ed.) J. Waldhauser, Das Keltische Gräberfeld bei Jenisuv Ujezd II, 106–13. Teplice: Krajské Muz. Archaeology down under, Popular Archaeology 2:6 (December), 29–33. The view from on top – annotations on a conference theme, Australian Archaeol. 11 (December), 81–89.

Emeritus Professor J. V. S. Megaw Bibliography 1981 Edited with contributions. The archaeology of musical instruments: World Archaeology 12:3. Artists-in-Residence, subjects or objects? Some first reactions to the Aboriginal Artists-in-Residence programme at the Flinders University of South Australia, Austr. Inst. Abor. Studies: Newsletter 15 (March), 50–62. Zum Fragment eines Schwertes mit maskenverzierten Knauf aus Herzogenburg. In (ed.) J.-W. Neugebauer, Herzogenburg-Kalkofen: Materialhefte zu den Fundberichten aus Österreich, 39–45. Kelti Kommt!, Antiquity LV, 125–7. A hundred books of Archaeology, Popular Archaeology 3:1 (July), 18–20. Attenborough’s eyes, RAIN 4:2 (Dec.), 2–3. Une ‘volière’ celtique... Rev. Arch. de l’Est et du CentreEst 32:3–4/ (ed.) S. Deyts, Etudes offertes à J.-J. Hatt, 137–43. 1982 Encountering Attenborough, Popular Archaeology 3:8 (Feb.), 41–43. Finding purposeful patterns: further notes towards a methodology of Pre-Roman Celtic art. In (eds.) P. M. Duval and V. Kruta, L’art celtique de la période d’expansion... 213–29. Paris: Ecole pratique des Hautes Etudes. Western Desert painting – artefact or art?, Art History 5:2 (June), 205–18. With A. Gibson. Ein Kriegergrab mit verzierter Schwertscheide aus dem Dürrnberg, Arch. Korrespondenzbl. 12:4, 491–6. An early La Tène Maskenfibel from Slovenské Pravno, okr. Martin, Etudes Celtiques 19, 7–34. With Christine Brade. The prehistoric flute: did it exist?, Galpin Soc. J. XXV, 138–50. Der Trichtinger Ring... (review article), Bonner Jahrb. 182, 608–9. 1983 From Transdanubia to Torrs: further notes on a Gabion of the late Jonathan Oldbuck. In D. V. Clarke and A. O’Connor (eds.) From the Stone Age to the Forty-five: studies in Scottish material culture presented to R.B.K. Stevenson, 127–48. Edinburgh: John Donald. (Review article) in Nature 305 (13 Oct.), 647–8. Sandor Gallus – archaeologist in two hemispheres, The Artefact 8, 2–8. The present past? a minimal view of ethnographic analogy and rock art. In (ed.) Moya Smith, Archaeology at ANZAAS 1983, 279–92. Perth: Western Australian Museum.

365

1984 Introduction to catalogue, Painters of the Western Desert (Contemporary Australian Art Exhibitions, Adelaide Festival of Arts), no pagination. Bone musical instruments from Medieval Exeter. In (ed.) J. Allen, Medieval and post-medieval finds from Exeter, 349–51. Exeter Arch. Rep. 3 (Exeter Excavations Committee). The archaeology of rubbish or rubbishing archaeology: backward looks and forward glances, Australian Journal of Historical Archaeology 2, 7–12. Dot and circle – paradox, politics and ‘Papunya painting’, Art Network 13, 55–61. A fragmentary bone end-blown pipe from a medieval house site at Stonar, near Sandwich, Arch. Cantiana 99, 266–8. 1985 ‘Introduction’ and ‘Meditations on a Celtic hobby-horse’ in (eds.) T. C. Champion and J. V. S. Megaw, Settlement and society: aspects of West European prehistory in the first millennium BC 1–8, 161–92. Leicester: Leicester University Press. Paintings of the Western Desert: Calendar 1986 (Australian Information Service, Canberra). 1986 Adelaide’s Destitute Asylum: A cautionary tale of rescue archaeology and confrontation politics. In G. K. Ward (ed.), Archaeology at ANZAAS Canberra, 64–73. Canberra: Canberra Archaeological Soc. ‘Cross-Channel trade between Gaul and Britain . . . edited by Sarah Macready and F. H. Thompson’, Germania 64, 225–9 (review article). With M. Ruth Megaw, Early Celtic Art in Britain and Ireland. Princes Risborough: Shire Publications, reprinted with amendments 1990; revised edition 1994. Something, but for whom? ethics and transitional art (rev. version), Cultural Survival Quarterly 109:34, 12–15. Contemporary Aboriginal art: Dreamtime discipline or alien adulteration?, COMA: Bulletin of the Conference of Museum Anthropologists 18, 31–42. With Janet Maughan. The Dreamtime today: a survey of contemporary Aboriginal arts and crafts. SA Jubilee 150 Exhibition catalogue. Bedford Park: The Flinders University of SA. With Janet Maughan and Jenny Zimmer (eds.). Dot and Circle: a retrospective survey of the Aboriginal acrylic paintings of Central Australia. Melbourne: RMIT. 1987 Tibor Kovács, Éva F. Petres and M. Szabó (English text revised and edited by J. V. S. Megaw), Corpus of

