Late Roman Precious Metal Deposits c. AD 200-700: Changes over time and space 9781407329574, 9781841719344

Table of contents :
Front Cover
Title Page
Copyright
Table of Contents
List of Tables
List of Figures
Preface and Acknowledgements
Chapter 1: Aims: The Study of Late Roman Precious Metal Deposits
Chapter 2: Methodology: The Construction of the Database
Chapter 3: Survey of Material
Chapter 4: Long Range Patterns with Reference to the Whole Database
Chapter 5: Interpreting Patterns of Precious Metal Deposition
References
Appendix
Precious Metal Deposits, c. AD 200-700
Index of Deposits

Citation preview

BAR S1504 2006 HOBBS LATE ROMAN PRECIOUS METAL DEPOSITS, c. AD 200-700

B A R

Late Roman Precious Metal Deposits c. AD 200-700 Changes over time and space

Richard Hobbs

BAR International Series 1504 2006

Late Roman Precious Metal Deposits C. AD 200-700 Changes over time and space

Richard Hobbs

BAR International Series 1504 2006

Published in 2016 by BAR Publishing, Oxford BAR International Series 1504 Late Roman Precious Metal Deposits c. AD 200-700 © R Hobbs and the Publisher 2006 The author's moral rights under the 1988 UK Copyright, Designs and Patents Act are hereby expressly asserted. All rights reserved. No part of this work may be copied, reproduced, stored, sold, distributed, scanned, saved in any form of digital format or transmitted in any form digitally, without the written permission of the Publisher.

ISBN 9781841719344 paperback ISBN 9781407329574 e-format DOI https://doi.org/10.30861/9781841719344 A catalogue record for this book is available from the British Library BAR Publishing is the trading name of British Archaeological Reports (Oxford) Ltd. British Archaeological Reports was first incorporated in 1974 to publish the BAR Series, International and British. In 1992 Hadrian Books Ltd became part of the BAR group. This volume was originally published by Archaeopress in conjunction with British Archaeological Reports (Oxford) Ltd / Hadrian Books Ltd, the Series principal publisher, in 2006. This present volume is published by BAR Publishing, 2016.

BAR PUBLISHING BAR titles are available from:

E MAIL P HONE F AX

BAR Publishing 122 Banbury Rd, Oxford, OX2 7BP, UK [email protected] +44 (0)1865 310431 +44 (0)1865 316916 www.barpublishing.com

TABLE OF CONTENTS • List ofTables ......................................................................................................................... • List ofFigures ........................................................................................................................ Preface and Acknowledgments ..............................................................................................................

vi vii ix

Chapter 1 Aims: The study of late Roman precious metal deposits ............................................................................

1

The background to the research ............................................................................................................. Past research into the study of precious metal deposits .................................................................................. The approach of this study ..................................................................................................................

1 2 .4

Chapter 2 Methodology: the construction of the database .............................................................................................

6

Why only precious metal deposits? ......................................................................................................................................... What constitutes a precious metal deposit? ............................................................................................................................. Sources of data ............................................................................................................................... Coin deposits ................................................................................................................................. Non-coin artefacts ............................................................................................................................ Types of deposit ............................................................................................................................. Dates of discovery ........................................................................................................................... The database of deposits (Appendix) .................................................................................................... Chronology: the deposition periods ........................................................................................................ Spatial considerations ....................................................................................................................... The value of deposits:comparing deposits in termsof size .. ......................................................................... Date range .................................................................................................................................... Summary ......................................................................................................................................

6 6 7 8 9 10 10 11 11 15 17 24 24

Chapter 3 Survey of material. .........................................................................................................................

26

Introduction ...................................................................................................... : .......................... Period 1 ( 193-221): Discussion and analysis ........................................................................................... Period 2 (222-37): Discussion and analysis ............................................................................................. Period 3 (238-59): Discussion and analysis ............................................................................................. Period 4 (260-74): Discussion and analysis ............................................................................................. Period 5 (275-95): Discussion and analysis ............................................................................................. Period 6 (296-3 I 7): Discussion and analysis .......................................................................................... Period 7 (318-29): Discussion and analysis ............................................................................................ Period 8 (330-47): Discussion and analysis ............................................................................................ Period 9 (348-63): Discussion and analysis ............................................................................................ Period 10 (364-94): Discussion and analysis ............................................................................................ Period 11 (395-410): Discussion and analysis .......................................................................................... Period 12 (411-24): Discussion and analysis ............................................................................................ Period 13 (425-56): Discussion and analysis ............................................................................................ Period 14 (457-90): Discussion and analysis ............................................................................................ Period 15 (491-526): Discussion and analysis .......................................................................................... Period 16 (527-64): Discussion and analysis ............................................................................................ Period 17 (565-81 ): Discussion and analysis ............................................................................................ Period 18 (582-609): Discussion and analysis .......................................................................................... Period 19 (610-40): Discussion and analysis ............................................................................................ Period 20 (641-67): Discussion and analysis ............................................................................................ Period 21 ( 668-84): Discussion and analysis ....... : .................................................................................... Period 22 ( 685- 711): Discussion and analysis ................................................................................ , ......... Broader changes in distribution of precious metal artefact types and the integration of poorly dated deposits .................................................................................................................... Broadly the third century: AD 193-317 ............................................................................................. Broadly the fourth century: AD 318-410 ............................................................................................

IV

26 26 34 35 36 38 .40 .46 .48 .49 51 53 59 64 66 67 72 73 74 75 81 82 84 85 85 92

Broadly the fifth century: AD 411-526 ........ , ..................................................................................... Broadly the sixth century: AD 527-609 ............................................................................................. Broadly the seventh century: AD 610-711 ..........................................................................................

Chapter 4 Long range patterns with reference to the whole database .......................................................................

94 95 96

99

Total number of finds and total EGW .................................................................................................... Regional differences in the size and nature of deposits ............................................................................... Regional variations in levels of deposition over time ................................................................................. The overall rate of deposition per year for all regions ................................................................................. The west (regions 1-5) ......................................................................................................................... The east (regions 6-10) .................................................................. : ................................................ Outside the frontiers (regions 11-15) ................................................................................................... Conclusion .................................................................................................................................. Summary ...................................................................................................................................

99 100 106 108 109 112 115 119 119

Chapter 5 Interpreting patterns of precious metal deposition ...............................................................................

120

A hoard in context: Patching, West Sussex, 1997 .................................................................................... Interpreting the wider patterns ............................................................................................................... The interpretation of individual finds .................................................................................................... The interpretation of chronologically and geographically distinct groups of finds .............................................. The pattern of hoarding 'hotspots' is nothing to do with ancient patterns .................................................... The patterns observed can be directly equated with changing levels of wealth and prosperity ........................... The patterns should simply be correlated with threat and response ................................................................ Levels of hoarding simply relate to changes in the availability and quality of the coinage ................................. Deposition is related to none of the above: rather it is related to social behaviour .......................................... A unified theory of hoarding? ..............................................................................................................................................

120 121

References .................................................................................................................................

. 135

Appendix Precious metal deposits, c. AD 200-700 ..............................................................................................

152

Index of deposits .........................................................................................................................

