<%BANNER%>

Marginal Money

MISSING IMAGE

Material Information

Title:
Marginal Money Coins, Frontiers, and Barbarians in Early Byzantium (6th-7th Centuries)
Physical Description:
1 online resource (651 p.)
Language:
english
Creator:
Gandila, Andrei
Publisher:
University of Florida
Place of Publication:
Gainesville, Fla.
Publication Date:

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
History
Committee Chair:
Curta, Florin
Committee Members:
Geggus, David P
Effros, Bonnie
Sterk, Andrea Louise
Gillespie, Susan D

Subjects

Subjects / Keywords:
archaeology -- byzantium -- coinage -- frontiers -- warfare
History -- Dissertations, Academic -- UF
Genre:
History thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract:
The dissertation explores the nature of the early Byzantine frontier on the Lower Danube both from a historical and an archaeological perspective. The main thesis is that the frontier is a multifaceted concept which cannot be encapsulated into a single definition as most students of the frontier question have done so far. Instead, the same frontier can act both as a political/ military frontier of exclusion and as a cultural frontier where ideas, fashions, and people can circulate more or less freely. The nature and extent of cultural contact can be properly understood only through the study of Byzantine artifacts found beyond the political frontier. Rather than being a comprehensive treatment of all categories of imports, the dissertation addresses a selected number of diagnostic items which lend themselves to a versatile analysis due to their wide distribution, variety of types, and multiple implications in the realm of social, economic, and religious development of the frontier area and the regions beyond. Such artifacts are amphorae, lamps, jewelry, and coins. Among these artifacts early Byzantine coins represent one of the most interesting, yet unexplored media whereby the Empire and barbaricum connected at multiple levels. The study of numismatic evidence clearly points to the fact that economic, cultural, and political borders are not coterminous and the Empire’s influence can be traced far beyond its administrative limits. One of the main arguments is that economic frontiers existed inside the Empire as well, due to economic fragmentation and different levels of prosperity and integration. The dissertation challenges some of the previous interpretations of early Byzantine coins found beyond the frontier by arguing that an ethnic interpretation of finds is problematic. Drawing on anthropological studies, I also argue that symbolic and ritual functions should be taken into account along with the economic role of coins, usually overemphasized in previous scholarship. At a deeper level, the dissertation aims to contribute to the methodology of using numismatic evidence as the primary source of a monograph which attempts to weave numismatics, archaeology, and history into a homogeneous narrative.
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility:
by Andrei Gandila.
Thesis:
Thesis (Ph.D.)--University of Florida, 2013.
Local:
Adviser: Curta, Florin.
Electronic Access:
RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2015-08-31

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
lcc - LD1780 2013
System ID:
UFE0045718:00001

MISSING IMAGE

Material Information

Title:
Marginal Money Coins, Frontiers, and Barbarians in Early Byzantium (6th-7th Centuries)
Physical Description:
1 online resource (651 p.)
Language:
english
Creator:
Gandila, Andrei
Publisher:
University of Florida
Place of Publication:
Gainesville, Fla.
Publication Date:

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
History
Committee Chair:
Curta, Florin
Committee Members:
Geggus, David P
Effros, Bonnie
Sterk, Andrea Louise
Gillespie, Susan D

Subjects

Subjects / Keywords:
archaeology -- byzantium -- coinage -- frontiers -- warfare
History -- Dissertations, Academic -- UF
Genre:
History thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract:
The dissertation explores the nature of the early Byzantine frontier on the Lower Danube both from a historical and an archaeological perspective. The main thesis is that the frontier is a multifaceted concept which cannot be encapsulated into a single definition as most students of the frontier question have done so far. Instead, the same frontier can act both as a political/ military frontier of exclusion and as a cultural frontier where ideas, fashions, and people can circulate more or less freely. The nature and extent of cultural contact can be properly understood only through the study of Byzantine artifacts found beyond the political frontier. Rather than being a comprehensive treatment of all categories of imports, the dissertation addresses a selected number of diagnostic items which lend themselves to a versatile analysis due to their wide distribution, variety of types, and multiple implications in the realm of social, economic, and religious development of the frontier area and the regions beyond. Such artifacts are amphorae, lamps, jewelry, and coins. Among these artifacts early Byzantine coins represent one of the most interesting, yet unexplored media whereby the Empire and barbaricum connected at multiple levels. The study of numismatic evidence clearly points to the fact that economic, cultural, and political borders are not coterminous and the Empire’s influence can be traced far beyond its administrative limits. One of the main arguments is that economic frontiers existed inside the Empire as well, due to economic fragmentation and different levels of prosperity and integration. The dissertation challenges some of the previous interpretations of early Byzantine coins found beyond the frontier by arguing that an ethnic interpretation of finds is problematic. Drawing on anthropological studies, I also argue that symbolic and ritual functions should be taken into account along with the economic role of coins, usually overemphasized in previous scholarship. At a deeper level, the dissertation aims to contribute to the methodology of using numismatic evidence as the primary source of a monograph which attempts to weave numismatics, archaeology, and history into a homogeneous narrative.
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility:
by Andrei Gandila.
Thesis:
Thesis (Ph.D.)--University of Florida, 2013.
Local:
Adviser: Curta, Florin.
Electronic Access:
RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2015-08-31

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
lcc - LD1780 2013
System ID:
UFE0045718:00001


This item has the following downloads:


Full Text

PAGE 1

1 MARGINAL MONEY: COINS, FRONTIERS, AND BARBARIANS IN EARLY BYZANTIUM (6TH 7TH CENTURIES) By ANDREI GANDILA A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIRE MENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2013

PAGE 2

2 2013 Andrei Gandila

PAGE 3

3 Memoriae Patri

PAGE 4

4 ACKNOWLEDGEMENTS While preparing this dissertation I have incurred many debts, which I am happy to acknowledge. This study would not have been po ssible without, help, advice, and cooperation from a large number of people. I am particularly grateful to Florin Curta, my dissertation advisor and mentor for his constant guidance and intellectual rigor. I would like to thank the members of my committee Bonnie Ef fros, David Geggus, Susan Gille s p ie, and Andrea Sterk for their advice and encouragement. The research of this study, conducted in Europe and the United States, was made possible with generous support from Dumbarton Oaks Research Library and Col lection, the Medieval Academy of America, the American Numismatic Society, the International Association of Byzantine Studies, and the Center for the Humanities and the Public Sphere, University of Florida. This dissertation is the outcome of a long and ex citing journey. The second chapter is the result of research conducted in 2010 at the American Numismatic Society in New York where I have taken advantage of its unparalleled numismatic library including rare books from the Age of Enlightenment The third chapter was written at Dumbarton Oaks in Washington DC during a short term pre doctoral residency in 2011, but essentially grew out of two early seminar papers from 2007 and 2008. The fourth chapter was researched in Florida in 2011 and in Romania in 2012 with generous database has provided a goldmine of information, without which the research period devoted to this chapter would have become agonizingly long. Archaeological resea rch conducted at Capidava since 2001, on the site of an important early Byzantine fortress

PAGE 5

5 University o f Bucharest has provided indispensable hands on experience with archaeo logical material from the frontier region. Research for the fifth chapter was done in two stages. The beginnings of the project date back to my time as an Assistant Curator at the National History Museum of Romania (2005 2007), where I have benefited from the guidance and advice of Ernest Oberl n der Trnoveanu who initiated me into the fascinating world of numismatics. The chapter took shape in New York during the 2009 Summer Seminar generously sponsored by the American Numismatic Society, where I had the privilege to study the largest collection of Early Byzantine coins in the world. I enjoyed much help and ho spitality from the ANS curators and I would like to mention the numerous numismatic conversations with Robert Hoge, Peter van Alfen, and Richard Wits chonke. As the project expanded I have benefited from close collaboration with scholars from Bulgaria, Macedonia, Turkey, and Israel who generously gave me access to unpublished manuscripts and offered precious advice. I would like to thank Georges Abou Di wan, Gabriela Bijovsk i, Zeliha Demirel Gkalp, Maj a Hadji Maneva, Stoyan Mihailov and Alena Tenchova I would also like to thank Alan Stahl for commenting on an early draft of this chapter. The final chapter, which is also the most substantial, was resea rched in Romania with generous funding from Public Sphere. The origin of the project dates back to a paper presented at an international conference in Cracow in 2007 where I realized the intellectua l potential of the topic. I wrote the chapter in the genial atmosphere of Dumbarton Oaks during my one year Junior Fellowship in 2012 2013. The superb library of the Center invited a

PAGE 6

6 much more ambitious undertaking than initially envisaged. I have also ben efited from numerous thought provoking conversations about history, archaeology, and numismatics with Ccile Morrisson, Julian Baker, Rebecca Darley, Kuba Kabala and Axel Nielsen. I have presented early versions of this chapter and of several others at sc holarly meetings in Europe and the United States. Conference audiences in Kalamazoo, Glasgow, Cracow and Sofia have made many helpful comments for which I am grateful. Last but not least I would like to thank my wife and my mother for their love and unco nditional support during my graduate career.

PAGE 7

7 TABLE OF CONTENTS page ACKNOWLEDGEMENTS ................................ ................................ ............................... 4 LIST OF TABLES ................................ ................................ ................................ ............ 9 LIST OF FIGURES ................................ ................................ ................................ ........ 10 LIST OF ABBREVIATIONS ................................ ................................ ........................... 14 ABSTRACT ................................ ................................ ................................ ................... 18 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 20 2 ARCHAEOLOGY, AND NUMISMATICS ................................ ................................ 30 2 .1 A Circuitous Road: Byzantine History, Archaeology, and Numismatics in Western Scholarshi p from Du Cange to World War 2: ................................ ........ 30 2 .2 Post War Developments: Searching for Common Ground ............................... 41 2 .3 Coins Beyond Frontiers: The Use and Abuse of the Numismatic Evidence in National and Regional Studies ................................ ................................ ............ 50 3 THE EMPIRE'S FRONTIER ON THE LOWER DANUBE ................................ ....... 60 3.1 The Frontier Question in Late Antiquity ................................ ............................. 60 3.2 The Early Byzantine Frontier on the Lower Danube: Topos and Reality ........... 66 3.3 Putting t he Danube into Perspective: The Early Byzantine Frontier in the East ................................ ................................ ................................ ..................... 77 4 CULTURAL ENCOUNTERS ON THE LOWER DANUBE FRONTIER AND BEYOND: AN ARCHAEOLOGICAL PERSPECTIVE ................................ ............. 82 4.1 The Cultural Background ................................ ................................ .................. 82 4.2 Categories of Imports and Imitations in barbaricum ................................ .......... 91 4.2.1 Amphorae ................................ ................................ ................................ 93 4.2.2 Lamps ................................ ................................ ................................ .... 101 4.2.3 Molds and Metallurgical Tools ................................ ............................... 121 4.2.4 Brooches and Buckles ................................ ................................ ........... 129 4.2.4.1 Fibulae with bent stem ................................ ................................ 129 4.2.4.2 Cast fibulae with bent stem ................................ .......................... 137 4.2.4.3 Sixth to Seventh Century Bow Fibulae ................................ ........ 146 4.2.4.4 Byzantine buckles ................................ ................................ ........ 157 4.3 Christianity North of the Danube ................................ ................................ ..... 165

PAGE 8

8 4.4 Conclusion: Contact and Separation on the Lower Danube Frontier .............. 177 5 CENTER AN D PERIPHERY: COIN CIRCULATION IN EARLY BYZANTIUM ...... 192 5.1 The Role of Money in the Early Byzantine Economy ................................ ...... 192 5.2 Coin Productio n, Coin Circulation and the Nature of the Evidence ................. 215 5.3 A Comparative Approach to Early Byzantine Coin Circulation: The Balkans, Anatolia, and Syria Palestine ................................ ................................ ............. 234 5.3.1 The Reform of Anastasius and the Pre 538 Coinage ............................ 243 5.3.2 The Post Reform Coinage of Justinian I ................................ ................ 252 5.3.3 Inflationary Tendencies and Decline (565 616) ................................ ..... 263 5.4 Conclusion: Money in the Early Byzantine World ................................ ........... 286 6 MONEY, BULLION, AND PRESTIGE: THE FUNCTION OF EARLY BYZANTINE COINS BEYOND THE FRONTIER ................................ .................. 323 6.1 General remarks ................................ ................................ ............................. 323 6.2 Geogr aphical and Chronological Patterns of Coin Finds in barbaricum .......... 332 6.2.1 The Land of Bronze: The Lower Danube ................................ ............... 332 6.2.2 The Land of Silver: Transcaucasia ................................ ........................ 369 6.2.3 The Land of Gold: The Carpathian Basin ................................ .............. 401 6.3 Money and "barbarians": Same Coins, Different Functions ............................ 432 7 CONCLUSIONS ................................ ................................ ................................ ... 470 APPENDIX A CORPUS OF EARLY BYZANTINE COIN FINDS IN BARBARICUM .................... 478 A.1 Context ................................ ................................ ................................ ........... 478 A.2 Inventory ................................ ................................ ................................ ......... 485 A.2.1 Single Finds ................................ ................................ .......................... 485 A.2.2 Hoards/ Collective Finds ................................ ................................ ....... 543 B STATISTICAL ANALYSIS OF COIN FINDS IN BARBARICUM ............................ 558 LIST OF RE FERENCES ................................ ................................ ............................. 570 Primary Sources ................................ ................................ ................................ ... 570 Secondary sources ................................ ................................ ............................... 572 BIOGRAPHI CAL SKETCH ................................ ................................ .......................... 651

PAGE 9

9 LIST OF TABLES Table page B 1 Single finds by metal, percent of each, and number of coins per year of reign 558 B 2 Single finds from the main geographic regions. ................................ ................ 558 B 3 Single finds of gold coins in barbaricum. ................................ .......................... 558 B 4 Single finds of copper coins ................................ ................................ ............. 559 B 5 Single finds of copper coins: mints. ................................ ................................ .. 563 B 6 Single find s of copper: denominations. ................................ ............................. 564 B 7 Single finds of copper coins found north of the Lower Danube: mints. ............. 564 B 8 Single find s of copper coins found north of the Lower Danube: denominations. ................................ ................................ ................................ 565 B 9 Single finds of copper coins found in Transcaucasia: mints. ............................ 565 B 1 0 Single finds of copper coins found in Transcaucasia: denominations. .............. 566 B 11 Single finds of copper coins found in the Byzantine bridge heads: mints. ........ 566 B 12 Single finds of copper coins found in the Byzantine bridge heads of: denominations. ................................ ................................ ................................ 567 B 13 Single finds of copper coins per year of reform. ................................ ............... 568 B 14 Hoards of early Byzantine coins from barbaricum : finds of gold, silver and copper coins, mixed hoards of genuine Byzantine coins and imitations and Byzantine silver hexagrams and Sasa ni an silver drachms .............................. 568 B 15 Hoards of early Byzantine coins from barbaricum : archaeological context, grave finds, hoards found in ceramic or metal receptacles, hoards which include old coins a nd hoards only partially retrieved. ................................ ....... 568 B 16 Hoards of early Byzantine coins from barbaricum : last coin in the hoard. ........ 569

PAGE 10

10 LIST OF FIGURES Figure page 4 1 Early Byzantine frontier provinces and regions in barbaricum referred to in the text (with dotted line the former province of Dacia). ................................ ..... 86 4 2 "Non Roman" hand made pottery from the frontier provinces. ........................... 88 4 3 Amphora finds in barbaricum . ................................ ................................ ............ 97 4 4 "Da nubian" lamps. ................................ ................................ ........................... 103 4 5 Anatolian lamps and local imitations. ................................ ............................... 109 4 6 North African lamps and imitations.. ................................ ................................ 113 4 7 Palestinian lamps and imitations. ................................ ................................ .... 117 4 8 Handmade lamps.. ................................ ................................ ........................... 118 4 9 Metallur gical implements . ................................ ................................ ................ 128 4 10 Fibulae with bent stem. ................................ ................................ .................... 132 4 11 Cast fibulae with bent stem found in barbaricum and their typologi cal analogues in the frontier provinces. ................................ ................................ 141 4 12 Sixth seventh century "Slavic" bow fibulae in the frontier region and their parallels in barbaricum. ................................ ................................ ................... 150 4 13 Main types of Byzantine buckl es found in the frontier region.... .......... .. ........... 158 4 14 Distribution of bronze liturgical lamps, clay lamps with cross shaped handle or cross/Chi Rho on discus/nozzle, and Menas flasks. .......... ............. ......... ....... 170 4 15 Distribution of crosses and molds for the production of c rosses. ..................... 172 4 16 Main categories of Byzantine artifacts found in barbaricum ............................ 182 4 17 Early Byzantine lamps and imitations ................................ .............................. 188 4 18 Sixth to seventh century molds. ................................ ................................ ....... 189 4 19 S ixth to seventh century fibulae ................................ ................................ ...... 190 4 20 Byzanti ne artifacts in barbaricum and parallels from the Empire ..................... 191 5 1 Percent of nummia per year of reign (Justinian I and Justin II).. ....................... 235

PAGE 11

11 5 2 Percent of nummia per year of reign (Maurice and Heraclius). ......................... 236 5 3 Early Byzantine coin finds : major sites and local museums ............................. 237 5 4 Percentage of nummia per year of reign (498 616). ................................ ......... 238 5 5 Percent of nummia per year (538 616). ................................ ............................ 239 5 6 Coin s per year of reform ................................ ................................ ................... 240 5 7 Thessalonica mint (565 602) ................................ ................................ ............ 277 5 8 Nummia per year of reign. ................................ ................................ ................ 294 5 9 Nummia per year of reign. ................................ ................................ ................ 295 5 1 0 Mints (498 616).. ................................ ................................ .............................. 296 5 11 Mints (498 512).. ................................ ................................ .............................. 297 5 12 Mints (512 518).. ................................ ................................ .............................. 298 5 13 Mints (518 527).. ................................ ................................ .............................. 299 5 14 Mints (527 538 ).. ................................ ................................ .............................. 300 5 15 Mints (538 542). ................................ ................................ ............................... 301 5 16 Mints (542 550).. ................................ ................................ .............................. 302 5 1 7 Mints (550 565). ................................ ................................ ............................... 303 5 18 Mints (565 570).. ................................ ................................ .............................. 304 5 19 Mints (570 578).. ................................ ................................ .............................. 305 5 20 Mints (578 582).. ................................ ................................ .............................. 306 5 21 Mints (582 602).. ................................ ................................ .............................. 307 5 22 Mints (602 610).. ................................ ................................ .............................. 308 5 23 Mints (610 616). ................................ ................................ ............................... 309 5 24 Denominations (498 616). ................................ ................................ ................ 310 5 25 Denominations (498 512). ................................ ................................ ................ 311 5 26 Denominations (512 518). ................................ ................................ ................ 312

PAGE 12

12 5 27 Denominations (518 527) ................................ ................................ ................. 313 5 28 Denominations (527 538) ................................ ................................ ................. 314 5 29 Denominations (538 542).. ................................ ................................ ............... 315 5 30 D enominations (542 550) ................................ ................................ ................. 316 5 31 Denominations (550 565) ................................ ................................ ................. 317 5 32 Denominations (512 518 ................................ ................................ ................. 318 5 33 Denominations (578 58 2) ................................ ................................ ................. 319 5 34 Denominations (582 602) ................................ ................................ ................. 320 5 35 Denominations (602 610) ................................ ................................ ................. 321 5 36 Denominations (512 518) ................................ ................................ ................. 322 6 1 Sixth seventh century Byzantine coin finds in barbaricum. .............................. 327 6 2 Mints ................................ ................................ ................................ ................. 328 6 3 Denominations ................................ ................................ ................................ .. 328 6 4 Byzantine gold, silver, and bronze hoards in barbaricum. ................................ 329 6 5 Sixth seventh century Byzantine gold coins in barbaricum .............................. 329 6 6 Early Byzantine coin finds north of the Lower Danube. ................................ .... 336 6 7 Early Byzantine coin finds north of the Black Sea. ................................ ........... 337 6 8 The chronology of early Byzantine coin finds from the Danube frontier and barbaricum. ................................ ................................ ................................ ....... 338 6 9 Mints and denominations in barbaricum (Lower Danube and Black Sea). ....... 339 6 10 Mints and denominations in the Byzantine bridge heads (Sucidava, Drobe ta and Dierna). ................................ ................................ ................................ ...... 339 6 11 Early Byzantine Transcaucasia. ................................ ................................ ....... 373 6 12 Early Byzantine coin finds in Trancaucasia. ................................ ..................... 374 6 13 The chronology of early Byzantine coin finds from eastern Anatolia and Transcaucasia. ................................ ................................ ................................ 375

PAGE 13

13 6 14 Mints and denominations in Transcaucasia ................................ ...................... 376 6 15 Mints and denominations in eastern Anatolia (Amasia and Melitene). ............. 3 76 6 16 Early Byzantine coin finds in the Carpathian Basin. ................................ ......... 403 6 17 Early Byzantine coin finds north of the Middle and Upper Danube. .................. 404 6 18 Early Byzantine gold, silver, and copp er coin finds in the Carpathian Basin (491 680). ................................ ................................ ................................ ......... 405 6 19 First to third century mummy portraits wearing coin necklaces and coin pendants. ................................ ................................ ................................ .......... 463 6 20 Late nineteenth century Algerian woman and her d aughter wearing coin necklaces and late twentieth century Andeans dugging up the community treasure made of old foreign coins and banknotes ................................ .......... 464 6 21 Pierced Spanish Reales from the New World used in Madagascar .................. 465 6 22 Bracelet and necklace from Madagascar made of early modern European coins. ................................ ................................ ................................ ................ 466 6 23 Byzantine solidi of Heraclius mounted into rings found in Austria at Emling and in German y at Hausen and Gromehring and necklace with a mounted solidus from the joint reign of Justin I and Justin ian I fr om Dzhiginka, Russia . 467 6 24 Head jewelry from Madagascar made of early modern European coins dress jewelry made of seventh century Byzantine solidi from the Malo Pereschepyne hoard and c ut folles of Justi nian from Haskovo (Bulgaria) ..... 468 6 25 Early Byzantine gold, silver and bronze coins from Hungary, Serbia, Romania, and Georgia, pierced, cut, and mounted to be worn as pe ndants. ... 469 A 1 Early Byzantine coins in barbicum (6 th 7 th centuries) ................................ ........ 557

PAGE 14

14 LIST OF ABBREVIATIONS AA: American Anthropologist AAASH: Acta Archaeologica Academia e Scientiarum Hungaricae AAC: Acta Archaeologica Carpathica AB: Archaeologia Bulgarica AJN: American Journal of Numismatics AM: Arheologia Moldovei AMN: Acta Musei Napocensis ANS: Amer ican Numismatic Society Atlante : Carandini, A. ed. Atlante dll e forme ceramice, I. Ceramica fine romana nel Bacino Mediterraneo (medio e tardo Imperio). Rome: Istituto de lla enciclopedia Italiana, 1981 BAR Int.: British Archaeological Reports. International Series BASOR : Bulletin of the Americ an Schools of Orient al Research BASOR Suppl.: Bulletin of the American Schools of Oriental Research Supplemental Studies BMC : Wroth, W. Catalogue of the imperial Byzantine coins in the British Museum Vol. 1. London: Printed by order of the Trustees, 1908 BMGS : Byzantine and Modern Greek Studies BNP : Morrisson, C. Catalogue des monnaies byzantines de la Bibliothque Nationale (491 I er Justinien II (491 711) Pari s: Bibliothque Nationale, 1970 Broneer : Broneer, O. Terracotta Lamps. Co rinth, IV, Part II Cambridge, MA: Harvard University Press, 1930 BS NR: Numismatice Romne CAB: CN: DOC : Belling er, A. R., and P. Grierson. Catalogue of the Byzantine Coins in the Dumbarton Oaks C o llection and in the Whittemore C ollection Vol. I

PAGE 15

15 II. Washington, DC: Dumbarton Oaks Center for Byzantine Studies, 1966 1968 DOP : Dumbarton Oaks Papers EN : Ephemeris Napocensis FHG: F ragmenta Historicorum Graecorum Hayes: Hayes, J. W.. Excavations at Sarahane in Istanbul. Vol 2. The Pottery Princeton: Princeton University Press, 1992 IAI: Izvestiia na Arkheologicheskiia Institut Iconomu: Iconomu, C. romane arheologie, 1967 INMV: Izv estiia na Narodniia Muzei Varna INJ: Israel Numismatic Journal JRA: Journal of Roman Archaeology JRA Suppl.: Journal of Roman Archaeology. Supplementary Series JRS: Jou rn al of Roman Studies KOD: Althoff, R. Sammlung Khler Osbahr. Band V/1 2. Byzantinische Mnze n und ihr Umfeld. Duisburg, Kultur und Stad thistorisches Museum, 1998 1999 Kuzmanov: Kuzmanov, G. Antichni lampi. Kolekciia na Nacionalniia Arkheologicheski M uzei Sofia: Izdatelstvo na Bulgars kata Akademiia na Naukite, 1992 MAIET: Materialy po arkheolog ii, istorii i etnografii Tavrii MCA : Mater MGH: AA : Monumenta Germaniae Historica: Auctores Antiquissimi MGH: SRM: Monumenta Germaniae Historica : Scriptores Rerum Merovingicarum MGH: SS: Monumenta Germaniae Historica: Scriptores Rerum Germanicarum in Usum Schola rum Separatim Editi MIB: Hahn, W. Moneta Imperii Byzantini 3 vols. Vienna: Verlag der sterreichischen Akademi e der Wissenschaften, 1971 1981

PAGE 16

16 MIBE : Hahn, W., and M. Metlich, Money of the Incipient Byzantine Empire. Vienna : stereichische Forschungsgesel lschaft fr Numismatik am Institut fr Numismatik und Geldgeschichte, 2000 MIBEC : Hahn, W. Money of the Incipient Byzantine Empire Continued (Justin II Revolt of the Heraclii 565 610). Vienna: sterreichische Forschungsge sellschaft fr Numismatik, 2009 M iltner: Miltner F Das Cmeterium der Sieben Schlfer. Forschungen in Ephesos IV Baden: Rudolf M. Rohrer, 1937 Mnzfunde: Noeske, H. C. Mnzfunde aus gypten I. Die Mnzfunde des gyptischen Pilgerzentrums Abu Mina und die Vergleichsfunde aus den Dio cesen Aegyptus und Oriens vom 4. 8. Jh. n.Chr 3 vols. Berlin: Mann 2000 NC: Numismatic Chronicle NZ: Numismatische Zeitschrift PACT: Carcasonne, C., and T. Hackens, ed. Statistique et numismatique: table ronde organise par le Centre de mathmatiqu e sociale de l'Ecole des hautes tudes en sciences sociales de Paris et le Sminaire de Numismatique Marcel Hoc de l'Universit Catholique de Louvain, Paris, 17 19 sept. 1979 Strasbourg: Conseil de l'Europe Assemble parlementaire, 1981 Perlzweig: Perlz weig, J. The Athenian Agora. Lamps of the Roman Period VII. Princeton: P rinceton University Press, 1961 Ratto: Ratto, R. Lugano, 1930; reprint by J. Schulman. Amsterdam, 1959 REB: Revue des Etudes Byzantines RESS: Revue europenne des sciences sociales RIN: Rivista Italiana di Numismatica e Scienze Affini RN: Revue Numismatique Sabatier: Description gnrale des monnaies Byzantines. Paris: Chez Rollin et Feuardent, 1862 SCIV(A) : Studii SCN:

PAGE 17

17 Tolstoi: Vizantiiskiia monety Saint Pete rsburg: R. Galiksi A. Vilbori 1912 1914 Trsors: Morrisson, C., V. Popovic and V. Ivanisevic, ed. Les Trsors montaires byz antins des Balkans et d'Asie Mineure (491 713) Paris : Lethielleux, 2006 VV: Vizantiiski Vremennik

PAGE 18

18 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy MARGINAL MONEY: COINS, FRONTIERS, AND BARBARIANS IN EARLY BYZANTIUM (6TH 7TH CENTURIES) By Andrei Gandila August 2013 Chair: Florin Curta Major: History The dissertation explores the nature of the early Byzantine frontie r on the Lower Danube both from a historical and an archaeological perspective. The main thesis is that the frontier is a multifaceted concept which cannot be encapsulated into a single definition as most students of the frontier question have done so far. Instead, the same frontier can act both as a political/ military frontier of exclusion and as a cultural frontier where ideas, fashions, and people can circulate more or less freely. The nature and extent of cultural contact can be properly understood on ly through the study of Byzantine artifacts found beyond the political frontier. Rather than being a comprehensive treatment of all categories of imports, the dissertation addresses a selected number of diagnostic items which lend themselves to a versatile analysis due to their wide distribution, variety of types, and multiple implications in the realm of social, economic, and religious development of the frontier area and the regions beyond. Such artifacts are amphorae, lamps, jewelry, and coins. Among the se artifacts early Byzantine coins represent one of the most interesting, yet unexplored media whereby the Empire and barbaricum connected at multiple levels.

PAGE 19

19 The study of numism atic evidence clearly points to the fact that economic, cultural, and politic traced far beyond its administrative limits. One of the main arguments is that economic frontiers existed inside the Empire as well, due to economic fragmentation and different levels of pros perity and integration. The dissertation challenges some of the previous interpretations of early Byzantine coins found beyond the frontier by arguing that an ethnic interpretation of finds is problematic. Drawing on anthropological studies, I also argue t hat symbolic and ritual functions should be taken into account along with the economic role of coins, usually overemphasized in previous scholarship. At a deeper level, the dissertation aims to contribute to the methodology of using numismatic evidence as the primary source of a monograph which attempts to weave numismatics, archaeology, and history into a homogeneous narrative.

PAGE 20

20 CHAPTER 1 INTRODUCTION Andr Piganiol closed his Empire Chrtien with two sentences destined to become famous in the following decades: belle morte. Elle a t assassine 1 Half a century ago the frontier question was not really a question at all. The decline and fall of the Roman Empire was the dominant interpretation and had been in place since the days of Edward Gibbon. Frontiers loomed large in this fatalistic scenario according to which wave after wave of uncivilized barbarians managed to throw Europe into the Dark Ages. And yet there were important signs announcing a major his toriographic shift. The articulation of Late Antiquity as an independent field and an unprecedented turn towards the study of social and economic phenomena forever changed our understanding of ancient frontiers. The Anglo American school of Late Antique st udies, the French Annales School, and indeed, the creation and expansion of the European Union have brought a renewed interest in frontiers as well as new and exciting vistas for the study of cultural contact between different cultures and civilizations. However, only certain areas of the Roman frontier, which stretches on three continents, have been thoroughly researched in the past decades. This is certainly the case of western parts of the Roman Empire, which received great attention because of the nee d to reassess the transformation of the Roman world and the transition to the Middle Ages. The early Byzantine frontier in North Africa has long been a central focus of historians interested in the transition from the Vandal to the Byzantine, and later Ara b, domination. In the Near East our understanding of the role and function of the 1 A. Piganiol, 395) (Paris: Presses universitaires de France, 1947), 422.

PAGE 21

21 oriental frontier with Persia is still very much influenced by the seminal thesis of Benjamin Isaac and the subsequent works built on his concept of an open frontier. On the other hand, there is a conspicuous gap in English language scholarship on the evolution of the Lower Danube frontier in Late Antiquity, especially in what concerns the cultural contact between the Empire and barbaricum. With few exceptions, scholars have f ocused their attention on the military and political history of the region and archaeological research mostly targeted the Late Roman network of fortifications built to protect the provinces of the Balkans. Three major arguments develop from the main them es covered by this dissertation, the Early Byzantine frontier, the monetary economy of Early Byzantium, and the role of Byzantine money in barbaricum By looking at the Danube from a historical and archaeological perspective I argue that the frontier can b e perceived as a political and military border of exclusion, due to its strategic advantages as a natural obstacle, but also as a cultural frontier open to the circulation of ideas, objects, and people. Communities living beyond the frontier competed for a ccess to Byzantine goods and one of the main hypotheses is that several other cultural frontiers can be identified beyond the classical antithesis, Empire vs. "barbarians". Nevertheless, the circulation of ideas, fashions, and people across the Danube was closely related to the militarized nature of the frontier. In this case inclusion and exclusion are complementary rather than antithetical notions. Second, I argue that the monetary economy of Early Byzantium was far more diverse than previously envisaged and coin circulation in militarized frontier regions such as the Danube had less to do with the existence of markets and more with the needs of the garrisons. Finally, the analysis of coin finds

PAGE 22

22 reveals the non economic function of Byzantine money outside the Empire. The social value of Byzantine coins resides in their direct association with the Roman way of life and the emulation of Roman practices. Against previous interpretations, I will argue that coins had little or no monetary function. The overarchi ng goal of this dissertation is to contribute to the methodology of using numismatic evidence as the primary source of a monograph which attempts to weave numismatics, archaeology, and history into a homogeneous interdisciplinary narrative. The main object ives are threefold. The first is to bring into focus the Danube frontier in Late Antiquity, a region which remained largely unexplored by western scholarship until the last decade and to compare it with the frontier in the East in order to draw some concl usions about the role of the periphery in Early Byzantium The second is to employ a comparative framework for the study of the Early Byzantine monetary economy, by analyzing coin finds from the Balkans, Anatolia, and Syria Palestine. Finally, I will striv e to demonstrate that the circulation of Byzantine artifacts and especially coins found beyond the Danube frontier is connected with the multicultural nature of the Late Antique frontier and wa s driven by chiefly non economic motivations. Above everything this dissertation is an invitation to dialogue between archaeologists, anthropologists, historians, and numismatists. A common thread of all chapters is the realization that only by pulling together various strands of information, often the province of q uite diverse disciplines and specializations, can we build a nuanced and multifaceted narrative of frontier history. The reader will not find a brand new theoretical framework for analyzing frontiers and cultural contact but a long

PAGE 23

23 overdue fusion between c oncepts and definitions which often seemed mutually exclusive in previous scholarship. Indeed, t he frontier in its complexity and chameleonic nature defies total encapsulation by any one model. The sources which form the basis for discussion in the follow ing chapters were drawn from archaeology, written accounts, and ethnographic research, as well as from my own research in numismatic collections from various museums and the field work undertaken since 2001 in the Late Antique Capidava, an important fortre ss on the Lower Danube frontier. The nature of the evidence characterized by a slender body of texts about barbaricum and incomplete or sometimes non existent archaeological evidence forced me to change the geographic scope of each chapter depending on the major themes addressed. More abundant sources concerning the frontier policies of Early Byzantium invites broader parallels between the Lower Danube frontier and more distant peripheral regions such as Arabia or Transcaucasia. To the contrary, the archaeo logical record is much more diverse in quantity, quality, and indeed regarding the methods and questions which concerned various archaeological schools. The cultural diversity is all the more baffling as the main concern here has been the study of societie s living outside the Empire, which did not benefit from the relative institutional standardization of Byzantine settlements. On the other hand, coins are perhaps the most standardized and well distributed Byzantine artifact and this fact alone afforded a c omparison between distant regions such as the Carpathian Basin, the Lower Danube, the Northern Black Sea steppe and Transcaucasia.

PAGE 24

24 In chronological terms the dissertation covers the final period of Late Antiquity or the sptere Sptantike as Peter Brown t ermed it. 2 More exactly it covers the period from the late fifth century to the late seventh century. The weight of the dissertation rests on the long sixth century from the accession of Anastasius to the reign of Heraclius, which roughly corresponds to the renewal of the Byzantine frontier on the Danube and its ultimate demise, respectively. To be sure, both dates are flanked by important transitional periods. The historical framework of the project begins with the dissolution of the Hunnic power in the Danube region followed by the recovery of the Byzantine Empire in the Balkans reaching its zenith under Justinian. At the other end we witness a more dramatic transition marked by the de facto retreat of the Empire from the northern Balkans during the rei gn of Heraclius and the important cultural transformations preceding the creation of the Bulgar state in 680. The dissertation has a thematic structure. The second chapter follows the parallel evolution of Byzantine history, archaeology, and numismatics si nce their antiquarian beginnings in the Age of Enlightenment. The main argument is that numismatics rarely converged with other disciplines as historians perceived numismatics as an arcane field, while numismatists rarely ventured out of their narrow field of enquiry. The accumulation of material at unprecedented speed in the past few decades makes this a pressing matter. Although it would be pointless to belabor differences of approach resulting from material unavailable to earlier scholars, the methodolog ical rift between disciplines is a long standing issue which deserves highlighting in order to identify adequate solutions in the future. Indeed, the social and economic history of E arly Byzantium as reflected in 2 P. Brown, "SO Debate: The World of Late Antiquity Revisited," Symbolae Osloenses 72, no. 1, (1997): 28.

PAGE 25

25 handbooks and treatises can benefit from th e inclusion of numismatic evidence in the larger narrative, something that still remains a desideratum. The third chapter is a reinterpretation of the Early Byzantine frontier. The historiography of the past few decades has described frontier rivers as pri marily facilitating communication and cultural contact, and less as borders of exclusion. On the other hand archaeologists continue to emphasize the military dimension of frontiers. To be su re, this is a direct result of the lack of proper conversation bet ween historians and archaeologists. Under the spell of the postmodernist turn the former have emphasized the cultural, intellectual and symbolic dimensions of frontiers, while the latter remained entrenched in the traditional focus on fortifications and l ines of defense. The chapter is an attempt to re emphasize the strategic role of frontier rivers such as the Danube during Late Antiquity. By doing so it is necessary to intrude upon several of the major and still outstanding questions of Byzantine history What was the Byzantine worldview regarding frontiers? How was it different from the early Roman period? A careful study of accounts referring to various frontier regions of the Empire reveals enduring literary topoi but also a n ongoing concern to use riv ers as political frontiers able to act as convenient barriers against "barbarians". The fourth chapter is an archaeological interpretation of the Danube frontier as an interface between the Empire and barbaricum based on a variety of Byzantine goods found outside the Empire. Exhaustiveness is neither claimed nor achieved. The chapter discusses a selected number of diagnostic items that lend themselves to a multifaceted analysis due to their wide distribution, variety of types, and multiple implications in the realm of social, economic, and religious realities of the frontier area and the regions

PAGE 26

26 beyond. The study of archaeolog ical evidence clearly points to the fact that economic, ce can be traced far beyond its administrative limits. The surge of Byzantine artifacts across the frontier, such as amphorae, lamps, brooches, and buckles, points to different channels of distribution and particular preferences associated with the creatio n of identity and social prestige in relation to Byzantium. The distribution of Byzantine artifacts north of the Danube substantiates the claim that the river was not the only frontier. Consequently, communities living in the shadow of the Empire should no t be lumped together under the broad label "barbarians". The fifth chapter is a reinterpretation of the role of money in the Early Byzantine economy, by demonstrating that current views have been shaped by parallel developments in anthropology and economi c history half a century ago. Drawing on language scholarship, with Michael Hendy being its most prominent advocate. Although the role of the state in the distribution of coin and the regulation of the monetary economy cannot be denied, this chapter takes a more balanced approach, one of the main contentions being that some of the tools and theoretical underpinnings of formal economics can be applied to the study of ancient economic systems. Working under the assumption of a relative uniformity, scholars have drawn sweeping conclusions regarding the monetary economy of Byzantium. In reality, a close comparative analysis of coin circulation in the Balkans, Anatolia, and the N ear East reveals a great degree of variation, which also informs the way in which we should address the flow of Byzantine coins beyond the frontier Local developments rather than a standardized policy from

PAGE 27

27 Constantinople dictated the evolution of the mone tary economy. Unlike large coastal settlements on the Mediterranean and the Black Sea, frontier settlements were not dominated by a market economy, in the Late Antique version of the concept. Coin circulation in frontier fortresses was closely linked wit h the presence of garrisons who received regular payments. The sixth chapter is an attempt to re evaluate the function of Early Byzantine coins beyond the frontier by analyzing their distribution on a wide geographic area from Central Europe to the Caucas us. While precious metal coins have been connected with political payments, current interpretations of Byzantine copper coins found outside the frontier are chiefly economic. Given the fiduciary nature of Byzantine bronze coins the question therefore inelu ctably arises as to how they could act as monetary media of exchange outside the Empire. Previous arguments have been couched in preconceived notions regarding the Early Byzantine monetary economy, on one hand, and the tendency to view parts of barbaricum as following the same conditions applicable to the Empire on the other Drawing on anthropological parallels for the use of monetary instruments by primitive societies I argue that coins served mainly non economic purposes. From an economic, but non mon etary, perspective coins were more attractive for their intrinsic value as raw material for the production of jewelry. As one moves farther from the border, the non economic value of coins as amulets, souvenirs, and objects of prestige increases. A note o n terminology is required Historical periodization can be notoriously confusing and the Late Antique period makes no exception. Depending on region, language, and intellectual tradition Roman, Late Roman, Early Byzantine or Byzantine

PAGE 28

28 may be used as labels for the period covering the sixth and seventh centuries. Archaeological terms such as Roman Byzantine or even Late Byzantine in Near Eastern archaeology, Early Medieval or Late Migration for cultures outside the Empire only add to the general confusion to which the non specialist may easily succumb. In the following chapters only Roman and Early Byzantine will be used, the former for general cultural characteristics (e.g. Roman way of life, Roman tradition) and the latter for chronologically sensitive cont exts and in relation to the fact that sixth seventh century coinage is universally described as Early Byzantine. Latin terminology has been employed for names or the English equivalent long established in scholarly works (e.g. Anastasius, Justinian) and La tin or Greek terms for coinage, reflecting the evolution of the terminology during Late Antiquity (e.g. solidus hexagrammon ). Copper will be used most frequently to designate the low value Byzantine coinage, although the metal itself is a copper alloy, so metimes described as "bronze" in the numismatic literature, where the terms are used interchangeably. For the sake of brevity, barbaricum will be used to designate regions outside the Empire, although no cultural uniformity must be expected. Throughout the dissertation regions beyond the frontier are sometimes labeled "Gepidia," "Avaria," or "Sklavinia," but their boundaries are hard to define in the ever changing world of barbaricum. On the other hand, the term "barbarians" will be used sparingly and only in abstract contexts, lest the reader be left with the impression that various ethnic groups may be lumped together under the same cultural umbrella. Finally, modern g eographic terminology had to be brought to a common denominator by dropping the term desi gnating the administrative unit (e.g. Jude

PAGE 29

29 given the different administrative division of the countries covered by the inventory of coin finds.

PAGE 30

30 CHAPTER 2 ARCHAEOLOGY, AND NUMISMATIC S 2 .1 A Circuitous Road: Byzantine History, Archaeology, and Numismatics in Western Scholarship from Du Cange to World War 2: The incorporation of numismatic evidence as a meaningful element of the historical narrative about early Byzantium is a more recen t development of the last few decades. The wealth of numismatic material brought to light by archaeological excavations conducted in various parts of the early Byzantine Empire in the past decades has enriched the existing corpus of early Byzantine coins a nd attracted unexplored areas and much unpublished material, for Asia Minor in particular, economic historians can no longer ignore the numismatic evidence as an indication o f how the Empire functioned during the long sixth century and how it interacted with populations and polities beyond its frontiers. However, many historians who use material culture still feel uncomfortable using the numismatic evidence and fitting it into the larger narrative. What is at stake here is the dialogue between numismatists, archaeologists, and historians, on one side, and developments within each discipline, on the other. The interest in Byzantine economic studies, where a numismatic discuss ion is likely to fit, is essentially a post war development, although older general histories of Byzantium often included a chapter on economy, trade, and finance. The branch of Byzantine numismatics itself underwent a slow development and the antiquarian phase was left behind only in the twentieth century. The interest in Byzantine coins as currency, i.e. as an instrument of the Byzantine economy and administration, grew around the time when the historical discipline itself turned from a focus on political narrative to social and

PAGE 31

31 economic developments and, more importantly, with the publication of the first major archaeological reports. Therefore, in order to explain the marginal role of money in the historiography of the early Byzantine period one needs to go back in time to the very beginning of the discipline and follow the parallel developments of Byzantine history, archaeology, and numismatics. Western scholarship and general histories of Byzantium have to be treated separately from national scholarship in countries whose territory was once part of the early Byzantine Empire or whose relation with Byzantium has a bearing on the national history, such as Hungary, Romania or Russia, to name a few. Surprisingly, the first true Byzantinist, Charles Du Cange (1610 1688), had an interest in coins and among his numerous works there is also a valuable numismatic treatise, De imperatorum constantinopolitanorum seu inferioris aevi vel imperii uti vocant numismatibus first included in his dictionary of Medieval Lat in, 1 and later published as a separate work. 2 The Age of Enlightenment, unfortunately, did not bring a real progress in the field and the example set by Du Cange, who approached Byzantine history using various tools such as genealogy and numismatics, was n ot readily followed. The eighteenth century was instead dominated by a derogatory view of Byzantium as a decadent society described by Voltaire as an "histoire horrible et dgoutante de brigands obscures," 3 and in quite similar terms by Montesquieu and lat er 1 C. Du Fresne Du Cange, Glossarium ad scriptores mediae et infimae Latinitatis, in quo latina vocabula novatae significationis, aut usus rarioris, bar bara et exotica explicantur, eorum notiones et originationes reteguntur (Paris: Louis Billaine, 1678). 2 C. Du Fresne Du Cange, De imperatorum constantinopolitanorvm seu inferioris aevi vel imperii uti vocant numismatibus (Rom: Typis J. Mari Salvioni, 17 55). 3 Cited by G. I. Br tianu, Etudes byzantines d'histoire conomique et sociale (Paris: Geuthner, 1938), 25.

PAGE 32

32 by Hegel. 4 Decline is what Gibbon himself was working on his monumental history. Out of six volumes there is barely one paragraph about the reorganization of mints and a footnote on the exchange rate between silver and gold. 5 Economy itself has only a m arginal place in his work. Clearly Gibbon's interests lay elsewhere and he turned out to be extremely influential during the nineteenth century, his work being translated into most European languages. Nonetheless, as part of the encyclopedic fervor dominat ing the Age of Reason, a number of important contributions were made to the study of ancient and medieval numismatics, Byzantine coins being included in the larger frame of ancient coinage in the works of Anselmo Banduri, Girolano Tanini, and Joseph von Ec khel. 6 The nineteenth century saw the first major catalogues of Byzantine coins. French scholarship has been in the forefront of these important developments. N. D. Marchant began to reassess some of the attributions previously made by Du Cange and Domenic o Sestini and also added a number of new types followed by historical commentary. 7 However, the first attempt at classifying the Byzantine series we owe to Flicien de Saulcy, who was also the first to suggest that the study of Byzantine coinage 4 A. A. Vasiliev, History of the Byzantine Empire 324 1453 vol 1 (Madison: University of Wisconsin Press, 1961), 6 7. 5 E. Gibbon, The History of the Decline and Fall of the Roman Empire vol. 1, ed. David Womersley (New York: Penguin, 1994), 636, n. 178, and vol. 3, 993. That innovation and not decline is what characterizes the Late Roman monetary economy, is a realization reached only by Gunnar Mickwitz in the interwa r period, G. Mickwitz, Geld und Wirtschaft im rmischen Reich des vierten Jahrhunderts n. Chr (Leipzig: Harrassowitz, 1932). 6 G. Tanini, Numismatum imperatorum romanorum a Traiano Decio ad Constantinum Draconem : ab Anselmo Bandurio editorum supplementu m ( Rome: A. Fulgonium, 1791); A. Banduri, Numismata imperatorum Romanorum a Trajano Decio ad Palaeologos Augustos (Paris: Luteti, 1718); J. H. von Eckhel, Doctrina numorum veterum 9 vols. (Vienna: Sumpt. J.V. Degen, 1792 1798). 7 N. D. Marchant, Mlange s de numismatique et d'histoire (Metz: Lamort, 1818) ; D. Sestini, Lettere e dissertazione numismatiche sopra alcune medaglie rare della collezione Ainslieana (Livorno: Stam T. Masi, 1789).

PAGE 33

33 should be gin with the Anastasian reform of 498. 8 The beginning of the Byzantine period in numismatic terms is thus very different from the historical one, which usually placed the beginning of the Byzantine age proper either with the reign of Constantine and the fo unding of Constantinople, or sometime in the seventh century, when the imperial administration collapsed in the Balkans and the Near East. 9 However, Anastasius' momentous reform, which was almost never left unnoticed in general histories of Byzantium, as w e shall see, remains to this day the standard numismatic date for the beginning of the Byzantine period. This demonstrates once again the great complexity of this historical period, presently known under the generic name Late Antiquity, and also the differ ence between historians and numismatists, who often operate with distinct periodizations. The nineteenth century is also the time when large collections began to be assembled. Byzantine numismatics, like archaeology, was still in its antiquarian phase. Th e collection of M. Soleirol formed the basis of the general catalogue previously published by de Saulcy, but the most important work in this respect would become Justin Sabatier's Description gnrale des monnaies Byzantines (1862), which remained the stan dard catalogue for almost half a century. 10 This is also the first attempt at a comprehensive study, since Sabatier did not just rely on his personal collection, or even French collections alone, but also consulted numismatic collections 8 F. de Saulcy, Essai de classification des suites montaires Byza ntines (Metz: Lamort, 1836), xi. 9 J. C. Royou, Histoire du Bas Empire, depuis Constantin jusqu' la prise de Constantinople, en 1453 2nd ed. (Paris: Chez L'Auteur, 1814); G. Finlay, A History of Greece from its Conquest by the Romans to the Present Tim e B.C. 146 to AD 1864, vol. 1, Greece under the Romans B.C. 146 A.D. 716 2nd ed. (Oxford: Clarendon Press, 1877), 354. 10 M. Soleirol, Catalogue des monnaies byzantines qui composent la collection de M. Soleirol (Metz: Lamort, 1853) ; J. Sabatier, Descri ption gnrale des monnaies Byzantines (Paris: Chez Rollin et Feuardent, 1862).

PAGE 34

34 from London, Vienna Berlin, and Copenhagen. Sabatier arranged the currency of the Byzantine period in his catalogue by "metals" and also included market prices, which bespeaks his intention to facilitate its use by collectors. Nevertheless, Sabatier also included a historic al commentary with references to works such as M. G. Finlay's Greece under the Romans (1 st Historie de Bas Empire and clearly alluded to the necessity of using the numismatic evidence for a better understa nding of Byzantium's general history. 11 Within this frame it does not seem surprising that Sabatier ignored de Saulcy's decision to start the Byzantine series with Anastasius and instead chose to go back to the reign of Arcadius, arguing that the division o f the Roman Empire after the death of Theodosius I should be taken as the moment when Byzantine history starts. In other words, Sabatier deliberately rejected the perfectly plausible numismatic division between Roman and Byzantine made by de Saulcy in favo r of a purely historical chronology. Nineteenth century scholars varied when it came to assigning the early Byzantine period either to antiquity or to the Middle Ages. T. E. Mionnet included the Byzantine series at the end of his popular De la raret et d u prix des mdailles romaines 12 which went through several editions during the nineteenth century, while the well known Danish scholar Christian Jrgensen Thomsen published the Byzantine coins from his large collection of ancient and medieval coins in the second part entitled Les monnaies 11 L. P. Comte de Sgur, Histoire de Bas Empire: e de Constantinople 7th ed., v.1 (Paris: Didier, Libraire diteur, 1943). 12 T. E. Mionnet De la raret et du prix des mdailles Romaines, ou recueil contenant les types rares et indits des mdailles d'or, d'argent et de bronze frappes pendant la dure de la Rpublique et de l'Empire Romain 3rd ed. (Paris: Auguste Aubry, 1847).

PAGE 35

35 de Moyen Age. 13 George Finlay, whose History of Greece proved to be extremely influential in the nineteenth century, regarded it as a gradual process of transformation, beginning with the reign of Heraclius and ending with the accession of Leo the Isaurian (A.D. 717). However, the transformation continued at different levels and a purely Greek monarchy was established as late as the reign of Alexios I Comnenos. What is interesting about Finlay's argument is the fact that it relies on coinage. In this respect Finlay supersedes even twentieth century historians and it should be noted that Finlay embarked on a much larger task than the history of Byzantium, his massive seven volume work covering the history of Greece from the Ro man conquest to 1864. Finlay argued for a transition from the Roman (Latin) to the Byzantine (Greek) period by looking at the way coin legends evolved in time. He noticed that Anastasius introduced Greek numerals as indications of value on his copper curre ncy and thus began a slow and gradual process whereby Latin was replaced by Greek on the imperial coinage. This process finally came to an end with the accession of Alexios I, described by Finlay as the first Emperor of the East who was entirely Greek. Fin lay's knowledge of numismatics was indeed remarkable. 14 The reader of his 30 page appendix on coinage will soon find out that Finlay had a good command of the numismatic literature with references to all the important works of the previous centuries, Du Can ge, Tanini, Marchant, de Saulcy, Sabatier, and even the special study of Friedlnder and Pinder on 13 C. Jrgensen Thomsen, Catalogue de la collection de monnaies de feu Christian Jrgensen Thomsen. Seconde partie: Les monnaies du moyen ge tome I ed. K. Erslev ( Copenhagen: Imprimerie de Thiele, 1873). 14 Some of his academic interest in monetary history might have derived from the fact that he was also a collector and he even donated a few interesting Byzantine pieces to the British Museum, Finlay, A History of Greece 445 46, n. 6.

PAGE 36

36 the coinage of Justinian I. 15 He also displayed a very good understanding of economic phenomena seen through coinage, such as the debasement of the copper coi nage during the sixth century and the importance of the changing exchange rate between the three metals, gold, silver, and copper. On the continent, the great Byzantine scholar of the late nineteenth century was Gustave Schlumberger, who made major contri butions in the fields of Byzantine history, archaeology, numismatics, and sigillography. 16 Although he was mostly interested in the later Byzantine period, his scholarly work is a fine example of multidisciplinary preoccupations. However, Schlumberger is an exception and most historians were quite unaware of developments in the field of Byzantine numismatics. 17 Charles Diehl, who wrote an important history of the Justinianic age, does not comment on the monetary policy of the Empire in the sixth century, alth ough he had some interest in economic history. He discusses commerce in particular, but only long distance trade in luxury items, which is best documented in written sources such as Cosmas Indicopleustes. 18 In 1906 Warwick Wroth, curator at the British Mus eum, was deploring the lack of dialogue between Byzantine historians and numismatists: 15 J. Friedlnder and M. Pinder, Die Mnzen Justinians (Berlin: Nicolaischen Buchhandlung, 1843). 16 G. L. Schlumberger, Les principauts franques du Levant d'aprs les plus rcentes dcouvertes de la numismatique (Paris: E. Leroux, 1877); G. L. Sc hlumberger, Numismatique de l'Orient latin (Paris: E. Leroux, 1878); G. L. Schlumberger, Sigillographie de l'Empire byzantin (Paris: E. Leroux, 1884); G. L. Schlumberger, Mlanges d'archologie Byzantine. Monnaies, mdailles, mreaux, jetons, amulettes, bu lles d'or et de plomb, poids de verre et de bronze, ivoires, objets d'orfvrerie, bagues, reliquaires, etc. ( Paris: E. Leroux, 1895). 17 Another major exception is Karl Krumbacher whose comprehensive bibliographies on various disciplines dealing with the By zantine period included an excellent section on Byzantine numismatics, both general and special studies since the time of Du Cange, for which see. K. Krumbacher, Geschichte der byzantinischen Litteratur von Justinian bis zum Ende des ostrmischen Reiches, vol. 2, 2nd ed. (Munich: Beck, 1897; reprint New York: Burt Franklin, 1958), 1128 32. 18 C. Diehl, Justinien et la civilisation byzantine au VI e sicle 2 vols, 1st ed., 1901 (New York: Burt Franklin, 1968), 533 45.

PAGE 37

37 In the case of Byzantine numismatics, on the other hand, the present number of workers is infinitesimal, and it is hard to find the ideal union of numismatic experien ce and of general Byzantine scholarship. 19 Wroth himself embarked on the important task of publishing the large collection of Byzantine coins in the British Museum. The catalogue published in 1906 largely superseded the work of Sabatier and remained the s tandard reference for more than half a century. 20 This effort, however, remained without any significant echo in scholarly histories of Byzantium written in the following decades. Nevertheless, developments in the field of Byzantine numismatics are visible in the work of J. B. Bury, the author of one of the first histories of the Late Roman Empire. Much like Sabatier, Bury chose to begin his history with the reign of Arcadius. The influence of Byzantine numismatic studies, still in their infancy, was limited since Bury did not discuss monetary history at all in his initial work on the Later Roman period, published in 1889. 21 A much improved edition appeared in 1923 when Bury made room for a short numismatic discussion of the Anastasian reform of 498 and the e xpansion of the mint system under Justinian, for which he cited Wroth's catalogue. 22 It appeared that historians would no longer shy 19 W. Wroth, "The Study of Byzantine Numis matics," in Corolla Numismatica, Numismatic Essays in Honour of Barclay V. Head ed. G. F. Hill (London, New York: Oxford University Press, 1906), 335. For of great use. We retain his request for a new and updated standard work on Byzantine numismatics, "vraiment 1905," in tudes Byzantines (Paris: Al phonse Picard et fils, 1905), 96. See also the earlier agenda of S. P. Lambros, "Byzantinische Desiderata," BZ 1 (1892): 185 201, esp. 193 94. 20 W. Wroth, Catalogue of the Imperial Byzantine Coins in the British Museum vol. 1. London, 1908. s handbook would also serve as an additional resource for scholars and collectors alike, see H. G. Goodacre, A Handbook of the Coinage of the Byzantine Empire 3 vols. (London: Spink, 1928 1933). 21 J. B. Bury, A History of the Later Roman Empire, from Arc adius to Irene (395 A.D. to 800 A.D.) 2 vols. (London/ New York: Macmillan, 1889). 22 J. B. Bury, History of the Later Roman Empire from the Death of Theodosius I. to the Death of Justinian (A.D. 395 to A.D. 565) 2 vols. (London/ New York: Macmillan, 1923 ), 357 with n. 5, 446 47.

PAGE 38

38 away from a numismatic incursion into the economic history of the early Byzantine period. I. I. Tolstoy's catalogue publish ed between 1912 and 1914 added to the slowly growing body of western scholarship. His splendid collection became a major reference for Byzantine numismatic studies in the first half of the twentieth century, although his decision to publish it in Russian r endered it somewhat inaccessible to western scholars. Nevertheless, it is surprising that A. A. Vasiliev did not even mention Tolstoy in the otherwise useful historiographical introduction to his History of the Byzantine Empire 23 Although he included a det ailed section on Byzantine studies in Russia, Vasiliev made no reference of Tolstoy's contributions in the field of Byzantine history, archaeology, and numismatics. 24 Moreover, his section on Byzantine periodicals and general references does not mention any works on numismatics or sigillography, although a good number of important works had by then already been published. 25 The first decades of the twentieth century also witnessed the first important publications of excavation finds by western archaeological teams in Greece, Turkey, and Palestine. The excavations at Constantinople, 26 Corinth, 27 Sardis, 28 Ephesus, 29 23 A. A. Vasiliev, History of the Byzantine Empire. I. From Constantine the Great to the Epoch of the Crusades (A.D. 1081 ) (Madison: University of Wisconsin Press, 1928). 24 For a detailed biography of Tolstoy, see V. V. Guruleva's introduction to I. I. Tolstoy, Byzantine Coins, 10th issue (Bernaul: Den', 1991), i xvi. 25 branches of Byzantine studies, including numismatics; see Vasiliev, H istory 42. 26 A. H. M. Jones, "The Coins," in Preliminary Report upon the Excavations Carried out in the Hippodrome of Constantinople in 1927 (London: Oxford University Press, 1928), 46 50. 27 A. R. Bellinger, Catalogue of the coins found at Corinth, 1925 ( New Haven: Yale University Press, 1930); K. M. Edwards, Corinth VI: Coins, 1896 1929 (Cambridge, MA: Harvard University Press, 1933). 28 H. W. Bell, Sardis vol. XI, Coins, part I, 1910 1914 (Leiden: Brill, 1916).

PAGE 39

39 Priene, 30 Pergamum, 31 Jerusalem, 32 Scythopolis, 33 and Gerasa 34 yielded an important number of coins. However, only occasionally did publications include a h istorical interpretation of coin finds. In the introduction to the catalogue of coins found in Corinth in 1925, Alfred Bellinger pleaded for the historical significance of coins found during excavations arguing that "coins reflect the life of the common pe ople, which is generally too vulgar to have been noticed in the formal histories." 35 The publication of hoards witnessed a rapid development in the interwar period and the most important contribution is undoubtedly Sawyer McA. Mosser's attempt at a corpus o f Byzantine coin hoards. His work constituted a great research instrument above anything else, since the historical interpretation was almost lacking, the study consisting of an inventory followed by bibliographic references. 36 Nevertheless, his most import ant achievement was that of bringing together a large body of material scattered in dozens of local journals, less accessible to western scholars. The works of Bury and Vasiliev, mentioned above, were just two of the general scholarly histories of Byzantiu m published in the first decades of the twentieth century. 29 J. G. hesus," NC (1925): 385 91. 30 K. Regling, Die Mnzen von Priene (Berlin: Mann, 1927). 31 K. Regling, "Mnzfunde," in Altertmer von Pergamon: Stadt und Landschaft ed. A. Conze, vol. I, part 2 (Berlin: Reimer, 1912), 355 63. 32 G. M. Fitzgerald, "The coins," in Excavations in the Tyropoeon Valley, Jerusalem 1927 ed. J. W. Crowfoot and G. M. Fitzgerald (London: Palestine Exploration Fund, 1929), 103 32. 33 G. M. Fitzgerald Beth Shan Excavations 1921 1923. The Arab and Byzantine Levels (Philadelphia: Penn Uni versity Press, 1931). 34 A. R. Bellinger, Coins from Jerash, 1928 1934 (New York: American Numismatic Society, 1938). 35 Bellinger, Catalogue of the coins vii. 36 S. McA. Mosser, A Bibliography of Byzantine Coin Hoards Numismatic Notes and Monographs 67 ( New York: American Numismatic Society, 1935).

PAGE 40

40 Historians such as Diehl, Nicola Turchi, and Jan Romein approached the Byzantine period from different angles, and although economic history is not completely lacking, there is almost nothing about monetary economy. 37 In addition to scholarly works, it is perhaps worth mentioning that histories of Byzantium written for a general audience would not even discuss economic issues, political narrative being their major focus. 38 That historians did not have a strong foundation to rely on for a numismatic discussion can bee seen in their frequent appeals to presentist comparisons. R. Byron did not include a detailed discussion of the Byzantine monetary system in his Byzantine Achievement but instead he argue d that "the Byzantines conducted a finance of almost modern dimensions." 39 Along similar lines, A. Andreades, in one of the first attempts at discussing price levels and their evolution in Byzantium, compared wheat prices and interest rates with those of mo dern Greece. 40 Despite this climate, one has to look with appreciation at Fritz Heichelheim 's massive ancient economic history. Written in a heavy German style and somewhat lacking the proper structure for such an ample undertaking, Heichelheim's work has often been ignored. However, his detailed section on money, "Geldverhltnisse, Zins und Kapital," 41 included many insightful comments 37 C. Diehl, and G. Marcais, Histoire de Moyen Age. Tome II: le monde oriental de 395 a 1081 (Paris: Presses Universitaires de France, 1936); J. Romein, Byzantium; Geschiedkundig overzicht van staat en beschavin g in het oost Romeinsche Rijk (Zutphen: W. J. Thieme & CIE, 1928); N. Turchi, La civilt bizantina (Torino: Fratelli Bocca, 1915). 38 B. Diener, Imperial Byzantium (Boston: Little, Brown & co, 1938); A. Bailly, Byzance (Paris: Artheme Fayard, 1939). 39 R. Byron, The Byzantine Achievement 1st ed., 1929 (London: Routledge, 1987), 56. 40 A. Andreades, "De la monnaie e de la puissance d'achat des mtaux prcieux dans l'empire byzantine," Byzantion (1924): 7 50. 41 F. M. Heichelheim, Wirtschaftsgeschichte des Alt ertums vom Palolithikum bis zur Vlkerwanderung der Germanen, Slaven und Araber (Leiden: Sijthoff, 1938), 774 96.

PAGE 41

41 on salaries, prices, and inflation, relying especially on Egyptian papyri, the bread and butter of many present discussions about the role of money in early Byzantium. 42 2 .2 Post W ar Developments: Searching for Common Ground Works on Byzantine history, archaeology, and numismatics increased with astonishing rapidity in the decades after the war. V. Laurent's Bulletin de numism atique byzantine is perhaps the best reflection of the early developments. Laurent went far beyond the scope of a purely historiographical update of works in the field of Byzantine numismatics. Divided into nine sections, his study offered a complete pictu re of the problems touched by the discipline, such as the historical significance of coins, questions of metrology, imitations, hoards and hoarding behavior, and circulation inside and outside the Empire's borders. Even more important, his extensive biblio graphies arranged by topic gathered less accessible publications from Eastern Europe as well as reports of recent finds. 43 In a sense, this tour de force came as a response to Gyula Moravcsik's earlier appeal, urging the collection and publication of numism atic and sigillographic material as important sources for Byzantine history. 44 By the mid 1950s the great Hungarian Byzantinist was already noticing an important shift in Byzantine studies from an emphasis on political history towards a greater interest in social and economic relations. 45 Ernst Stein's Histoire du Bas Empire was part and parcel of this 42 Two recent examples are J. Banaji, Agrarian Change in Late Antiquity: Gold, Labour, and Aristocratic Dominance (Oxford/New York: Oxford Uni versity Press, 2001), and P. Sarris, Economy and Society in the Age of Justinian (Cambridge/New York: Cambridge University Press, 2006). 43 V. Laurent, "Bulletin de numismatique byzantine (1940 1949)," REB 9 (1951): 192 251. 44 G. Moravcsik, "Les tches act uelles de la byzantinologie," Byzantinoslavica 10 (1949): 1 10. 45 G. Moravcsik, "L'tat et les tches de la byzantinologie," Byzantinoslavica 16 (1955): 17, "Le point de sociale)." See also the collection of studies edited by N. Baynes and H. Moss which included contributions on economic

PAGE 42

42 new tendency. His treatment of Byzantine monetary issues in the sixth century is, however, restricted to the policy conducted by Peter Barsymes and the effects of the depreciation of the copper coinage. 46 To his credit, Stein cited not only Wroth but also the Ratto sale catalogue, 47 which remains to date the most important single sale of Byzantine copper coins and whose dissemination in the 1930s had greatly incre ased the popularity of the Byzantine coin series. George Ostrogorsky published in 1940 his Geschichte des byzantinischen Staates which went through several revised English editions in the following decades and soon became one of the standard histories of Byzantium. In the introduction to the second English edition he advocated the importance of numismatic and sigillographic sources: The evidence of coins and seals has much to contribute towards the solution of all kinds of problems, especially those relat ing to the internal history of Byzantium, and is indeed often decisive. Numismatic material is indispensable for any study of the currency and economic history on the one hand, and of the symbolic representation of imperial authority on the other. [...] Al l the same, in spite of remarkable progress, general knowledge of this important source material does indeed lag far behind the needs of scholarship. 48 However, Ostrogorsky was mostly interested in political history, so that the section on the Late Roman p eriod, 324 601, contains little information about economy, while the life, currency, budget, and administration, N. Baynes and H. Moss, ed., Byzantium; An Introduction to East Roman Civilization (Oxford: Clarendon Press, 1948). 46 E. Stein, Histoire du Bas Empire II De la disparition de l'empire d'occident la mort de Justinien (476 565) (Paris/ Bruxelles/Amsterdam: Desclee de Brouwer, 1949), 766 69. 47 R. Ratto, (Lugano, 1930; reprint by J. Schulman, Amsterdam, 1959). 48 G. Ostrogorsky, History of the Byzantine State (New Bru nswick, N.J.: Rutgers University Press, 1969) 16 7, and 26 7, for a bibliography of major works on Byzantine numismatics since Sabatier.

PAGE 43

43 monetary history itself is covered by a short description of the Anastasian reform of 498 and its general implications. 49 Ostrogorsky's pessimistic remark about the branch of Byzantine num ismatics lagging far behind the needs of scholarship is yet another example of the lack of dialogue between historians and numismatists. In 1968 Joel Malter was able to assemble a bibliography of c. 300 titles of works in the field of Byzantine numismatics published between 1950 and 1965. 50 Without being comprehensive, his effort shows the extent to which interest in this topic developed after World War II. Much of this interest was due to a parallel development in Late Roman and Early Byzantine archaeology. An ever increasing number of reports and monographs were being published, many of which included special sections on coin finds, and even book length studies, which was the case for large metropoleis like Antioch, Athens, and Sardis. 51 In a series of semin al early articles, Philip Grierson, arguably the greatest Byzantine numismatist of the last century, laid out the benefits as well as limitations of coins as source material, but what transpired unequivocally from his assessment was the need to integrate t he growing 49 Ibid., 65. However, see Ostrogorsky's influential study about cities during the "Dark Age," in which the numismatic e vidence played a crucial role, Ostrogorsky, "Byzantine Cities in the Early Middle Ages," DOP 13 (1959): 45 66. His article is a piece of a larger polemic concerning the survival of cities, which involved prominent Byantinists like A. Khazdan and P. Grierso n, all of whom relied heavily on the presence or absence of coin finds in order to support their arguments, for which see A. P. Khazdan, "Vizantijskie goroda v VII XI vv," Sovestkaja archeologija 21 (1954): 164 83; P. Grierson, "Coinage and Money in the By zantine Empire 498 c. 1090," in (Spoleto: Presso La Sede del Centro, 1961) 445 46. 50 J. L. Malter, Byzantine Numismatic Bibliography, 1950 1965 (Chicago: Argonaut, 1968). For similar bibliographies and useful critical c ommentary see P. Grierson, "Byzantine Numismatics," in A Survey of Numismatic Research 1960 1965. II. Medieval and Oriental Numismatics ed. K. Skaare and G. C. Miles (Copenhagen: International Numismatic Commission, 1967), 52 62; M. Restle, "Forschungen z ur Byzantinischen Numismatik, 1950 1960," Byzantinisch neugriechische Jahrbcher 19 (1966): 225 59, includes the less accessible literature from Bulgaria, Romania, and Russia. 51 D. Waage, Antioch on the Orontes vol. IV, part 2: Greek, Roman, Byzantine and Crusaders' Coins (Princeton: Department of Art and Archaeology of Princeton, 1952); M. Thompson, The Athenian Agora 2. Coins from the Roman through the Venetian Period (Princeton: The American School of Classical Studies at Athens, 1954); G. E. Bates, Byz antine Coins (Cambridge MA: Harvard University Press, 1971).

PAGE 44

44 body of numismatic material, excavation finds and hoards, into the larger narrative about early Byzantium. 52 The stage had been set for more ambitious goals, such as a better understanding of the early Byzantine monetary circulation in certain ar eas of the Empire. Michael Metcalf, who was a pioneer in this respect, pointed to the conceptual distinction between Byzantine coinage and Byzantine currency, which also meant a shift from the study of the former, as a purely numismatic enquiry, to an emph asis on the latter, which deals with money and its functions integrated into a larger historical framework. The eastern Mediterranean was the focus of early studies, 53 no doubt because of the larger number of published single finds and hoards, 54 and also bec ause of the existence of a larger body of complementary literary sources, which allowed for a more nuanced interpretation of the numismatic material. 55 Nevertheless, Metcalf's comparative analysis of coin circulation in Sirmia and Slavonia unveiled the grea t potential of such studies in the Balkans as well. 56 Although many single finds and 52 P. Grierson, "Coinage and Money in the Byzantine Empire 498 c. 1090," in medioevo (Spoleto: Presso La Sede del Centro, 1961), 411 53; Grierson, "The Interpretation of Coin Finds," NC 5 (1965): i xiii; Grierson, "The Interpretation of Coin Finds (2)," NC 6 (1966): i xxi; Grierson, "Byzantine Coinage as Source Material," in Proceedings of the XIII International Congress of Byzantine Studies, Oxford, 5 10 September 1966 e d. J. M. Hussey, D. Obolensky, and S. Runciman (London: Oxford University Press, 1967), 317 33. 53 E. J. Prawdzic Golemberski and D. M. Metcalf, "The Circulation of Byzantine Coins in the South Eastern Frontiers of the Empire," NC 123 (1963): 83 92; D. M. Metcalf, "Some Byzantine and Arab Byzantine Coins from Palaestina Prima," INJ 2, n. 3 4 (1964): 32 47. 54 For a comprehensive bibliography, see H. C. Noeske, Mnzfunde aus gypten I. Die Mnzfunde des gyptischen Pilgerzentrums Abu Mina und die Vergleichsf unde aus den Diocesen Aegyptus und Oriens vom 4. 8. Jh. n.Chr vol. 1 (Berlin: Mann, 2000), 34 71. 55 For a brilliant integration of both types of evidence, see P. Grierson, The Monetary Reforms of Anastasius and their Economic Consequences," in Internati onal Numismatic Convention, Jerusalem 1963; The Patterns of Monetary Development in Palestine and Phoenicia in Antiquity ed. A. Kindler (Tel Aviv/Jerusalem: Schocken, 1967), 283 302. 56 D. M. Metcalf, "The Currency of Byzantine Sirmia and Slavonia," Hambu rger Beitrge zur Numismatik

PAGE 45

45 hoards were being published, especially in Bulgaria, the linguistic barrier remained a major deterrent for most western scholars. When A.H.M. Jones published his magisteria l Later Roman Empire there was already enough information to rely on in order to integrate the numismatic evidence into an economic survey of the Late Roman period. Nevertheless, Jones had little interest in archaeology, which he famously ignored in his wo rk, while he relied almost exclusively on written sources. 57 Consequently, his discussion of monetary issues was restricted to comments about salaries, prices and the standard of living as it transpired from literary sources, as well as discussions pertaini ng to the exchange rate between copper and gold and its economic implications. 58 In the English speaking world scholarship would still pay tribute to this tradition of writing the monetary history of early Byzantium based almost exclusively from literary so urces, even though the increasing number of finds from excavations called out for closer scrutiny. Michael Hendy's monumental work Studies in the Byzantine Monetary Economy c. 300 1450 is perhaps the perfect embodiment of this type of approach. 59 Although he 14 (1960): 429 44. 57 A.H.M. Jones, The Later Roman Empire 284 602. A Social, Economic and Administrative Survey 2 vols. (Norman: University of Oklahoma Press, 1964). For a recent assessment of his work, see B. Ward Perkins, "Jones and the Late Roman Economy," in A. H. M. Jones and the Later Roman Empire, ed. D. M. Gwynn (Leiden/Boston: Brill, 2008), 193 208. 58 Jones, The Later Roman Empire 435 448; Surprisingly, as a young scholar Jones had an interest in coinage, being invo lved in the publication of finds from the excavations conducted in the hippodrome of Constantinople in the 1920s (see above, n. 28). He further encouraged the use of the numismatic evidence arguing that "by the classification of coins they [numismatists] h ave made a substantial contribution to history, especially in the area and periods where the literary evidence is scanty or untrustworthy." See A. H. M. Jones, "Numismatics and History," in Essays in Roman Coinage Presented to Harold Mattingly ed. R. A. G Carson and C. H. V. Sutherland (Oxford: Oxford University Press, 1956), 13. 59 M. Hendy, Studies in the Byzantine Monetary Economy c. 300 1450 (Cambridge: Cambridge University Press, 1985).

PAGE 46

46 made use of site finds and hoards he did so only as an ancillary to his main arguments economy had more historical depth than any previous study and it remains to date t he monetary study most widely cited by historians. In fact, Hendy pointed to the time honored division between numismatist and historian in the introduction to his influential book. In his view, both were responsible for the lack of proper dialogue. The nu mismatist rarely ventured out of his narrow numismatic enquiry, while the historian was reluctant to accept the historical conclusions drawn by looking at the numismatic evidence, if he did not dismiss the discipline of numismatics altogether as arcane and focused on minute details, irrelevant to the larger narrative. Hendy noted: In any case, the effect of these several tendencies and divisions current tendencies in the study of numismatics, inherited divisions in the study of history, and a general d ivision between numismatists and historians has been little short of disastrous for the study of the Byzantine monetary economy as a whole. 60 Byzantine History introduced the western readers t o the standard view of Byzantium in Greek scholarship, which placed greater weight on material culture. This emphasis was clearly reflected in her introduction on the sources of Byzantine history, which included a section dedicated to numismatic finds and a short bibliography comprising standard catalogues, general handbooks, site finds and hoards. 61 60 Ibid., 11. His observation echoes a similar criticism made b y P. le Gentilhomme some forty years earlier: "Si la numismatique, entendue dans son sens le plus large, peut se dfinir comme la science de la monnaie dans ses rapports avec l'histoire, cette science se limite trop souvent, par la faute de numismates, l a connaissance descriptive de la monnaie, considre comme un document archologique plutt que comme un moyen d'change," P. le Gentilhomme, "Le monnayage et la circulation montaire dans les royaumes barbares en occident (V e VIII e sicle)," RN 7 (1943): 45. 61 A. Christophilopoulou, Byzantine History: I. 324 610 trans. W. W. Phelps (Amsterdam: Adolf M. Hakkert, 1986), 50 53. The original Greek edition dated back to 1975.

PAGE 47

47 Under the influence of the Annales school, French scholarship on Early Byzantium had a larger emphasis on social and economic history and the life of the common people using a variety of sources. velyne Patlagean's well known book Pauvret conomique et pauvret sociale Byzance, 4 e 7 e sicles is a good example of the integration of the numismatic evidence into the narrative about social life in early Byzantium 62 Patlagean used both literary sources, especially saints' lives, and excavation finds to describe the social and economic conditions of the period. Unlike other historians, Patlagean had an excellent knowledge of the literature pertaining to Byzantine co ins. She acknowledged the importance of site finds for estimating the economic prosperity of a region and seemed more interested in the functions of money beyond state related affairs, which was the primary focus of English scholarship. Her interpretation of finds used a historical lens as she explained the scarcity of coins by looking at invasions or demographic changes caused by earthquakes and the sixth century plague. By that time, the standard catalogues of Wroth and Tolstoy had been largely superseded by the publication of the large collections in Washington (1966 8) and Paris (1970), which opened a new era in terms of the methodology behind the study of Byzantine coinage. 63 Both works included detailed numismatic and historical introductions and also references to general works, single finds, and hoards. Relying on 62 E. Patlagean, Pauvret conomique et pauvret sociale Byzance, 4 e 7 e sicles (Par is: Mouton, 1977). 63 A. R. Bellinger and P. Grierson, Catalogue of the Byzantine Coins in the Dumbarton Oaks Collection and in the Whittemore Collection vol. I II (Washington, DC: Dumbarton Oaks Center for Byzantine Studies, 1966 1968); C. Morrisson, Cata logue des monnaies byzantines de la Bibliothque Nationale (491 1204) Tome premier: D'Anastase Justinien II (491 711) ( Paris: Bibliothque Nationale, 1970). One could add here a useful handbook of Byzantine coins by Hugh Goodacre, a 1959 shortened versi on of his earlier massive work (see above, n. 20), largely superseded by P. Grierson, Byzantine Coins ( Berkeley: University of California Press, 1982).

PAGE 48

48 the major public collections, public sales, but also excavation finds, Wolfgang Hahn created a corpus of Early Byzantine coins with a rich introduction on iconography, and the nature of coi n production and distribution. 64 The number of excavation reports itself multiplied in the last quarter of the twentieth century. A full treatment is beyond the scope of this chapter, but the steady increase in find reports and studies can be followed in a number of specialized bibliographic updates. 65 Almost every archaeological monograph included a separate chapter on coins, but unfortunately very few went beyond the basic cataloguing of coins followed by a number of general observations. 66 Sometimes this wa s due to the still insufficient number of finds which did not allow for a full treatment of the numismatic material in the context of the site itself or of an entire circumscribed region. This situation can be disconcerting for the historian interested in including such finds in his/her narrative. Nevertheless, historians can rely on a larger number of analytical studies written in the past few decades either in English 64 W. Hahn, Moneta Imperii Byzantini 3 vols. (Vienna: Verlag der Osterreichischen Akademie der Wissensch aften, 1973 1981). 65 Numismatic Literature issued by the American Numismatic Society strives to gather new contributions in the field of numismatics; the Byzantine section offers a good indication of the number and variety of recent finds. Relevant bibli ographies can also be found in Byzantinische Zeitschrift and in Survey of Numismatic Research published with the occasion of the International Numismatic Congress, usually held once every seven years. For a useful annotated bibliography, see also J. L. M alter, Byzantine Numismatic Bibliography, 1966 1994 (Encino, CA: Malter Galleries, 1995). 66 See, however, a number of book length studies on the monetary circulation in towns from the Balkans and the Near East, T. Marot, Las monedas del Macellum de Gerasa (Yaras, Jordania): aproximacin a la circu (Madrid: Museo Casa de la Moneda 1998); K. Sheedy et al., Pella in Jordan, 1979 1990: The Coins ( Sydney: Adapa, 2001); J. DeRose Evans, The Joint Expedition to Caesarea Maritima. Excavation Reports.Volume VI. The Coins and the Hellenistic, Roman, and Byzantine Economy of Palestine ( Boston: American Schools of Oriental Research 2006); K. Butcher, Archaeology of the Beirut Souks 1. Small Change in Ancient Beirut: The Coin Finds from BEY 006 and BEY 045: Iron Age, Hellenistic, Roman and Byzantine Periods," Berytus 45 46, 2001 2002 (Beirut: American University of Beirut, 2003); E. Theoklieva Stoycheva, Medieval Coins from Mesemvria (Sofia: Agato, 2001).

PAGE 49

49 or in another international language, drawing on the significant number of single finds and hoards accumulated in almost every region. 67 Despite such developments, even historians who use material culture in their work still feel uncomfortable handling the numismatic evidence. Above all it is still unclear what type of economic functions were performed by coins and how exactly they can assist economic historians in mapping regional, inter regional and even international trade. 68 Scholars who do engage with this type of source material usually do it from an indirect angle, by extracting the role of money in the early Byzantine state from written sources, rather than referring to the body of numismatic finds from excavations. 69 In the previous decades, this type of exercise had already inspired an approach to money as a means of state expenditure t o its main beneficiaries, the army and the administration. 70 Nevertheless, it is encouraging to see that economic histories 67 C. Morrisson, "La monnaie en Syrie Byzantine," in Archologie et histoire de la Syrie II. La Syrie de ed. J. M. Dentzer and W. Orthmann (Saarbrucken: Saarbrucker Druckerei und Verlag, 1989), 187 204; Morrisson, "La circulation montaire dans les Balkans enne et post justinienne," in Acta XIII Congressus internationalis archaeologiae christianae ed. N. Cambi and E. Marin, vol. II (Vatican: Pontificio Istituto di Archeologia Cristiana, 1998), 919 30; F. Curta, "Invasion or Inflation? Sixth to Seventh Centu ry Byzantine Coin Hoards in Eastern and Southeastern Europe," Annali dell'Istituto Italiano di Numismatica 43 (1996): 65 224; Mnzfunde ; A. Walmsley, "Coin Frequencies in Sixth and Seventh Century Palestine and Arabia: Social and Economic Implications," Jo urnal of the Economic and Social History of the Orient 42, n. 3 (1999): 326 50; C. Lightfoot, "Byzantine Anatolia: Reassessing the Numismatic Evidence," RN 158 (2002): 229 39; F. Curta, "Byzantium in Dark Age Greece (The Numismatic Evidence in its Balkan C ontext)," BMGS 29, n. 2 (2005): 113 46; C. Morrisson Les Trsors montaires byzantins des Balkans et 713) (Paris: Lethellieux, 2006); A. G Circulation in the Byzantine Provinc e of Scythia during the 6th and 7th Century," in Numismatic, Sphragistic and Epigraphic Contributions to the History of the Black Sea Coast ed. I. Lazarenko, vol. 1 (Varna: Zograf, 2008), 301 30. 68 C. Wickham, Framing the Early Middle Ages: Europe and th e Mediterranean, 400 800 (Oxford: Oxford University Press, 2005), 702, n. 16; M. Mundell Mango, "Byzantine Trade: Local, Regional, Interregional and International," in Byzantine Trade, 4th 12th Centuries: The Archaeology of Local, Regional and Internationa l Exchange: Papers of the Thirty eighth Spring Symposium of Byzantine Studies, St John's College, University of Oxford, March 2004 ed. M. Mundell Mango (Burlington, VT: Ashgate, 2009), 5. 69 Banaji, Agrarian Change in Late Antiquity ; Sarris, Economy and So ciety 70 Hendy, Studies in the Byzantine Monetary Economy ; R. Delmaire, Largesses sacres et res privata:

PAGE 50

50 of Byzantium, such as the monumental three volume work edited by Angeliki Laiou, include important sections on monetary production an d circulation which rely on both literary and archaeological sources. 71 Moreover, the fact that Ccile Morrisson, a professional numismatist, is the co author of a recent monograph on the Byzantine economy is yet another indication that coins will become mo re visible in the Byzantine economic landscape. 72 2 .3 Coins Beyond Frontiers: The Use and Abuse of the Numismatic Evidence in National and Regional Studies Unlike western scholarship, national and local histories had a more contextualized approach to the nu mismatic material, although interest in the Byzantine period grew much later. Typically, in the countries whose territory was once part of the Byzantine Empire or whose proximity to Byzantium can be seen in the material culture, the antiquarian phase is le ss prominent than in the West. Except for Tolstoy, whose major collection was partly acquired from Western sources, there is no catalogue of Byzantine coins published in Eastern Europe, Turkey or the Near East. Historical conditions, namely the birth of mo dern states and decolonization explain why interest in the Byzantine period as a part of national history came into focus only in the twentieth century, and in some cases, only in the past few decades. Nevertheless, if western scholars, whose enquiries enc ompassed vast geographic regions, were primarily looking to answer questions on a macro scale, at the local level, where Byzantine history was intimately entwined with a nation's past, the questions posed were of a L'aerarium imprial et son administration du IV e au VI e sicle (Rome: Ecole Franaise de Rome, 1989). 71 A. E. Laiou, ed., Economic History of Byzant ium 3 vols. ( Washington, D.C.: Dumbarton Oaks Research Library and Collection, 2002) 72 A. E. Laiou and C. Morrisson, The Byzantine Economy (Cambridge: Cambridge University Press, 2007).

PAGE 51

51 different nature. Unable to rely on a sat isfactory body of written accounts, historians have been forced to turn to material culture for answers. In many cases, as we shall see, the numismatic material played an important part in constructing theories about local economic development, social orga nization, and the establishment of chronologies for cultural diffusion. In other words, coins had a more concrete application as local historians were trying to pull together all the possible sources to reconstruct a piece of the nation's past. The presen ce of Byzantine coins outside the Empire's borders was itself of little interest to western scholars if it did not contribute to a better understanding of their own local history. 73 Byzantine coins were being used either for dating purposes, 74 to illustrate the transition to the beginnings of the early medieval coinages, especially in the Merovingian and Visigothic realms, 75 or as catalogues of museum collections meant to illustrate the monetary history of a region or, more often, local antiquarian initiative s since the Age of Enlightenment. 76 73 See for instance the inventory of Byzantine coins found in Gaul wi th bibliographic references going back to the late 19th century, J. Lafaurie and C. Morrisson, "La pntration des monnaies byzantines en Gaule Mrovingienne et Visigotique du VIe au VIIIe sicle," RN 29 (1987): 38 98. For an updated inventory see J. Lafau rie and J. Pilet Lemire, Monnaies du Haut Moyen ge dcouvertes en France (V e VIII e sicle) (Paris: CNRS ditions, 2003). For Germany, see K. Christ, Antike Mnzfunde Sudwestdeutschlands. Mnzfunde, Geldwirtschaft und Geschichte im Raume Baden Wrttenberg s von keltischer bis in alamannische Zeit (Heidelberg: Quelle & Mayer, 1960); R. Laser Berlin: Akademie, 1980). 74 J. Werner, Mnzdatierte austrasische Grabfunde (Berlin/Leipzig: 1 935). 75 le Gentilhomme, "Le monnayage et la circulation." 76 France: M. Amandry et al., Catalogue des monnaies d'or grecques, gauloises, romaines du V e sicle, ostrogothique, franques, mrovingiennes, lombardes et byzantines (Nmes: Muses d'art et d'histoi re de Nmes, 1989); C. Morrisson et al., Les Monnaies byzantines du muse Puig (Perpignan: Muse numismatique Joseph Puig, 1991); Belgium: F. de Callata, "Le dveloppement de la collection de monnaies byzantines du cabinet des mdailles de la Bibliothque Royale Albert I er ," Revue belge d'archologie e d'histoire de l'art 65 (1996): 259 98. Germany: S. Schultz, Byzantinische Mnzen in der Frhchristlich Byzantinischen Sammlung im Bode Museum (Berlin: Staatliche Museen zu Berlin, 1977);

PAGE 52

52 In Turkey, the Byzantine period never really captivated the interest of local historians, despite the fact that Anatolia represented the Byzantine heartland and was thus one of the most promising regions for archaeologica l research. Aside from the ongoing research of western archaeological teams in Asia Minor, which resulted in the publication of an important number of Byzantine coins, single finds and hoards, 77 there is only one major Turkish publication, the coins found d uring the excavations at Side, 78 a major museum collection, 79 and an unpublished dissertation about the early Byzantine coins from Pisidian Antioch. 80 In what concerns the Empire's eastern frontier there is only recent concern with economic relations in Ar abia, the main results being expected from the major international archaeological project Limes Arabicus 81 and an interest in the history of J. Raeder Die byza ntinischen Mnzen im Kestner Museum Hannover (Hannover: Kestner Museum, 1987); S. Burmeister, Rmische und byzantinische Mnzen der Universitt Rostock (Rostock: Institut fr Altertumswissenschaften, 1999); A. S. Sommer, Katalog der Byzantinischen Mnzen ( Gttingen: Universitatsverlag Gottingen, 2003). Switzerland: W. Hummel and L. Specker, Historisches Museum St. Gallen: Katalog der byzantinischen Mnzen (St. Gallen: Historisches Museum, 1982); Austria: R. Gbl, "Die byzantinischen Munzen in der Sammlung d er Mechitharisten Congregation in Vienna," NZ 79 (1961): 11 27; M. Alram, R. Denk, and W. Szaivert, Die Mnzsammlung des Augustiner Chorherrenstiftes Klosterneuburg sterreichischen Akademie der Wissenschaften 1989); M. Alram et al., Die Mnzsammlungen der Benediktinerstifte Kremsmnster und St. Paul im Lavanttal Die Mnzsammlungen de r Zisterzienserstifte Wilhering und Zwettl Wissenschaften 1975); W. Szaivert, Die Mnzsammlungen des Stiftes Gttweig 77 See abov e n. 28 31, and below, Chapter 5. 78 S. Atlan, 1947 (Ankara: Trk Tarih Kurumu 79 O. Tekin, Yap Kredi Koleksiyonu, Bizans sikkeleri (Istanbul: YKY, 1999). 80 Z. Demirel Gkalp, Yalva c ve Isparta arkeoloji muzelerinde bulunan Bizans sikkeleri Ph.D. diss., Anadolu Universitesi Sosyal Bilimler Enstitusu, 2007. 81 T. Parker, ed. The Roman Frontier in Central Jordan. Final Report on the Limes Arabicus Project, 1980 1989 2 vols. (Washingt on DC: Dumbarton Oaks Research Library and Collection 2006). See also below, Chapter 3.3.