366

Emeritus Professor J. V. S. Megaw Bibliography

Celtic finds in Hungary 1 (Transdanubia 1). Budapest: Akadémiai Kiadó. An early La Tène sword ‘from Hallstatt’ in the Evans Collection, Ashmolean Museum, Oxford. Arch. Korrespondenzblatt 17, 97–101. With Andrew Fitzpatrick. Further finds from the Le Câtillon hoard, Proceedings of the Prehistoric Society 53, 433–44. 1988 With Janet Maughan. Yuelamu Dreaming, Artlink 8:1, 21. With Janet Maughan, Ruth Megaw and Polly Sumner. Figures in a landscape ...? Artlink 8:2, 24–29. With Kerry Giles and M. Ruth Megaw. The Cutting Edge: new art of the Third and Fourth Worlds: an exhibition for NADOC Week 1988. EAF/Flinders University. Aboriginal Australians: Visual arts. In J. Jupp (ed.) The Australian people: an encyclopedia of the nation, its people and their origins, 168–74. Angus & Robertson for the Australian Bicentennial Authority. Archaeology without politics? Some European experiences, Australian Archaeol. 26, 52–64. UISPP and WAC: a view from Down Under and across the Main, Archaeological Review from Cambridge 7:2, 218–29. With M. Ruth Megaw. Alting er politik’: Australsk infødt politisk kunst: plakater og farvetryk. University of Aarhus. With M. Ruth Megaw. A La Tène neckring of unknown provenance, Helinium 28:2, 271–7. With M. Ruth Megaw. The stone head from Msecké Zehrovice: a re-appraisal, Antiquity LXII, 630–41. With M. Ruth Megaw. The Dreamers awake: contemporary Australian Aboriginal art. In Anna Rutherford (ed.) Aboriginal Culture Today – Kunapipi, 158–83. University of Aarhus. 1989 With M. Ruth Megaw. Celtic Art from its beginnings to the Book of Kells. London: Thames & Hudson, p/b reprint with corrections (1990, 1994). The Emperor’s new clothes: the new music archaeology?. In Ellen Hickmann and David W. Hughes (eds.) The archaeology of early music cultures: Third International Meeting of the ICTM Study Group on Music Archaeology, 343–53. Bonn: Verlag für systematische Musikwissenschaft. With M. Ruth Megaw. Trevor Nickolls 1973–1989. Catalogue, Deutscher Gertrude Street Gallery, Fitzroy, Vic. With M. Ruth Megaw, Albert Namatjira and the Hermannsburg painters. In A myriad of Dreamings: twentieth century Aboriginal art 101–11. Melbourne: Lauraine Diggins Fine Art.

With M. Ruth Megaw and J.-W. Neugebauer. Zeugnisse frühlatènezeitlichen Kunsthandwerks aus dem Raum Herzogenburg, Niederösterreich, Germania 67:2, 477– 517. With M. Ruth Megaw. Further notes on the Early La Tène openwork mounts from Čižkovice, okr. Litomerice, Arch. Roz. 41:5, 543–8. With M. Ruth Megaw. The Italian Job: some implications of recent finds of Celtic scabbards decorated with dragonpairs, Mediterranean Archaeology 2, 85–100. With M. Ruth Megaw. Horse fibulae in the Early La Tène period, Arqueologia 19, 133–42. 1990 With M. Ruth Megaw. The Basse-Yutz find: Masterpieces of Celtic art. Society of Antiq. Research Report 46. With M. Ruth Megaw. Italians and Greeks bearing gifts: the Basse-Yutz find reconsidered. In Jean-Paul Descoeudres (ed.) Greek colonists and native populations, 579–605. Oxford: Clarendon Press, for the Australian Academy of the Humanities. Bone whistles and related objects. In M. Biddle (ed.) Object and economy in medieval Winchester, Winchester Studies 7.ii, 718–23. Oxford: Oxford University Press. The Anglo-Australian excavations on the Dürrnberg bei Hallein 1978–81: Interim Report, Germania 68, 509–49. A La Tène armring ‘from Hungary’ in the collections of [the] Royal Ontario Museum, Toronto, Communicationes Archaeologicae Hungariae 1988, 41–7. The Dreamtime comes to Park Avenue: the surveyors surveyed (review article), Canberra Anthropologist 13:2, 75–90. Art as identity: aspects of contemporary Aboriginal art. In Allan Hanson and Louise Hanson (eds.) Art and identity in Oceania, 282–92. Honolulu/Bathurst NSW: University of Hawaii Press/Crawford House Press. With A. P. Fitzpatrick. Le dépôt de monnaies celtiques et de parures du Câtillon, Jersey, Isles anglo-normandes, Bull. d’Information: Ass. Manche Atlantique pour la Recherche Archéologique dans les Iles 2, 1989, 55–60. 1990–91 With M. Ruth Megaw. Continental Celtic Art (script for an audio-visual production to accompany a UNESCO travelling photographic exhibition of Celtic art). 1991 ‘La musica celtica’ in I Celti (Bompiani, Milan), 643–8: English version published as ‘Music archaeology and the ancient Celts’ The Celts (Bompiani, Milan/Thames and Hudson, London), same pagination. With M. Ruth Megaw. Questioning the Time Machine (review article), Antiquity 65, 138–41.