275

V

121

124 125 127 128

130 132 134

LIST OF TABLES 1. Summary of different types of deposit included in this study ....................................................................... 2. Deposits discovered in the sixteenth and seventeenth centuries ................................................................... 3. Dating phases employed in this study ................................................................................................. 4. Regions included in the study area .................................................................................................... 5. Mean weights of the aureus and the denarius in the early Empire ................................................................ 6. Relative values of gold and silver between the early Empire and the early fifth century AD ................................. 7. Metrology of the Roman and early Byzantine gold and silver coinage .......................................................... 8. Analyses of silver plate items from the third to seventh centuries AD ........................................................... 9. The ten largest deposits of period 1 (193-221) ...................................................................................... 10. Date range of deposits of period 1 (193-221), by region ......................................................................... 11. The ten largest deposits of period 2 (222-37) ...................................................................................... 12. Date range of deposits of period 2 (222-37), by region ........................................................................... 13. The ten largest deposits of period 3 (238-59) ...................................................................................... 14. Date range of deposits of period 3 (238-59), by region ........................................................................... 15. The twenty largest deposits of period 4 (260-274) ................................................................................. 16. Date range of deposits of period 4 (260-74), by region ........................................................................... 17. The ten largest deposits of period 5 (275-95) ....................................................................................... 18. Date range of deposits of period 5 (275-95), by region .......................................................................... 19. The ten largest deposits of period 6 (296-317) .................................................................................... 20. Date range of deposits of period 6 (296-317), by region ........................................................................ 21. Wcs tern deposits containing antoniniani and nummi of period 7 ............................................................... 22. The six largest deposits of period 7 (318-29) ..................................................................................... 23. Date range of deposits of period 7 (318-29), by region .......................................................................... 24. The six largest deposits of period 8 (330-47) ...................................................................................... 25. Date range of deposits of period 8 (330-47), by region .......................................................................... 26. The ten largest deposits of period 9 (348-63) ....................................................................................... 27. Date range of deposits of period 9 (348-63), by region ........................................................................... 28. The ten largest deposits of period 10 (364-94) ..................................................................................... 29. Date range of deposits of period 10 (364-94), by region ......................................................................... 30. The ten largest deposits of period 11 (395-410) .................................................................................... 31. British hoards of the period 395-410 and Equivalent Gold Weights ............................................................ 32. Categories of precious metal artefact groups per region in Britannia ........................................................... 33. Date range of deposits of period 11 (395-410), by region ........................................................................ 34. The ten largest deposits of period 12 (411-25) ..................................................................................... 35. Date range of deposits of period 12 (411-25), by region .......................................................................... 36. The ten largest deposits of period 13 (425-56) ......................................................................... : ........... 37. Date range of deposits of period 13 (425-56), by region .......................................................................... 38. The ten largest deposits of period 3 (457-90) ...................................................................................... 39. Date range of deposits of period 14 (457-90), by region .......................................................................... 40. The ten largest deposits of period 15 (491-526) ............................................... ·..................................... 41. Date range of deposits of period 15 (491-526), by region ........................................................................ 42. The ten largest deposits of period 16 (527-64) ..................................................................................... 43. Date range of deposits of period 16 (527-64), by region ......................................................................... 44. The ten largest deposits of period 17 (565-81 ) ..................................................................................... 45. Date range of deposits of period 17 (565-81), by region ......................................................................... 46. The ten largest deposits of period 18 (582-609) ................................................................................... 47. Date range of deposits of period 18 (582-609), by region ........................................................................ 48. The ten largest deposits of period 19 (610-40) .................................................................................... 49. Date range of deposits of period 19 (610-40), by region ......................................................................... 50. The ten largest deposits of period 29 (641-67) .................................................................................... 51. Date range of deposits of period 20 (641-67), by region ......................................................................... 52. The ten largest deposits of period 21 (668-84) ..................................................................................... 53. Date range of deposits ofperiod 21 (668-84), byregion ......................................................................... 54. All the deposits from period 22 (685-711) ........................................................................................... 55. Date range of deposits of period 8 (685-711), by region ......................................................................... 56. Total quantities of third century silver plate ........................................................................................ 57. Total quantity of third century gold and silver jewellery ......................................................................... 58. Total quantity of third century gold and silver coin, bullion and Hacksilber ............. .....................................

Vl

8 10 13 17 19 20 22 25 33 33 34 35 35 36 37 38 39 .40 .46 .46 47 .47 .47 48 .48 50 51 52 53 55 56 57 59 64 64 65 65 66 67 71 71 72 73 74 74 75 75 80 81 82 82 83 83 84 84 91 91 92

59. Total quantities of fourth century silver plate ....................................................................................... 60. Total quantity of fourth century gold and silver jewellery ........................................................................ 61. Total quantity of fourth century gold and silver coin, bullion and Hacksilber ................................................ 62. Total quantities of fifth century silver plate ........................................................................................ 63. Total quantity of fifth century gold and silver jewellery .......................................................................... 64. Total quantity of fifth century gold and silver coin, bullion and Hacksilber ............. ...................................... 65. Total quantities of sixth century silver plate ........................................................................................ 66. Total quantity of sixth century gold and silver jewellery ......................................................................... 67. Total quantity of sixth century gold and silver coin, bullion and Hacksilber .................................................. 68. Total quantities of seventh century silver plate ..................................................................................... 69. Total quantity of seventh century gold and silver jewellery ..................................................................... 70. Total quantity of seventh century gold and silver coin, bullion and Hacksilber ............................................. 71. Total number of finds and total EGW by region .........................•......................................................... T2. Total number of deposits, in rank order by region, with Equivalent Gold Weights of over 250g ......................... 73. Deposits with an Equivalent Gold Weight of more than 250g ............................ ; .................................... 74. The Cunetio hoard (period 4 (260-74): an assessment of its Equivalent Gold Weight. ..................................... 75. Total number of deposits included in this study .................................................................................. 76. Total Equivalent Gold Weight (EGW) calculation for all deposits in this study ............................................. 77. Debasement of silver coinage in comparison to overall deposition levels .................................................... 78. Hoards containing gold coin .........................................................................................................

92 93 93 94 95 95 95 96 96 97 98 .. 98 99 I 00 .101 105 106 107 131 132

LIST OF FIGURES 1. Dates of discovery of the deposits included in this study ........................................................................... 2. Example database entry (Appendix) ................................................................................................... 3. The study area ............................................................................................................................ 4. Spatial distribution of deposits of period 1 (193-221) ............................................................................... 5. Spatial distribution of deposits of period 2 (222-37) ................................................................................ 6. Spatial distribution of deposits of period 3 (238-59) ................................................................................ 7. Spatial distribution of deposits of period 4 (260-74) ................................................................................ 8. Spatial distribution of deposits of period 5 (275-95) ............................................... , ................................ 9. Spatial distribution of deposits of period 6 (296-317) ............................................................................... 10. Spatial distribution of deposits of period 7 (318-29) .............................................................................. l l. Spatial distribution of deposits of period 8 (330-47) .............................................................................. 12. Spatial distribution of deposits of period 9 (348-63) .............................................................................. 13. Spatial distribution of deposits of period 10 (364-94) ............................................................................. 14. Spatial distribution of deposits of period 1 (395-410) ............................................................................ 15. British deposits of period 11 (395-410) .............................................................................................. 16. Region 1 (Britannia) deposits of Period 11 (395-411) by total EGW ........................................................... 17. Spatial distribution of deposits of period 12 (411-24) ............................................................................. 18. Spatial distribution of deposits of period 13 (425-56) ............................................................................. 19. Spatial distribution of deposits of period 14 (457-90) ............................................................................. 20. Spatial distribution of deposits of period 15 (491-526) ............................................................................ 21. Spatial distribution of deposits of period 16 (527-64) ............................................................................. 22. Spatial distribution of deposits of period 17 (565-81) ............................................................................. 23. Spatial distribution of deposits of period 18 (582-609) ............................................................................ 24. Spatial distribution of deposits of period 19 (610-40) ............................................................................. 25. Spatial distribution of deposits of period 20 (641-67) ............................................................................. 26. Spatial distribution of deposits of period 21 (668-84) ............................................................................. 27. Spatial distribution of selected deposits of periods 1-6 (193-317) ............................................................... 28. Spatial distribution of selected deposits of periods 7-11 (318-410) ............................................................. 29. Spatial distribution of selected deposits of periods 12-15 (411-526) ............................................................ 30. Spatial distribution of selected deposits of periods 16-18 (527-609) ............................................................ 31. Spatial distribution of selected deposits of periods 19-22 (610-711) ............................................................ 32. Deposition rate per period, all regions ................................................................................................ 33. EGW rate per period, all regions .................................................................................................... 34. Deposition rate per period, region 1 (Britannia) ................................................................................... 35. EGW rate per period, region 1 (Britannia) ........................................................................................ 36. Deposition rate per period, region 2 (Gallia) ...................................................................................... 37. EGW rate per period, region 2 (Gallia) ............................................................................................