PAGE 53

53 the Arab foederate tribes, which created a buffer zone between Persia and Byzantium. 82 In the region of Transcaucasi a, however, the interest in Byzantine finds grew quite early as Georgian and Armenian historians were trying to understand the intricate history of the country during the sixth century when both the Byzantine and the Sasanian empires were trying to gain co ntrol of this important strategic region, regarded as a buffer zone between the two powers. The Byzantine influence is most visible in Georgia, where the kingdom of Lazica had been under the cultural, economic, and political influence of Byzantium during L ate Antiquity. Coin finds published by Georgian numismatists have been used by historians in conjunction with written sources to illustrate this influence in major Lazi centers, such as Vani, Pitsunda, Archaeopolis, and Kutaisi. 83 Furthermore, the publicati on of museum collections, composed of coins found locally has offered a body of material which allows for statistical analysis and comparison with other regions. 84 In Armenia the heavier influence of Sasanian Persia resulted in fewer Byzantine coin finds. C onsequently, hoards have a mixed composition, of coins issued by Byzantium and Sasanian Persia, whose interests collided more heavily in Armenia. 85 82 The most comprehensive work, if almost exclusively based on literary sources, remains I. Shahd, Byzantium and the Arabs in the Sixth Century 2 vols. (Washington, D.C.: Dumbarton Oaks Research Library and Collection, 2005 2009). 83 O. D. Lorthkhiphanidze Vani archeologiceskie raskopki vol. 4 (Tbilisi: Mecniereba, 1979); A. M. Apakidze et al., vol. 2 (Tbilisi : Mecniereba, 1977); Zakaraia,ed., Nokalakevi Arkeopolisi. Arkeologiuri gatkhrebi 1978 1982 (Tbilisi: Mecniereba, 1986); N. Inaishvili, (Tbilisi: Mecniereba, 1993). 84 T. Abramishvili, Sakartvelos sakhelmts'ipo muzeumis bizant'iuri monet'ebi (Tbilisi: Metsniereba, 1965); T. Abramishvili, Sakartvelos sakhelmts'ipo muzeumis bizant'iuri monet'ebi (1966 1984 ) (Tbilisi: Metsniereba, 1989); I. Tsukhishvili and G. Depeyrot, History and Coin Finds in Georgia: Late Roman and Byzantine Hoards (4th 13th c.) (Wetteren: Moneta, 2003); M. Tsotselia, Coin Finds in Georgia: (6th Century BC 15th Century AD) (Wetteren: Moneta, 2010). 85 The Byzantine coins have been published by Khatchatur Musheghian in a series of works starting from the 1960s, Kh Musheghian, Denezhnoe obrashchenie Dvina po numizmaticheskim dannym (Erevan: Izdatel'stvo Akademii Nauk Armianskoi SSR, 1962) ; Musheghian, Denezhnoe obrashchenie v Armenii (V v. do n.e. XIV v.n.e.) (Erevan: Izdatel'stvo Akademii Nauk Armianskoi SSR, 1983 ); Kh. Musheghian, The

PAGE 54

54 Unfortunately, with publications from sites in eastern Turkey still insufficient, 86 it is hard to gain a more complex understanding of economic relations on the Transcaucasian frontier and historical narratives still rely heavily on written sources. 87 The region north of the Danube and of the Black Sea, which is the main focus of this book, has produced by far the largest number of early Byzantine coins to be found in barbaricum and it is also the region with the most interesting historiography on the use (and abuse) of the numismatic evidence. Very often ideology has played its part in the interpretation of coin f inds along ethnic lines as the Soviet, Romanian, Bulgarian, and Hungarian archaeological schools have struggled to substantiate different theories by looking at the same body of evidence. In Romanian scholarship a general interest in Byzantine coin finds d ates back to the interwar period, 88 but the first scholarly attempt at analyzing the Byzantine coin finds north of the Danube came only a few decades later. 89 In an ideological climate emphasizing notions such as autochthony and cultural continuity, C. Pred a explained the significant presence of early Byzantine coins north of the Danube border by the Numismatics of Armenian History (Erevan: Anahit, 1997); Kh. Musheghian et al., History and Coin Finds in Armenia: Coins from Ani, Capital of Armenia, 4th c. BC 19th c. AD (Wetteren: Moneta, 2000); Kh. Musheghian et al., History and C oin Finds in Armenia: Coins from Duin, Capital of Armenia, 6 7th c.: Inventory of Byzantine and Sasanian Coins in Armenia, 6 7th c (Wetteren: Moneta, 2000). 86 For eastern Turkey there is only one unpublished catalogue of the finds from the important fort ress of Melitene by Zeliha Demirel Gkalp whom I thank for allowing me to study the manuscript, Z. Demirel Gkalp, Malatya Arkeoloji Mzesi bizans sikkeleri kat 87 Tsukhishvili and Depeyrot, History and Coin Finds 17 28; For an example of an exclusively written source approach, see M. Whittow, The making of Byzantium, 600 1025 (Berkeley: University of California Press, 1996), 194 220. 88 I. Minea, "Influen lea n BSNR 27 28 (1933 1934): 97 114. 89 I. Dimian, "Cteva descoperiri monetare bizantine pe teritoriul RPR," SCN 1 (1957): 189 216.

PAGE 55

55 cultural integration of this territory into the Byzantine world. 90 Coins were thus given the same economic function as the one performed in the Empire and the ex istence of a true monetary economy was predicated upon this principle. In addition, for the sake of historical unity, the author disregarded any regional variations and particular cultural conditions in various parts of today's Romania, e.g. proximity to B yzantium in the south, Gepid and Avar influence in the west, Slavs in the east. This fundamental notion of a rather uniform Romanized population has led to the interpretation of coin finds as evidence of a "civilized" society which tried to avoid contact w ith barbarian invaders migrating from the East, while at the same time maintaining strong ties with the Byzantine world. Regardless of the region under discussion, numismatists and historians relying on the numismatic evidence have tried to prove that the existence of early Byzantine coins and other imports indicate the presence of a sedentary Romanic population. 91 Although the existence of a genuine monetary circulation in regions that were not part of the Empire has been rejected in recent studies, it is s till unclear what were the functions performed by early Byzantine coins once they crossed the political border of the Empire. 92 90 C. Preda, C SCIV 23 (1972): 375 415. 91 D. G. Teodor, Romanitatea carpato XI e.n 38 ; S. Dolinescu Ciurel (V e VII e sicle). La situation en Valachie," Dacia N.S. 28 (1984): 120 ; Societatea carpato danubiano XI. Structuri demo politice (Bucharest: 1997), 174 ; I. Corman, Co nistrian n epoca evului mediu timpuriu (sec. V VII d.Chr.) Cartdidact 1998), 98 9 ; IV VIII d.Hr.)," Carpica 31 (2002): 59 78. 92 vestul Romniei," Ephemeris Napocensis 12 (2002): 239; E. Oberlnder Trnoveanu, Barbaricum apropiat in (secolele VI X) 4 (2004): 346.

PAGE 56

56 In the steppe north of the Black Sea and northern Caucasus, early Byzantine coin finds were mainly used by Soviet archaeologists to date burial assemblages and to build relative chronologies for early medieval artifacts. In a seminal early article A. K. Ambroz had argued that the dating of burials based on coins should not take into consideration the date when the coin was issued, s ince coins arrived in the region much later and at least a hundred years must be added in order to reach the approximate date of the burial. 93 This was not treated as a mere hypothesis but as a true axiom that needed to be applied to every case and it was m ainly rooted in the fact that early Byzantine coins were sometimes found together with later Islamic coins. However, since it was not founded on a careful study of the monetary circulation, his rather arbitrary "delay" principle influenced his huge chronol ogical sequence that covered the territory from the Ural Mountains to the Middle Danube. Such an interpretation of the numismatic finds has thus created a large gap between the chronology of the Black Sea region and the one established by the Hungarian arc haeologists working on the Avar period, who have criticized Ambroz's principles. 94 Aside from chronological issues, there have been very few attempts to determine the economic significance of coins found in the region. 95 In her monograph about the monetar y circulation in the north western Black Sea region Elena Stoliarik 93 A. K. Ambroz "Problemy rannesrednevekovoi khronologii Vostochnoi Evropy," Sovetskaia Arkheologiia n. 2 (1971): 96 123. 94 Cs. Blint, "ber die Datierung der osteuropischen Steppenfund e des frhen Mittelalters. Schwierigkeiten und Mglichkeiten," Mitteilungen des archologischen Instituts der ungarischen Akademie der Wissenschaften 14 (1985): 137 47; Blint, "Probleme der archologischen Forschung zur awarischen Landnahmen," in Ausgew hlte Probleme europischer Landnahmen des Frh und Hochmittelalters ed. M. Mller Wille and R. Schneider (Sigmaringen: Jan Thorbecke, 1993), 195 273. 95 V. V. Kropotkin, Ekonomicheskie sviazi Vostochnoi Evropy v I tysiacheletii nashei ery (Moskow: Nauka, 1967).

PAGE 57

57 concluded that a sedentary population maintained a small scale monetary economy, but also suggested that copper coin finds should be treated separately from gold coins, which performed dif ferent functions. 96 A corpus of Byzantine coin finds on the territory of the former U.S.S.R. was attempted by V. V. Kropotkin in 1962 but unfortunately it has been seldom updated in the last decades. 97 The Middle Danube region was another major recipient of early Byzantine coins. Starting from the second half of the sixth century the rise of the Avar khaganate as t he most important power in the T ransdanubian "Barbaricum" brought a huge quantity of gold coins in the form of political payments sent from Constan tinople. Based on the written sources, historians have advanced the figure of six million solidi (c. 83.000 pounds) for the total amount of gold paid to the Avars until the failed siege of Constantinople in 626. 98 A major question in Hungarian historiograph y is what happened to this huge accumulation of gold, given that the number of surviving solidi is rather small. 99 Another central debate is the transition from the early to the middle Avar period, which relies heavily on the nature of the numismatic eviden ce. As early as 1955 96 E. S. Stoliarik, Essays on Monetary Circulation in the North western Black Sea Region in the Late Roman and Early Byzantine Periods: Late 3rd century Early 13th Century A.D. (Odessa: Polis Press, 1993), 68 69. 97 V. V. Kropotkin, Klady vizantii skikh monet na territorii SSSR ( Moskow: Izd vo Akademii nauk SSSR, 1962). A recent French edition is unfortunately a mere translation of the 1961 original Russian version, with no attempt to update either the historical commentary or the inventory itself, Kropotkin, Les trouvailles de monnaies byzantines en. U.R.S.S. ed. G. Depeyrot (Wetteren: Moneta, 2006). 98 J. Iluk, "The Export of Gold from the Roman Empire to Barbarian Countries from the 4th to the 6th Centuries," Mnstersche Beitrge zur antichen Ha ndelsgeschichte 4, n. 1 (1985): 93. 99 I. Bna, "Byzantium and the Avars: the Archaeology of the First 70 years of the Avar Era," in From the Baltic to the Black Sea. Studies in Medieval Archaeology ed. David Austin and Leslie Alcock (London: Unwin Hyman, 1990), 113 18. A corpus of Byzantine coin finds in "Avaria" has been published by P. Somogyi, Byzantinische Fundmnzen Der Awarenzeit (Innsbruck: Universittsverlag Wagner, 1997), with a recent update, Somogyi, "Byzantinische Fundmnzen der Awarenzeit. Ei ne Bestandaufnahme, 1998 2007," AAC 42 43 (2007 2008): 231 298.

PAGE 58

58 Gyula Lszl had suggested that the cultural changes were brought about by the arrival of a new group in Pannonia in 670 680 led by Kuber after the Bulgars had been forced westwards under the pressure of the Khazars. 100 The theory was fu rther developed and refined by Istvn Bna who sustained the immigrationist thesis by looking at coin hoards from the Carpathian basin and the overall reduction in coin finds towards the end of the seventh century. 101 Although it has been largely refuted sin ce then, 102 the numismatic evidence remains one of the central points of debate, as well as the existence of striking cultural parallels in Ukraine for the Middle Avar period. 103 Numismatic evidence is therefore one of the key elements in reconstructing a piec e of the nation's past in countries touched by the Byzantine influence. The stakes are even higher for regions outside the Empire's borders, which received little attention in written sources and mostly related to the Empire's external politics, be it warf are or diplomatic action. The student of social and economic relations in the marginal regions of the Empire and beyond has to make extensive use of the archaeological evidence in order to make sense of the societies living in the shadow of the Empire and their relations to Byzantium. It is precisely to the nature of frontiers and the archaeological 100 Gy. Lszl Die Reiternomaden der Vlkerwanderungszeit und das Christentum in Ungarn," Zeitschrift fr Kirchengeschichte 59 (1940): 126 146. 101 I. Bna, Grave of an Avar horseman at Ivancs a," 97 (1970): 243 261. 102 Cs. Blint, Der Beginn der Mittelawarenzeit und die Einwanderung Kubers," Antaeus 29 30 (2008): 29 61, with the historiographic background of the debate. 103 See recently P. Somogyi, "New Remarks on the Flow of Byzantine Coins in Avaria and Wallachia during the Second Half of the Seventh Century," in The Other Europe in the Middle Ages: Avars, Bulgars, Khazars and Cumans ed. F. Curta (Leiden/Boston: Brill, 2008), 83 149; P. Somogyi, "Neue berlegungen ber de n Zustrom byzantinischer Mnzen ins Awarenland (Numismatischer Kommentar zu Csand Blints Betrachtungen zum Beginn der Mittelawarenzeit)," Antaeus 29 30 (2008 2009): 347 393; C. Blint, "Antwortschreiben an Pter Somogyi," Antaeus 29 30 (2008 2009): 395 4 01.

PAGE 59

59 evidence that we need now turn before addressing the role and functions performed by Byzantine coins in the Byzantine world and beyond.

PAGE 60

60 CHAPTER 3 THE EMPIRE'S FRONTIER ON THE LOWER DANUBE 3.1 The Frontier Question in Late Antiquity For more than a century historians have struggled to define the concept of frontier in Roman history and, paradoxically, one of the major challenges has not been related to understand ing ancient perceptions on their own terms, but to manipulate and negotiate modern concepts of borders and boundaries which stood against a proper understanding of what frontiers meant in the Roman world. Running the risk of falling into the trap of presen tism, a number of scholars based their argumentation on implicit or explicit comparisons with modern frontiers. Nineteenth century efforts to redraw the map of Europe based on nationalistic views of political, cultural, and linguistic separation have promp ted historians to rethink the notion of Roman frontiers by projecting modern concepts to ancient contexts. 1 The ensuing debate among specialists in Roman history and archaeology shaped two schools of thought. The main point of debate has been the very natu re and definition of the Roman frontier. Students of the Roman frontier had to grapple with an unwieldy question: was the Roman frontier a linear barrier separating two worlds or was it an area of economic, cultural, and religious contact? Is there more th an one definition of Roman frontiers? For the later Empire the answer to this question is fundamental for shaping our understanding of the relations between the Empire and barbaricum The idea of frontier as a line of demarcation between different civiliz ations had already been put forward by Edward Gibbon in the eighteenth century and it underlay nineteenth century thinking, influencing scholars like the American historian Frederick 1 W. Pohl, "Conclusion: The Transformation of Frontiers," in The Transformation of Frontiers. From Late Antiquity to the Carolingians ed. W. Pohl et al. (Leiden: Brill, 2001), 248.

PAGE 61

61 Turner. Even to this day, in almost any study dealing with the conceptual ization of Roman frontiers, it is de rigueur Most importantly, he argued that the westward expansion was the driving force behind American development, while the frontier itself represented the meetin g point between savagery and civilization. 2 Such ideas had been a source of inspiration for prominent Roman historians writing in the first half of the twentieth century. In 1908, Theodor Mommsen defined the limes as a double defensive line, having the sha pe of a strip of land with an inner and outer border. 3 A few decades later, Andreas Alfldi famously described the Danube and the Rhine as "moral barriers" separating Roman civilization from barbarians, 4 a strong echo of another well known defensive line, the Great Wall of China, described by Owen Lattimore as separating the Chinese civilization from the barbarian "outer darkness." 5 Alfldi's essay opened the first edition (1949) of a series of conferences dedicated to the study of Roman frontiers ( Limeskon gresse ), held periodically to this day, in which historians and archaeologists present their work on different sections of the Roman and early Byzantine limes 6 The results of this prestigious conference perpetuate the concept of a linear fortified border, as the 2 F. J. Turner, The Frontier in American History (New York: Holt, Rin ehart and Winston, 1893, reprint, New York: Holt, Rinehart and Winston, 1962). 3 Th. Mommsen, "Der Begriff des Limes," Gesammelte Schriften 5 (1908): 456 64. 4 A. Alfldi, "The Moral Barrier on Rhine and Danube," in The Congress of Roman Frontier Studie s, 1949 ed. E. Birley (Durham: Durham University, 1952), 1 16. 5 O. Lattimore, Inner Asian Frontiers of China (Boston: Beacon Press, 1962), 238. Comparisons between the Chinese Wall and the Roman frontier system are still in fashion although of course ad apted to more recent developments in frontier studies both in China and in the West, for which see T. Wilmott, "Towers and Spies on Chinese and Roman Frontiers," in The Army and Frontiers of Rome. Papers Offered to David J. Breeze on the Occasion of his Si xty fifth Birthday and his Retirement from Historic Scotland ed. W. S. Hanson, JRA Suppl. 74 (Portsmouth, R.I. : Journal of Roman Archaeology, 2009), 127 33. 6 The term itself, limes is a modern adaptation of the Roman term which had a different connotat ion, for which see especially B. Isaac, "The Meaning of the Term Limes and Limitanei ," JRS 78 (1988): 125 47.

PAGE 62

62 program of the conference has only recently begun to include studies on aspects unrelated to military strategy, such as the role of economy, religion, and group identity. 7 Surprisingly, the most influential, yet controversial, thesis has come from Edward Luttwak, a modern military strategist during the Cold War era. His "Grand Strategy" of the Roman Empire, and more recently, his "Grand Strategy" of the Byzantine Empire, has championed the view of a rational, pragmatic, and well informed frontier p olicy, which adapted to external threats. 8 Ac c ording to Luttwak, the frontier strategy of the late Empire was characterized by an in depth defense system, based on a network of fortifications able to absorb the shock of barbarian invasions. Perhaps the gre atest achievement of his episodic incursions into Roman history has been the body of scholarship produced to refute his arguments. 9 In doing so, scholars involved in the frontier debate have reshaped and refined the notion and function of Late Roman fronti ers. 10 In the light of recent interest in social, economic, and religious issues, 7 For an review of the problem, see D. H. Miller,"Frontier Societies and the Transition between Late Antiquity and the Early Middle Ages," in Shif ting Frontiers in Late Antiquity ed. R. W. Mathisen and H. S. address exchange patterns beyond the frontier, M. Wheeler, Rome beyond the Imperial Frontier (Lo ndon: Bell, 1954). 8 E. N. Luttwak, The Grand Strategy of the Roman Empire. From the First Century A.D. to the Third (Baltimore: Johns Hopkins University Press, 1976); E. N. Luttwak, The Grand Strategy of the Byzantine Empire (Cambridge, MA: Harvard Univ ersity Press, 2009). 9 The merits of an in depth defense strategy have been questioned by archaeological research conducted in the Near East, for which see B. Isaac, The Limits of Empire. The Roman Army in the East (Oxford: Clarendon Press, 1990). For Nort h Africa, see D. Pringle, The Defence of Byzantine Africa from Justinian to the Arab Conquest. An account of the Military History and Archaeology of the African Provinces in the Sixth and Seventh Centuries BAR International Series 99 (Oxford: B.A.R., 198 1), 94 109; D. Cherry, Frontier and Society in Roman North Africa (Oxford: Clarendon Press, 1998), 24 74. For the Balkans, see A. Poulter, "Cataclysm on the Lower Danube: The Destruction of a Complex Roman Landscape," in Landscape of Change: Rural Evolutio ns in Late Antiquity and the Early Middle Ages ed. N. Christie (Aldershot: Ashgate, 2004), 223 53. 10 B. Isaac wrote the most influential rebuttal of the Grand Strategy concept, which led to a rethinking of the entire limes concept, from a military standp oint; see Isaac, The Limits of Empire

PAGE 63

63 military strategy has become a secondary concern. 11 The role and function of "frontier societies" looms large in recent studies, as part of a larger intellectual shift, which s tresses the transformation of the Roman world in which romanitas and barbaritas blend together in a formative process. 12 and civilization. Despite the peculiarity of the Ameri can frontier, the versatility of his model resides in the ability to define frontiers as non static cultural areas, which fits, in a limited sense, the "transformation" theory. 13 The pendulum swing in the recent historiography has a main thrust and a number of subsidiary channels. Although the frontier question is far from being solved, it is now widely accepted that Late Roman frontiers were neither static lines of defense, nor barriers holding back barbarian tides. 14 11 In the last two decades collective volumes on the topic of late ancient and medieval frontiers have started to include separate sections on intellectual and religious frontiers, see for example A. Rousselle, ed. Fr (Paris: Diffusion De Boccard, 1995); Mathisen and Sivan, Shifting Frontiers ; O. Merisalo, ed. Frontiers in the Middle Ages: Proceedings of the Third European Congress of Medieval Studies (Jyvskyl 10 14 June 2003) (Louvain la Neuve: 2006). 12 The Transformation of the Roman World a massive project sponsored by the European Science Foundation, had a major role in shaping this new approa ch back in the 1990s. That it soon became the leading paradigm is evinced by recent volumes on the same topic, see for instance R. W. Mathisen and D. Shanzer, ed., Romans, Barbarians, and the Transformation of the Roman World. Cultural Interaction and the Creation of Identity in Late Antiquity (Farnham: Ashgate, 2011). 13 The most prominent difference would be the fact the Roman government played the major role in shaping the frontier, whereas in the case of the American frontier frontiersmen were the main agent; for a brief discussion see I. Kopytoff, "The Roman Frontier and the Uses of Comparison," in Frontires Nature et signification des frontires romaines ed. P. Brun et al. (Nemours: Edition de Archologique en Ile de France, 1993), 144. 14 Mathisen and Sivan, Shifting Frontiers ; Pohl, et al., The Transformation of Frontiers ; F. Curta, ed., Borders, Barriers, and Ethnogenesis. Frontiers in Late Antiquity and the Middle Ages (Turnhout: Brepols, 2 005); Isaac, The Limits of Empire ; C. Whittaker, Frontiers of the Roman Empire. A Social and Economic Study (Baltimore: Johns Hopkins University Press, 1994). This interpretation is not unanimously accepted. Derek Williams maintained that once the Empire r eached a military stalemate in all directions it had to accept de facto its territorial limit and to conduct the frontier policy based on the principle of an empire enclosed by perimeter barriers, see D. Williams, The Reach of Rome. A History of the Roman Imperial Frontier 1 st 5 th Centuries AD convey the notion of linearity, see for example P. Parker, The Empire Stops Here. A Journey along the Frontiers of the Roman World (London: Jonath an Cape, 2009). In addition, Mark Graham has

PAGE 64

64 The theory of the frontier as a homogene ous contact zone which included regions on both sides of the military installations was developed by C. R. Whittaker in a series of seminal works. 15 According to this interpretation, borders and boundaries are two distinct notions, which need to be treated by using distinct methodological tools. 16 While borders refer to political and military edges of society, boundary studies address the interaction at the frontiers and the social transformation produced by long periods of contact, leading to the creation of the frontier as a unique area, with a different outlook than the core regions of the Empire. While convincingly arguing against the notion of linear frontiers, either natural or artificial, historians have largely overlooked the subtleties of the evoluti on of frontiers from the early Roman period to the early Byzantine. There is also a marked tendency to generalize conditions throughout the Roman world based on research focusing mainly on the northern frontier of the Western Empire. Although Roman ideolog y may have remained unchanged in terms of an imperium sine fine with an added Christian dimension in Byzantium, the reality on the ground changed dramatically. 17 The endemic civil war of the third century and the collapse of the tetrarchic system in the f ourth convincingly argued that historians have overcorrected in their desire to avoid presentist notions of frontiers and by doing so they have rejected altogether the concept of static frontiers, see M. Graham, News and Frontier Consciousness in the Late Roman Empire (Ann Arbor: The University of Michigan Press, 2006), 50. 15 Whittaker, Frontiers of the Roman Empire ; C. R. Whittaker, Rome and Its Frontiers: The Dynamics of Empire (New York: Routledge, 2004). 16 Whitta ker, Rome and Its Frontiers 4; S. W. Green and S. M. Perlman, "Frontiers, Boundaries, and Open Social Systems," in The Archaeology of Frontiers and Boundaries ed. S. W. Green and S. M. Perlman (New York: Academic Press, 1985), 3 13; K. G. Lightfoot and A Martinez, Frontiers and Boundaries in Archaeological Perspective, in Annual Review of Anthropology 24 (1995): 471 92. For a similar approach of the Danube frontier, see E. Zanini, "Confine e frontiera: il limes danubiano nel VI secolo," in MILION. Stud i e ricerche d'arte bizantina ed. C. Barsanti et al. (Rome: Biblioteca di Storia Patria, 1988), 257 71. 17 For the Byzantine ideology on frontiers, see recently G. Dagron, "Byzance et la frontire. Idologie et ralit," in Frontiers in the Middle Ages 3 03 18.

PAGE 65

65 century transferred the military initiative to various barbarian tribes and the Empire was often forced to fight on several fronts. Conquering new territories or subduing new peoples was no longer as high a priority as preserving the Empire and keepi ng the barbarians at bay. Consequently, unlike early historians, writers of the late Empire referred to frontiers and the loss of territories (as opposed to conquering peoples). The often cited account of the fourth century anonymous author of De rebus bel licis is a case in point. Above all it must be recognized that wild nations are pressing upon the Roman Empire and howling about it everywhere, and treacherous barbarians, with the s taken by the State for its own advantage there is also the effective care of the frontier works which surround all the borders of the Empire. Their safety will be better provided for by a continuous line of forts constructed at intervals of one mile with firm walls and very powerful towers. 18 Based on this new ideology emperors often received praises for protecting the Empire. Constantius I was held in high regard in a panegyric because he "reached the Rhine unexpectedly and protected that whole frontie r", while another panegyric mentions the camps of cavalry units and infantry toto Rheni et Histri et Eufrate limite restituta by the Tetrarchy. 19 In the sixth century, Procopius spoke of the defenses with which Justinian "surrounded the farthest limits of the territory of the Romans," and of "all the fortifications whereby this Emperor preserved the Empire, walling it about and frustrating the attacks of the barbarians on the Romans." 20 As the Empire grew weaker, 18 De rebus bellicis 6.1 and 20, ed. and trans. E. A. Thompson (New York: Arno Press, 1979) 113 and 122 23. 19 Panegyrici Latini 8.13.3 and 9.18.4, ed. R. A. B. Mynors; trans. C. E. V. Nixon and B. S. Rodgers (Berkeley: University of California Pres s, 1994), 549 and 562. 20 Procopius, De Aedificiis II, 1.3, ed. J. Haury and G. Wirth, vol. 4 (Leipzig: Teubner, 1964), 46; trans. H.

PAGE 66

66 frontier consciousness grew stronger. 21 The si xth century historian Agathias was fully aware of the meaning of territorial losses in his own time when he referred to the shameful defeat of Julian in Persia "confining thereafter the extent of his Empire within new frontiers, whittling away its far flun g corners." 22 Especially the border with Persia, the archrival and the ideological antithesis of Rome, became increasingly important in the later centuries. Procopius showed that the need for establishing a firm frontier had a long tradition: "[...] it was forbidden in the treaty which the Emperor Theodosius once concluded with the Persian nation, that either party should construct any new fortress on his own land where it bordered on the boundaries of the other nation." 23 As a matter of fact, during the sixt h century frontier forts such as Dara in Mesopotamia became central in the military and diplomatic relations between Byzantium and Sasanian Persia. The capture of the fortress by the Persians literally drove Emperor Justin II insane. 24 3.2 The Early Byzanti ne Frontier on the Lower Danube: Topos and Reality A discussion of the Early Byzantine frontier on the Lower Danube, which will be the focus of the following pages, raises again the issue of rivers as frontiers of B. Dewing, vol. 7 (Cambridge MA: Harvard University Press, 1940), 97 8. The propagandistic tone does not reduce the hist defense is the well known inscription found in Byllis, Albania mentioning the fact that Viktorinos, ts of Scythia, Moesia, Thrace, and provinces balkaniques," Bulletin de la Socit nationale des Antiquaires de France 1988, 136 46. 21 For a recent dis cussion of this complex phenomenon see Graham, News and Frontier part I. 22 Agathias, Historiae, IV, 25.6 7, ed. R. Keydell (Berlin: De Gruyter, 1967), trans. J. D. Frendo (Berlin: De Gruyter, 1975), 128. 23 Procopius, De Aedificiis II, 1.5, ed. Haury and Wirth, vol. 4, 46; trans. Dewing, vol. 7, 99. Moreover, the regulation of trade at the Persian border in a rescript from 408/9 suggested that boundaries should be strictly enforced and traffic closely monitored, since no trade was permitted beyond Nisibis, Callinicum, and Artaxata, see Corpus Iuris Civilis 4.63.4 (408/9), ed. P. Krueger, vol. 2 (Berlin: Weidmann, 1929), 188 89. 24 John of Ephesus, Historia Ecclesiastica 3.4, ed. E. W. Brooks (Louvain: Ex Officina Orientali et Scientifica, 1936), 92.

PAGE 67

67 exclusion. If Alfldi's "moral frontier" i s no longer tenable, the role of rivers as primarily facilitating transport, communication, and supply is equally problematic. 25 A middle ground between these opposing views has been proposed by Martijn Nicasie and Mark Graham who looked at river frontiers from different standpoints, but reached a similar conclusion, that rivers had both a symbolic and a practical role of separation. 26 More often than not historians engaged in the frontier debate operate with multiple definitions of the frontier or tend to em phasize specific facets of this decidedly versatile concept. One major achievement of the past decades is the differentiation between political and cultural (including social and economic) frontiers. Depending on such angles, the same natural frontier can act as a cultural frontier of inclusion and at the same time play the role of exclusion in the shape of a fortified border. The Danube frontier is an excellent case for testing such assumptions against the available literary and archaeological evidence. Se veral scholars have argued that the Danube did not represent the logical frontier of the Empire in strategic terms. The argument proceeds on the premise that if the Danube had been the natural frontier, then it would have been strategic to maintain that li ne. Instead, Emperor Trajan 25 C. M. Wells, The German Policy of Augustus (Oxford: Clarendon Press, 1972), 24; Isaac, The Limits of Empire 410 3; Whittaker, Frontiers of the Roman Empire 171. Historians often cite Lord Curzon's 1907 lecture on frontiers as a historiographic foundation fo r their claim that rivers connect rather than separate; see Lord Curzon, Frontiers. Romanes Lecture (Oxford, Clarendon Press, 1908), 21. On that same page, however, Lord Curzon also states that rivers provided "a convenient line of division, easily capable of defence" and specifically referred to the cases of the Rhine and the Danube in the Roman period. 26 M. J. Niacasie, Twilight of Empire. The Roman Army from the Reign of Diocletian until the Battle of Adrianople (Amsterdam: J. C. Gieben, 1998), 123 5; Graham, News and Frontier 56 72. See also Parker, The Empire Stops Here 4. For the role of water defenses see recently N. Christie, "From the Danube to the Po: the Defence of Pannonia and Italy in the Fourth and Fifth Centuries AD," in The Transition to Late Antiquity 570 73. For the Danube frontier, see recently C. S. Sommer, "Why there? The Positioning of Forts along the Riverine Frontiers of the Roman Empire," in The Army and Frontiers 103 14.

PAGE 68

68 established a new province, which took many legions to defend and had to be abandoned in the third century. 27 Although the Danube was by no means a definitive frontier, from a Roman perspective it was still considered the leg al frontier of the Empire as late as the seventh century when the administration of the northern Balkans virtually collapsed. 28 The Danube offered clear strategic advantages and we have plenty of information to show that they were exploited both by Romans o n the right bank, and barbarians on the left bank. Since the establishment of the province of Moesia, Rome regarded the Danube as a line separating them from the peoples living north of the river, especially the Dacians, whose inroads south of the Danube a re described by Annaeus Florus: "whenever the Danube froze and bridged itself [ gelu Danuvius iunxerat ripas ], under the command of their King Cotiso, they used to make descents and ravage the neighboring districts." 29 A similar reference to the frozen river is made on Trajan's column, in the depiction of the heavily armed Dacian and Sarmatian cavalry crossing the frozen Danube for the Dacian counteroffensive in the winter of 101. 30 This view is strengthened by the fourth century writer Libanius who mentioned the raids of the "Scythians" which 27 Whittaker, Rome and Its Frontiers 34 5. 28 E. Chrysos, "Die Nordgrenze des byzantinischen Reiches im 6. bis 8. Jahrhundert," in Die Vlker Sdosteuropas im 6. bis 8. Jarhrhundert ed. B. Hnsel, (Munich: Selbstverlag der Sdosteuropa Gesellschaft, 1987), 27 40. 29 Annaeus Florus, Epitome II, 28, ed. T. E. Page et al., trans. E. S. Forster (London: William Heinemann Ltd., 1929), 334 35. For a discussion of the frozen Danube as an ancient literary topos in Greek and os," Gymnasium 64 (1957): 154 61, and more recently S. Patoura, "Emporio kai synallages ste dounabike He methorios tou Dounabe kai o kosmos tes sten epoche tes metanasteuses ton laon (4os 7os ai.) ed. S. P atoura Spanou (Athens: National Hellenic Research Foundation, Institute for Byzantine Research, 2008), 195 221. 30 L. Rossi, Trajan's Column and the Dacian Wars (Ithaca, N.Y.: Cornell University Press, 1971), 146 48. For a discussion of the Danube as symbol izing the frontier on the Column see also M. Galinier, "La Colonne trajane: images et imaginaire de la frontire," in Frontires terrestres 274 76.

PAGE 69

69 he "could not bear to look upon", praying that "the ice on the Danube should not be frozen solid, to be sufficient for their crossing 31 To be sure, the situation remained unchanged in the sixth century when Agathias was writing about the successful invasion of the Cutrigurs in 559: As usual, with the approach of winter, the river froze to a considerable depth and the ice was already hard enough to be crossed on horseback. Whereupon Zabergan, the leader of the Cotrigurs g alloped across the frozen waters with a huge force of cavalry and crossed over without difficulty into the territory of the Romans. 32 Such descriptions indirectly reveal that the unfrozen river was perceived as an efficient if not entirely reliable barrier 33 which became especially convenient in Late Antiquity, when the offensive stance of the Empire on the Lower Danube had drastically diminished. The role of the river as a communication artery cannot be denied, especially during the existence of the trans Danubian province of Dacia and also later, 34 but it would be a mistake to ignore the strategic value of the river in regulating access from and to the left bank. 35 The strategic importance of the Danube was not lost on Late Roman historians writing about fr ontiers, although propagandistic purposes, either 31 Libanius, Oratio 59 90, ed. R. Foerster, vol. 4 (Leipzig: Teubner, 1908), 253; trans. M. H. Dodgeon et a l. (New York: Routledge, 1996), 185. 32 Agathias, Historiae V, 11, ed. R. Keydell; trans. J. D. Frendo, 146. 33 Contra Isaac, The Limits of Empire 413, who turns the argument on its head by suggesting that such accounts prove that the Danube was not an eff icient natural barrier. 34 I. Barnea, "Le Danube, voie de communication byzantine," in He epikoinonia sto Byzantio. Praktika tou B' diethnous symposiou, 4 6 oktobriou 1990 ed. N. Moschonas (Athens: Kentro Byzantinon Ereunon, 1993), 577 95. 35 It is importan t to note that a new age of expansion in the late tenth century brought the Byzantines back to the Danube, where some of the old fortifications were reused and the river was once again guarded by the imperial fleet, see 10, n o 1 2 (1999): 41 55.

PAGE 70

70 positive or negative, should be taken into account. In the geographical introduction to his Historiarum adversus paganos Orosius described the territory which the "Danube separates from the land of the Barb arians in the direction of Our Sea." 36 Interestingly, centuries earlier of the fact that "the Danube separates what is Roman from what is Sarmatian" as he ridiculed the "bounda ries of mortals." 37 In addition, Zosimus noted that according to the peace terms imposed on the Goths by Valens, "the barbarians were forbidden to cross the river or ever again to set foot inside the Roman borders," 38 while Jerome lamented that the Danube fr ontier had been shattered by barbarian invasions ( fracta Danubii limite ), after which "Rome had to fight within her own borders not for glory but for bare life". 39 During the dark days of the Hunnic domination at the Danube Attila managed to extend his cont rol south of the river and according to Priscus of Panium he moved the frontier to Naissus, "which he had laid waste and established as the border point [ chorion ] between the Scythian and the Roman territory." 40 Later he re acknowledged the Danube as the pr oper frontier as he vowed to "keep the peace on the same terms" and to "withdraw from the Roman territory bordering on the Danube." 41 36 Orosius, Historiarum Adversus Paganos Libri VII 1, 2.54 ed. C. Zangemeister (Leipzig: Teubner, 1889), 9; trans. R. J. Deferra ri, 13. 37 Seneca, Naturales quaestiones I.9 (praefatio), ed. H. M. Hine (Leipzig: Teubner, 1996), 4 ( "Danuuius Sarmatica ac Romana disterminet") 38 Zo simus, Historia Nova 4.11, ed. L. Mendelssohn (Leipzig: Teubner), 167; trans. J. J. Buchanan and H. T. Davis (San Antonio: Trinity University Press, 1967), 146. 39 Jerome, epistula 123.17, ed. P. Schaff, trans. W. H. Fremantle (Peabody: Hendrickson Publishers, 1994), 237. 40 Priscus of Panium, History frag. 11.1, ed. and trans. R. C. Blockley (Liverpool: Fra ncis Cairns, 1981), 242 43. 41 Priscus of Panium, History IV, 15.4, ed. and trans. Blockley, 298 99.

PAGE 71

71 The status quo remained quite unchanged in the sixth century when Procopius was writing that the Danube "clearly forms th e boundary between the barbarians, who hold its left bank, and the territory of the Romans, which is on the right." 42 It is no less true that most accounts either refer to situations when the Danube was crossed by barbarians or constitute opportunities to p raise some emperor for securing the border, which indirectly suggests that it had been crossed quite often in the past. However, this is not an attempt to determine whether the frontier was successfully defended at all times, but to make the case that the river created a serious problem for groups trying to cross into the Empire. In addition, a chain of fortifications built on the right bank was meant to supplement the strategic value of the natural barrier. 43 At least one of their purposes was very clear t o contemporaries, even as they looked back to the state of the frontier before the onslaught of the Huns: The Roman emperors of former times, by way of p reventing the crossing of the Danube by the barbarians who live on the other side, occupied the entir e bank of this river with strongholds, and not the right ban k of the stream alone, for in some parts of it they built towns and fortresses on the other bank. 44 42 Procopius, De Aedificiis IV, 5.10, ed. Haury and Wirth, vol. 4, 125; trans. Dewing, vol. 7, 267. 43 Moreover, the fortifications on the Danube were suppl emented with walls and towers placed at the outlets of the tributaries into the Danube to prevent circulation upriver, see F. Curta, The Making of the Slavs (New York: Cambridge University Press, 2001), 184. The fact that the river itself and the adjacent fortifications were a convenient frontier line is evidenced by a Byzantine return to the Danube in the tenth century, for which see E. Condurachi et al., "Nouvelles recherches sur le Limes byzantin du Bas Danube aux X e XI e sicles," in Proceedings of the X IIIth International Congress of Byzantine Studies. Oxford, 5 10 September 1966 ed. J. M. Hussey et al. (London/New York/Toronto: Oxford University Press, 1967), 179 93. 44 Procopius, De Aedificiis IV, 5.2, ed. Haury and Wirth,124; trans. H. B. Dewing, vol 7, 265. For the view and terminology ( choria eschatiai ) of Procopius on frontiers, see, J. P. Arrignon and J. F. Duneau, "La frontire chez deux auteurs byzantins: Procope de Csare et Constantin VII Porphyrognte," Geographica Byzantina (1981): 17 30

PAGE 72

72 No doubt Procopius had in mind Justinian's recent renovatio imperii and the complete overhaul of the Danube frontier in the first half of the sixth century, including the control over a few strategic bridgeheads on the left bank, such as Litterata and Recidiva, mentioned in the Novella XI of 535. 45 The importance of the river and the adjacent fort ifications can be clearly seen in Menander's description of the Avar campaign against the Sclavenes, ordered by Tiberius II Constantine in the summer of 578. The Avar horsemen were ferried into the Empire and escorted along the right bank of the Danube onl y to cross the river again to attack the Slav ic tribes living in eastern Wallachia 46 It clearly follows that the right bank was safer and easier to protect and more suitable to stage various military maneuvers. Rivers also had a symbolic meaning for the Ro mans as boundaries that should not be crossed, with the Rubicon being the classical example, or as natural frontiers between two peoples, acting as neutral territory. This can be deduced from the fact that peace negotiations were sometimes held symbolicall y close to an important river. Romans had met the Persians on the Euphrates for negotiations in the first century, in the fourth century Valentinian negotiated with the Alamans on a ship in the middle of the Rhine, while Valens met Athanaric on the Danube. 47 After the dissolution of the Hunnic empire, the sons of Attila came to Emperor Leo I to ask that Romans and Huns should meet at the Danube for a peace treaty, "in the old manner" ( to palaion ethos ), and 45 Corpus Iuris Civilis Novella XI.2 (535 A.D.), ed. R. Schoell and W. Kroll, vol. 3 (Berlin: Weidmann, 1928), 94. 46 Menander, Historia fr. 21, ed. and trans. R. C. Blockley (Liverpool: Francis Cai rns, 1985), 192 95. 47 Ammianus Marcellinus, Res Gestae 30.3.4 5; 31.4.13, ed. J. C. Rolfe, vol. 3 (Cambridge, MA: Harvard University Press, 1939), 316 and 408; Tacitus, Annales 2.58, ed. E. W. Bowen (Boston/New York: B. H. Sanborn & co, 1913), 86 87.

PAGE 73

73 establish a market. 48 No doubt Priscus had in mind t he fourth century precedent with the Goths playing the role of the barbarian. In the sixth century the Avars having reached "the banks of the Scythian Hister" and threatening to invade the Roman province, sent envoys to Constantinople requesting stipends f rom Emperor Justin II. 49 Finally, in the early seventh century Georgios Pisides contemplated the role of the Danube against the threat of the Avars in the same vein as sixth century historians had, the river being described metaphorically as a "boundary to savagery" and "a fence and a new wall." 50 The border between symbolical and practical was easy to cross and the Danube often became a convenient separation line: Until now the river provided by unwritten (natural) law a border (to the Empire), (nonetheless) how much damage has the state suffered from the wicked barbarians On the one hand, they always tried to break through the established frontier, on the other hand, the power of the Romans prevented them from invading our land. 51 Especially when the Empir e was not strong enough to assert its influence north of the river, the Danube was specifically indicated as the northern border of Byzantium. Such was the case in the treaty sealed by Emperor Maurice with Baian, the khagan of the Avars according to which "t he Ister was agreed as intermedium between Romans 48 P riscus of Panium, History VI, 46, ed. and trans. Blockley, 352 5 3. 49 Corippus, In laudem Iustini Augusti minoris 3. 300, ed. and trans. A. Cameron (London: Athlone Press, 1976), 108. 50 Georgios Pisides, Bellum Avaricum 30 35, ed. and trans. A. Pertusi ( Ettal: Buch Kunstverlag, 1959), 177. 51 Georgios Pisides, Bellum Avaricum 30 40, ed. and trans. Pertusi, 177 78 and For a discussion of this particular work of Pisides, see P. Speck, Zuflliges zum Bellum Avaricum des Georgios Pisides (Munich: Institut fr Byzantinistik, Neugriechische Philologie und Byzantinische Kunstgeschichte der Universitt, 1980). However, these particular lines (30 40) concerning the Danube frontier received little historical attention. For a brief mention see Patoura, "Emporio kai s ynallages," 399, n. 7.

PAGE 74

74 and Avars." 52 Indeed, the Empire was sometimes in the strategic position of initiating offensive expeditions north of the Danube. 53 However, the very fact that the Empire was able to embark on such expedit ions only twice during the sixth century shows that the Danube was not necessarily the desired frontier, but certainly the more convenient and affordable one at that time. One could even argue that the few expeditions north of the Danube did not bring anyt hing good in the long term, but rather caused additional trouble for the frontier provinces. After a few years of successful Roman incursions north of the river led by Chilbudios, during which "not only did no one succeed in crossing the Ister against the Romans, but the Romans actually crossed over to the opposite side many times", in 533 the general was killed in battle and the result was that "the river became free for the barbarians to cross at all times just as they wished and the possessions of the Ro mans were rendered easily accessible." 54 It took more than six decades for the Romans to venture again north of the Danube. This time, after a series of successful expeditions, Petrus, Emperor Maurice's brother, was ordered to spend the winter in enemy terr itory in 602, which led to a general mutiny ending with the deposition of Maurice and added chaos at the Danube frontier. 55 On the rare occasions when the Empire was on the offensive beyond the Danube, the main reason for action 52 Theophylact Simocatta, Historia VII, 15.14, ed. C. de Boor and G. Wirth (Stuttgart: Teubner, 1972), 273; trans. M. Whitby and M. Whitby (Oford: Oxford University Press, 1986), 201. For the Danube as the political border of Byzant ium in the sixth century see also Chrysos, "Die Nordgrenze." 53 The early Byzantine fleet, well attested until the seventh century, played an important role at the Danube, for which see O. Bounegru and M. Zahariade, Les forces navales du Bas Danube et de la Mer Noire aux I er VI e sicles (Oxford: Oxbow Books, 1996), 108 9. 54 Procopius, Bella VII, 14.3, and VII, 14.6, ed. Haury and Wirth, vol. 2, 354; trans. Dewing, vol. 4, 263 65. 55 Theophylact Simocatta, Historia VIII, 6, ed. de Boor and Wirth, 293 95. T he idea that the Danube limes finally collapsed in 602 has been long refuted, see A. Barnea, "Einige Bemerkungen zur Chronologie des Limes an der unteren Donau in sp tr mischer Zeit ," Dacia 34 (1990): 283 90. However, the old dating of

PAGE 75

75 was not a desire to annex n ew territories but to pacify the region by retaliating against the barbarian groups who were often crossing into the Empire for plunder. The river itself was of course insufficient for keeping the frontier under control. A dense network of fortifications w as meant to supplement the strategic value of the natural obstacle. The program of reconstruction in the Balkans was initiated most probably before Justinian, who was responsible for taking it to an unprecedented scale. The defensive nature of the frontier fortifications is made apparent by the shape of the towers and the fact that in many cases walls were thickened and elevated. 56 The strategic planning which presupposed the existence of three layers of fortifications in the Balkans, whose existence was co nfirmed by archaeology, was meant to block the access to Constantinople and the large towns of Greece. 57 A massive program of fortifications in the Balkans and at the Danube, such as the one implemented by Justinian, does not necessarily go against the curr ent definition of the frontier as an area including not only a fortified natural or artificial border, but also the supporting frontier provinces and regions beyond the border. However, this is a cultural definition of frontiers rather than a political one the collapse has not yet disappeared from scholarship, see recently L. Ellis, "Ellusive Places: A Chorological Approach to Identity and Territory in Scythia Minor," in Romans, Barbarians, and the Transformation 251. 56 For studies of the early Byzantine fortifications of th e frontier region in the Balkans, see M. Biernacka The Roman and Early Byzantine Fortifications of Lower Moesia and Northern Thrace Zanini, "Confini e frontiera"; R. Ivanov, ed., Roman and Early Byzantine Settlements in Bulgaria vol. 2 (Sofia: Ivray, 2002); S. Torbatov, Ukrepitelnata sistema na provintsiia Skitiia (kraia na III VII v.) protobyzantin dans la province de Msie Premire," Starinar 45 46 (1994 1995): 41 Studien zu den Militrgrenzen Roms III. 13. internationaler Limeskongre Aalen 1983. Vortrge (Stuttgart: Konrad Theiss Verlag, 1986), 542 60; V. Dinchev, "The Fortresses of Thrace and Dacia in the Early Byzantine Period," in The Transition to Late Antiquity 479 546. 57 Curta, The Making of the Slavs ch. 4.

PAGE 76

76 Some historians believe that the permanent contact between different populations gradually led to the creation of a homogeneous and distinct cultural area at the frontier. 58 The applicability of such a definition to the Danube frontier in the sixth centu ry will be tested against the archaeological evidence in the next chapter. Suffice it to say at this point that the Byzantine administration itself did not see this region as entirely homogeneous and a clear difference was made between the Danubian provinc es and the lands stretching north of the river. According to Procopius, Justinian reorganized the Danube frontier "wishing, as he did, to make the Ister River the strongest possible line of first defense before them and before the whole of Europe" 59 In add ition he covered the Balkans with a network of fortifications claiming that "each farm either has been converted into a stronghold or lies adjacent to one which is fortified." 60 This must have created a clear sense of separation from what was happening nort h of the Danube, despite the existence of a number of bridgeheads on the left bank. Moreover, Edict 13 from 538 stipulated that officers unwilling to assist in the collection of taxes in Egypt would be punished by being sent north of the Danube to defend t he border, and this was undoubtedly meant to be a severe punishment. 61 58 See especially Whittaker, Frontiers of the Roman Empire 223. 59 Procopius, De Aedificiis IV, 1.33, ed. Haury and Wirth, vol. 4, 106; trans. Dewing, vol. 7, 229. The river alone, as a natural obstacle (rather than a barrier), was of course insufficient for the proper protection of the Balkans. This is clearly implied in the advice that Priscus received from the Emperor Maurice, urging hi m to protect the Danube because "the barbarians would not remain quiet unless the Romans kept a very strict guard on the Ister," Theophylact Simocatta, VI, 6.2, ed. de Boor and Wirth, 230; trans. Whitby and Whitby, 167. 60 Procopius, De Aedificiis IV, 1.35 ed. Haury and Wirth, 107; trans. H. B. Dewing, vol. 7, 229. 61 Corpus Iuris Civilis universa cohors e regione mota in loca quae ultra Istrum sive Danubium sunt transferatur, ut illis limitibus custodi ae causa adhaereat ."The banishment of criminals or even higher ranked officials fallen into disgrace to peripheral regions seems to have been common practice, as evidenced by the case of Petra in Palaestina Tertia in the sixth century, for which see Z. T. Fiema, "Late antique Petra and its Hinterland: Recent Research and New Interpretations," in The Roman and Byzantine Near East. Volume 3: Late antique Petra, Nile Festival

PAGE 77

77 3.3 Putting the Danube into Perspective: The Early Byzantine Frontier in the East The Empire does not seem to have had a "Grand Strategy" at the frontier, in the manner described by Lutt wak. The lack of a standardized policy, however, should not lead to the opposite conclusion that different frontiers of the early Byzantine state had very little in common in terms of scope and organization. 62 More often than not it was local geography and the challenge of dealing with very different populations that shaped the Byzantine response at the respective frontier. At a time when the Empire was relying on a dense network of fortifications in the Balkans and at the Danube, in the East the Limes Arabi cus was being stripped of any military function and ultimately abandoned. The thorough exploration of the Arabian limes in central Jordan and in the Dead Sea region has yielded important evidence regarding the function of the frontier in the sixth century and is a useful point of comparison with the Danube frontier. Unlike fortresses at the Danube, the forts of the Arabian frontier show no evidence of coin circulation after c. 550 and other categories of finds point to the conclusion that they had been aba ndoned in the second half of the sixth century. Out of sixteen forts studied by Complex, and Other Studies ed. J. H. Humphrey, JRA Suppl. 49 (Portsmouth, RI : Journal of Roman Archaeology L.L.C., 2002), 193 and n. 20. 62 The comparison between different frontier regions during Late Antiquity has gained more ground in the recent past, see for example J. Crow, "Amida and Tropaeum Traiani: a Comparison of Late Antique Fortress Cities on the Lower Danube and Mesopotamia," in The Transition to Late Antiquity 435 55; Comparisons between the Roman frontiers and frontiers from other historical periods is also a worthy exercise, for which see recently Wilmott, "Towers and Spies," 127 33; P. Mayerson, "The Saracens and the Limes ," BASOR 262 (1986): 39; S. L. Dyson, "The Roman Frontier in Comparative Perspective: the View from North America," in Frontires d'empire 149 57, as well as his earlier article, S. L. Dyson, "The Role of Comparative Frontier Studies in Understanding the Roman Frontier," in Actes du IX e Congrs international d'tudes sur les frontires romaines, Mamaia 6 13 sept. 1972 ed. D. M. Pippidi (Bucharest/Cologne: Editura Academiei Republicii Socialiste Romnia/Bhlau, 1974), 277 83. For a more skeptical view of comparative methods, seen from an anthropological perspective, see Kopytoff, "The Roman Frontier," 143 48.

PAGE 78

78 T. S. Parker and his Limes Arabicus team, only three yielded early Byzantine remains, while none of the observation towers yielded any early Byzantine pottery. 63 It has already been established that Strata Diocletiana leading from the Euphrates to Damascus and the old Via Nova Traiana leading to Aila on the Red Sea served the double purpose of controlling and monitoring traffic and ensuring the rapid redeployment of troops towar ds the front in Mesopotamia. Watchtowers played a major part in policing the frontier trunk road, and the lack of evidence of sixth century occupation automatically leads to the conclusion that the proper functioning of the military border was no longer po ssible. To the south, however, the frontier south of Jebel Druz shows signs of occupation at least until the end of the sixth century, while the "soldiers" archive from Nessana mentions the transfer of troops in the region towards the end of the century. 64 In addition, papyrological evidence from sixth century Petra alludes to the presence of troops that received the regular annona. 65 West of the Dead Sea, at En Boqeq and Upper Zohar, the coin circulation ended c. 550, much like in the case of El Lejjun, stud ied by Parker. 66 Nonetheless, the pottery found in the two Palestinian forts shows some sort of occupation up to the Ummayad period. As a rule, more and more settlements produced sixth and seventh century pottery assemblages as one proceeds westward of th e frontier road. On the other hand, no specifically 63 Parker, ed. The Roman Frontier 48. 64 R. Alston, "Managing the Frontiers. Supplying the Frontier Troops in the Sixth and Seventh Centuries," in The Roman Army and the Economy ed. P. Erdkamp (Amsterdam: J. C. Gieben, 2002), 400 1. 65 Fiema, "Late Antique Petra ," 230. 66 M. Gichon, ed., En Boqeq: Ausgrabungen in einer Oase am Toten Meer (Mainz am Rhein: P. von Zabern, 1993); R. P. Harper, ed., Upper Zohar: An Early Byzantine Fort in Palaestina Tertia: Final Report of Excavations in 1985 1986 (Oxfor d: Oxford University Press, 1995); Parker, ed. The Roman Frontier 48 9.

PAGE 79

79 military items have been found at Upper Zohar and the conclusion of the excavator was that the fort appeared to be a temporary campsite, as seemed to be the case with many settlements in the region. 67 In t he northern sector, along the Euphrates, the fortifications were meant to block the passage towards northern Syria and Mesopotamia, and that was clearly aimed against the Persian threat. 68 The Euphrates had been the main defensive line against the Parthian s since the early Empire and Procopius described in detail the careful work of fortifying and securing this region. 69 A general pattern of defending the frontiers emerges from the panegyrical work of Procopius but in reality it was the escalation of hostili ties on various fronts that eventually made the evolution of early Byzantine frontiers quite different. As we have seen, the Limes Arabicus lost its strategic function around 550 at the latest. Although, the date dovetails quite nicely with the claim made by Procopius that Justinian demobilized the limitanei it was in fact a much longer process going back to the fifth century. 70 What Procopius failed to tell us in his diatribe against Justinian is that the emperor abandoned the limes for both economic and military reasons. The system was not able to fight efficiently against frequent raids of the Lakhmids and it was also a big burden on an increasingly impoverished imperial 67 Parker has recently suggested that the two forts were built to offer some protection for the fertile lands to the west; they might stand as indirect evidence that the defensive line across Negev did no longer exist at the beginning of the sixth century, see S. T. Parker, "The Roman Frontier in Southern Arabia: A Synthesis," in The Army and Frontiers 150. 68 S. P. Kowalski, "The Defenses of the Syrian Frontier of the Roman Empire in t he 4th 6th Cent. A. D," In The Roman Frontier at the Lower Danube 4th 6th Centuries, ed. M. Zahariade (Bucharest: Romanian Institute of Thracology), 37. 69 Procopius, De Aedificiis II, ed. Haury and Wirth, vol. 4, 45 82. For the role of the Euphrates, see Nicasie, Twilight of Empire 134 with n. 51. 70 Procopius, Historia Arcana 24.12 14, ed. Haury and Wirth, vol. 3, 148 49. For the evolution of the frontier, see G. Fisher, A New Perspective on Rome's Desert Frontier," BASOR 336 (2004): 49 60.

PAGE 80

80 treasury. With a bold and radical move, the limes system was replaced with a Ghassan id superphylarchy, which was expected to be more efficient and less costly. 71 Indeed, the state of affairs at the Danube was rather different. Not only that the coin circulation did not end during the reign of Justinian, but the frontier garrisons seem to have been strengthened by Justin II, whose opposition to securing peace through external payments is well known. Most fortresses remained functional until the last decade of the sixth century or even later, although the success of the frontier system with its in depth defensive layers is more problematic, especially if we believe Agathias' claim that Zabergan's Cutrigurs proceeded towards Constantinople in 559 after "finding the area deserted and advancing unopposed." 72 What seems even more interesting is th e role of the buffer areas located in the immediate proximity of frontiers which, on many occasions, were part and parcel of the frontier policy. Looking at the northern frontier of the Western Roman Empire, Lotte Headeger has offered a model of interactio n which supposes the existence of three interrelated structures: a Roman system (the peripheral provinces), a system of client kingdoms acting as a buffer zone between Rome and the Germanic tribes, and the independent Germanic system. The buffer zone maint ained strong connections with the Empire, which can be seen in the large quantity of Roman artifacts and coins. It is known from literary sources that the Ghassanids in Arabia and the petty kingdoms of Transcaucasia represented such buffer zones between ea rly Byzantium and the Sassanid Empire, although the material evidence for these regions has not yet been 71 For the ro le of the Ghassanids, see Shahd, Byzantium and the Arabs vol. 1, 35 42. 72 Agathias, Historiae V, 11, ed. R. Keydell; trans. J. D. Frendo, 146.

PAGE 81

81 fully explored. In order to gain a proper understanding of interaction on the frontier it is important to test for the existence of such a buffer zone north of the Danube against the archaeological evidence.

PAGE 82

82 CHAPTER 4 CULTURAL ENCOUNTERS ON THE LOWER DANUBE FRONTIER AND BEYOND: AN ARCHAEOLOGICAL PERSPECTIVE 4.1 The Cultural Background After abandoning the trans Danubian province of Dacia the political a nd military frontier retreated on the Danube, which became once again a convenient line of separation and protection against external threats. Indeed, early Byzantium held its position on the Lower Danube for more than three centuries after Emperor Aurelia n decided the tactical retreat south of the Danube between 271 275 Nevertheless, it is important to recognize that the Danube would never constitute a cultural frontier of exclusion. A major communication artery in the second and the third century after t he Empire conquered Dacia, the Danube continued to facilitate contact between its two banks as attested by fourth century peace negotiations between the Romans and the Goths who insisted on maintaining acc ess to Roman goods through trade points along the L ower Danube. The river acted as an interface between Byzantium and barbaricum not only for customary items required by the tribal aristocracy, but also for item s prohibited from export like precious me tals, weapons, and goods like oil and wine, whose flow north of the Danube could not be prevented. 1 Although c ultural interaction on the Danube frontier should not be necessarily seen as part and parcel of the Empire's policy toward the populations from Barbaricum the desire for Roman goods was oft en exploit ed for political gain Despite the lack of any written records attesting the regularization of trade in the sixth century, similar to the ones available for the fourth and fifth centuries, it is safe to assume that the issue did 1 S. Patoura, "Ho Dounabes stis istoriographikes peges kata ten periodo tes metanasteuseos ton laon; mythoi k ai pragmaikoteta," 9 (2002): 410 and n. 55.

PAGE 83

83 not escape Justinian's thor ough administrative overhaul of the Lower Danubian provinces. Much of the exchange was probably conducted through the Byzantine bridge heads on the northern bank of the Danube which could have acted as trade ports. At least part of the trading was also th e result of individual pursuits, with Byzantine merchants looking to make a profit by bringing coveted Byzantine items to communities in b arbaricum, or mercenaries, traders, or prisoners f rom such communities being in closer contact wi th the Empire The ea rly Byzantine government was unable to monitor such traffic so as to exclude any cultural contact through private initiative and there is little indication that such a separation was ever intended in the first place. Although interaction took place, the r egions north and south of the Danube were by no means developing similar cultures during the long sixth century (Figure 4 1). The dissolution of the Chernyakhov culture and the subsequent chaos brought by the Hunnic onslaught in the fifth century led to in creased cultural fragmentation in barbaricum The imperial machine regained momentum toward the end of the fifth century and repaired most of the damage produced by the Huns in the Danubian provinces. North of the river, however, the situation was quite di fferent and the remnants of the Hunnic confederation could not achieve the homogeneity and spread of the Chernyakhov cultural horizon. In this climate of political and cultural confusion a number of cultural traditions came to be recycled, though they had never been totally abandoned, relying on a mix of Roman provincial, Carpic, Sarmatian, and Chernyakhovian influences.

PAGE 84

84 Surprisingly in the case of pottery, which is the most common artifact found in the frontier region, there is not much borrowing from co ntemporary wheel made pots produced in the Danubian provinces. A study based on the ceramic assemblage from Iatrus on the Lower Danube has shown that the ceramic types typical for the frontier fortresses on both sides of the Danube did not influence the ce ramics produced in barbaricum 2 Different regions in b arbaricum, even the ones closest to the Danube, display distinct patterns of ceramic production. In most cases, the ceramic shapes in the lands north of the Danube follow the different cultural traditio ns developed by the mos aic of populations that dominated the region at one time or another since the early Roman period For example, a t Boto ana, in Moldavia, there is a clear dichotomy between the hand made pottery of so called Slavic tradition and the wheel made pottery which clearly follows a Roman tradition, whi le in other settlements from Buk ovina the study of ceramic assemblages has led to the conclusion that the region was already dominated by Slavic influences in the sixth century. 3 Moreover several additional influences can be traced to the Chernyakhov tradition and the Carpathian barrow culture in settlements from Bukovina such as Ra shkiv and Kodyn. 4 The co existence of hand and wheel made pottery is best exemplified by the finds from Dulceanca in Wallachia. The most significant indication is the fact that both types were clearly produced together, as evidenced by a kiln found in th e settlement 2 veacului al VII 70. http://www.mnir.ro/publicat/TTW/index_est.html (Visit of 03.07.2011). 3 D. G. Teodor, Suceava) (Bucharest: Editura Academiei VI VII d. H.", AM 17 (1994): 223 51. 4 Teodor, "Ceramica de uz comun," 95 97.

PAGE 85

85 Dulceanca I, where a large wheel made bowl was associated with eleven hand made pots. 5 To be sure, almost half of the ceramic assemblage from Dulceanca was made po ttery is predominant, including some gra y gritty pots typical for the Byzantine frontier fortresses. 6 The pottery becomes more "Roman" as one moves from east to west, into the teritory of the former province of Dacia. For instance, the storage capacity of the ceramic containers from Grop ani seems to follow the Roman system. 7 Moreover, Wallachia represents a cultural horizon clearly dominated by wheel made pottery whose initial phase has been traced back to the end of the fifth century and the beginning of the sixth. 8 Clearly the regions west of the Olt, which were once part of the province of Dacia had a closer connection to developments from the Empire. The ceramic assemblages from the area of modern Bucharest show a perplexing diversity of types and influences given their close proximity. The pottery from Ciurel sometimes considered to be early Slavic has been attributed recently to a Chernyakhov tradition, the of the Carpi who dominated Moldavia in the second and third centuries. At Lunca we encounter one of the rare occasions when the Byzantine influence can be 5 S. Dolinescu Ferche, (Bucharest, 1974), 90. 6 Dolinescu Ferche, fig. 174 175. 7 Teodor, "Ceramica de uz comun," 242 43. O. Toropu et al., "Noi descoperiri arheologice n Oltenia", Drobeta 2 (1976): 98 102. 8 P. Roman and S. Dolinescu Ferche, materiale autohtone din sec. al VI lea n Muntenia)", SCIVA 29, no.1 (1978): 89 90.

PAGE 86

86 Fig ure 4 1. Early Byzantine frontier provinces and regions in barbaricum referred to in the text (with dotted line the former province of Dacia).

PAGE 87

87 traced on the local production of wheel made pottery. 9 In addition, the "Slavic" cemetery Monteoru, by far t he largest in central and south eastern Europe, boasts a series of ceramic urns, whose shape have been recently ascribed to a local tradition despite the funerary ritual, which is believed to be early Slavic. 10 In what concerns the dating of the complexes two coins of Justinian found at Bucharest settlements toward the middle of the sixth century, although the arch a eological context of the coins invites caution. 11 A common characteristic i s the higher concentration of wheel made pottery in the first half of the sixth century, although this tends to become a circular argument as dating often relies on the proportion between hand and wheel made pottery. The assumption is that forms beca me mo re simplified in the course of the sixth cent ury as the Slavic influence became more pronoun ced and hand made pottery started to dominate the cera mic assemblages. This tendency wa s not exclusive to the Lower Danube region. A similar process seems to have b een at work in Cyprus where good quality ceramics co existed with an increasing number of hand made cooking vessels, which was seen as an adaptive response to the changing social and economic realities. 12 9 Ciurel: S. Dolinescu Ferche, "Ciurel, habitat des VI e VII e sicles des notre re," Dacia 23 (1979): 179 Ciurel?" Istros 10 (2000): 491 Lunca, see Teodor, "Ceramica de uz comun," 117 18, 122. 10 Teodor, "Ceramica de uz comun," 133 34. 11 Both coins were found outside the sunken featured buildings, in the so called "cultural layer." See M. CAB 2 (1965): 182 and 189, fig. 93. 12 M. Ra utman, "Handmade Pottery and Social Change: The View from Late Roman Cyprus, Journal of Mediterranean Archaeology 11, no. 1 (1998): 95.

PAGE 88

88 Figure 4 2. "Non Roman" hand made pottery from t he frontier provinces. 1. Dinogetia; 2. Niculi 3. Tulcea; 4. Halmyris; 5. Troesmis ; 6. Beroe 7. Halmyris; 8. Callatis (Mangalia); 13. Dervent; 14. Tropaeum Traiani (Adamclisi); 15. Golesh; 16. Durostorum (Silistra); 17. Ni griniana; 18. Nova Cherna; 19. Iatrus (Krivina); 20. Sturmen; 21. Beroe (Stara Zagora); 22. Karasura (Rupkite); 23. Kosloduy; 24. Augusta; 25. Vulchedrum; 26. Jakimovo; 27. Kabyle; 28. Dorticum (Vruv); 31. Veles nica.

PAGE 89

89 If the Empire's influence is rarely seen in sixth century ceramics produced in barbaricum many early Byzantine fortresses in the frontier region have yielded a variety of hand made pots, typical for the regions north of the Danube (Figure 4 2). The ceramic types are not restricted to the well known Penkovka and Prague groups usually associated with the Antes and the early Slavs, respectively. 13 The inventory of hand made pottery includes shapes deriving from the pre Roman tradition or influenced by th e Sarmatic or Germanic groups settled in the frontier region. Based on such influences, fortresses located in the same province show major differences in the hand made pottery assemblages found on the site. In Scythia the majority of shapes from Dinogetia, Beroe, and Halmyris belong to the Penkovka type, while at Capidava the Sarmatic tradition is much more powerful. 14 Since hand made pots appear at a great number of sites from the northern Balkans, the hypothesis that the garrisons were partly recruited fr om barbaricum or a multi ethnic environment gains more credibility. 15 13 For the Prague type, see F. Curta, "The Prague Type: A Critical Approach to Pottery Classification," AB 5 (2001): 73 1 06. For the Penkovka culture, see recently B. S. Szmoniewski, "The Antes: Eastern 'Brothers' of the Sclavenes?" in Neglected Barbarians ed. F. Curta (Turnhout: Brepols, 2010), 53 82. 14 E. S. Teodor, "Handmade Pottery from the Late Roman Fortress at Capida va," in Between the Steppe and the Empire. Archaeological Studies in Honour of Radu Harhoiu at 65th Anniversary ed. A 23; F. Topoleanu and E. S. Teodor, "Handmade Pottery from Halmyris and Its Cult ural Context," Peuce 7 (2009): 347 60; D. Vlceanu and A. SCIVA 26, no. 2 (1975): 209 aves au sud du Danube durant les VI e VII e sicles d'aprs quelques donnes archologiques de Dobroudja," in I 18 IX 1965 Ossolineum, 1970), 322 i romano bizantine din secolele VI VII la geneza ponto danubian," in dacilor 1980), 193 VII ss.," in Istoriia i kul'tura drevnik i srednevekovykh slavian ed. V. V. Sedov (Moscow: Editorial URSS, 1999), 301 13. 15 Wedrowka i etnogeneza w starozytnosci i sredniowieczu ed. M. Salamon and J. Strzelczyk (Cracow: Towarzystwo Wydawnicze Historia Iagellonica, 2004), 205; S. Angelova and R. Koleva, "Archologische Zeugnisse frhslawischer Besiedlung in Bulg arien," in Post Roman Towns, Trade and Settlement in Europe and Byzantium. Vol. 2: Byzantium, Pliska, and the Balkans ed. J. Henning (Berlin/New York: Walter de Gruyter, 2007), 281 307.

PAGE 90

90 The diverse traditions behind the production of hand made pottery, as in the cases presented above, might suggest that the "barbarians" themselves, recruited to defend the Empire's front ier, belonged to different cultural backgrounds. To be sure, there is a risk in ascribing the hand made pottery to outside groups alone. The example from Cyprus warns us that more profound social and economic transformations are likely behind the decline o f ceramic production and the shift toward a more localized production of hand made pottery. 16 The archaeological context in which the hand made pots have been found at Capidava in Scythia, for instance, presents us with a perplexing association of such "inf erior" types with a large quantity of regular Roman amphorae and cooking wares. The complex, which was destroyed probably in the late 570s, reflects such cultural changes, in which the new recruits from barbaricum may have compounded the general economic d ecline of the Empire. 17 The apparent lack of dialogue between the regions south and north of the Danube in the case of sixth century ceramics can be ascribed to the more conservative nature of ceramic production and the greater resist a nce in replacing tradi tional ceramic shapes. In addition, a variety of objects could fill the need to follow "fashions" from the frontier provinces. Indeed, in order to look more "Roman", populations in barbaricum chose to imitate different categories of items like amphorae, la mps, buckles, and brooches or to import them directly from the Empire. A closer investigation of the main categories of imports and imitations will shed more light on the nature of cultural contact in the frontier region, and the factors favoring or limit ing its extent. Cultural contact could 16 A similar interpretation has been long suggested as a possible expl anatio n for the appearance of hand 17. 17 CN 12 13 (2006 2007): 103.

PAGE 91

91 mean commerce in some cases although t rade remains an elusive activity on the sixth century Danube frontier. The exact goods offered by "barbarians" in return for Roman products are very hard to identify in the archae ological record due to their perishable nature. A far more promising avenue by which to explain the circulation of Byzantine artifacts is a closer examination of the movement of people in the context of the political, diplomatic and military developments i n the frontier region, documented both by archaeological and literary sources. 4.2 Categories of Imports and Imitations in barbaricum For lack of a better term imports will be defined here as any object produced in the Empire and brought to the territories beyond the Danube limes through economic or non economic channels, either as basic goods or luxury/prestige items. Besides coins, which will be discussed in a separate chapter, ceramics (amphorae, lamps), metal items (fibulae, buckles, jewelry), and a num ber of Christian objects that run across these categories constitute the main types of imports found north of the Danube by archaeologists Much has been made of the presence of such objects in barbaricum in an attempt to demonst rate the intensive nature o f relations between the Empire, its former trans Danubian province of Dacia, and the region east of the river Olt and east of the Carpathians, which had been exposed to Roman culture for a long time. 18 Since many 18 M. Constantiniu Elemente romano bizantine n cul tura material autohtone din partea Munteniei n secolele VI VII e.n. ", SCIVA 17, no. 4 (1966): 665 78; D. G. Teodor, "Elemente bizantine n Moldova, n sec. VI XI ," SCIV 21, no.1 (1970): 97 128; Z. Szkely, "lments byzantins dans la civilisation matrielle des VI VII sicles dans le Sud Est de la Transylvanie," Dacia 15 (1971): 353 58; D. G Teodor, Romanitatea carpato n veacurile V XI ( : Junimea, 1981); I. Barnea, "Sur les rapports avec By zance du territoire situ au Nord du Bas Danube durant la priode Anastase I er Justinien I er (491 565), tudes Byzantines et Post Byzantines 2 (1991): 47 57; D. G. Teodor, "lments et influences byzantins dans la civilisation des VI e VII e sicles aprs J. Chr. au nord du Bas Danube," tudes Byzantines et Post Byzantines 2 (1991): 59 72.

PAGE 92

92 such interpretations were connected more wit h certain ideological purposes than a balanced and impassionate assessment of hard evidence, it is necessary to commence with a survey of the main categories of Byzantine artifacts and their geographic distribution in barbaricum. As already demonstrated, the Danube was the desired political frontier, but it did not act as a cultural frontier of exclusion, as objects, ideas, and fashions traveled more or less freely north and south of the river. Rather than being a comprehensive treatment of all categories of imports, the following sections address a selected number of diagnostic items that lend themselves to a multifaceted analysis due to their wide distribution, variety of types, and multiple implications in the realm of social, economic, and religious rea lities of the frontier area and the regions beyond. To be sure, such an approach might obscure the diversity of items that found their way north of the Danube during the long sixth century. Various small finds of Byzantine origin, usually pieces of jewelry sometimes made of precious metal have been found in barbaricum but the lack of context reduces their usefulness for answering the type of historical questions which interest us most. Although most of the objects have been published, sometimes repeatedly and with additional comments, especially in the case of artifacts with religious significance, the archaeological context is most of the time unknown. The main categories of imports mentioned above are themselves not free of such problems but there are, h owever, sufficient finds in secure contexts to allow for certain historical interpretations and conclusions. In addition, aside from genuine imports from the Danubian provinces or more distant corners of the Empire, it is important to note the presence of local imitations of

PAGE 93

93 almost all categories of imports as well as metallurgical tools as evidence of local production. Although these are not Byzantine objects per se they can still be treated as indirect evidence of contact with the Empire. The unavailabili ty of imports to meet the demand or the the lack of means to procure genuine Byzantine imports make such imitations relevant for a better understanding of the circulation of both Roman goods and the ones modeled after the original prototypes. 4.2.1 Ampho rae The early Byzantine amphora is one of the most common artifacts found in frontier fortresses from the northern Balkans, being primarily related to the state controlled annona system designed to supply the frontier garrisons. 19 Unfortunately, very few si tes have so far benefited from quantitative studies of well stratified ceramic assemblages. Based on the available archaeological monographs, the amphora occasionally accounts for almost half of the ceramic remains from a site. I n most cases either LR1 or LR2 is the most frequent type. 20 This is hardly the case north of the river, 19 Amphora types Late Roman 1 (LR1) and Late Roman 2 (LR2) are the most common finds in ceramic assemblages from fortresses of the northern Balkans. For the role of the LR2 amphora on the Danube frontier see O. Karagiorgou, "LR2: a Container for the Military annona on the Danubian border?" in Economy and Exchange in the East Mediterranean during Late Antiquity. Proceedings of a conference at Somerville College, Oxford 29 th May, 1999 ed. S. Kingsley and M. Decker (Oxford: Oxbow Books, 2001), 129 66. 20 Capid VI p. Chr.) Halmyris: F. Topoleanu, "Ceramica", in A. Suceveanu et alii, Halmy (Cluj Napoca: Nereamia Napocae, 2003), 226 27. Sadovec: M. Mackensen, "Amphoren und Spatheia von Golemanovo Kale," in S. Uenze et al., Die Sptantiken Befestigungen von Sadovec (Bulgarien) (Munich: andlung, 1992), 239 54. lampes," in ouest de la ville haute ed. B. Bavant, V. Kondic, and J. M. Spieser (Rome: Ecole Franaise de Rome, 1990), 174 76 and pl. XXI. Aquis: . J Podunavski deo oblasti Akvisa u VI i pochetkom VII veka 53. Ratiaria: G. Kuzmanov, "Ceramica del primo periodo bizantino a Ratiaria," Ratiarensia 3 4 (1987), 115 16. Dichin: G. bitova keramika ot Gradishteto (severozapaden sector)," in V. Dinchev et al., Bulgaro britanski razkopki na gradishteto pri s. Dichin, Velikoturnovska oblast, 1996 2003: rezultati ot prouchvaniiata na bulgarskiia ekip (Sofia: Bulgarska akademiia na naukite 2009), 174 78; V. G. Swan, "Dichin (Bulgaria): Interpreting the Ceramic Evidence in its Wider Context," in The Transition to Late

PAGE 94

94 in barbaricum With the exception of the major bridge head of Sucidava, the rural settlements in barbaricum have produced only a small qu antity of amphora fragments (Figure 4 3). F rom an economic standpoint it is surprising that a high density of amphora finds can be seen not only in the close proximity of the Danube, but also a few hundred kilometers to the north, in Moldavia. The territory of the former province of Dacia has also produced a significant number of amphora finds, which along with other categories of impor ts discussed below, points to easier access to goods from the Empire. As for the region west of the river Olt, it is a common caveat in archeology that blank spots on a map can often point to unexplored areas (or unpublished material) rather than to a real difference in distribution. Nevertheless, it has to be asked whether such amphorae with a large circulation in the Danubian provinces performed a similar function in barbaricum If such containers we re indeed related to annona militaris their scarcity or absence in barbaricum can be explained by the lack of any military presence beyond the bridge heads established on the northern bank of the Danube. Moreover, the few finds of LR1 and LR2 north of the river are closer to what can be described as evidence of trading activity than the amphorae found in the Lower Danube fortresses, which were state controlled shipments and not the result of commerce properly speaking. Unl ike dress accessories, discussed below, amphorae were too large to Antiquity on the Danube and Beyond ed. A. G. Poulter (Oxford: Oxford University Press, 2007), 260 65. Nicopolis ad Istrum: R. K. Falkner, "The Pottery," in A. G. Poulter, Nicopolis ad Istrum: A Roman to Early Byzantine City. The Pottery and Glass (London: Leicester University Press, 1999), 117, fig. 8.4. Sacidava: C. Scorpan, "Ceramica romano Pontica 8 (1975): 263 313. Iatrus: B. Btger, "Die Gefkeramik aus dem Kastell Iatrus," in B. Btger et alii, Iatrus Krivina. Sptantike Befestigung und frhmittelalterliche Siedlung an der unteren Donau. Band II: Ergebnisse der Ausgrabungen 1966 1973 (Berlin: A kademie Verlag, 1982), 33 148. At Viminacium LR2 and LR1 types Contribution to the Study of the Early Byzantine Viminacium," Starinar 38 (1987), 36. On the other hand, at Sarahane, in the Byzantin e capital, LR1 and LR2 account for ca. 25 percent of the sixth and seventh century deposits, for which see Hayes 62 71.

PAGE 95

95 count as "souvenirs" and could not be di splayed (i.e. worn) as objects of prestige. On the other hand, a ccess to oil or wine from the Empire could, of course, enhance one's social status. A n economic function can also be ascribed to amphora finds from the western edge of the Balkans, in Slovenia, where a good number of sites of the frontier region, (Capris/Justinopolis, on the Adriatic coast) produced a conspic uously large quantity of spatheia customarily produced in the Western Mediterranean. 21 Similar channels of distribution might be responsible for finds of spatheia from Klked Feketekapu, an Early Avar settlement on the right bank of the Middle Danube. This is also reflected in Feketekapu suggests a route coming from the Western Balkans and even Italy, rather than the usual provenance from the Lower Danube provinces and the wes tern Black Sea region. 22 Klked Feketekapu seems to be at a crossroads in that respect, since it also yielded LR1 and LR2 amphorae. These types are also found in burial assemblages of the Middle Avar period from the region between the rivers Tisza and the D anube, such as Tiszavasvri and Kunbbony. 23 These and other finds of amphorae in "Avaria are associated with ceramics produced in the western Black Sea area, which arrived after the collapse of the Danubian frontier system, since most of the finds are 21 For Spain and North Africa in particular, see D. P. S. Peacock and D. F. Williams, Amphorae and the Roman Economy. An Introductory Gu ide Mediterrane Feinkeramikimporte des 2. bis 7. Jahrhunderts n. Chr. im oberen Adriaraum und in Slowenien (Espelkamp: M. Leidorf, 1996). 22 A eulogy flask with no analogies in the eastern Balkans ad ds weight to this proposition, see Z. Hajnal, Communicationes Archaeologicae Hungariae (2005): 477 80. 23 T. Vida, Die Awarenzeitlische Keramik I. (6. 7. Jh.) (Berlin/Budapest: Balas si Kiad, 1999), pl. 38/3 and pl. 39/3.

PAGE 96

96 ass ociated with material securely dated to the Middle Avar period. 24 Tivadar Vida has suggested that such amphorae represent a symbol of prestige for the Avar nobility who received such items in their burial assemblages, but they also reflect the nature of the relations between the Empire and the Avar khaganate. 25 A similar interpretation was recently suggested for the fragments of amphorae found in settlements belonging to the so called Ipote ti presence of amphorae must be connected to a local elite that chose to display its economic power through such artifacts imported from the Empire. 26 Indeed, some amphora fragments have been retrieved from houses that stand out as being larger and richer, as it is clearly the case of the building B10 at Bucharest Soldat Ghivan, where fragments of several imported amphorae were associated with a local imitation of an early Byzantine amphora, a ladle used to pour metal, and a mold used to produce jewelry. 27 Although the outstanding house at Soldat Ghivan displays the largest number of amphorae to be found in a single building north of the Danube, there are other 24 For other amphora finds see Vida, Die Awarenzeitlische Keramik 242 43, and . Garam, Funde byzantinischer Herkunft in der Awarenzeit vom Ende des 6. bis zum Ende des 7. Jahrhunderts eum, 2001), 166 68, pl. 123 24. 25 The practice of including amphorae in burials is not restricted to the Avar khaganate; an early Byzantine amphora was found in an inhumation grave (M 21) from Kushnarenkovo (Bashkortostan, Russia) together with amber bead s, bronze and silver fibulae, and a buckle, for which see V. F. Gening, "Pamiatniki u s. Kushnarenkovo na r. Beloi (VI VII vv. n.e.)," in Issledovaniia po arkheologii Iuzhnogo Urala ed. R. G. Kuzeev, N. A. Mazhitov, and A. Kh. Pshenichniuk (Ufa: Institut istorii, iazyka i literatury AN SSSR, Bashkirskii filial, 1977), 102, fig. 9/1. 26 Seventh Century in the Extra Carpathian Area," in Potestas et Communitas. Interdiciplinary Studies of the Constitution and Demonstrati on of Power Relations in the Middle Ages East of the Elbe ed. Akademii Nauk & Deutsches Historisches Institut Warschau, 2010), 80 81. 27 S. Dolinescu Ferche and M. Constantinescu, "Un tabl issment du VI e sicle Bucarest (Dcouvertes de la rue Soldat Ghivan)," Dacia 25 (1981): 320, with fig. 17/16 and 323, fig. 19/1.

PAGE 97

97 Figure 4 3 Amphora finds in barbaricum 1. Bucharest Soldat Ghivan; 7. Vedea; 8. 25. Tiszavasvri; 26. G tr; 27. Kunbbony; 28. Klked Feketekapu; 29. Fntna

PAGE 98

98 sunken fea tured buildings in the same settlement that provide artifacts that could very well have play ed the role of prestige items, such as fibulae or even the hand made lamp found in the building B5, an artifact rarely found in barbaricum Since any artifact impo rted from the Empire could in principle bear a certain social value, what is at stake here is the relative value of amphorae among other imported items. Following this line of thought it is worth mentioning that Soldat Ghivan is just one of the sixth centu ry settlements in Bucharest to provide such finds. Indeed, no less than five other locations in Bucharest have yielded amphora fragments, 28 of which only the sunken featured an, being larger than the other houses. The inventory, however, is less spectacular, although it boasts three storage jars of 100 liters each. 29 The presence of metallurgical implements at Soldat Ghivan (B10) can also be traced in other settlements as well south and east of the Carpathians, at Dulceanca amphora fragments are associated with scrap pieces of bronze and in the case of Dulceanca, the sunken featured building B2 also produced a stone mold for jewelry. 30 Moreover, their association of amphora fragments with brooches or buckles in houses from Dulceanca, Olteni, and Hansca 28 CAB 1 (1963): 39 with fig. 6 (1968), 125 fig. 6/8. Bucharest 1959), 33 34, 93 96, pl. LXXIII/1 3; Bucharest 91 94. 29 CAB 2 (1965): 185, with fig. 83/5. 30 Dulceanca: Dolinescu Ferche, ele V Carpica 18 19 (1986 1987): 247, fig. 14/4.