Emeritus Professor J. V. S. Megaw Bibliography With M. Ruth Megaw. Felix Müller, Die frühlatènezeitlichen Scheibenhalsringe’ (review article), Bonner Jahrbücher 190, 1990, 618–22. The decoration on bronze ring No 165. In C. R. Musson et al. The Breiddin hillfort: a later prehistoric settlement in the Welsh Marches, 139–41. CBA Research Report 76. With M. Ruth Megaw. The great treasure of Snettisham, Celtic Connections: Newsletter of the Celtic Studies Foundation (Sydney) 6, n.p. With M. Ruth Megaw. ‘Semper aliquid novum...’ Celtic dragon-pairs re-reviewed, Acta Archaeologica (Hungary) 42, 1990, 55–72. M. Ruth Megaw with contributions by J. V. S. Megaw and John Morton. The Heritage of Namatjira (Australian Exhibitions Touring Agency, Melbourne), exhibition catalogue. 1992 With Jane Hardy and M. Ruth Megaw (eds. and introduction). The Heritage of Namatjira: The watercolourists of Central Australia. Melbourne: William Heinemann Australia. With M. Ruth Megaw. Acrylbilder aus Zentral-Australien: Eine junge und doch uralte Kunst. In C. Kaufmann (ed.) TJUKURRPA: Zeitgenössische Malerei der Aborigines Australiens, 4–12. Basel: Museum für Völkerkunde, exhibition catalogue. With Pauline Beswick, M. Ruth Megaw and Peter Northover. A decorated late Iron Age torc from Dinnington, South Yorkshire, Antiq. J. 70, 1990, 16–33. With M. Ruth Megaw. Die Verzierung des Siebtrichters von Hoppstädten, Hundert Meisterwerke Keltischer Kunst, 99–104. Trier: Rheinisches Landesmuseum, exhibition catalogue. With M. Ruth Megaw. The Celts: The first Europeans? (review article), Antiquity 66 (March), 254–60. 1993 With M. Ruth Megaw and Hans Nortmann. Neue Untersuchungen zum frühlatènezeitlichen Siebtrichter von Hoppstädten, Trierer Zeitschrift 55, 1992, 105–28. With M. Ruth Megaw. The discovery and history of the Basse-Yutz flagons, Archaeologia Mosellana 2; Actes du XIe Colloque de l’AFEAF, 331–5. With M. Ruth Megaw. Pacific encounters: visual art exhibitions and the Fifth Pacific Arts Association Symposium, Artlink 13:2, 74–76. ‘Something old something new’: further notes on the Aborigines of the Sydney district as represented by their surviving artefacts, and as depicted in some early European representations. In Jim Specht (ed.) F. D. McCarthy, Commemorative Papers (Archaeology, Anthropology, Rock Art), 25–44. Records of the Australian Museum Supplement 17.

367

With M. Ruth Megaw. The swans of Radovesice reviewed. In Jiří Waldhauser, Wolfgang Dehn, Václav Bareš et al. (eds.)Die Hallstatt- und Latènezeitliche Siedlung mit Gräberfeld bei Radovesice in Böhmen, 227–34. Archeologický výzkum v severních Čechách 21. With M. Ruth Megaw. Cumulative Celticity and the human face in insular pre-Roman Iron Age art. In J. Briard and A. Duval (eds.) Les représentations humaines du Néolithique à l’Âge du Fer – Actes du 115e Congrès national des Sociétés savantes (Avignon, 1990), 205–18. Paris: Éditions du Comité des Travaux historiques et scientifiques. With M. Ruth Megaw. The earliest insular Celtic art: Some unanswered questions. In Venceslas Kruta (ed.) Actes du IXe Congrès int. d’études celtiques Paris, 8–12 juillet 1991: première partie ‘Les Celtes au IIIe siècle avant J.-C., 283–307. Études Celtiques 28. With M. Ruth Megaw. Black art and white society: some Bicentennial observations on contemporary Australian Aboriginal art. In Artistic heritage in a changing Pacific; Proceedings of the Fourth International Symposium of the Pacific Arts Association, Honolulu, Hawai’i, 162–72. Honolulu: University of Hawaii Press, in assoc. with Bathurst NSW: Crawford House Press. With M. Ruth Megaw. Cheshire Cats, Mickey Mice, the new Europe and ancient Celtic art. In C. Scarre and F. Healy (eds.) Trade and exchange in prehistoric Europe, 219–32. Oxbow Monographs 33: Oxbow Books for the Prehistoric Society and the Société Préhistorique Française. 1994 With M. Ruth Megaw. Through a window on the European Iron Age darkly: 50 years of reading early Celtic art, World Archaeology 25:3, 287–303. With Kerry Smallwood and M. Ruth Megaw (eds.). Looking towards the future: Contemporary Aboriginal art. Flinders University Art Museum, Bedford Park: exhibition catalogue. With Carl Heron, Robert Jones, M. Ruth Megaw, Peter Northover and Robert Trett. A Late Iron Age cast bronze head from Chepstow, Antiq. J. 72, 54–75. With M. Ruth Megaw. Celtic Art: a reply to the Editor, Ulster J. Archaeol. 54/55, 1991–2, 164–5. ‘There’s a hole in my shield . . .’: a textual footnote, Australian Archaeol. 38, 35–37. Advisory Editor and, with M. Ruth Megaw, contributor, David Horton (general ed.). The Encyclopaedia of Aboriginal Australia. Canberra: Aboriginal Studies Press. Bone pipes from medieval Winchester. In Catherine HomoLechner and Annie Bélis (eds.) La Pluridisciplinarité en archéologie musicale; 4e Rencontres intern. d’Archéol. musicale de l’ICTM, 433–40. Paris: Éditions de la Maison des Sciences de l’Homme. With M. Ruth Megaw. Excavating in museums: two La Tène

368

Emeritus Professor J. V. S. Megaw Bibliography

‘heraldic’ spearheads. In Claus Dobiat (ed.) Festschrift für Otto-Herman Frey, 395–404. Marburger Studien zur Vor- und Frühgeschichte 16.

773–4 in Jane Taylor (ed.) The Macmillan Dictionary of Art. London Macmillan. Nécrologie: Stuart Piggott, Studia Celtica 30, 299–303.

1995 With M. Ruth Megaw. The nature and function of Celtic art. In Miranda Aldhouse-Green (ed.) The Celtic world, 345–75. London: Routledge. Editor, Why? An exhibition of contemporary Aboriginal art from the collections of the Flinders University Art Museum. Bedford Park: Flinders University Art Museum: exhibition catalogue. With Barry Raftery and Val Rigby (eds.). Sites and sights: essays on fieldwork and museum research presented to Ian Mathieson Stead. Oxbow Monograph 56. Oxford: Oxbow Books. With M. Ruth Megaw. An unprovenanced La Tène linch-pin with ‘Vegetal’ decoration in the British Museum. In J. V. S. Megaw, Barry Raftery and Val Rigby (eds.). Sites and sights: essays on fieldwork and museum research presented to Ian Mathieson Stead, 139–48. Oxbow Monograph 56. Oxford: Oxbow Books. With M. Ruth Megaw. The prehistoric Celts: identity and contextuality. In M. Kuna and N. Venclová (eds.), Whither archaeology? Papers in honour of Evzen Neustupny, 230–45. Prague: Institute of Archaeology, Czech Academy of Sciences. With M. Ruth Megaw. Paper tigers, tilting at windmills and Celtic Cheshire cats: a reply to Tim Taylor, Scottish Archaeol. Review 9–10, 248–52.