vu

11 11 16 27 28 29 30 31 32 .41 .42 .43 44 .45 55 56 60 61 62 63 68 69 70 77 78 79 86 87 88 89 90 108 ! 08 109 109 110 110

38. Deposition rate per period, region 3 (Hispania) .................................................................................. 39. EGW rate per period, region 3 (Hispania) ......................................................................................... 40. Deposition rate per period, region 4 (Africa) ..................................................................................... 41. EGW rate per period, region 4 (Africa) ............................................................................................ 42. Deposition rate per period, region 5 (Italia) ....................................................................................... 43. EGW rate per period, region 5 (Italia) .............................................................................................. 44. Deposition rate per period, region 6 (Illyricum) ..................................................................................... 45. EGW rate per period, region 6 (Illyricum) ........................................................................................ 46. Deposition rate per period, region 7 (Thrace & Dacia) ......................................................................... 47. EGW rate per period, region 7 (Thrace & Dacia) ................................................................................ 48. Deposition rate per period, region 9 (Pontica) .................................................................................... 49. EGW rate per period, region 9 (Pontica) ........................................................................................... 50. Deposition rate per period, region 10 (Oriens) .................................................................................... 51. EGW rate per period, region IO (Ori ens) .......................................................................................... 52. Deposition rate per period, region 12 (Scandinavia) ............................................................................. 53. EGW rate per period, region 12 (Scandinavia) ................................................................................... 54. Deposition rate per period, region 13 (Germania) ................................................................................ 55. EGW rate per period, region 13 (Germania) ...................................................................................... 56. Deposition rate per period, region 14 (Eastern Eurasia) ......................................................................... 57. EGW rate per period, region 14 (Eastern Eurasia) ................................................................................ 58. Deposition rate per period, region 15 (Russia) .................................................................................... 59. EGW rate per period, region 15 (Russia) ......................................................................................... 60. Diagrammatic representation of the debasement of silver coinage in relation to deposition ...............................

vm

111 111 111 112 112 112 113 113 .113 114 114 114 115 115 116 116 116 117 117 118 118 .119 131

PREFACE AND ACKNOWLEDGEMENTS

This publication is based upon my doctoral thesis, which was accepted for the degree of PhD at the University of London in 1997. Due to a range of other commitments I was unable to return to the process of publication of this research until recently. In its initial form the study included details of around 1,800 deposits, so the work required bringing it to publication included adding about 200 new deposits relevant to this research which had been published in the intervening years. I was pleased to discover however that the new data did not significantly modify the patterns I had observed in the survey of the material conducted during the 1990s. This is a vindication of my assertion that there are distinct fluctuations of hoarding levels of precious metals in the late Antique period which can be identified and require explanation. For that reason, although the interpretation section of this volume (Chapter 5) has been updated and expanded, the conclusions I drew in 1997 have not substantially altered. During my research I have had considerable support from a number of individuals. I particularly want to thank my PhD supervisor Richard Reece whom I owe a considerable debt of gratitude. In addition Martin Millett, my second supervisor, who offered helpful comments during my doctoral studies. I must also thank Peter Guest for allowing me to use a substantial amount of data which he gathered for his own doctoral research (Guest 1994); Roger Bland, Jonathan Williams, John Oma Ornstein and Richard Abdy in the department of Coins and Medals at the British Museum, for providing me with various hoard publications, some in advance of publication, and for pointing me in the right direction where necessary; Catherine Johns, my predecessor in the department of Prehistory and Europe at The British Museum, for her continuing encouragement and support. In Russia during April 1995 I conducted a study trip to St Petersburg and Moscow to gather data on late Antique material held in Russian collections, with the aid of a grant from the British Academy. I owe a great debt of gratitude to Dr. Alexander Nikitin, for his time and patience; Dr. Marschak, and Dr. Zeimal for the use of his department and library; and Dr. Jerusalem Zalleskaya for access to details of Russian hoards little known in the West and to the contents of the Hermitage Treasury. To my hosts in St. Petersburg, Dr. Praslov, Seraphima and family, and Kathy Judelson for helping arrange the trip. To Natalya Smirnova at the Pushkin Museum in Moscow, for attempting to gain me access to the Kremlin Armoury (sadly failed), and providing me with details of other Russian deposits. I would also like to thank my colleagues in the British Museum for their help and support in bringing this volume to publication, namely Stephen Crummy for overhauling the maps and figures; Ralph Jackson, J D Hill and Richard Abdy for their comments on earlier drafts; and Leslie Webster for allowing me study time to complete the necessary work. Richard Hobbs March 2006

IX

CHAPTER l AIMS: THE STUDY OF LATE ROMAN PRECIOUS METAL DEPOSITS

This monograph examines the deposition of precious metal artefacts in the late Roman and early Byzantine periods (from c. AD 200 to AD 700) within and beyond the frontiers of the Roman Empire and its successor states. The primary foci of the study are the size, date range and spatial distribution of these finds, with less emphasis on specific aspects of artefacts themselves (for example, cpigraphic and art historical features of late Antique silver plate) and the specific contexts in which individual deposits were found (for instance, localised archaeological backgrounds to individual finds or find groups). The immense chronological and regional scope allows broad changes in deposition patterns to be presented and examined, and a variety of possible interpretations of these patterns are offered in the final chapter.

expansion of settlement, the embellishment of private estates by the addition of mosaics and wall paintings, and the importation of specialised pottery. Hence a more realistic and practical approach needed to be found. Ultimately, it was decided that realising such an ambitious aim was possible, but that research had to proceed in stages. It was thus concluded that initially, precious metals, in all its various forms, would be one of the main classes of material which would be an integral part of such a study (meaning worked metals, not the raw material: although the closest worked material to source, ingots, have been included in this study). It was postulated that there was likely to be a link between the ability to procure precious metals (in all forms), and the economic success of different regions at different times. To assess which regions were able to command precious metal resources, reliance had to be placed on where deposits of precious metals had been discovered, as that would potentially provide the vital link between a localised population, and its economic success at specific periods in time. I accept that a straightforward concept such as 'supply and demand' is a simplistic way of viewing the late Roman exchange mechanism. Gift exchange was as likely to have been in operation as commercial transaction. This, however, can still be linked with the acquisition of precious metal resources, and hence 'command' is being used here in the broadest possible sense.

The background to the research

Originally, this study was initiated with a rather different and more ambitious aim in mind. The aim was to assess whether changes in the economic prosperity of different regions in the later Roman Empire were related to a wider underlying economic process. It was the intention to perceive if regional economic developments could be tracked across the whole Roman Empire, purely from the perspective of surviving artefactual evidence, and to develop ways of interpreting these regional variations once identified from the point of view of economic change. This was inspired by the work of researchers working from a macro-economic perspective such as Braudel (1984), who related fluctuations in European com prices during the late l'vliddle Ages to broader economic processes, and wide ranging chronological and archaeological studies such as the approach taken by Hodges and Whitehouse (1983). However, the most influential study, as it was based purely on the analysis of archaeological material, was that of Going (1992), who conducted an examination of fluctuations in the regional development of RomanoBritish pottery industries over the whole period of Roman administration of the island. In the latter study, Going suggested that it was possible to identify the rise and fall of different pottery industries, which he interpreted as being a factor of an underlying economic process and which were considered to be in the form of a 'wave'. The changing fortunes of different regional pottery industries were suggested therefore as being linked to periods of 'boom and bust' in the province.