PAGE 99

99 seems to suggest a higher social status for their owners. 31 However, in many instances amphora fragments are found in inconspicu ous buildings, while some of them do not preserve a record of their original archaeological context, making it hard to determine whether the amphora, and more precisely its content (wine, oil), 32 had a higher symbolic value for those communities in relation to other categories of imports. 33 Amphora finds are not restricted to the region north of the Danube. A LR1 was found as far east as Klimovka on the upper Volga, the easternmost find of early Byzantine amphora e A number of Pontic amphorae belonging to th e type Kuzmanov XVI/ Scorpan IX are known from the region of Kiev. 34 A distribution of amphorae in barbaricum based on typology has already been attempted and it clearly shows that most finds north of the Black Sea, especially on the Dnieper, belong to a s eries of 31 Dulceanca: Dolinescu Ferche, figs. 85/6 8 and 92. Olteni: S. Dolinescu din sec. VI e.n. de la Olteni MCA 10 (1 973): 205. Hansca: I. A. Rafalovich, Slaviane VI IX vekov v Moldavii (Kishinew: Shtiinca, 1972), 32 fig. 3/1 and 38 fig. 8/8. 32 For the possible content of LR1 and LR2 amphorae, see Karagiorgou, "LR2: A Container," 146 49. 33 For other amphora finds north of the Danube, see S. Dolinescu com. Vedea (jud. Teleorman)," MCA VII)," in Sesiunea vol. 2 (Bucharest, 1971), 130 fig. 6/4; G. Popilian and M. Nica, (Bucharest: Institutul Romn de Tracologie, 1998), 23 5; S. Dolinescu Ferche, "Un complex din sec olul al VI SCIV 18, no. 1 (1967): 130 fig. 3/5; G. Popilian and Drobeta 15 (2005), pl. VII/8 (LR2); O. Toropu, carpat XI) (Craiova: Scrisul Romnesc, 1976), 205 17, ns. 52, 53, 76, and 92, and pl. V (LR1 and fragments); D. G. Teodor, Romniei, 1997), ns. 49, 59, 106, 195, 343, 512, and 756; I. Tentiuc, "Siturile din secolele V VII de la Ialoveni (Republica Moldova)," AM 21 (1998): 209; N. Gudea and I. Ghiurco, Din istoria (Oradea: Editura Episcopiei Ortod oxe Romne a Oradiei, 1988), 197. 34 Kiev: A. M. Shovkoplias, "Ranneslavianskaia keramika s gory Kiselevki v Kieve," in Slaviane nakanune obrazovaniia Kievskoi Rusi ed. B. A. Rybakov (Moscow: Izd vo Akademii Nauk SSSR, 1963), 140 fig. 2/1; O. M. Prikhod niuk, Arkheologychny pam'iatki seredn'ogo Pridnyprov'ia VI IX st. n.e. (Kiev: Naukova Dumka, 1980), 130; 63 fig. 44/9. Svetil'ne: Shovkoplias, "Ranneslavianskaia keramika," 140. Klimovka: V. V. Priimak, "Kulturnye transformatsii i vzaimovliianiia v Dneprov skom regione na iskhode rimskogo vremei Doklady nauchnoi konferentsii, posviashchennoi 60 letiiu so dnia rozhdeniia E. A. Goriunova (Sankt Petersburg, 14 17 noiabria 2000 g.) (Sankt Petersburg: Petersburgskoe vostokovedenie, 2 004), fig. 3B.

PAGE 100

100 amphorae produced in the Pontic area, 35 which could suggest trading connections with the Empire, possibly through the Crimea. In contrast, the region north of the Lower Danube produces mostly finds of LR1 and LR2, the typical annona containers foun d in the frontier fortresses of the Danube frontier, especially after the creation of qu a estura exercitus in 537, which brought under the same administrative umbrella some of the production centers of these amphora types. 36 The apparent scarcity of imported amphorae from the first half of the sixth century in the Crimea and the northern Balkans points to major difficulties in supplying the Danube frontier and was probably one of the main reasons for the creation of this peculiar administrative division. 37 Suc h developments can explain the dearth of LR1 and LR2 containers north of the Black Sea, while their presence north of the Danube can be related to the Justinianic reforms and the short lived prosperity of the Danube frontier. Among the amphorae found north of the Danube two appear to be local imitations. One of them was already mentioned, the fragment from the B10 building at Bucharest Soldat Ghivan, found in association with a Byzantine imported amphora, while the second was found at Bratei in Transylvan ia, in the sunken featured building B5, also associated with an imported amphora. 38 Such imitations strengthen the hypothesis that imported amphorae represented one way in which communities in 35 Curta, The Making of the Slavs 244, fig. 37. 36 For quaestura exercitus see recently A. Gkoutzioukostas and X. M. Moniaros, He peripheriake dioiketike anadiorganose tes byzantines autokratorias apo ton Ioustiniano A' (527 565): he periptos e tes quaestura Iustiniana exercitus (Thessaloniki: Banias, 2009). See also S. Torbatov, "Quaestura exercitus: Moesia Secunda and Scythia under Justinian," AB 1 (1997): 78 87. 37 Swan, "Dichin (Bulgaria)," 263. 38 Soldat Ghivan: Dolinescu Ferche and Constan tinescu, "Un tablissment du VI e sicle," 318 19, fig. 16/1, possibly imitating an early Byzantine LR2; Bratei: E. Zaharia, "La station n. 2 de Bratei, dp. de Sibiu (VI e VIII e sicles), Dacia 38 39 (1995), 305.

PAGE 101

101 barbaricum imitated the Roman way of life, while their owners g ained local prestige for being able to procure oil or wine from the Empire. 4.2.2 Lamps Byzantine lamps constitute one of the most important, yet insufficiently explored, categories of imports in barbaricum They were traditional ly associat ed with the Rom an way of life and often connected to Christian practice which demanded such liturgical items. 39 Clay lamps provide the greatest diversity and potential for a study of geographical distribution, but bronze lamps are also important due to their higher valu e and their excellent parallels in the Byzantine world. The Christian bronze lamp from Luciu on the left bank of the Danube, the only such find outside the former province of Dacia, has to be mentioned for its perfect resemblance with lamps found at Archar and Athens, in the Balkans, while the Christian lamp from Tpigyrgye in "Avaria" has good analogies in Dardania at Kalaja, in Asia Minor at Ephesus, and in Crete. 40 Other fifth to sixth century bronze lamps have been found in the former province of Dacia at 39 A EN 5 (1995): 255 99, remains the only synthetic treatment of this category of artifacts, valuable for its contribution to the typology of lamps found in barbaricum 40 Luciu: I. Barnea Christian Art in Romania 1 (3 rd 6 th centuries) (Bucharest: Publishing House of the Bible and Mission Institute of the Romanian Orthodox Church, 1979), pl. 112. Archar: Kuzmanov n. 443. Athens: Perlzweig n. 2984. Another specimen from Greece is now in the Vatican Museum, see M. C. Le Lucerne di bronzo (Vatican: Musei della Biblioteca Apostolica Vaticana, 1986), 68, n. 29. A slightly different lamp was found at Sadovec, for which see Uenze, Die Sptantiken Befestigungen p l. 143/3. A lamp very similar to the one from Luciu was part of a spectacular hoard of coins, jewelry, and liturgical objects recently found in Gaza and buried at the beginning of the seventh century, K. Golan, "Why hide? Hoarding in Late Antiquity in View of the Early Byzantine Hoard from the Gaza Area," presentation at the 22nd International Congress of Byzantine Studies, Sofia, 2011. Tpigyrgye: Garam, Funde byzantinischer Herkunft pl. 133 and XL/4. Ephesus: Miltner n. 356. Kalaja: E. Shukriu, "Frhc hristliche Lampen aus der antiken Provinz Dardania," Mitteilungen zur Christlichen Archologie 9 (2003): fig. 1. Crete: the collection of the Chania Byzantine Museum, http: //www.crete kreta.com/files/u2/ByzantineCollectionBronzeLamp.jpg (visit of 09.23.2011).

PAGE 102

102 origin. 41 This does not necessarily mean tha they were imported directly from Egypt, as such finds are quite frequent in Greece as well. The largest percentage of clay l amps was produced in the Balkans They are usually referred to in the literature as "Danubian lamps because they are mostly found in settlements of the Lower Danube provinces (Figure 4 4). 42 However, it seems that some t ypes were either produced or had pr ototypes in Constantinople as evidenced by numerous finds from the excavations at Sarahane in Istanbul 43 Although this is the most common type of lamp found in Barbaricum "Danubian lamps are not showing up in the quantity we would expect given their hig h frequency in almost every fortress of the frontier region. The lamp from Bumbe ti in Oltenia, dated to the sixth century has a perfect 4 20/5 6). 44 Here, and also to the south along the Black Sea coast, at Kranevo, as well as on the Danube, at Sacidava, molds for this type have been unearthed proving the local production of these lamps in the Eastern 41 Gudea and Ghiurco, Din istoria 143 47. Many similar lamps found their way in western museums, such as the Rmisch Germanischen Zentralmuseum zu Mainz or the British Museum. For a comprehensive list of analogies, see recently M. Xanthopoulou, que palochrtienne (Turnhout: Brepols, 2010). 42 For the typology, see especially Iconomu ; Kuzmanov ; Hayes In the classification created for the lamps found on the Yassi Ada shipwreck, Danubian lamps are found under the label "Balkan type", see K. D. Vite lli, "The Lamps", in G. E. Bass and F. H. van Doorninck, Jr., Yassi Ada. Volume I. A Seventh Century Byzantine Shipwreck (College Station: Texas A&M University Press, 1982), 196 99. The name was also adopted by scholars from the Balkans, for which see K. K ostova and D. Dobreva, "Roman, Late Roman and Early Byzantine Lamps from National Archaeological Reserve "Deultum Debelt"," in Lychnological Acts 2. Trade and Local Production of Lamps from the Prehistory until the Middle Age. Acts of the 2nd International Congress on Ancient and Middle Age Lightning Devices Cluj Napoca, 13th 18th May 2006) Napoca: Mega, 2008), 164. For their frequency in the northern Balkans, see for example Capidava where local lamps account for 66 percent of the total number of finds, see Opr 167, fig 10. A high proportion (over 50 percent) was also noted for the late occupation levels at Halmyris, for which see Topoleanu, "Ceramica ", 244, table 8. 43 Hayes type 11 12. 44 pl. V/1 and the analogy from Tomis at pl. V/2.

PAGE 103

103 Figure 4 4. "Danubian" lamps. Kovilovo Grad; 5. Tropaeum Tra iani (Adamclisi); 6. Novae (Svishtov); 7. Iulia (Apulum); Sadovec; 16. Shumen; 17. Izvoarele; 18. Iatrus (Krivina); 19. Ha Vodenica; Grad; 28. Odurtsi; 29. Veliko Turnovo; 30. Sarahane (Istanbul); 31. Oltina; 32. Vienna; 33. Chersone ssus; 34. Kertch; 35. Drobeta; 36. Romuliana (Gamzigrad).

PAGE 104

104 Balkans, although a similar type is known from Anemurium in Clicia. 45 Very similar lamps have been found at Noviodunum, Durostorum, Halmyris, Dinogetia and Veliko Turnovo, 46 while a late seventh centu ry lamp with a cross at the base of nozzle has been found during excavations at Sarahane, showing the longevity of this design. 47 Another Danubian lamp was found at Alba Iulia (Apulum) and represents one of the most popular types of the period, its main ch aracteristic being the handle in the form of a human head. Alexandru Diaconescu considered that the closest analogies are to be found in the Danubian provinces, as was to be expected for this category of lamps. 48 Indeed, lamps with human shaped handles have been found at Adamclisi, Capida va (Figure 4 Iatrus in Moesia Secunda, and Mokranjske Stene, Romuliana and Aquis in Dacia Ripensis. 49 The circulation of this type wa s not restricted to the Lower Danube region 45 Constan Pontica 9 (1976): 135 45. Kranevo: G. Toncheva, "Keramichna rabotilnica krai s. Kranevo," Izvestiia Bulgarskoto Arkheologichesko Dru zh estv o 9 (1952): 81 104. Sacidava: C. Scorpan, "Descoperiri arheologice diverse la Sacidava," Pontica 11 (1978): pl. VI. Anemurium: H. Williams and P. Taylor, "A Byzantine Lamp Hoard from Anamur (Cilicia)," Anatolian Studies 25 (1975): 80, fig. 4. 46 Noviodunum: Peuce 9 (1984): pl. Bucharest, 1992, 177 78. Murighiol: Topoleanu, "Ceramica", pl. XLVII/ 27; I. Barnea, "Perioada Dominatului (sec. IV VII)," in R. Vulpe and I. Barnea, Din istoria Dobrogei v. 2 (Bucharest: Editura Academiei, 1968), fig. 51/3. V. Turnovo: Kuzmanov n. 305. 47 Hayes type 16, no. 134. Lamps with cross on the nozzle have been labele d "North African" by Kuzmanov, but similar types are found among the Attic lamps from the Athenian agora. It is well known that Greek lamps of the Early Byzantine period were inspired by African types, so that similar lamps produced in the Balkans could ha ve followed the same practice, see Kuzmanov type XXXVI; Perlzweig no. 2933. See also below, n. 70. 48 d. 49 a et alii Tropaeum Traiani. I. Cetatea (Bucharest: Editura Academiei, 1979), fig. 173/ 9.5 and 9.6. Capidava: Z. Capidava," Pontica 24 (1991), fig. 2/ 1 3, 10. Oltina: M. Irimia, "Cuptoarele romano bizantine de ars ceramica de la Oltina," Pontice Iconomu p. 149, n. 778. Beroe: l. LIX/10

PAGE 105

105 since similar examp les appear further to the east, in the Crimea and to the west in Pannonia. 50 However, the specimen from Alba Iulia departs from the common design of the lamps mentioned above, most of which display short strokes on the rim, so typical for "Danubian lamps. Instead, we find a zigzag pattern with small globules reminiscent of the popular lamps produced in Asia Minor during the fifth and sixth centuries and sometimes imitated in the northern Balkans. 51 Anatolian influences on the specimen from Apulum should not be surprising, since lamps with human shaped handles have been found in Constantinople, which was perhaps one of the main destinations of this particular type, as well as at Sardis. 52 The lamp discovered at the nalogies in Scythia, at Tomis (Figure 4 17/2), Capidava, Histria, and Halmyris, to name just a few. 53 and LVI/9. Tomis: I. Barnea et alii, (Bucharest: Editura Meridiane, 1971), 171, n. 344. Novae: A. Dimitrova, "Arkheologicheskie raskopki v vostochnom sektore Nove v 1963 godu," IAI 28 (1965), 60 fig. 31. Iatrus: G. V. Blow et alii, Iatrus Krivina. Sptantike Befestigung und frhmittelalterliche Siedlung an der unteren Donau. Band VI : Ergebnisse der Ausgrabungen 1992 2000 Podun avski deo oblasti Akvisa from the End of 4th to 11th Centuries A. D.," in Keszthely Fenkpuszta im Kontext sptantiker Kontinuittsforschung zwischen Noricum und Moesia ed. O Heinrich Tamska (Leipzig: Marie Leidorf, Podunavski deo oblasti Akvisa, pl. XI/3. 50 Chersonesus: L. Chrzanovski and D. Zhuravlev, Lamps from Chersonesus in the State Historical Museum Moscow hneider, 1998), 172 74, ns. 109 110. Kertch: T. I. Makarova, Bospor Korchev po arkheologicheskim dannym," in Vizantiiskaia Tavrika. Sbornik nauchnykh trudov (v XVIII kongressu vizantinistov) ed. P. P. Tolochko (Kiev: Naukova Dumka, 1991), 135, fig 15/2. Vienna: D. Ivanyi, Die Pannonischen Lampen: eine typologischchronlogische bersicht (Budapest: Numismatic and Archaeological Institute, 1935), n. 1224, pl. LXVII/8 (very similar to the specimen found in Mokranjske Stene, for which Podunavski deo oblasti Akvisa pl. XI/5). 51 A mold for such imitations was found at Halmyris, for which see F. Topoleanu, romano Muzeale, 2000), 211 14. 52 Hayes type 11, pl. 21/64 65; J. Stephens Crawford, The Byzantine Shops at Sardis (Cambridge MA: Harvard University Press, 1990), fig. 93 (shop W 8). 53 Apulum 31 (1994), pl. VII/7. Capidava: Covace century

PAGE 106

106 A similar lamp was found at Sucidava, the most important Byzantine settlement on the left bank of the Lower Danube in Oltenia 54 Several lamps of Balkan type have been found there, among which a few lamps with cross shaped handle, a very popular type on the Lower Danube during the early Byzantine period. One of them was found in the early Byzantine church together with coins dated 587/8 and 596/7, which securely dates the terminus post quem of this type to the end of the sixth and the begining of the seventh century. 55 Similar lamps with different cross shapes and designs on the handl e have been found throughout the northern Balkans, at Halmyris (Figure 4 17/3), Tomis, Capidava, Novae, Aquis, Kovilovo Grad, and on the northern bank of the Danube, at Drobeta. 56 The type is also present in abundance in Constantinople, which might indicat e a possible center of production, especially since the type appears among finds from Anatolian sites, such as Amorium. 57 MCA 7 (1961): fig. 15/2. Halmyris (described as an "oriental" type): Topoleanu, "Ceramica", pl "Ceramica," fig. 173. Ulmetum: V. Prvan, Cetatea Ulmetum III Ist orice 37 (1914 1915): pl. VI/3. 54 Materiale arheologice privind istoria veche a RPR vol. 1 (Buch arest, 1953): fig. 12/f. 55 Tudor, "Sucidava IV," fig. 11/d. 56 Halmyris: Topoleanu, "Ceramica," pl. LXVII/ 10. Tomis: Barnea et alii, 170, n. 338. ra unei categorii," fig. 1/11 12. Novae (2 specimens are associated with a sixth century coin): D. P. Dimitrov et alii, "Arkheologicheskie raskopki v vostochnom sektore Nove v 1962 godu," IAI 27 (1964): fig. 18; D. P. Dimitrov, et alii, "Arkheologicheskie raskopki v vostochnom sektore Nove v 1963 godu," IAI 28 (1965): fig. 30; L. Press et alii, "Novae sektor zachodni, 1971. Sprawozdanie tymczasowe z wykopalisk ekspedycji archeologicznej Uniwersytetu Warszawskiego," Archeologia 24 (1973): fig. 55 and 65. A Podunavski deo obrasti Akvisa pl. XI/2. Kovilovo Grad: Ibid., pl. XI/1. Drobeta: Barnea, Christian Art in Romania pl. 114/1. 57 Constantinople: Hayes type 11, pl. 22 and 24. Amorium: M. A. V. Gill and N. T. "Roman and Early Byzantin e Terracotta Lamps. With a Discussion and Two Appendices by C. S. Lightfoot," in Amorium Reports II. Research Papers and technical Reports ed. C. S. Lightfoot, BAR International Series 1170, 2003, pl. III/5 and III/31.

PAGE 107

107 Another common group of lamps circulating during the sixth century has a palmette or leaf shaped handle, and sometimes a stylized "Tr ee of Life Models displaying a superior craftsmanship have been found in Asia Minor, suggesting a possible source of inspiration for their Danubian counterparts. 58 One specimen with a handle in the shape of an upright sprig from Sucidava has good analogie s at T omis (Figure 4 17/4) and Halmyris in Scythia and Sarahane in Istanbul, 59 while a second specimen resembles finds from Histria, Novae, Sadovec and Shumen, some of them associated with sixth century coins. 60 One last "Danubian lamp found in Barbaricum at Feldioara R zboieni in Transylvania (Figure 4 17/5), has a perfect analogy at Tomis in Scythia and belongs to the type with grapevine motif s, 61 whose prototype might have been an earlier lamp from Asia Minor (Figure 4 17/6), 62 sometimes found in the border province of Scythia (Adamclisi, Sacidava and Halmyris) and also imitated in Greece during the sixth century. 63 A similar design on a lamp 58 Miltner type IX. Such lamps are o ften found in Samos; for their typology see N. Poulou Papadimitriou, "Lampes palochrtiennes de Samos," Bulletin de correspondance hellnique 110, no. 1 (1986), fig. 53. Amorium: Gill and "Roman and Early Byzantine," fig. III/8 ns. 39 41 and figs. III/25 and III/26. For various hypotheses regarding the origin of this type, see P. Dyczek, "Lamps of the 3rd 6th Century AD from the Civil Architecture in Sector IV at Novae," in Lychnologic al Acts 2 76. 59 Sucidava: Barnea et al, 171, n. 348. Tomis: Iconomu fig. 183. Halmyris: Topoleanu, "Ceramica," pl. XLVII/11. Sarahane: Hayes type 11, pl. 22/82. 60 Sucidava: Barnea et al., 172, n. 350; Histria: Popescu et al., arheologic Histria ," fig. 15/3; Sadovec: S. Uenze, ed., Die sptantiken Befestigungen pl. 46/22. Shumen: trakiisko to, rimskoto i rannovizantiiskoto selishte)," Izvestiia na Narodniia Muzei Shumen 6 (1973): fig. 1; Novae: Dyczek, "Lamps of the 3rd 6th Century," pl. 29/11. 61 V/5a c; for the analogy with Tomis see pl. V/6 62 Miltner type X. Also on lamps from a large hoard found at Anemurium, for which see Williams and Taylor, "A Byzantine Lamp Hoard," 81, form II, type 15. 63 Scorpan, "Descoperiri arheologice diverse," pl. XV/10. Halmyris: Topoleanu, "Ceramica," pl. XLIII/19. A similar lamp was found

PAGE 108

108 sixth century foundation, and on lamps from the Yassi Ada shipwreck proves that the type remained popular for a long time. 64 The second major category of lamps described here were either imported from Asia Minor or borrowed motifs from the most popular models produced in the Byzantine heartland and circulated in the Balkans duri ng the earl y Byzantine period (Figure 4 5). Only one such lamp has been found north of the Danube, at Sucidava, on the left bank of the river, while in barbaricum proper no sixth century lamps from Asia Minor have been reported to date. The lamp found at Sucidava bel ongs to the most common type, with its main charac teristic being the small globules on the rim. 65 An attempt was made to divide this type into two chronological groups: first lamps with a round shape and no channel towards the nozzle, dated to the late fift h and early sixth century, and second lamps with a more elongated shape and a small channel from discus to nozzle, dated to the later sixth century (Figure 4 17/7). 66 Such a division finds a possible confirmation in the finds from the ea rly seventh century shipwreck from Yassi Ada where only the later type has been found. 67 As it is, the specimen from Sucidava belongs to Hayes type 2, dated to the later sixth century. As a matter of fact, most lamps with globules on the rim imported from Asia Minor or imitat ing this popular design found in the Danubian at Stobi, in Macedonia, see M. Glumac, "Glinene svetiljke iz kasnoantichke zbirke Narodnog Muzeja u Beogradu," Zbornik Narodnog Muzeja 17, no. 1 (2001 ): 218, n. 8. 64 XXII/5. Yassi Ada: Vitelli, "Lamps", 190, L1, and fig. 9/2. 65 D. Tudor, "Sucidava III. Quatrime (1942), cinquime (1943) et sixime (1945) campagnes de fouilles et de recherches arch Dacia 11 12 (1945 1947), fig. 20/4. 66 Hayes 82. 67 Vitelli, "Lamps", 193, L12, and fig. 9 3.

PAGE 109

109 Fig ure 4 5. Anatolian lamps and local imitations. 1. Sadovec; 2. Pernik; 3. Beroe (Piatra Karasura (Rupkite); 8. Sacidava; 9. Sucidava; 10. Halmyris; 11. Tomis Cari

PAGE 110

110 provinces belong to Hayes type 2. 68 Molds found at Halmyris in Scythia prove that the type was imitated and produced locally. 69 It is, therefore, highly probable that the specimen from Sucidava did not come from Asia Minor bu t from a closer production center in the northern Balkans. Another important category of lamps is constituted by original lamps imported from North Africa or imitations produced in Greece or in the Danubian region (Figure 4 6). 70 A qu ick look at the map re veal s the fact that North African models found in barbaricum were almost as popular as the common "Danubian types found in great numbers in the Byzantine provinces of the northern Balkans. North African lamps type Hayes 2B/ Atlante X, most typical for the sixth century, have b een found at Turda (Potaissa) (Figure 4 17/8), Alba 68 Sadovec: Uenze, ed., Die sptantiken Befestigungen pl. 44/8. Pernik: V. Liubenova, "Selishteto ot rimskata i rannovizantiiskata epokha," in Pernik I. Poselishten zhivot na khulma Krakra ot V khil. pr. n.e. do VI v. na n.e. ed. T. Ivanov (Sofia: BAN, 1981), fig. 68/1 3. Beroe (associated with sixth century coins): Vlceanu and Barnea, "Ceramica lucr Podunavski deo oblasti Akvisa Po ntica 15 (1982): pl. VII/1. Karasura: V. Stoikov, "Kusnoantichni lampi ot Karasura (Predvaritelno suobshtenie)," in Paleobalkanistika i starobulgaristika. Vtori esenni mezhdunarodni cheteniia Profesor "Ivan Gulubov", Veliko Tunovo, 14 17 noemvri 1996 ed. I. Kharalambiev et al. (Veliko Turnovo: Universitetsko izdatelstvo "Sv. sv. Kiril i Metodii", 2001), pl. 1/4 5. Sacidava: Scorpan, "Descoperiri arheologice diverse," pl. IV/17. Capidava: pl. XLIII/427 portuare antice de la Capidava (II), C 5 7 (1988 1989), fig. 9/7. Iatrus: M. Wendel et alii, Iatrus Krivina. Sptantike Befestigung und frhmittelalterliche Siedlung an der unteren Don au. Band III: Die mittelalterlichen Siedlungen (Berlin: Akademie Verlag, 1986), pl. 59/494. 69 Topoleanu, Ceramica 211 4. have been described as local imitations, for which see Bjelaj ac, "La cramique et les lampes," 188. 70 Many such lamps are Attic products, for which see Perlzweig pl. 38 sq., or Corinthian, for which see Broneer types XXVIII XXXIII. Several such imitations were found in a villa urbana from Nea Anchialos in a cont ext dated with coins from Justinian I, see P. Lazaridis, "Nea Ankhialos," Archaiologikon Deltion 20 Stobi, in Macedonia, see Glumac, "Glinene svetiljke," 216, n. 6. A number of locally produced African imitations have been found at Novae, on the Lower Danube, for which see recently Dyczek, "Lamps of the 3rd 6th Century," 74. A mold for African lamps was found at Dinogetia, for which see I. Barnea, Les monuments palochrtiennes de Roumanie (Vatican: Pontificio Istituto di Archeologia Christiana, 1977), 241.

PAGE 111

111 centers of the Roman province of Dacia (Figure 4 20/5 6). 71 North African lamps are more often found in the Western Balkans, 72 alt hough a number of finds have been recorded at Tomis, Capidava, Dinogetia, Halmyris, Adamclisi, and Novae, some o f them being local imitations (Figure 4 17/9). 73 However, no direct parallels can be found between the type of lamps found in barbaricum and the lamps from Scythia, which might i ndicate that the Western Black S ea region was not the main supplier of such items, but probably the Aegean or the Dalmatian coast. 74 Nonetheless, in the central and western Balkans we find closer typological parallels, such as the African lamp type Atlante analogue at Koper (Capris), on the Adriatic coast. Moreover, it is closer to the lamps with Chi Rho ( ) insignia on the discus, the main type found nor th of the Danube. 75 A lamp type Atlante VIII A1a found at Tomis has a good analogy at Siscia (Sisak), which testifies once again to the importance of sea r outes in the diffusion of North African 71 d and 3a Acta Musei Porolissensis 10 (1986), fig. 3b. 72 For finds of African lamps in the Adr iatic region, see Prttel, Mediterrane Feinkeramikimporte 69 81; B. For the distribution of African lamps type Atlante VIII and X in the Danube region, se e recently V. diffusione delle lucerne nordafricane dei sec. IV VI D. C. nelle ex provincie central e sud danubiane. La lucerna del tipo Atlante VIII X con chrismon," in Lychnological Acts 2 197 206, and pl. 135. 73 Tomis import la Tomis," Pontica 9 (1976): 201 VIII/ 1 3, 5. Adamclisi: Barnea et al., Tropaeum pl. 173/ 9.2. A. Diaconescu argued that the lamp from Adamclisi has monetary impressions of Theodosius II and must be dated after 430 see Diaconescu, Halmyris: Topoleanu, "Ceramica," pl. LVII/13. Capidava: pl. XLIII/433. Dinogetia: Barnea, Christian Art in Romania pl. 107. Novae: Dyczek, "Lamps of the 3rd 6 th Century," pl. 28/4. 74 African lamps were being imitated on a large scale in Greece throughout the sixth century. Many of the types commonly found in the Danube region have analogies at Delphi, where a late sixth century cemetery has produced a large qua ntity of well preserved imitations of African lamps, for which see P. Petridis, La cramique protobyzantine de Delphes: une production et son contexte (Athens: Ecole franaise d'Athnes, 2010), 85 92. 75 Od Rimljanov do Slovanov Predmeti (Ljubljana: Narodni Muzej Slovenije, 2001), fig. 92.

PAGE 112

112 models. 76 Original models imported in towns close to the sea co ast were imitated in the Danube region or in the heart of the Balkans, where genuine imports were more difficult to procure. A perfect example is offered by two similar lamps, type Atlante X C2 with palm on the discus, one being a genuine import found at D ionysopolis on the Black Sea coast and the other a local imitation from Kozloduy, in Dacia Ripensis. 77 Well known for their Christian symbolism, the African lamps brought to the former province of Dacia might be related to the long distance circulation of C hristian artifacts. A route from Italy or through the Adriatic is very probable, given the high density of finds in the western Balkans, as well as on the Middle Danube. 78 The lamps could have traveled by sea with larger cargoes from North Africa, such as s patheia often found in the Adriatic region, and then redistributed on land routes and rivers to regions beyond the frontier. The lamps found in Scythia, on the other hand, were more easily brought via the Black Sea and the Aegean together with Greek imitat ions, such as the specimen found at Tomis. 79 Lamps imitating North African models are also present north of the 76 Tomis: Barnea, Christian Art in Romania 42, n. 304 and pl. XVII/7. Both lamps seem to be local imitations. The type originates most probably in Tunisia, for which see A. Ennabli, Lampes chrtiennes de Tunisie (Muses du Bardo et de Carthage) (Paris: Centre Nationa l de la Recherche Scientifique, 1976), pl. XLVII/865 and pl. LI/923. Imports of African Chi Rho lamps were also reported in Crimea together with imitations, probably following the same route on the Aegean and the Black Sea, for which see Chrzanovski and Zh uravlev, Lamps from Chersonesus 158, n. 96. 77 For both lamps, see Kuzmanov 42, no. 302 and 303. 78 For possible routes of distribution see 140. For the role of the Adriatic in the diffusion of western imports, see V. Vidrih Perko, "Seaborne Trade Routes in the North East Adriatic and their Connections to the Hinterland in the Late Antiquity," in L ed. 77. 79 Barnea et al., 169, n. 331. The image on the discus has been described as a rabbit but in fact i s a stag, a common depiction on African lamps type Atlante X C2. For a close parallel see Ennabli, Lampes chrtiennes de Tunisie pl. XX/391. Imitations of this type have been found in Chersonesus, for which see Chrzanovski and Zhuravlev, Lamps from Cherso nesus 162 64, ns. 102 103.

PAGE 113

113 Figure 4 6. North African lamps and imitations.1. Alba Iulia (Apulum); 2. Turda (Potaissa); 3. Drobeta (Turnu Severin); 11. Sarmizegetusa; 12. Sirmium; 13. Osijek; 14. ny; 18. Obuda; 19. Novae (Svishtov); 20. Nicopolis ad Istrum; 21. Dionysopolis (Balchik); 22. Kozloduy; 23. Deultum; 24. Siscia (Sisak); 25. Vinkovci; 26. Poetovio (Ptuj);

PAGE 114

114 Danube, at Drobeta, and further to the north at Sarmizegetusa. 80 A route from the Western Black Sea was not the only way in which such objects could find their way north of the Danube, since they could have easily arrived directly from Greece or from one of the Danubian provinces. Palestinian lamps are among the most exotic finds in the Lower Danube region, but also very desirable ones because of their connection to the Holy Land ( Figure 4 7). Few such lamps are reported among finds from the Balkans, but it may well be that some are either misattributed or dated to an early period. Alexandru Diaconescu has re dated the specimens found in barbaricum based on the typology established for lamps found in Syria called "candlestick type p roduced in the Holy Land from the sixth century onward (Figure 4 17/10). 81 castrum has a more problematic attribution (F ig ure 4 20/7 8). 82 Although it has been assigned to the early "Jerash type, the typol ogical parallels from Palestine are less convincing. 83 80 Drobeta: Barnea et al., Cultura 172, n. 353, very similar to a lamp from Athens, for which see Perlzweig n. 2591. b. A third Greek lamp was found at Apulum, being a typical product of Athens and Corinth, see Ibid., pl. VI/4a c. Greek imitations after African lamps are also common in Bulgaria, for which see Kostova and Dobreva, "Roman, Late Roman, and Early Byzantine," pl. 121, and in Serbia, see J. C. Rubright, "Lamps from Sirmium in the Museum o f Sremska Mitrovica," in Sirmium III. Archaeological Investigations in Syrmian Pannonia 1973), pl. V/56. 81 VII/1a c and VII/6. 82 Ibid. pl. VIII/1a c. 83 T. Scholl, "The Chronology of Jerash Lamps: Umayyad Period ," Archeologia 42 (1991): 65 84. dubious since the type boasts handles ending with zoomorphic heads, impossible to verify in the case of ose handle is now missing. The pattern on the rim is also different. Finally, the dating of this type to the second half of the seventh century is yet another reason to look elsewhere for parallels, since it is highly unlikely that Umayyad lamps were expor ted north of the Danube in the late seventh Kuzmanov type XXXII, who recognizes the

PAGE 115

115 A certain degree of skepticism is also warranted by closer analogies from the Balkans, with similar "barge shaped lamps having been found at Karasura, Haskovo, Iambol, Deultum, and Beroe (Stara Zagora) in Bulgaria, but also in Dobrudja and southern Russia. 84 Their dating to the sixth century is not certain as they may well belong to an earlier period. 85 Indeed, some of them might constitute local imitations of Syro Palestinian late antique models in the shape of a boat with a tall and curved handle, although Anatolian influences may also be present in the decoration. Handmade lamps have been largely neglected although they are an excellent indicator of the degradation of Roman life in urban centers of th e early Byzantine provinces and an example of imitatio in the case of lamps produced in barbaricum ( Figure 8). Indeed, where a modified Roman brick was transformed into a makeshift lamp, and at Sacidava on the Danube where a handmade cup was used for this purpose. 86 A third lamp, st ill in the form of a "Getic" cup, was found in a cremation grave from Nalbant in the northern part of Scythia. 87 The decline of urban production toward the end of the sixth century is influence of fourth century Syro Palaestinian models. The concentration of this type of lamps in central and south e astern Bulgaria suggests a possible center of product ion in that region (see below). 84 Karasura: Stoikov, "Kusnoantichni lampi ot Karasura," pl. I/1 2; Kuzmanov n. 291. Iambol: N. Tantcheva Vassilieva, "Antichni glineni lampi ot Muzeia v Iambol," Izvestii a na Muzeite ot Iugoiztochna Bulgariia 5 (1982), pl. III/18. Deultum: Kostova and Dobreva, "Roman, Late Roman, and Early Byzantine," 166, with the bibliography and further analogies in Dobrudja and southern Russia, and pl. 119 120; Stara Zagora: K. Kalchev "Antichni i kusnoantichni lampi ot Stara Zagora," Izvestiia na Muzeite ot Iugoiztochna Bulgariia 5 (1982): 7 22. 85 For a recent typology, see Kostova and Dobreva, "Roman, Late Roman, and Early Byzantine," 164 66, with a dating spanning the entire Late An tique period. 86 Sacidava: C. Scorpan, romane," Pontice 2 (1969), fig. 15 17. 87 G. Simion, "Necropola feudal timpurie de la Nalbant (jud. Tulcea)," Peuce 2 (1971), pl. V/3

PAGE 116

116 probably responsible for the existence of such improvised lamps in the fr ontier provinces of the Balkans. As for the handmade lamps produced in Barbaricum most of them are found in Wallachia east of the river Olt. An important contrast becomes apparent: imported lamps manufactured in various centers of the Empire concentrate i n the former province of Dacia, while handmade lamps are most often found in Wallachia. All handmade lamps have been found in sunken feature d buildings in sixth to seventh century contexts, none of which suggesting a particularly high social status. 88 The l amps from Br atei, Bucharest Soldat Ghivan (Figure 4 17/12), and Cnde shape and constitute attempts to imitate early Byzantine lamps. Interestingly, good analogies for these lamps can be found at Dunajvros (Intercisa) and Klked Feketekapu ( Figure 4 17/11), in Avaria, while no such lamps can be found in between, from Bratei to the Middle Danube, the demand for lamps being satisfied through imports from the Empire. 89 Given the rather unprepossessing nature of the buildings in which handmade lamps have been found, their production can be explained as a need to find a convenient alternative to early Byzantine lamps, unavailable (or unaffordable) in barbaricum A few conclusions emerge from the survey of lamps found in barbaricum The geographical distribution of the Balkan, or so called Danubian, lamps indicat es that in all cases the finds are coming from the territory of the former province of Dacia. Not a 88 M. Constantinescu, i (building H7): Zaharia, "La station n 2 de Bratei," fig. 13/12. Malu trg, 7 (1984), pl. II/2. Bucharest Soldat Ghiva n (building B5): Dolinescu Ferche and Constantiniu, "Un tablissement du VI e sicle," fig. 16/7a b. 89 31 (1907), fig. 75c. Klked Feketekapu (buildings 120 and 135): Ha jnal, "K

PAGE 117

117 Figure. 4 7. Palestinian lamps and imitations. 3. Gherla; 6. Haskovo; 7. Deultum; 8. Iambol; 9. Beroe (Stara Zagora).

PAGE 118

118 Fig ure 4 8. Handmade lamps. 2. Dunajvros (Intercisa); 3. Klk ed Bratei; 9. Bucharest Soldat Ghivan.

PAGE 119

119 single imported lamp has been b l e oil) were treated there as luxury items, rather than basic commodities, as was the case in frontier fortresses south of the Danu be. In fact, with the exception of the Palestinian concentrate in the former province of Dacia and most often in former urban centers, such as Sarmizegetusa, Potaissa, and Apulum. It has also been suggested that lamps should be associated mainly with Christian liturgy and the significance of light in Christian symbolism. 90 Indeed, many lamps bear Christian symbol s, especially the ones of North Afr ican inspiration and some of those produced in the Balkans. Nevertheless, a number of cheaper alternatives were available for liturgical purposes, so the lamp, a typical Byzantine import, must have possessed an important social value beyond its practical use or religious symbolism. T he fact that few or no lamps are found in the regions east of the former province of Dacia does not necessarily disprove the existence of a Christian population in contact with the Empire. It may be that communities from this region used other types of art ifacts to express their religious identity and to emulate the Roman way of life, as we shall soon find out. Concerning the origin of the Byzantine lamps exported to Barbaricum one cannot be more precise than making a general attribution to one of the prov inces adjacent to the Lower Danube. Indeed, some authors have suggested closer 90

PAGE 120

120 connections to the province of Scythia, 91 but such artifacts were probably more easily distributed through bridge heads on the northern bank of the Danube, especially Sucidava, w here a good variety of "Danubian lamps have been found. 92 Many African and Anatolian models were imitated locally and sometimes molds have been found as direct evidence of local production centers, but most of the time it is the local clay that betrays the practice of imitation along with a stylistic simplification or alteration of the original models. Nevertheless, N orth African and Palestinian lamps found in barbaricum as well as the Coptic liturgical lamps made of bronze, stand a greater chance of bein g genuine long distance imports, which testify to the existence of trading routes to the Western and Eastern Mediterranean, in which the former province of Dacia was still included. The Western connection was probably interrupted more rapidly, since after ca 550 African Fine Ware becomes rare even on parts of the Dalmatian coast, a traditional recipient of African imports during Late Antiquity. 93 This must have surely affected the supply of goods from the Western Mediterranean, which had been arriving north of the Danube probably through the mediation of intermediary trading ports in the western Balkans. 94 91 A. Madgearu, (Bucharest: All, 2001), 78. Contra hat Scythia might be too far to exert a significant influence over the former province of Dacia 92 Sucidava had been an important transit point for goods going north of the Danube since the fourth century, Poulter, Nicopolis ad Istrum 42 43. 93 J. P. Sodin i, "The Transformation of Cities in Late Antiquity within the Provinces of Macedonia and Epirus," in The Transition to Late Antiquity 330. A possible decline in production in North Africa is testified by the increased presence of eastern imports, such as LR1, for which see J. A. Riley, "The Coarse Pottery, in Excavations at Sidi Khrebish, Benghazi (Berenice) ed. J. A. Lloyd, v. 2 (Tripoli: Department of Antiquities, 1979), 121, fig. 2. In central and southern Greece there seems to be a persistence of A frican imports as late as the seventh century, for which see recently Petridis, La cramique protobyzantine 126 2 7. 94 Gudea and Ghiurco, 128. Cf. D. Protase, Autohtonii n Dacia II: Dacia

PAGE 121

121 4.2.3 Molds and Metallurgical Tools The nature of metallurgical activity north of the Danube has long captured the attention of scholars trying to assess the cultural influence of the Empire in barbaricum. 95 Molds in particular have been considered one of the most fascinating, yet unwieldy types of imports. 96 Indeed, the very notion of import attached to molds is under serious doubt as many of them could ha ve been produced outside the Empire, together with other metallurgical implements which are often found in association in archaeological contexts (Figure 4 9). It is imitatio rather than genuine import in many cases, with some of the items produced with th e molds being of Byzantine inspiration, mostly small pi eces of jewelry and ornaments (Figure 4 18/3). The technique itself, pseudo granulation and lost wax casting in particular, might have been borrowed from the Empire. Moreover, as we shall soon find out many of the items were in fact produced for further "exporting" somewhere else in barbaricum The task of the (Cluj Napoca: Risoprint, 2000), 80, arguing that the territory north of the Danube was isolated from the western world due to the presence of Gepids a n d later Avars. 95 nistrean (secolele V XIII) (Bucharest: Editura Academiei Romne); D. G. Teodor, "Atelie rs byzantins des VI e VIII e sicles au nord du Bas Danube," Etudes Byzantines et Post Byzant ines 5 (2006): 198 201; Societatea carpato danubiano XI. Structuri demo politice (Bucharest: 1997); D. G. Teodor, X I 9 (1972): 73 99. 96 VII carpatic ," 1 (2009): 149 57; B. 8th Centuries Metallurgical Activity from Budureasca Valley. The Molds," AMN 41 42, no. 1 (2004 2005): 291 318; D. G. Teodor, "Tipare din secolele V XI d. Hr. n regiunile carpato nistrene," AM 28 (2005): 159 74; B. S. Szmoniewski, "Production of Early Medieval Ornaments Made of Non Ferrous Metals: Dies from Archaeological Finds in North East Romania," AAC 37 (2002): 111 35; E. Iu. Tavlinceva, "K voprosu o meta llicheskom bisere v riazno okskikh mogil'nikakh (po materialam Shoshinskogo mogil'nika)," in Nauchnoe nasledie A. P. Smirnova i sovremennye problemy arkheologii Volgo Kam'ia. Materialy nauchnoi konferencii ed. I. V. Belocerkovskaia (Moscow: Gosudarstvenny i istoricheskii muzei, 2000), 109 Krenz, "The Beginning of Metal Casting in Polish Territories in the Early Middle Ages (the 6th 13th centuries)," in Trudy V Mezhdunarodnogo Kongressa arkheologov slavistov, Kiev 18 25 sentiabria 1985 g. ed. P. P. Tolochko (Kiev: Naukova Dumka, 1988), 83 VI, descoperite pe teritoriul Romniei," 101 (1983): 557 61.

PAGE 122

122 researcher trying to make the most of the molds under study is to go beyond the mere description of the objects as seen on the molds themselves and actually attempt to trace them in the archaeological record. More often than not, objects resembling the ones carved on a mold are completely unknown in the region where the mold was found, which has important implications for our understanding both of the production and the circulation of such items. The question of craftsmen is also somewhat problematic, as many scholars would have us believe that Byzantine itinerant craftsmen had a crucial role in the metallurgical production in barbaricum Much has been made of the account of Procopius who suggested that many merchants and craftsmen were fleeing the country because of the many abuses which reduced them to poverty. Leaving aside the fact that the Secret History a vitriolic Kaiserkritik of epic propo rtions, should rarely be taken at face value, Procopius does not even refer specifically to the Danube region, but to the East: And in the other cities practically the whole population found itself suddenly reduced to beggary. For the mechanics and the han d workers were naturally compelled to struggle with hunger, and many in consequence changed their citizenship and went off as fugitives to the land of Persia 97 The account was conveniently truncated in an early seminal study and later adopted uncriticall y to suggest that such craftsmen may be the ones responsible for the presence of molds north of the Danube. 98 Such itinerant craftsmen may very well have operated north of the Danube, attracted by a larger demand for jewelry of Byzantine inspiration or co mmissioned to produce certain types of items. The ethnicity 97 Procopius, Historia Arcana 25.25, ed. Haury and Wirth, vol. 3, 1 5 7; trans. Dewing, vol. 6, 301. 98 ," 152.

PAGE 123

123 of such craftsmen is, however, impossible to know in the absence of any solid evidence regarding their activity. There is reason to believe that local production was perhaps more important than th e presence of itinerant specialists. Although the buildings where metallurgical implements have been uncovered look no different than the ordinary sunken featured residences so common in barbaricum the association of molds with crucibles, ladles, scrap me tal, and specific tools such as tweezers and car vers point to the existence of local metallurgical activity. To be sure, the re is less evidence of centralized production of jewelry in specialized workshops. Rather, it seems to be a household activity or at least not restricted to a single workshop in the community, judging by the fact that in settlements like the ones at Budureasca, and Zimne, metallurgical implements and molds in particular have been found in several residences. 99 The notion that the presence of molds is essentially associated with the frontier region and a stronger influence of Byzantium is weakened or even altogether disproved by the presence of such artifacts on a huge geographical area from the Volga region to Central Europe. To be sure, there is a larger concentration of molds in the territories adjacent to the frontier, w hile molds used to produce certain objects, such as crosses, are usually found in this area. By contrast a considerably smaller number of molds have been found in the Byzantine provinces of the Balkans, which begs the question of 99 98, fig. 19/6; 99 fig. 20/3 and fig. 20/1; 100 fig 21/1 a c. secolele V "Issledovaniia rannesl avianskikh poselenii v Moldavii," in Arkheologicheskie issledovaniia v Moldavii v 1970 1971 gg. ed. G. F. Chebotarenko (Kishinew: Shtiinca, 1973), 140 fig. 3/1 and 3/2; I. A. Rafalovich and V. L. Lapushnian, "Raboty Reutskoi arkheologicheskoi ekspedicii," in Arkheologicheskie issledovaniia v Moldavii v 1972 g. "The 6th 8th Centuries." Zimne: V. V. Aulikh, Zimnyvs'ke gorodishche slov'ians'ka pam'iatka VI VII n.e. v zakhydnyi Voliny (Kiev: Naukova dumka, 1972), pls. XIV/1 3, 5 and fi g. XV/1 7.

PAGE 124

124 whether metallurgical pr oduction using molds was an essentially "barbarian" technology. The answer must be negative, since many types of Byzantine jewelry and ornaments were being produced using the same technique, while production itself is well attested in towns like Sadovec, Cari 100 It may well be that molds, as part of organized imperial workshops, were carefully discarded or removed and are thus harder to trace in the archaeological record. The few molds found in the Balkans offer precious clues to the con nection between metallurgical activity on the frontier and what was transmitted beyond, in barbaricum There is in fact very little to connect the "fashions" in the two regions, as only a few of the items appearing on molds from the Empire can be found on molds from barbaricum The triangle shaped pendant from the mold found in the late sixth century occupation level at Adamclisi in Scythia has good parallels at Budureasca 3, Bernashivka, and as far north as Loosi in Estonia ( Figure 4 20/21 22), 101 while the belt some degree artifacts found in burial assemblages from Kerch and Lucistoe, although the fashion itself can also be traced in the Avar khaganate. 102 The pyramidal earring pe ndants from a mold found near Oescus have good analogies in Early Avar 100 Sadovec: Uenze, Die sptantiken Befestigungen in 220 24. Adamclisi: Barnea et al., Tropaeum Traiani 218, fig. 169/10/14. 101 Budureasca: 8 th Centuries," 315, fig. 7 Bernashivka: I. S. Vinokur, Slov'ians'kyi iuveliry Podnistrov'ia. Za materialamy doslidzhen' Bernashivs'kogo kompleksu seredyny I tys. n. e. (Kam'ianets' Podil's'kyy: Oium 1997), fig. 23. Loosi: M. Schmiedehelm and S. Laul, "Asu stusest ja etnilistest oludest Kagu Eestis I aastatuhande," in Studia archaeologica in memoriam Harri Moora ed. M. Schmiedehelm et al. (Tallinn: Kirjastus "Valgus", 1970), 160 and fig. 4; pl. I V. 102 Bavant, "Les petits objects," 220 24; B. Bavant, "Un mo ule d'orfvre protobyzantin au British Museum," in Mlanges Jean Pierre Sodini Travaux et Mmories 15 (Paris: CNRS 2005), 643, fig. 6.

PAGE 125

125 assemblages (earrings type Deszk) but also in the steppe north of the Black Sea. 103 Similar connections with Avaria can be traced on the mold found accidentally at Gorna Sekirna, which bo asts a model for earrings imitating the shape of twisted wires, with analogies at Klked Feketekapu. 104 The granulated column pendants found on the mold from Argamum represent a widespread type of ornamentation found on molds across barbaricum (Figure 4 20/1 5 16). Pseudo granulation, a typically Byzantine technique, can be found on very similar molds from Budureasca 4 ( Figure 4 18/1 2), Cristuru, Bucharest Str Russia, dated from the sixth to the eleventh century. 105 Finally, the mold for Maltese crosses from Sadovec represents a special case as both molds and actual crosses are found north of the Danube. Molds designed to produce Byzantine crosses have been found at Olteni, Celei, Izvoru Dul ce, Davideni, Boto 106 With two exceptions, all molds from this 103 M. Daskalov and D. I. Dimitrov, "Za prozvodostvoto na nakiti prez VI VII v. v bulgarskite zemi," Ar kheologiia 42 (2001): 69, fig. 1; Cs. Bllint, "Probleme der archologischen Forschung zur awarischen Landnahme," in M. Mller Wille and R. Scheider, ed., Ausgewhlte Probleme europischer Landnahmen des Frh und Hochmittelalters (Sigmaringen: Thorbecke, 1993), 237 38; Garam, Funde byzantinischer Herkunft pl. 10. 104 M. Daskalov, "Kalapi za metalni nakiti i kolanni ukrasi (VI VII v.) ot Iuzhna Balgariia," Arkheologiia 45 (2004): 91, fig. 1; A. Kiss, Das awarenzeitlich gepidische Grberfeld von Klked Fekete kapu A. (Innsbruck: Universittsverlag W agner 1996), 268, fig. 45/187. 105 8th Centuries," 318, fig. 10/9 10. Cristuru: Z. VIII n bazinul superior al Trnavei Mari," SCIVA 39 ( 1988), no. 2, 182 and 186; fig. 19/6. Bucharest Vinokur, Slov'ians'kyi iuveliry fig. 22, 27, 36, 37, 38, 40. Russia: Tavlinceva, "K voprosu o metallicheskom," 112, fig. 3. 106 Olten i: C. Preda, "Tipar pentru bijuterii din secolul al VI lea e.n., descoperit la Olteni (r. Videle, reg. SCIV 18, n. 3 (1967): 513 15; Celei: Izvoru Dulce: I. Miclea and R. Florescu, Daco romanii vol. 2 (B ucharest: Meridiane, 1980), 209; pl. 760. Davideni: D. G. Teodor, lea 99 100 fig. 20/1 and 21/1 a c. Bucharest Ialoveni, de Stat din Moldova s umane (1998): 29 31, fig. 1.

PAGE 126

126 category are found outside the former province of Dacia, which stands in sharp contrast with the distribution of Byzantine lamps with Christian motif s, very common on the territory of the former province and rather rare south and east of the Carpathians. However, it should be noted that the few molds found in a good archaeological context are not associated with any other Christian artifacts. The mold s, the finished products, as well as the typical metallurgical implements are not restricted to the frontier region and the near Barbaricum but seem to have a lot in common with similar artifacts found on a very wide geographical area. The mold from Bucha rest not only at Budureasca and Davideni, but also at Bernashivka in Ukraine and at Supruty and Kuzhendeevsky in Russia, where they have been found in graves applied on clothing and not on pieces of jewelry. 107 The dating of such items is very loose as they are sometimes found in later medieval contexts showing the longevity of such ornaments, which should not be surprising due to their basic shape The interesting file carved on a mold fro m Budureasca 5 has very good parallels both in Avaria at Vc Kavicsbnya and at Zimne in Ukraine, 108 while tweezers similar to the one from Bucharest Soldat Ghivan have been found at Bancerovsko and Usugorsk in Russia. 109 The rare occurence of molds made of bo 107 For Bucharest 106. Supruty and Kuzhendeevky: Tavlinceva, "K voprosu o metallicheskom," 112, fig. 3/1 3. 108 8th Centuries," 316, fig. 8. Vc Kavicsbnya: S. Tettamanti, Das awarenzeitlische Grberfeld in Vc Kavicsbnya (Budapest: Magyar Nemzeti Mzeum, 2000). Zimne: Aulikh, Zimnyvs'ke gorodishche 74 and pl. XV/1. 109 Bucharest Soldat Ghivan: Dolinescu Ferche a nd Constantiniu, "Un tablissement du VI e sicle," 321 and fig. 18/13. Bancerovsko: A. G. Mitrofanov, "Bancerovskoe gorodishche," Belorusskie drevnosti (1967): 255. Usugorsk: A. M. Murygyn, "Poselenie Usugorsk III na srednei Mezeni," Materialy po arkheolog ii Evropeiskogo Severo Vostoka 8 (1980): 81.

PAGE 127

127 at Zimno and Pastyrske in Ukraine, while crucibles and ladles, some of which display extremely similar shapes, can be found in most early medieval metallurgical complexes in Central and Eastern Europe, from the Czech Republic, to the Balt ic and the Upper Volga region. Perhaps the most interesting observation resulting from the geographical distribution of metallurgical implements (Figure 4 9) is the fact that many of the more elaborate items produced with th e mol ds have clear analogies in the Avar world. items such as punched appliqus, pendants, belt ornaments, and earring components identifiable in the archaeological repertoire of t he Early and Middle Avar milieu. 110 In recently been re dated based on analogies with artifacts belonging to the Middle Avar period. 111 What is then the e xplanation for the presence of molds used to produce objects of Avar inspiration but Byzantine technique? Molds from barbaricum as well as those from the frontier region of the Empire have a strong connection with the presenceof troops and a militarized so ciety. However, the fact that molds found in the extra Carpathic region point to a strong connection with the Avar khaganate is a good indicator of a diminishing influence of the Empire in the territories north of the river, especially in the seventh centu ry when the Danube ceased to be the political frontier of the Empire in the Balkans. 110 8th Centuries," 297 99. 111 Szmoniewski, "Production of Early Medieval," 122 27.

PAGE 128

128 Figure 4 9. Metallurgical implements 1. Tropaeum Traiani ( Adamclisi); 2. Aldeni; 3. Bucha rest Casa Armatei; 11. Bucharest Soldat Ghivan; 12. Bucharest Tei; 14. Bucharest eava; 40. Bani; 46. Bernashivka; 47. Gorecha; 48. Kodin; 49. Mikhailyvka; 50. Novoselitsia; 51. Parievka; 52. Pastyrs'ke; 53. Rashkiv; 54. Samchincy; 55. Semenki; 56. Skibincy; 57. Teterev ka; 58. Volos'ke; 59. Gorna Sekirna; 60. Oescus (Ghigen); 61. Sadovec; 62.

PAGE 129

129 4.2.4 Brooches and Buckles 4.2.4.1 Fibulae with bent stem Among the various imports of Byzantine origin brought to the lands north of the Danube during the long sixth century, fibulae are by far the most diverse group from a typological perspective. The fibula with bent stem and its derivation, the cast fibula with bent stem, have a long reputation of resisting any attempt at creating a coherent and universally accepte d classification based on style, size, and chronology. The earlier type seems to be the fibula with bent stem which the German archaeologist Syna Uenze has rightly associated with the fourth c entury fibula with bent stem typical for the Sntana de Chernyakhov culture known from sites excavated north of the Danube. 112 Although there are certain missing links between the two types, namely specimens from the fifth century, the fibula with bent stem typical for the Justinianic and post Justinianic period in the Balkans has a number of constitutive similarities with the fourth century version that cannot be ignored. To be sure, there are at least two major differences between the fourth century f ibu la with bent stem and its sixth century counterpart. Unlike the early forms, the sixth century type is larger and has a U shaped stem, which along with other stylistic details, notwithstanding a chronologically sensitive archaeological context, where av ailable, safely dates it to the sixth century. 113 The second major difference lies in the distribution of the two types, as the sixth century fibula with bent stem is no longer a creation of the lands north of the Danube frontier, but a typical product of t he frontier fortresses acquired in barbaricum as a Roman 112 Uenze, Die Sptantiken Befestigungen 146 54. 113 I. O. Gavritukhin, "Fibula iz Luki Kavechinskoi v kontekste slaviano vizantiiskih sviazei," in ed. M. Dulinicz (Lublin/Warszawa: Uniwersytetu Marii Curie 201.

PAGE 130

130 import. Consequently, the overwhelming difference in the number of finds in the Balkans compared to the regions outside the Empire's borders comes as no surprise (Figure 4 10). The chronology of the fibulae with bent stem raises several problems for the archaeologist trying to narrow it down to a more precise interval (i.e. first or second half of the sixth century). In the past two decades the number of finds has incre a sed five times but this spect acular growth of the corpus is due less to archaeological excavations and more to the intensive use of metal detectors, which often leaves museum curators with the frustrating task of making sense of important artifacts with no archaeological context to gi ve them historical meaning. Even so, there are sufficient datable finds to allow for a tentative chronological attribution for the main categories of fibulae with bent stem. 114 Syna Uenze has advanced a date in the first haf of the sixth century for the fibu la with bent stem, which was replaced in the second half of the century by its cast version. 115 However, evidence from various regions of the Balkans, the Crimea, and also from barbaricum clearly suggests a continuous use of fibulae with bent stem throughout the sixth century and perhaps even more intensely in the second half of that century. Unfortunately, there are no coin dated fibula finds, although in at least two cases, Nea Anchialos in Macedonia and Histria in Scythia, fibulae with bent stem have been found in proximity of coins struck for Justin II (565 578), but no stratigraphic al correlation can be made between the a rtifacts to secure the dating to the second half of the sixth century. 116 114 Century Fibulae with Bent Stem," Starinar forthcoming, fig. 29. 115 Uenze, Die Sptantiken Befestigungen 149 51. 116 Nea Anchialos: Lazaridis, "Nea Anchialos," 326

PAGE 131

131 In a third case, a fibula with bent stem was found near a house in the Sadovec Golemannovo kale fort, an area which also produced coin finds, the latest of which are issues of Justin II, but once again no secure correlation can be made between the fibula and the coins. 117 Nevertheless, in several cases the archaeologica l context indirectly suggests a later dating than the one proposed by Uenze. Such a case is the as semblage in grave 112 in Stari Kostolac which includes the remains of a wooden bucket with iron handle and hoops, a practice unknown before the settlement of the Avars in Pannonia ( ca. 570). 118 At Bucharest Militari a fibula with bent stem was found inside a sunken featured building in association with fragments of handmade pottery typical for the second half of the sixth or the early seventh century. 119 To be su re, there a re also clear examples of specimens produced in the first half of the sixth century such as the fibula found in grave 155 in Suuk Su (Crimea). The fibula was associated with tury. 120 In the Balkans, a fibula from grave 100 in Stari Kostolac was associated with a shield on Century," n. 48. Histria: Condurachi MCA 4 (1957): 16 21. For the coins, see C. Preda and H. Nubar, Histria III. Descoperirile monetare de la Histria 1914 1970 (Bucharest: Editura Academ iei, 1973), 209 16. 117 A hoard of gold coins was also found in that house, with the later coins struck for Emperor Maurice, see Uenze, Die Sptantiken Befestigungen 116, 118 119, 302, 332 333, 403, 477 478. 118 Les ncropoles de Viminacium l'poque des Grandes Migrations (Paris: Association des Amis du Centre d'Histoire et Civili sation de Byzance, 2006), 162. 119 Militari," CAB 1 (1963): 60 and 63; M. Sgbea, "Fibule din sec. arheologice de la Militari," CAB 1 (1963): 373 and 378 80; 379 pl. II/1; Teodorescu, "Centre 120 N. I. Repnikov, "Nekotorye mogil'niki oblasti Krymskikh gotov," Zapiski Odesskogo obshchestva istorii i drevnostei 27 (1907), 117 118. I. P. Zaseckaia, "Datirovka i proishojdenie palchatih fibul bosporskogo nekropolia rannesrednevekovogo perioda," MAIET 6 (1997): 450.

PAGE 132

132 Fig ure 4 10. Fibulae with bent stem. 1. Volos'ke; 2. Zvonets'ke; 3. Igren'; 4. Kizlevo; 5. 10. Poian; 11. Szolnok; 17. Keszthely.

PAGE 133

133 tongue buckle, which also suggests a dating within the first half of the sixth century. 121 Only ca. ten percent of the total number of finds are f rom outside the Empire. This does not seem to affect the distribution of fibulae with bent stem in barbaricum which covers a very large area from Lake Balaton to the Middle Dnieper. It does translate, however, in a much reduced variety of types compared to the Balkans where we encounter all variants and sub variants of fibulae with bent stem. Out of twelve classes identified in a recent classification, only four are present in barbaricum 122 Whether or not the preference for certain classes reflects fashions outside the Empire is a matter of speculation, but a comparative analysis of the typological parallels among finds from barbaricum and the Balkans, respectively, can shed some light on possible routes of distribution and areas of contact between the two wo rlds. The largest category of finds belongs to the group of fibulae with the bow wider than the stem, with zig zag ornament, which is, unsurprisingly, one of the most numerous classes of fibulae with bent stem from the Balkans. Most analogies are with fin ds from the provinces of Moesia II and Scythia, in the north eastern Balkans. The more distant finds from the Dnieper region have clear analogies in the two frontier provinces. The specimen from Zvonets'ke (Figure 4 19/1) has a close parallel at Zdravkovec in Mo esia II, the latter displaying surprisingly crude craftsmanship compared to the fibula from barbaricum. 123 The fibula from Igren', in the same Dnieper 121 Les ncropoles de Viminacium 160. For the dating of shield on tongu e buckles, see J. Cseh et al. Gepidische Grberfelder im Theissgebiet II (Budapest: Mag yar Nemzeti Mzeum, 2005), 154. 122 Century Fibulae." 123 A. V. Bodianskii, "Arkheologicheskie nakhodki v Dneprovskom nad porozh'e," Sovetskaia arkheologiia (1960), no. 1, 274 and 275; 273 fig. 1/7. K. Koicheva and A. Kharalambieva, "Fibuli ot Istoricheskiia muzei v Gabrovo (III VII vek)," Godishnik na muzeite ot Severna Bulgariia 19 (1993): 70; pl.VII/9.

PAGE 134

134 region, has a common decoration found on several speci mens from Scythia and Moesia II (Adamclisi, Gen eral Kantardzhievo, and Golesh), but also from Bucharest, north of the Danube. 124 No precise details are known regarding the archaeological context, except for the fibula foun d at Adamclisi in the sixth century settlement and the specimen from Bucharest foun d inside a clay oven in a sunken featured building. Yet another site find is the iron fibula from Davide ni, which has a good stylistic parallel at Pet Mogili in Moesia II, although the latter was made of copper alloy. 125 As we move closer to the frontier th e parallels of finds from barbaricum come to include provinces west of Moesia II, Dacia Ripensis in particular, but also Dacia Mediterranea. Bucharest seems to have be en at a crossroads in this respect, as a second fibula found in a neighboring settlement has good analogies at Stari Kostolac, Aquis, and Korbovo in the Iron Gates region, but also at Sucidava, the most important Byzantine bridge head on the left bank of the Danube. 126 A number of specimens of this class have been found at Pernik in western Bulg aria, which sug g ests that this particular variant was mainly produced in the diocese of Dacia. 127 Interestingly, the specimen found in a female grave at Szolnok on the Middle Danube, the westernmost find of 124 Igren': O. M. Pr ykhodniuk, Pen'kovskaia kul'tura. Kul'turno khronologicheskii aspekt issledovaniia. (Voronezh: Voronezhskii universitet, 1998), 140 fig. 71/2. Adamclisi: Barnea et al., Tropaeum Traiani 223 fig. 174/10.3. General Kantardzhievo: A. Kharalambieva, "Dva tipa kusnoantichni fibuli vuv Varnenskiia muzei," INMV 25 (1989), 34 and 37; pl. III/2. Golesh: G. Atanasov, "Martyrium et dans le castel bas byzantin prs du village de Golech, rgion de Silistra (communication prliminaire)," in Von der Scythia zur ed. Kh. Kholiolchev et al. (Vienna: Verlag "Freunde des Hauses Wittgenstein", 1997), 127 129; 138 fig. 5/4. Bucharest: Sgbea, "Fibule din sec. III," 373, 378 380; 379 pl. I I. 125 Mitrea, 138 39; 329 fig. 68/1. A. Kharal ambieva and G. Atanasov, "Novopostupili fibuli ot III VII v. v Shumenskiia muzei," INMV 28 (1992): 100 pl. III/5. 126 Buharest Les ncropoles de Viminacium 160 and 161 pl. 10/T100.2. Aquis: Korbovo: Podunavski deo oblasti Akvisa 215; 126 fig. 70. Sucidava: D. Tudor, "Sucidava III," 197; 196 fig. 41/14. 127 Liubenova, "Selishteto ot rimsk ata," 168 170; 171 fig. 110/7.

PAGE 135

1 35 fibulae with bow wider than the stem and zig zag or nament from Barbaricum has no parallels in the Iron Gates region, being somewhat similar to a specimen from Venchan in Moesia II. 128 The second largest group of finds from barbaricum is represented by fibulae with bows and stems of similar width. This is on e of the most interesting groups with at least two finds having no close analogies in the Empire the specimen from Hradyz'ke (Figure 4 19/2) on the middle Dnieper and the one from Keszthely in Hungary. The latter's bow is somewhat similar with the fibulae from Kramolin at the border between Moesia II and Dacia Ripensis and Suuk Su in Crimea, but the connection is not secure and in fact both finds may have been produced in barbaricum. 129 In addition, two fibulae found in Moldavia have loose analogies in the B alkans, the specimen from Davideni being a basic one with no decoration, very common in Moesia II, while the fibula from Moldoveni finds its closest parallel at Pernik in western Bulgaria. 130 Interestingly, the fibula from Davideni was found in a sunken feat ured building together with amphora fragments and a ladle, usually associated with metallurgical production, all suggesting that the owner had close ties to the Byzantine world. 128 I. Bna and M. Nagy, Gepidische Grberfelder am Theissgebiet I. (Budapest: Magyar Nemzeti Mzeum, 2002), 220; 376 pl. 102.8. Kharalambieva, "Dva tipa kusnoantichni, 33, 34, 37 and 39; pl. III/6. 129 Hradyz'ke: L. M. Rutkivs'ka, "Arkheologicheskie pamiatniki IV VI vv. v raione Kremenchugskogo moria (Ukraina)," Slovensk Archeolgia 27 (1979): no. 2, 358; 341 fig. 22/9. Keszthel y: R. Mller, "Sgi Kroly Zalai Mzeum 9 (1999): 158 and 173 fig. 4/23.1. Kramolin: Koicheva and Kharalambieva, "Fibuli ot istoricheskiia," 70; pl. VI/4. Suuk Su: Repnikov, "Nekotorye mogil'n iki," 116 17; 146 fig. 111. 130 Davideni: Mitrea, 121 122; 326 fig. 66/6. Moldoveni: I. Mitrea, "Cteva fibule romano bizantine descoperite n Moldova," SCIV 24 (1973), no. 4, 663 and 665; 664 fig. 1/1. Pernik: Liubenova, "Selis hteto ot rim skata," 168 70; 171 fig. 110/5.

PAGE 136

136 The third class of fibulae with bent stem found in barbaricum features a trape ze shaped stem as its most distinctive feature. Perhaps not accidentally such finds are most common in the Dnieper region, at Volos'ke and Hradyz'ke, both with parallels in north eastern Bulgaria. 131 A third specimen was f ound at Poian in Transylvania (Figur e 4 19/3), somewhat resembling the shape of a specimen from the L'viv region in Ukraine and another one from Gabrovo, south of the Danube. 132 The fibula from Poian is also important for establishing the chronology of this variant, being found in a house toge ther with hand made pottery and fragments of clay pans, which have been dated to the second half of the sixth century. 133 Finally, the last category of fibulae with bent stem found beyond the Empire's frontier displays a triangular bow section, which is als o one of the most conspicuous similarities with the fourth century fibula with bent stem. Only one such find has been recorded so far, at Kavetchina on the Dniester (Figure 4 19/4). 134 The iron fibula from Kavetchina stands out as the longest known fibula of its class, 135 but its basic shape, with no decoration, resembles several specimens from Bulgaria and Serbia, many of which, however, have been found on sites located west of Moesia II, the main supplier of fibulae with bent stem in barbaricum. 131 Volos'ke: Prykhodniuk, Pen'kovskaia kul'tura 156; 142 fig. 74/4. Hradyz'ke: Rutkivs'ka, "Arkheologicheskie pamiatniki," 358; 341 fig. 22/9. NE Bulgaria (unknown location): A. Kharalambieva, "Fibulite ot I VII v. v muzeia na Dulgopol," INMV 32 33 (1996 1997): 113; 128 pl. XII/107. General din secolele VI XI p. Ch. n bazinul Oltului superior," SCIVA 43 ( 1992): no. 2, 263; 269 fig. 17. 132 Gabr ovo: Koicheva and Kharalambieva, "Fibuli ot istoricheskiia," 69, fig. II/4. 133 Curta, The Making of the Slavs 296. 134 L. V. Vakulenko and O. M. Prykhodniuk, Slavianskie poseleniia I tys. n.e. u s. Sokol na Srednem Dnestre (Kiev: Naukova Dum ka, 1984), 82 and 57 fig. 32/9. 135 Century Fibulae," f orthcoming.

PAGE 137

137 Prolong ed p roduction and circulation of such fibulae beyond the fir st half of the sixth century have important implications for their presence in barbaricum at a time when the northern Balkans became a major theater of military operations. Despite the fact that many of the finds have no known archaeological context, their association mainly with hilltop sites is unquestioned Although the late fibulae with bent stem may have been fashionable in the Balkans as early as the first decades of the sixth century, the growin g militarization of the frontier provinces in the la ter sixth century and the connect ion between fibulae with bent stem and frontier fortresses and settlements offers the necessary explanation for their popularity, which lasted until the fall of the Danube frontier in the first decades of the seventh century. The few finds with a chronologically relevant context from the lands north of the Danube suggest that fibulae with bent stem continued to be brought to barbaricum in the second half of the century. The ir production may have ceased, or at least diminished, after the cast fibula with bent stem started to become more fashionable in the second half of the sixth century. However, there is sufficient reason to believe that both types were contemporary, if not equally p opular, during the troubled decades announcing the end of the Byzantine domination on the Danube. 4.2.4.2 Cast fibulae with bent stem The cast fibula with bent stem is often named "early Byzantine," "Roman Byzantine," Danubian Byzantine," or "ca st fibula with fake spiral" ( gegossene Fibel mit Scheinumwicklung ). Indeed, its name in the literature derives from the fact that it is most often found in the frontier region of the northern Balkans, being a typical product of the military forts on the Da nube and one of the most popular types of brooches in barbaricum as well ( Figure 4 11). Establishing a firm typology and identifying production

PAGE 138

138 centers have been the main objectives of archaeologists working with the cast fibulae with bent stem. Syna Uenze was the first to notice the heavy concentration of finds in the area of the Iron Gates of the Danube which pointed to an important production center in the region. 136 This hypothesis was confirmed by the publication of the casts found in the Theodora Tower of the Late Roman fort in Drobeta, which Adrian Bejan identified as an important production center on the northern bank of the Danube. 137 The existence of such production centers north of the river might explain the popularity gained by this type of fibula i n b arbaricum, although the prisoners taken from the provinces of the Balkans might have also contributed to its significant presence beyond the frontier. 138 However, a growing body of finds from the last decades has contradicted the earlier hypothesis that the cast fibulae with ben t stem originated in the border lands between Moesia II and Dacia Ripensis. Many finds from northern and northeastern Bulgaria and especially two specimens with unfiled edges from the region of Shumen led Anna Kharalambieva to the j ustified conclusion that such fibulae were in fact produced in several centers of the frontier provinces. 139 A number of finds from central 136 S. Uenze, "Gegossene Fibeln mit Scheinumwicklung des Bgels in den stlichen Balkanprovinzen," in Studien zur vor und frgeschichtlichen Archologie: Festschrift fr Joachim Werner zum 65. Geburt stag ed. G. Kossack and G. Ulbert (Munich, C H. Beck, 1974), 483 94. 137 A. Bejan, "Un atelier metalurgic de la Drobeta Turnu Severin," AMN 13 (1976): 257 68. 138 VII vekov i slaviane," in Rapports du III e Congres Inte rnational d'archologie slave, Bratislava 7 14 Septembre 1975 vol. 1 (Bratislava: VEDA, 1980), 171. 139 A. Kharalambieva, "Production of Dress Ornaments in the Fortresses and Small Settlements in North Bulgaria during the Period from the 5th till the 7th C entury AD," in The Roman and Late Roman City. The International Conference (Veliko Turnovo 26 30 July 2000) ed. L. Ruseva Slokoska, R. T. Ivanov, and V. Dinchev (Sofia: Akademichno izdatelstvo "Prof Marin Drinov", 2002), 393 97.

PAGE 139

139 Bulgaria, as well as a mold designed to produce cast fibulae with bent stem found at 140 Due to the lack of standardization in the production of this type of fibu la there is no consensus regarding the precise criteria for establishing a firm typology, and in fact none is completely satisfactory. 141 Adrian Bejan distinguished four typological variants ology on the section of the bow as well as the length of the fibula. 142 is based mainly on the specimens found at Aquis (Prahovo) where he distinguished between fibulae produced locally and "barbaric" imitations. Equa lly limited is Dan G. Teodor's typology based solely on specimens found in b arbaricum, thus ignoring the far larger number of finds from the Danubian provinces. 143 A more nuanced classification was attempted by Anna Kharalambieva who had access to the growin g number of finds from north eastern Bulgaria. Her five variants of cast fibulae with bent stem were based mainly on the shape of the bow: rectangular, trapeze, triangular, and semicircular and a peculiar type of stem ending in the shape of a cross. 144 Igor bow and the length of the fibula. By such means he distinguished three series named Drobeta, Varna, and Golemannovo, respectively, in order to acco unt for three areas of 140 bizantine," MCA 4 (2008) : 141. 141 Critical Approach to the Classification and Interpretation of Cast Fibulae with Bent Stem, AB 15, n. 3 (2011): 51 81. 142 nie fibuli VI VII vekov," 173. 143 bizantine din secolele V danubiano pontic," AM 12 (1988): 197 223. 144 Kharalambieva, Dva tipa kusnoantichni," INMV 25 (1989): 29 40.

PAGE 140

140 production and/or demand: the Iron Gates segment of the Danube frontier, northeastern Bulgaria, and the plain between the Danube and the Stara Planina Mountains. 145 ogies as based on inconsistent criteria and instead favored a classification entirely and solely based on decoration, thus distinguishing no less than fifteen groups. 146 A refined chronology of the production and use of such fibulae is one of the most impor tant results to come out of successive attempts to create a coherent typology. Syna Uenze has long established a dire ct connection between the sixth century cast fibula with bent stem and the early fibula with bent stem, the latter being a dress accessory frequently found in assemblages of the third to fourth century Sntana de Chernyakhov culture. 147 As we have seen, such fibulae continued to be produced in the sixth century, and in all probability they were contemporary with cast fibula e Indeed, mos t datable assemblages with cast fibulae with bent stem indicate a date within the second half of the sixth century, which seems to have been the "golden age" of such fibulae. Although it was supposed that production ceased around 600, extended the chronology of the cast fibulae with bent stem into the first three decades of the seventh century, based on the finds from Klked Feketekapu. Such a proposition needs further confirmation in order to gain full credibility, especially since no evidence exists of cast fibulae with bent stem in well 145 I. O. Gavritukhin, "Fibuly vizantiiskogo kruga v Vostochnoi Evrope (litye dunaisko illiriiskie)," MAIET 9 (2002): 232 33. 146 bizantine," 100 11. 147 Uenze, "Gegossene Fibeln ," 483.

PAGE 141

141 Fig ure 4 11. Cast fibulae with bent stem found in barbaricum and their typological analogues in the frontier provinces. 1. Volos'ke; 2. Zvonets'ke; 3. Hansca; 4. vsrhely; 16. Klked Feketekapu.

PAGE 142

142 dated assemblages in the lands still under imperial control in the early seventh century. Moreover, while the presence of such fibulae in seventh century assemblages seems to suggest that they were still in use, their production in the northern Balkans was brou ght to an end by the collapse of the frontier in several stages from the last decade of the sixth century to the beginning of the seventh. Unfortunately, there is no way of refining the chronology at the level of variation in shape, length and decoration o f the cast fibulae with bent stem. The largest group is represented by fibulae with no decoration on the bow or foot, which seems to be the standard type and possibly the longest in production. Local fashions seem to have be en responsible for the appearanc e of several types of decoration on the bow and/or foot, some of which can be localized with some precision based on the geographic distribution of finds, but their exact chronology is impossible to determine. 148 While typological debates remain important as a tentative foundation for interpreting future finds, they are only marginal for our purposes here. They will be used only to identify those types that do not appear in barbaricum as well as to highlight certain geographical patterns of distribution and t he extent to which they differ from the situation observed in the Danubian provinces. Ultimately, there is no certainty that the order brought by modern scholars into the wide variety of types and sub types and their subsequent distribution on a map reflec t with any accuracy the intentions of sixth century producers and users of cast fibulae with bent stem. One of the most interesting cast fibulae with bent stem found in barbaricum is that from Davideni, which belongs to the group of fibulae with human fac e s on the bow (Figure 4 20/10). The fi nd has been 148 bizantine," 111 13.

PAGE 143

143 interpreted as a badge of Christian identity under the assumption that the human faces are portraits of saints. The prototype may indeed have been designed to portray religious images but the symbolism was lost on many such fibulae which display a very schematic decoration, such as the one from Chornivka in Ukraine, a more distant find from barbaricum than the one from Davideni. 149 Indeed, the fibula from Davideni might have actually been produced in the front ier provinces of the Balkans, as the closest analogies come from Pernik, where no less than four specimens of this type have been found. 150 That the fibula from Davideni had religious meaning cannot be completely ruled out, but the archaeological context of house 51 where it was found in association with fragments of wheel and hand made pottery, is not explicitly Christian. In addition, no fibulae with cross shaped foot have been found so far in barbaricum proper although some may have been found on the nort hern bank of the Danube in the Iron Gates area. 151 Fibulae with a rectangular section of the bow are very common in the northern Balkans especially in the provinces of Scythia and Moesia Secunda, but are rarely found in barbaricum. One of the two known spe cimens in barbaricum is the one found in house 25 from 4 19/5), in Moldavia together with a belt buckle of the Sucidava class, typical of military assemblages from the frontier region, suggesting a higher status of the owner. 152 It is not impr obable that the owner of the fibula from 149 Gavritukhin, "Fibuly vizantiiskogo," 239 40; 250 fig. 6/1. 150 Liubenova, "Selishteto ot rimskata," 172 fig. 112/4 7; 173 fig. 113/1 4. 151 Catalog," Analele Banatului 9 (2001), 202 pl. IV/2. 152 AM 12 (1988), 249 250; 250 fig. 1/1. The other specimen was found ve ry far from the frontier, at Bukhlichskii

PAGE 144

144 was a mercenary involved in the numerous military conflicts in the Danube region in the second half of the sixth century. The shape of the bow resembles specimens such as the ones found at Ibida and Venchan, but it is clearly an imitation produced in barbaricum with modifications like the absence of the fake spiral and the rectangular shape of the bent stem. Imitation seems to have been ordinary practice in barbaricum. The specimens from 4 1 9/6) display multiple striations on the foot, a type of decoration produced only outside the Empire as no analogies are known so far from the imperial provinces of the Balkans. 153 In that respect the fibula from Zvonets'ke comes closer to the genuine specimens from the Balkans due to the presence of the fake spiral which is differentiated from the f its foot, resembles another imitation from b arbaricum, th e iron fibula with bent stem (not cast) from Davideni. 154 It is possible that imitations were occasionally made after earlier local imitations, closer to home, rather than after new fashions from south of the Danube. Fibulae with a semicircular section of th e bow belong to the type most commonly found in the frontier region, with a correspondingly significant presence in barbaricum i n a wide geographical area from the Middle Danube to the middle Dnieper region. Indeed, two fibulae have been found in Klked in burial assemblages. One of them is a child burial (grave 492, cemetery A), the fibula being associated with bronze and iron chains Khutor, near Petrikov, Mozyr region (Belarus), for which see V. S. Viargei, "Poseleniia prazhskoi kul'tury Belorusskogo Poles'ia," in Problemy slavianskoi arkheologii ed. by Valentin V. Sedov (Moscow: Institut Ark heologii R AN, 1997), 35 36; 35 fig. 2/3. 153 D. G. Teodor, The East Carpathian Area of Romania in the V XI Centuries A.D. (Oxford: BAR International Series 81, 1980), 12; fig. 11/3. Prikhodniuk, Pen'kovskaia kul'tura 157; 142 fig. 74/10. 154 Mitrea, Comunit 329 fig. 68/1.

PAGE 145

145 as well as two silver earrings and glass beads with eye shaped inlays. The second, a burial of a female, has a rich inventor y including a silver fibula of the Cividale class, typical for the early seventh century, earrings, belt mounts, and strap ends, all made of silver. 155 A good parallel for this practice can be found in the province of Scythia, where a child burial from Piatr a Frec ei (grave E143) produced a pair of cast fibulae with bent stem in association with two silver earrings with star shaped pendants and a large n umber of glass and lead beads (Figure 4 19/7). 156 Three fibulae with a semicircular section have been found east of the Carpathians at Hansca, Bor and Bac latter being found in a typical association for the late sixth century, with handmade pottery and clay pans. 157 No perfect analogies can be found south of the Danube, although their shapes are somewh at similar to many of the cast fibulae with semicircular section of the bow found in the frontier provinces at Novae, Kapitan Dimitro vo, or 155 A. Kiss, Das awarenzeitlich gepidische 132; 503 pl. 89/492.3; A. Kiss, Das awarenzeitliche Grberfeld in Klked Feketekapu B (Budapest: Magyar Nemzeti Mzeum/Magyar Tudomnyos Akadmia Rgszeti Intzete, 2001), 35; pl 30/40. An iron fibula of this type was found in Budapest, still in a burial assemblage, for which see M. Nagy, Awarenzeitliche Grberfelder im Stadtgebiet von Budapest (Budapest Magyar Nemzeti Mzeum MTA Rgszeti Intzete, 1998), 29; pl. 33/21. A cast fibula with zig associated with a belt buckle with rectangular plate with embossed decoration, for which see Bna and Nagy, Gepidische Grberfelder 45; 346 pl. 72/2. 156 A. Petre, La romanit en Scythie Mineure (II e VII e sicles de notre re). Recherches archologiques (Bucharest: AIESEE, 1987), 79 and pl. 145/239d. The association of glass beads with cast fibulae with other graves produced such fibulae, in one case, a cremation burial, the cast fibula with bent stem being found together with unidentifiable fragments of other fibulae, for which see U. Fiedler, Studien zu Grberfeldern des 6. bis 9. Jahrhunderts an der u nteren Donau (Bonn: Habelt, 1992), 80 83, figs. 11/8, 11/11, and 11/16. 157 Hansca: I. A. Rafalovich, "Issledovaniia ranneslavianskikh," 153; 152 fig. 10/2. : I. Mitrea, nistrean," Zargidava 1 (2002), 24; 39 fig. 9/7. : Carpica 4 (1971): 236; 242 fig. 13/1. A variation of the fibula with semicircular bow, having two vertical incisions has been recorded in the same region, at Suceava, for which see D. G. Teodor, "La pntration des Slaves dans les rgions du sud est de l'Europe d'aprs les donn es archologiques des rgions orientales de la Roumanie," Balcanoslavica 1 (1972), fig. 4/3.

PAGE 146

146 Accres (Cape Kaliakra) (Figure 4 19/8). 158 The fibula from Hansca has a closer parallel at per, and could point to a connection between the two regions, since the fi similarities. 159 Surprisingly, most cast fibulae with bent stem found in barbaricum do not find their closest parallels in the Iron Gates area, which was initially considered to be the main area of prod uction, but in the north eastern Balkans, in the provinces of Moesia II and Scythia. This distribution also explains the larger number of finds in Moldavia, a region with an easier access to goods from the two Danubian provinces. It is no accident that fro ntier forts defending these provinces also produced the largest quantity of "non Roman" hand made ceramics, often attributed to outside populations, Slavs in particular, and dated to the second half of the sixth century and the beginning of the seventh. Si nce the golden age of the production and distribution of cast fibulae with bent stem seems to have been the second half of the sixth century, this dovetails nicely with other categories of finds It suggest s that the movement of mercenaries recruited from b arbaricum to man the garrisons on the Lower Danube may have be en responsible for the distribution of cast fibulae with bent stem in the region east of the Carpathians. 4.2.4.3 Sixth to Seventh Century Bow Fibulae The stylistic variety of the Byzantine fibulae with bent stem which created so many problems to archaeologists trying to establish a firm typology turns into nightmare in the case of the so called "Slavic" bow fibulae, whose diversity defies any attempt to 158 Kapitan Dimitrovo: A. Kharalambieva, "Fibuli ot I VII v. v Dobrichkiia muzei," Dobrudzha 9 (1992), 137 38; 134 pl. III/7. Accres: Kharalambieva, "Dva tipa kusno antichni," 39; pl. VI.7. Novae: S. Stefanov, Izsledvaniia v pamet na Karel Shkorpil ed. K. Miiatev and V. Mikov (Sofia: 26. 159 Prikhodniuk, Pen'kovska ia kul'tura 156; 142 fig. 74/9.

PAGE 147

147 create a universally accepted classif ication. Given that additional finds have done nothing but increase that confusion, it is not surprising that the most influential cl assification is also the oldest. Joachim Werner's typology is based on the shape of the terminal lobe, either a human face or an animal head, and the firm conviction that the bow fibulae should be considered an ethnic badge of the early Slavs. 160 For all its methodological shortcomings, Werner's typology is still employed by historians and archaeologists as a common foundation f or further discussion of the bow fibulae, most of wh om depart from Werner's system of classification to create a new typologic al arrangement no less subject to methodological pitfalls and arbitrary criteria. To be sure, Werner's classification as well as h is interpretative framework came under attack soon after publication, but serious attempts to replace it have been made only in the past twenty years. 161 Entire dissertations have been devoted to this topic by Liudmil Vagalinski and Christina Katsougiannopou lou, 162 while Florin Curta has embarked on redefining each of Werner's classes in a series of articles. 163 Despite the great efforts of 160 J. Werner, "Slawische Bgelfibeln des 7. Jahrhunderts," in Reinecke Festschrift zum 75. Geburtstag von Paul Reinecke am 25. September 1947 ed. G. Behrens and E. Schneider (Mainz: E. Schneider, 1950), 150 72. 161 H. Khn, "Das Problem der masurgermanischen Fibeln in Ostpreussen," In Documenta Archaeologica Wolfgang La Baume dedicata 8.II.1955 ed. O. Kleeman (Bonn: L. Rohrscheid, 1956), 79 108; I. Nestor, "L'tablissment des Slaves en Roumanie la lumire de quelques dcou vertes archologiques recentes," Dacia 5 (1961): 429 49; A. Petre, bizantine la geneza unor VII e.n.," SCIV 17, no. 2 (1966): 255 75; D. Pallas, "Donnes nouvelles sur quelques b oucles et fibules consideres comme avares et slaves et sur Corinthe entre le VI e et le IX e s., Byzantino bulgarica 7 (1981): 295 318; D. G. Teodor, "Fibule digitate din secolele VI VII pontic," AM 15 (1992): 119 52. 162 L. Vag alinski, Zur Frage der ethnischen Herkunft der spten Strahlenfibeln (Finger oder Bgelfibeln) aus dem Donau Karpaten Becken (M. 6. 7. Jh.), Zeitschrift fr Archologie 28 (1994): 261 305; C. Katsougiannopoulou, "Studien zu ost und sdosteuropischen B gelfibeln," Ph. D. Dissertation, University of Bonn, Bonn 1999, 1 112. 163 Estonian Journal of Archaeology 16, n.1 (2012): 1 44; F. Curta, "Zhenshchina iz Danchen' ili k voprosu o fibulakh tipa II C po Verneru," Tyragetia 5, n. 1 (2011): 153

PAGE 148

148 Bal kan historians to bring order to a very large and exaspera tingly diverse body of finds, general consensus is yet to be re ached. 164 Two major historical issues are at stake, aside from the taxonomical inclinations of archaeologists: the chronology and the ethnic attribution of the sixth to seventh century bow fibulae. Much of the debate surrounds the cultural links between the groups of fibulae found in the Dnieper region, the Crimea, the areas north of the Danube, and Eastern Prussia, a problem beyond the scope of this enquiry which aims to establish the nature of the relations between Byzantium and barbaricum before the fall of the Danube frontier. Nevertheless, the origin of the so called Slavic bow fibulae, as well as their limited presence south of the Danube (compared to the north), raises some important questions related to the new fashions developing in the frontier regi on during the long sixth century (Figure 4 12). One of the issues is the stylistic origin of the Slavic bow fibula, which is clearly not an original creation of the sixth century. It is generally accepted that the late bow fibula derives from the East Ge rmanic fibula of the late fifth century, with semicircular head and terminal lobes, itself depending on previous shapes combining Gothic (Chernyakhov) and Roman provincial elements. Its "Slavic" counterpart is no less a mix of "barbaric" and early Byzantin e elements, which has led Class II A," in (Bucharest: Renaissance, 2 010), 149 75; F. Curta, "A Note on the 'Slavic' Bow Fibulae of Werner's Class I J," Archaeologia Baltica 12 (2009): 124 36; F. Curta, "Neither Gothic, nor Slavic: Bow Fibulae of Archaeologia Austriaca 95 (2009): 45 77; F. Curta, "Once Again on Bow Fibulae of the 'Pietroasele Type' (Werner's Class I F)," AAASH 59 (2008): 465 92; F. Curta, "Some Remarks on Bow Fibulae of Werner's Class I C," Slavia Antiqua 49 (2008): 45 98; F. Curta, "Slavic Bow Fibulae? AA ASH 57 (2006): 423 74; F. Curta, "A Contribution to the Study of Bow AM 29 (2006): 93 124; F. Curta, "Female Dress and Slavic Bow Fibulae in Greece," Hesperia 74, n. 1 (2005): 101 Bow Fibulae Revisited," AB 8, n.1 (2004): 59 78. 164 See recently U. Fiedler, "Die slavischen Bgelfibeln von Joachim Werners Gruppe I. Bemerkungen yum Forschungsstand unter besonderer Berck sichtigung des Typs I C ," in n onoarea lui Radu Harhoiu 52; F. erratum corrigendum cum commentariis EN 21 (2011): 63 110.