1997 With M. Ruth Megaw. Triangular impressed clay fragment (lost). In E. Hostetter and T. N. Howe (ed.) The RomanoBritish villa at Castle Copse, Great Bedwyn, 260–6. Bloomington, Indiana: Indiana University Press. With M. Ruth Megaw and J.-W. Neugebauer. Notes on a silver pendant from Pottenbrunn-Ratzersdorf, Niederösterreich and its wider European connections, Germania 75:2, 717–35. Stuart Piggott, Anthropology Newsletter (Sept.), 43. Visualising archaeology: Has the past a future? Professorial inaugural address, Australian Archaeol. 44 (June), 39–51. Sándor (Alexander) Gallus, Australian Archaeol. 44, 56–57. With M. Ruth Megaw. Ngarrindjeri warrior: Kerry Giles (‘Kurwingi’) 1959–1997, Art Monthly Australia 103 (September), 33–34. Ngarrindjeri soldier: Kerry Giles (‘Kurwingie’) 1959–97, Artlink 17:3, 81–82.

1996 ‘Comments: Angela Piccini,’ Filming through the mists of time: Celtic constructions and the documentary, Current Anthropology 37: Supplement, 102–4. With M. Ruth Megaw. Ancient Celts and modern ethnicity, Antiquity 70, 175–81. With M. Ruth Megaw. From the Tyrol to Texas: Lost and found in Middle La Tène. In Marc Lodewijckx (ed.) Archaeological and historical aspects of West-European societies: Album amicorum André Van Doorselaer, 59–66. Acta Archaeologica Lovaniensia Monographiae 8. With Doreen Mellor. The land made visible – ‘Native titled now’, Artink 16:4 (Summer), 6–8. From boards to the boardroom: the most ancient new art in the world. In Christine Adrian Dyer (ed.) Icons of the Western Desert: the John W. Kluge Papunya Tula twentieth anniversary commission (Museum Art International Pty Ltd, Adelaide, 1996), 31–37. ‘Aboriginal Australia: contemporary art’ v.1, 62–66 and, with M. Ruth Megaw, ‘Celtic Art’ v.6, 152–64; ‘Dürrnberg’ v.9, 453; ‘La Tène’ v.18, 828 and ‘Waldalgesheim’ v.32,

1998 With M. Ruth Megaw. The stone head from Msecké Zehrovice, okr. Rakovník: an essay on the human head in early Celtic art. In N. Venclová (ed.) Msecké Zehrovice in Bohemia: Archaeological background to a Celtic hero 3rd – 2nd cent. BC; chronothèque 2, 281–92. Sceaux: Kronos Editions. With M. Ruth Megaw. Cartoons, crocodiles and Celtic art: Images from a scholar’s notebooks, Oxford J. of Archaeol. 17:1, 121–6. With M. Ruth Megaw. ‘The mechanism of (Celtic) dreams?’: A partial response to our critics, Antiquity 72, 432–5. With M. Ruth Megaw. Do the Ancient Celts still live? An essay on identity and contextuality, Studia Celtica 31, 107–23. With M. Ruth Megaw. Cheshire Cats in the Tyrol. In A. Müller-Karpe et al. (eds.) Studien zur Archäologie der Kelten, Römer und Germanen in Mittel- und Westeuropa, 389–400. Rahden/Westf.: Verlag Marie Leidorf. With M. Ruth Megaw. The case of the vanishing celts (sic): or ‘what’s in a name?’ In M. Pearce and M. Tosi (eds.) Proc. European Association of Archaeologists: Third Annual Meeting, Ravenna, Italy, Sept. 1997, 290–3. BAR Intern. Ser. 717. Oxford: BAR. With M. Ruth Megaw. Early celtic art 50 years after Paul Jacobsthal, Proc. IUPPS XIII Congress: Forlì, Italy 4:12 – The Iron Age in Europe, 691–7. Forlì: ABACO Edizioni.