Before any relationship between precious metal deposits and any underlying economic process could be postulated, fundamental areas of primary research needed to be addressed for the following reasons:

It was soon realised that such a goal could not be easily achieved within the constraints imposed by the research programme, simply because it would require the compilation of a vast database of information, and the assessment of a myriad of different factors. The main problem, of course, is how is economic success monitored from a material perspective? Economic success could potentially manifest itself in myriad of ways in the archaeological record. Examples might include the

•

No previous researchers had provided an adequate survey of precious metal deposits. Specialisation meant that most research had been conducted on specific classes of precious metal artefact, for instance, coin hoards and silver plate. Research had also tended to be restricted either geographically, chronologically, or both. The reintegration of data to provide a survey of material therefore became a research objective in its own right.

•

The construction of a database capable of moving towards realising the initial aim was complex, because making each deposit consistent - and hence comparable with other finds - required material to be gathered from a wide range of sources.

In addition, having constructed the database, a number of other questions of a more pragmatic nature arose which required examination: for instance, why were deposits buried in the first place? That is not to say that the original aim was abandoned: some of the reasons that emerged to account for the patterns in the data can be seen to relate

1

AIMS & PAST RESEARCH

theoretically to underlying economic changes (Chapter 5, p. 127).

ultimately ended up, curators became responsible for the care, acquisition and study of particular classes of artefact of particular regions and periods. Departments formed around certain classes of object or particular convenient parameters of time or space; for instance, the British Museum has a 'Coins and Medals' department, which deals with all forms of currency and medals for all historical periods and regions, and a department of 'Greek and Roman antiquities', which deals with all the material culture from the Hellenistic and Roman periods from all Mediterranean regions, but not coins, and not Roman Britain (which is dealt with by another department, the department of 'Prehistory and Europe' (formerly the departments of 'Prehistoric and Romano-British Antiquities' and 'Medieval and later Antiquities').

The initial theoretical aim, as set out above, thus became a different set of objectives after the initial assessment of the practicality of the project had been conducted. These objectives can be listed as follows: •

To examine briefly how precious metal deposits have been studied in the past, and develop new strategies appropriate to the aims of this study.

•

To provide a survey of precious metal deposits (gold and silver) produced and buried in the later Roman Empire and the successor Byzantine Empire, irrespective of where they have been discovered, but within the chronological parameters of AD 193 to AD 711, by dividing the material up chronologically and geographically. This effectively fixes every find of late Roman precious metals into a background context provided by all other deposits in the database.

•

•

To identify and discuss broader changes in spatial distribution of these deposits, and relationship of these changes to variations in size and content of the surveyed deposits over whole study period.

In order to address the aims of this research, it was fundamental that all deposits in the data set were as complete as possible, in order to allow them to be compared. Deposits do not respect the divisions imposed on their artefactual assemblages which result from the different objectives of curation or specialisation. If we therefore wish to discuss and compare deposits as whole groups in their own right, a number of material and scholarly problems arise:

the the the the

Material. Because of the divisions imposed on precious metal finds, artefacts become separated between disciplines; which usually means, in the modem context, between different museum departments. Hence, for instance, the material from Hoxne (discovered in 1992) when acquired by the British Museum was not kept as a whole assemblage: the coins are preserved in the Coins and Medals department, and the other artefacts (namely jewellery and silver tableware) in the department of Prehistory and Europe. This in itself does not usually present a problem in modem times, as whole deposits will normally be published in their entirety, even if the divisions are adhered to within these publications (e.g. Cahn and Kaufmann-Heinimann's 1984 publication of the Kaiseraugst treasure). However, major problems are presented by the past practice of 'dividing up the spoils' between museum departments, and subsequent publication in different volumes or papers. As an example this practise made it difficult to reconstruct a series of Russian deposits, necessitating a study visit to the Hermitage, as the assemblages - mostly discovered in the late nineteenth and early twentieth centuries - had been divided up between different parts of the museum's collections and published separately as a result.

To explore theoretical ways of interpreting these broad patterns as observed and described.

Past research into the study of precious metal deposits

The aim of this section is to discuss past approaches to the study and interpretation of precious metal finds, and relate these to the aims of this study as expressed above. It is my contention that the failure of academics to adopt an interdisciplinary approach to the study of precious metal finds from the late Roman period has limited our understanding of wider issues related to the deposition of these artefacts. The past research environment

In the post-Darwinian academic world of late nineteenth century Europe, strategies for the classification of artefacts were one of the principal developments in the study of the material culture of the past. Pitt Rivers, for example, is credited with the invention of 'typology', the fundamental aim of which was to classify objects in a similar manner to biological species (Bowden 1991: 55). This in time led to the formation of separate disciplines within archaeology: for example, numismatics (the study of coins), and lithics (the study of stone artefacts). It. would probably not be contentious to state that a major factor in the formation of these disciplines derived from the preferences and prejudices of individual scholars. This environment enabled expertise to develop to the levels which only specialisation can hope to engender. Within the context of museums, where the bulk of archaeological objects

Shelton's (1981) admirable reconstruction of the material from the Esquiline treasure (Appendix, no. 2006) demonstrates an even more difficult problem, namely that of material not just being divided up within one museum, but between several international collections. In cases where some objects are also in private hands, it makes the process of reconstructing the original contents of precious metal assemblages at the very least exasperating, and more often than not entirely hopeless. In the latter case, Shelton

2

AIMS & PAST RESEARCH

is forced to concede that the exact number of original pieces in the treasure will never be known (ibid.).

order to survey it as a coherent group, irrespective of the imposed material and scholarly divisions.

Scholarly. Specialisation has meant that precious metal deposits have not usually been discussed as a coherent group in their own right. As outlined above, the material itself has invariably been divided up into separate categories within museum collections, largely due to antiquarian ideals of material typology. In the case of hoards, this has usually meant division by material type: coins in one place, silver plate and jewellery in another. (Bullion, such as ingots, and coins made into pendants, present particular problems for museums, and will probably be found in different parts of the collections).

Numismatic research on coin hoards

Numismatic research on coin hoards has tended to be limited either chronologically, geographically, or both. Few specialists in the field have the luxury of being able to devote much research time to the wider implications of coin hoarding, particularly in terms of comparing coin deposits in the wider archaeological and historical framework. This is because of the fact that the publication of the material itself requires a great expenditure of time in order to keep apace with the actual discovery of new material. Given that in most countries, the laws concerning the discovery of precious metals draw these finds into the remit of national and local museums, this in effect leads to publications invariably restricting themselves to factual analysis of finds (for example, lists of coins by type), as curators are under pressure to deal with the material quickly.

Over time, this division of material led in tum to a tendency for academics to specialise within these particular find groups. This is not necessarily a bad thing, as different approaches need to be taken towards different types of material; a numismatist will apply a different set of skills to the study of a set of coins than a jewellery historian will to the study of a repousse bracelet. But the consequences of such specialisation can lead to problems when the wider questions of precious metal deposition in the later Roman Empire are addressed. This was highlighted by Martin Millett back in 1994, who stated: 'The curse is not specialisation itself but the failure to reintegrate results once studies have been completed to create an overall understanding'. (Millett 1994a: 103). Millett was suggesting that if material is studied in isolated groups, the interpretations will naturally come from a narrow perspective; thus those who specialise in the study of coins will tend to interpret deposits in one way, whilst those who specialise in, for instance, silver plate, will tend to interpret them in another. This is ably demonstrated by an important collection of papers published in Antiquite Tardive in 1997, where precious metal deposits from the late Empire are discussed from a number of different specialist approaches: for instance art historical (Mango), numismatic (Bland), and archaeological (Kiinzl).