PAGE 149

149 some scholars to conclude that the Crimea was a major center for the production of such fibulae, regardless of whether the Crimean Goths played a role in the creation of this type. 165 Alternatively, it has been suggested that the ma in centers for the production of such fibulae should be located south of the Danube, based on the production of the earlier "Germanic" type in the frontier region, as well as the typical early Byzantine decoration patterns on some "Slavic" types, the natur alistic human faces in particular. 166 Indeed, bow fibulae of a superior craftsmanship which are often gilded have been found in the Balkans at Lezh, Liuliakovo, and Istanbul, all belonging to Werner's class I B. Such specimens are also recorded in barbari cum (Figure 4 19/9), the most 167 In the absence of molds or workshops as direct evidence for the production of such dress accessories in the Balkans, s tylistic criteria as well as finds of bow fibulae of a higher artistic complexity, of which subsequent "Slavic" types might have derived, remain the only criteria for attributing their origin to early Byzantine production centers in the Danube region. Out of the entire c orpus of sixt h and seventh century bow fibulae, only ca. fifteen percent have been found south of the Danube, and indeed the highest proportion is noted in the case of the types with human masks on the appendix, whose production was often located in the Balkans. By co ntrast the production of bow fibulae in barbaricum can be show with certainty after the find of prototypes at Felnac, 165 B. A. Rybakov, "Drevnie rusi. K vop rosu ob obrazovanii iadra drevnerusskoi narodnosti v svete trudov I. V. Stalina," Sovetskaia Arkheologiia 17 (1953): 23 104. 166 Petre, bizantine," 267 75 ; Teodor, "Fibule digi tate din secolele VI VII," 123. 167 I. Nestor and C S. Nicolaescu Germania 22, n. 1 ( 1938): 33 35 and pl. 7.

PAGE 150

150 Fig ure 4 12. Sixth seventh century "Slavic" bow fibulae in the frontier region and their parallels in barbaricum. 1. Velesnica; 2. Korb ovo; 3. Aquis (Prahovo); 4. n Grad; 21. Shumen; 22. Sturmen; 23. Iatrus (Krivina); 24. Vardim; 25. Iambol; 35. Istanbul (?); 36. Nea A nchialos; 37. Demetrias; 38. Dion; 39. Tropaeum Traiani ( Adamclisi); 40. Carevec; 41. Sv. Erazmo ; 42. Chufut Kale;

PAGE 151

151 Bucharest Tei, and in Banat and especially after the spectacular find of 64 molds in a bu ilding from Bernashivka, one of which was designed to produce a type of bow fibu la so far unknown among finds (Figure 4 18/5). 168 Chronology is perhaps the only chance of bringing some order to the larger corpus of bow fibulae, given that their stylistic var iation prevents any classification based on shape and artistic style. 169 Unfortunately, more than half of the total number of bow fibulae found in Eastern Europe lack s any archaeological context. Being one of the artifacts which survived the great transforma tions of the early seventh century, which marked the Byzantine retreat from the Lower Danube, such a historical event cannot be used as a chronological benchmark for dating the "Slavic" bow fibulae. In addition, fibulae belonging to the same class are ofte n found in different chronological contexts, i.e. before and after the fall of the Danube frontier. To make matters worse, each bow fibula is unique, with no identical pairs and therefore any chronological arrangement based on Werner's classes, or any clas sification for that matter, is difficult. For the same reasons, the usual method of extrapolating a secure dating of one specimen from a class in order to date the entire class cannot be employed in the case of "Slavic" bow fibulae. Analyzing each context on its own terms is perhaps the only acceptable course of action, even if the approach is not going to be completely satisfactory for answering broad questions regarding this category of artifacts. Given that many bow fibulae found 168 E. A. Shablavina and B. S. Szmoniewski, "The Forming Model of the Kertch Type Finger Shaped fibula," Sprawozdania Archeologiczne 58 (2006 ): fig. 5; Vinokur, Slov'ians'kyi iuveliry 57, fig. 18. 169 In his 1950 contribution Werner dated all "Slavic" bow fibulae to the seventh century, but later he allowed for an earlier dating going back to the second half of the sixth century, for which see J Werner, "Neues zur Frage des slawischen Bgelfibeln aus sd osteuropischen Lnders," Germania 38 (1960): 114 20. A more radical chronology was suggested by Uwe Fiedler based on similarities between the late fifth century "Germanic" bow fibulae and "Slav ic" bow fibulae, which Fiedler dated to the sixth century, see Fiedler, Studien zu Grberfeldern 101.

PAGE 152

152 south of the Danube are accidental or stray finds, there is no certainty that they had been produced or worn before the Empire lost its administrative control over the northern Balkans. To be sure, many of the "Slavic" bow fibulae found in a clear archaeological context south of the Danube were almost certainly lost or deposited after the fall of the limes in the Balkans ( ca 615). Such is the case of the fibula type I B found in grave 36 and the fibula type I C from the grave 32 at Lezh in Albania, a cemetery dated to the middle and especially the second half of the seventh century. A strikingly similar specimen was found at Dubovac in Serbia, unfortunately a stray find, which might also date from this period. 170 The pair of fibulae class I C from grave 28 at Kruje, also from Alban ia, can be dated to the same period based on its association with a buckle of the Corinth class. 171 A close analog is the fibula found accidentally at C 4 19/10), which might be contemporary with the pair of fibulae from Kruje. 172 The gilt fibula from Liuliakovo in Bulgaria resembles the one probably found in Istanbul and by extension with the spectacular fibula from ovenii de Jos, found together with silver earrings with star shaped pendants, typical for the seventh century. 173 An even later dating should be accepted for the two class I C fibulae from the hoard found at Kamenovo in northern Bulgaria, which also included a 170 F. Prendi, "Nj varrze e kulturs arbrore n Lezh," Iliria 9 10 ( 1979 1980 ): 129 and 166 67, pl. 20/3 and 21/2; D. Dimitrijevi ma Vojvodine u vreme doseljenja Slovena, in Simpozijum 26. oktobra 1968 u Mostaru ed. A. Benac (Sarajevo: Akademija nauka i umjetnosti Bosne i Her cegovine, 1969), 88 and fig. 1. 171 S. Anamali and H. Spahiu "Varrza arbrore e Krujes," Iliria 9 10 (1979 1980 ): 61 62 and pl. 7/11 and 11/1. 172 D. G. Teodor and C. Chiriac "Noi fibule digitate din Dobrogea," Peuce 3 4 (2005 2006): 241; 249 fig. 2. 173 S. Mikhailov, "Die Bgelfibeln in Bulgarien und ihre historische Interpretation," in Archologie als Geschichtswissenschaft. Studien und Untersuchungen ed. J. Herrmann (Berlin, Akademie Verlag, 1977), 317 18 and pl. 7; J. Werner, "Neues zur Frage des slawischen Bgelfibeln aus sd ost europischen Lnders," Germania 38 (1960): 119 and pl. 2.

PAGE 153

153 pair of fibulae class I F, the only ones of this type found in the Danube region, but also a strap end with scrollwork decoration, most typica l for the Late Avar period ( after 700). 174 To be sure, an equally large number of fibulae can be dated with some c ertainty before the fall of the Danube frontier. No doubt the safest dating is that of the I H fibula found during excavations in the Danubian fortress of Iatrus, in the buildin g 66/23N, together with a sixth century coin, whose exact dating is unfortunat ely unspecified in the publication. 175 The excavators have dated the complex to the last phase of the fortress which also covers the first two decades of the seventh century. In addition, the pair of I H fibulae found in the inhumation burial B 42 at Piatra recently dated to the sixth century, based on the association with copper alloy bracelets typical for this period. 176 A similar dating can be advanced for the fibula found in an inhumation burial at Tulcea in northern Scythia, which was found together with a ceramic pitcher and nine amber beads. 177 Another coin dated fibula was found at Dervent, close to the Danube, having a terminus post quem 570/1, but the stratigraphical association between the two artifacts is not secure. It belongs to Werner s class II A, as do a few other finds from the same region, the three stray finds from Somova ( Figure 4 20/11) and the specimen found during archaeolo gical 174 R. Rashev, Prabulgarite prez V VII vek 15. A fibula class I F was also found in an unknown location in Bulgaria, S. Mikhailov, "Rannosrednovekovni fib uli v Bulgariia," IAI 24 (1961): 43 and 41 fig. 3/1. 175 J. Herrmann, "Die archaologischen Forschungen des Zentralinstituts fur Alte Geschichte und Archaologie 1973 und im ersten Halbjahr 1974," Ausgrabungen und Funde 19, n. 6 ( 1974): 303 and pl. 44/b. 176 A. Petre, "Predvaritel'nye svedeniia Dacia 6 (1962): 226 fig. 12/1 b/1; F. Curt a, "Werner's class I H," 67 68. 177 G. Simion, "Un nouveau groupe de fibule digitales dcouvertes dans la rgion du Dobroudja," Studia antiqua et medievalia. Miscellanea in honorem annos LXXV peragentis Professoris Dan Gh. Teodor oblata ed. D. Aparaschivei (Bucharest: Editura Academiei Romne, 2009 ), 412, fig. 1.

PAGE 154

154 a scrollwork motif whose quality of execu tion suggests an early date in the sixth century. 178 An equally secure dating before the collapse of the frontier is offered by two settlement finds of cl ass II C fibulae from Carevec (Figure 4 stem. 179 Less certain is the dating of the pair of fibulae class I D found in an inhumation burial at Edessa in Greece, and associated with a buckle of the Sy racusa class, whose dating spans the entire seventh century. 180 A fibula of the same class was found in the early Byzantine settlement of Dinogetia in northern Scythia whose closest parallel is a fibula from Bashtanovka in Crimea, whose association with a Ge rmanic rectangular buckle with eagle head dates it to the second half of the sixth century. 181 Although not much can be said about the bulk of the finds of "Slavic" bow fibulae from the Balkans, it is clear that their presence spans a long period of time, fr om before and certainly continuing after the fall of the military frontier on the Lower Danube. Recent distribution maps as well as inventories of finds show that most of the "Slavic" bow fibulae known until now from Eastern Europe dated ca 500 have been found in 178 SCIV 13, n. 2 (1 962): 447 48 and fig. 1. Somova: Simion, "Un nouveau groupe," 414; 415 16, fig. 2 3. SCIVA 41, n. 4 ( 1990): 37 38 and 48 fig. 19/37. 179 A. Kharalambieva, "Bgelfibeln aus dem 7. Jh. sdlic h der unteren Donau," in Actes du XII e Congrs international des sciences prhistoriques et protohistoriques, Bratislava, 1 7 septembre 1991 ed. J. Pavuj (Bratislava, Institut Archologique de l'Acadmie Slovaque des Sciences, 1993), 25 and 26 fig 1/1; V. Popovi "Byzantins, Slaves et autochtones dans les provinces de Prvalitane et Nouvelle Epire," in Villes et peuplement dans l'Illyricum protobyzantin. Actes du colloque organis par l'cole Franaise de Rome, Rome 12 14 mai 1982 (Rome: Ecole Franaise de Rom e, 1984), 175 and 176 fig. 188. 180 P. Petsas, "Archaiotetes kai mnemeia Kentrikes Makedonias," Archaiologikon Deltion 24 (1969): 307 and fig. 320/ 181 I. Nestor, "L'tablissement des Slaves en Roumanie la lumire de quelques dcouvertes archologiques rce ntes," Dacia 5 (1961): 440 and 444 fig. 3/1a b; I. S. Pioro, Krymskaia Gotiia (Ocherki etnicheskoi istorii naseleniia Kryma v pozdnerimskii period i rannee srednevekov'e) (Kiev: Lybid', 1990), 135 and 111 fig. 31. For eagle headed buckles found in Crimea, see E. A. Shablavina, "O rannesrednevekovoi produkcii Bosporskikh iuvelirov (na primere orlinogolovykh priazhek," Arkheologicheskie vesti 13 (2006): 230 51.

PAGE 155

155 barbaricum making it a typical dress ornament of the populations beyond the frontier. The multi ethnic nature of the frontier region explains the presence of such items in the northern Balkans. A female dress accessory, the "Slavic" bow fibula was probably worn by the wives of mercenaries recruited from barbaricum to defend the provinces of the Lower Danube. Whether or not such fibulae were an exclusive social and ethnic marker of "foreigners" from b arbaricum, they were clearly not as popular in fr ontier forts as they were north of the Danube, judging by the number of finds in both regions, and they certainly did not surpass the popularity of the fibulae with bent stem (including the cast ones). Identifying the bow fibula as the ethnic badge of a ce rtain ethnic group, as Werner has suggested by attributing them to the Slavs, has become increasingly difficult if not altogether impossible since the corpus of finds has come to include a variety of contexts, many of which are unrelated to what archaeolog ists traditionally describe as early Slavic. The map reveals a greater concentration of finds in the north eastern Balkans, in the provinces of Moesia II and Scythia, where a larger quantity of Penkovka and Korchak ceramics has been found, usually associat ed with the Antes and the Slavs, respectively. Is this an additional piece of evidence to suggest that the Slavs were in fact the main ethnic group using bow fibulae? The answer must be negative, since many of the contexts are not otherwise indicative of w hat may be viewed as Slavic material culture. Bow fibulae have been found in early Byzantine settlements such as Di Mediterranea and even further west at Butrint in Albania. An interesting concentration of finds may be observ ed in the Iron Gates region, on both sides of the river, an area

PAGE 156

156 under con stant Avar pressure, especially after the fall of Sirmium (582). An even larger number of specimens have been found in inhumation burials at Adamclisi, Tulcea, and as in a number of cemeteries from Crimea. 182 One might argue that perhaps Slavs recruited to defend the frontier provinces adopted local customs including the practice of inhumation, but such finds are also reported north of the Danube. To be sure, Slavic bow fi bulae are sometimes found with cremations such as those of the large cemetery 183 Found in a variety of contexts, associated with the Slavs, Avars, Ge pids, and the early Byzantine fortresses and cemeteries in the Balkans, the sixth to seventh century bow fibula can be hardly described as an ethnic badge. It was, however, a fashionable dress accessory on both sides of the river around the time when the B yzantines lost control of the frontier, but even more so in regions far removed from the Danube, in the Dnieper region and in Mazuria. 182 acest tip," Pontica 20 (1987): 207, 209 210; 208 fig. 1/a b. "Les Slaves du centre balkanique du VI e au XI e sicle, Balcanoslavica 1 (1972): 47 and fig. 1/3. Sv. Erazmo: V. Malenko, "Ranosrednovekovnata materijalna kultura vo Okhrid i Okhridsko, in Okhrid i Okhridsko niz istorijata ed. M. Apostolski (Skopje: Sobranie na opshtina Okhrid, 1985), 289 and pl. VI/4. A l arge number of "Slavic" bow fibulae have been found in cemeteries in Crimea at Luchistoe, Skalistoe, Chufut Kale, Eski Kermen and Suuk Su, for which see especially A. I. Aibabin, Khronologiia mogil'nikov Kryma pozdnerimskogo i rannesrednevekovogo vremeni, MAIET 1 ( 1990): 5 68; G. F. Korzukhina, "Klady i sluchainye nakhodki veshchei kruga "drevnostei antov" v srednem Podneprov'e. Katalog pamiatnikov," MAIET 5 (1996): 352 435, 586 705. 183 Fiedler, Studien zu Grberfeldern 83 fig. 11. Bratei : L. Brzu, Ein gepidisches Denkmal aus Siebenbrgen. Das Grberfeld 3 von Bratei (Cluj Napoca: Accent, 2010), 208; 301 pl. 23/G130/2 (and several other specimens). Rafalovich and Lapushnian, "Raboty Reutskoi," 139 and 131 fig. 9/2. Rafalovich, Dancheni. Mogil'nik cherniakhovskoi kul'tury (Kishinew: Shtiinca, 1986), 25 26 and pl. 14/1. Pruneni: M. donaulndischen Raum im 6. 7. Jahrhundert," Zeitschrift fr Archologie 7 (1973) : 210.

PAGE 157

157 4.2.4.4 Byzantine buckles The great variety of types and variants noticed in the case of fibulae is also shared by sixth to seventh century Byzantine buckles found in the frontier region and beyond (Figure 4 13). This category of artifacts is, however, less homogeneous and only a few types have been studied and classified. While a large number of buckles of basic shape and no decoration found in barbaricum are often named Byzantine, their dating largely depends on the archaeological context of their find. Many of them may well be local products, possibly imitations, but their connection with contemporary buckles from the Dan ube region is uncertain. There are, however, at least four well established types in the literature, dated to the sixth and seventh centuries, which will constitute the main basis for assessing the role of Byzantine buckles in the regions north of the Lowe r Danube. The most widespread category of Byzantine buckles north of the Lower Danube is by far the Sucidava class, with a number of spin offs known as Beroe in the literature, named so after the sixth cen tury cemetery excavated at Piatra 184 The type was defined by Joachim Werner in 1955 based on finds from Sucidava, the most important Byzantine bridge head on the left bank of the Lower Danube, which he dated to the second half of the si xth century. 185 Much like in the case of fibulae, the early 1990s brought a renewed interest in the typology of the Sucidava class of Byzantine buckles. Syna Uenze and Dan G. Teodor introduced new classifications, 184 A. Madgearu, The Sucidava Type of Buckles and the Relations between the Late Roman Empire and the Barbarians in the 6th Century," AM 21 (1998): 217 22. For specimens found at Piatra Petre, La romanit en Scythie Mineure. 185 J. Werner, "Byzantinische Grtels chnallen des 6. und 7. Jahrhunderts aus der Sammlung Diergardt," Klner Jahrbuch fr Vor und Frhgeschichte 1 (1955): 39 40.

PAGE 158

158 Fig ure 4 13. Main types of Byzantine bu ckles found in the frontier region: Sucidava Beroe (circle), Salona Histria (triangle), Ppa (square), and Syracusa (star). a) Sucidava Beroe: ac; 5. Alba Iulia (Apulum); 6. Sucidava; 7. Drobeta (Turnu Severin); 8. Diern 9. Pecica; 10. srhely Kishomok; 13. Szentes Nagyhegy; 14. Tiszafred; 15. Jnosh i da; 16. Tatabanya; 17. Klked Feketekapu; 18. Cristuru Secuiesc; 19. Sukhanova; 20. Volos'ke; 21. Kecskemt; 22. Pcs (Sopianae); 23. Szegvr; 24. Trkblint; 25. b) Salona Histria: 1 Cskberny; 10. Szekszrd; 11. K lked Feketekapu; 12. Szeghegy. c) on; 5. Klked Feketekapu; 6. Gyd; 7. Pcs Kztemet; 8. Keszthely; 9. Krnye; 10. Ppa. d) Syracuse: 1. Szelevny; 2. Szeged Fehrt; 3. Sz zhalombatta; 4. Cskberny; 5. Dunapentele; 6. Klked Feketekapu; 7. Bratei.

PAGE 159

159 while Vladimir Varsik, Uwe Fiedl er, and Alexandru Madgearu published new studies and inventories of finds. 186 The most recent inventories are a good indicator of the increasing number of finds of Sucidava Beroe buckles, many of them the result of the use of metal detectors, a mixed blessin g affecting the archaeology of north eastern Bulgaria in particular. Students of early Byzantine buckles can now rely on a large inventory of ca 150 buckles from the Middle and Lower Danube regions, almost half of them having been found in sixth century s ettlements. 187 In addition, a number of Sucidava buckles are recorded in Western Europe, as far as Normandy, evidencing the popularity and wide circulation of this type. 188 The large concentration of finds in the northern Balkans clearly indicates a local prod uction of the Sucidava class of buckles, although their presence in Constantinople, in Asia Minor and as far as Palmyra in Syria led some scholars to conclude that multiple workshops existed for the production of this popular type. 189 While this seems plausi ble, it is equally possible that Sucidava buckles 186 Uenze, Die Sptantiken Befestigungen 184 87; D. G. Teodor, "Piese vestimentare bizantine din secolele VI o pontic," AM 14 (1991): 118 125; V. Varsik, "Byzantinische Gurtelschnallen im mittleren und unteren Donauraum im 6. und 7. Jahrhundert," Slovensk Archeolgia 40, n. 1 (1992): 78 80; Fiedler, Studien Grberfeldern 71 73; A. Madgearu, Continuita VIII (Bucharest: Editura Univ 187 Madgearu, "The Sucidava Type," 218; M. Daskalov and K. Trendafilova, "Kolanut v iuzhnodunavskite vizantiiski provintsii prez V I VII v.," in The Bulgarian Lands in the Middle Ages 7th 18th Centuries. International Conference, a Tribute to the 70th Anniversary of Prof. Alexander Kuzev, Varna, Septembre 12th 14th, 2002 ed. V. Iotov (Varna: Regionalen Istoricheski Muzei, 2005), 7 1 8. 188 M. Schulze Drrlamm, Byzantinische Grtelschnallen und Grtelbeschlge im Rmisch Germanischen Zentralmuseum. Teil I: Die Schnallen ohne Beschlg, mit Laschenbeschlg und mit festem Beschlg des 5. bis 7. Jahrhunderts (Mainz: Verlag des Rmisch Germ anischen Zentralmuseums, 2002), 150, fig. 54. The Beroe buckles, deriving from the Sucidava type, have a wider circulation in the east as far as Petropavlovo in Udmurtia, Russia and Brody close to the Ural mountains for which see V. A. Semenov, "Petropavlo VII vv. V iuzhnoi Udmurtii," Voprosy arkheologii Urala 7 (1967): 170, pl. II/22; R. D. Goldina and N. V. Vodolago, (Irkutsk: finds are also attested in the Caucasus, for instance in the cemetery of Tsibilium in Abkhazia, for which see Y. Voronov, Tsibilium: la ncropole apsile de Tsibilium (VII e av. J. C. VIII e ap.J. C). (Abkhazie, Caucase) (Oxford: John and Erica Hedges Ltd., 2007), 72 and 258, fig. 146 (grave 313). 189 Madg earu, "The Sucidava Type," 218.

PAGE 160

160 were brought to these regions by troops transferred from the Balkans to theaters of operation in the Near East, especially in the second half of the sixth century. The production of Sucidava buckles probab ly took off sometime in the first decades of the sixth century, as suggested by specimens found in association with coins. At Mokranjske Stene in eastern Serbia a Sucidava buckle was found together with a coin of Anastasius, while at Izvor, close to Pernik in western Bulgaria, a similar buckle was associated with a follis minted for Justin I. 190 Further south, at Nea Anchialos a buckle of the Sucidava Beroe II sub type was found together with a coin of Justinian. 191 In all three cases the association is not dir ect, in that the buckles and the coins were not found within the same assemblage. Instead, the coins simply date the "layer" in which the buckles were found. Indeed, in all cases the buckles might date from a later period since the coins remained in circul ation during the entire sixth century. At any rate, Sucidava buckles are mostly associated with early Byzantine hilltop sites from the frontier region along the Danube, from Belgrade through Aquis, Oescus, Novae, Durostorum to Dinogetia in northern Scythia and the larger towns of the western Black Sea coast, Histria, Tomis and Odessos. 192 There is a consensus that the production of Sucidava buckles ended with the Byzantine control of the Danube at the beginning of the seventh century, which indirectly links this fashion with the presence of garrisons defending the Byzantine frontier in the Balkans. 190 Varsik, "Byzantinische Grtelschnallen," 78; Daskalov and Trendafilova, Kolanut v iuzhnodunavskite," 9. 191 Lazar idis, "Nea Ankhialos," pl. 394. 192 Daskalov and Trendafilova, Kolanut v iuzhnodunavskite," 10, fig 1.

PAGE 161

161 Sucidava buckles were also fashionable outside the Empire, but unlike other categories of artifacts there is a conspicuous scarcity of finds in Wallachia and Molda via, the traditional recipients of imports from the Empire. Aside from the Byzantine bridge heads from the left bank of the Danube, Dierna, Drobeta, and Sucidava, none of the settlements that produced a significant number of finds of amphorae or fibulae y ielded Byzantine buckles of the Sucidava class, or any other type for that matter. In Moldavia three finds have been recorded so far, including (house 25), which seems to be a local imitation rather than a genuine import (Figure 4 20/17 193 G iven the great density of finds in the provinces of Scythia and Moesia II, the scarcity of Byzantine buckles in the corresponding regions across the Danube appears even more perplexing. The same people who were so eager to obtain Byzantine artifacts such as fibulae, crosses, and various types of jewelry appear to have been totally uninterested in expressing their identity or enhancing their social prestige by wearing Byzantine buckles. The same is true for the other three important classes of Byzantine buc kles which can be dated to before the fall of the Danube frontier: Salona Histria, Ppa, and Syracuse. 194 The first two classes have the same function, that of attaching a small 193 Romni a 1983 1992. A XXXI (Buchare st: CIMEC, 1997), 105 6, n. 70. 194 For Ppa buckles, see S. Uenze, "Die Schnallen mit Riemenschlaufe aus dem 6. und 7. Jahrh.," Bayerische Vorgeschichtsbltter 31, n. 1 2 (1966): 142 8 1. For Salona Histria and Syracusa, see Varsik, "Byzantinische Grtelschnallen," 80 82; Daskalov and Trendafilova, "Kolanut v iuzhnodunavskite," 11, fig. c Orgam/Argamum. Supplementa I. la recherche d'une colonie. Actes du Colloque International l'occasion du 40 me anniversaire des fouilles Orgam/Argamum, Bucarest

PAGE 162

162 purse to the belt, the difference between the two being the presence of a pair o f stylized animal heads on the Ppa type. The production of Ppa buckles postdates the Salona Histria type, from which it derives, although both may have been in use around year 600. Their presence in similar, Early Avar assemblages in Krnye (M 66, M 106, and M 109) and Klked Feketekapu (M 223, M 259, M 385, M 425, and M 647) substantiates this hypothesis. 195 In addition, both types of buckles have been found in the large 20/18 19). 196 In th e early Byzantine provinces of the Balkans Salona Histria and Ppa buckles display the same pattern of distribution as the Sucidava type, although Ppa buckles are so far missing from the provinces of Dacia Ripensis and Moesia I. Unlike Sucidava buckles, w hose popularity brought them to Western Europe, the circulation of Ppa and Salona Histria buckles was restricted mostly to the Middle and Lower Danube regions and theCrimea, where they are common finds in all the major sixth to seventh century cemeteries. We are facing the same paradox, with finds concentrated in "Avaria," the north eastern Balkans, and the Crimea, while they are almost completely absent north of the Danube in Wallachia and Moldavia. This is even more surprising since a workshop for the p roduction of Salona Histria buckles has been found at Drobeta on the Tulcea Jurilovca, 3 5 octobre 2005 teanu (Bucharest: AGIR, 2006), 348 52 and pl. V. For the distribution of Syracusa and Ppa buckles outside the Balkans, see Schulze Drrlamm, Byzantinische Grtelschnallen 176, fig. 62 and 226, fig. 82. 195 Krnye: A. Salamon and I. Erdly, Das vlkerwander ungszeitliche Grberfeld von Krnye (Budapest: Akadmia Kiad, 1971), 24, pl. 19 (Salona Histria) and 20, pl. 9/6 (Ppa); Klked Feketekapu: Kiss, Das awarenzeitlich gepidische 75 76, pl. 57 (Salona Histria) and 165 66, pl. 102 (Ppa). In both cases the S alona Histria buckles were associated with reused Late Roman coins (Valentinian I Gratian). 196 Uenze, "Die Schnallen," fig. 1/6 and 5/35.

PAGE 163

163 left bank of the Danube, which could have easily facilitated the distribution of such buckles further north in barbaricum. 197 The production of the last type of buckle under discussion, kno wn as Syracuse in the literature, took off close to the time when the Empire was losing control of the Danube frontier. Finds of Syracusa buckles in the Balkans are scarce compared to the other three types and only a few are known from the provinces of Moe sia II and Scythia, where the Empire retained some influence in the first decades of the seventh century due to the presence of the fleet in the Black Sea. Indeed, most of the finds are from coastal settlements such as Mesembria, Odessos, Accres, and Histr ia. Although V. Varsik included this type in the Mediterranean Dalmatian group, the largest density of finds can be found in Crimea, where almost every cemetery excavated yielded buckles of the Syracuse class. It is no less true that among all early Byzan tine buckles the Syracuse type is the most widespread, with finds reported as far as Britain, North Africa, and Iran. The large number of finds in the Middle Danube region compared to the Balkans shows that Syracuse buckles, and perhaps other types of Byza ntine buckles as well, might not have originated from the Lower Danube region but from the Dalmatian coast or even the Crimea, given the wide political influence of the Avars in barbaricum The impressive density of finds of Byzantine buckles of all types in the regions controlled by the Avars, the Middle Danube, the Tisza valley, and Transylvania has been explained in the context of the political and diplomatic contacts between the Empire and the early Avar khaganate. 198 Since most buckles were associated wi th 197 Bejan, "Un atelier metalurgic," pl. VI. 198 A. Madgearu, "A Buckle of Ppa Type Found in the Early Byzantine Fortress Halmyris (Murighiol, Tulcea County)," Peuce 15 (2003), 171 72; Varsik, "Byza ntinische Grtelschnallen," 89.

PAGE 164

164 weapons in Avar and Gepid graves they were most probably pieces of military equipment, although occasionally they are found in female graves. 199 If the majority of finds of Sucidava buckles are from the frontier provinces, Salona Histria and Ppa buckles were equally popular in Barbaricum the latter type being more often found in "Avaria" than in the Danubian provinces. The concentration of Ppa buckles in the regions under Avar control has led to the conclusion that they were locally produced outside the empire, perhaps by Byzantine craftsmen working for the Avars. 200 In reality, none of the sixth to seventh century Byzantine buckles can be associated with a certain ethnic group since they have been found in a variety of contexts, inhumation and cremation, male and female graves, hilltop settlements from the frontier region, towns on the Black Sea coast, as well as away from the frontier region, in Asia Minor. Such contexts have been associated with the Byzantines, the Slavs, the Avars, the Gepids, and the C rimean Goths, making these buckles elements of military fashion on a wide geographic area. Indeed, the Avars extended their influence from the Middle Danube to the steppes north of the Black Sea. However, the scarcity of such buckles south and east of the Carpathians may be seen as a form of resistance to the Avar domination, a conscious rejection of some of the items that identified Avar warriors toward year 600. The Avar influence is undeniable in Transylvania where the Gepids under Avar rule in and politically more assimilated into the Avar confederation than communities in Wallachia and Moldavia. 199 Klked Feketekapu: Kiss, Das awarenzeitlich gepidische Grberfeld 75 76, pl. 57 (M 425); Kiss, Das awarenzeitliche Grberfeld 144, fig 43 (M 45 7). 200 lea," SCIVA 44, n. 2 (1993): 171 83.

PAGE 165

165 4.3 Christianity North of the Danube Christianity is one of the m ajor cultural interfaces connecting the frontier provinces with the region s north of the Danube. A series of monographs written in the past decades have established an almost unchallenged narrative of the process of Christianization in barbaricum and espec ially in the territory of the former province of Dacia. 201 However, this common wisdom was recently contested based on a different interpretation of the archaeological evidence. The main contention is that the existence of Christian artifacts in barbaricum s hould not be seen as direct evidence for the existence of a Christian population. Items such as flasks, lamps and crosses may have not arrived north of the river for religious considerations but more probably as items required by local elites, part of the cultural process described as imitatio imperii 202 Indeed, the number of potential liturgical items such as the bronze lamps found in the former province of Dacia is unfortunately insufficient to argue in favor of the existence of a strong Christian communit y, since most of them are chance finds with no archaeological context to give them more meaning. Moreover, the absence of any documented churches dated to the sixth century adds to this uncertainty. It has been repeatedly suggested that sixth century churc hes north of the Danube could have been made of wooden structures or other perishable materials, impossible to trace in the 201 Barnea, Les monuments palochrtiens ; Gudea and I. Ghiurco, ; D. G. Teodor, lea (Bucharest: I nstitutul Romn de Tracologie, 1997); Madgearu, 202 F. Curta, "Limes and cross: The Religious Dimension of the Sixth century Danube Frontier of the Early Byzantine Empire," Starinar 51 (2001): 64 65.

PAGE 166

166 archaeological record. 203 While this hypothesis cannot be completely ruled out, the lack of any documented churches remains problemati c, especially since such examples exist in barbaricum the most spectacular being the basilica and extramural cemetery at Fenekpuszta, on the western shore of Lake Balaton. 204 The basilica with three apses from Fenekpuszta has good analogies in the Lower Dan ube frontier region, which begs the question as to why such religious buildings are so far missing from the region north of the Danube. The rarity of Christian inhumation burials north of the Danube is also striking. With the exception of the burial assem W est E ast orientation and a number of Byzantine grave goods, other inhumation graves south and east of the Carpathians canno t be described as Christian with any certainty. 205 As a matter of fact, sixth century cemeteries from barbaricum display a large variety of practices, which prevent us from drawing any meaningful conclusions at this time. Nevertheless, the lack of any notabl e inhumation cemeteries to contrast with large cremation cemeteries such a Monteoru adds to the uncertainty regarding the dominant religion north of the Danube or the nature of its practices. 203 Teodor, 82; Barnea, Les monuments palochrtiens 27; Madgearu, 82; Zugravu, Ge 337; Protase, Autohtonii n Dacia 77. 204 R. Mller, Die Grberfelder vor der Sdmauer der Befestigung von Keszthely Fenkpuszta (Budapest/ Leipzig: Magyar Tudomnyos Akadmia Rgszeti Intzet/Geisteswissenschaftliches Zentrum Geschich te und Kultur, 2010). 205 Teodor, no. 17, 26, 31, and 32. The religious symbolism of the fibula with cross uncertain. A similar grave, found at Dancheni in Moldova, contained two "Slavic" bow fibulae, glass beads and a hand made pot, for which see Rafalovich, "Dancheni," 25 27.

PAGE 167

167 What cannot be denied for the sixth century, however, is the absolute increase in the number and diversity of objects b earing Christian symbols found north of the Danube, many of which are imports from the Empire. 206 While most Christian objects dated to the fourth and fifth centuries concentrated in the former province of Dacia, the long sixth century ( ca 500 to ca 620) witnessed a dramatic increase in such finds in Wallachia and Moldavia. The lack of any documented missions from Byzantium leaves out any potential explanation based on an official policy of Christianization. 207 Since the Empire used missions as a political device during the Justinianic age in various peripheral areas such as the Caucasus, Yemen, and Nubia, we can only conclude that the Empire had little to gain politically from such initiatives north of the Danube. 208 As previously argued, the Christianization of Slavs or Avars was not taken into consideration by sixth century emperors, which come s in sharp contrast with fourth century interests in the area when Ulfilas was preaching the Gospel to the Goths. 209 Novella XI regarding the archbishopric of Justinia na Prima (Dacia Mediterranea) remains the single most important official document regarding th e sixth century religious policies in the Danube region. 210 The creation of an archbishopric at Justiniana Prima 206 Madgearu, pl. II and V. 207 Contra Istoria Bisericii Ortodoxe Romne vol. 1 (Buch de Misiune al Biseric ii Ortodoxe Romne, 1980), 166. 208 I. Engelhardt, Mission und Politik in Byzanz: ein Beitrag zur Strukturanalyse byzantinischer Mission zur Zeit Justins und Justinians (Mnchen: Institut fr Byzantinistik und Neugriechische Philologie der Universitt, 1974). 209 Curta, "Befor e Cyril and Methodius," 186 89. 210 Corpus Iuris Civi lis Novella XI, 3.94, ed.R. Schoell and W. Kroll, vol. 3 (Berlin: Weidmann, 1928), 94 Historians have also drawn attention to a later account of Theophanes Confessor mentioning the help offered by a Christian Gepid during a campaign north of the Danube i n 593. Since no Gepids were living

PAGE 168

168 eform meant to reshape the ecclesiastical hierarchies and administrative apparatus. 211 According to Novella XI the jurisdiction of the new archbishopric of Justiniana Prima extended north of the Danube to include the Byzantine bridge heads of Litterata and Recidiva. 212 Scholars have often used this account to suggest that the archbishop of Justiniana Prima was also charge d with the welfare of the Christian population living north of the Danube, well beyond the area controlled by the Danubian fortresses mentio ned in the official document. A similar role was envisaged for the bishoprics of Scythia whose authority would have extended beyond the Danube, deep into barbaricum 213 This can hardly be the case with such a highly unstable frontier as the Lower Danube Eve n if any such initiatives were present they were short lived and soon curtailed by the growing insecurity in the region toward the middle of the sixth century. The lack of any evidence of attempts to institutionalize Christianity north of the river, marked by the absence of churches, as noted, warrants a in north eastern Wallachia, it has been suggested that the Christian scout used by the Roman army actually belonged to a local Romanized community. While this is a plausible assumption, it is of little use for a broader u nderstanding of Christianity north of the Danube. See Theophanes, Cronographia a. 6085 (593), ed. C. de Boor (Leipzig: Teubner, 1883), 271; Teodor, 61. 211 V. Velkov, Cities in Thrace and Dacia in Late Antiquity (Amsterdam: Adolf M. Hakkert, 1977), ch. 2. 212 The location of Recidiva is still controversial with some scholars identifying it as Arcidava or even Sucidava, th e most important Byzantine settlement on the northern bank of the Danube, while more recently it has been located at Stari Dubovac. See D. Tudor, (Bucharest: Editura Academiei RSR, 1978), 466; Barnea, "Sur les rapports," 56; A. Madgearu, "Th e 6th Century Lower Danubian Bridgeheads: Location and Mission," EN 13 (2003): 297. For the location of Litterata (Lederata) Roman Limes on the Middle and Lower Danube de: Archaeol ogical Institute, 1996), 69 72. 213 Barnea, Les monuments palochrtiens Istoria Bisericii Ortodoxe Romne 165; D. G. Teodor, 60 61; N. Zugravu, secole le II VIII," Pontica 28 29 (1995 1996): 176 79; Protase, Autohtonii n Dacia 87; For a more skeptical interpretation, see Gudea and Ghiurco, 128; Madgearu, Rolul 80.

PAGE 169

169 high degree of skepticism as to the precise influence of Byzantine bishoprics in the regions beyond the Danube. Nevertheless, the presence of an unprecedented number of Christian objects still requires a m ore satisfying explanation. While their presence cannot be ascribed to a encourage imitatio in barbaricum and m umenical ambitions, which might have taken different forms. Justinian's frontier policy in the Balkans included not only the program of fortification, but also alliances with the Lombards, the Antes and the Utigurs. It is quite probable that Justinian's strategy of bringing the populati ons north of the Danube closer to the Empire included such cultural tactics. A recent inventory of Christian objects shows a great variety of items showing up north of the Danube even very far from the frontier. 214 Lamps, flasks, pectoral crosses, and molds for the production of crosses represent the most frequent and also the most interesting Christian objects imported or produced in barbaricum (Figure 4 14). former province of Dacia, many of which bear Christian symbols. The bronze liturgical lamps are part icularly interesting, as are some of the "Danubian and African clay lamps, most of them found in formerly Roman urban centers. Very few lamps have been found south and east of the Carpathians, while the often cited lamp from Luciu, close to the left bank of the Danube, is more probably part of the booty taken from one of the Roman fortresses in Scythia. Unfortunately, the scarcity of lamps coming from a clear stratigraphical context makes any religious interpretation purely impressionistic. 214 Madgearu, Rolul 120 27.

PAGE 170

170 Fig ure 4 14. Distribution of bronze liturgical lamps (square) clay lamps with cross shaped handle or cross/Chi Rho on discus/nozzle (circle), and Menas flasks (triangle). a) Bronze lamps:1. Luciu; 2. Tpigyrgye; 3. Ratiaria; 4. Dej; 5. Lipova; 6. Moigrad (Porolissu Scupi (Skopje); 11. Kalaja; 12. Nova Palanka; 13. Histria. b) Clay lamps:1. Sucidava ; Tropaeum Traiani (Adamclisi); 6. Novae (Svishtov); 7. Halmy ris; 8. Capidava; 9. Axiopolis (Cernavoda); 10. Pirdop; 11. Drobeta (Turnu Severin); 12. Callatis (Mangalia); 13. Dinogetia; 14. Ulmetum; 15. Dionysopolis (Balchik); Iulia (Apulum); 19. Turda (Romula); 21. Sirmium (Sremska Mitrovica); 22. Osijek; 23. Gyor; 24. Kekkut; 25. Budapest (Aquincum); 26. Vinkovci; 27. Poetovio c) Menas flasks:1. Moigrad (Porolissum); 2. Dierna

PAGE 171

171 A quick look on the map will reveal different patterns of distribution. I have already pointed to the large concentration of early Byzantine la mps in the territory of the from the Holy Land and Egypt, especially Menas flasks, seem to have been popular north of the Danube if one considers the finds from Moigrad (Porolissum), Dierna (Or ova), and Szombathely. 215 Menas flasks are very common in Italy and the Dalmatian coast, but also north of the Alps in the Rhine region, although the local provenance of some of the finds is not always certain. 216 The flasks from Szombathely, Moigrad, and perhaps that from Dierna may have arrived from the west, the Dalma tian coast being the most probable source, although a route from Constantinople cannot be completely ruled out. 217 Four Menas flasks have been found at Tomis and Capidava, but Scythia seems a less probable supplier of such items to barbaricum 218 Nevertheless, conspicuous differences in the pattern of distribution remain important for understanding the variety of preferences and fashions in barbaricum If Christian lamps are rarely found outside the former province of Dacia, pectoral cro sses are very often foun d in Wallachia and Moldavia (Figure 4 15). Most of them are Maltese 215 I. Barnea, "Menasampullen auf dem Gebiet Rum niens, in Akten des XII. Internationallen Kongresses fr Christlische Archologie ed. E. Dassmann, K. Thraede, and J. Engemann, vol. 1 (Mnster: Aschendorffsche Verlagsbuchhandlung, 1995), 509 14 and pl. 61; Z. Kadar, Die Menasampulle von Szombathely ," in Dassmann et al, ed. Akten 886 88; For the long distance circulation of Menas flasks see recently S. Bangert, "Menas Ampullae : A Case Study of Long Distance Contacts," in Incipient Globalization? : Long dist ance Contacts in the Sixth Century ed. A. Harris (Oxford: Archaeopress, 2007), 29 33. For a discussion about pilgrim routes to Europe based on ampullae see C. Lambert and P. P. Demeglio, "Ampolle devozionali ed itinerari di pellegrinaggio tra IV e VII se colo," Antiquit Tardive 2 (1994): 205 31. 216 P. Linscheid, "Untersuchungen zur Verbreitung von Menasampullen nrdlich der Alpen," in Akten 982 86; Demeglio, "Ampolle devozionali,"fig. 6. For the caveat, see Bangert, "Menas Ampullae," 29. 217 Curta, "Limes a nd Cross," 59, n. 136; Lambert and Demeglio, "Ampolle devozionali," 219; contra Madgearu, "Rolul," 52 53, who argued that the region north of the Danube had little or no connection with the West. 218 Contra Barnea, "Mena sampullen auf dem Gebiet," 514.

PAGE 172

172 Fig ure 4 15. Distribution of crosses (circle) and molds for the production of crosses (square). a) Crosses:1. Moigrad (Porolissum) ; Romni; 3. Davideni; (Izvoarele); 13. Golesh; 14. Bucharest; 15. Novi Banovci; 16. Ram; 17. Bezhanovo; 28. Odessos (Varna); 29. Callatis (Mangalia); 30. Histria; 31. Axiopolis (Cernavoda); 32. Beroe (Piatra Fr Izvoru Dulce; 7. Olteni; 8. Sucidava; 9. Sadovec.

PAGE 173

173 crosses Rugino asa, Valea Voievozilor, and in "Avaria at Balatonfzf Szalmssy and Szkkutas. 219 It seems that the parallel between the Middle Danube region, on one hand, and Moldavia and Wallachia, on the other which was made apparent by the study of hand made lamps is also valid for Maltese pectoral crosses, with the former province of Dacia following a different pattern. Unlike lamps, many crosses have been found in a clear archeological context, in association with other sixth century artifacts and sometimes with m olds used to produce such crosses. Molds for Maltese crosses Ol teni and they clearly point to local production of such items. 220 Unlike the genuine Byzantine imports found i n "Avaria," north of the Lower Danube the unaffordable Byzantine crosses were replaced by cheaper local imitations. However, these were mass produced as evidenced by the large number of molds, as well as actual crosses found north and south of the Danube. 221 How can we distinguish between imitatio imperii and genuine adherence to the Christian creed? Crosses were but one of many items and ornaments produced with molds found in Barbaricum while the totality of Byzantine imports finding their way north of the Danube does not indicate a special 219 For recent discussions, see F. Curta, "New Remarks on Christianity Beyond the Sixth and Early Seventh Century Frontier of the Empire," in Keszthely Fenkpuszta im Kontext sptantiker Kontinuittsforschung zwischen Noricum und Moesia ed. O. Heinrich Tamska ( Budapest/Leipzig: Marie Leidorf, 2011), 305 09; VII," AM 30 (2007): 129 36; F. Curta, Limes and Cross 58 62. Similar crosses are often found south of the D anube, see Curta, "Before Cyril and Methodius," 185 and fig. 8.5. There are many Avar period crosses from Hungary usually found in female graves, of which the lead crosses are the closest analogies for the ones found in Romania, see Garam, Funde byzantinis cher Herkunft 64, fig. 6 and pl. 40. 220 60. 221 For finds of crosses in the frontier region, see Curta, "Before Cyril and Methodius," fig. 8.1 8.8.

PAGE 174

174 preference for Christian objects. On the contrary, some of the common dress accessories from the Balkans which bear Christian symbols are rare or nonexistent among finds north of the Danube, the most important being the Sucidava buckles with cross shaped open work decoration, fibulae with bent stem with Christian inscriptions (Minna, Amin, Maria, etc.) and cast fibulae with bent stem with cross shaped foot. Christian symbols on local hand made pottery have also been conne cted with the religion of local communities in Barbaricum especially south and east of the Carpathians. 222 Recent attempts to classify the incisions made on local pots were met with mixed conclusions and a certain degree of skepticism regarding their relig ious significance 223 To be sure, crosses appear not only on traditional pots whose shape can be traced back even to the pre Roman period, but also on typically Slavic wares of the so called Praga Korchak typ e found in the sixth to seventh century settlement s from 224 Moreover, swastikas and other symbols appear on hand made pottery from Bucharest and Dulceanca in Wallachia, and they might have had a totally different cultural symbolism than the iconography associated with Christianity. 225 What is particularly important about this practice, which seems to be so common in Barbaricum is the fact that it occurs only on hand made pottery. Despite 222 For a recent inventory see Madgearu, 127 33, 135 7. 223 E. S. Teodor and I. Stanciu, "About Cr osses on Wet Clay as Cultural Markers," EN 19 (2009): 129 55; Madgearu, 83 4. Previous publications offered a more optimistic interpretation, for which see especially Teodor, 82 84. 224 Teodor, fig. 14/4 and 15/1, 4. 225 Dolinescu Ferche, fig. 70/1; Dolinescu Ferche and Constantiniu, "Un tablissment du VI e sicle," fig. 9/6.

PAGE 175

175 recent attempts, its cultural symbolism related to identity formation is far from b eing fully understood 226 The absence of churches, despite the existence of liturgical objects such as the lead ampulla from Ia or the more controversial bird headed distaff from Craiova, suggests that Christianity was adopted at the individual level and not in an institutionalized form. 227 Conversion to Christianity or adherence to at least some Christian traditions firmly in place since the fourth century was most likely not synonymous with a full adherence to Christian norms and precepts, since inhumation seems to have been practiced only on a small scale north of t he Danube. Few of the key liturgical artifacts invoked for the Christianization of barbaricum have a good stratigraphical context, while others were acquired from private collectors and their original provenance is seldom known. Crosses and molds for cross es are often found north of the Danube, most of the time in a good archaeological context, but the distinction between cultural imitatio and true belief is hard to discern Local elites north of the Danube may have very well expressed their status through 226 Teodor and Stanciu, "About crosses," 141. 227 SCIVA 32, n. 2 (1981): 299 302; Periam: Arheologie 26 (1983 h analogies from the Danubian frontier region. Bird headed distaffs, some probably used as dress pins are common in the northern Balkans, for instance at Novi Banovci, Popina, Prahovo, Mokranjska Stena, Donje Butorke, Kladovo, Iatrus, Sadovec, Goleche, and also in Avaria at Szekszrd Bogyiszli, for which Situla Podunavski pl. XVIII; Uenze, Die sptantiken pl. 6 1 5; G. Gomolka Fuchs, "Die Kleinfunde vom 4. bis 6. Jh. aus Iatrus," in Iatrus Krivina II. Ergebnisse des Ausgrabungen 1966 1973 (Berlin: Akademie Verlag, 1982), pl. 64/28 6; G. Atanasov, "De nouveau sur la localisation de la fortresse Bas byzantine St. Ciril en Scythie Mineure," in Prinos lui Petre Diaconu la 80 de ani ed. I. Cndea et al. (Br ), fig. 6; Garam, Funde byzantinischer Herkunft pl. 41/3. They have been often associated with the heresy of Bonus, although there is no consensus regarding their use and symbolism. A dress pin bearing the inscription BONOSA was found at Keszthely Fenkpuszta horreum (grave 5), for which see Ibid., pl. 41/3.

PAGE 176

176 common items such as bronze or lead crosses. 228 Settlement patterns north of the Danube have very little to offer in terms of spectacular residences, while gold and silver objects with or without religious symbolism, are extremely rare finds. 229 Indeed, some individuals or communities living in the shadow of the Empire might have decided to adopt the religion of Byzantium, but the decision was taken at a personal level and actual practices might have departed from the official canons, as they sometimes did sou th of the Danube. Objects with religious designation arrived north of the Danube, either as prestige objects for local elites or as Christian artifacts for believers, some of whom may have been Christian prisoners from the Empire, allowed to practie their religion while in captivity. Some of the items produced in provinces far from the Danube, such as the North African lamps or Egyptian ampullae could also show social and economic connections with regions such as Dalmatia or the northern Balkans or even a d irect access, through pilgrimage, in case of Menas flasks. Communities from the former province of Dacia had been exposed to Christianity since the fourth century and important foundations were already in place. The large number of Roman prisoners taken no rth of the Danube could have also facilitated cultural contribution is hard to assess. At any rate, Christianity was hardly a general phenomenon in sixth century barbaricum and it will take solid archaeological evidence to change this conviction 228 Pace M adgearu, "Semnif 229 Phase of the Early Middle Ages," AAC 38 (2003): 111 36.

PAGE 177

177 4.4 Conclusion: Contact and Separation on the Lower Danube Frontier The study of written and archaeological sources pertaining to the Lower Danube frontier during the long sixth century p resents an ostensible paradox. Often used as a rhetorical device by sixth century writers, the Danube was presented as the main line of defense which capable emperors were able to hold against barbarian tides. No attempts were made to conquer regions north of the river, except for efforts to re establish a number of bridge heads which had the strategic mission of act ing as a spearhead in barbaricum and perhaps facilitating trade. By contrast archaeologists have revealed a long chain of fortifications rebui lt in the first half of the sixth century, leaving the impression that the Empire had a clear priority in the Northern Balkans: to prevent a massive crossing of the Danube that could endanger Constantinople and to discourage small bands of raiders set on p lundering the frontier provinces. The role of the Danube fleet, recently re assessed by Florian Himmler, 230 adds a new layer to the strateg ic system devised by the Empire. The Danube was truly "the first line of def ense" in the words of Procopius neither m etaphorically, however nor as a natural barrier, but through the fleet monitoring the river. Indeed, very little room was left for cultural contact, or at least this is the daunting impression left by the multi layered system whereby the Empire tried to secure its frontier in the Northern Balkans. Despite such precautions, a wide variety of objects of early Byzantine origin found their way north of the Danube throughout the sixth century, a paradox which needs further explanation. Whether imports in the e conomic sense of the term or prestige items procured by the local aristocracy, the large quantity and diversity of Byzantine artifacts 230 F. Himmler, Untersuchungen zur schiffsgesttzen Grenzsicherung auf der sptantiken Donau (3. 6. Jh. n. Chr.). BAR Int. 2197 (Archaeopress: Oxford, 2011)

PAGE 178

178 testifies to the fact that the Danube did not prevent the crossing of objects and people. The Danube was by no means an Iron C urtain" as th e Empire had neither the streng th nor the energy to insure a clear separation from barbaricum. To be sure, allowing cultural contact with communities outside the frontier was a double edged sword. The ever increasing need for Byzantine i mports seen as prestige objects inevitably led to a series of plunder expeditions which exposed the weakness of the defense system devised by Justinian, whose theoretical virtues would soon be tested by invading Cutrigurs, Slavs, and Avars. What makes the Lower Danube a particularly interesting frontier is the fact that the lands north of the river do not fall under the standard definition of Barbaricum as much of the region had been a Roman province for more than 150 years. With all the controversy regar ding the level of Romanization and the demographic impact of the Roman retreat on the Danube in the last quarter of the third century, it is impossible to question the fact that the region was deeply permeated by Roman culture. Romanized communities sought to maintain a certain degree of Roman life, which explains the desire to acquire Ro man objects from the Empire. This provincial society was not the only one seeking Roman goods. The fourth century Chernyakhov culture, whose degree of homogeneity is impres sive given its presumed multi ethnic background, developed in the shadow of the Empire and the high priority given to commercial relations in treaties between Late Roman emperors and the Goths is a case in point. Moreover, o n the eve of the dissolution of their father's empire, Attila's sons decided that their chief priority was to convince Emperor Leo I to resume trad ing relations with

PAGE 179

179 the Huns, indirect evidence of the fact that access to Roman goods was crucial for securing and maintaining power in barba ricum. The Empire was slow to recover from the dark days of the fifth century but with the reign of Anastasius we witness the first major attempt to re establish a firm imperial control on the Lower Danube. Facing a much more divided political landscape n orth of the river, the E mpire lacked a partner for negot iations at least until the arrival of the Avars, which may in part explain the lack of information regarding the regulation of trade on the Danube frontier during the sixth century. Out of this histor ical background, leading back to the abandon ment of Dacia by Emperor Aurelian, emerged the communities which would be the main recipients of Byzantine goods in the sixth century. Some of them lived in the former province of Dacia, now mostly under the cont rol of the Gepids whose political relations with Constantinople remained precarious, while others were located south and east of the Carpathians, a buffer zone between the Empire to the south, Gepids and later Avars to the west, and Slavs, Antes, and Cutri gurs to the east. This region was an important reservoir of manpower and potential recruits for manning the garrisons defending the fortresses of the Lower Danube rebuilt by Justinian. In addition, the advice given by the author of the Strategikon repeated ly refers to grain and food supplies as one of the major resources to be taken from the lands north of the Danube. There was undoubtedly a mutual interest in maintaining cultural contact. T he presence of Byzantine goods i n this large region between the Mid dle Danube and the steppes north of the Black Sea reflects both the desire to maintain a life style that depended heavily on emulating Roman ways and the political developments in the region, with the Avars slowly recreating the scenario of the fifth

PAGE 180

180 centu ry by draining the Empire's wealth towards Pannonia in exchange for an unsettled peace. Influenced by the discourse of sixth century writers, historians have often viewed this frontier region in terms of a long struggle between "Romans" and "barbarians" wi th different names but similar agendas. In reality, the frontier provinces of the Balkans had been a multi ethnic envirnoment long before the arrival of the Slavs and Avars in the sixth century. The shared Roman culture did not lead to a complete assimilat ion through the process of Romanization. Various traditions going back to the pre Roman period could still be traced on sixth century hand made pottery found in the Lower Danube provinces. Especially after the 530s when the Empire found itself engaged in w ars on multiple fronts, the defense of the Balkans witnessed a great shortage of manpower, which Justinian tried to overcome by forging alliances with various groups, such as the Antes or the Utigurs, or by manipulating conflicts within barbaricum The eas iest way to become "Roman" was to serve in the garrisons stationed o n the frontier. Here much of Roman identity revolved around wearing a certain military costume and living in a fortress where life depended on the arrival of the state controlled annona wine and oil received in standardized containers, mostly amphorae known as LR1 and LR2. These are precisely the artifacts most often found in the lands north of the Danube: fragments of amphorae, oil lamps, and dress accessories associated with the militar y environment of the Lower Danube frontier ( Figure 4 16). Being Roman on the sixth century Danube front ier was a fluid concept, a negot iated identity whose main coordinates revolved around serving in Roman garrisons, adopting the provincial life style and perhaps adhering to Christianity.

PAGE 181

181 While some of these items may have been purchased directly from one of the trade centers on the Lower Danube, the most probable candidates being the Byzantine bridge heads on the left bank of the Danube, many Byzantine ite ms reflect ed the circulation of people and the specific tastes of each area. Historians have tended to lump together regions from barbaricum conveying a false sense of homogeneity in their contacts with Byzantium. In reality, their access to various catego ries of goods from the E mpire depended on a series of conditions, such as distance from the frontier the availability of certain goods in specific areas of the frontier, itself lacking a perfect homogeneity or standardization, and of course the social fab ric of communities in Barbaricum which might have placed different values on artifacts of Roman origin. In light of such preconditions, it comes as no surprise that amphorae are commonly found in the close proximity of the Danube, north of Sucidava, the main Byzantine possesion on the left bank of the river. However, no Byzantine fibulae or buckles are found in this region, although they were readily available from Sucidava or from any one of the Byzantine forts in the Iron Gates area. The only possible c onclusion is that communities in Oltenia had no interest in expressing their identity in such manner or adopting fashions connected to the military hilltop sites on the Lower Danube. By contrast amphorae are often found in Moldavia, some 150 miles away fr om Dinogetia, the closest Byzantine fortress, which implies that communities east of the Carpathians placed a great value on gaining access to wine and oil, the main products transported in amphorae. The same communities sought to procure Byzantine dress a ccessories such as fibulae with bent stem and cast fibulae with bent stem, as well as

PAGE 182

182 Fig ure 4 16. Main categories of Byzantine artifacts found in barbaricum : amphorae (circle), clay lamps (square), fibulae (triangle), buckles (star), Christian objects (cross).

PAGE 183

183 Byzantine buckles of the Sucidava class. This region permeated by Slavic culture in the second half of the sixth century emulated the early Byzantine provincial way of life more than any other region in the lands north of the Lower Danube. Many of the mercenaries recruited by Byzantium to defend the fortresses of the province of Scythia, whose presence can be seen in the hand made pottery found there, may have originated in the communities from Moldavia. Such mercenaries brought home Roman fashi ons mostly related to a military environment, typical for the Lower Danube region, and increased the demand for Roman goods and the social value placed on such artifacts. Some of them came to be produced locally as testified by the significant number of mo lds and metallurgical implements found in the region, including molds used to produce crosses, a possible sign that Christianity was important for creating a social identity based on Roman values. Between Oltenia, the land closest to the Lower Danube, and Moldavia lies a region of mixed cultural influences best seen in the local ceramics briefly discussed at the beginning of this chapter. Interestingly, there are no known settlements closer than twenty miles from the Danube. The existence of a buffer zone i n the Danube valley could be ascribed to the Empire's policy to discourage frequent crossings of the Danube or the creation of power centers too close to the military frontier. The cultural profile of the area is dominated by the controversy regarding the role of the Slavs, the great Monteoru, and the mosaic of settlements in Bucharest which produced diverse ceramic shapes and settlement patterns, but also an important number of Byzantine imports betraying close ties to the Byzantine wo rld. Almost all categories of Byzantine artifacts can be found in those settlements evi ncing an easy

PAGE 184

184 access to goods, which can be explained by the ir relative proximity to the Danube ( ca 50 miles). The metallurgical activity of the region is also impressi ve with no less than six settlements in Bucharest yielding metallurgical tools; mean while several settlements on we re also notable for their metallurgical activity. Yet another region with strong local particularities is the territory of the former province of Dacia, espec ially Transylvania. Here, the stronger Roman tradition going back to th e days of the Roman occupation wa s coupled with another defining trait, a stronger domination of the Hunnic confederation in the fifth century and of the Gepids and Avars in the sixth. Romanian historians in particular have insisted on the great transformations produced by the Huns whose domination o f the Middle Danube brought an end to cultural contacts of Dacia with the western Roman world and eventually led to a reorientation toward Constantinople. However, the break with the West was a much longer process connected with developments in the western provinces during the first half of the fifth century, a scenario in which the Huns were but one episode, albeit the most notorious. In fa ct, a reorientation toward Constantinople could not have happened during the Hunnic domination with chunks of the western Balkans under the direct control of Attila. Once the Hunnic storm passed the former province of Dacia resumed its cultural contacts w ith the West, favored by the reconquest of Italy by Justinian. Some of the items virtually unknown in other parts of barbaricum discussed here, such as bronze lamps, Menas flasks, genuine North African lamps may have arrived in Dacia through the mediation of the Adriatic ports. Crosses or molds for the production of crosses, so often found south and east of the Carpathians are absent from the former

PAGE 185

185 province of Dacia. Common Byzantine dress accessories appear in Gepid and early Avar contexts and generally f ollow cultural trends from the Middle Danube and Tisza valleys, where pieces of military equipment of Byzantine origin are found in association with artifacts typical for Central Europe or following fashions from the eastern corners of the Avar khaganate, whose influence extended to the lands north of the Black S ea. Transylvania clearly followed a d ifferent pattern compared to Wallachia or Moldavia S ometimes cemeteries excavated in this region offer a glimpse of the wealth of the early Avar center in Panno nia Grave inventories combine Byzantine and Eastern elements, an exquisite taste which the Avar aristocracy was able to afford after the Empire sent an estimated total of ca. 80.000 pounds of gold in the form of tribute. The movement of people and perhaps the activity of Byzantine merchants north of the river insured a steady flow of Byzantine items in barbaricum. Such items had an important social value in the lands north of the Lower Danube, but it is important to note that not all categories of goods ty pical for the frontie r settlements have been found in barbaricum. Roman glass, an accessory of the military elite, was perhaps too fragile to be transported north of the Danube. Moreover various types of jewelry and dress accesories are missing in Barbar icum which seems to suggest that communities beyond the frontier received only a number of basic items, lead and copper crosses instead of silver, buckles and fibulae made of iron or copper alloy, and the most common ceramic containers in which soldiers r eceived their rations. Oil lamps were not so easy to come by, although they are found in abundance in frontier fortresses. Sometimes communities from Barbaricum except the ones from the former province of Dacia, had no alternative but to manufacture crude hand made imitations. Silve r and

PAGE 186

186 gold objects are extremel y rare and the ones recorded do not have a clear archaeological context and might constitute the result of plunder in the Balkans. To be sure, precious metal objects are also scarce in the frontier provinces, where Justinian introduced a special administrative unit, quaestura exercitus in a desperate attempt to feed the northern Balkans, which housed many garrisons but possessed few resources. Many of the Byzantine imports found in barbaricum are therefore nothing but a fragmentary reflection of the annona militaris which took on a different meaning after crossing the frontier. The existence of economic exchange on the frontier and beyond cannot be ruled out, but the Danube as a contact zone presu pposed a more important set of relations, especially those of a non economic nature. The movement of objects documented by archaeological means nicely dovetails with sixth century accounts documenting the movement of people, Herul i H uns," Slavs and Antes recruited in the Roman army and large numbers of prisoners taken from the provinces of the Balkans into captivity north of the Danube. This two way street of Romans forcibly taken to barbaricum and "barbarians" serving in the Roman army and presumably in frontier garrisons produced a largely non economic kind of cultural contact, whose material remains are the Roman objects found in Barbaricum the non Roman objects found in frontier provinces, as well as the multi cultural influences traceable on several categories of objects discussed in this chapter. The frontier region understood as the Danube with its military installations, supporting provinces, and lands north of the river was by no means homogeneous. Basic items of little value in the frontier fort resses took on a special social significance once they crossed into barbaricum Settlement patterns are in sharp contrast, the

PAGE 187

187 fortified hilltop sites of the Balkans being replaced by defenseless open settlements in the lands north of the river. The cerami c production in barbaricum did not follow contemporary shapes from the Empire. Instead, it relied on traditional pots, some of them going back to the Roman provincial types of the first centuries AD. Christianity was present north of the Danube but in modi fied forms, which depart ed from the traditional canons regarding burial rites and rituals and the practice in churches so far missing in barbaricum. All these convey an image of a group of societies living in the shadow of the Empire but adopting only cer tain cultural practices, a decision shaped by the political situat ion of the region and the degre e to which the Danube could function both as a military frontier and as a cultural interface between the Empire and the world beyond. Unfortunatel y none of th e categories of imports studied in this chapter is chronologically sensitive as to allow for a step by step analysis of the evolution of cultural contact in the frontier region. Coins are the only type of Byzantine artifact which can offer a precise dating being inscribed with the regnal year of the ruling emperor. Coins are also the most frequently found artifact of Byzantine origin in barbaricum and can answer multiple questions regarding their economic role in facilitating cultural contact between Byzan tium and the lands north of the Danube, as well as their non economic functions in barbaricum. However, a brief overview of the monetary economy of the early Byzantine Empire is needed before attempting to understand the role of coins outside the empire an d this will be the next topic of our discussion.

PAGE 188

188 Fig ure 4 17. Early Byzantine lamps and imitations: Danubian lamps (1 5 ), Anatolian lamps (6 7), North African lamp (8), imitation after North African lamp (9), Palestinian lamp (10), hand made lamps (11 12).

PAGE 189

189 Figure 4 18. Sixth to seventh century molds.

PAGE 190

190 Figure 4 19. Sixth to seventh century fibulae: fibulae with bent stem (1 4), cast fibulae with bent stem (5 8), bow fibulae (9 12).

PAGE 191

191 Figure 4 20. Byzantine artifacts in barbaricum and parallels fr om the Empire: Coptic bronze lamps (1, 2); North African (3, 4), Danubian (5, 6) and Palestinian (7, 8) clay lamps; fibulae with bent stem (9, 10), bow fibulae (11, 12), cast fibulae with bent stem (13, 14), molds (15, 16, 21, 22); Sucidava buckles (17, 18 ); Ppa buckles (19, 20).

PAGE 192

192 CHAPTER 5 CENTER AND PERIPHERY: COIN CIRCULATION IN EARLY BYZANTIUM 5.1 The Role of Money in the Early Byzantine Economy The nature of exchange in ancient societies is one of the most time honored and scholarly fruitful debates of the past century. Although the discussion initially focused on the classical period, students of the emergent fields of Late Antiquity and Early Byzantium made the debate their own as they were trying to shed new light on the role of trade and to determin e the prevalent form of exchange in the Mediterranean world from the fourth to the seventh century. It has become de rigueur to introduce the topic by referring to the now classical opposition between two major schools of thought, the neoclassical and the institutional in academic economics, the substantivist and the formalist in anthropology, the primitivist vs. modernist in history (or minimalist vs. maximalist ) and the way in which it shaped various humanistic disciplines, including history, arc haeology and anthropology. The debate itself has lost momentum in the past decades and it is doubtful that a strictly dichotomous discourse is going to further our understanding in any dramatic way. 1 The growing archaeological record and with it, its weigh t in the general narrative about Late Antiquity, has no doubt influenced the way in which students of the period chose to handle theoretical and methodological models. However, a brief introduction into the larger historiographic al issues surrounding the 1 For an early critique, see G. Dupr and P. P. Rey, "Reflections on the Pertinence of a Theory of the History of Exchange," in The Articulation of Modes of Production ed. H. Wolpe (London: Routledge, 1980), 128 60. In anthropology the debate was declared dead by Barry Isaac at the beginning of the 1990s but it managed to cast a much longer shadow, see B. L. Isaac, "Retrospective on the Formalist Substantivist Debate," i n Research in Economic Anthropology 14 ed. B. L. Isaac (London/Greenwhich CT: Jai Press Inc., 1993), 231, n.1: "At present, the only anthropologists who still try to address their work to that old debate are the few archaeologists who have not looked at t he subfield since graduate student coordinates have been reshaped; for a recent review, with applications to the Roman economy as well, see M. E. Smith, "The Archaeology of Ancient State Economies," Annual Review of Anthropology 33 (2004): 73 102.

PAGE 193

193 n ature of exchange and the role of the market with its monetary instruments is needed in order to situate current theories regarding the role of money in Early Byzantium in their proper scholarly context. Although the scholarly debate most often referred t o is essentially a product of the past century, sociological and anthropological theories of exchange have their roots as far back as the eighteenth century, Adam Smith being one of the central figures. The so called substantivist approach of the post war period is often attributed to Karl Polanyi who articulated the theory as well as its opposition to the formalist stance, but his money in pre capitalist societies. 2 Essent ially, and this is where the disciplines of history and anthropology converge, substantivists claim that an economy dominated by concepts such as supply, demand, and profit applies to formal economics and cannot be projected on pre capitalist societies in which social conditions qualified what by modern standards seem like rational decisions. Social motivations (and obligations) rather than economic self interest defined the pre modern world. Although the historical debate surrounding the nature of ancient economies had been around since the days of Karl Bcher and Eduard Meyer in the late nineteenth century 3 substantivist ideas 2 K. Polanyi, "The Economy as Instituted Process," in The Sociology of Economic Life ed. M. Granovetter and R. Swedberg (Boulder, CO: Westview Press 1957), 243 70; K. Polanyi, "The Semantics of Money Uses," in Primitive, Archaic, and Modern Economies: Essays of Karl Polanyi ed. G. Dalton (Garden City, NY: Doubleday, 1968), 175 203; G. Dalton, "Economic Theory and Primitive Society," AA 63, no. 1 (1961): 1 25; M. Sah lins, "On the Sociology of Primitive Exchange," in The Relevance of Models for Social Anthropology ed. M. Banton (London: Tavistock Publications, 1965), 139 236. For the influence on Roman monetary history, see M. de Cecco, "Monetary Theory and Roman Hist ory," Journal of Economic History 45, no. 4 (1985): 809 22; Banaji, Agrarian Change in Late Antiquity: Gold, Labour and Aristocratic Dominance (Oxford/New York: Oxford University Press, 2001), ch. 2; F. Carl, L'oro nella tarda antichit: aspetti economici e sociali (Torino: Silvio Zamorani, 2 009), 9 25 with the literature. 3 M. I. Finley, ed., The Bcher Meyer Controversy: An Original Anthology (New York: Arno Press, 1979). See more recently H. Schneider, "Die Bcher Meyer Kontroverse," in Eduard Meyer: Le ben und Leistung eines Universalhistorikers ed. W. M. Calder III and A. Demandt (New York: E.J. Brill, 1990), 417 55.

PAGE 194

194 floating in several humanistic disciplines were soon picked up by historians of the classical world, the most influential voice be ing that of a series of remarkable Ancient Economy is perhaps the most radical embodiment of the conviction that the overwhelmingly agrarian ancient economy cannot be treated by employing modern tools. 4 His disdain for archaeology and its potential contribution led to a sometimes cynical view of the economic possibilities of ancient states dominated by war and imperialism rather than by concerns for developing a market economy. By extension, coinage was seen as strictly non fiduciary as states had no coherent monetary policies, while minting was never the result of economic pressure from the market. As so often happens when borrowing theoretical models from neighboring disciplines, historians ended up by throwing out the baby with the bathwater and many Roman historians jumped in the primitivist boat reshaped by Finley They sometim es ignored the fact that the Roman economy had reached a far greater complexity than the societies usually studied by anthropologists or the ones reflected in literary sources mirroring an aristocratic perspective. 5 The application of such views on Late Antiquity had already been developed by A. H. M. Jones whose monumental work on the social and economic conditions of the 4 M. I. Finley, The Ancient Economy ( London, Chatto & Windus, 1973). 5 See however the critical review article of M. W. Frederiksen, "The ory, Evidence and the Ancient Economy," JRS 65 (1975): 165 71. Much more recently and based on archaeological evidence largely unavailable to Finley, see K. Greene, "Technological Innovation and Economic Progress in the Ancient World: M. I. Finley Reconsid ered," Economic History Review 53, no. 1 (2000): 29 59. Another problem was the fact that historians drew inspiration mostly from the very early stage of the substantivist/formalist debate in anthropology, when the major theoretical propositions were being articulated, without carefully following subsequent developments. As early as the 1960s it became obvious that a purely substantivist position had much less applicability than Polanyi envisaged a decade earlier (below, n. 9). Eventually, historians of the ancient world would acknowledge this fact independently after taking stock of the growing archaeological evidence which revealed a much more diverse economic landscape. This is, however, a recent development of the last decade (below, n. 16).

PAGE 195

195 Late Roman Empire cam e out of the same intellectual mo ld. A similar distrust of material evidence sealed the fate of the Late Roman economy, whose trajectory was seen as depending almost exclusively on state dirigism. The influence of Finley and Jones on subsequent generations of English language historians cannot be overestimated. This is also reflected in the treatment of ancient coinages and their exact function in the political and economic mechanism of ancient states. The role of money was of course one of the main topics of discussion, if not altogether the key to establishing the nature of ancient exchange patterns. Although early coinages featured as the third type of exchange in the hierarchy developed by Polanyi and Dalton, their all purpose function was not recognized by anthropologists of the substantivist school, but merely their role as a series of specialized instruments of exchange. 6 Plenty of ethnographic evidence of redistribution, reciprocity and gift exchange put together by a series of distinguished anthropol ogists and s ociologists going back to Marcel Mauss e nsured that the primitivist view of instruments of exchange prevailed over the anachronistic use of t erms such as all purpose money. Referring to the Roman monetary system, historians and numismatists who drew on such substantivist theories rejected the existence of all purpose money based on the failure of the Roman economy to display all the features recognizable in its modern oinage found its literary sources pertaining to the monetary economy of the Late Roman and Byzantine 6 For a deta iled discussion and its application to an early Medieval context, see R. Hodges, Primitive and Peasant Markets ( New York: Blackwell, 1988).

PAGE 196

196 Empire. 7 Largely ignoring the physical evidence of excavation finds and h oards, He argued that t he use of money was strictly related to the needs of the state. The prima ry mechanism behind the distribution of coin was the military and the public post, while most of the energies gravitated around the needs of Constantinople. The circulation of coin was superficial, Hendy going as far as to say that coinage was relatively l ittle used. 8 Indeed, historians who placed greater emphasis on the archaeological evidence tended to have a less monolithic approach to ancient economies seen as being governed by a series of dynamic forces often deriving from state controlled activity, b ut allowing for private initiative and a more or less developed private sector. This time, however, such developments in the treatment of the Roman and Late Antique economy were less the product of influence coming from other disciplines and had more to do with the growing body of evidence result ing from systematic excavations. To be sure, in anthropology the Polanyist school of economic thought, which came as a much needed reaction to modernist interpretations no doubt influenced by the first age of global ization in the early twentieth century, was vigorously criticized for its rigidity and doctrinaire nature by the new generation of formalists. Drawing on the work of early formalists such as Raymond Firth, D. M. Goodfellow, and Melville Herskovits, formali sts of the second 7 M. Crawford, "Money and Exchange in the Roman World," JRS 60 (1970): 40 48; Hendy, Studies in the Byzantine Monet ary Economy c. 300 1450 ( Cambridge: Cambridge University Press, 1985) 8 Hendy, Studies in the Byzantine Monetary Economy C. Morrisson, "Review of M. F. Hendy, The Economy, Fiscal Administration and Co inage of Byzantium (Northampton: Variorum Reprints, 1989)," RN 33 (1991): 307 10.

PAGE 197

197 half of the twentieth century emphasized the need to test the usefulness of formal economics tools based on the specificities of the socio cultural entities studied, but maintained the position that economization and maximization represen t a universal type of behavior. 9 The formalist version of ancient economies developed by historians of the ancient world is perhaps less modernist than often assumed. 10 Although Mikhail Rostovtzeff, one of the first and most enthusiastic advocates of an ancient economy mirroring many of the principles encountered in modern economic thinking, was perhaps too quick to find sophisticated mechanisms behind ancient economic decision making his approach clearly opened the way to a more nuanced enquiry whic h relied heavily on the archaeological evidence. 11 However, few students of the modernist school ever intended to project purely capitalist concepts on ancie nt economic realities. The ever growing archaeological record has improved our understanding of sh ort and long distance circulation of goods, although it is not yet clear where one draws the line between state controlled shipments, circulation between estates and commerce properly speaking. 12 Nevertheless, even primitivist historians of the Cambridge school 9 The literature is abundant and permeates economic anthropology at all levels. For early and influential reactions, see E. E. LeClair Jr., "Economic Theory and Economic Anth ropology," AA 64, no. 6 (1962): 1179 203; Economic Anthropology," AA 68, no. 2 (April 1966): 323 45; R. Burling, "Maximization Theories and the Study of Economic Anth ropology," in Economic Anthropology: Readings in Theory and Analysis ed. E. E. LeClair and H. K. Schneider (New York: Holt, Rinehart and Winston, Inc., 1 968), 168 87. 10 To be sure late nineteenth century historian Eduard Meyer was equating fifth century B C Greece with sixteenth century Western Europe, for which see I. Morris et al., "Introduction," in Cambridge Economic History of the Greco Roman World ed. W. Scheidel et al. ( Cambridge : Cambridge University Press, 2008), 2. 11 M. Rostovtzeff, The Social and Economic History of the Roman Empire 2 nd ed. ( Oxford, Clarendon Press, 1957). 12 There is an abundant literature on the topi c of trade and commerce in Late Antiquity. For early influential contributions which also take into account the archaeological evidence, see C. R. Whittaker,

PAGE 198

198 could no longer accept the static model proposed by Finley and were forced to work within a more flexible paradigm, to be sure still essentially part of the same orthodoxy. 13 The contributions of Keith Hopkins are perhaps the best exemplification of such developments. 14 Later studies of the use of credit and banking in the Roman world, the general contribution of archaeology, as well as attacks on the consumer city theory, one of the major pillars of the Finleyan economic model, channeled the discus sion back to the idea of a developed pre capitalist economy in the Roman Mediterranean world. 15 Indeed, any interpretation depends on a series of prior assumptions and this is very much true in the case of coin finds and their treatment as evidence of eithe r market controlled circulation or state directed payments. It is however significant to note that a recent Economic History of Byzantium a massive collective enterprise bringing together contributions by archaeologists, historians, and numismatists, test ifies that "Late Roman Trade and Traders," in Trade in the Ancient Economy ed. P. Garnsey et al. (London: Ho garth Pres, 1983), 163 80; C. Panella, "Gli scambi nel Mediterraneo Occidentale dal IV al VII secolo dal punto di vista di alcune merci ," in Tome I, IV e VII e sicle, ed. C. Morrisson and J. Lefort (Paris: Le thielleux, 1989), 129 41; C. Abadie Reynal, "Cramique et commerce dans le basin gen du IV e au VII e sicle," in Hommes et richesses 143 59. 13 Hopkins argued that the views of Jones and Finley amounted to a "new orthodoxy" in the field of ancient econo mic history; see K. Hopkins, "Introduction," in Trade in the Ancient Economy xi. 14 See especially his economic model based on taxation, K. Hopkins, "Taxes and Trade in the Roman Empire," JRS 70 (1980): 101 25. The search for suitable explanatory models h as led historians in various literature as well, only to conclude that its applicability to the Roman economy is limited. See. G. Woolf, "World Systems Ana lysis and the Roman Empire," JRA 3 (1990): 44 58. For the existence of a Roman world economy, see also A. Carandini, "Il mondo della tarda antichit visto attraverso le merci," in Societ romana e impero tardoantico. Volume terzo: Le merci, gli insediament i ed. A. Giardi na (Bari: Laterza, 1986), 3 19. 15 For banking, see especially J. Andreau, Banking and Business in the Roman World (New York: Cambridge University Press, 1999). For an early book length study on the role of archaeology in pushing Roman econo mic history in new, and essentially non Finleyan, directions, see K. Greene, The Archaeology of the Roman Economy (Berkeley/ Los Angeles: University of California Press, 1986). For the "producer city," see the essays in H. Parkins, ed., Roman Urbanism Beyo nd the Consumer City (New York : Routledge, 1997) and D. J. Mattingly and J. Salmon, ed., Economies Beyond Agriculture in the Classical World (London/New York: Routledge, 2001).

PAGE 199

199 trade, finances and coin circulation can no longer be forced to fit the primitivist/ modernist model but a much more complex paradigm, yet to be fully articulated, which draws its inspiration from both. 16 Without witnessing a revolutionary turn in the historiography of the topic, as many primitivist claims are hard if not impossible to refute with the current evidence, 17 a shift towards a more flexible interpretation of the Roman economy is evident in recent scholarship. 18 Economic historians and archaeologists are now much more comfortable 16 Laiou, ed., Economic is part of a larger shift, which also includes the rec ent Cambridge Economic History of the Greco Roman World Framing the Early Middle Ages as well as several edited volumes such as S. Kingsley and M. Decker, Economy and Exchange in t he East Mediterranean during Late Antiquity (Oxford: Oxbow, 2001); M. Mundell Mango, Byzantine trade, 4th 12th centurie : the archaeology of local, regional and international exchange (Burlington, VT: Ashgate, 2009) and most recently C. Morrisson, ed., Tra de and Markets in Byzantium (Washington DC: Dumbarton Oaks, 2012), with the potential of channeling the discussion in new directions in the next decades For a thoughtful discussion of present concerns, see also I. Morris and J. G. Manning, "Introduction," in The Ancient Economy. Evidence and Models ed. J. G. Manning and I. Morris (Stanford: Stanford University Press, 2005), 1 44; H. J. Drexhage et al., "Die Wirtschaft der rmischen Kaiserzeit in der modernen Deutung: Einige berlegungen," in Die konomie des Imperium Romanum: Strukturen, Modelle und Wertungen im Spannungsfeld von Modernismus und Neoprimitivismus ed. K. Strobel (St. Katharinen: Scripta Mercaturae Verlag, 2002), 1 66. The complementary nature of the two seemingly irreconcilable positions wa s already emphasized a few decades ago by anthropologists, for which see J. H. Dowling, "The Goodfellows vs. the Dalton Gang: The Assumptions of Economic Anthropology," Journal of Anthropological Research 35, no.3 (1979): 292 308; C. Smith, "Regional Econo mic Systems: Linking Geographical Models and Socioeconomic Problems," in Regional Analysis. Volume I: Economic Systems ed. C. A. Smith (New York: Academic Press, 1976), 44, and even earlier by F. Cancian, "Maximization as a Norm, Strategy and Theory," in Economic Anthropology: Readings in Theory and Analysis ed. E. LeClair Jr. and H. K. Schneider (New York: Holt, Rinehart and Winston Inc., 1968), 228 33. Economists are sometimes less optimistic about this fusion of ideas; see for example J. A. Elardo, "Re formulating the Debate between the Substantivists and Formalists in Economic Anthropology: Is the Neoclassical Model Suitable for Describing Conditions in Nonmarket Economies?" Ph.D. Diss., University of Utah 2003, 92. 17 For a recent defense of the Finley an approach, see P. F. Bang, The Roman Bazaar. A Comparative Study of Trade and Market in a Tributary Empire (New York: Cambridge University Press, 2008), esp. 17 36 for the theoretical underpinnings. 18 Many refer to these new developments as a more "optim istic" view of the Roman economy but I have deliberately avoided the term for being too charged with a nineteenth century evolutionist flavor. The turn itself, however, is very real. Economist Peter Temin went as far as to say that the early Roman economy can be described as an integrated market economy, and many of his arguments are compelling, see P. Temin, "A Market Economy in the Early Roman Empire," JRS 91 (2001): 169 81. For recent views on trade and commerce in Late Antiquity, see also J. Durliat, "L es conditions du commerce au VI e sicle," in The Sixth Century: Production, Distribution and Demand eds R. Hodges and W. Bowden (Leiden: Brill, 1998), 89 117; B. Ward Perkins, "Specialized Production and Exchange," in The Cambridge Ancient History. Volume XIV: Late Antiquity: Empire and Successors, A.D. 425 600 ed. A. Cameron et al. (New

PAGE 200

200 speaking about Roman and Byzantine markets than they were two decades ago. A shift toward a modernist vision of the Mediterranean economy is evident in recent scholarship, without entailing a return to the trad itional approach of Rostovtzeff. In fact, the most important outcome of the long lasting debate in economic history has been the adoption of a moderate interpretative framework that legitimizes the discussion of markets and interdependent markets with re spect to the Late Antique economy, but still successfully avoid s the trap of presentism. 19 The re conceptualization of the modernist approach can be briefly characterized as an attack on the overwhelming dominance of the state in economic affairs, an added emphasis on private initiative based on the archaeological research of the past few decades, and a much more critical assessment of ancient sources displaying the senatorial disdain for trade and profit making which conceal and obscure the diverse conditio ns of the marketplace. To be sure, numismatists who are generally more concerned with the physical evidence than with theoretical models drawn from anthropology or sociology have always tended to feel more comfortable when treating coin finds as money rath er than exploring its limitations or even non economic functions. The adoption of a sophisticated statistical apparatus in the past three decades which brought order into an ever growing body of evidence from site finds and hoards, as well attempts to calc ulate ancient coin production added a scientific flavor to numismatic studies, bringing them even closer to York: Cambridge University Press, 2000), 346 91; M. McCormick, Origins of the European Economy. Communications and Commerce AD 300 900 (Cambridge: Cambridge University Pr ess, 2001), esp. part I; C. Morrisson and J. P. Sodini, "The Sixth Century Economy," In The Economic History of Byzantium 171 220; S. T. Loseby, "The Mediterranean Economy," in The New Cambridge Medieval History. Volume 1, c. 500 c. 700 ed. P. Fouracre ( Cambridge: Cambridge University Press, 2005), 605 38. 19 See recently J. M. Carri, "Were Late Roman and Byzantine Economies Market Economies?," in Trade and Markets in Byzantium 13 26.