Emeritus Professor J. V. S. Megaw Bibliography With M. Ruth Megaw. Waldalgesheim variations [review article], Germania 76:2, 847–63. With M. Ruth Megaw. Celtic connections past and present: Ancient Celts and modern ethnicity. In R. Black, W. Gillies and R. Ó Maolalaigh (eds.) Celtic connections: Proceedings of the 10th Intern. Congr. of Celtic Studies 1, 19–81. East Linton: Tuckwell Press. Sandor Gallus (1907–1996), Acta Arch. Hung. 49, 279–83. With Marcia Langton. The prehistory of ‘prehistory’ – a further response, Australian Archaeology 48, 52–54. With Doreen Mellor (eds.). Twenty-five years and beyond: Papunta Tula painting Flinders Art Museum. Bedford Park SA: Flinders University, exhibition catalogue. With M. Ruth Megaw. Artists as performers: the Flinders University Aboriginal artists-in-residence programme re-reviewed, Art, performance and society – Proceedings of the Fifth International Symposium of the Pacific Arts Association, Adelaide, 66–73 and 303. Bathurst NSW: Crawford House Press and Honolulu: University of Hawai’i Press. With M. Ruth Megaw and Rosalind Niblett. A decorated Iron Age copper alloy knife from Hertfordshire, Antiq. J. 79, 379–87. With Thomas Stöllner and Graham Morgan. Mining the wetlands: recent montane archaeology of a medieval mining entrance at Bad Dürrnberg-bei-Hallein, Land Salzburg, Austria, NewsWARP 26, December, 12–20. With Thomas Stöllner. KG Dürrnberg, SG Hallein, VB Hallein, FÖ 37, 875–6. 2000 With Graham Morgan and Thomas Stöllner. Ancient saltmining in Austria, Antiquity 74, 17–18. With M. Ruth Megaw. Bibliography, Artlink 20:1; Reconciliation? Indigenous art for the 21st century, 106–11. Whistles. In Peter Ellis (compiled and ed.) Ludgershall Castle, Wiltshire: A report on the excavations by Peter Addyman, 1964–1972, 163–7. Wilts. Archaeol. and Nat. Hist. Soc. Monograph Ser. 2. With Rhys Jones. Confessions of a wild colonial boy, Australian Archaeology 50, 12–26. With Claus Dobiat and Thomas Stöllner et al. Siedlungsund Wirtschaftsgeschichte des Dürrnberges bei Hallein, Arch. Korrbl. 30:1, 78–81. With M. Ruth Megaw. From the South Pole to Irn Bru: the new Museum of Scotland, Scottish Archaeol. J. 22:1, 83–89. With M. Ruth Megaw. ‘Painting country: The Arrernte watercolour artists of Hermannsburg’, 197–204; ‘Ebatarinja, Cordula, 379; ‘Giles, Kerry’, 593; ‘Namatjira, Albert’, 655–6; ‘Pfitzner, Milika Darryl’, 673; ‘Nickolls, Trevor’, 663; ‘Shearer, Heather Kemarre’, 698, in Sylvia

369

Kleinert and Margo Neale (eds.) The Oxford companion to Aboriginal art and culture. Melbourne: Oxford University Press. With M. Ruth Megaw, Anne-Marie Adam, Le fibule di tipo celtico nel Trentino [review article], Germania 78:2, 502–7. A Childe of our time or is prehistory impractical? In Atholl Anderson and Tim Murray (eds.) Australian arcdhaeologist: Collected papers in honour of Jim Allen, 117–21. Centre for Archaeological Research and Department of Archaeology and Natural History, The Australian National University, Canberra with Department of Archaeology, LaTrobe University, Melbourne. 2001 With M. Ruth Megaw, Nikola Theodossiev and Nartsis Torbov. The decorated La Tène sword scabbard from Pavolche near Vratsa: some notes on the evidence for Celtic settlement in Northwestern Bulgaria, Archaeologia Bulgarica 4, 25–43. With M. Ruth Megaw. The Ancient Celts: The first European Community? (Plenary Lecture). In Geraint Evans, Bernard Martin and Jonathan Wooding (eds.) Origins and revivals: Proceedings of the First Australian Conference of Celtic Studies; Sydney Series in Celtic Studies 3, 559–97. Centre for Celtic Studies, University of Sydney. With M. Ruth Megaw. Celtic art from its beginnings to the Book of Kells. London and New York: Thames and Hudson, revised and enlarged ed. With M. Ruth Megaw. Indigenous Australians: Visual Arts. In James Jupp (ed.) The Australian people: An encyclopedia of the nation, its people and their origins, 97–101. Cambridge: Cambridge University Press. With M. Ruth Megaw. Expressiveness and communication: Insular Celtic art through six centuries [review article], Antiquity 75, 433–6. With M. Ruth Megaw. A cat-and-mouse game: tracing Iron Age elements in the Book of Kells. In Ernst Pohl, Udo Recker and Claudia Theune (eds.) Archäologisches Zellwerk: Beiträge zur Kulturgeschichte in Europa und Asien. Festschrift für Helmut Roth zum 60. Geburtstag, 209–53. Rahden/Westf.: Verlag Marie Leidorf. ‘Your obediant and humble servant’: notes for an Antipodean antiquary. In Atholl Anderson, Ian Lilley and Sue O’Connor (eds.) Histories of Old Ages. Essays in honour of Rhys Jones, 95–110. Canberra: Pandanus Books – Research School of Pacific and Asian Studies, The Australian National University. ‘Australian Aboriginal Cultures Gallery ‘The Speaking Land’: A review article’, Rec. South Australian Museum 34:2, 115–2.

370

Emeritus Professor J. V. S. Megaw Bibliography

2002 With M. Ruth Megaw. D’anciens Bretons à l’étranger? Notes sur deux bronzes émaillés du Ier siècle après J.-C., Revue du Nord 83:343, 51–58. With M. Ruth Megaw. Early Celtic art in the British Isles. By E. M. Jope [review article], Antiq. Jnl. 82, 355–7. Figuration and abstraction in early Celtic art: who is smiling now?. In Zuzana Karasová and Milan Licka (eds.) Figuration et abstraction dans l’art de l’Europe ancienne (VIIIème – Ier s. av. J.-C.; Sbornîk Národního Muzea v Praze A 56:1–4, 57–64. With M. Ruth Megaw. The Celts. In Tim Murray (ed.) Encyclopedia of archaeology: History and discoveries 1, 285–95. Santa Barbara, CA, Denver, Oxford: ABC Clio. [Emeritus Professor Rhys Jones: obituary] ‘Further reminisces of a back-looking curiosity’, Antiquity 76, 12–14. [items 28, 218 and 233 reprinted with commentary in:] Gillian Carr and Simon Stoddart (eds.) Celts from Antiquity; Antiquity Papers 2 (Antiquity Publications, Cambridge), 19–26, 47–50 and 83–90. With M. Ruth Megaw. Persians bearing gifts? A footnote on an old find. In R. Gicheva and K. Rabadiev (eds.) Pituh Studia in honorem Prof. Ivani Marazov, 486–492. Sofia: Anubis. Fresh sardines and stone knights [conference report], Antiquity 76, 304–5. Transformations: Appreciation, appropriation and imagery in Indigenous Australian art. In Anita Herle, Nick Stanley, Karen Stevenson and Robert L. Welsch (eds.) Pacific Art: Persistence, Change and Meaning, 370–38. Adelaide: Crawford House Publishing. With M. Ruth Megaw. Rudolf Echt, Das Fürstinnengrab von Reinheim [review article], Germania 80, 336–44. A late anthropoid griped sword in New York. In K. Kuzmová, K. Pieta and J. Rajtár (eds.) Zwischen Rom und dem Barbaricum: Festschrift fur Titus Kolník zum 70. Geburtstag, 407–18. Nitra: Archaeologica Slovaca monographiae; Communicationes 5. From Hungary to the Hunsrück-Eifel? The bronze buttons from Sitzerath, Kr. St.Wendel, Arch. Korrespondenzbl. 33, 229–37. 2003 Where have all the warriors gone? Some aspects of stone sculpture from Britain to Bohemia’, Madrider Mitt. 44, 269–86 and Taf.62–65 With M. Ruth Megaw. From Dürer to Dolly Daniels: A partial history of the Flinders University Art Museum. In Ian McLean and Ruth and Vincent Megaw, Big Country: works from the Flinders University Art Museum, 32–54. Flinders University Art Museum, Adelaide, exhibition catalogue.