Coin hoards, however, have received attention on a wider level, with the objective of contextualising the material. Duncan-Jones (1990) in Britain, and Depeyerot (1991) in France, are examples of scholars using the evidence of coin deposits, often in relation to literary evidence, to examine wider questions such as circulation, levels of taxation and the social distribution of wealth. In the latter study, for instance, Depeyerot uses a series of gold coin hoards and an extensive corpus of literary evidence from the Western Empire to suggest that there were declining levels of gold in the Western Empire from c.AD 370 onwards, as the state increasingly sought taxes in gold rather than silver. Other researchers have used the evidence of coin hoards to consider more fundamental questions concerning the nature of hoarding itself. For example, Guest (1994) gathered together coin hoards from the whole of the Westem Roman Empire (excluding the British Isles) of the Imperial period, in order to examine the validity of past interpretations of the phenomenon of hoarding. Aitchinson's (1988) paper called for coin hoards to be placed into their background archaeological context and introduced new ideas concerning the burial of hoards for potentially ritualistic reasons. Gricourt (1988) studied the spatial distribution of a specific set of late third century AD coin hoards from northern France in an attempt to track the movement of barbarian incursions in relation to relevant historical references. Banaji (1996) suggested that, by comparing the size of gold coin hoards of the late Roman period, changes in the distribution of wealth could . be indicated. ·

That is not to say, however, that inter-disciplinary research is never conducted; there are some fine examples of precious metal deposits being catalogued and discussed in an exemplary fashion, such as the extensive series of essays on all aspects of the new material from the Kaiseraugst treasure, including a complete re-examination of the context of the whole find (Guggisberg and Kaufmann-Heinimann 2003). Rather, it is to contend that the environment in which research must be conducted works against those wishing to look at wider aspects of deposition. Liaison between scholars from different specialist areas is not always as developed as it ideally should be, and the material itself is still largely divided up in to the same typological groups which it has been for the last 200 or so years.

A great deal of space could be devoted to discussing these works and many other papers concerned with Roman coin hoards, and indeed the phenomenon of hoarding itself as an historical activity. All the above papers touch upon issues raised within this study: none, however, have integrated coins with other classes of precious metal

To a large extent, the research contained within this monograph was driven by a desire to address some of these problems by bringing all this material together in

3

AIMS

& PAST RESEARCH

Roman period, it is worth briefly outlining some of the problems and questions which have arisen from one of the published debates mentioned above, that of Painter and Cameron.

artefacts, or studied finds m such a wide spatial and chronological framework. Research on late Antique silver plate and other non-coin artefacts

Painter (1988; 1993) and Cameron (1992) As is the case with coin hoards, specialisation, and the pressures associated with publication of new finds, has meant that much of the research work on areas such as silver plate and jewellery has concerned itself with the stylistic and typological aspects of the material. Detailed comparisons of different object types have been made within the contexts of find publications, for instance, Cahn and Kaufmann-Heinimann (1984 ), and Guggisberg and Kaufmann-Heinimann (2003). Both these works provide extremely detailed descriptions of the objects within the find from Kaiseraugst, covering such aspects as iconography, metallurgy and metrology. The authors also place specific object types within the spatial context provided by comparative material (for instance, fifth to seventh century distribution of European finds of wine strainers of all metals: 1984: 118, Fig. 57). Many corpora have been published to accompany exhibitions, for instance, Kent and Painter (1977) and Banek (1966; 1977), and these often include introductory essays which are generally concerned with describing the objects in iconographic and stylistic terms. Other research has concentrated on more specific aspects of the material. Examples are Dodd ( 1961) on Byzantine silver stamps, Hauser (1992), on the typology of late Antique silver spoon types, Griinhagen on Hacksilber (1954) and more recently Wiegels' corpus of silver ingots (2003). Others have seen silver plate as a facet of artistic expression that is able to provide an insight into how late Roman society functioned. For example, iconography has been the subject of analysis which aims to understand changing social attitudes in a world slowly modifying its religious and social ideologies (Schneider 'i983). A related study focusing on the meaning of the changing iconography of silver plate has been published more recently (LeaderNewby 2004). The wider contribution silver plate can make to our understanding of the late Roman period is an area which has really only just begun to be explored, and there are no instances of researchers attempting to reintegrate material with other categories of precious metal artefact, in particular, coins (although coin hoards are sometimes mentioned; an example is Johns (1996a)). Some of this recent research includes the changing function of silver plate items from the early to late imperial period (Martin-Kilcher 1984), the distribution and ownership of silver plate (Painter 1988; Cameron 1992; Painter 1993), and the interpretation of the reasons leading to the burial of silver plate in late Antiquity (Millett 1994a; Johns 1994; Johns 1996a). The important role of gold as a prestige substance in late Antiquity has been discussed by Janes ( 1998).

Painter ( 1988) addressed the issue of ownership and status of assemblages of silver plate. He argued that although the number of known production sites for silver vessels was small (only a very small proportion of extant vessels bear marks which can be associated with places of manufacture, namely Sirmium, Naissus, Nicomedia, Antioch, Thcssalonica, Heraclea in Thrace, Rome, Aquileia, Mainz, Trier and Cologne), the wide geographical spread of finds could be linked to the historical evidence for 'high status' owners having a number of estates throughout the Mediterranean world. Therefore, the diversity of mints evident in the material such as that from Kaiseraugst, support the belief that the material in the find was owned by a high ranking official with widely spread interests and a number of estates. Painter also argued that the high quality of_ the silver plate (usually 96-98% pure) (see TABLE 7), which made it difficult to work, implies that its production was more likely to have been subject to imperial control. Therefore, he argued, silver plate was likely to have been used as part of an official gifting system that was designed to maintain relations in the higher echelons of late Roman society (ibid.). Cameron (1992) took exception to Painter's 1988 paper. His principal objection was that so few pieces of late Roman silver can actually be linked with manufacturing sites, making a system of gift exchange unlikely. This was partly accepted by Painter (1993), who pointed out that the consistently high purity of silver plate of late Antiquity, even if few pieces bore mint marks, continued to imply that the manufacture of plate was centralised and probably under official control. Cameron argued for a more consumer-led approach to the acquisition of silver plate, with officials of high status and wealth anxious to acquire silver plate commercially to impress their 'inferiors' (Cameron 1992: 185).