PAGE 201

201 the tools of formal economics. 20 No doubt the early Byzantine monetary system had a great deal of sophistication, but even a cursory survey of the ac tual conditions of the sixth and seventh centuries will immediately reveal that advocating the existence of a monetary policy in the modern sense of the term would be totally anachronistic. On the other hand, early Byzantine administrators were building on a very long monet ary history and relied on a complex system inherited from the Roman Empire. Many of the regions under Byzantine control in the sixth century boasted an uninterrupted tradition of handling coins which spanned over a millennium of Greek, He llenistic, and Roman monetary history. It has been recently suggested that the level of monetization reached by the Byzantine economy was around 46 percent, which means that coinage permeated the early Byzantine society at multiple levels. This did not nec essarily reflect market transactions as the same author advanced a figure between 42 and 57 percent for the quantity of coin represented by taxes alone. 21 It immediately becomes evident that taxation (as the main form of revenue) and administrative and mili tary salaries (as the main form of disbursing coin) constituted both the main reason for issuing coinage and the most important channels of circulation in terms of the scale and value involved. It comes as no surprise that numismatists and economic histori ans have devoted much of their energy to aspects related to state finance and the ways in which ancient sources can help us understand Early Byzantine fiscal practices. 22 On the other hand, 20 See especially the essays in PACT ; W. Hahn and W. Metcalf, ed., Stud ies in Early Byzantine Gold Coinage (New York: American Numismatic Society, 1988); T. V. Buttrey, "Calculating Ancient Coin Production: Facts and Fantasies, NC 153: (1993): 335 51; F. de Callata, Calculating Ancient Coin Production: Seeking a Balance," N C 155 (1995): 289 311. 21 Morris son, "Byzantine Money," 949 50. 22 The literature is, again, very abundant. Several contributions from different intellectual traditions must be cited for their comprehensive treatment of the topic: I. E. Karayannopoulos, Das Finanzwesen des

PAGE 202

202 differentiating between the reasons why states issue coinage in the first place and why and how coinage actually circulates is a crucial methodological distinction which must be kept in mind at all times 23 Traditionally early Byzantine coinage has its starting point in 498 with the reform of Anastasius, which was essentia lly a successful overhaul of the bronze currency. The new system, however, inherited the stable gold coinage whose main features had been introduced by Diocletian and Constantine the Great two centuries earlier. Its resilience derived both from the fact th at its intrinsic value was equal to its nominal value and from a careful regulation of its production and circulation. 24 The Anastasian reform of the bronze coinage and its subsequent evolution in the sixth and seventh centuries are of crucial importance fo r a proper understanding of the function performed by low value currency in the early Byzantine financial and economic system. 25 Although the petty currency accounted for a small fraction of the total value of coins in circulation, dominated by the gold sol idus it bears upon the larger debate regarding the fiduciary frhbyzantinischen Staates (Mnchen: R. Oldenbourg, 1958); Hendy, Studies in the Byzantine Monetary Economy ; R. Delmaire, Largesses sacres et res privata: L'aerarium imprial et son administration du IV e au VI e sicle (Rome: Ecole Franai se de Rome, 1989) ; Carl, L'oro nella tarda antichit 23 For the concept see C. J. Howgego, Ancient History from Coins (London/ NewYork: Routledge, 1995), 33 38. See also E. Lo Cascio, "How did the Romans View their Coinage and its Function?" in Coin Finds and Coin Use in the Roman World ed. C. E. King and D. G. Wigg (Berlin: Mann, 1996), 282: "The character of coinage as a creation of the state for the satisfaction of its needs does not necessarily imply that it was not used extensively as a means of exch ange, nor considered by its users as accomplishing in the first instance this function." 24 For punishments in case of falsifying gold coins see CJ IX.24.2. The case of logothete Alexander nicknamed "the scissors" for his ability to clip down solidi is reve aling ( HA XXVI.29, discussed by Hendy, Studies 316). For a discussion of the larger implications, see P. Grierson, "The Roman Law of Conterfeiting," in Essays in Roman Coinage Presented to Harold Mattingly ed. R. A. G. Carson and C. H. V. Sutherland (Oxf ord: Oxford University Press, 1956), 240 61. For the interdiction of exporting gold coins outside the frontiers, see CJ IV.63.2. For exceptions, see Hendy, Studies in the Early Byzantine Monetary Economy 260 63. 25 For the role of money in Byzantium see th e more recent overview by Morrisson, "Byzantine Money," 909 66. For the evolution of the monetary system ca 500 700, see also K. Harl, Coinage in the Roman Economy, 300 BC to AD 700 (Baltimore: Johns Hopkins University Press, 1996), 191 203.

PAGE 203

203 nature of pre modern coinages, a bone of contention among numismatists and economic historians for many decades and by extension a topic relevant to the primitivist modernist polemic. 26 The refor m of 498 has been universally acclaimed as a successful measure against the decline of the petty currency during the decades preceding the accession of Anastasius. 27 It has not been sufficiently stressed, however, that the new coinage was revolutionary in m any respects. Not only did it e nsure that the value of the bronze coinage would not relapse to its state from the final decades of the fifth century, but for the first time in the monetary history of the Mediterranean world the nominal value of each denom ination featured prominently on the reverse. It is significant to note that the usual propagandistic imagery of the reverse was almost entirely replaced by the large numeral representing the value, flanked by other purely official markings such as the mint the officina and after 538, the date. In terms of design there is hardly a closer correspondent to modern coinages than the bronze currency introduced by Anastasius in two stages during his reign, in 498 and 512, respectively. 26 For the natu re of early Byzantine coinage, see C. Morrisson, "La monnaie fiduciarie Byzance ou vraiemonnaie , monnaie fiduciaire et fausse monnaie Byzance," Bulletin de la Socit Franaise de Numismatique 34 (1979): 612 y a kredytowy character Historia i wsplczesnosc 5 (1979) : 95 120; Pottier, Analyse 124 26; P. Yannopoulos, "Inflation," 126 28; MIBE 8; Banaji, Agrarian Change 70; Zuckerman, Du village l'empire autour du reg istre fiscal d'Aphrodit (525/526) (Paris: Centre de Recherche d'Histoire et Civilisation de Byzance, 2004 ), 82 83; Carl, L'oro nella tarda antichit 336 40. 27 For the reform of Anastasius and its impact see in particular R. P. Blake, "The Monetary Refor m of Anastasius and its Economic Implications," In : The Disciplines of the Humanities. American Council of Learned Societies Devoted to Humanistic Studies ( Menasha, WI.: George Banta Pub. Co., 1942 ), 84 97; P. Grierson, "The Monetary Reforms of Anastasius and their Economic Consequences," in International Numismatic Convention, Jerusalem 1963; The patterns of monetary development in Palestine and Phoenicia in Antiquity ed. A. Kindler (Tel Aviv/Jerusalem: Schocken, 1967), 283 302; D. M. Metcalf, The Origin s of the Anastasian Currency Reform (Chicago: Argonaut, 1969) ; Hendy, Studies in the Byzantine Monetary Economy 475 92.

PAGE 204

204 The new system was obviou sly designed to regain the confidence of the population and be used according to its nominal value. 28 Can the new bronze coinage be understood as token coinage or at least as leaving room for manipulations leading to a nominal value higher than the intrinsi c one? The evidence of texts, single finds, and hoards point to an affirmative answer although many scholars working within a primitivist paradigm would rather stick with the metallist point of view. 29 To be sure, the view of a strictly non fiduciary natu re of bronze coinage has permitted reconstructions of the exchange rate between the copper follis and the gold solidus during the sixth and seventh centuries, which would otherwise be impossible to estimate given the scarcity of written evidence regarding monetary policies. But there is something more important at stake than losing the ability to create tables reflecting fluctuations in the ratio between gold and copper. The relation between gold coinage and taxation has been discussed at length, as well as its role in the monetization of landlocked areas deep in the hinterland where people still had to acquire gold coins to pay their taxes, although they did not have easy access to an urban market, nor perhaps the need to use money on a daily basis. 30 While the role of gold seems clear, not the same can be said about the copper coinage. Based on official regulations, taxes could only be paid in the most stable and reliable form of currency, which was obviously the gold solidus. Taxes owed to the state 28 Initial popular reaction was negative, bordering on violent, Anastasius being accused of extreme avarice and of "hating all the poor ," for which see Sarris, Economy and Society 201; A. D. Lee, "The Eastern Empire: Theodosius to Anastasius," in The Cambridge Ancient History, vol XIV 55. 29 More recently, MIBE 8; Banaji, Agrarian Change 70. For a detailed discussion of the "metallist vs. "nominalist" interpretation of Roman coinage see K. Strobel, "Geldwesen und Whrungsgeschichte des Imperium Romanum im Spiegel der Entwicklung des 3. Jahrhunderts n. Chr. Wirtschaftsgeschichte im Widerstreit von Metallismus und Nominalismus," in Die konomie des Imperium Romanum 86 168. See also Lo Cascio "How did the Romans," 273 87. 30 Hendy, Studies in the Byzantine Monetary Economy 295 96; Carl, ch. 4.

PAGE 205

205 did not always amount to multiples of one solidus and sometimes its fractions, the semissis and the tremissis were themselves insufficient for settling the debt in gold only. Sixth century Byzantium lacked a silver coinage, other than ceremonial pieces with no r eal economic function. The reasons most often discussed are the scarcity of silver in the Byzantine Empire at a period which coincided with the greatest wealth of silver plate, possible export to Persia which relied almost exclusively on a silver coinage, and perhaps the insufficient difference between the ratio of coined metals and the market ratio. 31 This basically meant that there was not any kind of denomination to fill the gap between the tremissis and the follis (40 nummia ), which might be as wide as t he difference between 40 nummia and 8000 nummia at the exchange rate advanced by numismatists for the later sixth century. In cases when the gold coinage alone could not be used to pay the exact amount due, the taxpayer was forced to add a gold fraction an d whatever went beyond the required sum was returned in copper coin, folles and lower denominations at the current exchange rate between the solidus and the follis 32 In most cases the majority of taxpayers, either urban dwellers hiring out their labor or p easants selling their produce or their labor, had to produce a quantity of gold coin in order to pay their taxes, a process which often involved taking a larger sum of bronze coin to official money changers ( zygostatai ) in order to acquire the much needed gold. This rather straightforward circuit, which of course reflects nothing but the standard scenario devised by the Byzantine authorities, leaves room for speculation regarding the ways in which the transaction took place. Plenty of evidence exists 31 Hendy, Studies in the Byzantine Monetary Economy 451 and n. 1 1; P. Grierson, "The Role of Silver in the Early Byzantine Economy." In Ecclesiastical Silver Plate in Sixth Century Byzantium ed. S. A. Boyd and M. Mundell Mango (Washington DC: Dumbarton Oaks, 1992), 138 46. 32 Hendy, Studies in the Byzantine Monetary Ec onomy 286 87; MIBE 8.

PAGE 206

206 regar ding the careful inspection of gold pieces to make sure that they conformed to the official weight and to insure that whatever sum needed to be paid matched the corresponding quantity of gold expected from full weight solidi (72 solidi to the pound). 33 Agai n this entailed further calculations and further disbursing of copper coin as change. The crux of the matter is the way in which the sum of copper coins was calculated. If it was weighed, as the logic of the non fiduciary thesis would imply, then we would expect a careful regulation of coins in circulation to insure that they corresponded to the official weight, which was itself a reflection of the exchange rate between the solidus and the follis Although the sixth century enjoyed the most complex system o f low value currency since the days of the early Roman Empire, it was far less stable in time and serious fluctuations in weight can be noticed over the long sixth century ( ca 498 616). Eventually, the Anastasian system of denominations would gradually col lapse from the reign of Heraclius onward. However, the evidence of hoards from various regions of the early Byzantine Empire clearly demonstrates that the coin pool towards the end of the sixth century included coins of the same denomination but of very di fferent weight. One could handle a sum of coins which could include a follis of Anastasius from the first 33 Weighing gold coins seems to have been the norm as evidenced by the large number of Late Roman and Early Byzantine exagia known in museum collections and from excavations. The latter are particularly significant as they provide pr ecious information regarding the context in which weights were used. For sixth century exagia see especially M. M. Fulghum and F. Heintz, "A Hoard of Early Byzantine Glass Weights from Sardis," AJN 10 (1998): 105 20; C. Filipova, "Rannovizantiiski merki z a tejest ot c. Bistritsa, Dupnishko," Numizmatika, sfragistika I epigrafika 3 (2006): 185 90; A. Minchev, "Early Byzantine Weights Found in Northeastern Bulgaria and Some Notes on their Production, Distribution and Use," in Acta Musei Varnensis, VII/1, Num ismatic, Sphragistic and Epigraphic Contributions to the History of the Black Sea Coast ed. I. Lazarenko, vol. II (Varna: Zograf, 2008), 7 medievale descoperite recent n Dobrogea," Pontica 42 (2009): 671 82. Abuses are also known, including the use of heavier exagia by tax collectors, mentioned in C.Th XI.8.4. In order to preven t corruption Novella 128 from 545 encouraged taxpayers to request the official weights of the comes sacrarum largitionum For a discussion of such practices, see Delmaire, Largesses sacres 518 19; Hendy, Studies 332; C. Morrisson, "Weighing, Measuring, Paying: Exchanges in the Market and the Marketplace," in Trade and Markets in Byzantium 386 89. For the practice of weighing solidi see 116.

PAGE 207

207 reform weighing around 8g together with large folles of Justinian weighing over 20g and later coins of Justin II, Tiberius II, and Maurice weighing ar ound 11 12g, all displaying the large M indicating the nominal value of 40 nummia. The clear implication is that bronze coins were counted rather than weighed in transactions with the state (and elsewhere), based on a fragile consensus that the face va lue should prevail. Moreover, there would be no need for five different denominations of the bronze coinage if coins were to be weighed during what would arguably be the most important transactions of the annual cycle, the payment of taxes. The denominatio nal system was obviously geared toward the market, which would soon collapse if weighing coins were to be the preferred method of payment. The written sources, fragmentary though they are with respect to money, almost always refer to numbers when discussin g the petty currency. Thus we learn that a worker digging a well received five folles a day for his labor, bread was sold for 1 follis in Constantinople, Procopius criticized Justinian for allowing the number of folles to one solidus to fluctuate, while Ps ellos mentioned that small change was to be numbered and not weighed like bullion, testifying that the practice was maintained later in Byzantium. 34 In addition, there is no way of accepting two different functions of the copper coinage, a metallistic one for dealings with the state (weighing) and nominalistic in market transactions (counting). In such a case, the strategy of sellers and buyers would be to hold on to the heavier specimens in order to use them for the payment of taxes, while the owner of the well from the account of John Moschus would try to pay his worker in 34 For a brief discussion see Harl, Coinage in the Roman Economy 285 89. For detailed lists of prices and wages, see C. Morrisson and J. C. Cheynet, "Prices and Wages in the Byzantine World," in Economic History of Byzantium ,, 815 78. For weighing vs counting see recently Morrisson, "Weighing, Measuring, Paying," 379 98, esp. 379 89 for Late Antique practices.

PAGE 208

208 lighter coins, for similar reasons. Similarly, the state would withdraw from circulation those specimens which did not conform to the standard and ran the risk of being hoarded acco In conclusion, both the literary and the archaeological evidence points to the fiduciary nature of the petty coinage and to the fact that the term Gewichtgeldwirtschaft refers only to gold coinage. 35 The significant number of heavy coins of Justinian in late sixth century hoards discussed elsewhere clearly shows that we are dealing with a fiduciary coinage in the pre modern sense of the term, which allows for a certain overvaluation dependant on the confidence of the population as we ll as on the 36 This does not mean that the confidence of the population once acquired after half a century of insignificant changes in the aspect of the bronze coinage could never be lost again. There is ample evidence of c risis in the second half of the sixth century when the short term effects of the plague and the long term effects of waging war on several fronts translated into rampant inflation. Economic xplain inflationary mechanisms, but the matter might be settled quite simply by referring back to the needs of the state. A crippled demographic base of recruitment and taxation after the plague coupled with the increased militarization of frontiers meant that more money was needed to pay the troops and to meet higher prices. Excess of expenditure over income accounted for the issue of fresh coin as did the rate of attrition through hoarding, casual loss, or export across the frontiers. The type of response adopted by the administration 35 Carl, 476. 36 Consequences," RN 168 (2012): 363 402.

PAGE 209

209 when faced with a crisis, namely lowering the weight of the coins bespeaks the nominalist function of the coinage, as clearly more and more coins were needed, understood as units of 40 nummia and its fractions, in other wo rds understood as numbers not as quantity (bullion). Since the supply of metal was limited, now more than ever the Byzantine administration needed to overvalue its currency in order to meet the new challenges. That the attempt was not entirely successful, we know from the fact that heavier specimens were eventually withdrawn from circulation at the end of the century in the attempt to standardize the currency while gaining additional metal for striking fresh coin. The breakdown of the Anastasian system and the subsequent simplification of the copper currency are attested by both written and archaeological sources. The crisis accelerated in the second decade of seventh century and fractions of the follis below the half follis became rare. To be sure, the pro cess had been under way, albeit at a slower pace, since the mid sixth century when the nummus the lowest denomination of the early Byzantine system, disappeared almost completely from circulation, as attested by hoard evidence. 37 Pentanummia however, cont inued to be issued in considerable numbers at least until the reign of Maurice. The collection of the miracles of St Artemius describes a man who gathers all the small change he dropped on a street of Constantinople around 640. He managed to retrieve all t he coins down to the last half follis which would suggest that this was the smallest denomination available. 38 While 37 DOP 65 66 (2012): 45 111. 38 The M iracles of St. Artemios: A Collection of Miracle Stories by an Anonymous Author of Seventh Century Byzantium ed. V. S. Crisafulli and J. W. Nesbitt (Leiden and New York, 1997), 129.

PAGE 210

210 entirely accurate, it certainly illustrates the situation in provincial towns. 39 The breakdown of the currency system led to a simplification of urban market transactions as people could no longer rely on a steady supply of lower denominations. The early Byzantine administration dealt with a very fragile balance of revenue and expenditure and any serious disruptions could throw the entire fiscal system into collapse. 40 Multipl e crises starting from the 540s leave the impression of a long agony but also illustrate the resilience and the capacity to adapt, if not always successfully, to the new conditions. This much can be seen from the manipulation of the copper coinage, discuss ed above, but also from the introduction sometime in the 540s of a series of officially sanctioned light weight solidi worth 20, 21, 22, and 23 siliquae respectively. 41 The exact purpose of this coinage has been a longtime preoccupation of numismatists and historians, various theories being advanced such as trade, external payments, and internal regulations, but it seems that the latter is more likely closer to reality, namely the need to account for the worn solidi which no longer corresponded to the offi cial weight. 42 39 Excavations in Istanbul have produced many decanummia dated to the sec ond half of the seventh century. See M. F. Hendy, "The Coins," in R.M Harrison et al, Excavations at Sarahane in Istanbul vol. I (Princeton: Princeton University Press, 1986), 313 23; M. F. Hendy, "Roman, Byzantine and Latin Coins," in Kalenderhane in I stanbul. The Excavations ed. Philipp von Zabern, 2007), 222 35 and 275. 40 see In Laudem Iustini Augusti Minori II, ed. and trans. A. Cameron (Lo ndon: Athlone Press, 1976), 55. 41 The date of the introduction of light weight solidi is debated, see Hendy, Studies in the Byzantine Monetary Economy 493 (after 538/9); Delmaire, Largesses sacres 521 22 (547 550 ); MIBE 12 (538 542); Carl, 381 (546 547). 42 MIBE 12; Carl, 385. Constantine Zuckerman also favors an internal explanation for the use of light weight solidi but argues that these were meant to pay donativa and other types of irregular largesse, see C. Zuckerman, Du village l'empire 88. For older interpretations, see H. L. Adelson, Light Weight Solidi and Byzantine Trade during the Sixth and Seventh Centuries (New York, American Numismatic Soc iety, 1957). Regarding old gold coins in circulation and the legislation, see Hendy, Studies 366; Banaji, Agrarian 70 75; Harl, Coinage in the Roman Economy 196. Interestingly,

PAGE 211

211 The introduction of the silver hexagrammon by Heraclius is another similar example. As noted, silver had been a ceremonial coinage during the sixth century, although silver coins were mostly issued on the standard of the siliqua or the milia rensis and their fractions. Many displayed meaningless vota inscriptions typical for the Late Roman coinage, some featured the inscription TP ( tropaion ), clearly alluding to their purpose as awards, while others were nothing but token pieces featuring a la rge K on the reverse to be distributed around May 11 when Constantinople was celebrated. 43 The role of silver changed dramatically in the early seventh century during the equally dramatic events that forced Heraclius to seize the silver plate of the church in order to strike a brand new coinage. Hexagrams were intended for paying salaries at 50 percent of the former rate, while taxes would continue to be collected in gold, a clear intention to increase the state revenue during a time of deep crisis. 44 Michae l Hendy has argued that no less than 3/8 of the budget was used to pay the army and the administration. The budget itself was around 5 6 million solidi the highest estimate for the Empire during the Justinianic age. 45 Military payments are of outmost impor tance for understanding the particularities of coin circulation in the highly no light weight solidi have been found in Palestine, for which see G. Bijovs ki, "Monetary Circulation in Palestine during the Byzantine Period (Fifth Seventh Centuries CE)," Ph.D. Dissertation, Hebrew University Jerusalem, 2011. The scarcity of finds within the Empire compared to those in barbaricum leaves room for further specula tion regarding the fun ction of this peculiar coinage. 43 MIBE 27, 33 and 52; MIBEC 26 and 39. Banaji, Agrarian Change 44, argued that miliarenses were used to pay construction workers, but this does not seem convincing. 44 Hendy, Studies in the Byzantine Monetary Economy 494 95; For a book length treatment, see P. Yannopoulos, e sicle (Louvain la Neuve: Institut ) 45 Hendy, Studies in the Byzantine Monetary Economy 171 72.

PAGE 212

212 and irregular items. The annual salary was theoretically paid in gold but there are suff icient grounds to argue that at least in part it was paid in bronze. The ration allowance ( annona ) and fodder allowance for cavalrymen ( capitus ) had been commuted in coin ( adaeratio ) by Anastasius, and amounted to 4 solidi for the former and 4 5 solidi for the latter. 46 The evidence of military site finds and hoards as well as the output of mints supplying frontier regions show that bronze coin was issued in high quantities to satisfy the needs of the army. 47 During the revolt against Phocas and later during the war with Persia, Heraclius opened temporary mints in the E ast with the clear purpose of supporting the army. The case of Cyprus is perhaps the most revealing. 48 In addition, Wolfgang Hahn has identified and classified a series of anomalous issues labele d moneta militaris imitativa under the plausible assumption that they were issued by mobile mints traveling with the army in order to cover the need for coin when official issues were not available in sufficient quantity. 49 Such semi official imitations lar gely correspond to the weight and style of the regular coinage and their presence among 50 The fact that salaries were partly paid in bronze, which the soldiers needed for small transactions, explains the inflation of the petty currency in the second half of the sixth century when 46 Ibid., 646. 47 See for instance the activity of Thessalonica mint for which see D. M. Metcalf, The Copper Coinage of Thessalonica under Justinian I (Vienna: Verlag der Osterreichischen Akademi e der Wissenschaften, 1976), 8. Further evidence from Illyricum, discussed in the following section, substantiates this hypothesis. 48 D. M. Metcalf, Byzantine Cyprus 491 1191 (Nicosia: Cyprus Research Centre, 2009), 163. 49 MIB II 49 51, 58, and 74. 50 A. G "Sixth to Seventh Century Coin Circulation in Dobrudja," CN 9 11 (2003 2005): 109 66.

PAGE 213

213 frontier provinces of both the Balkans and the East came under constant pressure. It is less safe to argue that irregular distributions were also paid partly in bronz e, although the archaeological evidence seems to point in that direction at least for the Danube region. 51 The most important distributions were the donativum on imperial accession ( augustaticum ) which consisted of five solidi and a pound of silver durin g the early sixth century and nine solidi on the accession of Tiberius II, 52 and quinquenial donativa amounting to five solidi per man. 53 It is often argued that the latter was suppressed by Justinian, based on the testimony of Procopius, but the gold output of the mint of Carthage displays five year spikes as late as the reign of Maurice as do the copper coin finds from the Danube frontier during the second half of the sixth century, clearly evidencing the fact that the quinquenial donativum was not permanen tly discontinued. 54 In 596 Maurice modified the payment of salaries ( rogai ) and donativum ( epidosis ) by distributing it one third in coin and the other two thirds in equipment and weapons after he had already attempted to reduce the donativum by 25 percent in 587. 55 These 51 Five year spikes in coin finds from the Danube frontier may reflect distributions of donativa in copper, e Eastern Provinces: A Statistical Approach." American Journal of Numismatics 21 (2009): 208, fig. 3. Other irregular gifts ( dora ) of gold distributed to deserving soldiers who distinguished themselves in battle are discussed by Delmaire, Largesses sacres 554. 52 Hendy, Studies in the Byzantine Monetary Economy 177 argues that the pound of silver was commuted into gold at the exchange rate between the two metals. Silver, however, could have been distributed in the form of plate as suggested by a scene des cribed by Corippus in which Senators In laudem Iustini Augusti minoris IV, 109 111, ed. and trans. A. Cameron (London: Athlone Press, 1976). 53 Calc ulated at 750000 solidi for the age of Justinian based on the size of the army; see Hendy, Studies in the Byzantine Monetary Economy 177. 54 C. Morrisson, "Estimation du volume des missions de Solidi de Tibre et Maurice Tibre a Carthage (578 602)," in P ACT 282. 55 Theophylact Simocatta, Historia III, 1.2 and VII, 1.2, ed. C de Boor and G. Wirth (Stuttgart: Teubner,

PAGE 214

214 measures reveal the great stress placed on the imperial treasury by the need to finance war on two fronts. They also explain the reaction of the soldiers and of the population at large. The latter cared less about the imperial treasury left irrational largesse and mocked Maurice with appellations such as Maurice the Marcianist to his deposition in 602. 56 Fluctuations in the production of copper coinage at Constantinople, Nicomedia, and Cyzicus we re a direct reflection of political and economic crisis in the final decade of the sixth century. 57 Despite the wealth of information which the study of low value currency can contribute to a well rounded discussion of the early Byzantine economy, historians have seldom ventured beyond the analysis of the gold coinage, and even here rarely leaving the confines of the written sources to address the archaeological evidence. The assumption, which pervades Mic works, is that copper coinage had an insignificant value and as such, did not contribute much to the overall value of transactions involving coinage. This type of macro scale analysis emphasizes quanti ty and value at the expense of other variables such as frequency and integration. To be sure, copper coinage was used only in small scale transactions, serving the needs of the urban market and to a lesser extent that of monetized rural spaces, but at the same time low value currency reflected the daily 1972), 110 and 245 46. dvaluation et reevaluation la tra nsitions des mondes romain et byzantin," in Histoire conomique de l'antiquit: bilans et contributions de savants belges prsents dans une runion interuniversitaire ed. T. Hackens and P. Marchetti (Louvain la Neuve: Sminaire de numismatique Marcel Hoc Collge Erasme, 1987), 129. Delmaire disagrees with the reading of rogai as salaries, see Delmaire, Largesses sacres 559, with n. 81. 56 Byzantina (T hessaloniki) 11 (1982): 181 88. 57 86.

PAGE 215

215 activity of the majority, rather than the dealings of a wealthy minority or official business such as the payment of salaries and the collection of taxes. 58 The fact that the combined value of copper coin fin ds from excavations conducted in the past century amount only to the value of a few hundred solidi should not deter us from a sustained effort of understanding the role and function played by petty currency in the sixth century economy, which can be charac terized by a dualism between the circulation of gold through official channels and the much more dynamic marketplace dominated by high frequency transactions involving copper coins. 59 Indeed, a careful comparative study of coin finds in urban centers and ru ral settlements of the early Byzantine Empire, as well as the search for patterns of coin production and circulation through the study of hoards and large public and private collections can be brought to bear on the larger issue of trade, markets, and econ omic integration. 5.2 Coin Production, Coin Circulation and the Nature of the Evidence The numismatic evidence, whether in the form of hoards, site finds, accidental finds or museum collections is fragmentary and not without methodological problems. Hoards can tell us a lot about the lifespan of coins in circulation and sometimes they 58 For an engaging discussion of daily exchanges in the marketplace, see recently B. Pitarakis, "Daily Life at the Marketplace in Late Antiquity and Byzantium," in Trade and Markets in Byzantium 399 426 59 For the dual nature of the early Byzantine monetary economy, see Carl, 474. Commercial areas or "shops" have been excavated in many urban centers of the early Byzantine Empire. In cases where the coin finds have been publi shed, such as Sardis, Scythopolis, Berytus, and Tomis, a common characteristic is the significant number of smaller denominations of the follis which clearly points to the high frequency of low value transactions. To be sure, in most of these cases we are probably dealing with a mix of commercial, industrial and residential structures. For the finds, see M. D. Weishan, "Appendix 1: Conspectus of Mints," in J. Stephens Crawford et al, The Byzantine Shops at Sardis (Cambridge MA: Harvard University Press, 19 90), 126 28; G. Bijovski, "The Coins." In S. Agady et. al, "The Bet Shean Archaeological Project,"in What Athens Has to Do with Jerusalem. Essays on Classical, Jewish, and Early Christian Art and Archaeology in Honor of Gideon Foerster ed. L. V. Rutgers ( Leuven: Peeters, 2002), 507 12; Butcher, "Archaeology of the Beyrut Souks," 7 304; G. Poenaru Bordea et al., (Wetteren: Moneta, 2004).

PAGE 216

216 can provide accurate snapshots of the variety of coins in circulation in a certain time and place. Unfortunately, only a small percentage of hoards are found during archaeolog ical excavations, many being the result of chance find during agricultural works, construction work, and the use of metal detectors. Museum curators thus often find it difficult to establish the original number of coins in such hoards and the exact circums tances and context of the find. Site finds are much more useful for establishing patterns of coin loss over time, but a large enough sample is needed in order to reach statistically meaningful results, while at the same time similar samples must be availab le from other sites in the same region as well as further afield for comparison purposes. Moreover, few sites have been completely excavated. By necessity, the available numismatic evidence reflects the state of research as well as the function of the comp lexes thus far excavated, which may have a residential, economic, or religious function to name just a few. Finally, museum collections are often the result of amalgamating the types of evidence mentioned above, usually containing a high percentage of coin s with unknown provenance, in the case of large museums (e.g. British Museum), or with an uncertain provenance within a given region, in the case of local history museums from regions which were once part of the early Byzantine Empire. While acknowledging the nature and limitation of the sources, there is still much to learn from carefully navigating various types of numismatic evidence which can tell us a lot about the social, political and economic conditions of the early Byzantine Empire. Museum collect ions in particular hold great promise for gaining some understanding of fluctuations in coin production during the long sixth century. Although the collection and recording of Byzantine coins has been a longtime preoccupation of

PAGE 217

217 European scholars and intel lectuals which led to the creati on of large private and public collections, their value has been rarely recognized and modern numismatists have been too ready to dismiss them as irrelevant for answering broader historical questions. As early as the first h alf of the nineteenth century antiquarians and collectors became interested in developing means of organizing and systematizing the Byzantine coin series. The chief concern was to create suites montaires an attempt to gather all the known coin types is sued by the Byzantine state. The pioneering works of de Saulcy and Sabatier in the nineteenth century had been an important starting ground for the subsequent catalogue of the collection published by Count Tolstoi between 1908 and 1911. 60 However, the stand ard work for more than fifty years would become the catalogue of Byzantine coins in the British Museum. What made it atypical for this early period was the decision to publish an entire collection, whose purpose would be twofold: to fill the gaps in the By zantine coin series and to provide scientific access to an entire collection, including duplicates. 61 The breakthrough made after the publication of the major collection at Dumbarton Oaks, assembled through mass purchases, opened a new era in terms of the methodology behind the study of Byzantine coinage. 62 Alfred Bellinger and especially Philip Grierson embarked on the task of reassessing many of the old datings 60 F. de Saulcy, Essai de classification des suites montaires Byzant ines (Metz: Lamort, 1836); M. Soleirol, Catalogue des monnaies byzantines qui composent la collection de M. Soleirol (Metz: Lamort, 1853); Sabatier ; Tolstoi. Equally important are Thodore De la raret et prix des mdailles romai nes 2nd volume, 2nd ed. (Paris: De Bure, 1827), the second part of Christian Catalogue de la collection de monnaies de feu Christian Jrgensen Thomsen. Seconde partie: Les monnaies du moyen age, tome I (Copenhague: Imprimer ie de Thiele, 1873) as well as the contribution to the classification of Justinian's coinage I by J. Friedlnder and M. Pinder, Die Mnzen Justinians (Berlin: Nicolai, 1843) 61 BMC 62 DOC I

PAGE 218

218 and attributions in what became a seminal work for our understanding of the early Byzantine coi nage. Ccile Morrisson went a step further by cataloguing the old and important collection of the Bibliothque Nationale in Paris, enriched with the donations of leading scholar collectors such as Gustave Schlumberger. Equally significant have been the maj or private collections of Byzantine copper coins made available to the scientific world, such as those of Rodolfo Ratto and George Bates. 63 Creating a corpus of all known Byzantine coin finds, as the latter has pleaded for in the introduction to his catalog ue, might prove to be an illusory endeavor if we take into account the large number of coins currently on the market. Nonetheless, the number of coins in national or local museums from the Balkans, Turkey or the Middle East has greatly increased in the pas t fifty years due to extensive archaeological research, often performed by international teams of scholars at Apamea, Sardis, Berytus and Caesarea Maritima to name a few of the most important. In addition, the collections assembled by museums and universit ies in Western Europe contribute to the wealth of Early Byzantine coins available for study. 64 The difficult task of assembling all known Byzantine coin types was attempted by Wolfgang Hahn in his series Moneta Imperii Byzantini 65 However, few initiatives h ave been taken towards a statistical understanding of the monetary 63 Ratto ; G. E. Bates, A Byzantine Coin Collection (Boston: Pri vately Printed, 1981). 64 Most important are the collection of the Hunter Coin Cabinet in Glasgow, J. D. Bateson and I. G. Campbell, Byzantine and Early Medieval Western European Coins in the Hunter Coin Cabinet (London: Spink, 1998), the collection Khler Osbahr from the Duisburg Museum, KOD the collection of the University of Gttingen, A. S. Sommer, Katalog der Byzantinischen Mnzen (Gttingen: Universittsverlag Gttingen, 2003), and the collection of the Bottacin Museum in Padova, B. Callegher, Catalo go delle monete bizantine, vandale, ostrogote e longobarde del Museo Bottacin vol. I (Padova: Comune di Pado va, Musei e biblioteche, 2000). 65 MIB.

PAGE 219

219 circulation at the scale of the entire empire, 66 although the use of quantitative tools was promoted and employed for assessing local provincial patterns in coin circulation. 67 The use of st atistical tools in Byzantine numismatics is largely a post war development. The growing interest in elaborate means of quantification lies both in the need for a more complex method of analyzing the increasing number of coins and in the introduction of com puter based programs which facilitated such an approach. Mathematical tools have always been employed in numismatics; by necessity, coins needed to be counted and classified based on chronological and typological criteria, but no attempt was made to analyz e them statistically. 68 D. M. Metcalf has been a pioneer in this respect. 69 His study of Byzantine coins in Sirmia and Slavonia represents the first elaborate attempt to use statistics in order to understand the Early Byzantine coin circulation in that regio n. It also represented an opportunity to make use of comparative statistics, which permitted a number of generalizations at the scale of the Eastern Empire, based on the evidence available from the excavations at Corinth, Athens, 66 C. Morris son, "Byzantine Money," 909 66. 67 D. M. Metcalf, "Some Byzantine and Arab Byzantine Coins from Palaestina Prima," Israel Numismatic Journal 2, n. 3 4 (1964): 32 47; E. J. Pradwic Golemberski and D. M. Metcalf, "The Circulation of Byzantine Coins in the South Eastern Frontiers of the Empire," NC 123 (1963): 83 92; H. Pottier, e monnaies en bronze enfoui au VI e sicle en Syrie byzantine: contribution la mthodologie numismatique des VI e VII e sicles et leur circulation dans les Balkans," in Trsors ; Mnzfunde. 68 H. W. Bell, Sardis, vol. XI/ part I, 1910 1914: Coins (Leiden: Brill, 1916); K. Edwards, Corint h VI: Coins, 1896 1929 (Cambridge MA: Harvard University Press, 1933); S. McA. Mosser, A Bibliography of Byzantine Coin Hoards Numismatic Notes and Monographs 67 (New York: American Numismatic Society, 1935). A. R. Bellinger, Coins from Jerash, 1928 1934 (New York: American Numismatic Society, 1938); D. Waage, Antioch on the Coins (Princeton: Department of Art and Archaeology of Princeton, 1952); M. Thompson, The Athenian Agora 2: Coins from the Roman through the Venetian Period (Princeton: The American School of Clas sical Studies at Athens, 1954). 69 For an early methodological essay see D. M. Metcalf, "Statistische Analyse bei der Auswertung von Mnzfundmaterialen," Jahrbuch fr Numismatik un d Geldgeschichte 9 (1958): 187 96.

PAGE 220

220 Antioch, and Sardis. 70 Star ting from the early 1980s one can observe an explosion of studies employing more or less sophisticated statistical tools. The main impetus had been provided by the organization of a Round Table dedicated to the use of statistics in numismatics, in which re puted numismatists and professional statisticians collaborated for a better implementation of statistics in numismatic research. 71 The work diversified and the aims turned more ambitious, to analyses of metrology and calculations of mint output. Again, D. M. Metcalf should be mentioned for his role as a pioneer. His work on particular early Byzantine issues, the Anastasian small module coins and the Justinianic coinage from Thessalonica, represent early attempts to determine Byzantine mint output. 72 However, calculations of mint output based on die studies have not moved too far for the Byzantine series. Aside from the work of D. M. Metcalf, calculations have been attempted by W. E. Metcalf and Ccile Morrisson for small issues such as the joint reign solidi of Justin I and Justinian I and gold issues of Carthage, respectively. 73 The sheer size of the Byzantine base coinage has prohibited scholars from attempting any such calculations and the situation will probably remain the same in the foreseeable future. Co nsequently, students in the field 70 D. M. Metcalf, The Currency of Byzantine Sirmia and Slavonia," Hamburger Beitrge zur Numismatik 14 (1960): 429 44. 71 PACT. For the statistical method see also C. Carcassone and J. Guey, "Valeur statistique des petits chantillons," Revue Belge de Numismatique 124 (1978): 5 21; C. Carcassone, Methodes statistiques en numismatique (Louvain la Neuve: Sminaire de Numismatique Marcel Hoc, 1987). 72 Metcalf, The Origins of the Anastasian Currency ; Metcalf, The Copper Coin ag e of Thessalonica. 73 D. M. Metcalf, The Joint Reign Gold of Justin I and Justinian I," in Studies in Early Byzantine Gold Coinage 19 27; Morrisson, "Estimation du volume," 267 84; C. Morrisson, "The Moneta Auri under Justinian and Justin II, 537 578," in Studies in Early Byzantine Gold Coinage 41 64. Although removed from the chronological focus of this chapter, the comprehensive die study done by Franz Feg on the eighth century gold issues remains seminal for a general understanding of the Byzantine go ld coinage even if extrapolations can be problematic, F. Feg, "Die Solidusausgaben 717 803 in Konstantinopel," Revue suisse de numismatique 70 (1991): 35 54.

PAGE 221

221 of Byzantine numismatics have been less engaged in the lively debate of the last two decades centered on the question of mint output. 74 Steps have been taken, however, to understand the metrology of the multi denominational system of Byzantine coinage. The publication of the large collections of Dumbarton Oaks and Paris provided the opportunity for metrological calculations based on large samples. In addition, Henri Pottier contributed a seminal book for the metrological stu dy of Byzantine coinage but also for the monetary circulation in Syria, based on comparative statistics. 75 In the past two decades statistic methods in Byzantine numismatics have been used in almost any study dealing with a substantial sample of coins, eit her hoards or stray finds. The coin hoards from the Balkans, Anatolia, and the Middle East were processed mainly using a statistical apparatus. One should mention here a substantial article by Florin Curta on hoards from Eastern Europe with a thick appendi x of statistical results, the monumental trilogy of Hans Christoph Noeske on the Byzantine coin circulation in Egypt and the Near Eastern provinces, of which the last volume comprises a few dozen graphs derived from statistical calculations, and the collab ora tive work in the Balkans and in Anatolia. 76 At the same time, studies of stray and single finds from 74 W. Esty, "Estimating the Size of a Coinage," NC 146 (1986): 185 215; Buttrey, "Calculating Anci ent Coin Production," 335 351; T. V. Buttrey, "Calculating Ancient Coin Production II: Why it Cannot Be Done, NC 154 (1994): 341 52; T. V. Buttrey and S. E. Buttrey, "Calculating Ancient Coin Production, Again," American Journal of Numismatics 9 (1997): 113 35; R. Duncan Jones, Money and Government in the Roman Empire (New York: Cambridge University Press, 1994) with an extensive review by W. E. Metcalf, "Review of Duncan Jones 1994," Revue Suisse de Numismatique 74 (1995): 145 159; de Callata, "Calculat ing Anc ient Coin Production," 289 311. 75 Pottier, Analyse d'un trsor See more recently H. Pottier, "Nouvelle approche de la livre byzantine du V e au VII e sicle," Revue Belge de Numismatique et Sigillographie 150 (2004): 51 133. 76 F. Curta, "Invasion o r inflation? Sixth to Seventh Century Byzantine Coin Hoards in Eastern and Southeastern Europe," Annali di Istituto Italiano di Numismatica 43(1996): 65 224; Mnzfunde ; Trsors.

PAGE 222

222 major archaeological sites have included statistical analyses of rec ent finds and also previously published material. 77 The purpose of the following sections are twofold: first, it attempts to identify general fluctuations in the production of base metal coinage, based on the evidence of the major public collections and s econdly, it draws a series of comparisons between coin finds in the major geographical units of the Eastern Empire. This study deals exclusively with copper coins issued in the major Eastern mints, Constantinople, Thessalonica, Nicomedia, Cyzicus, and Anti och. The reasons behind this decision are both technical and practical; most of the Early Byzantine coin finds originate in the eastern part of the empire, from the Danube to the Eastern Mediterranean provinces, while the monetary system itself was not sta ndardized throughout the Empire. Egypt was largely self sufficient, Italy and the African mint at Carthage had different rhythms of coin production than the Eastern mints, and the ratio between gold and copper seems to have had regional particularities. 78 Therefore, in order to e nsure the accuracy of statistical parameters and ultimately of the historical conclusions drawn from this material, Western provinces, including North Africa were left out. In quantitative terms, this means working with more than 1 0,000 bronze coins located in five major collections 77 T. Marot, Las monedas del Macellum de Gerasa (Yaras, Jordania): aproximacin a la circulacin monetaria en la provincia de Arabia (Madrid: Museo Casa de la Moneda, 1998); K. Sheedy, "Byzantine Period Coins," in K. Sheedy, A. R. Carson, and A. Walmsley, Pella in Jordan, 1979 1990: The Coins (Sydney: Adapa, 2001), 129 45; Butcher, Archaeology of the Beirut Souks; J. DeRose Evans, The Joint Expedition to Caesarea Maritima. Excavation Reports. Volume VI. The Coins and the Hellenistic, Roman, and Byzantine Economy of Palestine (Boston: American Schools of Oriental Research, 2006); A. G "Some Aspects of the Monetary Circulation in the Byzantine Province of Scythia during the 6th and 7th Century," in Numismatic, Sphragistic and Epigraphic Contributions to the History of the Black Sea Coast edited by I. Lazarenko, vol. 1 (Varna: Zo graf, 2008), 301 30 78 MIB.

PAGE 223

223 Eastern Empire 79 The single finds resulting from systematic archaeological research are unevenly distributed over the geographical area under consideration. The material from the Balkans is by far the most abundant, followed by the Near East, where numerous Syro Palestinian sites have been excavated. Chris Lightfoot has sketched the current state of the Byzantine researc h in Anatolia and drew attention to the lack of interest for Byzantine archaeological layers in favor of the presumably more sophisticated classical period. 80 Recent research by Zeliha Demirel Gkalp has proven that the Turkish archaeological museums prese rve a wealth of Early Byzantine coins found in Anatolia, which awaits publication. 81 Although a few tentative steps have been taken towards a broad understanding of coin circulation in the Balkans and the Middle East, 82 little has been done with respect to gathering the numi smatic material for an in depth 79 The collections under consideration are Dumbarton Oaks and the Whittemore collection ( DOC ), Bibliothque Nationale in Paris ( BNP ), the collection of the British Museum ( BMC ), the collection Khler Osbahr in Duisburg ( KOD ), and the vast collection of the American Numismatic Society ( ANS ), still unpublished. They were chosen based on size and on the preponderance of copper issues, including numerous duplicates. The collections of Tolstoi and Ratto are not included in the statisti cal analysis, as the former published a type catalogue and the latter a sale catalogue and therefore neither was interested in including duplicates. Even if they remain outside the scope of this study, such collections retain a statistical significance for the understanding of mint output by looking at the varieties they were unable to find in their desire to assemble th e entire Byzantine coin series. 80 C. Lightfoot, "Byzantine Anatolia: Reassessing the Numismatic Evidence" RN 158 (2002): 229 39. 81 I wish t o express my gratitude to Zeliha Demirel Gkalp for allowing me to consult her unpublished PhD dissertation and two other unpublished catalogs of the Byzantine coins from the Malatya and Bolu museums. 82 C. Morrisson, "La monnaie en Syrie Byzantine," in Arc hologie et histoire de la Syrie II. La Syrie de ed. J. M. Dentzer and W. Orthmann (Saarbrcken: Saarbrcker Druckerei und Verlag, 1989), 187 204; C. Morrisson, "La circulation montaire dans les oque justinienne et post justinienne," in Acta XIII Congressus internationalis archaeologiae christianae ed. 30; A. Walmsley"Coin Frequencies in Sixth and Seventh Century Palestine and Arabia: Social and Economic Implications." Journal of the Economic and Social History of the Orient 42, no. 3. (1999): 326 "Some Aspects."

PAGE 224

224 comparative analysis, partly because of the still insufficient evidence. The case of the border province of Scythia is a unique situation, given that Romanian numismatists have constantly published comprehe nsive catalogues of recent finds and museum collections, 83 bringing the total number of Early Byzantine coin finds to a dazzling figure of more than 3,000 specimens. The relevance of site finds has been a debated issue. Philip Grierson has argued that the structure of site finds tends to favor the smaller coins because they were easily lost and not retrieved and thus they cannot offer a completely reliable image of the circulation. 84 What seemed acceptable at the time when Grierson was suggesting such an int erpretation of site finds is no longer tenable in the light of the material coming from the Balkans, Anatolia and the Near East. His conclusions were chiefly based on the major centers of the early Byzantine Empire, Corinth, Athens, Sardis, Antioch, and Co nstantinople where excavations had yielded a large number of small denominations. The idea that small coins were more easily lost due to their size seemed perfectly reasonable both because of the structure of finds and a sort of natural logic suggesting th at the smaller the coin (and the lesser the value) the higher the probability of it being 83 See the chronicles of the recent finds compiled by B. Mitrea and G. Poenaru Bordea from Dacia "Dcouvertes de monnaies antiques et byzantines en Roumanie" along with the ones regularly published in Pontica by Gh. Papuc, R. Oche descoperirilor monetare din Dobrogea". While the coins published in the Romanian journals were mainly single finds resulting from archaeological research, in Bulgaria such articles were devoted exclusively to Kolektivni nahodki na moneti" from Izvestiia na Arkheologiche skiia Institut 84 P. Grierson, "Circulazione monetaria e tesaurizzazione," in La Cultura bizantina, oggetti e messaggio: moneta ed economia ed. The Origins of the Anastasian C urrency 94, who argued that folles the same chance of being lost".

PAGE 225

225 lost. 85 This, however, does not help to explain why more than 80 percent of the coins coming from archaeological excavations in the Balkans and some Anatolian and Near Eastern sites are folles and half folles 86 The argument that excavators missed the smaller coins cannot be a valid explanation given the wealth of 4 th 5 th century material recovered from the same archaeological sites, many coins being half the size of a Byzantine penta nummium We must therefore accept the possibility that people were primarily losing coins based on availability, not size. Setting a pattern of the coins in circulation might be regarded as an endeavor both daunting and risky. While assessi ng the mint output based on die studies may be a fruitful approach for rare specimens or gold/silver issues, it is hardly a viable course of action in for the circulating base metal coinage. Even if the method was profitable it could only point to the pote ntial number of issues. The real number is effectively connected to minute mechanisms of the Byzantine monetary economy, whose purpose clearly was not to use dies until worn out, but to control the market through a regulated inflow of fresh currency, to pa y the army and the administrative apparatus, and to insure the collection of taxes. The lively debate initiated during the last decades has pointed to variables in determining mint output, which ultimately compromised the value of this method as a definiti ve tool in assessing absolute coin volumes. 87 The devastating criticism of T. V. Buttrey in a period when such applications were flourishing has 85 This interpretation held as a general applicable rule is still advocated, especially for sites in the Middle East. See more recently Sheedy, "Byzantine Period Coins ," 5. 86 G Coins from Jerash 95 119; Marot, Las monedas del Macellum 322; Z. Demirel Gkalp, "Yalva ve Isparta arkeoloji mzelerinde bulunan Bizans sikkeleri," Ph.D. Dissertation, Eski ir University, 2007. 87 See above, n. 20.

PAGE 226

226 precisely the merit of pointing to problems with this model. Although the discussion was centered on Greek and R oman coinage, his arguments were generally applicable to any ancient coin series. The skepticism regarding figures drawn for gold and silver series turns into total despair in the case of copper issues characterized by large die populations and high wear f actors due to intensive circulation. 88 Even if we could take advantage of the fact that most of the copper coins were dated with the regnal year starting with 538 and could hypothetically determine the mint output for a certain type based on the number of s urviving dies, we would still be nowhere near having a large understanding of the phenomenon of coin production. What would be needed, and is unfortunately illusory, is the absolute numbers for the variety of denominations, mints, officinae and dates in a given period. The only approach capable of spawning relevant statistical figures, insofar as they could be determined with our current body of knowledge, is the one based on the coin sample at our disposal. The representativity of museum collections first came into discussion as a central argument in the 1950s when scholars were trying to make sense of the transformations that brought the once prosperous empire into a Dark Age Using the numismatic material from the British Museum Alexander Kazhdan argue d that the number of bronze coins decreased dramatically towards the end of the seventh century and remained at a low level for the next two centuries. 89 George Ostrogorsky, on the other hand, using the same material from the British Museum, showed that th e gold coinage, which he took to be more important than the base currency, in fact 88 See also Hendy, Studies in the Byzantine Monetary Economy 7 8. 89 A. P. Kazhdan, "Vizantijskie goroda v VII XI vv.," SA 21 (1954): 164 83.

PAGE 227

227 witnessed an important numerical increase during the same period. Furthermore, he introduced new data in the equation by analyzing two other major collections, the ones in L eningrad and Washington. 90 What is important here is not the debate per se but the fact that the evidence provided by the largest collections of Byzantine coins was brought into question as a valid argument. Shortly after that, Philip Grierson, perhaps the highest authority in Byzantine numismatics at the time, completely refuted the concept that such collections can ever project a realistic picture of the monetary mass in circulation at any given time. 91 His position, reiterated in the following decades, wa s founded on the assumption that collectors contributing to what were to became the major public collections were driven by a general desire to gather full series of issues. 92 The numerous types of solidi introduced by the emperors of the house of Heraclius thus explained the abundance of seventh century gold coins in the major collections. Although he was mainly discussing gold in his attempt to respond to the claims made by Ostrogorsky, Grierson in fact drew a general conclusion regarding the statistical r epresentativity of public collections: Les grandes collections, malgr le nombre reprsentatif de la masse montaire un moment donn. 93 What was never taken into account, however, is the large body of sixth century material, namely copper issues, available in such collections, many of the common types being represented by dozens 90 G. Ostrogorsky, "Byzantine Cities in the Early Middle Ages," DOP 13 (195 9): 45 66. 91 P. Grierson, "Coinage and Money in the Byzantine Empire 498 c. 1090," in medioevo (Spoleto: Presso La Sede del Centro, (1961), 445 46; P. Grierson, "Byzantine Coinage as Source Material," in Proceedings of the XIIIt h International Congress of Byzantine Studies, Oxford, 5 10 September 1966 (London: Oxford U niversity Press, 1967), 323 24. 92 Grierson, Circulazione monetaria," 38 39. 93 Ibid., 39.

PAGE 228

228 of duplicates. Such coins elude the parameters set by the reputed Byzantinist for the gold seri es. Constantina Katsari has recently made a similar argument regarding the representativity of museum collections for statistical studies. Her focus was on Roman provincial bronze coins and her conclusion was that museum curators did not discriminate grea tly against particular types of bronze provincial coins, although in the past they may have shown a preference for certain types of silver and gold coinages. 94 The major collections included in the present analysis have the advantage of being heterogeneou s with respect to geographical sources of origin. Each of them in fact reunites smaller collections gathered at different times and in different places, and it is reasonable to suppose that they cover the entire eastern empire, albeit perhaps unequally so. The museums usually kept records of their purchases, visible in the background of the collection. Nevertheless, it is hard to trace back the mechanics of gathering a parti cular collection. 95 It is rather a " and one is often faced with the perspective of going back in time as far as the age of Enlightenment. Famous contributors have been recently honored by scholars and more information on their life and co llections is brought to light 96 However, first hand accounts on finding places are hard to obtain even for current acquisitions given the discretion of many collectors and 94 C. Katsari, "The Statistical Analysis of Stray Coins in Museums: The Rom an Provincial Coinage," Nomismatika Chronika 22 (2003): 52. 95 P. Grierson in DOC I, xiii xviii. 96 S. Bendall, "A Neglected Nineteenth Century Numismatist," Numismatic Circular 110 (2002): 261 64; P. Grierson, Memoir on the Coin Room (Washington DC: Dumbart on Oaks Center for Byzantine Studies, 1998); C. Morrisson, "La donation Schlumberger (1929)," in Trois donations byzantines au Cabinet des Mdailles: Froehner (1925), Schlumberger (1929), Zacos (1998) ed. D. Feissel et al (Paris: Bibliot Nationale, 2001), 21 50.

PAGE 229

229 their providers, let alone for those almost a century old. 97 Another methodologica l issue involves the so called collector behavior and is related to a more significant issue, namely that of establishing whether the sample of coins in various collections is representative for the total mass of coins produced in ancient times. To what extent can we safely trust the statistical results based on specimens going to color the big picture? A few arguments presented below point to the dramatic effect on the type of material selected for this analysis. This study is devoted exclusively to copper issues, which softens the effect of selectivity. By eliminating gold coins wh ich are always more carefully selected and more rigorously arranged in a collection, we are left with a bronze series subject to a more random selection. 98 One wonders how much selectivity there could have been in the creation of the Swiss collection purcha sed for Dumbarton Oaks which amounted to over 10,000 coins, mostly copper. The collector did not keep a personal record of the coins, and therefore any suggestion that such a collection had a clear direction in terms of its 97 P. Grierson, "The Interpretation of Coin Finds," NC 5 (1965): vi. 98 With few exceptions the major private collections focused on gold issues mixed with rare silver and bronze coins. Some of the outstanding collections falling under this cat egory are the ones gathered by The William Herbert Hunt collection. Highly important Byzantine coins, I, December 5 6 1990 New York; Sotheby's, The William Herbert Hunt collection. Highly important Byzantine coins, II, Ju ne 21, 1991, New York; Nadia Kapamadji (S. Boutin, Collection N.K. Monnaies des Empires de Byzance Byzantine Gold Coins from the P. J. Donald Collection, October 11 1995 London 1995, Hugh Goodacre (Chri The Goodacre Collection of Byzantine Coins, April 22, 1986 London), Anton C. R. Dreesmann (Spink, The Dr. Anton C. R. Dreesmann Collection of Ancient Coins, Part II: Byzantine and Early European Gold Coins, July 13, 2000 London), and the collecti A Catalogue of Standard Byzantine and Dark Age Gold Coins, December 3, 1980 London).

PAGE 230

230 structure is problematic. 99 The o ften huge number of copper coins in such collections suggests quantity and randomness as a major characteristic besides the basic desire to gather as many different types as possible. 100 Furthermore, some collectors were not even specialized in Byzantine coinage. 101 Perhaps the best example in this regard is the collection Khler Osbahr from the Duisburg Museum, in which the entire Byzantine series represents less than 5 percent of the 70.000 coins collected by Dr. Khler, which included ancient, medieval, a nd modern coins. The collection is particularly strong in Greek, German, and Asian coinage. Numismatics itself was just one focus of his collecting interests, as Dr. Khler assembled a very diverse collection of jewelry and minor arts covering a huge time period, from 3000 B.C. to the modern age. Ralf Athoff, who published the catalogue of Byzantine coins, confirmed the fact that Dr. Khler had no special interest in the Byzantine coins, whose purchase was less a process of systematic selection than a need to cover this important historical period in his huge collection. 102 Each large collection contains an important number of duplicates. Doubtless some selection occurs on the part of museum curators. The large collection at Dumbarton Oaks was subjected to th e removal of the poorly preserved duplicate s when the collection was published, but the state of preservation itself is often governed by 99 P. Grierson in DOC I, xvi. 100 gold and s ilver coins: "The collection of four hundred coins has been formed over many years on the basis of academic interest, rarity, style and chronology, rather than as so frequently happens in these ns of somewhat dubious quality." 101 Many of the collectors who donated their Byzantine coins, such as E. T. Newell, de Salis, Khler, and H. C. Lindgren had only a margin al interest in Byzantine coins. 102 I am grateful to Ralf Althoff from the Kultur und S tadthistorischen Museum Duisburg for the valuable information provided on this important collector.

PAGE 231

231 statistical principles. How curators define a duplicate is also important. For the purpose of this study a duplicate s ignifies a coin of the same denomination, date, mint, officina regardless of other variations pertaining to the use of different dies. Especially when large collections are involved curators may choose to define a duplicate as a coin struck with the same d ie(s), but this rarely occurs in the case of common coppers. The unpublished collection of the American Numismatic Society (hereafter ANS) meets the criteria for a statistically acceptable sample. The collection numbers more than 5000 early Byzantine copp ers from the eastern mints dated between 498 and 616 and is primarily the result of donations en masse in the last decades. 103 In this area duplicates were never cleaned, removed, sold, or exchanged regardless of their condition, as long as they were legible The major donations of bronze coins of this period are Lindgren (1984), 104 Milrod (1984, ex George Bates), 105 Clark (1972), Wales (1983) and Newell (1944), of which only the group belonging to E. T. Newell can be characterized as a sample selected with quali ty and workmanship as the main criteria, but not necessarily rarity. 106 Therefore, most coins are in mediocre condition at best and include numerous duplicates, which suggests a high degree of randomness. Most of the 103 The ANS collection is by far the largest; by comparison, the collection at Dumbarton Oaks, which is the second largest, has only ca 2800 pieces. 104 H. C Lindgren is best known for his collection of Roman provincial coins from Asia Minor, sold at public auctions. A few hundreds of the Early Byzantine coins donated to the ANS have an identical green patina and similar dirt incrustations which suggest that they were part of a large hoard. The age structure of the group is typical for the large hoards found in Syria, containing numerous pre 538 issues, very few dated coins of Justinian and a closing date in the first decades of the seventh century. This group of coins was excluded from the statistical a nalysis of the ANS collection. 105 Bates, A Byzantine Coin Collection 106 As a collector E. T. Newell is, of course, best known for his splendid collection of Greek coins, now at the ANS.

PAGE 232

232 few purchases made by the ANS date to t he early 1970s and the acquisition records point to a price range of $1 $2.5 per coin. Only Eastern mints are included in the study, leaving out the more desirable and sought after Western mints. Sixth to seventh century Byzantine coppers from mints suc h as Constantinople, Nicomedia, Thessalonica, Cyzicus, and Antioch are the most common and cheap coins on the market since the nineteenth century. 107 Mass purchases of such cheap types are typical for major collectors. Quality is not always an issue; all c ollections under scrutiny have poorly preserved specimens even from the most common types. They include specimens on which details such as the regnal year, the officina, or the mint mark are no longer legible. A large number of smaller denominations, less of the group, which suggests quantity not quality as a criterion. The hundreds of duplicates themselves point to the largely random nature of these collections. Admittedly, as Philip Grie rson has argued on several occasions, 108 some collectors intended to gather all the known (and hopefully unknown) types that they could find. However, the mere fact that none of them was able to achieve this ultimate goal, coupled with the fact that all of t hem seem to have been very successful at gathering the same particular issues (certain regnal years, mints, officinae, etc.) points to the fact that some types were more readily available than others. The fact that different 107 As early as the mid ni neteenth century Sabatier was pricing the Justinianic follis from the East at 2 10 francs while a follis from the Western mints ranged from 20 to 50 francs. The criterion is not so much style but degree of rarity and this considerable difference in evaluat ion evinces that collectors of Byzantine coins were already having a rough quantitative image of the Byzantine coin series and were therefore prices re Byzantine Coins and their Values (1987) confirm this difference of appeal up to this day. 108 Grierson, "Byzantine Coinage as Source," 323 24; Grierson, "Circulazione monetaria," 39.

PAGE 233

233 collectors had the same success with some types and shared a similar failure in finding others indicates a fluctuation in coin production which inevitably translates into the varying numbers of coins available today. Furthermore, there is a striking resemblance between the five major c ollections in terms of structure and consequent ly, of statistical results (Figure 5 4 ). What counts in the end is the observable similarity of these collections, even when they are tested at the detail level of annual fluctuations. We do not have sufficien t information on each of the major collectors in order to make meaningful comparisons, but even so it is very unlikely that they all shared the same collecting behavior. Finally, and most importantly in many respects, the archaeological evidence confirms t he general pattern of annual fluctuation. Unfortunately, the only samples comparable in size with the large museum collections are the single finds from the province of Scythia and a number of large circulation hoards from the Near East. 109 By necessity, the analysis has to follow the nature of the evidence. The age structure of the Near Eastern hoards makes them suitable for an analysis of the second half of the sixth century, for whic h the material is abundant (Figures 5 1 and 5 2 ). On the other hand, the c oins from Scythia are less useful for a close analysis of the last quarter of the century when the region was menaced by the attacks of the Slavs and Avars, but offers a good sample for the preceding decades (Figure 5 1 ). The comparison between the collect ions and the finds 109 The most important hoards for this purpose are Chyrrus, Tell Biss, Amman, Baalbek, Khirbet Fandaqumya, Quazrin, and a number of hoards with uncertain provenance in the Near East. See R. Todd, "A Late Sixth Century Hoard from Northern Syria," NC 147 (1987): 176 82; S. J. Mansfield, "Unknown (Near East ), 1994 or Before," NC 155: 348 54; S. J. Mansfield, "Unknown (Near East), 1993 or Before," NC 155 (1995): 354 58; D. T. Ariel, "A Hoard of Byzantine Folles from Qazrin," 29 (1996): 69 76; M. L. Bates and F. L. Kovacs, "A Hoard of Large Byzantine a nd Arab Byzantine Coppers," NC 156 (1996): 165 73; Pottier, Analyse ; Mnzfunde ; R. Naismith, "A Hoard of Byzantine Copper Coins Ending with the Last Year of Maurice," NC 164 (2004): 296 99.

PAGE 234

234 from Scythia and the Near East offer s a high degree of correlation and is perhaps the crux of the demonstration regarding the randomness of museum collections. 110 Even more, Figure 5 2 clearly shows that collectors did not limit their coll ecting behavior to the classical one coin of each type The first years of the reign of Heraclius shown here by way of example suggest that the museum collections follow the pattern of the single finds and hoards and not the variety of types. Obviously, both the single finds from Scythia and the hoards from the Near East represent a type of evidence that was never subjected to selection in the hands of collectors and museum curators. Such statistical similarities indirectly ascertain the acceptable degree of randomness in the large collections under consideration. 5.3 A Comparative Approach to Early Byzantine Coin Circulation: The Balkans, Anatolia, and Syria Palestine 111 The present analysis is not an attempt to determine the absolute number of coins produ ced by a certain mint or in a certain year. The graphs highlight the fluctuations in the quantity of fresh currency produced each year and therefore it has nothing to do with calculating the entire coin population in circulation at a given date. The eviden ce of hoards shows that coins issued by Anastasius were still circulating during the reign of Heraclius. Due to factors such as loss, hoarding, wear, and state policy of withdrawing certain issues, no precise calculations can be made in this respect. 110 The spikes observable on Figure 5 1 on coins from Scythia dated from 568 to 571 is explained by the significant number of coins issued at Thessalonica, a phenome non best illustrated by Fig. 4. 111 Unpublished catalogues of the Byzantine coins from the Isparta and Bolu museums and lists of coin finds from Pisidian Antioch and Melitene, repeatedly referred to in the following section, were kindly provided by Zeliha Demirel Gkalp from Anadolu University.

PAGE 235

235 A B Figure 5 1. Percent of nummia per year of reign (Justinian I and Justin II). A) Justinian I, Constantinople mint, B) Justin II.

PAGE 236

236 A B Figure 5 2. Percent of nummia per year of reign (Maurice and Heraclius). A) Maurice, Antioch mint, B) Heraclius (61 0 617).

PAGE 237

237 Figure 5 3. Early Byzantine coin finds: major sites and local museums.

PAGE 238

238 Figure 5 4. Percentage of nummia per year of reign (498 616).

PAGE 239

239 Figure 5 5. Percent of n ummia per year (538 616)

PAGE 240

240 A B Figure 5 6. Coins per year of reform. A) Nu mmia B) Solidi

PAGE 241

241 Private and public collections, single finds and hoards need to be employed as complementary types of numismatic evidence, moving away from the sterile debate over which is the more capable of spawning an accurate reflection of coin produ ction and circulation in ancient times. Large collections, provided that they were amassed with an acceptable degree of randomness, can provide important indications about the rhythm of production. It has been often shown that site finds, if they amount t o a statistically relevant sample are useful for observing the evolution of coin circulation in time in a circumscribed geographical area. Comparative analyses of site finds in a broader region provide a better understanding of the monetary economy in a la rger unit of analysis, such as an administrative province, as it has been shown for Scythia, Pisidia, and Arabia. 112 Finally, the evidence of circulation hoards, which has been privileged by prominent Byzantinists and numismatists, usually informs us about t he circulating medium at a certain time and in a certain place. Again, comparison is needed, in the sense of the exemplary analysis done by Henri Pottier and Hans Christoph Noeske for the Near East 113 and the team of scholars coordinated by Ccile Morrisson for the Balkans, 114 in which ideally a number of contemporary hoards concealed in the same geographical area are available for study. In the next sections the discussion will be based on several chronological and thematic parameters ranging from genera l to detail: nummia / year of reign (Figures 5 4, 5 8, and 5 9 ), quantity of nummia / year of reform (Figure 5 6 ), 115 quantity of nummia 112 Marot Las monedas del Ma cellum 3 0; Gkalp, "Yalva ve Isparta." 113 Pottier, Analyse d'un trsor ; Mnzfunde. 114 Trsors. 115 A conversion into solidi based on the ratios proposed by Wolfgang Hahn (see below, n. 133) is

PAGE 242

242 year by yea r (for the period 538 616) (Figure 5 5), mints (Figures 5 10 to 5 23), and denominations (Figures 5 24 to 5 36 ). Obviously, the collections differ in size. In order to make the comparison possible, a common denominator had to be provided and therefore all the numbers are percents from a given total, e.g. within each collection, the percentage represented by the numb er of nummia from the reform period 538 542 out of the total number of reforms from 498 to 616 (taking into account the time span of each reform), or the percentage of the volume of nummia from 565/6 out of the total quantity of nummia gn, in each collection. The chosen time span, 498 616, opens with the reform of Anastasius and ends with the abrupt decline in coin circulation after 616 in several major centers of the Eastern Empire. 116 The province of Scythia provides us with a number of coins that parallels the size of the large public collections and represents the only substantial sample of coins with a secure geographical provenance and usually with a clear archaeological context. For comparison purposes, hoards and various site finds or local museum collections from the Balkans, Anatolia, and the Near East will be used throughout the following discussion, the main criterion for inclusion being the total number of Early Byzantine coins available for study (Figure 5 3 ). provided in Figure 5 6, which gives a more accurate picture of the purchasing power of the base coinage in the early Byzantine p eriod. 116 Wastage rates are sometimes included in the analysis when long periods of circulation are involved. For the methodology and applications to Roman coinage see especially Duncan Jones, Money and Government ch. 14. However, the extrapolation of mod ern wastage rates to ancient coinages remains of somewhat dubious value. Moreover, Fig. i does not reveal any clear signs of wastage for the dated series of Justinian from Constantinople (27 years). Stray finds from Scythia, which are the direct result of wastage (i.e. casual losses) should theoretically contain a higher number of coins from the early regnal years (as a result of longer circulation) than circulation hoards, which should reflect the effects of wastage at the time when the hoards were conceal ed (i.e. fewer coins from the early regnal years).

PAGE 243

243 5.3.1 The Reform of Anastasius and the Pre 538 Coinage In 498 Anastasius introduced a new system for the copper currency, one that would put an end to the crisis of the fifth century, which rendered the petty currency almost worthless. 117 However, as shown by Fig. 2a, the n umber of small module coins struck after the reform does not seem to be very high, if compared with the quantity issued after a second reform in 512. In geographical terms, a larger number of small module issues can be found in the Danube area 118 and, to an even larger extent, in a number of urban centers in Syria Palestine (Scythopolis), and especially Berytus where a unique situation can be noted. 119 In Anatolia, urban centers such as Sardis, Sagalassos, Side, Pisidian Antioch, and 117 For the reform of Anastasius and its impact see above n. 27. 118 to seventh century coin circulation 129 44; S humen (4/16): Zh. Zhekova, Moneti i monetno obrashtenie v srednovekovnija Shumen ( Sofia: Iupi Tp 2006), 65 66). The highest concentration has been recorded in Constantinople, at Sarahane and Kalenderhane (19/26): Hendy, "The Coins," 285 87; Hendy, "Roman Byzantine and Latin Coins" 196 98. These types are scarcer among finds from the Western Balkans, in Byzantine Coins from the National Museum in Belgrade (Belgrade: National Museum Belgrade, 2006), 92 96), and Al bania (3/21): H. Spahiu, "Monedna bizantine t shekujve V XIII t zbuluara n territorin e Shqipris," Iliria 9 10 (1979 1980): 366 68, while in Greece the reform had little immediate impact: R. L. Hohlfelder, Kenchreai, Eastern Port of Corinth. III: The Coins (Leiden: Brill, 1978), 63; A. Bellinger, Catalogue of the Coins Found at Corinth, 1925 (New Haven: Yale University Press, 1930), 45; Edwards, Corinth VI 121; K. Edwards, "Report on the Coins Found in the Excavations at Corinth during the Years 1930 1935," Hesperia 6, n. 2: (1937): 249; J. D. Mac Isaac, "Corinth: Coins, 1925 1926. The Theater District and the Roman Villa," Hesperia 56, n. 2: (1987), 135; M. Thompson, The Athenian Agora 66 67. 119 D. M. Metcalf and S. Payne, "Some Byzantine and Arab Byz antine Coins Obtained in Jerusalem," Numismatic Circular 73 (1965): 130 31; Sheedy, "Byzantine Period Coins," 128 30; Marot, Las monedas del Macellum, 461 64; Bellinger, Coins from Jerash 95 97; G. Bijovsky, "The Coins," 511 12; Butcher, "Archaeology of t he Beyrut Souks," 257 64. One can add a few more sites or regions with a smaller quantity of Early Byzantine coins, of which a good proportion is made up of small module coins of Anastasius. Such cases are Capernaum: A. Spijkerman, Cafarnao III. Catalogo d elle monede de la citt (Jerusalem: Pubblicazioni dello Studium Biblicum Franciscanum, 1975), 29 and 31), Fayran: Mnzfunde 708, and Mesopotamia: E. J. Prawdzic Golemberski and D. M. Metcalf, "The circulation of Byzantine coins in the south eastern fronti ers of the Empire," NC 123 (1963): 90 92. A special case is the Israel Exploration Journal 37 (1987): 15 1, table 1.

PAGE 244

244 Melitene point to a rather reduced impact of the reform in the first period. 120 To return to the special case of Berytus, almost 70 percent of the total number of Early Byzantine coins is represented by small module issues. Berytus might well h ave represented an idiosyncratic circulating micro medium, a semi closed monetary environment which might have encompassed a larger area of Phoenice, 121 but it does, nevertheless, raise an important question regarding the withdrawal of these coins from circu lation, once a new reform in 512 doubled the weight of the copper coin. As Kevin Butcher showed in his discussion of the Anastasian coins from Berytus, many of the small module specimens were found in layers dating from the reign of Justinian, which means that the small coins were still in circulation at that time. 122 It is hard to determine with any 120 Sardis: Bell, Sardis ; G. E. Bates, Byzantine Coins (Cambridge MA: Harvard University Press, 1971); T. V. Buttrey, "Byzantine, Medieval and Modern Coins and Tokens," in T. V. Buttrey et al, Greek, Roman, and Islamic Coins from Sardis (Cambridg e MA: Harvard University Press, 1981), 212. Sagalassos: S. Scheers, "Catalogue of the Coins Found in 1992," in Sagalassos II: Report on the Third Excavation Campaign of 1992 ed. M. Waelkens and J. Poblome (Leuven: Leuven University Press, 1993), 254; S. S cheers, "Catalogue of the Coins Found in 1993," in Sagalassos III: Report on the Fourth Excavation Campaign of 1993, ed. M. Waelkens and J. Poblome (Leuven: Leuven University Press, 1995), 314; S. Scheers, S. et al, "Coins Found in 1994 and 1995," In Sagal assos IV: Report on the Survey and Excavation Campaigns of 1994 and 1995 ed. M. Waelkens and J. Poblome (Leuven: Leuven University Press, 1997), 332; S. Scheers, Coins Found in 1996 and 1997," in Sagalassos V: Report on the Survey and Excavation Campaigns of 1996 1997 ed. M. Waelkens and L. Loots (Leuven: Leuven University Press, 2000), 5 (ancient Amastris): S. Ireland and S. ins in Amasra Museum," in Studies in Ancient Coinage from Turkey ed. R. Ashton (London: Royal Numismatic Society, 1996), 132. Bolvadin (vicinity of ancient Polybotos): R. Ashton, C. Lightfoot, and A. zme, "Ancient and Medieval Coins in Bolvadin (Turkey), Anatolia Antiqua 8 (2000): 183. However, see a few specimens at Amasya (ancient Amaseia): S. Ireland, Greek, Roman and Byzantine Coins in the Museum at Amasya (London: Royal Numismatic Society, 2000), 101, and Pessinus: G. de Wilde, "Monnaies au Muse de Pessinonte," Epigraphica Anatolica 28 (1997), 107; J. Devreker, "Les monnaies de Pessinonte," in Les Fouilles de la Rijksuniversiteit te Gent a Pessinonte 1967 1973 ed. J. Devreker and M. Waelkens, vol. I (Brugge: De Tempel, 1984), 211. A few specimens f ound in Anatolia are now kept in the archaeological museum in Collection. Anastasius I (A.D. 497 518) Anastasius II (A.D. 713 715)," ja u Zagrebu 30 31 (199 7 1998): 143 45, no. 11 and 36. 121 See the hoard of small module folles found at Sarafand south of Beyrut, M. R. Arguelles, "A Hoard of Small Module Post Reform Coinage of Anastasius from Sarafand," M.A. Thesis, American University o f Beirut 1976. 122 The evidence of hoards seems to point in the same direction. Several hoards containing a large

PAGE 245

245 certainty their precise relation to the Anastasian and later, Justinianic, heavy standard. An analogy with a seventh century measure might reinforce a hypothesis established in the past decades. Special marks were placed on the reformed heavy coins introduced by Constantine IV (M on half folles K on folles ) indicating that the new coins were worth twice as much as the old, smaller ones. 123 Although no such clear marks are present on the heavy coins of Anastasius, the small module issues might have remained in circulation based on the same rationale of using them for a different face value. 124 To be sure, these small and ephemeral issues remained in circulation thro ughout the sixth century as testified by hoards found in the Eastern Empire 125 Despite their small size, such coins remain outside the danger of " as the state did not have to fear that the circulation of the newly introduced heavier specime ns might be disrupted by the existence of those lighter issues, as long as their face value was halved. The reuse of late Roman, early Roman and even Greek coins, number of small module folles were found in collapsed buildings associated with the earthquake of 551. Butcher, "Archaeology of the Beyrut Sou ks," 283 86; P. Belin, "A Hoard of Byzantine Folles from Beirut," NC 165 (2005): 314 2; G. Abou Diwan, "Un trsor montaire de Beyrouth. propos de la circulation des e sicle," NC 168 (2008): 303 2 0. 123 L. Schindler, "Die Reform des Kupfergeldes unter Konstantinos IV," NZ 86 (1955): 33 35. 124 See Mecalf, The Origins of the Anastasian Currency 41 43, followed by Pottier, Analyse d'un trsor 227 30, who suggested that the countermarks often found on small module coins from the Mid dle East are a sign that the state was attempting to regulate the use of the pre reform coins. The reduction of the face value was also accepted as a plausible hypothesis by P. Grierson, Byzantine Coins (Los Angeles: University of California Press, 1982), 60, and W. Hahn and M. Metlich, Money of the Incipient Byzantine Empire (Vienna: City Press, 2000), 30. Recently, Abu Diwan challenged this common wisdom by pointing to the abnormal circulating pattern of Berytus, which raises important questions regarding the uniform implementation of the monetary reforms throughout the Empire, Abu Diwan, "Un trsor montaire," 316 17. 125 In the Balkans and Anatolia out of 36 hoards containing coins of Anastasius, 7 include small module types. The latest of these hoards, C small module coins of Anastasius, Trsors 299. In the Near East small module coins occur occasionally in large hoards ending in the seventh century such as Tel Biss, Baalbek, Khirbet Fan daqumya and "Northern Syria:" Mnzfunde vol. II; "Lebanon:" M. Kruszynski, "A Group of Byzantine Coins from Lebanon." Notae Numismaticae 3 4 (1999): 221 42. supposition that the small module series was immediatel y withdrawn in 512, Mnzfunde vol. I, 150 51.

PAGE 246

246 based on a similar size and weight is not uncommon in the large centers of the empire and is also testified by hoards containing such specimens. The Collections confirm the high proportion of coins issued in Constantinople, over 75 percent in all cases, the rest being struck at the sub metropolitan mint of Nicomedia (Figure 5 11 ). There is a fairl y balanced proportion of folles and half folles both in the Collections and in the samples found during archaeological excavations, while the folles are themsel ves well represented (Figure 5 25 ). This phenomenon shows that the divisionary system was func tional and smaller denominations were used frequently in minor transactions. The Collections, however, hardly contain any specimens of the smallest denomination, the nummus, found especially in Greece (Athens, Corinth, Kenchreai), 126 Anatolia (Sardis, Sagala ssos), 127 and Palestine 128 and to a much lesser extent in the Balkans and at the Danube border. 129 The retrieval of large numbers 126 Thompson, The Athenian Agora 66; Edwards, "Report on the Coins," 249; J. E. Fisher, "Corinth Excavations, 1977, Forum Southwest," Hesperia 53, n. 2 (1984), 245; Hohlfelder, Kenchreai 64. See also the case of Nemea where John Mac Isaac argued that minimi continued to circulate during the sixth century, J. D. Mac Isaac, Early Christian and Later Coin Finds from Nemea," in J. D. Mac Isaac and R. Knapp, Excavations at Nemea III. The Coins (Los Angeles: Universi ty of California Press, 2005), 185. 127 Bates, Byzantine Coins ; Scheers, "Catalogue of the coins found in 1992," 254; Scheers et al., "Coins Found in 1994 and 1995," 332; Scheers, "Coin s found in 1996 and 1997," 525. 128 H. Hamburger, "Minute Coins from Caesar ea," 'Atiqot 1 (1956): 115 38); Evans, The Joint Expedition 180 203; Bijovsky, "The Coins," 507 Ramat Hanadiv Excavations. Final Report of the 1984 1998 Seasons (Jerusalem: The Israel E xploration Society, 2000), 413 table 4; R. Barkay, "Roman and Byzantine Coins," in Y. Hirschfeld et al., The Roman Baths of Hammat Gader. Final Report (Jerusalem: The Israel Exploration Society, 1997), 300. See also the hoard of minimi from Gush Halav, G. Bijovsky, "The Gush Halav Hoard Reconsidered," 35 (1998): 77 folles probably found in Lebabon, M. Phillips and S. Tyler Smith, "A Sixth Century Hoard of Nummi and Five Nummi Pieces." NC 158: (1998): 316 24. A signif icant number of one nummus pieces have been found during the excavations on the Limes Arabicus primarily T. Parker, "Preliminary Report on the 1985 S eason of the Limes Arabicus Project," BASOR, Supplemental Studies 25 (1988): 171 72. 129

PAGE 247

247 of minimi accumulated in special circumstances, like the water basins at Ramat Hanadiv and Hammat Gader, or the hoards found in Palestine, Greece, Dobroudja, and in Istanbul, 130 might signal the fact that we are largely underestimating the sheer quantity of petty coins still in circulation deep into the sixth century. 131 The con trasting image offered by Sardis awaits more information coming from other centers in Western Anatolia in order to determine whether this is a particular case or a more general phenomenon. Philip Grierson explained the paucity of minimi at Sardis by the ne gligence of the excavators, 132 but recent research in Anatolia, at Melitene and Pisidian Antioch shows that minimi are generally scarce. The period 512 538 is homogeneous in many respects, largely due to a stable ratio between the gold solidus and the cop per follis most probably 1:360. 133 Figure 5 1 points to an important increase in coin production during this period, although not a continuous one, the reign of Justin I usually providing a larger number of finds than the In the larger framework of the long sixth century, 130 Gush Halav: Bijovsky, "The Gush Halav Hoard," 77 106, with a comparative discussion of the circulation of minimi in the first ha lf of the sixth century; Trsors (Greece: several hoards in Thebes, Athens, Corinth, and Kenchreai; Thasos 1977, Argos 1892 1895, Hagios Nikolaos 1935, Kenchreai 1963, Kleitoria 1933, Megara before 1884, Patras 1938, Pellene 1937, Priolithos 1979, Spata 19 82, Trype 1935, Zacha, Chersonissos. Dobrudja: Constan a 1929; Histria 1974). Several small hoards in Istanbul, Hendy, "Roman, Byz antine and Latin Coins," 271 76. 131 See also the case of Gerasa where Teresa Marot has shown that late Roman coins are still present in sixth century archaeological contexts, Marot, Las monedas del Macellum, 304. 132 P. Grierson, "The Interpretation of Coin Finds," NC 5 (1965): xi. 133 The ratio between solidus and follis MIB I 27, MIB II 14 17, and MIB III 16 A consensus is yet to be reached regarding the calculation of this ratio and different propositions have been made in the past decades: J. Proceedings of the 9th International Congress of Numismatics, Berne, September 1979 ed. T. Hac kens and R. Weiller (Louvain la Neuve: Association internationale des numismates professionnels, 1982), 731 40; Pottier, Analyse d'un trsor 252; C. Morrisson, "Monnaie et prix a Byzance du V e au VII e sicle," in Hommes et richesses 248; Morrisson and Iv e VII e sicles, 51; Hendy, Studies in the Byzantine Monetary Economy 478.

PAGE 248

248 however, the coins minted between 512 and 538 stand at a lower point than the post reform coinage of Justinian and the inflationist peak reached during the reign of Justin II. The larger quantity of coin s from Justin I has already been noticed in the Eastern Mediterranean 134 although it is hardly a general phenomenon and the evidence is still too scant to permit a conclusion in this respect. 135 The phenomenon is conspicuous in the province of Scythia, where all the major sites without exception reveal a peak reached during the reign of Justin I. 136 This is by no means characteristic for the Balkan area as a whole. The neighboring province of Moesia II offers a contrasting image with a high occurrence of coins o f Anastasius. 137 A similar contrast is found in Greece in the cases of Corinth and Athens, while in the western Balkans, there is a fairly balanced proportion of the two periods, with somewhat higher numbers for Just in I (Figure 5 8 ). 138 In Anatolia the evide nce available from Sagalassos, Sardis, Side, Amaseia, Amastris, Pisidian Antioch, and Melit ene offers a mixed picture (Figure 5 9 ) 139 and so does the 134 Grierson, "The Monetary Reforms," 296; Walmsley, "Coin Frequencies in Sixth and Seventh Century Palestine 344. 135 Butcher, "Archaeology of the Beyru t Souks," 103, fig. 75. Almost half of the sites tabulated by Butcher provide a larger quantity of coins from Anastasius without the possibility of discerning between different provincial patterns of supply. 136 and 4, wher e ten major sites are compared. 137 S. Mihailov, "Vidovete nominali v monetnoto obrashtenie na bizantiiskite provintsii Scitiia i Vtora Miziia (498 681 g.)," in Numismatic, Sphragistic and Epigraphic Contributions 281, table 4. 138 Thompson, "The A thenian Agora," 67; Edwards, Corinth VI 121; Mac Isaac, "Corinth: Coins," 135; Byzantine Coins 92 1092) iz zairke Narodnog Myzeja y Pojarevci," 11 (1988): 90 94; Jankovic, Podunavski deo oblasti Akvisa u VI i po cetkom VII veka 139 Scheers, "Coins Found in 1996 and 1997," 525; Bates, Byzantine Coins 19 26; Atlan, 1947 1967 78; Ireland, Greek Roman and Byzantine Coins" 102; Ireland and "The An cient Coins in Amasra," 132 33.

PAGE 249

249 evidence from Cyprus, at Paphos, Salamis and Curium. 140 The major Syro Palestinian sites seem to be more corre lated, with the notable exception of Berytus (Fi gure 5 8 ). Overall, the apparent contrast between neighboring areas in the Balkans and Anatolia in particular suggests local patterns of circulation rather than a controlled macro economic policy. The Collec tions suggest a slight decrease in coin production during the reign of Justinian I, prior to his major reform in 538 (Figure 5 6 ). The archaeological evidence indicates that such a phenomenon is very clear in the Balkans 141 and to a large degree in Anatolia, 142 but seems to be somewhat irregular in the Near Eastern sites, where, without a clear distribution according to provinces, we find all three possible situations the prevalence of coins from 518 527 (Pella), 143 a balanced proportion (Gerasa, Nessana), 144 and a larger number of coins from 527 538 (Caesarea Mariti ma, Hama, Antioch) (Figures 5 8 and 5 9 ). 145 Aside from these fluctuations, the Eastern provinces yield the highest volume of finds dated to the pre 538 period. This characteristic is confirmed by the st ructure of the hoards found in the area, which contain a good 140 I. Nicolaou, Paphos II. The Coins from the House of Dionysos (Nicosia: Cosmos Press, 1990), 192 204; O. Callot, Salamine de Chypre. XVI Les monnaies. Fouilles de la ville 1964 1974 (Paris: Boccard, 2004, 41 43; D Cox, Coins from the Excavations at Curium, 1932 1953 Numismatic Notes and Monographs 145 (New York: American Numismatic Society, 1959), 77 78. 141 Byzantine Coins 110 25; Edwards, Corinth VI 121 22. The situation is somewhat balanced in Albania: Spa hiu, "Monedna bizante," 368 77. 142 Bates, Byzantine Coins 28 44; Ireland, Greek Roman and Byzantine, 102 05; Ireland and 133. 143 Sheedy, "B yzantine Period Coins," 130 31. 144 Bellinger, Coins from Jerash 98 102; Marot, Las monedas del Macellum, 465 71; A. Bellinger, "Coins," in Excavations at Nessana ed. H. D. Colt (London: British School of Archaeol ogy in Jerusalem, 1962), 71 72. 145 Evans, The Joint Expedition 183 88; R. Thomsen, "The Graeco Roman Coins," in A. Papanicolaou et al., Hama, fouilles et recherches, 1931 1938. III (Copenhague: Nationalmus eet, 1986), 62; Waage, Antioch on the Orontes, 149 55.

PAGE 250

250 number of pre reform coins, even if most of these hoards were concealed after 600. 146 As a general observation the quantity of Justinianic pre reform issues depends on the influence of the mint of Antioch, which is rather insignificant in the Balkans and most Anatolian sites (Pisidian Antioch and Amaseia b eing two major exceptions) (Figure 5 14 ). According to the structure of the Collections, Constantinople wa s the most important mint during the p eriod 512 538. Its influence, however, gradually diminishe d in favor of the Antioch mint, which greatly increase d its output during the first decade of ed a secondary role, whil e Cyzicus and Thessalonica, re opened by Justin I have only a modest output at this time (Figures 5 12, 5 13, and 5 14 ). The mints issued especially folles folles particularly during the reign. Except for this latter period, the follis was struck in smaller quantities. T he role of the follis appears to have been less significant during this period and it seems that, folles which fulfilled the role of small change on the market (Figures 5 26, 5 27, and 5 28 ). This phenomenon i s less visible in the Balkans, where, with the exception of Ahtopol (Agathopolis), 147 on the Black Sea coast, and of Constantinople, 148 the urban 146 Mnzfunde vol. II (Baalbek, Khirbet Fandaqumya, Syria 1974, Khirbet Dubel, Khirbet Deir Dassawi, Rafah, Amman), Pottier, Analyse d'un trsor ; W. E. Metcalf, "A Heraclian Hoard from Syria," Museum N otes 20 (1975): 110 12; S. J. Mansfield, "Unknown (Near East)," 348 50; Naismith, "A Hoard of Byzantine Copper," 296 97 147 I. Iordanov, A. Koichev, and V. Mutafov, "Srednovekoviiat Ahtopol VI XIII v. spored dannite numizmatikata i sfragistika," Numizmatika i Sfragistika 5, n. 2 (1998): 71, table I. 148 Hendy, "The Coins," 287 95; Hendy, "Roman, Byzantine and Latin," 197 206.