Rhys’ Day, Australian Archaeology 56, 59. Celtic foot(less) soldiers? An iconographic note, Gladius 23, 61–70. The Dürrnberg: Centre for craft skills or importer of luxury goods? In T. Stöllner et al., ‘The economy of Dürrnbergbei-Hallein: An Iron Age salt-mining centre in the Austrian Alps’, Antiq. J. 83, 177–9. With M. Ruth Megaw. The bronze mount from Drouzkovice, Northwest Bohemia, Památky archeologické 93, 173–94 2004 In the footsteps of Brennus? Further archaeological evidence for Celts in the Balkans. In B. Hänsel and Etela Studenikova (eds.) Zwischen Karpaten und Ägäis: Neolithikum und ältere Bronzezeit. Gedenkschrift für Viera Nemejcova-Pavukova, 93–107. Rahden/Westf.: International Archäologie: Studia honoraria 27, Verlag Marie Leidorf. ‘Tales of a Flying Dutchman’: An exaugural lecture, Australian Archaeology 58, 25–36. ‘Figuring it out . . . By Colin Renfrew’ [review article], Scottish Archaeol. J. 25:1, 83–101. With Ian Ralston. Beyond barbarian Europe: Stuart Piggott 1910–96: an appreciation. In I. A. G. Shepherd and G. J. Barclay (eds.) Scotland in Ancient Europe: The Neolithic and Early Bronze Age of Scotland in their European context, 16–19, 23. Edinburgh: Society of Antiquaries of Scotland. Banks and braes: notes for a film-buff Fellow. In Alex Gibson and Alison Sheridan (eds.) From sickles to circles: Britain and Ireland at the time of Stonehenge, 14–26. Stroud: Tempus. ‘There are kangaroos in Austria!’ Forty years of research at the salt-mining complex of Dürrnberg-bei-Hallein, Ld. Salzburg. In T. Murray (ed.) Archaeology from Australia, 387–401. Melbourne: Australian Scholarly Publishing. Treasures of the Nicholson Museum, the University of Sydney: [review article], Australian Archaeology 59, 72–73. 2005 With M. Ruth Megaw. Early Celtic Art in Britain and Ireland. Princes Risborough: Shire Publications, second enlarged and fully revised edition. The Dürrnberg-bei-Hallein: Life and death in an Iron Age salt-mining community, Sprawozdania z posiedzen; Komisji Naukowych (Polska Akademia Nauk, Oddzia w Krakowie 45:2, 1–4. Obituary: Emeritus Professor Derek Douglas Alexander Simpson, Proc. Soc. Antiq. Scot 134, 3–6. Celts in Thrace? A reappraisal. In Jan Bouzek and Lidia Domaradzka (eds.) The culture of Thracians and their neighbours, 209–14. Oxford: BAR IS 1350.

Emeritus Professor J. V. S. Megaw Bibliography With M. Ruth Megaw (trans. Patrick Galliou). Art de la Celtique: VIIe siècle av. J.-C – VIIIe siècle apr. J.-C. Paris: Editions Errance. The European Iron age with – and without – the Celts: a biographical essay, European J. of Archaeol. 8:1, 65–78. With Halina Dobrzanska and Paulina Poleska (eds.). Celts on the margin: Studies in European cultural interaction 7th century BC – 1st century AD dedicated to Zenon Wozniak. Kraków: Institute of Archaeology and Ethnology, Polish Academy of Sciences. Early Celtic art without Scythians? A review. In Halina Dobrzanska, Vincent Megaw and Paulina Poleska (eds.) Celts on the margin: Studies in European cultural interaction 7th century BC – 1st century AD dedicated to Zenon Wozniak, 33–47. Kraków: Institute of Archaeology and Ethnology, Polish Academy of Sciences. 2006 Notes on two belt-plates of early La Tène type from northern Poland, Pomorania Antiqua 20, 257–76. With M. Ruth Megaw. ‘art, Celtic [1] pre-Roman’ v.1, 89–97; ‘art, Celtic [2] post-Roman’, 98–106; ‘BasseYutz’, 184–86; v.3, ‘La Tène [2] the La Tène period’, 1071–6, in John T. Koch (ed.) Celtic culture: A historical encyclopedia. Santa Barbara CA/ Oxford: ABC-Clio. With M. Ruth Megaw. ‘Strike the lyre: some notes on an eastern Celtic motif’, Acta Arch. Hung. 57, 367–93. Liselotte Zimmer-Plank (Hrsg.), Kult der Vorzeit in den Alpen [review article], Germania 83, 413–8. 2007 With Ruth Megaw, Peter C. Ramsl and Birgit Bühler. From Austria to Arras: The gold armlets from Grave 115, Mannersdorf a.d. Leitha. In C. Gosden, H. Hamerow, P. De Jersey and G. Lock (eds.) Communities and connections: Essays in honour of Barry Cunliffe, 183– 216. Oxford: Oxford University Press. With Gilbert Burleigh. The Iron Age mirror burial at Pegsdon, Shillington, Bedfordshire: An interim account, Antiq. J. 87, 109–40. Imag(in)ing the Celts, Antiquity 81, 438–45. With M. Ruth Megaw. 4. Ancient Celts and modern ethnicity. In Raimund Karl and David Stifter (eds.) The Celtic world: critical concepts in historical studies I, 59–70. London and New York: Routledge. With M. Ruth Megaw. 6. Do the ancient Celts still live? An essay on identity and contextuality. In Raimund Karl and David Stifter (eds.) The Celtic world: critical concepts in historical studies I, 81–101. London and New York: Routledge. 2008 With Ruth Megaw. Celtic lyres on a Celtic kylix? A further