The approach of this study This debate, outlined briefly above, illustrates some of the pitfalls of relying upon only one class of precious metal find when addressing wider issues such as exchange mechanisms. My contention is as follows: how can we place a value on a particular class of precious metal artefact, or specific assemblages of material, when we do not understand how the various classes of object or individual finds relate to each other? How much importance should be placed on individual finds? For example, Painter (loc. cit.) refers to a biographer of Claudius Gothicus who describes a gift of silver of approximately 62-63lbs. He compared this to the Mildenhall find, which weighs c. 85 lbs., and concluded that such assemblages must have been owned exclusively

ln order to demonstrate the strengths of the approach adopted by this thesis, i.e. placing each find into the context of the deposition of all precious metals in the later

4

AIMS & PAST RESEARCH

by people of high status. I do not necessarily disagree with this view, but how important is an assemblage like Mildenhall in the context of late Roman society? We can only begin to answer this question by comparing Mildenhall not just with other assemblages of late Roman silver plate, but all deposits of late Roman precious metals. It could, theoretically, emerge that the Mildenhall find is actually rather insignificant in terms of its position in place and time. For the broad period in which it was deposited, there may be another region where there is a high concentration of gold coin hoards, which completely overshadow the Mildenhall find, and suggest that there were other areas of the Roman world where we could equally argue for the presence of 'high status' individuals, either receiving gifts (after Painter) or acquiring products commercially (after Cameron). This study, however, is not aimed purely at the issue of the status of individuals in the later Roman period. The philosophy behind the project was to allow the material to speak for itself. By organising the data as objectively as possible, in order to give every deposit equal status, and by ignoring divisions between artefact classes, changes in the distribution and nature of deposits can be tracked over the 500 year study period. Only after this has been established can the wider issues related to ownership, status, and reasons behind burial be addressed. To put this aim into practice, the following criteria relating to each precious metal deposit included in the database needed to be established: a. its geographical location, 1.e. its position in spatial terms; b. its chronological location, i.e. its position in time; c. its size in terms of precious metal content; d. the date range of material contained within it; e. a description of its content adequate enough to characterise it. In the following chapter, the methods which allowed these criteria to be established and the database of deposits to be constructed (Appendix) are set out.

5

CHAPTER2 METHODOLOGY: THE CONSTRUCTION OF THE DAT ABASE

This chapter is concerned with the means by which the database was constructed and the way in which the material was subsequently organised.

As for coinage, there arc some very sizeable bronze coin hoards from the late Roman empire (for instance Chapmanslade, Wiltshire, CHRB X: 339-69) but these too have been excluded from this study. Although the individuals burying these coins clearly saw them as having a value worth attempting to maintain, in actuality any value was entirely token. Bronze coins were produced in vast numbers, particularly in the fourth century AD, and would have facilitated small scale transactions in urbanised parts of the Empire. But they only functioned if it was possible to convert them into higher value precious metal coins via money changers, and that required the support of an the underlying economic infrastructure. Bronze coins were of little use outside the system, and there are numerous examples which demonstrate this. For instance payments to hostile tribes beyond the imperial frontiers were invariably made in silver or gold; Tacitus tells us that tribes beyond the Rhine/Danube frontier preferred older, and thus less debased, silver denarii to other forms of silver ( Germania, chapter 5), and copper coins would not have been accepted. Evidence drawn from an examination of a number of bog bodies from Denmark and free Germany also demonstrate a preference for imported Roman objects of gold and silver, principally jewellery and coin (van der Sanden 1995: 160).

Why only precious metal deposits?

This study restricts itself to gold and silver items, because from the earliest periods during which these substances arc fashioned into artefacts, they are held in far higher esteem within the societies that produce them than artefacts made of other metals. This is principally a factor of rarity. Whereas the sources of base metals were abundant and easy to locate, gold and silver sources were not. The invasion of Britain by Claudius in AD 43 for instance may have been partly motivated by the fact that the island was known to be a source of abundant lead ore and thus silver which could be cupellated (Pliny, Natural History, 34.49. 164), and gold scams were soon exploited in places such as Dolaucothi in Carmarthenshire (James 2003: 14-15). Throughout the Roman period, the imperial authorities sought to identify the sources of gold and silver and control their output; it has some resonance with the modem desire of the United States and Western Europe to secure and exploit precious oil reserves in regions of the world outside their immediate sphere of influence (Klare 2005).

In addition, trading with other parts of the world was only possible with silver and gold coin, for instance the purchase of pepper and precious stones from India (Shajan et al. 2004), where numerous Roman gold and silver coins have been found. Once again, copper coins would have been little use to Roman traders in far flung places.

In addition to rarity, gold and silver artefacts were often more carefully worked than base metal objects, and the most highly skilled craftsmen were employed to carry this out. The malleability of gold must have also elevated its prized status, in addition to the fact tliat in pure form it never tarnishes. In any case it is clear that base metal objects were not usually worked to the highest standards, for example pewter vessels (e.g. Poulton and Scott 1993), which only have rudimentary decoration and whose original forms were too easily bent out of shape. There are of course exceptions, such as bronze vessels exquisitely inlaid with enamel and glass, but as it is extremely difficult to place a 'value' on such objects, for reasons which will become clearer later in this chapter, these have not been included here.

It is also clear that when the coinage system breaks down,

it is gold and silver which becomes the principal means of facilitating exchange and making payments. This explains why coinage in Britain was clipped to make imitations as the supply of coinage dried up (with the clippings perhaps also used to make silver ingots, such as those in the Balinrees deposit (Appendix, no. 1472)), and why all over the former Empire pieces of silver tableware were broken down into exchangeable chunks (Hacksilber). For all the reasons outlined, base metal deposits have not been included in this study.

Gold was used when the emperor sought to indicate his elevated status, such as when commemorative medallions were struck. These were prestigious items produced in relatively small numbers, compared with the output of small module copper coins. Silver was another metal chosen to bestow imperial patronage, such as the commemorative plates produced to celebrate the decennalia of Licinius (Overbeck 1973). Objects made from both types of metal are illustrated in the oft-cited illustration in the fifth century Notitia Dignitatum showing the insignia of the post of Comes Sacrarum Largitionum, or Count of the Imperial Largesse (King 1980b).

What constitutes a precious metal deposit?

There are two means by which material can enter the archaeological record. The first is by being lost accidentally or discarded: instances where this can be applied are stray losses of coin, or discarded pieces of ceramic which have outlived their usefulness. The second is deliberate burial, a process usually referred to as 'hoarding'. This is applicable to groups of more than one artefact, which, through common internal features and/or other external indications (for instance, associated

6

METHODOLOGY: DATABASE CONSTRUCTION

containers), appear to have been deposited (in a single or number of episodes), and subsequently never recovered. Different researchers have adopted varying criteria for inclusion of material in their surveys. For example, in the series Coin Hoards from Roman Britain, single silver coins are not included, but single gold coins sometimes are, in the belief that single silver coins must have been lost accidentally, whilst single gold coins, being of much higher value, could potentially have been buried deliberately (i.e. hoarded). But others have adopted a different approach: Fagerlie (1967), in her corpus of Byzantine finds from Scandinavia, defined 'hoards', somewhat incongruously, as deposits which included five or more pieces.

Clearly, within these categories there are bound to be some anomalies. These can be summarised as follows: •

Bracteates are considered as jewellery even though they often copy late Roman gold coin types.

This study concentrates on the latter type of find, and to be included, deposits only had to have an element, however small, of precious metal (gold and/or silver). The data therefore ranges from deposits with a single base silver coin in a hoard of bronze nummi, to large deposits of ornately decorated silver plate. In addition, qualification for the database was not based upon theoretical motives for burial: hence finds associated with inhumations, for example, are included alongside coin deposits secreted in ceramic vessels, or complex deposits drawn from a wide range of sources. The following classes of gold and silver material· have therefore been included in this study: •

Coin: all deposits containing gold and/or silver coin, including pieces associated with bronze issues. Medallions have also been classed as coins;

•

Plate: gold and silver artefacts manufactured from thin to thick sheets of metal, used to make a whole range of products including domestic tableware items (for instance, dishes and spoons); toilet items (for instance, toothpicks); items of a ritual/ecclesiastical nature (including votive plaques, altar cladding from churches, statuettes of Classical deities, reliquaries); military items (for instance silver cladding used on ceremonial helmets); furniture fittings (for instance from boxes); horse trappings;

•

Bullion: gold and silver bullion, including ingots (official and local pieces), scrap metal (wire, fragments, pieces of sheet metal and so on), and Hacksilher (pieces of silver plate deliberately cut into smaller pieces - the nomenclature of the term is discussed by Johns (1996b));

•

Jewellery: gold and silver items of personal adornment, ranging from finger rings to bracelets and items of supposedly official regalia, such as belt sets.