PAGE 251

251 folles Especially in Scythia, the balanced proport ion between folles and folles indicates that the latter was the only fraction required on a market where the intensity of small transactions was relatively low. 149 In the western Balkans the proportion of folles wa s overwhelming, which could indicate that the severe disruption of urban life in the fifth century had long term consequences. 150 In Anatolia the most substantial evidence comes from Sardis, as usual, where the follis represented the main denomination in the period following the reform of 512, but its volume gradually decrease d in the following decades prior to 538 in favor of the follis At Amasya, Amastris, Side, Melitene and Pisidian Antioch the pattern of d enominations resembles the situation in the Balkans where the main role is played by the follis followed by the follis In the Near East we find once again a mixed picture. It can be argued that the smaller denominations are more present in the Oriental provinces, especially in Antioch where the follis is prevalent in this period, but also in other major sites like Caesarea Maritima, Nessana, and Berytus. 151 As Rammat Hanadiv, and Hamat Gader. In the last two cases most of the coins were found in the tunnel of a spring and a large bath complex, respectively, which might be less reflective of the real structure of denominations in circulation and more the habit of throwing small coins into the water as a symboli c offering. 152 In Syria II at Ham a 149 "Some Aspects," 318, table 5. 150 Spahiu, "Monedna bizantine," 366 Byzantine Coins 92 125; see also at the Podunavski deo oblasti Akvisa 66 table 3. 151 Evans, The Joint Expedition 180 88; Bellinger, "Coins from Jerash," 72; Butcher, "Archaeology of the Beirut Souks," 263. 152 See the discussion by Barkay, "The Coins of Horvat," 415 17.

PAGE 252

252 (Epiphania), in Palaestina II, at Pella, and in Arabia at Gerasa the structure resembles the one seen in the Balkans and in Anatolia, with ver y few small denominations (Figures 5 26, 5 27, and 5 28) A constant feature of the Oriental provinces is the larger role played by the Antioch mint than in Anatolia or the Balkans (Figures 5 12, 5 13, and 5 14) However, in centers like Pella, and especially Nessana, very far from Antioch, in Palaestina III, the mint of Antioch wa s less influent ial 153 5.3.2 The Post Reform Coinage of Justinian I The four year period following the reform of 538 is one of the most intriguing. It is also the only point at which the five major collections under scrutiny present a higher quantitative varia tion. Although there certainly was a dramatic increase in output immediately after the reform, we can also accept that a certain bias existed in favor of collecting the eye catching, impressively large folles of Justinian. This is highly visible in the cas es of DOC and B NP (Figure 5 5 ). Interestingly, the coin finds from Scythia, 542. In all cases, including Scythia, the numbers point to a continuous decrease in mint out put during the next two reform periods in the reign of Justinian, 542 550 and 550 565 (Fig ure 5 6 ). 154 The economy was not able to sustain a constant high output of heavy folles whose introduction in the first place must have relied on both economic and pro pagandistic agendas. 155 It is significant in that respect that the majority of coins both in the Collections and in Scythia are comprised of folles in a proportion usually hi gher 153 Walmsley, "Coin Frequencies," 337, table 4; Bellinger, "Coins," 71 72; at Nessana the influence of A lexandria, geographically much closer than Antioch, is more visible among the Early Byzantine coin finds. 154 For a possible explanation of this phenomenon see Pottier, Analyse d'un trsor 241 42. 155 402.

PAGE 253

253 than 70 percent (Figure 5 29 ). The situation changed dramatically in the secon d half of the 550s when a huge number of folles flooded the market (Figure 5 31 ). They were issued by the mint of Constantinople, but in even higher numbers by Nicomedia and Cyzicus. It is hard to determine what caused this sudden shift. It seems to corr espond to a wider set of measures taken by Justinian in the last years of his reign. According to the present information, Nicomedia and Cyzicus stopped minting folles and folles after 561 and concentrated almost exclusively on striking folles while T hessalonica abandoned its idiosyncratic denominational system and began issuing folles in 562. 156 These measures might have been caused by a need of small denominations after the market had been overwhelmed by a high number of folles for two decades. Furth ermore, the folles produced in high numbers between 512 and 538 began to be issued in lesser quantities after 538. The increased production of folles can be observed in all the regions of the Eastern Empire. At Noviodunum, on the Danube, 57 percent of the coins fr om 550 to 565 are folles ; at Tomis, on the Black Sea, they represent 75 percent; at Corinth, 73 percent; at Sardis in Lydia, almost 70 percent of the finds, and at Antioch 55 percent (Figure 5 31 ). Even when very few coins are reported for this time int erval we find folles among them. Such is the case at Capidava on the Danube, Sagalassos and Side in Anatolia, Curium and Salamis in Cyprus, Berytus, Gerasa, Caesarea Maritima, Hammat Gader, Rammat Hanadiv, Dibon in the Near East. 157 156 MIBE 56 62. 157 16; Scheers, "Coins Found in 1996 and 1997," 525; Atlan, 1947 79 81; Callot, Salamine de Chypre 44; Cox, Coins from the Excavations 78 ; Butcher, Archaeology of the Beirut Souks 268 69; D. Ariel, "The Coins," in L. I. Levine and E. Netzer, Excavations at Caesarea Maritima, 1975, 1976, 1979 Final Report (Jerusalem: Hebrew University of Jerusalem, 1986), 142; P. Lampinen, "The Coins, Pre liminary Report, 1990," in Caesarea Papers:

PAGE 254

254 The mint of Constantino ple gradually reduced its output in favor of Nicomedia, Cyzicus, and especially Antioch during the l ast reform period, 550 565 (Figure 5 17 ). Thessalonica still had a secondary role largely restricted to supplying the area of the western Balkans (Figure 5 17 ). 158 A geographic anomaly can be noted in the case of the Antioch mint: for reasons that are not clear, Antioch is extremely present in the collections of the museums in Amasra and Amasya, while closer to Antioch, at Side the Syrian mint is less well re presented. In the Near Eastern provinces, as was to be expected, Antioch plays a more important role, although still up to half the total number of coins came from the central mint in Constantinople. 159 The mint of Antioch appears to have served primarily th e needs of the city but its influence was far reaching as shown by the cases of Amasra and Amaseia. The higher presence of coins from Antioch in urban centers located close to the sea, such as Caesarea in Palestine and Amastris on the Black Sea could point to the distribution of coins through commercial activities. The monetary reform of 538 raises a number of interesting issues regarding the use and function of the large copper coins in a monetary system in which the mass of coins in circulation was up to 25 percent lighter. Even more problematic in the ed. R. L. Vann (Ann Arbor, MI: University of Michigan, 1992), 172; Evans, The Joint Expedition 186 87; Barkay, "The Coins from Horvat," 413, ta ble 4; A. D. Tushingham, 1952 53 (Cambridge MA: The American Schools of Oriental Research, 1972), 199. 158 Byzantine Coins "Monedna bizantine," 376 77. The mint is rather under represented in Greece proper and D. M. Metcalf has explained its geographical distribution by restricting its role to military expenditure at the Balkan border, Metcalf, The Copper Coinage 8. 159 D. T Ariel, "A Survey of Coin Finds in Jerusalem, Liber Annuus 32 (1982): 326; C. Morrisson, "La diffusion de la monnaie de Constantinople: routes commerciales ou routes politiques?" in Constantinople and its Hinterland ed. C. Mango and G. Dagron (Aldershot: Ashgate, 1995), 79; Walmsley, "Coin Frequencies," 337, table 4 5; Evans, The Joint Expedition 48, fig. 17; Butcher, Archaeology of the Beirut Souks ," 257 69.

PAGE 255

255 circulating scheme of the long sixth century is the role of such heavy specimens after the weight standard of the copper coin began to slide until it was finally established at half the weight of the Justi nianic large follis 160 Bad money drives out good was an economic principle well understood in early Byzantium. The reform of Constantine IV, briefly mentioned above, is a case in point. If the small module follis of Anastasius posed no serious circulating problems, the state would certainly have been interested in recalling the large coins of 538 542, either by coercion or by discouraging potential hoarding by temporarily raising their market value until they could be withdrawn from circulation. Certainly this represents only a logical, yet speculative, scenario and the actual process of withdrawing certain issues remains obscure in any detail. The complexity of the early Byzantine monetary economy should be neither under nor over estimated by adding a pre sentist flavor to its functionality. Both single finds and hoards suggest that the state had a good control over its major urban centers and was less able to impose its economic policies at the periphery. The intensive excavations at Sarahane and Kalender hane in Istanbul have yielded close to 500 coins dated 491 616 and not a single one of them was a heavy follis or a half follis of Justinian. In Antioch, out of more than 2300 Early Byzantine coins, only two folles and four half folles are dated to 538 542 Large cities where imperial mints were located, as in Constantinople and Antioch, certainly had more tight control of what circulated in their urban areas. In the Balkans, both hoards and single finds point to an abundance of such heavy coins and, more significant, their persistence until the last decade of the sixth 160 BNP I, 61.

PAGE 256

256 century. In Scythia, coins from 538 542 represent more than 10 percent of the entire group of Early Byzantine coins, while the proportion is much higher in Moesia II and in the north western Balkans, in Serbia. 161 It is interesting that the major urban centers of Scythia Tomis, Histria, and Noviodunum, yielded a smaller number of large folles while none of the four hoards found at Histria contains such coins. 162 On the fortresses defending the Danube frontier the situation is different. At Durostorum, 40 percent of the coins of Justinian are heavy issues from 538 42. 163 A small hoard recently found at Capidava contains coins up to Tiberius II, and yet one third are heavy folles of Justinian. The coins were kept in a small textile container and were found overlapped on the floor of a room destroyed by fire. The lack of intentionality allows a glimpse of an ordinary purse of coins probably handled by a soldier on the Danube frontier in the early 580 s. 164 Such examples suggest that the process of withdrawing the heavy series was more readily applicable in the major centers where state control was stricter. Nevertheless, the coin hoards from the Balkans, as a general characteristic, contain heavy specim ens as late as the 580s, as testified by such finds across the peninsula, in Greece, Serbia, Bulgaria, and Romania. 165 It is significant that, with one exception (Veliki Gradac), no such coins seem to appear in any of the hoards concealed in the 590s, a poss ible sign that the big coins of Justinian had been almost completely 161 Byzantine Coins ; Shumen museum (22.18 percent): Zhekova, Moneti i monetno obrashtenie. 162 Trsors 170 74. 163 Unpublished collection of the National History Museum of Romania in Bucharest 164 n tezaur de monede bizantine timpurii descoperit la Capidava," CN 15 (2009): 87 105. 165 Trsors ; most significant hoards are Koprivec, Zhalad, Adamclisi 1908, Athens 1908, and Eleusis 1893.

PAGE 257

257 removed from circulation by the end of the century. 166 Another phenomenon might suggest that the 590s represented a time of intensive withdrawal of heavier issues, namely the overstriking o f Maurice folles on previous Justinianic folles after the flan was trimmed to meet the demand of the new weight standard. Several public collections 167 and catalogues of site finds and hoards 168 contain such overstruck specimens. Most of them date form the ea rly 590s and correspond to the period when the Justinianic large folles disappear from hoards in the Balkans. Such a late date of withdrawal might be related to the difficulty encountered by Justin II and Tiberius II in collecting the taxes from the borde r provinces of the Balkans, which received particular mention in the legislation of 566 and 575. 169 The collection of taxes was also an opportunity to regulate the circulating mass, and a disruption of this system could have delayed the process of calling in the heavy Justinianic coinage. We may also use a later account from Theophanes Confessor who argued that the imperial treasury could no longer sustain the regular payment of the troops, so the state was forced to cut of the salaries in 587. 170 The decisio n to resize and overstrike larger issues, thus gaining 166 Trsors ; hoards ending in the 590s: Reselec, Rakita, Sofia, Hist barbaricum 167 Sommer, Katalog der byzantinischen Mnzen 59, n. 288; DOC 307, n. 33e2; BNP 185, n. 16; BMC 160 161, n. 138, 139; Ratto 51, n. 1105; KOD 111, n. 123. The ANS collection contains fourteen overstruck coins from this period, of which ten clearly show Justinianic undertypes. An even larger number of coins, of every denomination have trimmed planchets indicating a reval uation exercise. 168 Byzantine Coins 68, n. 562; Caesarea Maritima: Ariel, "The Coins," 143, n. 67; Evans, The Joint Expedition 193, n. 2472 (half follis ); Tell Biss hoard: E. Leuthold, "Monete bizantin e rinvenute in Siria," RIN 54 55(1952 1953): 39; Quazrin hoard: D. T. Ariel, "A Hoard of Byzantine Folles," 75, no. 6. 169 Les villages dans l'Empire byzantin (IV e XV e sic le) ed. J. Lefort, C. Morrisson, and J. P. Sodini ( Paris: Lethielleux, 2005), 379. 170 See the discussion by Yannopoulos, "Inflation, dvaluation et rvaluation" 129.

PAGE 258

258 additional metal and insuring the payment of troops in new coin, can be ascribed to the difficult financial situation mentioned in the written sources. In Anatolia the big coins are less presen t than in the Balkans but still represent an important proportion of the total number of Early Byzantine coins. Excavations at Side, Pergamum, and Sagalassos have yielded a number of specimens while the local museums in Bolvadin, Amasra, and Amasya also co ntain heavy folles dated 538 542. There is also variation: at Pisidian Antioch 25 percent of the coins are heavy issues while at Melitene they represent only 3 percent, to provide only the two extremes. Much like the Balkans, the hoards concealed in the 5 90s lack any large coins of Justinian. 171 There is an apparent scarcity of such coins in the Near East. D.M. Metcalf has long suggested that the post reform coinage of Justinian was not introduced in Palaestina and Arabia. Philip Grierson ascribed their sca rcity to their withdrawal from circulation, while Henri Pottier and Ccile Morrisson have pointed to the downfall in circulation between 538 and 565 and suggested that wars and natural disasters we re important factors explaining this situation. 172 More recen tly P. J. Casey attempted a closer analysis of the post reform coinage by looking at the evidence coming from site finds and hoards across the Eastern Empire. His point of departure was a written source, Procopius's Secret History in which the Byzantine h istorian claim ed that Justinian stopped paying the limitanei on the Eastern frontier. Seeking to assess the veracity of this statement by analyzing the numismatic and archaeological evidence 171 Unfortunatelly the information comes from a single major source, Sardis, where at least 4 hoards (found 1913, 1958, 1961, and 1968) ending after 590 are relevant for this discussion. Another hoard, from Anemurium in Isauria, ends in 602 and has no coins prior to 578. See Tresors 172 Grierson, "The Monetary Reforms," 296; Pottier, Analyse d 'un trsor 55; Morrisson, "La monnaie en e VII e sicles," 52.

PAGE 259

259 from Syria Palestine, Casey concluded that such circumstances may indeed explain the virtual absence of post reform coins from Palestine, but are less compelling in the case of Syria. 173 Finally, disregarding the evidence from the Balkans and Anatolia, Noeske has recently suggested that the post reform coinage was struck in limited quantities and was unsuited for the circulating medium of the Near Eastern provinces due to its heavy weight standard. 174 Although coins dated 538 542 are indeed conspicuously hard to find, some nonetheless have been reported at Jerusalem, Caesare a Maritima, Antioch, Berytus, Pella, and Nessana in six different provinces of the Near East. 175 Coins issued during 565 were more common and they they have been found in almost all excavations conducted in the region, and in a number of hoards. 176 It is thus fair to conclude that the post reform coinage did penetrate into the Oriental provinces, perhaps in smaller quantities than in the Balkans. This contrast should not be exaggerated, however, if we take into considerat ion the level of urbanization in the two regions. As noted, the major towns in Scythia yielded fewer 173 J. Casey, "Justinian, the limitanei, and Arab Byzantine relations in the 6th c.," JRA 9 (1996): 220. 174 Mnzfunde vol. I, 152 53. 175 G. M. Fit zgerald, "The Coins," in J. W. Crowfoot and G. M. Fitzgerald, Excavations in the Tyropoeon Valley, Jerusalem 1927 (London: Palestine Exploration Fund, 1929, 117; Ariel, "The Coins," 142; Waage, Antioch on the Orontes 152 54; Butcher, Archaeology of the Be irut Souks 266 68); Sheedy, "Byzantine Period Coins ," 132; Bellinger, "Coins," 72. 176 From the major site finds discussed here only Nessana and Hama failed to produce any finds from 542 565. The relevant hoards are "Northern Syria" (Pottier), Syria 1974, K hirbet Dubel, Tell Biss, Baalbek, Khirbet Deir Dassawi, see Mnzfunde II. Qazrin: Ariel, "A Hoard of Byzantine Folles," 70, table 1. "Northern Syria:" Todd, "A Late SIxth Century Hoard," 178 79. "Near East 1993:" S. J. Mansfield, "Unknown (Near East), 19 93 or Before," NC 155 (1995): 355. "Near East 2003:" Naismith, "A Hoard of Byzantine Copper," 297, and a Near Eastern hoard which includes Arab Byzantine issues as well, Bates and Kovacs, "A Hoard of Large Byzantine," 166.

PAGE 260

260 heavy coins and a tighter control of the coins in circulation can certainly be envisaged in a highly urbanized region like Palestine, for instance. Antioc h, the mint whose chief purpose was to serve the major Syrian city and its vicinity, issued coins in this period in especially high numbers s tarting from the late 540s (Figure 5 17 ), of which only a small percentage reached more distant parts of the Empir e such as the Balkan provinces. Coins minted in Constantinople and Nicomedia are extremely common among finds in the Near East and, judging by their increased output immediately after the reform in 538, it is hard to imagine that the coins were artificiall y kept out of the Eastern provinces. Doubtless catastrophic events such as the plague, the Persian invasions starting from 540, the Samaritan revolt in 555, and major earthquakes such as the one of 551 affected the circulation, but a long term disruption o f the influx of new coinage seems rather improbable. 177 The argument advanced by Casey might be acceptable for the frontier region only, but is unsuited for explaining the coin circulation in urban centers unrelated to any frontier business. As a matter of f act, only one, remote, Palestinian center, Nessana, notwithstanding his discussion of the hoards, concealed late in the sixth or early in the seventh century and consequen tly less relevant for the discussion. 178 The urban record is still decidedly thin, but in the light of the new evidence, mostly but not completely inaccessible to Casey, it is more plausible 177 The mint of Antioch ceased mint ing coins in years 14 DOC 143. For a list of the major earthquakes in Palestine see K. W. Russell, "The Earthq uake Chronology of Palestine and Northwest Arabia from the 2nd through the mid 8th Century A.D.," BASOR 260 (1985): 37 59. However, once the crises were o vercome, the mint was reopened. 178 Casey, "Justinian, the limitanei and Arab Byzantine Relations," 217

PAGE 261

261 to suggest that the heavy specimens did circulate in the area, but were more efficiently withdrawn from circulation at a later period. I have shown that the big coins disappear ed from hoards in the Balkans in the last decade of the sixth century, although the process might have started even earlier. It is hard to say if the Oriental provinces followed the same pattern, largely because the major coin hoards from this region, with the exception of Rafah and a North Syrian hoard 179 have a closing date after 595. The hoard of Rafah included a closing coin dated 573/4, but despite the early date of closing it contains no post reform coins of Justinian. The North Syria Hoard however, end ed in 584/5 and ha d 16 post reform coins out of a total of 60 pieces, namely more than 25 percent of the entire hoard. Almost half the co ins from this hoard were issued in Antioch so it might be safe to conclude that it was formed in the region and not brought from a more distant province of the Empire. The hoard found in the synagogue of Meroth in Palestine is particularly interesting for this discussion. It was found in a secret chamber where the treasury of the synagogue was kept and represents a slow and gradual accumulation throughout the sixth century and into the seventh. 180 The hoard contains 55 copper coins of Justinian of which 16 ar e post reform issues, meaning almost 30 percent of the total. Six of the post reform coins belong to the heavier standard. 181 Because of its special nature, as an open savings hoard, fresh coins were constantly fed into the treasury 179 Mnzfunde 634 39; Todd, "A Lat e Sixth Century Hoard," 176 82. 180 A. Kindler, "The Synagogue Treasure of Meroth, Eastern Upper Galilee, Israel," in Proceedings of the 10th International Congress of Numismatics, London, September 1986 ed. I. A. Carradi ce (We tteren: Cultura, 1986), 315 20. 181 I owe this information to Gabriela Bijovsky from the Israel Antiquities Authority whom I thank once again for allowing me to study the still unpublished catalogue of the coins from the Meroth hoard.

PAGE 262

262 and many were never tak en out. This is probably the reason why these heavy folles escaped the process of withdrawal. It also confirms once again the presence of the post reform coinage of Justinian in Palestine, possibly in much greater numbers than we are inclined to believe, b ased on the surviving specimens from site finds and later hoards. It seems so far that a policy of withdrawing the heavier issues was implemented in the Near East even earlier and more efficiently compared to the Balkans and even Anatolia. This would expla in the pronounced scarcity of the big coins among finds in Syro Palestinian sites, given the fact that they circulated for a shorter period. The fact that the large coins were withdrawn from circulation can also be indirectly reinforced by the unusual numb er of pierced coins. 182 Without attempting to be comprehensive, I assembled the most significant instances where pierced folles of Justinian have been signaled, both in public collections 183 and among site finds from the major geographical areas of the Easter n Empire, the Balkans, 184 Anatolia, 185 and the Near East. 186 Significantly, most of the coins we 182 Pierced coins from the following decades, 542 616, are seldom found in public collections or among site finds. An interesting case was signaled at the early Byzantine church from Khirbat al Karak, where several tombs contained holed sixth century coins possibly pointing to a habit of wearing coins as pieces of jewelry, P. Delougaz, "Coins," in P. Delougaz and R. C. Haines, A Byzantine church at Khirbat Al Karak (Chicago: University of Chicago Press, 1960), 51 and plate 46. 183 MIB I,112, plate 22; Sommer, Katalog der byzan tinischen Mnzen 38, no. 106, plate 2; several specimens in BNP DOC and in the ANS collection, the latter including a gold plated piece; Ratto 26, no. 495 and 30, no. 583; KOD 56, no. 333, plate XV; Bateson and Campbell, Byzantine Coins 13, no. 13, pl ate 2; E. Arslan, Catalogo delle monete bizantine del Museo Provinciale di Catanzaro (Catanzaro: Amministrazione Provinciale di Catanzaro 2000), 38, no. 14, plate III. 184 Byzantine Coins 150, n. 102, plate 8; Poenaru et. al, Monnaies byzantines 35, no. 206; Hohlfelder, Kenchreai 65, no. 1017, plate IV. 185 Bell, Sardis 77, no. 639; C. Mor rison, "Die byzantinischen Mnzen," in H. Voegtli, Pergamenische Forschungen 8: Die Fundmnzen aus der Stadtgrabung von Pergamon (Berlin: Walter de Gruyter, 1993), 55, no. 765, plate 8; Callot, Salamine de Chypre 191, fig. 18, no. 274. 186 Sheedy, "Byzantin e Period Coins," 132, n. 021 and 023, plate 10; Ariel, "The Coins," 142, no. 54, plate I; Metcalf and Payne, "Some Byzantine and Arab Byzantine Coins," 185, no. 48.

PAGE 263

2 63 from Pella is holed six times and was probably s ewn to a textile garment. Many were found in a clear archaeological context and therefore the hypothesis that such coins might have been found and pierced at a much later date is not plausible. The sheer number of cases itself points to a period closer to the time of their striking. No less than 7 percent of the total number of folles dated 538 542 in the Dumbarton Oaks and the ANS collections we re pierced, so this is hardly an isolated phenomenon from a later period. It is unlikely that such coins, once de monetized, would be taken out of the necklace and reintroduced in circulation. It is more probable that such large coins began to be transformed into pieces of jewelry only after the entire series was officially withdrawn from circulation. Although it is n ot entirely impossible for the two phenomena to coexist, the symbolic value of the coin turned into a pendant is much more powerful when the hundreds or thousands of similar pieces were no longer showing up in local market transactions. Furthermore, the ow ner of such a coin would have acknowledged its special nature only after Justin II had introduced a follis half its weight. The big coins of Justinian were therefore highly regarded by the common people and perhaps reminded them of an ambitious age of mili tary achievements and building programs, both lacking in the decades when such coins were probably being pierced. 5.3.3 Inflationary Tendencies and Decline (565 616) The reign of Justin II witnessed a dramatic increase in coin output with a peak reache d in the interval 570 575, after a general tendency of accretion during the first five years. The prominent peak from 574/5 might be artificially produced by the inclusion of numerous types described by Hahn as Moneta Militaris Imitativa which bear the re gnal year 10 (type MIB 89 93). The last years of reign mark a sharp downfall in coin

PAGE 264

264 production, which coincide d with the adoptio n of Tiberius as co regent (Figure 5 5 ). These fluctuations in coin output derived from the study of the Collections are parall eled by the numerous finds for Scythia, where the massive contribution of the mint of Thessalonica force d a more dramatic i ncrease until 570 (Figure 5 7 ). Numismatists have long drawn attention to the inflationist tendencies of the reign of Justin II, in d irect relation ship with the devaluation of the follis which went down from 216 folles / solidus to 525 and then 720. The huge volume of coins issued during this period is sometimes interpreted as a sign of crisis not of economic prosperity or increasing co mmercial activities. 187 prodigal policy of expenditure on warfare and buildings, as well as of demographic decline caused by natural disasters such as large epidemics and intensified seismic activity. However this was by no means a crisis of catastrophic proportions. The monetary economy remained fairly stable until 616, at least if we judge by the follis / solidus ratio, and Tiberius II Constantine was ambitious enough to attempt a return to the Justinianic s tandard. Moreover, the difference in mint output between the previous reform period (550 565) and the reign of Justin II as a whole wa s higher than the diff erence in purchasing power (Figure 5 6 ). This means that the volume of coins produced supersede d the theoretical level of inflation triggered by the devaluation of the follis This can be interpreted either in economic terms suggesting that a certain level of prosperity still existed, or in relation with the military situation of the Empire and the need to pay the army. The high level of coin output might have also been related to the policy of withdrawing the heavy coins of Justinian, which was a more or less successful 187 G. Poenaru Bordea, "Problmes historiques de la Dobroudja (VI e VII e sicles) la lumire des monnaies byzantines traites par des mthodes statistiques," in PACT 5 374 75.

PAGE 265

265 process, as we have seen. At any rate, such a procedure would have provided both the means (raw material) and the need to issue a large quantity of fresh coins. The mint of Constantinople wa s the most active in the first reform period, 565 570, covering approximately 40 percent of the total coin output. An important development wa s the im portance gained by Thessalonica and its folles issued in great numbers durin g these years (Figure 5 18 ). A few major changes occu r r ed in the second reform period, 570 578, when Constantinople, while still the major supplier, wa s closely followed by Antio ch, which increased its output probably due to the conflict with Persia. Cyzicus became more important after a period of low activity, while Thessalonica drastically reduced its output for reasons discusse d in the following section (Figure 5 19 ). More than half the coins issued during the reign of Justin II were folles The follis wa s less present, a sign that the monetary policy sustained by Justinian in his last years of reign was discontinued. The mint of Antioch alone continued to issue folles in si gnificant quantities. 188 The decline in the production of smaller denomination is considered a general characteristic of the second half of the sixth century, 189 but this is not entirely accurate. Although folles we re indeed rather scarce, the production of folles maintain ed and even surpassed the levels of the preceding dec ades (Figure 5 32 ). Their presence in urban settings, as it will be shown below, is an indication of a still vi brant monetary economy (Figure 5 32 ). There wa s a sudden influx of coins in decision to abandon the policy of regular payments sent to the northern barbarians in 188 See also the observations of Pottier, Analyse d'un trsor 186. 189 Ibid., 150.

PAGE 266

266 order to secure the border. It is very probable that such a shift implied the arrival of additional troops to be stationed in th e border fortresses along the Danube. The enlarged garrisons brought about an increased number of coins and this phenomenon is clearly visible in fortresses such as Noviodunum, Dinogetia, Capidava, Durostorum, Aquis, Viminacium, and Sirmium. 190 This is not o nly a frontier related phenomenon. Numerous finds from this period have also been reported in major urban centers, such as Corinth (and Kenchreai), Athens, and Tomis, and to a lesser degree in rural areas, 191 which gradually become isolated from the urban mo netary economy. The extensive mint output at Thessalonica explains the large number of coins in the provinces of the Balkans. Thessalonica wa s especially influent ial in the western half of the peninsula, in Greece, Albania, and Serbia 192 and to a lesser degr ee in the east, at Odurtsi, Agathopolis and in the region of Shumen. 193 Although the evidence for Anatolia is still insufficient for broad generalizations, there is strong indication of a general increase in the volume of coins during the reign of Justin I I. Apparently surprising from a geographical standpoint Thessalonica wa s a major supplier of coins at Sardis, where, at least in the first stage, 565 570, the coins 190 Podunavski deo oblasti Akvisa 66, anne x 3; 93 Srem," in Dj. Sirmium VIII (Rome Belgrade: cole Franaise de Rome Institut Archologi que de Belgrade, 1978), 181 85. 191 Bellinger, Ca talogue of the Coins 46 47; Edwards, Corinth VI 125 27; Mac Isaac, "Corinth: Coins," 135 36; Hohlfelder, Kenchreai 68 71; Thompson, The Athenian Agora 68 322 23, table 4; E. Oberlnder rural byzantin des Balkans Orientaux un essai de synthse au commencement du XXI e sicle," Peuce 14 (2003): 383 84. 192 Edwards, Corinth VI 125 27; Hohlfelder, Kenchreai 68 71; Thompson, The Athenian Agora 68 69; Spahiu, "Monedna bizantine," 378 81; Rad Byzantine Coins 132 35. See also the composition of hoards found in these areas: Trsors 193 72, table 4; Zhekova, Moneti i monetno obrasht enie 79.

PAGE 267

267 struck by the Macedonian mint cover almo st 35 percent of the finds (Figure 5 18 ). 194 This se ems to be a general characteristic of towns from western Anatolia, close to the coast, judging by the similar finds from Pergamum, Ephesus, and Side, 195 whereas further to the west the proportion dwindles, comprised 15 percent at Pisidian Antioch, while no coins of Thessalonica have as of yet been recorded at Amaseia, Pessinus, and Melitene. 196 Surprisingly, no such coins were found in the region of Bolu and Amasra close to the Black Sea, so we are still far from establishing a clear pattern. Coins from the se cond half of the reign are abundant at Melitene, which was a strategic position in the Armenian campaigns organized by Tiberius, now co emperor with Justin II. In the Near Eastern provinces the heavy influx of coins from Justin II has been often noted, esp ecially because of the contrast with the post reform period of Justinian I, less prominent among finds. The number of finds is conspicuously high at Gerasa in Arabia and in Palaestina at Pella, Hammat Gader, Caesarea Maritima and to a lesser extent at Jeru salem and Nessana. 197 Although they are fairly well represented at Antioch, Hama, and Apamea, 198 no coins of Justin II have been reported among the 199 and they have a generally weaker 194 Bates, Byzantine Coins 54 55. 195 NC (1925): 390; Atlan, 1947 81 82. 196 Ireland, Greek Roman and Byzantine 105 06; de Wilde, "Monnai es au Muse de Pessinonte," 107; De vreker, "Les monnaies," 195 96. 197 Bellinger, Coins from Jerash 103 13; Marot, Las monedas del Macellum, 472 80; Sheedy, "Byzantine Period Coins," 134 36; Barkay, "The Coins of Horvat," 299, table II; Evans, "The Joint Ex pedition," 188 90; Ariel, "A Survey of Coin Finds," 326; Bellinger, "Coins," 72 73. 198 Waage, Antioch on the Orontes 155 tude comparative," in Apame de Syrie: bilan des recherches archologiques 1 973 1979: aspects de l'architecture domestique d'Apame: actes du colloque tenu Bruxelles les 29, 30 et 31 mai 1980 ed. J.

PAGE 268

268 presence in the rural settlements from Syria, such as atal Hyk and Dhs 200 which parallels the situation observed by Ernest Oberlnder Trnoveanu in the case of the eastern Balkans. 201 This also accords with the observations made by Clive Foss for rural settlements in Syria where the archeological evidence suggests a period of decline after 550. 202 The large number of coins from Justin II at Gerasa has been described by Alfred Bellinger as the most salient feature of the Early Byzantine coin finds in this important city of the Decapoli s 203 The situation was rightly ascribed to the high presence of coins from Nicomedia, partially confirmed by the subsequent finds from the Macellum No clear explanation is given for this peculiar development. 204 Dealing with a similar situation at Pella, Kenneth Sheedy has suggested that it might reflect a new deployment of troops in the East for another episode of the war wit h Persia in the early 570s (Figure 5 8 ). 205 There is no such parallel at Antioch, but indeed at Apamea, the number of coins Balty (Bruxelles: Centre belge de recherches archologiques Apame de Syrie, 1984), 240 44; Thomsen, "The Graeco Roman Coins, 62. 199 G. Hennequin l Faraj, (Damascus: Institut Franais de Damas, 1978), 7 8. 200 T. Vorderstrasse, "Coin Circulation in Some Syrian Villages (5th 11th Centuries)," in Les villages dans l'Empire byzantin 498; C. Morrisson, "Les monnaies," in J. P. Sodini et al, Dhs (Syrie du nord) campagnes I III (1976 (Paris: Librairie Oriental iste Paul Geuthner, 1980), 279. 201 Oberlnder Trnoveanu, "La mon naie dans l'espace rural," 348. 202 C. Foss, "Syria in Transi tion, A.D. 550 750: An Archaeological Approach," DOP 51 (1997): 189 269. 203 Bellinger, Coins from Jerash 13. 204 This characteristic can be also noticed in most Syro Palestinian hoards: J. Baramki, A Hoard of Byzantine Coins," Quarterly of the Department of Antiquities in Palestine 8 (1939): 83 84; E. Leuthold, "Monete bizantine rinvenute in Cirrestica," RIN 73 (1971): 15; Mansfield, Unknown (Near East)," 350 51; Ariel, "A Hoard of Byzantine Folles," 70; Bates and Kovacs, "A Hoard of Large Byzantine," 166; N aismith, "A H oard of Byzantine Copper," 298. 205 Sheedy, Byzantine Period Coins ," 49.

PAGE 269

269 from Nicom edia is overwhelming. 206 Such an explanation is, however, weakened by a similar situation noticed in two distant sites, Amaseia and Pisidian Antioch, where although no invasions are recorded, the mint of Nicomedia ha d an unusually significant presence (Figu re 5 18 ). 207 The Collections point to a high output of pentanummia during the reign of Justin II (Figure 5 32 ). This is indeed confirmed by finds in Constantinople and Pessinus where more than 75 percent of the finds we re 5 nummia pieces; 208 at Tomis, Sardis, and Antioch they represent ed approximately one third of the total, while in Greece they cover less than 20 percent (Figure 5 32 ). 209 The cluster of small change in large urban centers points to a necessity of the market, which seems to have been less strong l y felt in small towns and fortresses and ev en less so in rural contexts. Less than 15 percent of the coins found in Scythia we re pentanummia while in Moesia Secunda they amount ed to a mere 4 percent. 210 No such coins are recorded among the published finds f rom Albania, Amaseia, Melitene, Caesarea Maritima, Pella, and Gerasa to name only the most important. 211 206 Balty, "Monnaies b yzantines des maisons," 240 44. 207 Ireland, Greek, Roman and Byzantine 105 06. 208 Hendy, "The Coins," 297 300; Hendy, "Roman, Byzantine and Latin," 208 11; de Wilde, "Monnaies au Muse de Pessinonte," 107; Devreker, "Le s monnaies de Pessinonte," 211. 209 Institutului de Arheologie "Vasile Prvan"," in Simpozion de nu Byzantine Coins 49 61; Waage, Antioch on the Orontes 156. 210 dovete nominali," 281, table 4. 211 Spahiu, "Monedna bizantine," 378 381; Ireland, Greek, Roman and Byzantine 105 06; Ariel, "The Coins," 142; Lampinen, "The Coins," 172; Evans, The Joint Expedition 188 90; Sheedy, Byzantine Period Coins 134 36; Bellinger, Coins from Jerash 98 102; Marot, Las monedas del Macellum, 465 71.

PAGE 270

270 In 579 Tiberius II, now sole ruler, attempted an ambitious reform designed to celebrate his consulship. The weight of the follis was lifted to a Justinia nic standard and the Collections as well as the numerous finds from Scythia indicate a high mint outp ut for this special series (Figure 5 5 ). His measure, no doubt popular with the masses, was short lived and most likely was never intended as a true reform meant to re establish the heavy standard of Justinian. Albeit less spectacular, the heavier folles introduced by Maurice in 602 with the oc casion of his consulship testifed to the irregular nature of these special issues. As usual, the mint of Constantino ple issued more than half the coins put in circulation, followed by Antioch and Nicomedia. The mint of Thessalonica is ranked higher than Cyzicus, probably because of the increasing military activity at the Danube border (Figure 5 20 ). The production of th e peculiar 30 nummia introduced by Tiberius II was perhaps less impressive than the Collections would let us believe. This scarcer presence in the Collections, on avera ge amounting to 10 percent of the entire number of coins att ributed to Tiberius II (Figure 5 33 ). Scythia offers a more realistic proportion, with the 30 nummia accounting for less than 4 percent. 212 In the three major regions of the Eastern Empire the influ x of coins issued by Tiberius II wa s characterized by a high degree of variation. At the Danube border it appears that the coins of Tiberius II made little impact as can be seen in the catalogue of finds from Dinogetia, Capidava, Aquis, Viminacium, and the Belgrade museum. 213 On 212 "Some Aspects," 318, table 5. 213 Podunavski deo oblasti Akvisa 66, annex 3; ntijski novac," Byzantine Coins 23, table 4.

PAGE 271

271 the western sector of the limes the coins from Thessalonica played an important role, and indeed in Greece, at Corinth and Athens, where close to 40 percent of the coin finds we re issued by the Macedonian mint. 214 The mint of Antioch be c ame more important in the Balkans, and it is possibly a sign that some troops were brought from 215 There is a generally higher number of coins of Tiberius found on the Blac k Sea coast, Histria, Callatis, and Accres Castellum being a few major examples. 216 Another characteristic of the Balkan settlements is the fact that in most cases when the coin circulation dropped during the reign of Tiberius II it never recovered in the fo llowing decades, a sign of the gradual disintegration of urban life in the area. Very few coins have been found in Constantinople, at Sarahane and Kalenderhane, 217 and a similar situation may be seen in the northern part of Anatolia judging by the coins fr om the museums in Amasya and Bolu, as well as in Pisidia at Antioch, Sagalassos and in the area of modern Isparta. 218 Sardis and Melitene, far apart on the map of Anatolia, share the same tendency and it seems so far that only Side provides a larger number of coins from Tiberius II (Fi gure 5 9 ). 219 Jumping to the island of Cyprus one notices a contrasting image: at Salamis and 214 Edwards, Corinth VI 128 29; Mac Isaac, "Corinth: Coins," 136; Thompson, The Athenian Agora 69. 215 M. Whitby, The Emperor Maurice and His Historian: Theophylact Simocatta on Persian and Balkan Warfare (Oxfo rd: Clarendon Press, 1988), 87. 216 Some Aspects," 322 23, table 3. 217 Hendy, "The Coins," 300 01; Hendy, "Ro man, Byzantine and Latin," 212. 218 No coins of Tiberius can be found among the four excavation reports from Sagalassos mentioned in n. 120. 219 Bates, Byzantine Coins 63 66; Atlan, 1947 84.

PAGE 272

272 Curium the number of coins of Tiberius II wa s conspicuously high given his short reign. 220 Considering that Caesarea Maritima in Palesti ne has yielded a similarly high number of coins from this period, we can advance a provisional hypothesis that the coins of Tiberius II circulated more intensively on sea routes. 221 The observation regarding the circulation of smaller denominations made fo r the reign of Justin II remains valid for the shor t reign of Tiberius II (Figure 5 33 ). Aside from a few urban centers such as Constantinople, Antioch, Tomis, Corinth, Pisidian Antioch, and Sardis the 10 and 5 nummia pieces become scarcely used across the Eastern Empire (Figure 5 33 ). Antioch and Constantinople remain the main mints issuing small denominations, no doubt partly because of the local needs of the two metropoleis. During the reign of Maurice the value of the copper follis remained stabilized a t 600 folles per solidus. The Collections point to a general decrease in mint output during the last two decades of the sixth century, when only two peaks reached in 589/90 and 602 resemble the quantity of coins issued by Justin II (Figure 5 5 ). The first peak owed partly to the high output of Antioch. During regnal year 8 the old type inherited from Tiberius II continued to be struck along with the new type introduced by Maurice, which increases the total number of coins from this year. The other peak, in 602, coincide d with the consulship assumed by the emperor. A special type was struck for this special occasion having the emperor represented in consular robes instead of with the usual military cuirass. A large number of coins were issued in a very short time interval, which explains why many specimens we re overstruck on previous issues. It also points to a 220 Callot, Salamine de Chypre 46 47; Cox, Coins from the Excavations 79. 221 Evans, The Joint Expedition 190 91.

PAGE 273

273 crisis of raw material for striking fresh coins, a typical phenomenon in the first two decades of the seventh century. The outstanding feature in mint activity is the high output of Antioch throughout the period, sometimes surpassing the production of Constantinople ( Figure 5 21 ). The intense military activity which characterize d the reign of Maurice is an important factor in explaining this phenomenon. The high number of troops involved in the war against Persia in the 580s and in the Balkans in the 590s increased the demand for fresh coins. The major role of Antioch even after the eastern front was closed, coupled with the unusually low output of Consta ntinople in the last years of the sixth century when the Empire was waging war against the Slavs and Avars is somewhat perplexing. Between 7 and 22 percent of the coins of Maurice in the northern Balkans, including the Danube fortresses were issued in Anti och. Especially in the western sector of the Lower Danube the coins from Antioch have been found in larger numbers. Conversely, in towns located on the Black Sea coast, such as Tomis, Callatis, Accres Castellum and Agathopolis, very few such coins have bee n found. 222 This is also true for the coins found in Constantinople, at Kalenderhane and Sarahane (Figure 5 21 ). 223 It is unlikely that the mint of Antioch was ever commissioned to e nsure the payment of troops stationed in the Balkans. It is more probable th at the coins were brought by the large number of troops transferred by Maurice after the war with Persia was brought to an end. This does not help explain the low activity of the mint of Constantinople especially between 597 and 602 (excepting the consular type in 602) 222 Aspects," 322 23, table 4; Iordanov, Koichev, and Mutafov, "Srednovekoviiat Ahtopol," 73, table 6. 223 Hendy, "The Coins," 301 05; Hendy, "Roman, Byzantine and Latin," 213 15.

PAGE 274

274 when the war against the Avars was in full motion. Michael Whitby considers that the 602 had specific military purposes, contra Theophylact who suggested t hat financial considerations were behind this decision. 224 Although multiple factors might have been at play, the low mint output of Constantinople and also Nicomedia, Cyzicus, and Thessalonica point to a serious financial crisis which must not be underestim ated, The coin finds from the Balkans concentrate on the eastern part, in the provinces adjacent to the Black Sea. The payment of the troops stationed in the Danubian fortresses seems to have been a serious problem. The numerous finds from the province of Scythia show a high level of coin loss for the regnal years 5 and 10, which seem to coincide with the distribution of the quinquennial donativa to the troops. One would have expected a similar peak in 597 as well, which is hardly the case. 225 Its absence wa s due to the low activity of the mints of Constantinople, Nicomedia, Cyzicus and Thessalonica, already mentioned, and coincide d with some serious military setbacks at the Danube fronti er, which was menaced by the Avars who reached as far as Tomis, the capital of Scythia, besieged in 597 598. 226 The low ebb in coin production during the last five years of the century indicates that military payments in the Balkans 224 Whitby, The Emperor Maurice 165 69. 225 "Some Aspects," 311. 226 A. Madge aru, "The Province of Scythia and the Avaro Slavic Invasions (576 626)," Balkan Studies 37, n. 1 (1996): 50.

PAGE 275

275 were delayed; discontent was certainly building up in the frontier garrisons and would eventually turn into rebellion in 602. 227 Especially the mint of Thessalonica, whose coinage was a major source of payment for the troops stationed on the Danube, reduced its output dramatically during the last years of the sixth century. The activity of this mint has been a debated issue in the last decades and it is primarily connected with the dating of the siege of Thessalonica. 228 Figure 5 7 represents a comparison between the coins from the Collections(336 coins) and the finds from Scythia (248 coins). The trend is clearly similar, while the peculiarities of Scythia are marked by the high peaks of 568/9/70 and 574/5, which are consistent with the situation in Serbia: the collection of the Nat ional Museum in Belgrade provides forty six specimens from Justin II of which eighteen are dated 569/70 and thirteen, 574/5. 229 The year by year fluctuations bring forth even more interesting observations about the Thessalonican mint output after 578. D. M. Metcalf maintained that 580 was a critical moment when the mint activity was virtually paralyzed. 230 The graph below around year 580, although a few specimens are ava ilable in the Belgrade collection as well as in Scythia. While this might be true for the Balkans (579 582 provides a striking difference in the comparison chart), the major collections provide 6 specimens from 227 See above n. 55. 228 e Rome 87, n. 1 (1975): 459 622; D. M. Metcalf, "Avar and Slav Invasions into the Balkan Peninsula (c. 575 625): The Nature of the Numismatic Evidence," JRA 4 (1991), 142. 229 Byzantine Coins 132 35. 230 Metcalf, "Avar and Slav Invasions," 142.

PAGE 276

276 580/1 and less for the next three years. Ther statement is true as a local feature and not necessarily a problem of mint output. that the mint activity at Thessalonica virtually ceased in 585/6, and yet we find four specimens in the Collections and no less than five (only single finds included) in Scythia. 231 century, 597 600, and is a more general phenomenon of the monetary economy. The statistic al value of coin finds of Maurice in Anatolia is usually double the value e stablished for the Balkans (Figures 5 8 and 5 9 ). The frequent invasions and the general devastation of the Danube provinces were certainly among the major reasons for this striking difference. Much like in the Balkans, however, the numerous coin finds from urban sites like Sardis, Amaseia, Pisidian Antioch, Side, and Malatya usually form a continuous sequence until 595. 232 The scarcity of coins from the later years and the similarity with the Balkans force us to conclude that economic reasons affecting the mint output should be held responsible for this situation and not military activity, pervasive in the Balkans, but almost inexistent in most of Ana tolia. Another distinctive characteristic was the significant influence of the Antioch mint in Anatolia. At Amaseia and Melitene around 50 percent of the coins were struck in Antioch, while the average for most of the towns for which we have sufficient inf ormation wa s more than 20 percent. 231 232 Bates, Byzantine Coins 67 78; Ireland, Greek, Roman and Byzantine 106 07; Atlan, 1947 Side, 84 86.

PAGE 277

277 Figure 5 7. Thessalonica mint (565 602) (percent of coins per year).

PAGE 278

278 Two prominent exceptions are Sardis and Amasra where the mint of Antioch wa s less significant or not present at all. In both towns, however, the c oins from Thessalonica represent ed an important proportion of the group, suggesting different channels of coin distribution in Anatolia, not necessarily ba sed on geographic location (Figure 5 21 ). 233 The collection of the archaeology museum in Bolu (ancient Claudiopolis), in the ancient province of Bithynia, far from Antioch, has no coins of Maurice from Thessalonica, but many from Antioch, amounting to 30 percent of the total. This is a general characteristic of the collection for the entire sixth century. J umping to the island of Cyprus we notice a combined influence of Antioch and Thessalonica at Salamis and Curium, no doubt because of the maritime dimension of both the towns and the mints. 234 In Syria Palestine the proportion of coins issued by Maurice wa s s lightly higher than in Anatolia. Antioch itself provide d a high number, due to the presence of the mint in the city and its influence wa s also felt in the vicinity, if less overwhelming, at Hama, Dhs, Apamea, and atal Hyk. 235 Southward, on the coast, at Berytus and Caesarea, a large number of coins from Antioch we re recorded among finds, perhaps a sign of commercial activities and also in towns from the Palestinian inland, such as Jerusalem and Nessana. 236 The large coin hoards found in the Near Eastern provinces testify to the fact that Antioch played a more important role in this region, compared to 233 Ireland and a," 124; Bates, Byzantine Coins 72 73. 234 Callot, Salamine de Chypre 48 51; Cox, Coins from the Excavations 80. 235 Waage, Antioch on the Orontes 159 160; Thomsen, "The Graeco Roman Coins," 62; Morrisson, "Les monnaies," 279; Balty, "Monnaies byzantines d es maisons," 240 44; Vorderstras se, "Coin Circulation," 498 99. 236 Butcher, Archaeology of the Beirut Souks ," 271 73; Ariel, "The Coins," 142 43; Evans, The Joint Expedition 193 94; Ariel, "A Survey," 326; Bellinger "Coins," 73.

PAGE 279

279 the Balkans and Anatolia. 237 T he decline in coin circulation wa s now felt in some previously prosperous towns, most importantly at Pella and Gerasa. Kenneth S heedy has explained this situation by the impoverishment of the two centers, 238 while Alan Walmsley has suggested that state consignments ended in this period. 239 It is hard to make any generalizations at the scale of the entire province of Arabia or Palaesti na II since the coin samples for most settlements are too small to observe any clear tendencies in coin circulation. It should be noted, however, that in most cases finds from Maurice Tiberius are present. The lower denominations continued to be struck in limited numbers and their use was generally restricte d to the urban economy (Figure 5 34 ). Both the Collections and folles which remain ed abundant only among the finds from Constantinople and to a lesser degree at Sardis and Ephesus. 240 The follis is found in a wider variety of settlements, in major urban centers such as Tomis, Pisidian Antioch, Salamis, and Antioch, and occasionally in rural settlements like Dhs, in Syria. 241 A new increase in coi n output may be observed during the reign of Phocas, which is another reason to reconsid er the merits of his reign (Figure 5 6 ). Historians in the past have relied perhaps too heavily on written sources biased against Phocas to 237 Mnzfunde vol. II; Tr sors 238 Sheedy, Byzantine Period Coins 52. 239 Walmsley, "Coin Frequencies," 345. 240 Hendy, "The Coins," 301 05; Hendy, "Roman, Byzantine and Latin," 213 15; Bates, Byzantine Coins 67 78; Milne, "J. T. Wood's Coins from Ephesus ," 390. 241 Isvoranu and Poenar u Bordea, "Monede bizantine de la Tomus," 153, table 3; Callot, Salamine de Chypre 51; Waage, Antioch on the Orontes 159 60; Morrisson, "Les monnaies," 279.

PAGE 280

280 describe his reign in overly negative terms. The Collections prove that his coinage was abundant, although it must be noted that a good proportion of his copper coinage wa s made of overstrikes, usually on coins of Maurice. This is not so much a case of damnatio memoriae as a direct r esult of a shortage of copper building up in the course of the sixth century as a consequence of inflation, hoarding, and casual loss. Chiefly folles were overstruck and mainly in Constantinople, Nicomedia, and Cyzicus. Based on the evidence of the Collect ions, the overstruck group represents on average c. 25 percent of the total, while the carefully published finds from Sardis reveal that c. 22 percent of the folles of Phocas we re overstruck. 242 The mint of Antioch is excluded from these calculations because it rarely overstruck its issues. Except for the collection in Paris 243 there is a general tendency of decline in mint output in the last regnal years, probably because of the turmoil created by the Heraclian revolt. The mint of Antioch continue d to be th e second most important after Constantinople and its output was related to the new offensive initiated by the Persian king after the deposition of Maurice. Surprisingly, the two major mints we re closely followed by Cyzicus, which became extremely active af ter having a minor role th roughout the sixth century (Figure 5 22 ). The military conditions in the Balkans worsened after the rebellion of 602. Although the thesis of the collapse of the Danube limes in 602 is no longer tenable, 244 242 Bates, Byzantine Coins 67 78. 243 The collection of the Bibliothque Nationale shows a prominen t peak in 609/10, apparently inexplicable (Figure 5 belonged to the collection Schlumberger; most of them were minted in Antioch and were purchased in Aleppo by the French s chola r and might be part of a hoard. 244 A. Barnea, "Einige Bemerkungen zur Chronologie des Limes an der unteren Donau in sptrmischer Zeit," Dacia 34 (1990 ): 283 90.

PAGE 281

281 numerous fortifications, p articularly on the western sector, were severely affected. There is a marked difference in coin circulation among the settlements on the border, such as Capidava, Novae, Aquis, and Viminacium, 245 where few or no coins of Phocas were found and towns further a way from the military operations, such as Tomis, Accres Castellum, Corinth, Athens, and Constantinople itself, where the volume of finds mark ed a visible increase compar ed to the reign of Maurice (Figure 5 8 ). 246 A very similar tendency may be observed in An atolia, where most of the urban settlements have yielded a large number of f inds from Phocas (Figure 5 9 ). One major exception is Melitene situated close to the front line after the Persians had occupied the main strategic towns in Upper Mesopotamia. The m int of Antioch, less visible in the Balkans, cover ed more than 30 percent of the coins found in Anatolia, an important increase compared to the reign of Maurice. As expected, the finds from the Near Eastern provinces show ed an even more pronounced influenc e of Antioch, especially at Gerasa and Caesarea, probably itself a sign of the threat p osed by the Persian armies (Figure 5 22 ). 247 However, the influence of Antioch stop ped being so pervasive in the Near East. At Hama, Jerusalem, and Nessana, where Antioch had always been an important 245 23, table 4; K. Dimitrov, "Poznorzymskie i wczesnobizantyjskie monety z odcinka iv w Novae z lat 294 612," Novensia Podunavski deo oblasti Akvisa 246 G 23, table 4; Bellinger, Catalogue of the Coins 46 47; Edwards, Corinth VI 130 31; Fisher, "Corinth Excavations, 1977, Forum Southwest," Hesperia 53, n. 2 (1984), 245; Mac Isaac, "Corinth: Coins," 136 ; Thompson, The Athenian Ag ora 69 70; Hendy, "The Coins," 305 08; Hendy, "Roman, Byzantine and Latin," 216 18. 247 Bellinger, Coins from Jerash 114 16; Marot, Las monedas del Macellum, 481 83; Ariel, "The Coins," 143; Evans, The Joint Expedition 195 96.

PAGE 282

282 supplier of fresh coins, no such finds have been reported. 248 This is not only a matter of distribution but also a problem of coin supply, since in many towns from Syria Palestine, unlike what we have seen in the Balkans and Anat olia, the volume of fresh coins stagnate d or decrease d during the reign of Phocas (Figure 5 8 ). The extreme scarcity of coins from Thessalonica, which had been a minor but steady supplier of coins since the reign of Justin II, wa s possibly another sign of decline. The low number of smaller denominations adds to this picture of downfa ll in coin circulation (Figure 5 35 ). Besides Antioch no other urban center in Syria Palestine provides such issues. 249 The rare hoard of small change found in Aleppo could origin ate from the circulating medium of the great Syrian metropolis. 250 By comparison, Tomis and Odartsi in the Balkans, 251 Sagalassos, Pisidian Antioch, and Sardis in Anatolia, 252 and Salamis in Cyprus 253 continued to receive lower denominations, folles folles It is often thought that the long reign of Heraclius mark ed the end of Antiquity. The empire lost its eastern provinces to the hands of the Arabs and its influence in the Balkans under the repeated attacks of Slavs and Avars. By necessity, this comparative drop ped in Scythia, our main element of comparison with the Collections. The first years 248 Thomsen, "The Graeco Roman Coins," 62; Ariel, "A Survey of Coin Finds;" Bellinger, "Coins," 73 74. See, however, the group of coins purchased in Jerusalem in 1963 in which 30 percent of the coins of Phocas were minted in Antioch, Metcalf and Payne, "Some Byzantine and Arab Byzantin e Coins," 209. 249 Waage, Antioch on the Orontes 161 62. 250 S. J. Mansfield, "A Hoard of Twenty Byzantine Copper Coins," NC 163 (2003): 354 55. 251 Isvoranu and Poenaru Bordea, "Monede bizantine de la Tomis," 153, table 3. 252 Scheers, "Catalogue of the Coins F ound in 1993," 314; Bates, Byzantine Coins 88 89. 253 Callot, Salamine de Chypre 52 53.

PAGE 283

283 of his reign were characterized by an abundant coinage with a hig h peak reached in 612 614 (Figure 5 5 ). This coincide d with an important change in iconography, the frontal bust of the emperor being replaced with the standing figures of Heraclius and his son Heraclius Constantine. One could argue that the successive chan ges in iconography we re the reason why the Collections possess so many coins from this time interval. This is not the case, however, since the abundant material from Anatolia and Syria shows a very simi lar peak in the same years and wa s, of course, unaffec ted by selection in the hands of co llectors or museum curators (Figure 5 2 ). 254 Even from these early years of reign, the complexity of the early Byzantine monetary system began to break down slowly. The mint of Antioch was closed, never to be reopened for r egular issues, while the mints in Cyzicus (615/6) and Nicomedia (618/9) were temporarily shut down. The multi denominational system ceased to be folles and concentrated c. 90 percent of their activity on the pr oduction of folles (Figure 5 36 ). Over 90 percent of the coins from the first six years of reign are overstruck, especially the ones belonging to the 2 nd type 255 The conversion of large quantities of coins issued by Maurice an d Phocas affects the quantitative estimation for these two reigns. To give one example, the number of undertypes of Phocas in DOC would increase by 30 percent the total number of folles issued during the reign of Phocas. The Balkans witness a severe downfa ll in coin circulation in the second decade of the seventh century. Leopard spots of Byzantine control usually located around urban 254 Bell, Sardis 82 95; Bates, Byzantine Coins 95 109. 255 DOC II/1, 226.

PAGE 284

284 centers from the Black Sea coast or Danubian fortresses still held by the Empire continued to receive fresh coins. With the exception of Durostorum, the number of coins found at Capidava, Sacidava, Novae, and Viminacium are too meager to represent position at the Danube. 256 Urban life continued its course to some extent at Tomis and Callatis and especially at Corinth and Athens, to name the most important centers (Figure 5 8 ), 257 but later in the century the eastern Balkans were menaced by a new and long lasting enemy of the Byzantine state, the Bu lgars. Most of the coins found in the Balkans were issued either in Constantinople and Nicomedia, while Cyzicus and Thessalonica have a negligible presence (Figure 5 23 ). As expected, most of them are folles over 80 percent on average, and half folles ; folles are lacking, even among finds from Constantinople (Figure 5 36 ). In Anatolia we encounter a totally different distribution The coins of Heraclius are among the most common Early Byzantine coins found during excavations or in local archaeology museu ms. In statistical terms they usually represent between 20 and 40 percent of the total number of finds, which is in sharp contrast with the picture offered by the Balkans. Excavations at Sardis and Side yielded a particu larly high number of finds (Figure 5 9 ), while the coins preserved in the Bolu and Isparta museums point to a 256 E. Oberlnder Trnoveanu, Monnaies byzantines des VII e X e sicles dcouvertes a Silistra dan CN 7 (1996): 97 to Seventh Century Coin Circulation," 140 ; Dimitrov, "Poznorzymskie i 257 23, table 4; Edwards, Corinth VI 131 32; Mac Isaac, "Corinth: Coins," 136; Thompson, The Athenian Agora 70.

PAGE 285

285 massive influx of coins in the respective areas. 258 It should be noted however that the abundance of coins in these early years predate d the Persian invasion, which initially affected the south eastern part of Anatolia but soon had an impact on the Byzantine heartland where the Persians sacked Caesarea, Ancyra, and Sardis and took the island of Rhodes. 259 The abundance of early Heraclian issues is most striking in the islands, in Cyprus, 260 at Salamis, Curium, and Paphos and on Samos 261 in the Aegean, where an impressive number of coins have been recovered from the Tunnel of Eupalinos, among which some rare folles The developments in Cyprus have been ascribed to an increased strategic impor tance of the island after Antioch was occupied by the Persians, which might also explain the ephemeral presence of an official mint on the island. 262 The first decades of the seventh century brought an unprecedented series of invasions led by the Persians a nd later Arabs which sealed the fate of the Byzantine provinces in Syria Palestine. The increased number of hoards testifie d to the growing insecurity in the area after 602. 263 Site finds, however, provide us with a mixed picture. 258 Bates, Byzantine Coi ns 95 109; Bell, Sardis 82 95; Atlan, 1947 88 92. 259 C. Foss, "The Persians in Asia Minor and the End of Antiquity," The English Historical Review 90 (1975): 721 47. 260 Callot, Salamine de Chypre 54 75; Cox, Coins from the Excavations 80 81; Nicolaou, Paphos II 194 99. 261 U. Jantzen, Die Wasserleitung des Uupalinos. Die Funde (Bonn: Rudolf Habelt GmbH, 2004), 156, 160. 262 D. M. Metcalf, "Monetary Recession in the Middle Byzantine Period: The Numismatic Evidence," NC 161 (2001), 135. 263 The relevant hoards are Khirbet Dubel, Tell Biss, Baalbek, Khirbet Fandaqumya, Khirbet Deir Dassawi, Cyrrhus, "Syria", and Deir Dassawi: Mnzfunde vol. II. Quazrin: Ariel, "A Hoard of Byzantine Folles," 69 76. "Northern Syria": Mansfield,"A Hoard of Tw enty Byzantine," 354 55. "Lebanon": Kruszynski, "A Group of Byzantine Coins," 221 42, and probably also three hoards with uncertain Near Eastern provenance buried after 602, Mansfield, "Unknown (Near East)," 348 54; Mansfield, "Unknown (Near East)," 354 58 ; Naismith, "A Hoard of Byzantine Copper," 296 99; For a recent catalogue of hoards from Palestine, see M. Waner and Z. Safray, "A Catalogue of Coin Hoards and the Shelf Life of Coins in Palestine Hoards during the Roman and Byzantine Periods," Liber Annuu s 51 (2001): 305 36.

PAGE 286

286 Relatively few early coins of Heraclius found their way into Antioch and Apamea, 264 the two major cities of the region, the finds being two or even three times fewer than in Anatolia, somewhat resembling the situation encountered in the northern Balkans. Surprisingly, at Hama 265 the fin ds we re much more numerous and correspond with an unexpected period of reconstruction late in the sixth century. 266 In Palestine we notice a sensible increase in coin finds, at Caesarea, Jerusalem, and Pella, 267 but they are by no means characteristi c to the region as a whole (Figure 5 9 ). 268 5.4 Conclusion: Money in the Early Byzantine World The statistical validity of large collections of Early Byzantine coins can no longer be overlooked. The comparison of five major collections with the numerous finds from the province of Scythia and the hoards from the Near East has revealed a number of quantitative similarities, which need to be addressed. The purpose of analyzing museum collections from a statistical perspective is not to provide us with absolute figures. Unfortunately the statistical tools often used in numismatics are far more sophisticated and precise than the sampled evidence, which is most of the time fragmentary and problematic. There are too many lacunae in our knowledge of the monetary policies con ducted by Early Byzantium to attempt any definitive propositions. 264 Waage, Antioch on the Orontes 162 64; Balty, "Monnaies byzantine des maisons," 240 45. 265 Thomsen, The Graeco Roman Coins," 62 63. 266 Foss, "Syria in Transition," 259. 267 Evans, The Joint Expedition 197 98; Ariel, "The Coins," 326; S heedy, Byzantine Period Coins ," 139 41. 268 Few or no early Heraclian coppers have been reported among the fairly large number of finds from Hammat Gader, Chorazin, Tel Jezreel, and Samaria, see Barkay, "Roman and Byzantine Coins," 279 300; G. Kloetzli, "Co ins from Chorazin," Liber Annuus 20," (1970, 367 369); T. S. N. Moorhead, "The Late Roman, Byzantine and Umayyad Periods at Tel Jezreel," Tel Aviv 24, n. 1 (1997), 162 63; W. J. Fulco and F. Zayadine, "Coins from Samaria Sebaste," Annual of the Department of Antiquities Jordan 25 (1981): 221 23.

PAGE 287

287 The nature of the evidence and the inherent methodological limitations are an invitation to caution. 269 Any statistical results will need to be confirmed and re confirmed by future evidence be fore attempting any conclusive remarks. Relative fluctuations can, however, be discerned at this point and the large number of copper coins in the major museum collections offer a solid base on which to re construct the rhythm of mint output and from which to draw a number of general remarks. Accounting for variation in the volume of output is of course insufficient and much more needs to be done in the realm of interpretation, of explaining such fluctuations. Such an understanding cannot be accomplished on ly by studying the de contextualized coins from the public collections. Archaeological excavations in the Balkans, Anatolia, and the Near East offer the most promising perspectives for understanding regional patterns in a comparative fashion. Nonetheless, the high correlation of the major collections of Early Byzantine coins remains instrumental for a better understanding of annual coin production. Where significant site finds are available they should be analyzed against this pattern and if anomalies (i.e. local particularities) are spotted they need to be explained within a geographical and historical framework. Many interpretations based on political/military events need to be reassessed, because abnormal levels in coin circulation can not be properly d etected and understood without basic knowledge of the normal pattern. It will soon become apparent that low points on the statistical curve of a region are very 269 See A. S. Robertson, "The Accidents of Survival," in Proceedings of the 10th International Congress of Numismatics, London, September 1986 ed. I. A. Carradice (Wetteren: Cultura, 1989), 315 20, for a methodologic al discussion and an invitation to caution in the case of Roman coins, which could be easily extrapolated to the Byzantine period as well.

PAGE 288

288 often reflections of coin production and distribution at the center and less the result of pr ovincial developments alone. 270 I have shown in the previous sections that, in spite of the still insufficient evidence, the analysis can be undertaken at the inter regional level, which is undoubtedly the most appropriate course of action, enabling us, at least provisionally, to examine, understand, and explain regional peculiarities and different levels of monetization and economic integration. Studying coin circulation in a single province without reference to the circulating medium in other corners of th e Empire can lead to false generalizations and unreliable interpretations of the numismatic material. The coin finds from a major urban site will not inform us sufficiently about the coin circulation in the whole province, while the finds from a province w ill not be necessarily relevant for an entire region. Therefore, I have tried to paint, perhaps in overly broad strokes, a comparative tripty ch of the Early Byzantine coin circulation, with one panel devoted to the Balkans and two complementary ones for An atolia and Syria Palestine in the hope that future studies will soon correct and improve this provisional effort. The broad outlines drawn for Anatolia, and to a certain extent for the Near East, are subject to change as new information surface s At least in the northern Balkans the current body of evidence is large enough to e nsure the stability of present analyses although, of course, at the level of detail the overall scheme will certainly require minor adjustments. The common features observed in many u rban centers of the Near Eastern provinces are likely to endure, although new data is expected to color the grey areas in the big 270 The activity of provincial mints should be part of the general explanation regarding local particularities. Indeed the mint of Thessalonica will heavily influence the coin distribution in the western Balkans, while the mint of Antioch will have a similar effect in Syria (Figures 5 10 to 5 23). Given the different channels of distribution, towns from Syria and Macedonia will certainly differ to some extent in their circulation patterns.

PAGE 289

289 picture and perhaps to bring more homogeneity in what seems like a very diverse landscape. Anatolia is by far the most sensiti ve region to future developments and constitutes the most promising avenue to test both the uniting and the distinctive features of the coin circulation in the major geographical units of the Eastern Empire. There can be no discussion about the function o f early Byzantine coins beyond the frontier before a proper understanding of its role in the Byzantine economy as well as the reflection of that role in finds from different provinces, and indeed, various frontier regions of the empire, as illustrated by t he previous sections. Although the time honored debate surrounding the nature of the Roman and Early Byzantine economy has not been settled, an important observation seems to have gained important ground in the past decades, namely the definition of the ec onomy in terms of a mo s aic of interconnected markets. Embracing such a proposition also entails a redefinition of terms such as center and periphery, to the extent that each market can be described in such terms. At the macro level we can still talk ab out Constantinople as the center draining many of the economic resources of the empire and the militarized frontiers which placed an overwhelming pressure on state finances. Nevertheless, zooming in on a specific province or region occasions a small scale reflection of this picture. The coin finds from the provinces of the eastern Balkans, some of them bordering on both the Black Sea and the Danube, reveal the existence of local centers and peripheries. Large coastal towns such as Tomis, Histria, and Varna were better connected to the Aegean and by extension to the Eastern Mediterranean world. Perhaps as a direct consequence of this connectivity such urban centers had a much more developed market reflected in folle s ) found accidentally or during

PAGE 290

290 systematic excavations. The same can not be said about military settlements on the Danube characterized by the predominance of higher denominations ( folles and folles ). Similar highly monetized centers may be identified al ong the coasts of the Aegean and the Mediterranean as well as in the hinterland at reasonably close distances to the sea. The latter were sometimes both large towns and administrative centers (e.g. Scythopolis, Sardis), but in all these cases the high perc entage of small denominations of the follis is an indication of low value market transactions. The existence of separate markets, sometimes in isolation, is clearly suggested by the case of Berytus, whose monetary economy is overwhelmingly dominated by sma ll module folles of Anastasius, which otherwise disappeared from circulation in most other regions. The fact that Berytus is a coastal site, rather than a remote landlocked town, makes this case even more remarkable. Although connected to the larger Medite rranean exchange system as testified by other categories of finds, Berytus kept an idiosyncratic coin pool for reasons which have not been sufficiently explored. At a larger scale, it resembles the circulation of large module coins of Justinian in the Balk ans until late in the sixth century, at a time when they were conspicuously absent in centers from Syria Palestine. Different patterns of coin circulation in different centers of the Mediterranean world do not necessarily bespeak the lack of integration or contact, but rather a deliberate policy which may reflect both local decision making and larger state policies. To be sure, the presence of the military and the duality between large civil settlements and military fortresses had an important impact on the monetary economy, the quantity and type of coins in circulation. Unfortunately, we know more about the

PAGE 291

291 vertical circulation of coinage to and from the imperial fiscal offices in the form of taxes and salaries than we do about the horizontal transmission b etween regional economic centers and their circumscribed peripheries, either rural settlements or military fortresses. When such information is available, it becomes clear that we are dealing with different degrees of monetization depending on the type of settlement and we cannot escape the conclusion that outside the confines of large urban centers, exchange facilitated by coinage would have been one, often not the dominant, manner in which exchanges took place. 271 In what concerns the monetary economy of fr ontier regions our best source of information is the Danube frontier where we encounter the largest number of excavated sites and, unsurprisingly, the highest density of coin finds. Recent research on the Limes Arabicus has revealed an unusually low number of sixth century coin finds, while sites in northern Syria and eastern Turkey have been insufficiently explored to provide a clear picture of the economic conditions at the border with Sassanid Persia. The economy of the Balkans in the sixth century as we ll as the nature of coin finds point to an artificial coin circulation, far from the monetary economy usually invoked by numismatists and historians. 272 Coin finds from hill top sites of the Danube valley reflect military payments rather than the volume of m arket exchange, although the latter is notoriously hard to quantify. Given the collapse of large scale agricultural production in the Balkans, state controlled annona was the main form of supplying the Danube frontier. Provinces of the heartland included i n Quaestura Exercitus the peculiar 271 For the use of coins in rural settlements, see especially the essays in Lefort et al., Les villages dans l'Empire byzantin 272 For the economic conditions, see Poulter, "Cataclysm on the Lower Danube," 223 53.

PAGE 292

292 administrative unit created by Justinian, had the task of sustaining the frontier provinces bordering on the Black Sea. On the other hand, soldiers received part of their regular salaries as well as irregular payments su ch as the quinquenial donativum in bronze coin, or at least had the option of obtaining copper coins in exchange for gold, an operation facilitated by official money changers. It is, therefore, safe to assume that some low value transactions took place, bu t they seldom involved the use of low denominations and they were probably erratic in nature rather than part of a coherent system fitting the typical town countryside model. It may be that sites on the Danube had access to a wider variety of goods cheaper to transport on the river and even to imports channeled through larger towns on the Black Sea coast, as evidenced by the presence of imported ceramics. Some of these transactions might have involved the use of bronze coins. All things equal, the striking resemblance between the statistical curve of site finds from well researched provinces of the Balkans such as Scythia and Moesia Secunda, points to constant military payments including five year spikes which can be ascribed to quinquenial donativa This si milar pattern of coin loss reveals the insularity of monetary exchanges in the northern Balkans and the low speed of coin circulation. The fact that most coins issued in Carthage have been found in settlements from the Black Sea coast testifies to the exis tence of long distance contacts and confirms the economic diversity of coastal towns, as well as the lack thereof in hill top military sites of the Danube valley. The coin samples are characterized by the overwhelming presence of coins issued at Constantin ople, in the case of provinces from the diocese of Thracia, and the importance of the Macedonian mint of Thessalonica in settlements from Illyricum.

PAGE 293

293 Archaeologists, numismatists, and economic historians interested in questions of cultural contact between the early Byzantine Empire and communities living north of the Danube based their interpretation of coins found beyond the frontier against a preconceived assumption that the Byzantine provinces of the Balkans had a developed monetary economy. In their vie w, this market economy based on money often extended north of the Danube to include territories which the Empire intended to include in its sphere of political, economic and cultural influence. The presence of coin finds beyond the frontier is undeniable, but the range of functions performed, which will be explored in the next chapter, goes beyond the monetary value and indeed the use of Byzantine coins as money beyond the Danube frontier of the Empire is perhaps the least significant of the roles it played in barbaricum.

PAGE 294

294 A B Figure 5 8. Nummia per year of reign. A) Balkans, B) Near East

PAGE 295

295 A B Figure 5 9. Nummia per year of reign. A) Co n stantinople mint, B) Anatolia

PAGE 296

296 A B Figure 5 10. Mints (498 616). A) collections, B) sites.

PAGE 297

297 A B Figure 5 11. Mints (498 512). A) collections, B) sites.

PAGE 298

298 A B Figure 5 12. Mints (512 518). A) collections, B) sites.

PAGE 299

299 A B Figure 5 13. Mints (518 527). A) collections, B) sites.

PAGE 300

300 A B Figure 5 14. Mints (527 538). A) collections, B) sites.

PAGE 301

301 A B Figure 5 15. Mints (538 542). A) collections, B) sites.

PAGE 302

302 A B Figure 5 16. Mints (542 550). A) collections, B) sites.

PAGE 303

303 A B Figure 5 17. Mints (550 565). A) collections, B) sites.

PAGE 304

304 A B Figure 5 18. Mints (565 570). A) collections, B) sites.

PAGE 305

305 A B Figure 5 19. Mints (570 578). A) collections, B) sites.

PAGE 306

306 A B Figure 5 20. Mints (578 582). A) collections, B) sites.

PAGE 307

307 A B Figure 5 21. Mints (582 602). A) collections, B) sites.

PAGE 308

308 A B Figure 5 22. Mints (602 610). A ) collections, B) sites.

PAGE 309

309 A B Figure 5 23. Mints (610 616). A) collections, B) sites.

PAGE 310

310 A B Figure 5 24. Denominations (498 616). A) collections, B) sites.

PAGE 311

311 A B Figure 5 25. Denominations (498 512). A) collections, B) sites.

PAGE 312

312 A B Fi gure 5 26. Denominations (512 518). A) collections, B) sites.

PAGE 313

313 A B Figure 5 27. Denominations (518 527). A) collections, B) sites.

PAGE 314

314 A B Figure 5 28. Denominations (527 538). A) collections, B) sites.

PAGE 315

315 A B Figure 5 29. Denominations (538 542). A) collections, B) sites.

PAGE 316

316 A B Figure 5 30. Denominations (542 550). A) collections, B) sites.

PAGE 317

317 A B Figure 5 31. Denominations (550 565). A) collections, B) sites.

PAGE 318

318 A B Figure 5 32. Denominations (512 518). A) collections, B) sites.

PAGE 319

319 A B Figure 5 33. Denominations (578 582). A) collections, B) sites.

PAGE 320

320 A B Figure 5 34. Denominations (582 602). A) collections, B) sites.

PAGE 321

321 A B Figure 5 35. Denominations (602 610). A) collections, B) sites.

PAGE 322

322 A B Figure 5 36. Denominations (512 518). A) collections, B) sites.

PAGE 323

323 CHAPTER 6 MONEY, BULLION, AND PRESTIGE: THE FUNCTION OF EARLY BYZANTINE COINS BEYOND THE FRONTIER 6.1 General remarks To the modern scholar ancient coinage presents two unique qualities: standardization and high freque ncy. Unlike the bewildering variety of artifacts ordered by archaeologists according to somewhat artificially defined classificatory schemes, coinage needs no such arrangements. Byzantine money has already been classified by the issuing authority into diff erent denominations, mints, and dates. Apart from local imitations of Byzantine coins, the same money circulating in the Empire can be found outside the frontier. To be sure, the function performed by Byzantine coins differed from place to place and from o ne context to another, but the distribution of coin finds on a map affords the same kind of observations whether those coins have been found inside or outside the political borders of the early Byzantine Empire. Indeed, such comparisons are most welcome, a s communication routes and directions of circulation can be deduced by analyzing the mints responsible for striking the coins found in a certain area. Much of the argument of the following sections is based on geographical and chronological analyses of fin ds in their respective historical and archaeological context. These may include clusters of small change in one region, the predominance of gold coins in another, the high frequency of "exotic" mints in certain areas, or comparisons between the age structu re of single finds and hoards. In geographic terms, the nature of the numismatic evidence invites a much broader discussion. Although the Lower Danube remains the main geographic unit of analysis, the functions performed by Byzantine coins cannot be prope rly understood without venturing into the regions flanking it to the west, the Carpathian Basin, and to

PAGE 324

324 the east, the northern Black Sea region and by extension Transcaucasia. An excursion of such geographic magnitude was not feasible for the entire archae ological record pertaining to the Byzantine influence outside the frontier, due to the fragmentary nature of the evidence, the level of publication, and indeed the cultural variety of the contexts in which Byzantine artifacts are found (Chapter 4). Coins, however, have been fairly well published, in quantitative if not always qualitative terms, in the wide region from the Baltic to the Caspian Sea (Figures 5 10 to 5 23). The dilemma posed by the cultural variety inevitably encountered in such a large geogra phic area is somewhat vindicated by the homogeneous distribution of similar samples of Byzantine coin finds, which facilitates comparison. The proximity to the Byzantine Empire and t o important war theaters is another major characteristic shared by the thr ee regions discussed in this chapter, the Lower Danube, Transcaucasia and the Carpathian Basin. Regular political payments, occasional imperial gifts, ransom for POWs and VIPs or simply plunder in the Byzantine provinces are the most important documented means through which coins traveled across frontiers. The contexts in which Byzantine coins have been found, singly or in hoards, and their usefulness for the integration of the numismaic evidence into the broader historical narrative requires some clarific ation. The inventory of Byzantine coin finds from barbaricum includes some 1200 single finds and a few more thousands in hoards, but few have been found in a clear archaeological context (Appendix B, Table B 15). Single finds from the frontier region of th e Empire usually resulted from systematic excavations in coastal towns and frontier fortresses, but the nature of the settlements in Barbaricum much smaller and usually unfortified makes them vulnerable to destruction during

PAGE 325

325 agricultural works and their i dentification by archaeologists is quite problematic. Consequently, very few coins have been found in settlements, and out of those rare occasions very few boast a good stratigraphic context. However, the fact that coin finds usually cluster along major ri ver valleys or in the proximity of known settlements remains important for drawing a number of general conclusions regarding their circulation. Furthermore, many coins which ended up in museums of modern countries far from the borders of the Byzantine Empi re, may not have been found locally, but collected from Byzantine provinces. Byzantine coins in England are the most notorious example but the controversy regarding coins from the National Museum in Budapest, closer to the Empire, illustrates some limitati ons of the numismatic evidence. 1 Whenever possible, unprovenienced coins must be judged against finds from secure contexts or confirmation should be sought in a wide frequency of similar finds from the same area. On the other hand, more than half of the h oards have been dispersed, either divided between finders or otherwise lost in unknown circumstances (Appendix B, Table B 16). This poses serious methodological problems for the researcher who tries to place their loss in a historical context, as the date of concealment is always approximated based on the date of the earliest coin in the assemblage. Unless there is a large concentration of hoards in a small region, such as the case of the string of incomplete hoards from Bohemia ending in the reign of Justi nian I, such incomplete finds offer very few certainties. Furthermore, few hoards have been found during archaeological 1 R. Abdy and G. Williams, "A Catalogue of Hoards and Single Finds from the British Iles c. 410 675," in Coinage and History in the North Sea World, c. AD 500 1250. Essays in Honour of Marion Archibald ed. B. Cook and G. Williams (Leiden: Brill, 2006), 11 73. For coins from Budapest, see below n. 291. Coins are not alone in eliciting such doubts. For similar concerns regarding early Christian flasks, see S. Bangert, "Menas Ampullae : A Case Study of Long Distance Contacts," in Incip ient Globalization?: Long distance Contacts in the Sixth Century ed. A. Harris (Oxford: Archaeopress, 2007), 27 33.

PAGE 326

326 excavations. Such hoards like the ones found at Magraneti, Pityus, and Archaeopolis in Georgia or Hrozov in the Czech Republic are impo rtant exceptions. In addition, for the majority of the hoards it is uncertain whether the accumulation was created by gathering coins from that same region or constituted a homogeneous group of coins brought from elsewhere, usually from the Empire. A numb er of general observations can be drawn from the statistical analysis of early Byzantine coin finds in barbaricum (Figures 6 1, 6 2, 6 3, 6 4 and 6 5) The largest number of finds are issues of Justinian I, but his reign is also the longest. By taking int o account the length of each reign the highest percentage of finds goes to Justin II, followed closely by Justin I (Appendix B, Tables B 1 and B 13). All types of Byzantine coins are found in Barbaricum gold, silver, and copper, but in varying proportions depending on emperor and region. The fewest gold coins are those struck for Justin I and Justin II, while the largest concentration of gold can be placed in the seventh century, although an important peak is also reached during the reign of Justinian (App endix B, Tables B 2 and B 3). The largest number of gold coins can be found in the Carpathian Basin and the neighboring regions, on the territory of the Avar Khaganate in particular, where many Avar age cemeteries have yielded Byzantine solidi Before the Avar age, the largest concentration of finds is represented by coins of Anastasius found in coastal areas of the Baltic and the Adriatic and by solidi and tremisses of Justinian found in Central Europe. In many cases across the vast region from Central Eur ope to the Caucasus, gold coins were pierced or modified to be used as jewelry.

PAGE 327

327 Figure 6 1. Sixth seventh century Byzantine coin finds in barbaricum.

PAGE 328

328 Figure 6 2. Mints Figure 6 3. Denominations

PAGE 329

329 Figure 6 4. Byzantine gold, silver, and bronze hoards in barbaricum. Figure 6 5. Sixth seventh century Byzantine gold coins in barbaricum

PAGE 330

330 Almost 80 percent of the silver coins are seventh century hexagrams, most of which are from the reign of Heraclius. Almost 50 percent of the hexagrams are issue s dated 625 629. This should not be surprising as the largest number of hexagrams against Persia in the 620s. In fact this region boasts the highest percentage of silver coins o ut of the total number of early Byzantine coins, 17 percent. This may seem rather unimpressive, but it should be remembered that sixth century silver was strictly ceremonial and is rarely found in barbaricum or the Byzantine provinces alike. Nevertheless, sixth century silver is quite frequent in Transcaucasia, given the rarity of these pieces and most of them date from the reign of Justinian, as do the few pieces found in the Carpathian Basin. Overall copper coins account for the majority of finds in barba ricum and this is no different from our picture of coin circulation in the Empire. Over 65 percent of the finds are copper coi ns and the proportion is somewha t distorted by the fact that gold predominates in the Carpathian Basin (Appendix B, Table B 2). Th e largest concentration of copper coins is to be found in the Lower Danube region ( ca 85 percent), the closest to a heavily militarized frontier. The largest concentration of copper was reached during the reign of Justin II (over 90 percent) but the frequ ency remains high throughout the long sixth century until the accession of Heraclius, after which copper coins account for less than a third of coin finds. Explanations should be sought not only in barbaricum and in circumstances related to the fall of the frontier in the Balkans, but also in Byzantine monetary policies and in fluctuations of mint output. The great advantage of copper coins, aside from their sheer number, is the fact that they are

PAGE 331

331 dated by regnal year (Appendix B, Table B 4). The chronologi cal breakdown of finds in barbaricum reveals a great concentration of coins issued between 512 and 538, which unsurprisingly corresponds with a period of great monetary stability in the Byzantine Empire. The best represented, however, is the coinage of 538 542, featuring the large folles of Justinian, apparently very popular outside the Empire. A general decline can be noticed after 578, while coins after 620 are rare throughout barbaricum. The distribution of coins based on the issuing mint usually reflect s the proximity to the large network of mints maintained by early Byzantine emperors, as well as the general importance of individual mints. Unsurprisingly, the central mint in Constantinople accounts for over 55 percent of the finds, followed by Nicomedia and Antioch (Appendix B, Table B 5). The Syrian mint is particularly influential in Transcaucasia, although it is far from being exotic in the Lower Danube region as well. The importance of provincial mints increased in the second half of the sixth centur y and in many contexts Thessalonica, Nicomedia, and Antioch surpass Constantinople. reconquista are more frequent in Central Europe, whose connections with the western Mediterranean were stronger. Finally, the breakdo wn by denomination reveals an overwhelming dominance of the follis (over 65 percent), while the small change used in low value transactions at the marketplace in the Empire, dekanummia (1/4 follis) and pentanummia (1/8 follis), accounts for ca 10 percent of the finds from barbaricum (Appendix B, Table B 6) It does not take much effort to notice that the flow of Byzantine coins during the sixth and seventh centuries was anything but linear. What are, then, the explanations for the chronological and spatial distribution of Byzantine coins in barbaricum ? What are

PAGE 332

332 the local and broader historical circumstances that influenced the availability of Byzantine coins beyond the frontier? What was the function of money and how were Byzantine coins used by communities living in the shadow of the Empire? These are some of the main qustions addressed in the following sections along with some attempts to clarify methodological issues regarding the use of the evidence. The distribution of gold, silver, and copper coins as well as wider historical implications require a separate discussion of coin finds from the three culturally interrelated regions: the Lower Danube, Transcaucasia, and the Carpathian Basin. 6.2 Geographical and Chronological Patterns of Coin Finds in barbar icum 6.2.1 The Land of Bronze: The Lower Danube From a strictly numismatic point of view the Byzantine coin series begins with the reform of Anastasius in 498. Consequently, numismatists discuss the significance of Byzantine coin finds outside the Empire by taking into account the flow of coins beyond the frontier from the reign of Anastasius onward. However, such a periodization seems artificial if we consider certain historical processes long in motion when Anastasius implemented his monetary policy. At least three major stages need to be taken into account, without necessarily extrapolating historical realities in place centuries before our main chronological focus. The first is the withdrawal of the Roman army and administration from the trans Danubian province of Dacia by Aurelianus. As a result, the cultural distinction between the territory of the former Roman province and the regions east and south of the Carpathians, which were never part of the Roman Empire, slowly became more blurred in the follow ing centuries (Figure 4 1). Coin finds of the fourth century still concentrate in the area of the former Roman towns in Transylvania and Oltenia but in the sixth century this is no longer the case. Notwithstanding the Byzantine

PAGE 333

333 bridge heads in Oltenia and Banat, coin finds in Moldavia and Wallachia are in fact more abundant in the sixth to seventh centuries. The second stage, which brought a certain degree of cultural uniformity to the region, is the development of the so called Chernyakhov culture, a mixture of Roman, Germanic, Sar matian and other local, sometimes pre Roman, influences. The fourth century, the period of its greatest extent and flourishing coincides with renewed Roman ambitions on the Lower Danube and beyond. The imperial policy is reflected in the peace treatises wi th the Goths, which also include provisions regarding trade and commerce often used by historians and numismatists to explain the presence of Late Roman coins north of the Danube. The third important stage is the cultural and political transformation of th e region under the domination of the Huns. The terms most frequently associated with the Hunnic storm in the Balkans and the Carpathian Basin are devastation, collapse, social, economic, and demographic decline. Indeed, there is a sharp decline in the coin flow north of the Danube in the first half of the fifth century but this also coincides with a more general decline of the Late Roman monetary system, reflecting a deeper internal weakness of the Empire at the turn of the fifth century. 2 Historically spea king a discussion of cultural contact in the Lower Danube region during the sixth century must begin with the transitional period from the dissolution of the Hunnic confederation to the accession of Anastasius. This is also the transition from 2 For Late Roman coins in north of the Danube, see D. Moisil, "The Danube Limes and the Barbaricum (294 498 A.D.): A Study in Coin Circulat ion," Histoire et Mesure 17, no. 3 4 (2002): 79 120; D. Protase, (Cluj Napoca: Risoprint, 2000). V. Butnariu, "Monedele romane postaureliene n teritoriile carpato pontice (anii 275 491): III Perioada 383 491 ," AM 14 (1991): 67 107; V. Butnariu, "Monedele romane postaureliene n teritoriile carpato pontice (anii 275 491): II. Perioada 323 383 ," AM 12 (1988): 131 96; V. Butnariu, "Monedele romane postaureliene n teritoriile carpato pontice (anii 275 491): I. Perioada 275 324 ," AM 11 (1987): 113 40.