371

note on the copy of an Attic red figure two-handled cup from Plzen-Roudna, Archeologické rozhledy 59, 799–804. With M. Ruth Megaw. A Celtic mystery: some thoughts on the genesis of insular Celtic art. In Duncan Garrow, Chris Gosden and J. D. Hill (eds.) Rethinking Celtic art, 40–58. Oxford: Oxbow Books. Vincent’s eye: 25 years of collecting for Flinders University Art Museum. In Identity and voices at Flinders University Art Museum: Two exhibitions celebrating Flinders University’s 40th Anniversary, 46–55. Adelaide: Flinders University (exhb. cat.). With M. Ruth Megaw. The antler implement from Sobesuky, feature no. 3472/91. An essay on early La Tène figural art in Bohemia, Archeologické rozhledy 60, 529–46. With Ruth Megaw. The early La Tène antler implerment from Sobesuky, okr. Chomutov, Czech Republic, Instrumentum 28, 12–13. With András Márton and Ruth Megaw. Decorated loomweights in the collection of Classical antiquities of Museum of Fine Arts, Budapest, Hungary. Archaeological Textiles Newsletter 47, 7–11. Engaging with Yolngu art: Howard Morphy, Becoming art: exploring cross-cultural categories, Art Monthly Australia 213, Sept. 2008, 42–43 (review article). 2009 With M. Ruth Megaw. Hare or hind? The decorated spear from Kandija grave K44. In George Tiefengraber et al. (eds.) Keltske studije II: Studies in Celtic archaeology. Papers in honour of Mitja Gustin: Protohistoire européenne 11, 163–72. With Birgit Bühler, M. Ruth Megaw and Peter C. Ramsl. Grab 115 des latènezeitlichen Gräberfeldes von Mannersdorf am Leithagebirge (Niederösterreich); Typologische, technologische und stilistische Studien zu den beiden Goldarmreifen, Germania 86, 103–34. With Jacqueline Nowakowska, Adam Gwilt and Susan La Niece. A late Iron Age neck-ring from Pentire, with a note on the find from Boverton Vale of Glamorgan, Antiq. Jnl. 80, 35–52. Blecic, M. et al. (eds.) Sripta praehistotica in honorem Biba Terzan, 2007. Arheoloski vestnik 60, 345–8 (review article). With M. Ruth Megaw. A style without genesis? The elusive birthplace of early Celtic art. In Gabriel Cooney et al. (eds.) Relics of old decency: archaeological studies in later prehistory, 291–306. Dublin: Wordwell. With Manuel Zeiler et al. Stempelgleiche FrühlatèneKeramik zwischen Traisental und Neusiedlersee. In Gabriel Cooney et al. (eds.) Relics of old decency: archaeological studies in later prehistory, 259–76. Dublin: Wordwell.

372

Emeritus Professor J. V. S. Megaw Bibliography

2010 Bearing the truth about Celtic art: Kunst der Kelten in Bern, Antiquity 84, 250–55. With M. Ruth Megaw. East and West in early Celtic art. The first stages once more reviewed. In Erzsébet Jerem, Martin Schönfelder and Günther Wieland (eds.) Nord-Süd, Ost-West: Kontakte während der Eisenzeit in Europa, 183–205. Budapest: Archaeolingua main series 17. With M. Ruth Megaw. The stamped sherd from Kanín (Central Bohemia): a further essay on early La Tène art style, Arch. roz. 62, 313–27. With M. Ruth Megaw. A world turned upsidedown. The bronze plaque from Stupava, okr. Malacky. In Sándor Berecki (ed.) Iron Age communities in the Carpathian Basin. Proceedings of the International Colloquium from Targu Mures 9–11 October 2009, 115–126. Bibliotheca Mvsei Marisiensis Seria Archaeologoca 2. 2011 With M. Ruth Megaw. The elusive arts: the study of Continental early Celtic art since 1944. In Gocha R. Tsetskhladze (ed.) The Black Sea, Greece, Anatolia and Europe in the first millennium BC. Coloquia Antiqua 1, 265–315. Leuven/Paris/Walpole MA. With R. Echt, Michael Matx, Wolf-Rüdiger Thiele, Luc van Impe and Leo Verhart. An Iron Age gold torc from Heerlen (prov. Limburg/NL), Arch. Korrespondenzblatt 41:1, 31–49. A world turned upside down: the bronze plaque from Stupava, okr. Malacky. In P. Kalábková, B. Kovár, P. Pavúk and J. Šuteková (eds.) PANTA RHEI. Studies on the Chronology and Cultural Development of South-Eastern and Central Europe in Earlier Prehistory Presented to Juraj Pavúk on the Occasion of his 75th Birthday. Studia Archaeologica et Mediaevalia 11, 607–622. Bratislava. With M. Ruth Megaw. The Kanín stamped sherd: a further note, Arch. roz. 63, 162–163. With Gilbert Burleigh. An Iron age mirror from Ruxox, Maulden, Bedfordshire, Antiq. J. 91, 51–58. Western Desert painting is not traditional. In Ian McLean (ed.) How Aborigines invented the idea of contemporary art, 98. Brisbane and Sydney: Institute of Modern Art and Power Publications. With Aurel Rustoiu. A foreign flowering in Transylvania: the vegetal style armring from Fântânele-Dealul Popii,