Looped and pierced coins: these are generally categorised as 'coin', even though in effect they have been taken out of circulation and turned into jewellery. However, in cases where coins have been incorporated into items of jewellery, for instance necklaces or chains, or in one case silver plate (from Toulouse) these are considered to be either jewellery or plate, as the largest part of the object in question fits into that category (i.e. the coin or coins are a composite component).

•

Folded items of plate: sometimes, items of silver plate were folded before burial, usually to save room in the container chosen place of burial. In these cases, these are usually categorised as silver plate rather than Hacksilber.

•

Fragments of silver plate, particularly spoon handles: once again, these objects will still be categorised as silver plate, because without examining each piece indivually, it is not possible to establish if the break is deliberate or accidental, for instance due to plough action. However, if fragments of larger silver objects are found in hoards which date to a period many decades after the likely date of original manufacture, these will be considered as Hacksilher as opposed to silver plate which may have been broken after deposition. It is also often unclear if other objects originally soldered together should be considered as Hacksilber or silver plate; but once again, it is considered that in most cases it is more likely that the solder has failed, and so these items are also usually categorised as silver plate.

Sources of data The study area covers both W cstcrn and Eastern parts of the Empire and areas outside the frontiers. As the emphasis of the study is on broad spatial changes in the deposition of precious metals, it was felt that as wide a geographical area as possible should be examined. The accessibility of regional corpora was a distinctive factor in the construction of the database. The only major area which has not been included is mainland Greece, mainly because there are no easily accessible corpora of hoards for this region. Some Greek hoards have made their way into the study, but no systematic attempt was made to improve the coverage of this particular region.

7

METHODOLOGY: DATABASE CONSTRUCTION

Published corpora of material were the main. source of data for this study. Exhaustive searches through periodicals would have been far too time consuming, and would not have yielded great rewards. This was because it was not essential to establish detailed descriptions of individual finds for the purposes of this research. Given that this study is concerned with broad patterns in the deposition of late Roman precious metals, it was not necessary to expend large amounts of research time for the sake of 'completeness'. Sufficient material is provided by the database as it stands for the aims of this study to be realised. As for the database itself, (Appendix) this is unique, as no previous study has organised material in this manner: it places deposits from the whole array of precious metal atiefact types alongside each other, purely on the basis of the date of the latest item in the find, and irrespective of provenance.

Single component deposits

Due to the tendency of researchers to divide artefacts into 'coin' and 'non-coin' groups (as discussed in Chapter 1, p. 3), sources of data were naturally diverse. There are also no international standards concerning strategies of data collection and presentation. Much relies on the energies and resourcefulness of individual scholars and institutions. Not surprisingly therefore, the more recent the publication, the less time which needed to be spent on establishing full details of particular finds. Despite this, even recent publications sometimes lack vital details concerning objects. For example, Shelton's (1981) publication of the Esquiline deposit failed to include object weights. Although corpora of coin hoards and assemblages have been systematically compiled for many regions, general corpora of precious metal deposits have not been conducted in the past. 'Coin' and 'non-coin' data sources are therefore best discussed separately.

Mixed component deposits (breakdown) No.

No.

%

Coin

1,697

83.0

Plate

92

4.5

Bullion

Coins, jewellery

%

119

53.6

Coins, jewellery, plate

22

9.9

23

1.1

Coins, jewellery, bullion

20

9.0

Religious

7

0.3

Coins, plate

20

9.0

Jewellery

3

0.1

Coins, bullion

15

6.8

0.0

Plate, religious

6

2.7

Coins, plate, bullion

4

1.8

Plate, jewellery

5

2.3

Coins, plate, religious

3

1.4

Bullion, plate

2

0.9

Coin hoard types

Coins, religious

1

0.5

Silver only

Military Mixed deposits

Total

Gold only Gold and silver Total

222

10.9

2,045

100.0

1,314

77.4

Coins, jewellery, plate, religious

319

18.8

Coins, plate, bullion, jewellery

1

0.5

64

3.8

Plate, bullion, religious

1

0.5

1,697

0.5

Plate, jewellery, bullion Plate, jewellery, bullion, religious

Total

0.5 1

0.5

222

100.0

TABLE 1. Summary of the different types of deposit included in this study. The first three columns list deposits with a single artefact type and provides a total for the number of mixed finds; the remaining columns provide details of these 223 deposits which have more than one component.

Martin Price at the British Museum, listing coin hoards of all regions and periods by consultation with mainly British and foreign numismatists; and the Roman Imperial Coinage project, some later volumes of which include listings ofhoards (for example: Kent 1994). Guest's (1994) doctoral research on Roman imperial coin hoards of the Continental western Roman Empire proved both an invaluable source for tracing references to lesser known coin finds, and provided much of the primary information

Coin deposits There are two types of corpora of coin finds: regional (invariably modem country; occasionally geographical zone, for example, Scandinavia); or by coin/hoard type (for example, gold coin hoards). Three important general works should be highlighted, as these are neither regionally, nor necessarily chronologically specific: Mosser (1935), a collection of Byzantine coin hoards from all regions; Coin hoards, a project initiated by the late Dr.

8

METHODOLOGY: DATABASE CONSTRUCTION

for deposits prior to AD 400. The rest of the data sources are best discussed by region.

region where the coverage of material is generally poor is northern Italy south of the Alps.

a. The West

North Africa. North Africa does not have any relevant systematic research projects, with the exception of Morrison (1987), a study which includes details of gold finds from the Vandal period.

Germany and central Europe. The most extensive and systematic programme is the Fundmunzen der Romischen Zeit project, established in 1960 initially in Germany, in order to detail all coin finds - both excavated site losses and hoards - of the Roman period by modem German district (1960-2003). Each assemblage of material is assigned a unique number and as much detail as possible is provided on content and the circumstances of discovery. Additional non-coin artefacts are mentioned, but do not tend to be listed systematically. Full references are given for each assemblage. The Fundmiinzen project has been extended to Luxembourg, Austria, the former Yugoslavia (but only for Slovenia) and Hungary.

b. The East A systematic corpus of Eastern coin finds is sadly lacking, particularly for Asia Minor and the Middle East. The picture is better for the Balkans; with vital work by Mirnik (1981) and Kos (1986; 1988) in Yugoslavia, Gerasimov in Bulgaria (various), and Bordea and Mitrea in Romania (various). Dacian hoards of the third and early fourth centuries have also been compiled by Gazdac (2002). Data on the eastern Mediterranean and the Black Sea States is drawn from a variety of sources, notably the general works previously cited, such as Kent (Zoe.cit.), and other publications ( e.g. Lightfoot 1991). Systematic searches were also conducted in the Coins and Medals department at the British Museum in the relevant finds sections of the library, where collected off-prints from periodicals are housed. Egyptian coin hoards from the fourth and fifth century have been summarised by Ford (2000).

Britain. Although a project of the nature of the Fundmunzen does not currently exist for Britain, coin hoards are nonetheless well served. Robertson's Corpus of Romano-British coin hoards (RNS Special Publications 20) details all British hoards based on literature searches of county and national journals. In addition, since 1979 the British Museum has overseen the production of the series Coin Hoards from Roman Britain, now in its twelfth volume. Access to details of some other unpublished British coin finds were provided by staff in the department of Coins and Medals at the British Museum. The important paper published by Archer (1979) was also invaluable, as it provides details of late Roman silver coin hoards from the province.

c. Outside the frontiers Scandinavia. Scandinavia is well served by the excellent corpus by Fagerlie (1967), although Norway is not included in this work.