PAGE 334

334 the Late Rom an monetary system to the new copper coinage introduced by Anastasius. The second half of the fifth century was marked by an almost complete breakdown of the low value currency system as a direct consequence of long term political and economic crisis. From an archaeological point of view the effects of the Hunnic onslaught north of the Danube are still hard to gauge. Fatalistic scenarios aside, both elements of continuity and major cultural change must be considered. The Gepids inherited the Hunnic dominati on in the Tisza basin and by extension in Transy lvania and Banat, but no coherent political unit existed in the Lower Danube region. This is partly why we lack precise information about economic and religious initiatives of sixth century emperors in the re gion, as the Empire lacked a political partner of the magnitude of the Gothic and Hunnic confederations, at least until the rise of the Avar khaganate in the last quarter of the sixth century. Nevertheless, historians often emphasize cultural continuity no rth of the Danube in order to explain the use of early Byzantine coins by making reference to a practice acquired centuries before by communities either living in the shadow of the Roman Empire or, indeed, part of the Roman Empire, in the case of Dacia. Su ch longue dure processes must be taken with caution as they often rely on brush strokes hiding rather than revealing important particularities and developments measured in decades rather than centuries. As a result of the major transformations of the late third to the late fifth century many of the cultural coordinates in place before the third century crisis, or even later during the Constantinian dynasty, were no longer dominant or even traceable on the accession of Anastasius. The Empire was not only st ruggling to re establish the frontier on the Danube but the Balkans often appear to archaeologists as a dense network of fortifications which lacked any support from the

PAGE 335

335 countryside and had to rely on official annona sent from elsewhere under the administr ative auspices of peculiar units such as quaestura exercitus. 3 Urban life can be found almost exclusively on the coasts of the Black Sea and the Aegean and the direct implication is that much of the monetary economy in the Northern Balkans reflects soldier numismatists continue to speak about intensive trade and coin use in barbaricum based on the assumption that the mere presence of coins must reflect a resuming of such activiti es in the sixth century. 4 A survey of sixth to seventh century coin finds is necessary in order to understand fluctuations in coin supply and ultimately the function of early Byzantine coins outside the Empire (Figures 6 6, 6 7, 6 8, 6 9, and 6 10) Early Byzantine bronze coins are dated with the regnal year of the ruler, which allows for a very precise analysis of coin circulation in time, but also conceals a number of methodological pitfalls to which many historians and numismatists have often succumbed. One of the most common is the assumption that coins arrived in a region immediately after they were issued, accompanied by a second logical fallacy stating that coins issued in a certain year could not end up in a given region decades after their issuing d ate. To be sure, the chronological sequence of coins in hoards suggests a steady release of fresh coin disbursed in the form of payment to soldiers defending the fortresses of the Balkans, but 3 A. G. Poulter, "Cataclysm on the Lower Danube: The Destruction of a Complex Roman Landscape," in Landscape of Change: Rural Evolutions in Late Antiquity and the Early Middle Ag es ed. N. Christie (Aldershot : Ashgate, 2004), 223 53. 4 A. M. Velter, Transilvania n secolele V XII (Bucharest: Paideia, 2002), 33 35; L. M. Stratulat, VIII d. Hr.)," Carpica 31 (200 2): 70 71; A. Madgearu, VIII

PAGE 336

336 Figure 6 6. Early Byzantine coin finds north of the Lower D anube (numbers on the map refer to numbers in Appendix A.2 ).

PAGE 337

337 Figure 6 7. Early Byzantine coin finds north of the Black Sea (numbers on the map refer to numbers in Appendix A.2 ).

PAGE 338

338 Figure 6 8. The chronology of early Byzantine coin finds from the Danube frontier and barbaricum.

PAGE 339

339 Figure 6 9. Mints and denominations in barbaricum (Lower Danube and Black Sea) (Abbreviations in Appendix B Table s B 4 and B 6). Figure 6 10. Mints and denominations in the Byzantine bridge heads (Sucidava, Drobeta and Dierna) (Abbreviations in Appendix B Table s B 4 and B 6).

PAGE 340

340 their free circulation prohibits any statement regarding the time when they were lost in barbaricum other than establishing a terminus post quem. The degree of wear which constitutes an import ant indication of prolonged circulation of a coin before its loss cannot be assessed in most cases, unfortunately, because coins are seldom illustrated in publications. This take s away some of the power of numismatic evidence but the reward is a much more responsible use of coins as a primary source for dating archaeological complexes and for assessing social and economic phenomena. Hoards closed in the seventh century often include early sixth century issues and notwithstanding the normal attrition rate coins of sixth century emperors remained in circulation for many decades. The early post reform coinage of Anastasius (498 512) is perhaps an exception as its weight standard was doubled in 512 making it anomalous and susceptible to being withdrawn from c irculation after this date. The fact that some late sixth century hoards from the Balkans still contain such early issues is an invitation to caution in generalizing the nature of its successful withdrawal soon after 512. 5 In any case, only two early issue s are so far known in Barbaricum found east of the Carpathians, one follis from Slobozia Mare (Cahul, Moldova) not far from the Danube and a follis from the area of Brlad (Vaslui, Romania). 6 As both finds are relatively close to the Byzantine province of Scythia one would be tempted to ascribe their presence to documented by 5 h." American Journal of Numismatics 21 (2009): 166 with n. 67. 6 Topografija kladov i nachodok edinichnykh monet Archeologicheskaia Karta Moldavskoi SSR 8 (Kishinev: Shtiintsa, 1976), 87, no. 46; E. Oberlnder Carpica 13, n. 2 (1992), 228, no. 1.

PAGE 341

341 archaeological sources. 7 The other two early reform issues of Anastasius have been found at Drobeta, the important By zantine bridge head in the Iron Gates region. 8 A well known inscription from Ratiaria if indeed dating from the reign of Anastasius, confirms 9 Based on the numismatic evidence it has been suggested that reco nstruction efforts in the Danube region had been initiated as early as the reign of Marcian (450 457), 10 but the process of restoration was indeed a long one and was hindered by renewed conflict with the Ostrogoths in the 480s and by the revolt of Vitalianu s during the reign of Anastasius. Mint output during the experimental stage which preceded the final reform of 512 was low compared to the following decades and the rarity of finds in barbaricum reflects the low number of finds in the Empire, for instance in the province of Scythia whose coin finds have been published in an exemplary fashion since the 1970s. 11 The post 512 coinage of Anastasius is in fact more abundant in barbaricum than it is in the Danubian provinces of the early Byzantine Empire. Most fin ds concentrate in the region close to the Danube, being so far absent from northern Moldavia. Interestingly, two half folles 7 I. Barnea, "Cont ributions to Dobrudja History under Anastasius I," Dacia 4 (1960): 363 74; A. Suceveanu and A. Barnea, La Dobroudja romaine (Buchares 72. 8 E. Oberlnder Trnoveanu, 375 565)," 16 (2004): 55, n. 50. 9 V. Velkov, "Napis na car Anastasiie (491 518) ot Ratiariia," Arkheologiia (1984), no. 1 2: 92 94. n ( anastasis protobyzantines de Bregovina," Starinar 40 41 (1989 1990): 283. 10 Oberlnder Trnoveanu, "Tr 11 Monetary Circulation in the Byzantine Province of Scythia during the 6th and 7th Century," in Numismatic, Sphragistic and Epigraphic Contributions to the History of the Black Sea Coast ,ed. I. Lazarenko, vol. 1 (Varna: Zograf, 2008), 306, table 2. Therefor e, no connection can be made with the attacks of "Bulgars" in 493, 499 and 502, for which see Oberlnder Trnoveanu, de la antichitate," 54.

PAGE 342

342 Romania), the latter from an inhumation grave. The coin was found in a pot attributed to the Gepids. 12 Such a connection might be supposed for the barbarous imitation of a follis bearing the name of Anastasius and found at Drobeta. Such imitations are known in Serbia and Bulgaria deep into the hinterland an d almost all use pre 538 coin types as models, a fact indicating that they were produced in the first decades of the sixth century. 13 This period coincides with the Gepid domination of the western Balkans but also with the presence of the Heruli whose cultu ral mark in the region has received increased attention in the past few years. 14 Gold coins were also imitated, a tremissis of Anastasius being found at Viminacium on the right bank of the Danube, perhaps issued at Sirmium where the Gepids opened a mint. T he post reform coins of Anastasius found in barbaricum along the Lower Danube, in the Carpathian Basin, and north of the Black Sea as far as Kiselivka Hill in 12 Velter, Transilvania n secolele V XII 291, no. 53; V. M. secolele VI VII n teritoriile carpato BSNR 13 1 133 (1983 1985): 219, no. 44. 13 I. Iurukova, "Imitation s barbares de monnaies de bronze byzantines du VI e sicle," Byzantinoslavica 30 (1969): 83 87; D. Gaj th Century in the Balkan Peninsula," Balcanoslavica 2 (1973): 95 100; La monnaie palobyzantine dans Mlanges Ccile Morrisson Civilisation de Byzance, 2010), 446 48. Some 40 imitations were found in the provinces of Thracia and Haemimons, mostly f ollowing Justin I prototypes, for which see A. Tenchova, "Monetna tsirkulatsiia prez VI VII vek v zemite na dneshna Iugoiztochna Bulgariia," Ph.D. Dissertation, National Institute of Arch aeology and Museum Sofia, 2011. 14 A. Kiss, "Heruler in Nordserbien, in Interaktionen der mitteleuropischen Slawen und anderer Ethnika im 6. 10. Jahrhundert. Symposium Nov Vozokany 3. 7. Oktober 1983 Archologisches Institut der slowakischen Akademie der Wissenschaften, 1984), 133 in, Scrinia slavonica 7 (2007): 7 64; A. Sarantis, "The Justinianic Herules: From Allied Barbarians to Roman Provincials," in Neglected Barbarians ed. F. Curta (Turnhout : Brepols, 2010), 361 402; R. Steinach er, "The Herules: Fragments of a History," in Neglected Barbarians 319 arheologicheskie sledy," Stratum+ (2010), 5: 141 57; I. Bugarski and V. oi imperii I varvarov: sistema oborony imperii ot Kutsiia do Lederaty," in Lesnaja i lesostepnaia zony vostochnoi Evropy v epohi rimskikh vliianii i velikogo pereseleniia narodov (Tula: Gosudarstvennyi muzei zapovednik "Kulikovo pole," 2012), 482 511.

PAGE 343

343 monetary syst em, but it is equally possible that some of the coins were brought to these regions and lost under his successors Justin I and Justinian I, sometime before 538 when Justinian operated another major monetary reform, or even later (Appendix B Table B 13). W e can be even less certain about the chronology of gold coins of Anastasius found in Barbaricum as Byzantine gold was more stable than the low value currency and Roman legislation itself encouraged the use of solidi issued by former emperors. 15 Consequentl y, the solidi found at Bucharest Romania), less than 50 miles from the Danube, may well reflect later sixth century subsidies or plunder expeditions in the provinces of the Balkans. 16 The period from 512 to 538 covering the fina l regnal years of Anastasius, the unprecedented monetary stability. Issues of Justin I are well represented in the Byzantine bridge heads of Sucidava and Drobeta, but also in the Roman castrum at among the fortresses reclaimed by the Empire on the right bank of the Danube. 17 Unfortunatelly most of the information for the latter comes from private collections of coins acquired from the area of ancient Dierna rather than from systematic 15 M. F. Hendy, Studies in the Byzantine Monetary Economy c. 300 1450 (Cambridge: Cambridge University Press, 1985), 366. 16 B. Mitrea, "Dcouvertes montaires en Roumanie: 1981, 1982 et 1983," Dacia 28 (1984): 188, no. 131; Butnariu, edelor bizantine," 218, no. 39. 17 Sucidava: 6; A. Vlcu and E. Nicolae, Monede bizantine descoprite la Sucidava," in 3. Drobeta: Oberlnder 61, n. 63. Pojejena: Butnariu, rna: Oberlnder de la antichitate," 62, n. 65.

PAGE 344

344 archaeological excavations. 18 Romania), a former Roman castrum including a pentanummium a rare denomination in Barbaricum which suggests a direct connection with the Byzantine provinces of the Balkans where such coins were more common. 19 A dekanummium of Justin I was found in the castrum at Pojejena, which was tied to the defense system of Moesia Prima. More p entanummia of Anastasi us have been found at Vadul lui Isac (Cahul, Moldova) and Kherson (Kherson, Ukraine) and might indicate similar connections with Byzantine Scythia and Crimea, respectively. 20 A similar pentanummium issued after the reform of 512 was found at Domne with the Carpathians. 21 i n finds from the first decad es of the sixth century. Coins of Justin I and Justinian I issued before the reform of 538 have been found in the area of modern day Bucharest, where several sixth century settlements have been excavated, some of them yielding coin and 22 Coins of 18 E. Oberlnder un punct de vedere numismatic," CN 8 (2002 ): 126. 19 125. 20 E. S. Stoliarik, Essays on Monetary Circulation in the North Western Black Sea Region in the Late Roman and Early Byzantine Periods: Late 3rd century Early 13th Century A.D. (Ode sa: Polis Press,1993), 132, no. 2; Ibid., 133, no. 9. 21 XI) descoperite pe SCN 9 (1989): 78, no. 85. 22 E. Oberlnder Trnoveanu, "Barbaricum apro piat (secolele VI X) 4 (2004): bizantine," 217, no. 19, 21 II," CAB 2 (1965): 174 189. 189, fig. 93/1 3. It has been long suggested that the region of Bucharest was "Les formations politiques (cnezats de valle) du VI e sicle s ur le territoire de la Roumanie," Prace i materialy. Muzeum Archeologicznego i Etnograficznego w Lodzi. Seria Archeologiczne 25 (1978): 109 17.

PAGE 345

345 which also include issu es of Justin II as well as the fact that the general area displays a heavy concentration of late sixth suspect that the coins from Ulmeni were in fact part of a dispersed hoard. 23 If coins of Anastasius did not travel very far from the Danube, coins of Justin I and early issues of Justinian are found in greater number in northern Moldavia in the Siret Prut interfluve but also in the highland region of the eastern Carpathians. This is also a region of hig h concentration of Late Roman amphorae and intensive metallurgical activity, whose beginnings might da te from this period (Figures 4 3 and 4 9). Two folles of Justinian from 527 537 have been found during archaeological excavations in the large settlement featured buildings. 24 In other cases, there is no stratigraphic context but the accidental find of coins together with sixth century pottery indicate s the existence of small settlements. Such situations were encountered i (Vaslui, Romania) and Crja (Vaslui, Romania) where coins of Justin I were found in association with remains of hand made pots. 25 In the same region, but closer to hidden in a ceramic container, ending with issues of Justin I. Its interpretation as 23 162; Oberlnder Trnoveanu, "Barbaricum apropiat," 332. 24 D. G. Teodor, colele V Suceava) (Bucharest: Editura Academiei, 1984), 61, fig. 30/1 2 and fig. 31/6 8. 25 D. G. Teodor, Istorie a Romniei, 1997), 64, no. 156 and 144, no. 619.

PAGE 346

346 reflecting the first Slavic attacks on the Danube frontier should be taken with caution. 26 Gold coins of Justin I are so far unknown in the region between the eastern Carpathians a nd the Dnieper while the two gold coins of Justinian I found at Borolea (Boto s long reign. 27 A solidus and a tremissis of Justin I have been found in Wallachia and in Oltenia respectively 28 A sol idus from as found accidentally and might already signal a center of power in the region not far from the large cemetery at ta Monteoru, if indeed the coin was lost in the first half of the sixth century. 29 Imitations of Late Roman amphorae and hand made lamps imitating Byzantine prototypes suggest that access to Byzantine goods played an important role in the creation of so cial identity and rank (F ig ure 4 8). A number of patterns emerge from the di stribution of coin finds dating from the first decades of the sixth century. The region west of the river Olt, in the former province of Dacia, was heavily influenced by the presence of Byzantine outposts on the left bank of the Danube and coin was distrib uted along the valleys of rivers Olt and Jiu through the mediation of the Byzantine fortresses on the Danube. In Wallachia Moldavia, and the north western Black Sea steppe the flow of coin was dictated by developments in the provinces of Moesia Secunda, S cythia and the Byzantine possessions in Crimea at 26 Trsors no. 351. E. Oberlnder mea sud XIV n lumina descoperirilor monetare," Suceava 26 28 (1999 2001): 317. 27 Oberlnder bizantine ," 217, no. 15 and 220, no. 76. 28 Butn 29 Ibid., no. 46 and 51.

PAGE 347

347 a time when the Empire was re establishing its domination in the northern Balkans through diplomatic action or reconstruction and entrenchment on the Danube. Most of the Byzantine coins crossing the Danube were copper issues from the central mint in Constantinople, which is also typical for the Balkans at a time when the early Byzantine monetary system was still in its infancy. Three coins of Justinian from Antioch were found not far from the Danube frontier and reflect communication routes in the Black Sea or perhaps the movement of soldiers from one theater of operation to another, if the coins were brought to the Danube region at a later date, when hostilities between the Byzantine Empire and Sasanian Pers ia were resumed. The finds of gold coins which concentrate in the Danubian plain may reflect the payment of soldiers and mercenaries but also plunder in the early Byzantine provinces or diplomatic gifts sent to petty chieftains north of the Danube. It is t empting to link these finds to the state of the frontier as described by Procopius who lamented the death of the capable general Chilbudios in 533, after which "the river became free for the barbarians to cross at all times just as they wished and the poss essions of the Romans were rendered easily accessible. 30 It must be mentioned that most gold coins from the territories north of the Danube are Justinianic solidi struck before 550, the majority being issues from 527 538. This might constitute a strong ind ication that they were lost sometime in the second quarter of sixth century. Finds of dated coinage of Justinian (538 565) display a remarkable series of contrasts. Previous interpretations rested on preconceived scenarios about the historical development of the Lower Danube region in which the numismatic evidence 30 Procopius, Bella VII, 14.6, ed. J. Haury and G. Wirth, vol. 2 (Leipzig: Teubner, 1963), 354; trans. H. B.

PAGE 348

348 had to play its designated role, as we shall soon find out. The first and most obvious observation one can make by looking at the distribu tion of finds in time (Figure 6 8 ; Appendix B Table B 4) is the unusual concentration of coins dated between 538 and 545, especially issues struck on the heavy standard employed after the reform (538 542), which can be noticed in all regions discussed. 31 The heavy coins of Justinian, whose role in the Byzantine e conomy was discussed elsewhere, 32 were extremely popular in barbaricum and account for almost 18 percent of all early Byzantine copper coins found between the Iron Gates of the Danube and the Dnieper. 33 The thirty find spots are not evenly distributed on the map, the largest number of finds being recorded in Moldavia and in the north western Black Sea region where Justinian sought the alliance of the Ant e s 34 Some of the coins were found close to the Danube and can be connected to coin circulation in the Byzan tine settlements in Oltenia while others were (Cherkasy, Ukraine) on the Middle Dnieper. 35 Constantinople remained the most important mint but the presence of issues from Nicomedi a and Cyzicus increased 31 Oberln der Trnoveanu, "Barbaricum apropiat," 338. His argument that coins from 527 538 are more abundant in Wallachia than later issues of Justinian is not valid since it does not take into account that the period covers eleven years while coins after 538 are da ted with the regnal year and statistical calculations are ma de on a one year basis. 32 Consequences," RN 168 (2012): 363 402. 33 The only hoard closed in this period is the small accumulation from Stary Biliary (Odesa, Ukraine), with the closing coin dated 542/3, for which see E. S. Stoliarik, "Klad vizantiiskikh bronzovykh VI v. iz Starye Beliary, Odesskoi oblasti," in trudov ed. G. A. Dzis Raiko (Kiev: Nauk dum ka, 1984), 136 38 wi th pl. 1 2. 34 S. Patoura, "Une nouvelle considration sur la politique de Justinien envers les peuples du Danube," Byzantinoslavica 58 (1997): 81 82. 35 D. G. Teodor, Teritoriul est carpatic n veacurile V blema Klady vizantiiskikh monet na territorii SSSR (Moskow: Izd vo Akademii nauk SSSR, 1962), 37, no. 275.

PAGE 349

349 testifying to the growing complexity of the monetary system which had relied almost exclusively on the central mint in the previous decades. This unprecedented increase in the number of finds from barbaricum came to an abrupt end c a 545 and there is a conspicuous hiatus in finds from 548 554. 36 The coin flow picked up again from 555 560 and a final decline occu r red until the end of s reign, marked by several gaps between 560 and 565 (Appendix B Table B 4). Historians and n umismatists have explained this situation relying on two preconceived notions: the first is the already mentioned assumption that coins must have arrived north of the Danube close to the time when they were issued and almost never at a later date, while th e second is the tendency to explain fluctuations in coin supply by reference to military events and the arrival of new ethnic groups, Slavs in particular. Recent analyses of the coin flow to barbaricum during the age of Justinian have linked the sharp decl ine in the second half of his reign to successive incursions of Slavs, Bulgars, and Cutrigurs in the diocese of Illyricum and Thrace and the permanent settlement of the Slavs east and south of the Carpathians. 37 Again, scholars did not allow for the possibi lity of a later arrival of coins and preferred to explain the gaps by searching for clues in the written sources for events taking place in those years devoid of coin finds in barbaricum. Archaeological evidence was adduced in support of this historical ex planation but the use of coins to date complexes often made the arguments 36 F. Curta, The Making of the Slavs: History and Archaeology of th e Lower Danube Region, ca. 500 700 the Balkans, but the hypothesis, which remains plausible, was based on an exaggerated interrupt ion of the coin fl ow (545 560). 37 Oberlnder Trnoveanu, "Barbaricum apropiat," 339 romane 12 (2003): 334; Madgearu, 68 89.

PAGE 350

350 appear circular. 38 In reality the explanation is quite different and has nothing to do with events in barbaricum The chronological gaps correspond to low points in mint output tracea ble in finds from the Balkans and from other corners of the Early Byzantine Empire. 39 Coin finds in barbaricum mirror fluctuations in output at the imperial mints and the supply of fresh coin to the frontier fortresses. Whether or not Slavic groups settled in the Danubian plain around 550 continues to be debated by historians and archaeologists of the Early Middle Ages but coins have little to offer toward a successful resolution of this historiographic question. The only meaningful observation is that whate ver changes occurred in the ethnic background of the region the connection to the Byzantine Empire, reflected in coin finds, was not broken until the Danube itself ceased to be under direct Byzantine rule. The heavy coins of Justinian from 538 542 and by extension, to 550 were popular not only in barbaricum but also in the Byzantine outposts from the left bank of the Danube, as testified by numerous finds resulted from archaeological excavations at Sucidava. 40 Some of the soldiers defending the garrisons o f the Byzantine bridge heads in barbaricum were probably recruited from the region north of the river and they helped Chilbudios himself, the general celebrated by Proco pius for his victories north of the Danube, had a non Roman origin and it is safe to assume that many of the soldiers under his command were also foreigners in the general sense of the term. Sucidava 38 M. Comsa, U nele date cu privire la Banatul de sud n sec. IV VII," in In Memoriam Constantini Daicoviciu ed. H. Daicovi ciu (Cluj : "Dacia," 1974), 96. 39 Coin Circulation," 208, fig. 3. 40 2 7, no. 11 17.

PAGE 351

351 also produced a significant number of dekanummia a low denomination typical for low value market transactions. Such denominations are rare in barbaricum and always worth highlighting. The three dekanummia from Calafat (Olt, Romania), Cioroiu Nou 41 The imperial mints issued high quantities of dek anummia in the last decade of barbaricum in this period, given the preference for heavier coins, usually folles Again, chronological gaps can be explained by reference to monetary policies i n the Empire rather than by highlighting military events on the Danube frontier, which supposedly hindered the circulation of coin across the river. Two dekanummia in Rome, an important fact pointing to a great diversification of mints in the age of Justinian, and renewed, probably military, connections to Italy. Mints other than Constantinople account for more than 45 percent of the f inds of Justinian north of the Danube and north west of the Black Sea (Appendix B, Table B 7). One pentanummium from Parutyne (Mykolayiv, Ukraine) was issued by the Crimean mint in Cherson, while a half follis from Salona was found at Drobeta in the Iron G ates of the Danube, possibly signaling the movement of troops during the Gothic wars in Italy. 42 Procopius informs us barbaricum for his campaigns in the west, and the presence of western mints north of the Danube 41 Calafat and Cioroiu Nou: Oberlnder Trnoveanu, "Societate, secolele VI XI," SCIV 21, n. 1 (1970): 121 fig. 9/4. 42 Stoliarik, Essays on Monetary Circulation, 137, no. 42; Oberlnder antichitate," 68, n. 70.

PAGE 352

352 seems to confirm his account. 43 Some of the gold coins of Justinian dated before 550 as well as the rare silver miliarens es from Oltenia and Banat may constitute stipends and gifts sent by Justinian to buy the loyalty of c ertain groups fr om the area dominated by Heruli and Gepids. 44 With the accession of Justin II we witness a heavy influx of Byzantine coins north of the Danube but also a drastic change in the imperial attitude toward barbarian groups. The new emperor decide d to discontinue diplomatic payments and rely instead on internal resources for securing the Danube frontier, an attitude best encapsulated in Cycle : let no barbarian, freeing himself from the yoke strap that passes under his neck, dare to fix h is gaze on our King, the mighty warrior. 45 As a result the Danube region became highly militarized and the direct consequence is an absolute increase in the number of coin finds from frontier fortresses of the Balkans. 46 The policy was coordinated with the imperial mints, Thessalonica being commissioned to strike huge quantities of half folles to pay the troops. The balance of power in barbaricum was also shifting with the arrival of the Avars in Pannonia ca 568 followed by the swift subjugation of the Gepid s and the Slavs. Against this background the increase of coin quantity in the 43 J. L. Teall, "T he barbarians in Justinian's armies," Speculum 40, no. 2 (1965): 294 322; H. Ditten, "Slawen im byzantinischen Heer von Justinian I. bis Justinian II.," in Studien zum 7. Jahrhundert in Byzanz. Probleme der Herausbildung des Feudalismus ed. H. Kpstein an d F. Winkelmann (Berlin: Akademie Verlag, 1976), 59 72; M. Whitby, "Recruitment in Roman Armies," in The Byzantine and Early Islamic Near East III: State, Resources and Armies ed. A. Cameron (Princeton, NJ: The Darwin Pre ss, 1995), 61 125, esp. 103 10. 44 BSNR 86 87 (1992 1993): 123 24; Oberlnder 70, n. 72; A. Vlcu, T. Isvoranu, and E. Nicolae, Les (We tteren, Moneta, 2006), 153, no. 460; A. Kharalambieva, Gepids in the Balkans: A Survey of the Archaeological Evidence," in Neglected Barbarians 245 262; Sarantis, "The Justinianic Herules," 361 402. 45 Agathias, Cycle IV.3.47 97,ed. and trans. W. R. Paton The Greek Anthology (New York: G P. Putnam & Sons, 1920), 119. 46 "Early Byzantine Coin Circulation," 179.

PAGE 353

353 northern Balkans was also felt in barbaricum Potential changes in the ethnic structure of the communities living north of the Danube cannot be traced in the numismatic evidence as coins seem to have followed the same paths as they had in the first half of the sixth century. The large production of half folles by Thessalonica is reflected in finds Ol ten coin was found in a sixth century settlement. 47 The year by year distribution of finds displays two major spikes corresponding to years 570 and 575, a clear reflection of the q uinquen n ial donativa promptly distributed to the garrisons defending the frontier. In this respect there is no significant difference between the chronology of finds from barbaricum and those from the Danubian provinces or the Byzantine bridge heads on the northern bank of the river. The only difference is the higher frequency of coins issued in Nicomedia, mostly folles in Barbaricum compared to those from Thessalonica, dominant in frontier fortresses of the Balkans, as we have seen. The explanation lies in the preference for higher denominations in Barbaricum folles of Nicomedia and Constantinople being favored against half folles of Thessalonica. The military nature of coin circulation in the Byzantine outposts north of the Danube is also testified by t he presence at Drobeta of coins issued by mobile mints, the so called Moneta Militaris Imitativa and of issues of Antioch at Sucidava pointing to the transfer of troops to the Balkans. 48 The constant presence of Antioch among finds throughout the sixth cent ury (over 6 percent, Figure 6 9 ; Appendix B Table B 7) might be also related 47 D. G. Teodor, Teritoriul est carpatic n veacurile V nimea, 1978), 23 and fig. 16/5. 48 Oberlnder Trnoveanu, "La bizantine," 227, no. 32; Vlcu and Nicolae, "Monede bizantine," 303, no. 22.

PAGE 354

354 to finds of Palestinian lamps and other devotional objects from the Holy Land in Barbaricum but a military explanation seems more plausible at this point. Recent attempts to mak e a distinction between Oltenia, Wallachia and Moldavia based on the chronology of finds from Justin II are not convincing. 49 The assumption that certain gaps in finds from Wallachia testify that the Slavs were disturbing cultural contact between local comm unities and Byzantium cannot be substantiated by reference to a still very small coin sample, substantial enough to make general remarks about a large geographic area from northern Serbia to southern Ukraine but too thin to trace annual developments in sma ller regions. 50 A common characteristic is the scarcity of gold coins of Justin II. Only one find is recorded, a solidus solidus Romania), while three other gold coins were found in unknown locations in historical Banat and might be connected to the Middle Danube region. 51 The rarity of gold coins gold outside the frontier. The virtual absence of gold from Justin II despite the accounts of contemporary writers describing the wealth brought by the Slav s north of the Danube 49 Oberlnder Trnoveanu, "Barbaricum apropiat," 348 49. 50 "Socio Ec onomic Organization of the Daco Romanic and Slav Population on the Lower Danube during the 6 th 8 th centuries," in Relations Between the Autochtonous Population and the Migratory Populations on the Territory of Romania an d P. Diaconu (Bucharest: Editura Academiei RSR, 1975), 171 200; S. Dolinescu Ferche, "La culture Ciurel e VII e sicles). La situation en Valachie," Dacia N.S. 28 (1984): 117 47. 51 Awarenzeit. Einem Bestandaufnahme, 1998 2007," AAC 42 43 (2007 2008): 292 294, no. 23 25.

PAGE 355

355 after plunder expeditions remains a mystery. 52 Gold issues of his predecessors may have been part of the booty but an explanation for the lack of coins of Justin II based on a supposed low gold mint output during his reign is improbabl e given the high frequency of his coins in gold coin hoards from the Balkans. 53 Political developments in the region, however, left their mark in the form of copper coin hoards whose number increases on both sides of the Danube. The only hoard from barbaric um echoing these events was Reka (575/6), Axiopolis (578/82) and Capidava (579/82). 54 These hoards as well as several others from the II into the early 580s can be connected with increased warfa re in the region, marked by Slavic incursions in the Balkans and retaliation of the Byzantines allied with the Avars. The account of Menander the Guardsman shows that insecurity on the Danube frontier was increasing and the contribution of the Avars to the destruction of the Slavic center of power in eastern Wallachia ca 579 did nothing but increase the Avar influence north 52 E.g. John of Ephesus, Histor ia Ecclesiastica VI, 25, ed. E. I. Brooks (Louvain: Ex Officina Orientali et Scientifica, 1936), 248 49. Gold coins accoun t for only ca nine percent off all finds from Justin II in the entire studied region from the Baltic Sea to the Caspian Sea (Appendi x, Table 1). 53 Almost all gold coin hoards hidden in the Balkans in the last quarter of the sixth century contain issues of Justin II. In one hoard from Thessalonica 85 out of 115 coins were issues of Justin II, while the proportion is lower, around one th ird of the hoard in most other cases, but still high enough to prevent any suggestion of a low mint output during his reign. For the hoards, see Trsors no. 19, 65, 82, 103, 117, 151, and 244 246. Moreover, the collection of the Romanian Academy in Buchar est includes many gold coins of Justin II, for which see A. Vlcu, II. Monnaies Byzantines (Wetteren: Mo neta, 2009), 49 54, no. 87 117. 54 Trsors descoperit la Capidava." CN 15 (2009): 87 105.

PAGE 356

356 of the Danube. 55 The hoard from Gropeni, hidden on the left bank of the Danube is a reflection of these events, but cannot tell us much about Byzantine coins from barbaricum. Most probably, given the proximity to the Byzantine frontier, the hoard from Gropeni represents a sum of money brought from the province of Scythia, hidden closer to home in the context of the conflict between Byzan tines, Avars, and Slavs, and never retrieved. 56 The period between 577 and 582 is marked by several gaps in the chronology of finds from barbaricum but this partly reflects the state of the Danube provinces affected by insecurity and warfare. Most of the co ins of Tiberius II have been found in Byzantine fortresses at Sucidava, Drobeta and Ostrovu Banului (Mehedin 57 Many are issues of Thessalonica, the mint responsible with sustaining the war effort in Illyricum. The pentanummium from Drobeta proves that low value exchanges were still taking plac e despite the political and finds can be placed with certainty outside the sphere of the Byzantine bridge heads on the left bank of the Danube, two folles 58 The 55 Menander, Historia fr. 21, ed. and trans. R. C. Blockley (Liverpool: Francis Cairns, 1985), 192 95; Madgearu, Conti nuitate 18 19; AM 16 (1993): 191 203. 56 A recent attempt to argue that only the coins of Justin II from the Gropeni hoard were brought from the Empire, the earlier ones being from the lo cal stock, is unconvincing, see Oberlnder Trnoveanu, "Barbaricum apropiat," 346. Another hoard, found in Bucharest before WW2 in unknown circumstances, ends with coins from 580 581, but it is uncertain whether the hoard is complete. See Trsors no. 350 57 304 05, no. 30 34; Oberlnder Drobeta: Ibid., 129, n. Ibid., 219, no. 67. 58 E. Oberlnder Trnoveanu and E. Mousaios 4 (1994): 319, n. 28.

PAGE 357

357 solidus nineteenth century hoard, and could have been brought north of the Danube after one of the numerous plunder expeditions lamented by contempo rary writers. We know for certainty that warfare in the Danube region, which led to the loss of Sirmium to the Avars in 582, did not stop the circulation of Byzantine goods north of the river. Increased warfare in the northern Balkans in the last quarter o f sixth century brought a large quantity of Byzantine cast fibulae with bent stem and buckles in Barbaricum which are typical for the military hilltop fortresses of the Danube provinces. The frontier region was also responsible for the development and spr ead of the the so called Slavic bow fibulae, a fashion that can be traced on both sides of the Danube. In addition, a steady influx of fresh coin to barbaricum can be noticed until ca. 600 with a short gap in 584 586 and 600 602. Coins from the last year generally scarce even in the Danubian provinces, the case of Scythia being the best documented. 59 It is most likely a problem of mint output in this period, as Antioch seems to have been the most active and the Syrian mint is never dominant in finds from the Balkans and by extension, in barbaricum. 60 Many finds of coins issued during the reign of Maurice have been recorded on the left bank of the Danube from the Iron Gates to the mouths of the river. Such coins found at Ostrovu Mare, Gogo and Vasylivka may be related to the Avar offensive in the 580s and the Byzantine counteroffensive which included a number of incursions north of the river, led by general 59 "Some Aspects," 320, table 1 60 It has been argued that the impact of the Byzantine campaigns against the Slavs in Wallachia led to the interruption of relations with Byzantium, but the presence of coins of Phocas and Heraclius in the same region makes this hypothesis improbable, see Oberlnder Trnoveanu, "Barbaricum apropiat," 354.

PAGE 358

358 s brother Petrus, between 594 and 602. 61 Both Drobeta and are recorded until 599, when Sucidava probably fell under the attacks of the Avars. The finds include gold coins, a tremissis at Sucidava and a light weight solidus of 20 siliquae at Drobeta, both coins probably reflecting military payments sent to the local garrison. 62 An unusual concentration of light weight solidi of Maurice can be noticed in the province of Scythia at a time when the Avar khagan was in Anchialus negociating an increase of the tribute to 100,000 solidi but preparing to plunder the rich towns along the Black Sea coast. 63 Three coins minted in Antioch found north of the Danube may signal the movement of troops after the conflict with Persia was brought to an end and Maurice focused his energy on the northern Balkans. One issue of Antioch and another of solidi are recorded in the region of Wallachia where several Slavic political centers are mentioned by Theophy lact Simocatta. 64 (Orhei, Moldova) and Pavlivka (Odesa, Ukraine). 65 The early Byzantine monetary 61 Curta, The Making of the Slavs 100; A. Madgearu, "The Province of Scythia and the Avaro Slavic Invasions (576 626)," Balkan Studies 37, no. 1 (1996): 42 51; M. Whitby, The Emperor Maurice and His Histo rian: Theophylact Simocatta on Persian and Balkan Warfare (Oxford: Clarendon Press, 1988), 140 65. 62 Oberlnder Vlcu, Les monnaies d'or 60, no. 148. 63 For the context, see Whitby, The Emperor Maurice 142 43. For the coins, see Vlcu, Les monnaies d'or 10. 64 Theophilact Simocatta, Historia VI, 6 10; ed. C. de Boor and P. Wirth (Stuttgart: Teubner, 1972), 230 42; Oberlnder Trnoveanu and Constantinescu, "Monede romane trzii," 317 19, n. 29 and n. 31; Ob erlnder Trnoveanu, "Barbaricum apropiat," 353, n. 40; B. Mitrea, "Dcouvertes rcentes et plus anciennes de monnaies antiques et byzantines sur le territoire de la Rpublique Populaire Roumaine," Dacia 7 (1963), 597, no. 52. 65 Topografiia kla dov i nakhodok 83, no. 14; Stoliarik, Essays on Monetary Circulation, 140, no. 59.

PAGE 359

359 economy was slowly breaking down in the Balkans, a process also felt in barbaricum where o nly one denomination below the half follis is so far recorded, a dekanummium found at Brlad in Moldavia. 66 More than two decades of warfare in the northern Balkans left a trail of hoards found on both sides of the Danube in different contexts, but hidden a nd lost under the same circumstances. The Avar offensive from the mid 580s led to the concealment of several copper and gold hoards in the Danubian provinces, at Koprivec, Provadiia, Adamclisi, Zulud, Zaldapa, and Sadovec. 67 The geographic distribution of finds shows that the diocese of Thracia was most affected by these attacks coordinated by the new Avar khagan in alliance with the Slavs led by Ardagast. 68 No echo of this hoarding frenzy can be trac ed north of the Danube, unless the hoard from Troianul in Wallachia 69 The hoard is an unusual combination of Late Roman coins with early Byzantine dekanummia and pentanummia of Justinian, Justin II and Maurice. The coin of Maurice cannot be dated with precision but the entire accumulation fits the pattern of small change used on a daily basis in market transactions in one of the large towns of the Black Sea coast or the Aegean. The next group of hoards dating to the las t decade of sixth century are connected with the Byzantine campaigns north of the Danube aimed at destroying the Slavic centers of power in Wallachia and southern Mold avia and at re establishing the balance of power in the Middle Danube region by checking the expansion of the Avars. Although 66 Oberlnder "Monede bizantine," 229, no. 5. 67 Trsors no. 47, 52, 61, 63, 79, 82 and 240 245. 68 Whitby, The Emperor Maurice 140 45. 69 Trsors no 364.

PAGE 360

360 most of the late hoards on the Danube frontier are from Illyricum (Bosman, Reselec and Rakita), the hoards from barbaricum were found across the province of Scythia. 70 The Danube and appears to be connected with the Byzantine early counteroffensive led by Priscus in Wallachia in 593 71 The second hoard was found quite far from the Danub e, to the same military events in barbaricum. The hoard itself is a collection of metal objects, a copper alloy pitcher, a bronze chain, and scrap pieces of bronze, perhaps gathered hastily by a refugee. Alternatively, it may be the result of plunder during this decade of insecurity. 72 Although a treaty in 598 was establishing the Danube as the natural frontier agreed between the Avars and the Byzantines, war resumed and the rebellion of 602 against Maurice was staged at a moment when the Byzantine army was camped north of the Danube. 73 Archaeological research in the past decades has confirmed the survival of certain sections of the frontier well after the rebellion of 602. In fact the new emperor Phocas sealed a new treaty with the Avars in 604, agreeing to increase the annual tribute to 140,000 solidi. 74 Although Phocas transferred the troops to the East to engage the Persians, no serious invasions occurred in the Balkans unti l the reign of Heraclius. The Danube remained de facto the frontier between Avars and Byzantines 70 Trsors no. 238, 239 and 260. 71 Ibid. no. 366. For the context, see Whitby, The Emperor Maurice 156 61. 72 Trsors no. 355. 73 Whitby, The Emperor Maurice 161 65. 74 Theophanes, Chronographia a. 6096, ed. C de Boor (Leipzig: Teubner, 1883), 292

PAGE 361

361 until ca 615 when the Balkans were left undefended in the face of new attacks of Avars and Slavs. The Persian conquest of Jerusalem in 614, the loss of Egypt i n 619, and the destruction of the large towns of the Byzantine heartland in Asia Minor taxed the remaining resources of an already impoverished Empire. The abandonment of the Danube frontier was nothing but an inevitable consequence. 75 Byzantine coins conti nued to cross the Danube without major interruption until 615, which is another confirmation of the fact that some Byzantine fortresses on the Lower Danube were still defended. No coins after 599 have so far been recorded at Sucidava and it is possible tha t its role as a military outpost ended after the Avar attacks in the last decade of sixth century. Drobeta, however, continued to receive Byzantine bronze coins well into the seventh century, as did a number of fortresses on the right bank, such as Novae, Durostorum, Capidava and Noviodunum, to name just a few. 76 Interestingly, no issues of Phocas have been recorded among coin finds from Drobeta, a long gap being noticed between 599 and 612, possibly pointing to early initiatives of Heraclius to strengthen h is position in the Iron Gates sector of the frontier after a decade of decline. Most of the strategic points in the western sector of the frontier had already been lost 75 For events in the East, see G. Greatrex and S. N. C. Lieu, ed., The Roman Eastern Frontier and the Persian Wars. Part II, AD 363 630. A Narrative Sourcebook (London/New York: Routledge, 2002), 187 97. For the Danube frontier, see Madgearu, Continuitate 24. 76 Novae: K. Dimitrov, "Poznorzymskie i wczesnobizantyjskie monety z odcinka iv w Novae z lat 294 612." Novensia 11 (1998): 110 11. Durostorum: E. Oberlnder Trnoveanu, Monnaies byzantines des VII e X e si cles dcouvertes a Silist Cabinet des Medailles du Histoire de Roumanie" CN 7 (1996): 117 20 ; Capidava: A. 12 13 (2006 2007): 118. Noviodunum: G. Poenaru Bordea et al., Contributions numismatiques aux VI e VII e si cles SCN 11 (1995): 157 61.

PAGE 362

362 and apparently never recovered judging by the interruption of coin circulation in i mportant fortresses such as Aquis, Viminacium, Ratiaria, and Singidunum. 77 Most coins of Phocas found in barbaricum are half folles a proportion noticed for finds from Scythia as well, 78 but since very few display the date we can no longer follow the annual influx of coins north of the Danube. Given the steady annual supply between 610 and 615 it can be supposed that the same was true for the reign of Phocas (602 610). Very few finds are recorded in Oltenia and Banat, the destruction of the Byzantine limes in Moesia Prima and Dacia Ripensis and the proximity to the Avar center of power being the main reasons for reduced cultural contact with communities living north of the Danube. Coins of Phocas are more abundant in Wallachia and Moldavia, usually close to the Danube, at Ol noe but also further to the north, in the Prut Dniester interf l uve. 79 Renewed efforts to re establish the Byzantine control on once again sent to the garrison stationed at Drobeta and perhaps at Dierna as well. 80 This policy was also felt in Barbaricum a significant number of finds being recorded in Oltenia in the traditional area of influence of the Byzantine bridge heads north of the Danube. All coins are copper issues dated between 610 and 616 when the 77 Podunavski deo oblasti Akvisa u VI i pochetkom VII veka Institut, 1981), 7 1092) iz zairke Narodnog Myzeja y Pojarevci," 11 (1988): 87 99. Ratiaria: B. Boshkova, "Coins from the Excavations of the Antique Town Ratiaria," Ratiarensia 2 (1984): 105 16. Bel 10 (1987): 88 107. For a synopsis of finds from northern 441 54. 78 "Some Aspects," 318, table 5. 79 Obe rlnder Trnovea nu, "Barbaricum apropiat," 356. 80 Oberlnder

PAGE 363

363 Danube frontier was still functional. Two solidi have an unknown finding place in Oltenia but we are better informed about gold coins from regions of barbaricum across the dioc ese of Thracia and north of the Black Sea. The power centers of the Slavs in southern Moldavia were the main recipients of gold coins as testified by the solidi found 81 A small hoard of solidi (Suceava, R omania) in northern Moldavia dated 616 625 belonged to a local big man, judging by the archaeological context in which the coins have been found. 82 The smaller communities in Moldavia continued to receive Byzantine copper coins. Two folles of which one is an issue of 619 620, a late date even for the provinces of the Balkans, were found together with sixth Romania). 83 river was hidden shortl y after 614 and has the typical structure of a sum of money withdrawn from circulation from one of the frontier provinces, including a number of dekanummia commonly used in the urban marketplace. 84 The most probable origin is one of the frontier towns of Sc ythia still guarding the lower sector of the Danubian limes in the second decade of seventh century, as testified by the numismatic evidence. 85 Three solidi found in burial assemblages in the region of Kirovohrad in central Ukraine are later issues from the third decade of seventh century and signal the major 81 Oberlnder Trnoveanu and Constantinescu, "Monede romane trzii," 3 11, no. 32; Oberlnder Trnoveanu, "Societate, eco 82 Suceava," Suceava 26 28 (1999 2001): 296 97 and 310, fig. 7/26 28. For the archaeological context, see Magazin Istoric 11 (1977): 49. 83 Teodor, Teritoriul est carpatic 23, fig. 16/6 7. 84 Trsors no. 358. 85 12.

PAGE 364

364 political transformations following the retreat of the Byzantine Empire from the northern Balkans. 86 At a time when Heraclius was desperately looking for allies against the Avars and Persians who assault ed the walls of Constantinople in 626, the steppe north of the Black Sea became very important for the balance of power in the region. The rising influence of the Bulgars in the Pontic region brought a large quantity of Byzantine coined gold and silver or jewelry and plate found in a number of spectacular seventh century burials and hoards. Diplomatic relations between the Byzantine Empire and the Old Great Bulgaria in the 630s, as described by Nikephor os and John of Nikiu, were concluded when Heraclius h onored Kubrat with the title of patricius 87 The most prominent archaeological reflection of such diplomatic contact s is the rich burial from Malo Pereschepyne (Poltava, Ukraine), which many historians attribute to Kubrat himself. 88 The treasure included som e 70 Byzantine gold coins, the majority being light weight solidi of 20 siliquae struck for Heraclius and Constans II, the latest issues being dated no later than 646. Most of the coins were either pierced or mounted into sets. The gold hoards from Kelegei a (Kherson, Ukraine), Maistrov (Zaporizhia, Ukraine), 86 Pechenaya: Kropotkin, Klady vizantiiskikh m onet 33, no. 196. Rivne: Stoliarik, Essays on Monetary Circulation, 141, no. 70 and fig. 14/3. Yosypivka: O. S. Beliaev and I. O. Molodchikova, "Pokhovannia kochyvnikov na r. Orel," Arkheolohiia 28 (1978): 86, fig. 2/4. 87 Nikephoros, Breviarum 22, ed. an d trans. C. Mango (Washington DC: Dumbarton Oaks, 1990), 70 71; John of Nikiu, Chronicle 120.47, ed. and trans. H. Zotenberg (Paris: Imprimerie Nationale, 1883), 460; V. Vachkova, "Danube Bulgaria and Khazaria as Parts of the Byzantine oikoumene ," in The O ther Europe in the Middle Ages: Avars, Bulgars, Khazars, and Cumans ed. F. Curta (Leide n/Boston: Brill, 2008), 343 45. 88 J. Werner, (Munich: C.H. V. Sokolova, "Monety Pereshchepinskogo klada," in Sokrovishcha Khana Kubrata ed. O. Fedoseenko (St. Petersburg: AO Slaviia, 1997), 17 41. However, there is no consensus regarding the ethnic attribution of the find, for which see especially O. V. Komar and V. M. Khardaev, "Zachepilovskii ("Novosanzharskii") kompleks rubezha VII VIII vv," in Stepi Evropy v epokhu srednevekov'ia. Khazarskoe vremia. Sbornik nauchnykh trudov ed. A. V. Evgelevskii (Donetsk: Doneckii Nacional'nyi universitet 2012), 243 96, esp. 278 281 for the discussion of coin finds.

PAGE 365

365 belong to the same cultural horizon of wealth accumulated through political payments sent to the regions north of the Bla ck Sea by Heraclius and Constans II. 89 Although the Empire had retreated from the Danube, the relations with the territories north of the river were not completely interrupted. The decline of the Avar khaganate after the failed siege of Constantinople in 62 6 and the Byzantine retrenchment after the initial shock of the Arab storm in the East permitted a more active policy in the Balkans. The few bronze coins of Constans II in barbaricum were found not very far from the Danube, at Re Wallachia and Brlad in Moldavia, while a solidus 90 Based on their dating, the c oins were probably lost in the 650s 6 60s. The specimen found at Novaci is an issue of Carthage, which should not appear surprising. Several seventh century coins issued in mints from Italy and North Africa have been found in the north eastern Balkans. 91 A hoard found at Tomis on the Black Sea coast included three coins from 89 Kelegeia: A. I. Semenov, "Vizantiiskie monety Kelegeiskogo kompleksa," Arkheologicheskii sbornik Gosudarstvennog Ermitazha 31 (1991): 121 30. Maistrov: Kropotkin, Klady vizantiiskikh monet, 31 32, no. 159. Dnipro Vizantiiskii Vremennik 11 (1956): 292 94. Zachepylivka: A. T. Smilenko, "Nakhodka 1928 g. u g. Novye Senzhary (Po materialam obsledovaniia A. K. Takhtaia," in Slaviane i Rus' ed. E. I. K rupnov (Moscow: Nauka, 1968), 158 166. For a recent discussion of finds of the second half of seventh century in the Black Sea region, see P. Somogyi, "New Remarks on the Flow of Byzantine Coins in Avaria and Wallachia during the Second Half of the Seventh Century," in The Other Europe in the Middle Ages 108 25. See also J. Smedley, "Seventh Century Byzantine Coins in Southern Russia and the Problem of Light Weight Solidi," in Studies in Byzantine Gold Coinage ed. W. Hahn and W. E. Metcalf (New York: Amer ican Num ismatic Society, 1988), 120 22. 90 monedelor bizantine," 220, no. 86. Brlad: Oberlnder "Monede bizantine," 230, no. 7 8. Cu rcani: Oberlnder Trnoveanu, "Barbaricum apropiat," 357, n. 48. It was recently argued that the solidus in question was in fact bought by a collector from Curcani, without necessarily being a local find, for which see Vlcu, Les monnaies d'or 13. 91 Coins of Heraclius and Constans II issued in Alexandria, Carthage, Ravenna, and Syracuse found in Moesia Secunda and Scythia, for which see F. Curta, "Byzantium in Dark Age Greece (The Numismatic Evidence in its Balkan Context)," BMGS 29, n. 2 (2005): 124 35; G Oberlnder Trnoveanu, Monnaies byzantines," 119 20. Coins of Constans II and Constantine IV from Syracuse found in Thracia and Haemimons, Tenchova, "Monetna tsirkulatsiia," no. 2323 and 2348. At

PAGE 366

366 Alexandria, one from Carthage and anoth er from Rome, dating from the reigns of Heraclius, Constans II and Constantine IV. 92 If the hoard from Tomis can be connected with the major campaign organized by Constantine IV against the Bulgars in 680, the hoard from Obr barbaricum was concealed in the late 650s and included issues of Alexandria, Carthage and Syracuse, which account for more than one third of the total number of coins in the hoard. 93 This suggests that Constans II was already makin g efforts to re establish a foothold in the northern Balkans by transferring troops and resources from the west. Theophanes informs us that the emperor staged a major campaign against the Slavs in the Balkans in 658 and his political ambitions never to ma terialize, might have included a much wider area up to the Danube, the traditional frontier of the Empire. 94 At least the Black Sea coast was still firmly in Byzantine hands until 680 judging by coin finds between Tomis and Akhtopol and so was perhaps Duros torum on the Lower Danube, although the often invoked hoard of silver coins in fact dates to the sixth century. 95 Late seventh century emperors the National History Museum of Romania in Bucharest I was able to inspect a half follis of Constans II from Carthage (652 657, type MIB 1 98a) found in northern Scythia. 92 Trsors no. 67. 93 Ibid. no. 359. A follis of Constans II from Syracuse was found further north at Pinsk in Belarus, for which see Kropotkin, Klady vizantiiskikh monet 38, no. 299. 94 Theophanes, Chronographia a. 6149, ed. de Boor, 347. 95 Curta, "Byzantium in Dark Age Greece," 124 35. For an updated catalogue of bronze coins, see S. Mihailov, "Seventh to Ei ghth Century Byzantine Bronze Coins from Northeastern Bulgaria," 26 (2008): 77 85. For Nesebur and Burgas, see recently Tenchova, "Monetna tsirkulatsiia," no. 2314 2350. For seventh century coin finds in the region of Durostorum (Silistra), see Oberlnder Trnoveanu, Monnaies byzantines," 97 127. The small silver hoard from Silistra was attributed to Constantine IV in the original pubication and later included in the historical context of the campaign organized by C onstantine IV against the Bulgars in 680. However, the two token siliquae date to the sixth century. F. Curta, "Invasion or Inflation? Sixth to Seventh Century Byzantine Coin Hoards in Eastern and Southeastern Europe," Annali dell'Istituto Italiano di Numi smatica 43 (1996): 169; Somogyi, New Remarks on the Flow," 113. For the redating to the sixth century, see Trsors no. 56.

PAGE 367

367 would continue to follow the same policy as testified by a follis of Tiberius III (698 705) minted in Ravenna fou nd at Berezeni (Vaslui, Romania). Most finds from this late period are located in southern Moldavia where the Empire was probably trying to find allies or recruits in order to re establish its strategic position at the mouths of the Danube. In addition, th e same general area produced a hexagram of Constantine IV found at 96 The most conspicuous development of the second half of the seventh century is the trail of hoards dating from the reign of Constantine IV which can be followed from southern Russia to Slovakia, with the largest concentration recorded in the Danubian latest coin is a hexagram of Constans II, the o ther hoards of Byzantine silver found at belong to the same wide historical context. 97 Since the date nicely dovetails with the events related with the migration of the On ogur Bulgars, most historians interpreted the finds as an illustration of the new power relations in the Lower Danube region. The large quantity of hexagrams reaching the territories north of the Danube was seen either as bribes and gifts for the Bulgars o r stipends sent to Slavic centers at a time when Byzantine emperors were desperately trying to find allies against the rising influence of 96 delor bizantine," 222, no. 137. 97 Byzantinische Fundmnzen Der Awarenzeit (Inns bruck: Universittsverlag Wagner, 1997), la Piua Pietrei," 135 136 (July December 1945): 53. Priseaca: B. Mitrea, "Date noi cu pri vire la secolul VII. Tezaurul de hexagrame bizantine de la Priseaca (jud. Olt)," SCN 6 (1975): 113 25; E. Oberlnder Trnoveanu, "From the Late Antiquity to the Early Middle Ages the Byzantine Coins in the Territories of the Iron Gates of the Danube fro m the Second Half of the 6 th Century to the First Half of the 8 th Century," Etudes Byzantines et Post Byzantines Somogyi, Byzantinische Fundmnzen, 128. n. 21.

PAGE 368

368 the Bulgars. 98 The two interpretations are not mutually exclusive as the geographic location of the hoards indicates t wo different areas of concentration. The hoards from province of Scythia. From the former provinces of the north eastern Balkans we know about a potential hoard from Valea T eilor and three other single finds of hexagrams from the reign of Constantine IV. 99 All these finds must be related to the presence of the Bulgars led by Asparuch in the area of the Danube Delta. On the other hand, the second area of concentration is some t hree hundred kilometers to the west, in Oltenia, demarcated by the Danube to the west and Olt river to the east. They may indeed constitute military payments and gifts sen t to power centers of mixed ethnic composition probably dominated by the Slavs, who constituted potential allies against the Bulgars. belongs to the same cultural horizon. 100 The creation of a Bulgar state in the Balkans after 680 marks the beginning of a new era in the history of the Lower Danube region. For more than two centuries from the dissolution of the Hunnic confederation to the arrival of the Bulgars, the Byzantine Em pire maintained cultural ties with peoples and centers of power located beyond its military frontier on the Danube. Regardless of the ethnic background of communities in Barbaricum some preserving elements of Roman provincial culture, others being mixed 98 Curta, "Invasion or Inflation," 114 16; Somogyi, "Ne w Remarks on the Flow," 115 35. 99 Curta, "Byzantium in Dark Age Greece," 130, no. 72, 74, 78 and 79. Another hexagram was found at Bu rgas on the Black Sea coast, for which see Tenchova, "Monetna tsirkulatsiia," 2343. 100 I. Nestor and C. S. Nicolaescu Germania 22, n. 1 ( 1938): 33 41.

PAGE 369

369 c ultural groups forged under the domination of the Huns and later of the Avars, maintaining contact with Byzantium was crucial for the creation of social status. On the other hand, whether dealing with Gepids, Heruls, Slavs or Cutrigurs, the high level prio rities of the Empire were dictated by political expediency and the need to secure the frontier in the northern Balkans, most of the time through diplomatic payments (tribute, gifts, and bribes) or the ability to create tension in barbaricum by playing off one group against another. Coins are the most chronologically accurate testament of these efforts in the Lower Danube region and together with other categories of finds can offer a broad picture of cultural contact in the frontier region and beyond. 6.2.2 The Land of Silver: Transcaucasia From a certain perspective the Lower Danube and the Transcaucasus seem to share many features. Both were deemed strategically important by the Byzantine Empire and for good reason. The Lower Danube guarded the access to th e provinces of the Balkans and to Constantinople itself which could be reached both by land and by sea. It also controlled access to important trade routes leading north to the Baltic Sea. As a consequence, emperors of the early sixth century saw fit to re build some of the old Roman fortifications on the northern bank of the Danube in order to use them as outposts in barbaricum Quite similarly, Transcaucasia guarded the Caspian Gates through which nomad raiders from the north could launch plunder expeditio ns into Asia Minor (Figure 6 11) The same passes were linking eastern and western tr a de route s including the Silk Road. 101 As a reflection of similar agendas fortresses such as 101 M. Kazanski and V. A. Mastykova, "Centry vlast i i togoviye pouti Zapadnoi Alanii V VI vv.," in Severnyi Kavkaz: istoriko arkheologicheskie ocherki i zametki ed. M. P. Abramova and V. I. Markovin (Moskva: nauk, I n t arkheologii, 2001), 138 61.

PAGE 370

370 Sucidava and Drobeta on the Lower Danube found their strategic equivalents at Pi tyus and Sebastopolis on the north eastern Black Sea coast. The political fragmentation of the region into a mosaic of tribes inhabiting the southern foothills of the Caucasus, subjected to the Lazian king, might not have been very different from the confe deracy of clients established by the Gepids and the Avars in the Danube region. Sources, however, are fa r more abundant for Transcaucasia. For it is the nature of the evidence that sets these two frontier regions apart. Stakes were much higher in Transcauc asia as the whole region became the main theater of operations for both Byzantium and Persia, and indeed often the main bone of contention in peace negotiations between the two superpowers. 102 The political and diplomatic priorities of Byzantium are always r eflected in the amount of space and energy that contemporary writers were willing to devote to the description of foreign peoples. If the Empire launched only two major campaigns north of the Danube during the sixth and seventh centuries, Lazica, Iberia, a nd Armenia saw continuous decades of warfare which feature prominently in the works of Procopius, Menander, and Agathias. The ideological conflict between two archenemies in a region fraught with centuries long political and cultural competition was more f ascinating than obscure events taking place in the post Hunnic barbaricum north of the Danube. Considering the differences between the Lower Danube and the Transcaucasus, the presence of Byzantine and Sasanian coins in Georgia and Armenia has slightly diff erent historical coordinates, although in both cases preserving imperial interests was the main priority. For reasons which have more to do with modern history and 102 For a recent overview, see B. Dignas and E. Winter, Rome and Persia in Late Antiquity: Neighbours and Rivals (New York: Cambridge University Press, 2007).

PAGE 371

371 scholarship another major difference between the two region s regards the recording and prese rvation of numismatic evidence. Although the region, Georgia in particular, produced a number of copper coin finds comparable to that yielded by the Lower Danube, in few cases the information about the finding place has been recorded (Appendix B Table B 2 ). Based on the geographic distribution of known finds we can responsibly speculate that most of the early Byzantine coins now in the Janashia Museum of Georgia were found in the western half of modern Georgia, on the territory of Late Antique Lazica (Figu re 6 12) 103 The chronological structure of coin finds largely corresponds to what we know about coin circulation in a number of Byzantine centers in eastern and north eastern Anatolia, although we do not have at our disposal a large number of publications to match the corpus available for the Danube frontier (Figure 6 13, 6 14, and 6 15) 104 On the other hand, the presence of rare sixth century ceremonial silver miliarens es and the massive quantity of seventh century hexagrams found in Transcaucasia truly ident ify this region as a land of silver, although gold is by no means absent, as a testament of the political stipends sent by Byzantium. A chronological, sometimes year by year, analysis of coin finds is still possible and is 103 The main source of information about hoards and single finds of Byzantine coins in Georgia remains the work of Tamara Abramishvi li, for which see especially T. Abramishvili, Sakartvelos sakhelmts'ipo muzeumis bizant'iuri monet'ebi (Tbilisi: Metsniereba, 1965); T. Abramishvili, Sakartvelos sakhelmts'ipo muzeumis bizant'iuri monet'ebi (1966 1984) (Tbilisi: Metsniereba, 1989). 104 Unfor tunately very few publications exist from eastern Turkey, the provinces of Armenia Prima and Armenia Secunda being closest to Transcaucasia. There is indication of a substantial collection in the museum from Erzurum but just a short synopsis is currently a vailable, for which see H. zyurt zcan, (Ankara: 2007), 1 16. The most reliable remain the catalogues of finds from Amasia and Melitene, although the samples from Sinope and Amasra can be useful for the larger implic ations of circulation in the Black Sea ; see S. Ireland, Greek, Roman and Byzantine Coins in the Museum at Amasya (London: Royal Numismatic Society, 2000); S. Ireland The Ancient Coins in Amasra Museum," in Studies in Ancient Coinage f rom Turkey ed. R. Ashton ( London: Royal Numismatic Society, 1996), 115 37; J. Casey, Sinope. A Catalogue of the Greek, Roman and Byzantine Coins in Sinop Museum (Turkey) and Related Historical and Numismatic Studies (London: Royal Numismatic Society, 201 0); Z. Demirel Gkalp, Malatya arkeoloji mzesi bizans sikkeleri

PAGE 372

372 matched by an almost equally pr ecise sequence of events and developments recorded in contemporary accounts. The first important observation is that the percent age of finds dating from 498 Danube. Unsurpri singly, these are decades of consolidation in Lazica, under the renewed suzerainty of Byzantium. Traditionally the Roman Empire claimed control of the eastern Black Sea coast from Trapezus to Pityus and extended its political influence in the immediate hin terland, which in Late Antiquity would become the kingdom of Albania, to the southeast, were under Persian control. This balance of power, however, had already been upse t in the second half of the fourth century when failure in the East and in the Balkans led to the humiliating death of two emperors, Julian and Valens, respectively. The Roman garrisons on the Black Sea coast fell under Persian sway, vinta ) and Sebastopolis (Sukhumi) were probably restored under Theodosius I. 105 As regards Pityus we at least know that it was the designated place of exile for John Chrisostomos in 407. 106 Even if Theodosius II regained a foothold in Lazica, the Roman positi on was still weak and it relied mostly on religious ties with 105 C. Zuckerman, "The Early Byzantine Strongholds in Eastern Pontus," Travaux et Mmoires 11 (1991): 52 7 40. 106 W. Seibt, "Westgeorgien (Egrisi, Lazica) in frhchristlicher Zeit," in Die Schwa rzmeerkste in der Sptantike und im frhen Mittelalter ed. R. Pillinger, A. Plz and H. Vetters (Vienna: Verlag der sterreichischen Akademie der Wissenschaften, 1992), 139.

PAGE 373

373 Figure 6 11. Early Byzantine Transcaucasia ( adapted after G. Greatrex and S. N. C. Lieu 2002 xxxii, map 5).

PAGE 374

374 Figure 6 12. Early Byzantine coin finds in Trancaucasia (numbers on the map re fer to numbers in Appendix A.2 )

PAGE 375

375 Figure 6 13. The chronology of early Byzantine coin finds from eastern Anatolia and Transcaucasia.

PAGE 376

376 Figure 6 14. Mints and denominations in Transcaucasia (Abbreviations in Appendix B Table s B 4 and B 6). Figu re 6 15. Mints and denominations in eastern Anatolia (Amasia and Melitene) (Abbreviations in Appendix B Table s B 4 and B 6).

PAGE 377

377 peoples of Transcaucasia who had converted to Christianity in the first half of the fourth century. 107 The kingdom of Lazica had strong ties with Sasanian Persia, something that sixth century writers in Byzantium tried to conceal as the political climate of the region turned in their favor. It is revealing, however, that Lazian rulers were appointed by the Persian king, a practice which remained in place in the second half of the fifth century and until 522 when Lazica re entered the Byzantine orbit. In a move typical for Transcaucasian politics in Late Antiquity the Lazian king Tzath decided to renounce his allegiance to Persia and throw in his lot with Byzantium. The Byzantine response was equally traditional: Tzath was baptized by Justin I, received a Byzantine wife, and returned to Lazica as a client king 108 The same decade saw the pacification of the Tzani, a troublesome tribe in the mountains of Pontus. Justinian built fortresses and churches in Tzanica a strategic region securing access to both Lazica and Armenia. In addition, the Iberian king rebelled against the Persian encroachment in local religious affairs and reached for help from the Byzantine emperor. The Byzantine position in Transcaucasia had improved dramatically. 109 One of the most important documents of this period is Novella XXVIII from 535 in which Justinian boasted his submission of the Tzani and his suzerainty ove r the Lazian Empire, a heterogeneous confederation which included the northern tribes of the 107 C. Haas, "Mountain Constantines: The Christianization of Aksum and Iberia," J ournal of Late Antiquity 1, no. 1 (2008): 101 26. 108 Greatrex and Lieu, ed., The Roman Eastern Frontier 80. 109 D. Braund, Georg ia in Antiquity: A History of Colchis and Transcaucasian Iberia, 550 BC AD 562 (New York: Oxford University Press, 1994), 289.

PAGE 378

378 Suani, the Abasgii, the Apsilii, the Scymni, and the Misimiani. Among the cities mentioned are Archaeopolis (Nokalakevi) the Lazian capital, Rhodopolis (Vartsi khe), Petra (Tzikhisdziri) a Justinianic foundation in Lazica, as well as two settlements which the document classifies as fortresses (rather than towns) Pityus and Sebastopolis, attached to the province of Pontus Polemoniacus. 110 These developments find an excellent reflection in coin finds from this period. Although still adhering to the caveat that some coins might have arrived at a much later date after their striking, it is still safe to assume that many of the coins issued during these decades (498 538) arrived in with expanding and consolidating the Byzantine influence in Lazica, the large number of coin finds from his predecessors might reflect earlier efforts, at least since Tzath sided with the Byzantine Empire in 522. Both Pityus and Sebastapolis produced a significant number of finds from Anastasius, while a half follis from Archaeopolis is actually an early reform issue from 507 512. 111 Two coins of Anastasius fo brand new foundation, were probably lost later, although older settlements are known in the area of Tzikishdziri. 112 In addition, one more coin was found at Batumi, possibly the location of ancient Portus Altus, strengthening the im age of a Byzantine presence along 110 Corpus Iuris Civilis Novella XXVIII (535 A.D.), ed. R. Schoell and W. Kroll, vol. 3 (Berlin: Weidmann, 1928), 212 13. 111 Pityus vol. 2, ed. A. Apakidze (Tbilisi: Metsnierieba, 1977), 314 15, no. 11 16 and 605, fig. 11 15. Sebastopolis: S. M. Shamba, Monetnoe obrashche nie na territorii Abkhazii: (V v. do n.e. XIII v. n.e.) (Tbilisi: Metsniereba, 1987), 77 78, no. 111 112. Archaeopolis: T. Abramishvili, "Nokalakevisa da Nokalakevi Arkeopolisi. Arkeologiuri gatkhrebi 1978 1982 ed. P. Zakaraia ( Tbilisi: M etsniereba, 1987), 277, no. 12. 112 I. Tsukhishvili and G. Depeyrot, History and Coin Finds in Georgia: Late Roman and Byzantine Hoards (4th 13th c.) (Wetteren: Moneta, 2003), 22.

PAGE 379

379 the Black Sea coast. 113 The defection of Iberia to Justin I in 525 was followed by an extension of Byzantine ambitions to the south, in Armenia, where Dorotheus and Sittas had led successful campaigns in the years preceding the "eternal peace" sealed in 532. 114 The large number of coin finds from Ani (Kars, Turkey) is probably a good indicator of the Byzantine military presence in Armenia. 115 The reign of Justin I which saw this major shift in political allegiance in Transcaucas ia brought an even larger quantity of coin to Lazica, especially to the main Byzantine strongholds on the coast. Excavations at Pityus produced no less than 18 coins of Justin I of which seven are pentanummia pointing to the fact that the fortress was fir mly connected to the Byzantine monetary system and perhaps to the trade network in the Black Sea. 116 Moreover, a hoard of 55 pentanummia all issues of Justin I, was found near the north wall during archaeological excavations in 1961. 117 The hoard is unique in the Black Sea and the wider Eastern Mediterranean region and the coins were clearly intended for use at the marketplace. Copper coins of Justin I have been found not only close to the coast, at Sebastopolis and Archaeopolis, but also in the hinterland, at Kutaisi and Urbnisi. 118 The specimen from Sebastopolis is an issue of 113 Abramishvili, Sakartvelos sakhelmts'ipo muzeumis (1966 1984) 11, no. 29. For the location of Portus Altus, see Braund, Georgia 275. The Byzantine presence on the north eastern Black Sea coast was also confirmed by a spectacular silver plate probably found in a fortress or cemetery in Abkhazia, for which see Y. Pyiatni sky, "New Evidence for Byzantine Activity in the Caucasus During the Reign of the Emperor Anastasius I," AJN 18 (2006): 113 22. 114 Greatrex and Lieu, eds, The Roman Eastern Frontier 94 96. 115 Mousheghian et al., History and Coin Finds in Armenia. Coins from Ani, Capital of Armenia (4th c. BC 19th c. AD) (Wetteren : Moneta, 2000), 72, no. 19 23. 116 18, no. 17 27 and 606 607, fig. 19 24, 26 28. 117 Tsukhishvili and Depeyrot, History and Coin Finds in Georgia 72 73, no. 4. 118 Sebastopolis: Shamba, Monetnoe obrashchenie 78, no. 113. Archaeopolis: Abramishvili,

PAGE 380

380 Antioch, a rare find in the Black Sea region, which points to the transfer of troops on the large theater of operations from the Caucasus to the Euphrates. The ceremonial silver heavy mil iarensis found at Zugdidi, close to the Lazian capital of Archaeopolis, is a clear reflection of diplomatic gifts necessary to cement the new alliance between Byzantium and Lazica. 119 Four solidi of which one has a known finding place at Gveso (Racha Lekhum i and Kvemo Svaneti, Georgia), may reflect similar political payments dating from the early stage of the Byzantine presence in Lazica. 120 Such practices were continued and even intensified in the first decade of were far more ambitious. Although Lazica administrative and political structures of Armenia and Lazica made him extremely unpopular with the local princes. The Lazi con federation of clients was particularly fragile and relied on the loyalty of the northern tribes, which was hardly unequivocal. The Abasgi, the Apsilii, the Mismiani, and the Suani, whose main strategic purpose was to guard the two major passes of Klukhor a nd Marukha in the western Caucasus, were lavished with occasional gifts to insure their pro Byzantine position. Later events would prove that their loyalty should not be taken for granted. In the 530s the Abasgi were "Nokalakevisa da nojikh 78, no. 13 15. Kutaisi: Tsukhishvili and Depeyrot, History and Coin Finds in Georgia 24, no. 2. Urbnisi: Abramishvili, Sakartvelos sakhelmts'ipo muzeumis 123, no. 33a. 119 Abramishvili, Sakartvelos sakhelmts'ipo muzeumis 40 41, no. 56 Ceremonial silver coins are a far more reliable chronological marker, since they were distributed by the emperor. They had no economic function and therefore it is likely that they arrived to their final destination immediately after they were issued. Co ntra Tsukhishvili and Depeyrot, History and Coin Finds in Georgia 25, who state that silver coins were intended for the foreign market. 120 M. Tsotselia, Coin Finds in Georgia: (6th Century BC 15th Century AD) (Wetteren: Moneta, 2010),130, no. 692; Abramishvili, Sakartvelos sakhelmts'ipo muzeumis 40, no. 53 55.

PAGE 381

381 converted to Christianity and their dua l monarchy abolished. 121 This decision did not come for free. Two ceremonial silver miliarensia dated between 527 and 538 were found in Apsilia and Abasgia, at Akhali Atoni and Tsebelda, respectively, while a solidus azia) might date from the same period. 122 A light miliarensis from Sochi (Krasnodar, Russia) has a wider dating (537 565) but may belong to the same historical context. 123 In addition, the Byzantine fortresses of the north eastern coast became more integrated into the Mediterranean world judging by the quantity of amphorae found in settlements of the coast and the finds of coins from Antioch, Alexandria, and Carthage at Pityus and Akhali Atoni in Abkhazia. 124 One of the folles found at Pityus is an imitation typi cal of the Balkans and may have been brought by the Bulgar" prisoners taken by Mundus and sent to Lazica and Armenia to defend 121 G. Greatrex, "Byzantium and the East in the Sixth Century," in The Cambridge Companion to the Age of Justinian ed. M. Maas (New York: Cambridge University Press, 2005), 497. 122 Akhali Atoni: Shamba, Monetnoe obrashchenie 78 Klady vizantiiskikh monet, 46, no. 477. Tsebelda: Shamba, Mon etnoe obrashchenie 79, no. 116. Two miliarensia and a solidus of Justinian were part of a belt found in a grave from Tsibilium. For the Abkhazii," Kra tkie Soobshcheniia 128 (1971): 100 105. For the dating of the assemblage, see C. Blint, Awarenforschungen ed. F. Daim (Vienna: Institut fr Ur und Frhgeschichte der Universitt Wien, 1992), 378 80; N. Khrisimov, Ein Rekonstruktionsversuch," Archaeologia Bulgarica 8 (2004), no. 1: 79 96. 123 Kropotkin, Klady vizantiiskikh monet 22, no. 22. 124 D. Braund, "Procopius on the Eco nomy of Lazica," The Classical Quarterly 41, no. 1 (1991), 223 with the older literature; Iu. N. Voronov and O. Kh. Bgazhba, issledovanii Tsibiliuma v 1978 1982 gg.) (Tbilisi: Metsniereba, 1985); N. Inaishvili and N. Vashakidze, "Typology and Chronology of Greek, Roman and Early Byzantine Amphorae from Petra Tsikhisdziri," in PATABS I: Production and Trade of Amphorae in the Black Sea: actes de la table ronde internationale de Batoumi et Trabzon, 27 29 avril 2006 e d. D. Kassab Tezgr and N. Inaishvili (Paris: Institut franais Georges Dumezil, 2010), 151 52. M. Zamtaradze, "Nokalakevshi aghmochenili amforebi," in Nokalakevi Arkeopolisi: III. Arkeologiuri gatchrebi 1983 1989 ed. P. Zakaraia (Tb ilisi: Metsniereba, 1993), 158 75. Three unprovenienced coins of Chersonesos in Abramishvili, Sakartvelos sakhelmts'ipo muzeumis, 49 51, no. 100 no. 42 43. Akhali Atoni: Shamba, Monetnoe obrashchenie 81, no. 123. Interestingly, no coins from African mints have been found at Sinope and Amasra, on the southern Black Sea coast, for which see Casey, A Catalogue of the Greek, Roman and Byzantine Coins 79 86; Ireland The Ancient Coins in Amasra," 132 35

PAGE 382

382 the frontier. 125 The coins from North Africa may well reflect trade but also the movement of people such as the Tzani who fought Justinian was also aiming further east in Iberia where a Byzantine presence can be felt in these decades, judging by the coin finds. Copper coins of Anastasius, Justin I, ustavi, Dmanisi, and Mtskheta. 126 A silver miliarensis pierced to be worn as a pendant was found in a grave from the Iberian capital of Mtskheta, possibly pointing to diplomatic gifts sent to Iberian rulers. 127 Gold coins of Anastasius had already reached the region controlled by Alans who guarded the Dariel Pass (Alan Gates) and Justinian went a step further by securing the allegiance of queen Boa of the Sabir Huns won over with gifts of imperial raiment and a variety of silver vessels and not a little money. 128 The money was well spent as the powerful queen defeated other Hunnic groups allied with the Persians. 129 Byzantine diplomatic activity is also evident northwards on the Black Sea coast, towards the Taman peninsula and Bosporus. By 520 Bosporus was under t he rule of Grod, the pro Byzantine leader of the Huns who dominated the region in cooperation with the Tetraxite Goths. The study of burial assemblages from the region points to strong 125 Abramishvili, Sakartvelos sakhelmts'ipo muzeumis (1966 1984) 16, no. 49. For the events, see Theophanes, Chronographia a. 6033, ed. de Boor, 219. The episode was placed around 530 after the defec tion of Iberia and the subjuction of the Tzani. 126 Sakartvelos sakhelmts'ipo muzeumis 39, no. 49. Rustavi: Ibid., 123, no. 33. Dmanisi: Ibid., 42, no. 63; Abramishvili, Sakartvelos sakhelmts'ipo muzeum is (1966 1984) 13, no. 36. Mtskheta: Kropotkin, Klady vizantiiskikh monet 44, no. 424. 127 Abramishvili, Sakartvelos sakhelmts'ipo muzeumis 44, no. 74. 128 Malalas, Chronographia 18.13, ed. L. Dindorf (Bonn: E. Weber, 1831), 431; trans. E. Jeffreys, M. Jeff reys and R. Scott (Melbourne: Australian Association for Byzantine Studies, 1986), 249. 129 Greatrex and Lieu, ed., The Roman Eastern Frontier 82.

PAGE 383

383 cultural ties between Bosporus and the northern Caucasus. 130 Further inte rnal turmoil in Bosporus was swiftly resolved by Justinian who sent an army across the Black Sea to retake the city in 534 535. Two solidi of Justin I found at Dzhiginka (Krasnodar, Russia) and Goryachi Klyuch (Krasnodar, Russia) may be related to these de velopments. The specimen from Dzhiginka is in fact a rare joint reign issue of Justin I and Justinian I struck in April August 527. The ceremonial nature of the coinage was doubtlessly recognized by the owner who attached the coin to a gold necklace with p recious stone pendants. 131 The successful Persian invasion of Byzantine Syria in 540 was not without who forced the Lazi to accept import duties and to s upport the Byzan tine troops by e nsuring the provision of supplies at low prices, king Gobazes turned to Chosroes for help. The Persians invaded Lazica and took the main Byzantine center of Petra, while the Byzantine garrisons from Abasgia withdrew from Pityus and Sebastop olis after razing the fortresses to the ground. Coins of Justinian found in the burned layer at 132 Sebastopolis was later rebuilt according to the same author who no longer mentions Pityus in his list, although coin finds f rom the second half of the sixth century prove that the fortress remained in use. 133 The Persian 130 I. O. Gavritukhin and M. Kazanski, "Bosporus, the Tetraxite Goths, and the Northern Caucasus Region during t he Second Half of the Fifth and the Sixth Centuries," in Neglected Barbarians 83 136. 131 Dzhiginka: Kropotkin, Klady vizantiiskikh monet 21, no. 9 and fig. 14. Goryachi Klyuch: Ibid., 21, no. 7. On finds of joint reign solidi of Justin I and Justinian I, see W. E. Metcalf, "The Joint Reign Gold of Justin I and Justinian I," in Studies in Early Byzantine Gold Coinage 19 27. 132 Tsukhishvili and Depeyrot, History and Coin Finds in Georgia 21. For the events, see Greatrex and Lieu, ed., The Roman Eastern Fron tier 115. 133 Tsukhishvili and Depeyrot, History and Coin Finds in Georgia 21, 25, associate the scarcity of finds

PAGE 384

384 king tried to impose Zoroastrianism by force in his intention to burn the most important cultural bridge uniting Byzantium and Lazica. In addition, Lazica found itself isolated from the Black Sea network, which affected its economy. Soon enough Gobazes II realized that Chosroes was even less accommodating than Justinian and he decided to make amends and reach back to the Byzantine emperor. War resumed in Lazica in 547. In his usual manner Justinian promised three centenaria of gold (21,600 solidi ) to the Alans and the Sabir Huns to attack Persian Iberia. 134 Three years of warfare did not settle the matter as the Persians still held Petra despite a number of victories won by the Byzantine Lazian armies. Another episode occurred in 550, typical of Transcaucasian politics. Byzantine taxation and political intervention in Abasgia became so unpopular that the Abasgi asked Chosroes for help. The Persians advanced from Iberi a then through Apsilia and restored the Abasgi monarchy under Persian suzerainty. In addition, a Lazi defector offered the Apsili fortress of Tzibile to the Persians. 135 Although Justinian finally recaptured and destroyed Petra in 551 and punished the rebell ious Abasgi, the Persian army led by Mermeroes was in control of eastern Lazica and Suania. An assault on western Lazica was successfully repelled after Justinian transferred more troops to the Transcaucasus. 136 The murder of king Gobazes II in 555 made the Byzantine position in Lazica even more tenuous as the Lazi withdrew their from the second half of the sixth century at Pityus with a general decline in mint output throughout the empire, which is obviously wrong, a nd can be refuted by looking at finds from any Byzantine province from the Danube to Palestine 134 Procopius, Bella II, 29.27 32, ed. Haury and Wirth, vol. 1, 293 94; trans. Dewing, vol. 1, 537. For the Byzantine centenarion see G. Dagron and C. Morrisson "Le Kentnarion dans les sources byzantines," RN 17 (1975): 145 62. 135 Braund, Georgia in Antiquity 300. 136 Greatrex and Lieu, ed., The Roman Eastern Frontier 118 21.

PAGE 385

385 military support, followed by a well timed Tzani revolt in 558. 137 There are only three coin finds dated between 558 and 565, but like in the case of the Lower Danube the explanation h Coin finds in Transcaucasia reflect increased warfare between 541 and 562 and the precarious position of Byzantium, which roughly corresponds with the dated coinage of Justinian (538 5 65). 138 The heavy series of 538 542, which is so well represented in the Lower Danube region, was not equally popular in Transcaucasia. To be sure, the period 538 544 produced almost the same number of finds as the last two decades of 565). Coins from the early years after the 538 reform have been found in Armenia, at Ani and Vanadzor and in Abkhazia on the Black Sea coast at Pityus and Sebastopolis. 139 The two important fortresses had been destroyed and abandoned in 541, as we have seen, but Procopius might have exaggerated this episode. At least at Sebastopolis five post reform issues of Justinian have been retrieved, of which three specimens date from the 540s. To be sure, some of them might have been lost later, together with coins of Justin II. The numismatic reflection of diplomatic initiatives during these decades of conflict left a few important traces in western Georgia. After the ill timed murder of Gobazes II, Justinian saw fit to entrust general Soterichus with distributing 4 c entenaria of gold (28,800 solidi ) among barbarian tribes from the Lazian confederation. The 137 Ibid., 122; Braund, Georgia in Antiquity 308. 138 Hoards of Sasanian drachms from Iber ia (Tolenji, Dusheti and Urbnisi) concealed during these decades must be associated with the same events; see M. Tsotselia, Hi story and Coin Finds in Georgia: Sasanian Coin Finds and Hoards ( Wetteren: Moneta, 2003), 49 57. 139 Ani: Mousheghian et al., History (Ani) 73 74, no. 32 33. Vanadzor: Ibid., 180, no. 2. Pityus: 22, no. 44 and 49. Sebastopolis: Shamba, Monetnoe obrashchenie 80 81, no. 122, 124 and 125.

PAGE 386

386 stakes were raised when the Misimians killed Soterichus and helped themselves to the money. A Roman army under generals Martin and John Dacnas punished the rebels a nd retrieved the gold. 140 Indeed, gold was not in short supply in Transcaucasia. John Dacnas was entrusted with 400 pounds of gold to be distributed to worthy soldiers fighting in Misimia, just as a few years before when the Byzantine army was struggling to to Lazica with gifts ( dora ) for the soldiers. 141 The dispersed hoards of solidi of Justinian from Chkhorotsqu (Samegrelo Zemo Svaneti, Georgia) and Kobuleti (Ajaria, Georgia) m ay belong to this context. 142 Two ceremonial silver miliarens es and one solidus found time when their neighbors, the Misimiani, had rebelled, the Abasgi had been barely pacified after their defection to Persia, and the Suani were yet to be punished after a similar defection in 552. 143 Justinian was in fact diplomatically engaged on the entire northern Black Sea front as he was striving to convince the Utrigur Huns and the Tetraxite Goths to attack the Kutrigurs in 551. The hoard from Ilych in Krasnodar made 140 Agathias, Historiae IV, 12 20, for the episode and IV, 20.9, for the sum of money involved, ed. R. Keydell (Berlin: De Gruyter, 1967); trans. J. D. Frendo (Berl in: De Gruyter, 1975), 113 22 141 Agathias, Historiae III, 2.3 4 and IV, 17.3, ed.Keyd ell; trans. Frendo, 69 and 118. 142 Abramishvili, Sakartvelos sakhelmts'ipo muzeumis 124, no. 42; Kropotkin, Klady vizantiiskikh monet 45, no. 444. 143 Voronov and Iushin, "Pog Monetnoe obrashchenie 79, no. 115. For the Suani, see recently G. M. Berndt, "Shifting Frontiers in the Caucasus Mountains: The Suani," in Shifting Cultural Frontiers in Late Antiquity ed. D. B rakke, D. Deliyannis, and E. Watts (Burlington, VT: Ashgate, 2012), 255 70; The Byzantine army used the fortresses of Apsilia as bases to attack Misimia, for which see Braund, Georgia in Antiquity 310. For archaeological evidence from the region, see M. K Empire," Travaux et Mmoires 11 (1991): 488 93.

PAGE 387

387 of 140 third to fourth century gold issues of Bosporan kings and solidi of Justinian hidden in an amphora may reflect such efforts. 144 Peace followed in 562, after Dagisth eus, magister militum per Armeniam won a major victory at Phasis (Poti) near the Black Sea coast in Lazica. The theater of operations moved southwards into Armenia during the reign of Justin II. In 571 the Armenians revolted against the Persian marzban wh o tried to impose Zoroastrianism in the Armenian capital of Dvin and limit the religious freedom of Armenians long granted by Shapur III (383 388). Vardan, the leader of the Armenian rebellion received help from Justin II and managed to take control of Dui n. In a manner typical of Transcaucasian politics, Iberia also revolted against Persia and between 572 and 574 the Roman armies advanced deep into Albania towards the Caspian Sea. The turn of events was so surprising that John of Biclar writing from distan t Spain was convinced that the emperor Justin made Armenia and Iberia Roman provinces. 145 The reality was, however, quite different and by 580 Persia regained the initiative in Transcaucasus. The situation was so desperate for the Romans that Tiberius II w as forced to recruit 15,000 Germans from the northern Balkans and send them to the East. 146 The transfer of troops from the Balkans is suggested by the significant number of half folles of Thessalonica found in 144 N. A. Frolova and E. Ia. Nikolaeva, "Il'ichevskii klad monet 1975 g.," VV 39 (1978): 173 79. For the historical and archaeo logical evidence, see M. Kazanski and A. Mastykova, Les peuples du Caucase du er VII e sicle apr. J. C.) (Paris: Editions Errance, 2003), 141. For the gifts sent by Justinian to the Utrigurs, see John of Antioch, Historia Chr onike 217, ed C. Mller FHG 4 (Paris: A. Firmin Didot, 1851), 621 22. 145 John of Biclar, Chronica 20 21, ed. T. Mommsen, Chronica Minora 2 (Berlin: Wedmann, 1894), 212 ("Iustinus imperator Armeniam et Hiberiam repulsis Persis Romanas provincias facit"). F or the events, see Greatrex and Lieu, ed., The Roman Eastern Frontier 138 49. 146 Greatrex and Lieu, ed., The Roman Eastern Frontier 151.