jud. Bistriţa-Năsăud, grave 62. In Despina Măgureanu et al. (eds.)Archaeology: making of and practice. Studies on honour of Mircea Babeş at his 70th anniversary, 217–237. Piteşti. 2012 The Flying Dutchman reaches port, Antiquity 86, 546–557. Foreword. In Roads cross: contemporary directions in Australian art, 7. Adelaide: Flinders University Art Museum (exh. cat.). ‘Go east young man!’ Antipodean thoughts on the earliest La Tène art in Slovakia (with particular reference to the fortified settlement of Horné Orešany). In Getrúda Brežinová and Vladamír Varsik (eds.) Archeológia na prahu historie. K žovotnému jubileu Karola Pietu. Archaeologica Slovaca Monographiae, 447–458. Nitra: Communicationes 15. With M. Ruth Megaw. Poles apart? Notes from the fringes of the ‘Hungarian’ Sword Style. In P. Anreiter et al. (eds.) Archaeological, cultural and linguistic heritage. Festschrift for Erzsébet Jerem in honour of her 70th birthday, 400–414. Budapest: Archaeolingua 25. With Emilov, J. Celts in Thrace? A re-examination of the tomb of Mal Tepe, Mezek with particular reference to the La Tène chariot fittings, Archaeologia Bulgarica 16:1, 1–32. ‘Dreidiemensional: Der Plastische Stil’, ‘Siegelwachsrot: Email am Gürtel’, ‘Die grinsende Katze’: Der Cheshire Style’ and ‘Mickey Maus im Grab: Mezek und der Disney-Stil’. In Ralph Rober et al. (eds.) Die Welt der Kelten. Zentren der Macht – Kostbarkeiten der Kunst, 304–307; 310–311; 314–315; 316–317. Stuttgart: Jan Thorbecke Verlag (exh. cat.). 2013 With Jordan Anastassov, Elina Mircheva and Ruth Megaw. Walt Disney comes to Bulgaria: a bronze mount in the Museum of Archaeology, Varna. In A. Colin et al. (eds.) L’âge du Fer en Europe. Mélanges offerts à Olivier Buchsenschutz, 551–565. Ausonius: Bordeaux. A Celtic cornucopia: review article. Antiquity 87.335, 280–284. Forthcoming With M. Ruth Megaw. Early Celtic Art. A Supplement. Oxford: Oxford University Press.

Colour Plates

Plate 1: Frontispiece, page ii.

Plate 2: Figure 3.1, page 22.

Plate 3: Figure 4.2a, page 31.

Plate 4: Figure 3.4, page 24.

Plate 5: Figure 4.7, page 35.

Plate 6: Figure 4.7, page 35.

Plate 7: Figure 5.7, page 45.

Plate 8: Figure 5.11b, page 47.

Plate 9: Figure 5.12. page 48.

Figure 10: 5.13. page 48.

Plate 11: Figure 6.1, page 57.

Plate 12: Figure 6.2, page 59.

Plate 13: Figure 6.3, page 60.

Plate 14: Figure 6.4, page 61.

Plate 15: Figure 6.5, page 62.

Plate 16: Figure 6.6, page 63.

Plate 17: Figure 7.2, page 70.

Plate 18: Figure 7.4, page 71.

Plate 19: Figure 8.6, page 83.

Plate 20: Figure 9.7, page 90.

Plate 21: Figure 12.1, page 108.

Plate 22: Figure 13.1, page 114.

Plate 23: Figure 14.2, page 123.

Plate 24: Figure 15.1, page 137.

Plate 25: Figure 17.1, page 149.

Plate 26: Figure 17.2, page 149.

Plate 27: Figure 18.1, page 153.

Plate 28: Figure 18.4, page 155.

Plate 29: Figure 19.1, page 160.

Plate 30: Figure 21.2, page 180.

Plate 31: Figure 23.4, page 198.

Plate 32: Figure 23.7, page 199.

Plate 33: Figure 24.1, page 207.

Plate 34: Figure 25.1, page 213.

Plate 35: Figure 26.1, page 225.

Plate 36: Figure 26.2a, page 225.

Plate 37: Figure 26.2b, page 225.

Plate 38: Figure 36.3, page 225.

Plate 39: Figure 36.4, page 225.

Plate 40: Figure 27.2, page 235.

Plate 41: Figure 27.4, page 236.

Plate 42: Figure 30.1, page 276.

Plate 43: Figure 30.2, page 276.

Plate 44: Figure 30.3, page 277.

Plate 45: Figure 30.12, page 282.

Plate 46: Figure 30.13, page 282.

Plate 47: Figure 31.4a, page 290.

Plate 48: Figure 31.4b, page 290.

Plate 49: Figure 31.6b, page 291.

Plate 50: Figure 31.7, page 292.

Plate 51: Figure 32.2, page 299.

Plate 52: Figure 34.1, page 318.

Plate 53: Figure 34.2, page 319.

Plate 54: Figure 34.3, page 319.

Plate 55: Figure 34.4, page 320.

Plate 56: Figure 34.5, page 320.

Plate 57: Figure 34.6, page 321.

Plate 58: Figure 34.7, page 321.

Plate 59: Figure 35.7, page 329.

Plate 60: Figure 35.11, page 335.