France, Belgium, the Netherlands. In France, a similar project to Britain exists, again initiated in 1982 by the Societe Fran9aise de Numismatique ( Corpus des Tresors Monetaires Antiques de la France), although this is organised more systematically by region in a similar manner to the Fundmunzen project, and provides details of finds from all historical periods, not just Roman. In addition there is also the series established by the Bibliotheque Nationale, Tresors Monetaires (1979 onwards). There are other important additional finds lists, notably Blanchet (1900), Lafaurie (1958), and Brenot and Loriot (1992). Belgium is served by Thirion's extensive but now out-of-date 1967 corpus. Van der Vin (1988) provides a useful listing of gold hoards of the late fourth century in the Netherlands.

Russia. The former Soviet Union has two important and unsurprisingly extensive corpora (given the vastness of geographical area) by Kropotkin (1962; 1966), which covers both the Roman and Byzantine periods, and Nudelman (1976), which deals purely with the Byzantine period. Non-coin art.efacts

The rarity of items of silver plate, gold and silver jewellery usually leads to their disposition in national museums, which fundamentally determines the nature of their publication. Individual assemblages tend to receive exclusive publication, due to their size, and this usually falls to the national museums and individual curators albeit often in collaboration with other specialists (for example: Kaiseraugst (Cahn and Kaufrnann-Heinirnann 1984; Guggisberg and Kaufrnann-Heinimann 2003); Notre Dame D'Allem;:on (Baratte 1981); Kaper Koraon (Mango 1986)). The other sources of data are exhibition catalogues, usually based around items from a particular museum (for instance, Kent and Painter 1977; Banek 1966; 1977), which illustrate particular pieces and provide brief details of their original provenance. The other sources are assemblages of certain types of artefact, for instance,

Iberia. There are no systematic compilations of finds from either Spain or Portugal. The two main sources of data are Castro Hipolito (1960), and Pereira et al. (1974)), with a more recent corpus of gold coin finds (Bost et al. 1992). Italy. Italian hoards come from a number of diverse sources, the main ones being Perantoni Satta (1954), and the Ritrovamentali Monetali de Eta Romana nel Veneta II/2 (1992). Important works have also been published by Panvini Rosati (1953; 1985) and Ungaro (1985). One

9

METHODOLOGY: DATABASE CONSTRUCTION

ingots (Wiegels 2003), or spoons (Hauser 1992). There are also papers and volumes which concentrate on a particular aspect of certain artefacts and which help to re-assemble deposits, notably silver picture plates (Toynbee and Painter 1986), and Byzantine stamps (Dodd 1961).

those with a number of different artefact types. The majority of hoards in the study are coin hoards not mixed with other components (83%). Finds which contain plate (usually silver) either as a single component or in mixed deposits account for 159 finds or about 8% overall. The commonest type of mixed hoards are those which contain coins and a jewellery component, usually a finger ring or rings (just over 50%).

During April 1995, a study visit was conducted at the Hermitage Museum in St. Petersburg. Under the guidance of Dr. Jerusalem Zalleskaya and Dr. Alexander Nikitin, details of a series of deposits containing Byzantine silver plate items from the former Soviet Union were established by reference to actual objects in the museum itself, museum registers, and a series of archaeological reports which were not readily accessible in Britain. An example of one such invaluable work was the Russian Archaeological Commission's gazetteer of ancient gold and silver metal ware (Smirnov 1909).

Dates of discovery

Types of deposit

When was most of the material within this study discovered? FIGURE 1 summarises the dates of discovery of all deposits, where date of discovery is known. Not surprisingly, the majority of finds included come from the post-war years (about 40%). This pattern undoubtedly reflects improved reporting of discoveries across all regions included in the study, as well as conscientious efforts to maintain better records of these finds.

TABLE 1 provides a summary of the types of deposit which have been included in this study. Details of 2,045 deposits have been collected. These can be divided into two groups, those with a single type of component, and

There are ten deposits from the sixteenth and seventeenth centuries, and it is worth pausing to assess how these survived long enough for it to be possible to include them in this corpus. TABLE2 summarises these finds.

Year of disc.

Deposit/ database no. Date

Comments

1568

200-300 610-40 300-400 300-400 395-410 348-63

Silver cup, Kunthistorisches Museum, Vienna. Gold coins and votive plaques, Lateran Palace?. AV coins, AR plate and jewellery; all lost. AR plate; all lost. AV and AR coins; dispersed. Four AV coins; present location not established.

411-24

AV coins; present location not established.

1653

Arras (no. 1946) Rome I (no. 1897) Laski (no. 1995) Trier I (no. 2011) Alcester I (no. 1468) Nielles- les-Calais (no. 1359) Villers-l'Hopital (no. 1624) Tournai (no. 1730)

457-90

1656

Avignon (no. 1978)

300-400

C. 1675

Venlo (no. 1592)

395-410

Childeric's tomb; Bibliotheque National, Paris; Tyroler Landesmuseum, Innsbruck. AR plate ('Scipio's shield'); Cabinet des Medailles, Biblioteque Nationale, Paris. AV coins; dispersed.

1587 1610 1628 C. 1638 1639 1646

TABLE 2. Deposits discovered in the sixteenth and seventeenth centuries.

There are clear distinctions in these ten deposits with regard to the fate of the objects found within them. The coinage was invariably lost or dispersed. In general this can be explained by the fact that coinage is more often than not of common types which cannot later be attributed back to their original sources on appearance, once they have been sold on through auction, even if they do end up in public collections. Being physically small, individual coins also tend to be more easily lost and would often have been melted down instead of retained, and even if they were retained their size made them easy to share discretely amongst interested parties.

plate could suffer the same fate of either being lost or melted down. The best example of this is the Trier hoard, which was found by novices in the grounds of a Jesuit monastery. What is extraordinary about this particular discovery is that although it was subsequently melted down in its entirety, it was fully recorded - including accurate weights - before this unfortunate act was carried out. What discoveries such as Trier demonstrate is the fact that we can be certain that large quantities of precious metals discovered in the Middle Ages would have been melted down soon after they were unearthed. There would have been a variety of reasons for this; in the case of the Trier assemblage, it was done because the Jesuit monks .feared that the silver would be discovered and confiscated, and could be made better use of in bullion form than as the original objects. But the fact that an effort was made, in this case by means of a scientific study far ahead of its

In contrast, items of silver plate can usually be linked back to their original provenance. A good example of this is the earliest recorded find in this list, from Arras in France, unearthed in 1568. This silver cup, remarkably, did survive, and is currently in Vienna, having been acquired from a French collection in 1809. But like coinage, silver

10

METHODOLOGY: DATABASE CONSTRUCTION

time, demonstrates that the archaeological and art historical value of the pieces within the hoard were at least recognised. This however was very much the exception to the rule: in most cases where treasure was discovered during this period, the finders would not have thought twice about the historical significance of their find, and

would have only considered the bullion value contained therein. The issue of non-reporting of material discovered is discussed further in Chapter 5 (p. 126).

12.0

10.0

i'l:i

8.0

" "I:!

6.0

01

.!! ~

8.

4.0

2.0 0.0

.,..,.,.....,..-r,"-t.,..,...,..,...,_.,....,.,..,..,..,..,..,..,..,..,.,..,..,..,..,...,.,.,..,..,.,..,..,..,..,..,..,..,..,..,..,.., ,"'"' ,~ ,'P' ,~"' ,~"' .... -;:,,.,."'._,P,re-"'"' ,""' .:t ..._'oOJ..._rorr.,..... ~ ..._ro-