PAGE 388

388 Georgia, given that the Macedonian mint was s upplying the troops from the Danube frontier (Appendix B, T able B 9). 147 Coin finds of Justin II are abundant in Transcaucasia, especially copper issues. His coins were found during excavations at Ani and Dvin and the majority were struck by the mint of Nico media, which was probably commissioned to support the war effort in Armenia. 148 Coins from Antioch were found at Vardenut (Aragatsotn, Armenia) and Vosketap (Ararat, Armenia), also present at Melitene, a major Roman base during this decade of conflict. 149 On t he other hand, things were quiet in Lazica and we can envisage a period of reconstruction and consolidation as testified by coin finds from Pityus, Archaeopolis, Kutaisi and from the many more specimens without a known finding place in Georgia. 150 A hoard of copper coins from Archaeopolis spanning the reigns of Anastasius to Justin II might have been concealed during renewed insecurity in Lazica in 574 575. 151 A half follis from a mobile military mint striking coins imitating the mint of Rome was found at Akhal tsikhe (Samtskhe Javakheti, Georgia) close to the 147 Thessalonica is less well represented in finds from Sinope, Amasia and Melitene, for which see Casey, A Catalogue of Greek, Roman and Byzantine 79 86; Ireland The Ancient Coins in Amasra," 132 35 ; Demirel Gkalp, Malatya arkeoloji," forthcoming. 148 Mousheghian et al., History (Ani) 74 75, no. 42 51; Mousheghian et al., History and Coin Finds in Arme nia: Coins from Duin, Capital of Armenia (4 13th c.); Inventory of Byzantine and Sasanian Coins in Armenia (6 7th c.) (Wetteren, Moneta, 2000), 62, no. 19 24. Contra Tsukhishvili and Depeyrot, History and Coin Finds in Georgia 19, who believe that the min t of Nicomedia had lost its importance since the reign of Justinian I. 149 Demirel Gkalp, Malatya arkeoloji," no. 184 193. Melitene was sacked by the Persians in 576, for which see R. W. Thomson, "Eastern Neighbours: Armenia (400 600)," in The Cambridge Hi story of the Byzantine Empire c. 500 1492 ed. J. Shepard (New York: Cambridge University Press, 2008), 168. 150 24, no. 50 53 and 609 10, fig. 50 53. Archaeopolis: Abramishvili, "Nokalakevisa d a nojikhevis monet'ebi," 278, no. 17 18. Kutaisi: Tsukhishvili and Depeyrot, History and Coin Finds in Georgia 24, no. 4. Georgia: Abramishvili, Sakartvelos sakhelmts'ipo muzeumis 50 52, no. 105 107 and 109 120. 151 Greatrex and Lieu, ed., The Roman Easter n Frontier 153.

PAGE 389

389 Mtkvari river and confirms the movement of troops from the west. 152 Pityus was perhaps the main base for such movements as another half follis of Tiberius II or Maurice from the military mint labeled Rome w as found during archaeological excavations in the fortress. 153 The find might be connected with the war effort in Albania and the revolt of Iberia supported by Justin II. A follis of 570/1 found in Iberia at Dmanisi (Kvemo Kartli, Georgia) is perhaps related to the same events. 154 As we have already seen Justin II was not very fond of sending gold to barbarians. However, the balance of power in the steppe north of the Caucasus was changing dramatically. By the end of his reign the Turks had become the masters o f this wide region after defeating the Sabir Huns, the Alans and the Utrigurs. 155 Solidi of Justin II and his predecessors as well as a copper coin of Justin II have been found in the Alan cemetery of Kamunta (North Ossetia Alania, Russia), a gold plated im itation was found in the cemetery from Mokraya Balka, and a solidus of Justin II with a suspension loop was part of the inventory of a grave from Pechanka (Kabardino Balkaria, Russia). 156 An exceptionally rare find from Mokraya Balka, a light weight solidus of 22 siliquae from the joint reign of Justin II and Tiberius II (26 September 578 4 October 578), pierced twice, reminds of the find from Dzighinka, and probably has the 152 Abramishvili, Sakartvelos sakhelmts'ipo muzeumis (1966 1984) 21, no. 73 with n. 114. 153 Ibid., 19, no 63. 154 Ibid., 20, no. 71 and pl. IV/71. A hoard of Sasanian drachms from Mtskheta may be related to the same events; see Tsotselia, History and Coin Finds in Georgia 61 64. 155 Kazanski and Mastykova, Les peuples du Caucase du Nord 150. 156 Kamunta: Kropotkin, Kla dy vizantiiskikh monet 30, no. 138/3, 11. Pechanka: Ibid., 30, no. 132. Mokraya Balka: E. V. Rtveladze and A. P. Runich, "Novye nahodki vizantiiskikh monet i indikatsii v okrestnostiakh Kislovodska," VV 37 (1976): 153.

PAGE 390

390 same significance. 157 All these finds, as well as others from the Kislovodsk basin, m ay be part of the political and diplomatic context of the 570s when Justin II managed to finally punish the Suani who had switched sides in favor of Persia two decades earlier. 158 Furthermore the Romans captured the Suani king and brought him as a captive t o Constantinople together with his royal treasure. 159 No doubt, the protection of the Dariel Pass remained a major strategic priority for both Byzantium and Persia and this much can be deduced from the fact that both Byzantine and Sasanian coins and artifact s are found in such cemeteries. 160 However, many gold coins date from the region during his reign. ontrol over Iberia and Albania. For almost a decade war would continue with fierce battles in the southern sector of the frontier, quite far from the Caucasus. Nevertheless, the region remained strategically important in the larger scheme. In 589 Maurice p aid the Iberians to invade Albania, trying to take advantage of the Persian weakness in the region after they were forced to transfer troops to the east to face an invasion of the Turks. 161 A solidus of Maurice from Saskhari (Mtskheta Mtianeti) may belong to this political 157 Rtveladze and Runich, "Novy e nahod ki," 151 and fig. 1/1 2. 158 Y. A. Prokopenko, "Byzantine Coins of the 5 th 9 th Century and their Imitations in the Central and Eastern Ciscaucasus," in Byzantine Coins in Central Europe Between the 5th and 10th Century ed. M. ces, 2009), 545 550 and fig. 1. 159 Berndt, "Shifting Frontiers," 269, discussing an account of John of Biclar. 160 Kazanski and Mastykova, Les peuples du Caucase du Nord 565. 161 Greatrex and Lieu, ed., The Rom an Eastern Frontier 171.

PAGE 391

3 91 context. 162 Returning general Bahram retaliated by launching a campaign against the Suani but his efforts were thwarted by a major defeat to the south, close to Araxes river. The subsequent rebellion of Bahram against Hormizd and the rightful heir Chosroes successful campaign against Bahram and the restoration of Chosroes under the terms imposed by Maurice meant that Byzantium would gain unprecedented influence in the east. In exchange for "a massive sum of money in addition to the military alliance" t he Persian king agreed to give up control of Persarmenia as far as Dvin and Lake Van and of Iberia as far as Tiflis. 163 Relations with Persia remained good until the death of Maurice, especially since the emperor was determined to deal with the Avars in the Balkans while Chosroes II had to establish his rule in Persia. 164 Coins of Maurice are found in fair numbers in Transcaucasia, particularly issues from the 590s afte r the war with Persia was brought to an end. Coins from Antioch are particularly abundant in Armenia at Ani and in Georgia at Pityus and Kutaisi. 165 Pityus continued to receive coins from Thessalonica in significant numbers signaling the continuous traffic f rom the Balkans. Two copper coins from 589/90 and 594/5 have been found in the most important centers of Iberia, at Mtskheta and Tbilisi, respectively, 162 Abramishvili, Sakartvelos sakhelmts'ipo muzeumis 55, no. 131. 163 Theophylact Simocatta, Historia V, 2.6, ed. de Boor and Wirth, 191; trans. M. Whitby and M. Whitby (Oxfor d: Clarendon Press, 1986), 134. 164 Whitby, The Emperor Mau rice 292 304. 165 Ani: Mousheghian et al., History (Ani) 75 76, no. 53 26, no. 54 57 and 610 11, fig. 54 57; Abramishvili, Sakartvelos sakhelmts'ipo muzeumis (1966 1984) 23 24, no. 81, 83 Kutaisi: Tsukhishvili and Depeyrot, History and Coin Finds in Georgia 24, no. 6.

PAGE 392

392 terms accepted by Cho sroes. 166 A small hoard found at Nekresi (Kakheti, Georgia) comprised of two silver Sasanian drachms and one siliqua of Maurice is another testament to the mixed Byzantine Persian influence in Iberia. 167 A silver light miliarensis and three solidi have no reco rded finding place, but they can be situated in the same historical context. 168 Although peace was restored in Transcaucasia and the emperor diverted his attention to the Balkans, a consolidation of the Roman presence in Lazica must have remained an importan t priority. A hoard of 23 solidi found during archaeological excavations at Archaeopolis is extremely important for understanding sources. Many of the solidi are die linked which means that they were brought fresh from the mint. The fact that they were not regular solidi but light weight issues worth 23 siliquae suggests a political payment sent to the Lazian capital. 169 A copper hoard found in Ochamchire (Abkhazia, Georgia), whose context was unfortunately poorly recorded, contained some 58 bronze coins from Justin I to Maurice, the last issue being dated 601/2. 170 Its concealment during the events following the death of Maurice seems plausible. 166 Abramishvili, Sakartvelos sakhelmts'ipo muzeumis 125, no. 58; Abramishvili, Sakartvelos sakhelmts'ipo muzeumis (1966 1984) 21, no. 74 and 23, no. 79. 167 Tsotselia, Coin Finds in Georgia 142, no. 755. 168 Abramishvili, Sakartvelos sakhelmts'ipo muzeumis (1966 1984) 21, no. 75 and pl. IV/75; Abramishvili, Sakartvelos sakhelmts'ipo muzeumis 54 55, no. 130, 132 and 134. 169 Tsukhishvili and Depeyrot, History and Coin Finds in Georgia 74, no. 5 and pl. 1; T. Ia. Abramishvili, "Nokalakevskii klad," VV 23 (1963): 158 165. 170 Akad. S. Janashias 24 B (1963): 57 59 and pl. I/1 6.

PAGE 393

393 War with Persia resumed during t he reign of Phocas and would continue for more than two decades. The deposition and murder of Maurice and his family was the perfect excuse for Chosroes II, whose gratitude towards Constantinople was already growing thin in the decade after his restoration to the Persian throne. The main objective was to reconquer Armenia but as Persian forces advanced deep into Byzantine territory the road was open for a much more ambitious expedition in the Byzantine heartland. 171 Phocas had already sealed a new deal with t he Avars in 604 and was probably trying to gain the loyalty of Caucasian tribes judging by the finds of gold and silver coins from the area. A ceremonial siliqua Georgia), and a solidus at Dranda in Abkhazia. 172 Grave finds f rom cemeteries such as Chmi and Kamunta in North Ossetia Alania included solidi of Phocas, sometimes pierced to be worn as jewelry. 173 Two light siliquae of Maurice and Phocas, respectively, were found in a grave from Mokraya Balka, indicating that such paym ents were also sent north of the Caucasus. 174 The Byzantine emperor was probably trying to compensate for the weak defense of the region, most of his troops no doubt having been moved to the main theater of operations in the south. It is important to note th at no copper coins of Phocas are so far recorded on the eastern Black Sea coast, although there are indeed quite a few unprovenienced coins of Phocas in Georgian museums. 175 171 Greatrex and Lieu, ed., The Roman Eastern Frontier 186 87. 172 Abramishvili, Sakartvelos sakhelmts'ipo muzeumis (1966 1984) 24, no. 86 and pl. VI/86; Kropotkin, Klady vizantiiskikh monet 43, no. 408. 173 Kropotkin, Klady vizan tiiskikh monet 31, no. 142/2 3; Ibid., 30, no. 138/6. 174 Rtveladze and Runich, "Novye n ahodki," 151 53 and fig. 1/3 5. 175 Abramishvili, Sakartvelos sakhelmts'ipo muzeumis 61 62, no. 170, 172 173, 176 180.

PAGE 394

394 in Byzantine history. Within a decade the Persians conquered Egypt and brought a trail of destruction in Syria Palestine and Asia Minor, while the Avars and the Slavs ravaged the Balkans. Heraclius contemplated moving the capital to Carthage but ended up by melting down church property to produce a brand new coinage, the hexagram, which he intended to use for paying the army and the administration at half the old rate. 176 The Byzantine recovery started in Transcaucasia and was made possible not only by the empe with the Avars in 620 Heraclius transferred the troops from the Balkans to Asia Minor. A f irst major achievement was the sacking of Dvin in 624 but then the war moved to the north. Moses of Dashkura tells us about the diplomatic initiatives of Heraclius seeking to gain the loyalty of Caucasian tribes as he was struggling to outmaneuver three su perior Persian armies. 177 The deal with the Turks proved decisive as the nomads poured through the Caspian gates and invaded Albania in 626. Having joined his Turkish allies at Tiflis in the spring of 527 Heraclius proceeded to the south through Armenia to a chieve a final and complete victory against Persia. 178 The coins of Heraclius found in Transcaucasia date mostly from this period of warfare. A first observation is the drastic reduction in the number of copper coins. 179 176 Greatrex and Lieu, ed., The Roman Eastern Frontie r 187 97. 177 Ibid., 202 05. 178 J. Howard 622 630," War in History 6, no. 1 (1999): 1 44. 179 Contra Tsukhishvili and Depeyrot, History and Coin Finds in Georgia 24 who argue that copper coins of the seventh century "vanished in West Georgia."

PAGE 395

395 Pityus continued to receive copper issu es at least until 630 but no other coins with a known finding place can be mentioned, although a dozen or more specimens are preserved in Georgian museums. 180 There is only one significant find from the hibati (Guria, Georgia) which contained some 2,000 coins of which only 124 have been preserved. 181 Much like the hoard of Maurice from Archaeopolis, which lies at a close distance to the northeast, most of the coins from Chibati are die linked. The coins of Phocas cover almost 85 percent of the hoard, which is an important indication of the fact that the hoard was shows that both Phocas and Heraclius were interested in m aintaining the support of Lazica. Moreover, their actions can be felt on a much larger front along the northern Black Sea coast, judging by the finds from Sennaya and Starodzhereliyevskaya in Krasnodar. 182 Local imitations of Byzantine silver and gold coins of Phocas and Heraclius from the Kislovodsk basin reveal the existence of Byzantine diplomatic initiatives in the Northern Caucasus, as well the creation of social status in relation to Byzantium. 183 180 Abramishvili, Sakartvelos sakhelmts'ipo muzeumis 64 68, no. 188 194, 196 197, 199 204, 206 211 and 213 214; Abramishvili, Sakartvelos sakhelmts'ipo muzeumis (1966 1984) 33, no. 127. 181 Akad. S. Janashias sachelobis 25 B (1968): 159 76; Tsukhishvili and Depeyrot, History and Coin Finds in Georgia 75 79 and pl. 2 6. We might ad d a contemporary hoard of Sasanian drachms from Urbnisi (Shida Kartli, Georgia) concealed after 617, for which see Tsotselia, H istory and Coin Finds in Georgia 77. 182 Kropotkin, Klady vizantiiskikh monet 22, no. 19a; Ibid., 22, no. 25. 183 Rtveladze and Runich, "Novye nahodki," 153; E. V. Rtveladze and A. P. Runich, "Nahodki indikatsii vizantiiskikh monet vblizi Kislovodska," VV 32 (1971): 220 and fig. 1/3.

PAGE 396

396 Copper coins are more common at Ani and Dvin in Armenia, w hexagrams are also abundant. Unsurprisingly the silver issues dated between 625 and 629 are most numerous in Armenia as this interval coincides with the Byzantine counteroffensive led by Heraclius and his Turkish allies. 184 Most hoards of the period, however, were concealed in Georgia, where Heraclius conducted an aggressive diplomatic activity aimed at gaining the support of the local tribes. Moses of Dashkura informs us that Heraclius sent a letter to convince Caucasian princes to join his c ause but a lot more than words was needed to build an alliance. It was also here that Heraclius bought the loyalty of the Turks after pledging to satisfy the thirst of the savage, gold loving people of long hair, in the words of the same chronicler. 185 The re are no gold coins in the region which can be dated to the 620s, aside from the solidus from Didi Chqoni (Samegrelo Zemo Svaneti, Georgia) in Lazica. 186 There are, however, many hexagrams found in Georgia, both single finds and ho ards most of which concent rate in Persian Iberia rather than in Lazica. In Iberia Heraclius had to fight against Stephen I who remained in communion with the church in Constantinople but chose a political alliance with Chosroes II. 187 Single hexagrams have been found in 184 Ani: Mousheghian et al., History (Ani) 77, no. 66 73. Dvin: Mousheghian et al., History 63, no. 26 51. Yerevan : Ibid., 197. 185 Regarding the attitude of Moses of Dashkura towards nomadic peoples of the steppes, see J. Howard J ohnston, "Armenian Historians of Heraclius: An Examination of the Aims, Sources and Working Methods of Sebeos and Movses Daskhurantsi," in The Reign of Heraclius (610 641): Crisis and Confrontation ed. G.J. Reinink and B. H. Stolte (Leuven : Peeters, 2002) : 41 62. 186 Abramishvili, Sakartvelos sakhelmts'ipo muzeumis 64, no. 186. 187 Greatrex and Lieu, ed., The Roman Eastern Frontier 17 9 and 209.

PAGE 397

397 (Mtskheta 188 Two small hoards found at Mtskheta and Sarachilo (Kvemo Kartli, Georgia) were concealed during the same events. 189 A third hoard of the period found at Odishi (Samegrelo, Georgia) is another typical example of political payments sent to the Lazian center of power around Archaeopolis. The hoard incudes two ceremonial siliquae of Maurice as well as hexagrams of Heraclius d ated 625 629, probably showing a slow accumulation of monetary gifts. 190 Not all finds should be associated with the war effort against Persia, as several more Byzantine coins from Georgia are clearly dated after 630 when the war was successfully concluded. A solidus Lekhumi) and a hexagram Mtianeti) date from 632 635 and 635 637, respectively an d a solidus from Shilda (Kakheti) dates from 641. 191 Two very large mixed silver hoards of Sasanian drachms and Byzantine hexagrams from Dedoplitskaro (Kakheti, Georgia) and Tbilisi, concealed sometime around 641 show that wealth in Iberia still relied on the dominant presence of Sasanian coins. 192 188 Sakartvelos sakhelmts'ipo muzeumis (1966 1984) Abramishvili, Sakar tvelos sakhelmts'ipo muzeumis Ibid., 126, no. 66. 189 Kropotkin, Klady vizantiiskikh monet 44, no. 425; Abramishvili, Sakartvelos sakhelmts'ipo muzeumis (1966 1984) 30, no. 111 113 and pl. VII/111 1 12; Tsukhishvili and Depeyrot, History and Coin Finds in Georgia 82, no. 9 and plate 7. 190 Tsukhishvili and Depeyrot, History and Coin Finds in Georgia 81, no. 8 and plate 6 and 29, arguing that the coins represent a homogeneous group; Abramishvili, Sakar tvelos sakhelmts'ipo muzeumis (1966 1984) 22, n. 120 and 22 31, no. 76 77, 88, 94 102, 119. 191 Sakartvelos sakhelmts'ipo muzeumis Shilda: Abramishvili, Sakartvelos sakhelmts'ipo muzeumis (1 966 1984) 25, no. 87 and pl. VI/87. Contra Somogyi, "New Remarks on the Flow," 121 who argued that the flow of Byzantine solidi to Lazica was interrupted after 613. Contra Tsukhishvili and Depeyrot, History 28 who argued that no gold coins of Phocas and Heraclius were found in Iberia 192 I. Dzhalagania, Inozemnaia moneta v denezhnom obrashchenii Gruzii V XIII vv. (Tbilisi: Metsnierebe, 1979), 10 33; I. Dzhalagania, Monety klady Gruzii: klad sasanidskikh i vizantiiskikh monet iz Tsiteli Tskaro (pervaia chas M. Tsotselia, History and Coin Finds in Georgia: Sasanian and Byzantine Coins from Tsitelitskaro (AD 641) (Weteren: Moneta, 2002), 86 87; Tsotselia, Coin 140,

PAGE 398

398 Similar developments can be traced to the south, in Armenia, wher e hoards of ( Turkey), Dvin, and Grtchi (Tavush, Armenia). The first hoard from the Persarmenian capital of Dvin ends with issues dated between 638 and 641, but the bulk of the hoard is made of issues from 625 629. The seco nd hoard from Dvin is another example of a mixed Sasanian Byzantine hoard in which the Persian issues are numerically superior. 193 Heraclius was well aware of the fact that Armenian loyalties were easily switched given the political climate of fragmentation which had characterized Armenia since the abolishing of the monarchy in 428. 194 Heraclius had little time to enjoy his otherwise glorious victory over the great archenemy because a new power would soon capitalize on the weakness of the two Near Eastern empir witnessed an intense campaign to attract Armenian princes by granting them honorary titles. 195 A solidus found at Masis (Ararat, Armenia) dated to 641 might indicate that loyalties were bo ught with money, not only with titles, especially since the Arabs got dangerously close after a successful raiding of Dvin in 640. 196 The Arab conquest of Transcaucasia was already under way. Constans II and Constantine IV made great efforts to maintain a f oothold in Georgia and Armenia no. 742. Tsukhishvili and Depeyrot, History and Coin Finds in Georgia 27 wrongly attribute these hoards to trade with Byzantium. For the presence of Sasanian coins in Georgia, see Tsotselia, History and Coin Finds in Georgia: Sasanian Coin Finds 193 Klady vizantiiskikh monet, 42, no. 370. Dvin: Mousheghian et al., History 107 108 and pl. 9; 131 133 an d pl. 9. Grtchi: Ibid., 194 95. 194 Thomson, "Eastern Neighbours," 160. See also M. Whittow, The Making of Byzantium: 600 1025 (Berkeley: University of C alifornia Press, 1996), 202 03. 195 T.W. Greenwood, "A Corpus of Early Medieval Armenian Inscriptions," DOP 58 (2004): 27 91. 196 Mousheghian et al., History 168 69; T. W. Greenwoo d, "Armenian Neighbours: 600 1045," in The Cambridge History of the Byzantine Empire 341; Whittow, The Making of Byzantium 209 10.

PAGE 399

399 through a combination of diplomacy and warfare when internal turmoil in the Caliphate offered good opportunities for action. Coin hoards from the 650s testify to the energetic activity of Constans II in Armenia especially aft er Muawiya offered very generous terms in 652 should Armenian nobles decide to switch sides. 197 Only one hoard was found in Georgia, in a burned layer from a seventh century residence in Magraneti (Mtskheta Mtianeti, Georgia), comprising an almost equal numb er of Sasanian drachms and Byzantine hexagrams. 198 The other four hoards were found in Armenia in the Mtkvari Arax interfluve, north of Dvin, at Echmiazdin, Gyumri, Kosh and Stepanavan and can be fforts to re establish the Byzantine domination in Armenia. 199 Six single finds of hexagrams of Constans II are known from Transcaucasia, most of which were found in Georgia, rather than in Armenia as we would have expected based on the geographic distributi on of hoards. Such coins were found both in Lazica at Zugdidi (Samegrelo Zemo Svaneti) and 200 Bracteates struck on thin gold foil imitating solidi from 654 667 were foun d in a grave from Dzhaga (Karachayevo Cherkesiya, Russia) evidencing continuing relations with tribes from Northern Caucasus. 201 197 Greenwood "Armenian Neighbours," 342. 198 T. Abramishvili, "Klad monet iz Magraneti," in Numizmaticheskii sbornik: posvias hchaetsia pamiati D. G. Kapanadze ed. V. A. Lekvinadze (Tbilisi: Metsnie reba, 1977), 73 82 and pl. VII. 199 Mousheghian et al., History 164 65 and pl. 23 24; 170 and pl. 25; 179 80; 182 83 and pl. 25. 200 Zugdidi: Abramishvili, Sakartvelos sakhelmts'ipo muz eumis (1966 1984) 35 Tsotselia, Coin Sakartvelos sakhelmts'ipo muzeumis 71, no. 229. Rustavi: Ibid., 123, no. 33. Contra Somogy, "New Remarks on the Flow," 121, who argued that no hexagrams from the second half of the seventh century have been found in the region. 201 Rtveladze and Runich, "Nahodki indikatsii," 220 21 and fig. 1/5 6.

PAGE 400

400 The Second F itna following the death of Muawiya was promptly exploited by Constantine IV who invaded Cilicia, while the new domin ant group of the northern steppe, the Khazars, invaded Armenia. Although the new caliph Abd al Malik offered very good terms the Byzantine position in Transcaucasia remained precarious. Despite the efforts of Justinian II to buy the loyalty of Armenian and Albanian nobles in the same manner as his predecessors, by the end of the century the Caliphate extended its influence even fa rther to the north. 202 Contrary to some opinions Byzantium did not lose control of the eastern Black Sea coast until the early eigh th century. 203 In 662 Constans II exiled monothelite heretics in Lazica, ecclesiastical centers were still in place at Phasis and Sebastopolis, while Smbat Bagratuni the leader of an Armenian revolt against the Arabs found refuge at Phasis in 705. 204 A solidus of Justinian II from his second reign (705 711) was found at Shemokmedi (Guria, Georgia) not far from Phasis. 205 Coin finds further substantiate this state of affairs. No hoards are known in Transcaucasia after 670 when the importance of the hexagram had al ready declined sharply, but single finds of solidi of Constantine IV, Justinian II, Leontius, and Tiberius III have been reported in Georgia and in the steppe north of the Caucasus as far as Serpovoe (Tambov, Russia). 206 The finds concentrate in Lazica (Pity 202 Greenwood "Armenian Neighbours," 344 45. 203 Whittow, The Making of Byzantiu m 210. 204 Greenwood, "Armenian Neighbours," 346; B. Martin Hisard, "La domination Byzantine sur le littoral oriental du Pont Euxin (milieu du VII e VIII e sicles)," Byzantinobulgaria 7 (1981): 144 46. 205 Abramishvili, Sakartvelos sakhelmts'ipo muzeumis (1966 1984) 37, no. 140 and pl. IX/140. 206 Abramishvili, Sakartvelos sakhelmts'ipo muzeumis 73 74, no. 240 241 and 244 246; Kropotkin, Klady vizantiiskikh monet 29, no. 125 126.

PAGE 401

401 207 An interesting concentration of late seventh century solidi can be found further to the north east of the Sea of Azov (Rostov na Donu, Russia), on the River Do n where several cemeteries have produced single and collective finds of solidi of emperors from Podgornenskii and Salovo which seem to share a date at the turn of the eigh th century, 208 A connection of these finds with gifts sent by Byzantium to the Khazar allies seems inescapable. 6.2.3 The Land of Gold: The Carpathian Basin Although the history of the Carpathian Basin in the sixth century is shaped by cultural and diplomatic contact with Byzantium, the ties with the Empire were much more volatile than those established in Transcaucasia and in the Lower Danube region. Distance is one factor, as c oncentrations of Byzantine artifacts, especially coins, are sometimes found 500 miles away from the nearest Byzantine province and might have 207 Pityus: Abramishvili, Sakartvelos sakhelmts'ipo muzeumis 127, no. 84. i: Ibid., 73, no. 242. Abramishvili, Sakartvelos sakhelmts'ipo muzeumis (1966 1984) 37, no. 140 and pl. IX/140. 208 Podgornenskii: A. Naumenko and S. I. Bezu glov, "j biznci s irni importleletek a Don vidk Mra Ferenc Mzeum vknyve. Studia archaeologica 2 (1996): 247 48. Salovo: V. E. Flerova, "Podkurgannye pogrebreniia vostochnoevropeiskikh stepei I puti slozheniia kul'tury Khazarii," in Stepi Evropy v epokhu srednevekov'ia. Khazarskoe vremia ed. A. V. Evgelevskii (Donetsk: Institut Arkheologii NAN Ukrainy/Donetskii Nacional'nyi Universitet, 2001), 172 M. Kosianenko, "Pogrebenie u slobody Bol'shaia Orlo vka rannii pamiatnik saltovo maiackoi kul'tury," in Problemy khronologii arkheologicheskikh pamiatnikov stepnoi zony Severnogo Kavkaza ed. V. Ia. Kiiashko (Rostov na Donu: Izdatel'stvo Rostovskogo universiteta, 1983), 115, Romanovskaya: A. I. Semenov, Vostochnoi Evropy," Arkheologicheskii sbornik Gosudarstvennogo Ermitazha 29 (1988), 109 Verbovyi Log: Naumenko and Bezuglov, "j biznci s irni," 250. For a general d iscussion of those finds of late coins, see E. V. Kruglov, "Obrashchenie vizantiiskikh monet VI VIII vv. v vostochnoevropeiskikh stepiakh," in Vizantiia: obshchestvo i cerkov' ed. S. N. Malakhov and N. D. Barabanov (Armavir: Armavirskii gosudarstvennyi pe dagogicheskii universitet, 2005), 168 77.

PAGE 402

402 exchanged hands many times traversing a wide region populated by a mosaic of Germanic tribes (Figure 6 16 and 6 17) Most of the time, centers of power in Bohemia might not have had direct access to Byzantine wealth. Although this appears similar with the case of the Suani or Abasgi on the foothills of the Caucasus who received stipends through the mediation of Lazi kings, Byzantine ambitions were much more limited north of the Alps. The distribution of finds reveals four geographic areas of concentration: the Tisza basin and the Middle Danube, corresponding with the core areas of the Gepid kingdom and later the Avar Khaganate, the Drava Sava interfluve, an important area of contact between the Empire and Barbaricum the upper Danube region, and finally Bohemia at the confluence between the late Germanic and early Slavic worlds. Another factor is chronology, in this case a somewhat arbitrarily defined unit of analysis but a good indicator of major trends and developments. From a historical perspective the flow of Byzantine coins, mainly gold, to the Carpathian Basin has at least to major stages, ca 488 568 and ca 5 68 680, respectively. The first stage can be divided into two phases, from 488 when Theodoric the Great took the Ostrogoths to Italy until 535 when Justinian launched his campaign in Italy and from 535 to 568 when the Avars settled in the Middle Danube reg ion and the Longobards migrated to Italy. The second stage is dominated by the Avar hegemony of the Carpathian Basin and can also be divided in at least two different phases corresponding to the Early and Middle Avar periods, respectively, divided by the t ransition after the defeat beneath the walls of Constantinople in 626. More than anything, political developments in Italy and the western Balkans dictated the nature of contact between Byzantium and the peoples of

PAGE 403

403 Figure 6 16. Early Byzantine coin fi nds in the Carpathian Basin (for Transylvania see above Figure 6 6 ) (numbers on the map refer to numbers in Appendix A.2 ).

PAGE 404

404 Figure 6 17. Early Byzantine coin finds north of the Middle and Upper Danube (numbers on the map refer to numbers in Appendix A.2 ).

PAGE 405

405 Figure 6 18. Early Byzantine gold, silver, and copper coin finds in the Carpathian Basin (491 680).

PAGE 406

406 Central Europe during the sixth and seventh centuries, which often presupposed the typical combination of warfare and diplomatic maneuvers, in cluding gifts, stipends and tribute. The particular nature of the relation between Byzantium and regions often far from the frontier is best reflected in the ratio between gold and copper coins ( Fig ure 6 18) Four to five times more Byzantine gold is to be found in Central Europe than in Transcaucasia or the Lower Danube region, at the expense of the base metal coinage (Appendix B, Table B 2). The absence of a hegemonic power in Central and Eastern Europe between the dissolution of the Hunnic confederation and the rise of the Avar Khaganate made the Carpathian Basin fertile ground for diplomatic initiatives. In the course of the sixth century Heruli, Lombards, and Gepids received political stipends and served as foederati or symmachoi in campaigns led by Byz antine emperors on three different continents. In addition, smaller and less homogeneous groups from Central Europe, which did not always draw the attention of contemporary chroniclers, were ful warfare. The age of Justinian in particular presented many such opportunities, closer to home in northern Italy and Dalmatia or further away in North Africa and in Persia. It comes as no surprise that Procopius complained about the blurring of the trad itional distinction between Roman stratiotai and barbarian foederates 209 After the battle of Nedao (454), the Gepids became masters of the regions east of the Middle Danube, into Transylvania and southwards to the Lower Danube. At the same time the Ostrogo ths were allowed to settle in Pannonia, but their contacts 209 Procopius, Bella I, 11.3 4, ed. Haury and Wirth, vol 1, 361; trans. De wing, vol. 1, 103.

PAGE 407

407 extended further north. A large quantity of solidi reached Pomerania, the Baltic islands, and Scandinavia in the second half of the fifth century only to end abruptly after the reign of Anastasiu s on the Baltic coast of Poland and perhaps a few decades later in Scandinavia. 210 Aside from a number of single finds of solidi hoards of solidi ending with issues of Anastasius have been found in Poland at Karsibr Malechowo, and 211 A hoard fo und at Karlino included solidi of Anastasius and also Scandinavian bracteates and jewelry. 212 In addition, a large hoard 150 gold coins from Mrzezino, included 130 issues of Anastasius, original solidi and Ostrogothic imitations hinting at the fact that most of the gold was channeled to the Baltic through the mediation of the Ostrogoths, the main recipients of Byzantine subsidies. 213 D. M. in furs on the Baltic is less convincing, although it may be part of the larger picture. 214 On the other hand, Renata in Pomerania reflect contacts within the German world rather than direct connections to Byzantium. 215 210 R. Der Zufluss von Solidi in die sdlichen Ostseegebiete," in Byzantine Coins in Central Europe 217 30; H. Hor snaes, "Late Roman and Byzantine Coins Found in Denmark," in Byzantine Coins in Central Europe 231 270; I. Hammarberg, B. Malmer and T. Zachrisson, Byzantine Coins Found in Sweden (London: Spink & Son, 1989). 211 Karsibr : kich z terenw Pomorza przechowywanie w zbiorach Muzeum Kulturalno Historycznego w Stralsundzie," 43 (1999): 176. Malechowo: VI wieku," in i historycznych 212 213 Numizmatyczne 42 (1998): 59 60. 214 D. M. Metcalf, "Viking Age Numismatics, 1. Late Roman and Byzantne Gold in the Northern Lands," NC 155 (1995): 413 441 215 Der Zufluss von Solidi," 224 26. See also Curta, The Making of the Slavs 195 96.

PAGE 408

408 In fact there is an uninterrupted chain of gold solidi or tremisses of Anastasius and Justin I running from the Baltic to the Adriatic coast which can be connected with federate groups. Unsurprisingly, south of Poland the largest concentration can be found in the Czech Republic and can be associate d at this time with Germanic settlements connected with the Longobard presence in Bohemia and perhaps also with the Thuringians on the Elbe. Few coins were found in archaeological context but at least the solidus however, later than the reign of Anastasius as testified by a tremissis of Justinian with a suspension loop found in another grave. 216 A tremissis with suspension loop of Anastas ius found at Chotusice in the nineteenth century may have also belonged to a destroyed grave. 217 Four other single finds of gold from Anastasius and Justin I found in the Czech Republic and Slovakia indicate that the region became strategically important for Byzantium especially after the Longobards defeated the Heruli in 508 and positioned themselves as a potential adversary of the Ostrogoths in Italy. 218 The gold trail follows east and south into the territory ruled by the Gepids and the Ostrogoths, respec tively. An early solidus of Anastasius from Balotaszlls (492 507), another solidus from the same period found in a grave from the Gepid cemetery in 216 J. Militk, "Finds of the Early Byzantine Coins of the Early Byzantine Coins of the Early Byzantine Coins of the 6 th and 7 th Century in the Territory of the Czech Republic," in Byzantine Coins in Central Europe 37 8 79, no. M28/1 2. 217 Ibid., 367, no. C2. 218 The best surveys on the Longobards remain N. Christie, The Lombards: The Ancient Lon gobards (Cambridge, MA: Blackwell, 1995) and K. Christou, Byzanz und die Langobarden: von der Ansiedlung in Pannonien bis zur e ndgltigen Anerkennung (500 680) (Athens: Historical Publications St. D. Basilopoulos, 1991). See also recently, G. Ausenda, P. Delogu and C. Wickham, ed., The Langobards Before the Frankish Conquest: An Ethnographic Perspectiv e (Rochester, NY: Boydell Pre ss, 2009). For sixth century Longobards, see W. Pohl, "The Empire and the Lombards: Treaties and Negotiations in the Sixth Century," in Kingdoms of the Empire: The Integraton of Barbarians in Late Antiquity ed. W. Pohl (Leiden/New York: Brill, 1997), 75 1 34.

PAGE 409

409 Kiszombor and a ceremonial silver miliarensis from a Gepid grave in Sz (all in Hungary) indicate polit ical payments sent to the Gepid center in the Tisza Mure interfluve or perhaps booty taken from Illyricum by the Gepids led by Mundo. 219 A copper half follis Romania) and another one found i n unknown circumstances at Gherla (Cluj, Romania) might signal connections with fortresses from the Danube where such small change was handled. 220 Finds extend to the west in Bosnia, Croatia and especially in Slovenia, closer to Italy, where four single find s of solidi and tremisses of Anastasius are known, possibly related to the Ostrogoths and the migration of various groups of Heruli after their defeat in 508. 221 Bronze coins of Anastasius and Justin I are not absent from Central Europe, with finds recorded at (Czech Republic), (Slovakia), Staasdorf and Reisberg (Austria) and further west on the upper Danube, but they tend to concentrate in the western Balkans, south of the Drava or in the sector of the Roman limes between Belgrade and the Iron Gates, where such coins were more readily available from Roman fortresses. 222 219 Byzantine emperors started to send annual subsidies to the Gepids after the dissolution of the Hunnic confederation, for which see Jordanes, Getica 264, ed. T. Mommsen, MGH: AA 5.1 (Berlin: Weidmann, 1882), 126 ( annua sollemnia ). Balotaszlls: A. K iss, "Rgszeti s numizmatikai adatok a Duna Tisza Mra Ferenc Mzeum Evknyve. Studia Archaeologica 4 (1998), 191. Kiszombor: D. Csallny, Archologische Denkmler der Gepi den im Mitteldonaubecken (Budapest: Verlag der ungarischen Akademie der Wissenschaften, 1961), 174; pl. CXV/11. Sz J. Cseh, et al., Gepidische Grberfelder im Theissgebiet II (Monumenta Germanorum Archaeologica Hungariae, 2)(Budapest: Magyar Nemzeti M zeum, 2005), 123. career, see B. Croke Mundo the Gepid : From Freebooter to the Roman General," Chiron 12 (1982): 125 35. 220 Csallny, Archologische Denkmler 145, 292 and pl. CCLXXII/2; Velter, Transilvania n secolele V XII 291, no. 53. 221 Kos, "The Monetary Circulation in the Southeastern Alpine Region ca 300 B.C. A.D. 1000," Situla 24 (1984 1985): 226, no. 1 4 (Ljubljana, Mokronog, Studeno and Trbinc). For the Heruli, see Sarantis, "The Justinianic Herules," 361 402. 222 Militk, "Finds of the Early Byzantine Coins of the Early Byzantine Coins, 367,

PAGE 410

410 The age of Justinian brought the opening of an important war front in Italy and renewed political alliances meant to isolate the Ostrogoths and secure the frontier of the Balkans. Already in 526 the Longobards took over Pannonia and rapidly strengthened their position, perhaps with the tacit approval of Byzantium and by taking advantage of s death and of subsequent instability in the Gothic kingdom. 223 Securing the western Balkans as a bridge between Thracia, Illyricum, and Italy was of paramount importance for Justinian as war broke out in Italy in 535. Diplomatic initiatives targeted all the important groups from the Carpathian Basin, the Gepids, the Heruli a nd the Longobards and it is possible that they all received stipends from Byzantium, especially since Justinian repeatedly exercised his ability to play off one barbarian group against ing off the Gepids and instead to bestow on the Longobards lands in Pannonia and Noricum as well as a very great amount of money ( chrema ). 224 His alliance with the Heruli settled in the area of Singidunum had already been sealed two decades before this eve nt when the Herul king was baptized in Constantinople and left home with many gifts, not before Justinian reiterated his intention to call on the Heruli when their military assistance would be required. 225 no. C4 and 374 A. Fiala, "Byzantsk mince na Slovensku (6. Slovensk Numizmatika 10 (1989): 57, no. 3. Staasdorf a nd Reisberg: W. Hahn, Die Fundmnzen des 5. 9. Jahrhunderts in sterreich und den unmittelbar angrenzenden Gebieten ," in Die Geburt Mitteleuropas. Geschichte sterreichs vor seiner Entstehung 378 907 ed. H. Wolfram (Vienna: Kremayr & Scheriau, 1987), 455 and 458. For the upper Danube, see J. Drauschke, "Byzantinische Mnzen des ausgehenden 5. bis beginnenden 8. Jahrhunderts in den stlichen Regionen des Merowingerreiches," in Byzantine Coins in Central Europe 288, fig. 12. 223 Christie, The Lombards 34 35 224 Procopius, Bella VII, 33.10, ed. Haury and Wirth, vol. 2, 4 43; trans. Dewing, vol. 4, 441. 225 Sarantis, "The Justinia nic Herules," 373 74.

PAGE 411

411 From an archaeological perspective the Heruli are s till hard to trace. T he same can not be said about the Gepids who left a deep cultural imprint in the archaeological the region as Gepidia ." 226 th the Gepids fluctuated depending on the balance of power in Barbaricum 540s. Numismatic evidence points to subsidies sent by Justinian to the Gepid centers of power in Tr ansylvania and Banat. Two light weight solidi customarily used in political payments have been found at Gyula (Bekes, Hungary) and Banatski Karlovac (South Banat, Serbia), both dated 538 542. 227 Three rare cer emonial silver pieces have been found at Gheorghe Doja and location in Arad County. 228 229 As we have seen in the case of Transcaucasia such s ilver coins indicate diplomatic gifts, less 226 Jordanes, Getica 74, ed. T. Mommsen, MGH: AA 5.1 (Berlin: Weidmann, 1882), 75; trans. C. C. Mierow (Cambridge: Speculum Historiale, 1960), 71. For "Gepidia" see also Geographus Ravennas, Cosmographia IV, 14, ed. I. Schnetz, Itineraria Romana II (Leipzig: Teubner, 1940), 53. On the Gepids, see I. Bna, "Gepiden in Siebenburgen, Gepiden an der Theiss," AAASH 23, no 1 2 (1979): 9 50; W. Pohl, "Die Gepiden und die Gentes an der mittleren Donau nach dem Zerfall des Attilareiche," in Die Vlker an der mittleren und unteren Donau im fnften und sechsten Jahrhundert ed. H. Wolfram and F. Daim (Vienna: Verlag der sterr eichischen Akademie der Wissenschaften, 1980), 239 305; A. Khraralambieva, Gepids in the Balkans: A Survey of the Archaeological Evidence," in Neglected Barbarians 245 62. 227 ntre ed. Renaissance, 2010), 147, fig. 2/5a b; Oberlnder Trnovea 70, n. 72. 228 XXVI.1. A. Lakatos, "Monede bizantine bizantine bizantine din perioada ava vestul Romniei." EN 12 (2002): 248, no. 13. Arad County: Oberlnder la antichitate," 69 70, n. 72. 229 K. Horedt, Grabungen in e. Vor u. Frhgeschichtl. Siedlung in Siebenbrgen (Bonn: Habelt, 1979 1984).

PAGE 412

412 as they had an two other from unspecified locations in Mure solidus which was perhaps part of a large hoard found at Apalina strengthen the hypothesis of stipends received by the Gepids controlling this area. 230 In addition, a large hoard (ca. 100 solidi and jewelry) attributed to the Gepids was found have been preserved with the most recent struck for Justin I. 231 To the west, in Banat a solidus of Justinian was found in a Gepid grave from Bo (North Banat, Serbia). 232 Copper coins found occasionally in T ransylvania were probably brought by those Gepids who served in the garrisons defending the Danube frontier or on other theaters as idicated by a 12 nummia 233 The gold coins found in unknown circumstances at Beclean, Byzantium or, more probably, redistributions in the Gepid kingdom, some perhaps resulting from a foedus signed in 552 in the eve o f a new offensive in Italy. 234 230 Lakatos, "Monede bizantine bizantine," 248, no. 15 and no .20 and 250, no. 26. Apalina: Szatmri (1747 1812)," SCN 3 (1960): 440. 231 P. Somogyi, "Der Fund von Kleinschelken (Siebenbrgen, 1856) im Lichte neuentdeckter Archivdaten," in Byzantine Coins in Central Europe 417 48. 232 Epoque prhistorique et protohistorique en Yougoslavie Recherches et rsultats ed. G. Novak et al.(Belgrade: Socit archologi que de Yougoslavie, 19 71), 191. 233 Lakatos, "Monede biza ntine bizantine," 252, no. 44B. 234 250, no. 27. Cetea: Velter, Transilvania 289, no. 23. Oarda de Jos: Alba," Apulum 38 (2001): 256 and fig. 2. Pecica: Oberlnder Mare: IX/10 and pl. XI/2.

PAGE 413

413 Gepids, Heruli and Longobards served in the Roman armies of Belisarius and Narses in Italy and some of the wealth accumulated in the service of the Empire is reflected in coin finds from their homelands. One method of tracing their movement is by looking at the mints issuing the coins found in the Carpathian Basin. Finds of coins from the Italian mints of Rome and Ravenna are recorded as far north as Poland. A quarter siliqua from Ravenna was found in a n unspecified location in Slovenia and a tremissis comes from a grave in Rifnik 235 More silver Byzantine coins from Italy have been reported in Bavaria where the interests of Byzantium, the Franks and the Ostrogoths colided during the age of Justinian. 236 Bronze dekanummia from Rome were found at Sisak (Sisak Moslavina, Croatia), Lovosice (Litom 237 In addition, two 12 nummia pieces from Alexandria have been found at Osijek (Osijek Baranja, Croatia) and Stillfried (Niedersterreich, Austria). 238 Regardless of their ethnic background men from acros s Central Europe were attracted by warfare in Italy. The fact that many of these coins are small denominations, as are other finds from eastern mints such as the pentanummia found at Vienna and Osijek and the already mentioned dekanummium 235 P. Kos, Die Fundmnzen der rmischen Zeit in Slowenien vol. 3 (Berlin: Mann, 1988), 583, no. 2; L. Bolta, (Ljubljana: Narodni muzei, 1 981), 36 and pl. 17/10. 236 Drauschke, "Byzantinische Fundmnzen," 289 fig 14 and 294 fig. 18. 237 "Byzantine Coins in the Zagreb Archaeological Museum Numismatic Collection. Anastasius I (A.D. 497 518) Anastasius II (A.D. 713 715)," Zagrebu 30 31: (1997 1998): 166, no. 329. Lovosice: Militk, "Finds of the Early Byzantine," 371, no. 7. Jh. In Polen," in Byzantine Coins in Central Europe 504, no. 14. 238 H. Gricke i Baranje," in Radovi XIII. Meunarodnog kongresa 1.10.1994) vol. 2, ed. N. Cambi and E. Marin "Die byzantinischen und karolingischen Fundmnzen aus dem sterreichishen Bereich der Avaria eine Neubearbeitung," in Byzantine Coins in Central Europe 341, no. 7 and 351, fig. 6/20.

PAGE 414

414 urban markets. 239 This is not to say that the large folles of Justinian, so popular in barbaricum north of the Lower Danube are absent in the Carpathian Basin. On the contrary, such coins ar e the most frequent Justinianic issues from Slovenia all the way north to Poland. 240 Soldiers fighting in Italy and perhaps political payments sent from Constantinople or redistributed in bilateral relations between different tribes in barbaricum are also re flected in single finds and hoards of gold coins. Tremisses which were the more popular gold denomination of Italy and the western kingdoms, unlike the case of Transcaucasia and the Lower Danube where they are rare finds, are well represented and their fre quency increases as one moves westwards from the Middle Danube to Bavaria. 241 Two hoards from Slovenia and Croatia, found at Vrh Pri Pahi and Sisak, respectively, included Byzantine issues as well as Longobard imitations after coins of 239 Winter, "Die byzantinischen und karolingischen," 343, no. 12b/1; G ricke tinijanov novac," 1151, no. 20. 240 Rifnik (Slovenia): Kos, "The Monetary Circulation," 226, no. 15. Vinkovci (Croatia): Gricke 3. Tn Nad Vltavou and Maratice (Czech Republic): Militk, "Finds of the Early Byzantine Coins," 371, no. C12 and 380, no. M30. 12 and 506 07, no. 20. Mitterndorf, Klosterneuburg and Carnuntum: Winter, "Die byzantinischen und karolingischen ," 337, no. 1/9 and 340, n o. 2, 4 with 350, fig. 6/17 18. 241 Osijek (Croatia): Gricke "Byzantine Coins," 149, no. 79. Ri fnik (Slovenia): Bolta, Rifnik 36 and pl. 17/1. Kranj (Slovenia): V. Stare, (Ljubljana: Narodni muzej, 1980), 107. Tolisa (Bosnia & ary): Winter "Die Byzantinischen und karolingischen," 346, no. 18 and 352, fig. 6/25. Pulst and St Paul im Lavanttal (Austria): Hahn, "Die Fundmnzen des 5. 9. Jahrhunderts," 455. Eisenstadt (Austria): Winter, "Die byzantinischen und karolingischen ," 344, no. 13 and 351, fig. 6/21. Luzice (Czech Republic): Militk, "Finds of the Early Byzantine Coins," 378 379, no. M28/2. For finds in Bavaria, see Drauschke, "Byzanti nische Fundmnzen," 285 fig. 8.

PAGE 415

415 Justinian, perhaps bet raying the ethnic background of the owner. 242 Other gold hoards concealed during the reign of Justinian have been found at Ktschach Laas (Carinthia, 243 The two miliarensia found at Rabka Zdrj (Nowy Targ, Poland) have a ceremonial nature and they date fr om the initial stage of the war in Italy, although they may be also connected with developments in barbaricum occasioned by the migration of the Longobards southwards into Pannonia. Unfortunately not enough is known about the circumstances of the find but both coins were struck with the same pair of dies and probably represent a fraction of a larger payment, a homogeneous group of coins sent directly from the mint of Constantinople to their original recipient in barbaricum 244 The density of finds in the Cze ch Republic, Bohemia in particular, is perhaps the most intriguing development of this period. A part from six single finds of gold and copper coins, the region yielded no less than six hoards concealed during the reign of Justinian. 245 Unfortunately, most of the hoards were found accidentally and were subsequently dispersed but the available specimens still offer a coherent p icture of the nad Szavou all other hoards include ancient Roman coins aside from the early 242 Ostrogothic Coinage from Collections in Croatia, Slovenia and Bosnia & Herzegovina (Ljubljana: Narodni Muzej, 1994), 227 and 229 31. 243 Hahn, "Die Fundmnzen des 5. 9. Jahrhunderts," 454; Militk, "Finds of the Early Byzantine Coins of the Early B yzantine Coins," 379 80, no. M. 29. 244 05, no. 16 and 508, fig. 6 6 7. Another silver coin of certain; see Ibid., 500, no. 4. 245 A n addi latest coin was struck for Justin I, but the hoard may in fact belog to the same horizon of hoards buried during the age of Justinian. See Militk, "Finds of the Early Byzantine Coins," 368 69, no. C7.

PAGE 416

416 Byzantine issues. 246 The hoard from Hradec Krlov includes a dekanummium fr om 540/1 and so does the hoard of 26 bronze coins from Prague, ending with a dekanummium from Carthage dated to the same year. 247 Interestingly, the earliest coin in the hoard is a first century issue of Augustus. Only two coins have been retrieved from the dekanummium minted in Ravenna in the 540s. 248 Finally, a hoard of 17 coppers from Turnov is made of small Late Roman AE4 pieces including a Vandal imitation, a small module half follis of Anastasius and two dekanummia of Justinian. 249 A few conclusions emerge from this enumeration. The owners of these small value accumulations obviously had access to small change and perhaps deliberately selected small change for hoarding. Some have associated their deposition with ritual pr actice or with monetary transactions restricted to the elite, but n either seems plausible. 250 Most of these hoards betray a western origin due to the presence of coins from Rome, Ravenna and Carthage. Contemporary hoards from North Africa and Italy are often large accumulations of small change reflecting the nature of the local monetary economy and it is conceivable that the similar but comparatively smaller hoards from the Czech Republic constitute groups of coins armies who fought in the west. 251 An ethnic 246 Ibid., 380, no. M31. 247 Ibid., 370 7 1, no. C11 and 369 70, no. C10. 248 Ibid., 369, no. C9. 249 Ibid., 362 and 371 73, no. C15. 250 Ibid ., 365. 251 For the hoarding of early Byzantine small change together with ancient Greek and Roman coins in the Mediterranean world, see recently F. Curta and "Hoards and Hoarding Patterns in the Early Byzantine Balkans," DOP 65 66 (2012): 59 62.

PAGE 417

417 attribution of these finds is notoriously difficult and Czech scholars prefer the neutral label late Germanic phase, although a more direct connection with the Longobards or groups associated with them is quite probable. 252 The rise of the Avar khaganate changed the bala nce of power in the Carpathian B asin. As early as 558 the Avars, still in the east, sent an embassy to Constantinople seeking an alliance and returned home with cords worked with gold, couches, si lken garment and a great many other objects. 253 Such precious gifts turned out to be only the preamble of the literally tons of gold absorbed by the Khaganate after the Avars migrated to the Middle Danube just a decade later. A yearly tribute of 80,000 soli di was agreed upon in 574 although Justinian might have sent additional subsidies between 558 and 565. An Avar offensive in the Balkans brought an increase of the tribute to 100,000 solidi in 585 and again to 120,000 in 598. The treaty signed by Phocas in 604 may have incurred an additional increase of the tribute and so may have the negotiations conducted by Heraclius, who finally ended up agreeing to a staggering 200,000 solidi in 623 when he desperately needed peace in the Balkans to stage his counteroff ensive in Transcaucasia. Historians generally agree that some six million solidi (ca. 25 tons of gold) were sent to Avaria between 568 and 626. 254 Part of this gold was kept in the royal treasury of the khagan and some of the wealth accumulated in this perio d might have beed seized by Charlemagne in 796. Einhard mentions omnis which might be taken 252 Christie, The Lombards 65. 253 Menander, Historia fr. 5.2, ed. Blockley, 48 51. 254 J. Iluk, "The Export of Gold from the Roman Empire to Barbarian Countri es from the 4th to the 6th Centuries," Mnstersche Beitrge zur antichen Handelsgeschichte 4, n. 1 (1985): 93; W. Pohl, Die Awaren: Ein Steppenvolk in Mitteleuropa, 567 822 n. Chr. (Munich: C. H. Beck, 1988), 502 for yearly tribute and 398, n. 32 for tota l estimations, with the previous literature.

PAGE 418

418 as a reference to the Byzantine solidi accumulated in the Early Avar period. 255 Some of the tribute, however, was distributed by the khagan to his warrior elite. Many coins were probabl y melted down to produce the spectacular jewelry often found in Avar age cemeteries, while others ended up as " in graves, or were remodeled into finger ring s or pendants. 256 Within a decade after their arrival in Pannonia, the Avars became the hegemonic power of the C arpathian B asin. Already in 566 the Longobard king Alboin renewed hostilities with the Gepids and achieved a decisive victory. The diplomatic init iatives related to this conflict would have long term consequences for the Danube region. the Empire but his failure to deliver after an initial joint success against the Longobards resulted in a neutral stance of the emperor in the final stage of the conflict. The defeated Longobards were forced to seek the alliance of the Avars, themselves eager to gain a foothold in the western Balkans. 257 The might of the Avars and possi bly other political calculations led Alboin to Italy and it is perhaps now that the Longobards gain ed direct control over the important fortresses in Slovenia. Although Justinian granted them the entire region north of Dalmatia, the Byzantine copper coin f inds from settlements such 255 Einhard, Vita Karoli Magni 13, ed. O. Holder Egger, MGH: SS (Hannover: impensis bibliopolii Hahniani, 1911), 16. For a discussion of the contemporary sources regarding the Avar treasure, see J. B. Ross, "Two Neglected Paladins of Charlemagne Erich of Friuli and Gerold of Bavaria," Speculum 20 (1945): 221 22. 256 A. Kiss, "Die Goldfunde des Karpatenbeckens vom 5 10. Jahrhundert (Angaben zu den Vergleichsmglichkeiten der schriftlichen und archologischen Quellen) ," AAASH 38 (1986): 105 45. On coins as obolus Archaeologiai 105 (1978): 206 16; I. Bon, "Studien zum frhawarischen Reitergrab von Szegvr," AAASH 32 (1980): 83 86. 257 Christou, The Lombards 98 106; Pohl, "The Empire and the Lombards," 96 98. On diplomacy, see F. Wozniak, "Byzantine Diplomacy and the Lombardic Gepid Wars," Balkan Studies 20 (1979): 139 58.

PAGE 419

419 as Rifnik, Ptuj, Sempeter and Celje are more numerous in the period before 568 and are dominated by small change, dekanummia and half folles some from mints such as Rome and Salona, indicating a connection with the war effort. The hoard of solidi from Arnoldstein in Carinthia, which included imitations after solidi of Justinian and a light weight solidus of 22 siliquae from Ravenna, was probably concealed during the events leading to the Longobard conquest of northern Italy. 258 Th Gepid cultural identity survived in a modified form under the rule of the Avars in Transylvania and under the Longobards in the west. 259 Few gold coins of Justin II have been found in Szentendre (Pest, Hungary) and Klked Feketekapu (Baranya, Hungary) and provide only a terminus post quem for the dating of those complexes to the Early Avar period. 260 Two times many more copper coins have been found in the region under the direct control of the Khaganate and Justin II is in fact the only Avar age emperor whose copper coins are more abundant than the gold coins. 261 This is the early stage of the Avar presence in the Carpathian basi tribute, a policy maintained for a decade (565 575). On the other hand, the recent 258 Hahn, "Die Fundmnzen des 5. 9. Jahrhunderts," 454. 259 On Gepid culture under the Avars, see R. Harhoiu, "Quellenlag e und Forschungsstand der Frhgeschichte Siebenbrgens im 6.7. Jahrhundert," Dacia 43 45 (2001), 130 31 with the important cemetery from Bratei, for which see R. Harhoiu, "Where Did All the Gepids Go? A Sixth to Seventh Century Cemetery in Bratei (Romania )," in Neglected Barbarians 209 44. See also Bna, "Gepiden," 45. On German culture under the Avars, see A. Kiss, "Germanen im awarenzeitlichen Karpatenbecken," in Awarenforschungen ed. F. Daim, vol. 1 (Vienna: Institut fr Ur und Frhgeschichte der Uni versitt Wien, 1992), 35 134. 260 Somogyi, Byzantinische Fundmnzen 56, no. 39, 79, no.66, and 87 88, no. 77. The cemetery from Szentendre also produced a solidus of Phocas, for which see Ibid., 88 89, no. 78. 261 Somogyi, "Byzantinische Fundmnzen," 234, tab le 1 and 237 38, table 3.

PAGE 420

420 recovery of Sirmium and the Gepid presence in Pannonia helped the distribution of copper coins to the north. Some of thes e coins are clearly connected with the military, such as the follis issued by a mobile military mint found at Carnuntum but also the half folles of Thessalonica from Carnuntum and Tiszakeszi (Borsod Abaj Zempln, Hungary) used to pay the soldiers defendin g the frontier in Illyricum. 262 In addition, the prevalence of coins from Nicomedia rather than Contantinople is also revealing. The same dominance of issues from Nicomedia has been observed in other militarized frontier regions such as Transcaucasia and Ara bia, but also in towns far from the major war theaters. 263 The arrival of the Avars did not completely disrupt connections between Central Europe and the Mediterranean world. A dekanummium from Carthage was found in wear, while folles from Nicomedia have been found at Sulejw (Poland) and in an unspecified location in Moravia. 264 In the Czech Republic a small hoard of silver and copper coins from Trajan Decius to Justin II (573/4) dition of accumulations containing older Roman coins. 265 Intrestingly, no other hoards are recorded in Poland, Slovakia, and the Czech Republic until the seventh century. Although the indirect connection with Byzantium was never completely severed, a drastic reduction in coin finds can be noticed in the last quarter of the sixth century. Some scholars attributed this development to the arrival of 262 Winter, "Die byzantinischen und karolingischen ," 338, no. 1/23; Somogyi, "Byzantinisch e Fundmnzen," 285 286, no. 16. 263 15. 264 02 no. 9 and 505, no. 17; Militk, "Finds of the Early Byzantine Coins of the Early Byzantine Coins," 381, no. M33. 265 Militk, "Finds of the Early Byzantine Coins," 374, no. C17.

PAGE 421

421 Slavic groups, but the hegemony exercised by the Avars is an equally plausible explanation, as the khagan had ever y reason to exploit the resources and the demographic potential of the region to his own benefit. 266 On the western fringes of the Avar khaganate, however, the connection with the Mediterranean was not interrupted as testified by the finds of three dodekanum mia of Maurice and Phocas minted in Alexandria at Maria Saal and Vienna and two dekanummia from Ravenna and Carthage, respectively, at Carnuntum (all in Austria). 267 Coin finds from the core regions of the Avar Khaganate confirm the flow of gold described by written sources. 268 Graves from Klked Feketekapu, Tiszagyenda Gi t, Pcs Alsmakr Tiberius II, Maurice, and Phocas. 269 Sometimes the coins are uncirculated (Szentendre) and many were used a s oboli judging by their position in the grave, regardless of the social status of the deceased. The specimen from Tc Gorsium is a hybrid imitation of solidi of Tiberius II and Maurice. 270 An imitation of Maurice was also found in 266 M. Woloszyn, Byzantine Coins from the 6 th and 7 th c. From Poland and their East Central European Context," In Roman Coins Outside the Empire: Ways and Phases, Contexts and Functions eds A. Bursche, R. Ciolek, and R. Wolters (Weteren: Moneta, 2008), germanischen Kulturen an der mittleren Donau und das Problem de Vordringes der Slawen," in Die Vlker an der mittleren und unteren Donau im fnften und sechsten Jahrhundert ed. H. Wolfram and F. Daim (Vienna: Verlag d er sterreichischen Akademie der Wissenschaften, 1980), 229 30; On the heavy tribute paid by the Slavs, see Fredegar, Chronicae IV.48, ed. B. Krusch, MGH: SRM 2:1 (Hannover: Impensis Bibliopolii Hahniani, 1888), 144; trans. J. M. Wallace Hadrill (New Yor k: Thomas Nelson and Sons Ltd, 1960), 40. 267 Winter, "Die byzantinischen und karolingischen ," 342 43, no. 11b/1 and 12a; Hahn, "Die Fundmnzen des 5. 9. Jahrhunderts," 454. 268 Hungary, north eastern Austria, south western Slovakia, wes tern Romania and Vojvodina in Serbia are considered to be part of the Avar khaganate proper. See Pohl, Die Awaren map 2. 269 Somogyi, Byzantinische Fundmnzen 56, no. 39, 57, no. 41/2, 87 88, no. 77 and 88 89, no. 78; Somogyi, "Byzantinische Fundmnzen," 2 79 280, no. 10 and 284 85, no. 15. 270 Somogyi, Byzantinische Fundmnzen 89 90, no. 80.

PAGE 422

422 Transylvania at Rupea ( Br the east. 271 Copper coins were usually deposited as oboli but they were also quite often part of waist purses as indicated by the burial assemblages from Jutas, Klked palota in Hungary. 272 They were found together with other small objects, but no other coins, so the existence of coin purses is out of the question. In other cases the coins were perforated to be worn on a necklace or perhaps attached to a piece of garment a s indicated by the finds from Nyregyhza Kertgazdasg (a light weight solidus perforated twice) and Tc Gorsium, but the fashion of using Byzantine coins as jewelry would become more popular on coins of the Heraclian house. 273 The proportion of surviving go ld coins, most of them from burial assemblages and stray finds, does not correspond to the evolution of the tribute recorded by Byzantine writers. Based on the available information and allowing for an almost uninterrupted flow of gold according to the tre aties, no less than 3.8 million solidi had already reached Avaria by 610. Even if we accept that payments were temporarily discontinued during years of intense warfare in the Balkans in the 580s but especially the late 590s when the Byzantine army was succ essful, still it is reasonable to suggest that 50 percent of the total amount of gold dispatched to the Avars had already been paid by the accession of Heraclius. How, then, are we going to account for the fact that the numismatic evidence from the core re gions of the khaganate present us with a disproportionate ratio weighing heavily in favor of coins from the reign of Heraclius? A 271 Velter, Transilvania n secolele V XII 298, no. 144. 272 Somogyi, Byzantinische Fundmnzen 38 39, no. 21, 48 49, no. 33 57, no. 41/1, and 93, no. 84. 273 Ibid., 67 68, no. 52

PAGE 423

423 recent updated inventory of gold coin finds from Avaria confirms the fact that only 36 finds can be dated from the period betw een 565 and 610, while no less than 49 coins date from the reign of Heraclius. In fact no less than 29 coins date from 616 625 which corresponds to the highest level reached by the tribute. 274 Starting from the second decade of the seventh century the Avars and Slavs ravaged the provinces of the Balkans and dealt a final blow to the already crumbling defensive system on the Lower Danube. 275 There is no clear indication that internal developments led to the decision to keep the wealth received from Byzantium in monetary form. To be sure, the khagan had to maintain a sizable treasury, especially after the flow of gold from Constantinople was interrupted in 626, as the loyalty of the peoples under his suzerainty depended on gifts and displays of authority. The onl y important development is an increasing tendency to use Byzantine coins as jewelry (rings, pendants), rather than melting them down, a practice documented from Transylvania to the eastern Merovingian world. 276 On the other hand, tribute was not the only mea ns of gaining access to Byzantine gold. Plunder and ransom for prisoners paid in coin were important alternatives, albeit harder to quantify by the modern historian. 277 However, plunder in the Balkans cannot fully account for the overwhelmingly superior nu mber of coins minted for Heraclius found in the Carpathian basin, especially 274 Somogy, "Byzantinisch e Fundmnzen," 234 37, table 1. 275 I. Bon, "Die Awarenfeldzge und der Untergang der byzantinischen Provinzen an der Unteren Donau," in Kontakte zwischen Iran, Byzanz und der Steppe im 6. 7. Jahrhundert ed. C. Blint (Budapest: Archologisches I nstitut der UAW, 2000), 163 83. 276 Somogyi, Byzantinische Fundmnzen 142; Militk, "Finds of the Early Byza ntine," 364; Drauschke, "Byzan tinische Fundmnzen," 283, 292. 277 A nephew of Heraclius was ransomed for an unspecified sum of money (chremata), which probably brought additional gold to Avaria in the decades following the failed siege of Constantinople. For the episode, see Nikephoros, Breviarum 21, ed. Mango, 70 71.

PAGE 424

424 since many finds are light weight solidi of 20 siliquae dated between 616 625, probably issued for external payments. 278 Petr Somogyi listed four other issues of light weight solid i of 20 siliquae of Phocas and three solidi of Maurice worth between 20 and 23 siliquae 279 which indicates that the practice was probably intensified after the peace of 584 when Maurice agreed to increase the tribute to 100,000 solidi 280 Such finds best refl ect the redistribution system in the Avar khaganate, but also in its wider area of so me 300 solidi 281 Some of the light weight solidi from the core areas of the Khaganate were used as jewelry such as the specimens from the burial assemblages uncovered at S npetru German, Tiszavasvri Kashalom Kiszombor Tanyahalom and Hajddorog Vroskert. 282 In regions west of the Avar center of power western mints continued to play an important role, Italy being the most probable mediator of contacts with the central and eastern Mediterranean world. North African copper coins of Phocas and Heraclius minted at Carthage were found at Sisak, Carnuntum, and Ostrwek (Otwock, Poland), while dodekanummia of Alexandria have been recorded at Carnuntum and Maria Saal 278 Somogyi, "Byz antinische Fundmnzen," 235 36. 279 Ibid., 234. 280 For the historical context, see Whitby, The Emperor Maurice 142. 281 Coins," 360 and 376, no. C23 24; 361 and 381 Byzantinische Fundmnzen 137 38. 282 Somogyi, Byzantinische Fundmnzen 43 44, no. 27 and 77, no. 65; S omogyi, "Byzantinische Fundmnzen," 274 275, no. 7, 283 28 4, no. 14, and 286 287, no. 17.

PAGE 425

425 in Austria. 283 A countermarked follis from Sicily a nd a dekanummium from Catania have been found Ptuj and Carnuntum, respectively. 284 Finally, a rare half follis minted at Alexandretta in Syria during the revolt of the Heraclii (608 610) was found in Slovakia at nd have not yet been published. 285 A hoard was found at Kluk (Nymburk, Czech Republic) in a flooded sand quarry but only one specimen has been preserved, a half follis minted at Carthage in 606 607. 286 The practice of burying hoards of small change brought fro m a Mediterranean milieu would continue in Bohemia until the second half of the seventh century. At least three major differences characterize coin finds from the Khaganate proper. First, no copper coin finds from western mints are so far known from the c ore areas of Avaria, although those are present in the Lower Danube barbaricum and further east in Transcaucasia. Only three gold coins issued by western mints are recorded in Avaria, but only the solidus of Constans II minted in Rome has a precise finding place as well as an archaeological context, having been found in a grave from the Avar age cemetery of Hajdnns Frjhalom jrs (Hajd Bihar, Hungary). 287 This is somewhat unexpected since the Avars raided as far as Cividale, sacked in 611, 283 Carnuntum: Winter, "Die byzantinischen und karolingischen ," 339, no. 25 26 and 28 29. Maria Saal: Hahn, Die Fundmnzen des 5.9. Jahrhunderts," 454. he Fundmnzen," 502 03, no. 11. 284 Kos, "The Monetary Circulation," 227, no. 28; Winter, "Die byzantinischen und karolingischen ," 339, no. 31. 285 J. Hunka, "Finds of Byzantine Coins from the 5 th 10 th Century From the Northern Part of the Carpathian Basin," in Byzantine Coins in Central Europe 397, table 1, no. 8. 286 Militk, "Finds of the Early Byzantine Coins," 375, no. C22. 287 Hajdnns Frjhalom jrs: Somogyi, "Byzantinische Fundmnzen," 272 273, no. 5. In addition, an unp rovenienced tremissis of Maurice from Rome in the collection of the museum in Miskolc and a solidus of Heraclius from Ravenna in the National Museum in Budapest, for which see Ibid., 291 92, no. 22; Somogyi, Byzantinische Fundmnzen 101 102, no. 91.

PAGE 426

426 where they cou ld have gained easy access to precious coins from western mints. 288 Second, no certain hoards are known either, which places the Khaganate in sharp contrast with the surrounding areas. To be sure, there is some indication of gold coin hoards found in the nin eteenth century at Bernecebarti (Pest, Hungary), Kupusina (West dispersed and no coins have been retrieved. Moreover, the available information suggests that none of them was later than 610. 289 Given the large quantity of Byzantine gold pouring in to the Avar khaganate, the absence of hoards in its core areas is quite perplexing. It is no less true that extremely few Avar settlements are known, most of the archaeological knowledge of the Avar period being derived from burial assemblages in which the deposition of large hoards is not customary. Finally, there is very little silver in Avaria, unlike the silver frenzy across Transcaucasia and the modest but still evident flow of hexagrams to the peripheral areas under the hegemony of the Avars. Hexagram (Arad, Romania), Linz (Obersterreich, Austria), and Szadzko (Stargard, Poland). 290 Wherever they are found in great quantities silver pieces are connected with major theaters of war or wi th political payments sent to allies, which the Empire perhaps did not consider important enough to require gold subsidies. The flow of Byzantine gold to Avaria dropped abruptly after 626, and despite the attempt of Attila Kiss to suggest that tribute cont inued to be paid in smaller amounts 288 P ohl, Die Awaren 238 40. 289 Somogyi, Byzantinische Fundmnzen 136 39. 290 Snnicolaul Mare and Arad: Oberlnder Hahn, "Die Fundmnzen des 5. 9. Jahrhunderts," Fu ndmnzen," 505 06, no. 18 and 511, fig. 6/13.

PAGE 427

427 after the failed siege of Constantinople, the numismatic evidence found in archaeological context indicates that the Avar khaganate no longer presented a serious threat for Byzantium. No coins struck between 625 and 655 have been found in archaeological context and the finds from burial assemblages dated from the second half of the seventh century indicate a drastically diminished flow of gold from Byzantium. 291 Still, the renewed infusion of Byzantine coins after a few dec ades of partial interruption requires an explanation. Continous, albeit reduced, tribute paid by following the death of Kubrat are not satistfactory solutions. 292 The m ost plausible so far region between the Middle Danube and the Caucasus around 650 may have led to renewed diplomatic relations between Byzantium and the weaker Avar khaganat e. The Avar embassy of 678 and the plausible assumption that Constans II and Constantine IV feared the growing power of the Onogur Bulgars provide sufficient grounds for this interpretation. 293 Gold solidi continued to be used as oboli as testified by the gr aves from Gyenesdis, Hajdnns Frjhalom jrs, and Kiskundorozsma in Hungary. 294 The scarcity of coins in the period after 626 often led to the use of symbolic round 291 A. Kiss, "Die "barbarischen" Knige des 4 7. Jahrhunderts im Karpatenbecken, als Verbndeten des rmischen bzw. byzantinischen Reiches," Communicationes Archaeologicae Hungaricae (1991): 122 23. For a convinc ing rebuttal, see Somogyi, Byzantinische Fundmnzen, 111 17, and more recently, Somogyi, "New Remarks on the Flow," 90 10 3 with the previous literature. 292 I. Bna, "Barbarische" Nachahmungen von byzantinischen Goldmnzen im Awarenreich," RIN 95 (199 3): 52 9 38. 293 Somogyi, "Ne w Remarks on the Flow," 125 35. 294 Somogyi, Byzantinische Fundmnzen 43, no. 26; Somogyi, "Byzantinische Fundmnzen," 272 273, no. 5 and 273 274, no. 6.

PAGE 428

428 shaped pieces of gold as oboli or the reuse of old Roman coins for this purpose as testif ied by burial assemblages from Hungary and Slovakia dating from this period. 295 Before the final assault on the Byzantine capital or perhaps shortly after, the Slavs rebelled under the leadership of Samo, who took advantage of the long term repercussions of the Avar defeat at Constantinple and established a 35 year long reign over what is often described as " stretching from the Elbe to the Middle Danube and the Drava. 296 The gold coins and imitations after coins of Heraclius and Constans II found Mostova, and s ephemeral political entity. 297 Seventh century finds from Poland have also been associated with a Slavic population. The follis of Heraclius (613/4) from Grodzisko Dolne was found in a Slavic settlement while the hexagr am from Szadzko and the solidus 298 of Byzantine inspiration as well as other Byzantine finds from the Czech Republic, Slovakia, and Poland sugges t some level of contact with Byzantine culture. 299 295 J. Zbojnk, "Antike Mnzen im Gebiet der Slowakei aus der Zeit des Awarischen Kh aganats," in Byzantine Coins in Central Europe 403 05; G aram, "A kzp avarkor," 210 11 296 Pohl, Die Awaren 256 61; Curta, The Making of the Slavs 109. Republic, see recently N. Profanov, "Awarische Funde in der Tsch echischen Republik Foschungsstand und neue Erkenntnisse," AAC 45 (2010): 203 70. For Avar archaeological evidence in Slovakia, see J. Zbojnk, "Das Awarische Kaganat und die Slawen and seiner nrdlichen Peripherie (Probleme der archologischen Abgrenzun g)," Slovensk Archeolgia 47, no. 1 (1999): 153 73. 297 Militk, "Finds of the Early Byzantine Coins," 381 82, no. M37; 376, no. C23 24; 378, no. C26; Somogyi, "New Remarks on the Flow," 94, n. 33; Somogyi, "Byzantinische Fundmnzen," 288, no. 18. s political center see recently F. Curta, "The Early Slavs in Bohemia and Moravia: A Response to My Critics." Archeologick rozhledy 61 (2009): 725 54. 298 01, no. 6 and 511, fig. 6/12; 505 06, no. 18 and 511, fig. 6 /13; 506, no. 19 and 511, fig. 6/14. 299 Ausgewhlte Probleme," in Byzantium and East Central Europe ed. G. Prinzing and M. Salamon

PAGE 429

429 Two hoards from the Czech Republic, found at 300 Slavic influence to the south, perhaps as far as Slovenia, expl ains the uninterrupted access to the Mediterranean. The structure of the hoards betrays a western origin. The small accumulation found during archaeological excavations on the left bank of the Hrozov river is made of four bronze coins, all issues from Car thage, including a Carthaginian issue of Zeugitania from the third century BC. The last coin is a rare follis of Constans II from Carthage, dated 662 667. Most probably the coins were brought as a homogeneous group from Tunisia, via Italy. Coins of Constan s II including issues from western mints have been found in Slovenia, at Celje and Ptuj, and in Austria, at Neulengbach and Wiener Neustadt indicating a possible communication bridge linking the Mediterranean, Bohemia, and Moravia. 301 y, of which only seven coins have been preserved, includes only early Byzantine coins, from Justin II to Constans II (last coin 651 657), and displays an array of mints but mostly western, from Catania, Rome, and Carthage. Once again an origin in North Afr ica is quite conceivable. This may be related with advantage of the First Fitna (656 6 61) and check the Arab expansion to the west, as he did in Transcaucasia. In fact the emperor established himself in Sicily in 663 and impos ed heavy taxation on southern Italy and North Africa, (Cracow: "Historia Iagellonica", Jagiellonian University, 2001), 49 Pr aehistorica 21 (1994): 69 103. See also A. Avenarius, Die byzantinische Kultur und die Slawen: zum Problem der Rezeption und Transformation (6. bis 12. Jahrhundert) (Vienn a: R. Oldenbourg, 2000), 20 49. 300 Militk, "Finds of the Early Byzantine Coins," 376 77, no. C25; 382, no. M38. 301 Ptuj: Kos, Die Fundmnzen vol. 3, 497, no. 916. Celje: Kos, "The Monetary Circulation," 227, no. 29 30. Neulengbach and Wiener Neustadt: Winter, "Die byzantinischen und karolingischen ," 341, no. 9; 340 41, no. 5 and 351, fig. 6/19.

PAGE 430

430 while at the same time trying to stabilize the political situation in Africa. 302 Communities in Bohemia maintained a century long connection with North Africa and Italy, undocumented by written evidence, and p robably established in the context of Finally, the hoard from Zemiansk Vrbovok (Zvolen, Slovakia), concealed after 669 674, is not only the northernmost collective find of Byzant ine silver but also a unique accumulation in many respects. 303 It is the upscale version of the hoard of coins century and can be compared only to the collections of Late Roma n coins and scrap silver from Gudme in Denmark. 304 Because of the association of coins with silver bowls, plate (Sassanian), jewelry, and scrap silver the hoard from Zemiansk Vrbovok was attributed to a silversmith, perhaps Byzantine, who worked for the Ava rs and catered to the tastes of the barbarian elite. 305 To be sure, the association of seventh century Byzantine silver coins with silver jewelry is not unique to Zemiansk Vrbovok. A contemporary but much larger hoard of hexagrams found at Priseaca in Olten ia included two star shaped silver earings. However, most of the coins from Zemiansk 302 Muslim Expansion and Byzantine Collapse in North Africa New York: Cambridge University Press, 2010), 166 99. See also D. Pringle, The Defence of Byzantine Africa from Justinian to the Ar ab Conquest. An account of the Military History and Archaeology of the African Provinces in the Sixth and Seventh Centuries BAR Int. 99 (Oxford: B.A. R., 1981), 46 48. 303 P. Radomersk, "Byzantsk mince z pokladu v Zemi anskm Vrbovku," Pamtky Archeologick 44 (1953): 109 22. A. Fiala, "K objavu miliarense Constansa II. z pokladu zo Zemianskho Vrbovku." Numismatick sbornk 17 (1986): 15 20; Avenarius, Die Byzantinische Fundmnzen, 37 with fig. 8 for the objects. 304 P. Vang Petersen, "Excavations at Sites o f Treasure Trove Finds at Gudme," in The Archaeology of Gudme and Lundeborg. Papers presented at a Conference at Svendborg, October 1991 ed. P. O. Nielsen, K. Randsborg, and H. Thrane (Copenhagen: Akademisk Forlag, 1994), 30 40. 305 Hunka, "Finds of Byzanti ne Coins," 396 97; Zbojnk, "Antike Mnzen," 406 08; Avenarius, Die Byzantinische Fundmnzen, 36 37.

PAGE 431

431 Vrbovok are not hexagrams but miliarenses ceremonial issues of irregular weight, two times lighter than the regular silver coinage. The coins are die linked and uncircul ated which confirms the supposition that they represent a diplomatic gift of freshly minted coins. 306 One early miliarensis of Constans II was found in Georgia, the most important region for the distribution of Byzantine ceremonial silver pieces. 307 The same r egion produced the two die linked gold hoards found at Chibati and Nokalakevi, which reprent homogeneous groups sent as payment from Constantinople, much like the collection from Zemiansk Vrbovok. Much closer to Slovakia, another miliarensis of Constans I I was found in a grave from Stejanovci (Vojvodina, Serbia). 308 type silver imitations after miliarenses struck for Constans II and Constantine IV prove that they had been widely known in the Avar khaganate. 309 In conclusion, the hoard from Zemiansk Vrbovok may indeed have belonged to a crafts man, but the coins were most probably part of an imperial largesse, which the recipient subsequently gave away as raw material for the production of jewelry and plate. The gift may have been occasioned by one of the Avar Byzantine contacts preceding the em bassy of 678 and it is quite probable that the hexagrams of Constans II and Constantine IV were random additions to the homogeneous group of miliarensia of Constans II sent to Avaria during 310 306 DOC II, 19. 307 Abramishvili, Sakartvelos sakhelmts'ipo muzeumis, 70, no. 220 and pl. XIII/220. 308 Somogyi, Byzantinische Fundmnzen, 78 79, no. 68. 309 Ibid ., 127. 310 predecessor is unlikely and might have been ill received by the Avar envoys. See Somogyi, "New Remarks on the Flow," 132.

PAGE 432

432 6.3 Money and barbarians : Same Coins Different Functions After discussing the historical contexts which facilitated access to Byzantine coins in barbaricum we should now turn to the other main problem, the function of money in societies living beyond the frontier of the early Byzantine Empi re. It is a widely accepted postulate that money has no economic function beyond the political borders of the issuing authority. That is not to say that coins cannot perform any economic function whatsoever. It basically means that the state no longer guar antees the value of the currency, the exchange rate, and the enforcement of other regulations in place. This fact alone affects Byzantine money in different ways depending on the metal. Gold and silver coins were hardly the currency used on a daily basis a t the marketplace. The main purpose of gold was to create a means for the payment of salaries and the collection of taxes, and of course, to serve as a medium of exchange in large scale transactions. As we have already seen, the main purpose of silver, at least until the introduction of the hexagram in the seventh century, was ceremonial. It served no economic monetary function although it was loosely struck on the weight standard of the siliqua or the miliarensis. Consequently, we should not expect any mon etary function of gold and silver Byzantine coins found in barbaricum. It is common knowledge that despite the Roman law prohibiting the export of precious coin, large quantities of coined gold were sent to barbaricum or to Persia in the form of political and diplomatic payments, almost never to return back to Byzantium. As already hinted in the previous sections Byzantine gold and silver coins were mainly used for non economic purposes, as jewelry or raw material for the production of jewelry and dress acc essories, and for the ritual deposition in graves.

PAGE 433

433 The function of copper coins, however, is much more problematic. This was the currency most frequently circulated in the Byzantine provinces, being the main facilitator of small scale transactions in urba n markets. Although the highly militarized environment of the northern Balkans dominated by hilltop fortresses rather than urban agglomerations was not characterized by a "market economy," copper coins still performed their designated monetary function. Th e major question, therefore, is what have preferred an economic interpretation of the se coins, part and parcel of what seemed to be an extension of the Byzantine monetary economy beyond the frontier. It is not surprising that most scholars working on this topic in the past decades have analized the presence of Byzantine copper coins almost exclusively in terms of an "intense monetary circulation" or at least an "active circulation" after the reform of Anastasius which brought an "economic impetus" in barbaricum 311 Most of the time the use of small value currency was ascribed to the local Ro manic population, in place since the time of the Roman province of Dacia, as an ethnic antithesis to waves of migratory peoples who did not have the benefit of a time 311 C. Preda, "The Byzan tine Coins An Expression of the Relations between the Empire and the Populations North of the Danube in the 6 th 13 th Centuries," in Relations Between the Autochthonous Population 221 (active coin circulation); Teodor, Romanitateacarpato 38 (i ntense coin 68 (coin circulation and frequent transactions). I. Corman, nistrian n epoca evului mediu timpuriu (sec. V VII d.Chr.) Cartdidact 1998) 98 99 (coin circulation); Teodor, 34 (uninterrupted coin circulation); Stoliarik, Essays on Monetary Circulation, 68 (real coin circulati Societatea carpato danubiano XI. Structuri demo politice 1997) 174 (coin circulation); A M. Velter, Transilvania n secolele V XII 33 35 (coin circula tion and active trade). For similar arguments in favor of monetary circulation in Georgia, see N. Iu. Lomouri, "Arkheopolis Tsikhegodzhi Nokalakevi," in Nokalakevi Arkeopolisi: III. Arkeologiuri gatchrebi 1983 1989 ed. P. Zakaraia (Tbilisi: Metsniereba, 1 993), 136 39.

PAGE 434

434 honored tradition of economic contact with the Roman Empire. Some scholars have gone as f ar as to suggest that Byzantine coins were also used for internal trade, i.e. horizontal exchanges between communities in Barbaricum apropiat, as opposed to vertical dealings with Byzantium. 312 Along these lines, coins were used for tracing the arrival of ne w ethnic groups, interruptions in "coin circulation" signifying the disruption of normal relations between the autochtonous population and the Byzantine Empire provoked by peoples ignorant of monetary transactions. 313 Although the uncompromising thesis of a true coin circulation has been undermined in the past decade, the larger paradigm has not been challenged. 314 Lip service has been given to alternative functions performed by Byzantine coins and to the realization that societies of quite different ethnic bac kground handled Byzantine coins, but such vistas have not been properly explored. 315 Unfortunatelly Byzantine sources tell us little about the use of Byzantine coins in barbaricum and there is virtually no useful information about trade. Information about th e political organization and the social structure of societies in barbaricum is fragmentary at best and comes exclusively from Byzantine writers who filtered realities 312 Corman, 99; Velter, Transilvania n secolele V XII 35. 313 Teodor, 12; Oberlnder, Barbaricum apropiat, 332 58. For a more nuanced interpretation see E. S. Teodor, VI lea de la Dulceanca, in Istro Pontica: Muzeul tulcean la a 50 a aniversare, 1950 la 45 de ani de activitate ed. M. Iacob, E. Oberlnder Trnoveanu, and F. Topoleanu (Tulcea: Consliul gued that Slavs could also be interested in handling Byzantine coins, only to conclud later that the interrupton of the coin flow may be connected with the massive arrival of C epoca lui Justinian," 12 (2003): 334. 314 See especially Oberlnder, "La monnaie byzantine," 177 81; Lakato s, "Monede bizantine," 239 46. 315 Oberlnder, La monnaie byzantine, 182. For the function of Byzantine coins as raw material for the production of bronze jewelry and dress accessories, see A. "Face Value or Bullion Value? Early Byzantine Coins beyond the Lower Danube Border," in Byzantine Coins in Central Europe 455 59.

PAGE 435

435 from barbaricum through their own cultural sieve. Archaeology can certainly shed more li ght on these issues, especially future research if conducted with such questions in mind. For the time being, however, the only avenue leading to a better understanding of the use of money by "barbarians" is the parallel with more recent traditional societ ies in contact with capitalist states using a carefully regulated currency. The methodological limitations of such an approach are readily apparent and difficult to overcome. Naturally, the economy of Europe from the Age of Discovery onward is quite differ ent from the economy of the early Byzantine Empire. Moreover, the particular culture of each "primitive" society studied by anthropologists, which also translates in the way foreign currencies were perceived and adopted, invites to caution in pushing compa rison beyond the level of generality. Nonetheless, some similarities make this a worthwile endeavor. Although Byzantine influence in barbaricum cannot be equated with the colonial administration of modern empires, the exposure of village societies to an a lien economic system based on the use of money remains quite similar and might have elicited comparable general responses regardless of time and space. To be sure, territories in barbaricum were never directly administered by Byzantium. Several attempts by Justinian to encroach upon the internal autonomy of Armenia and Lazica ended in rebellion and no similar administrative ambitions can be traced in the northern Balkans aside from the settlement of various ethnic groups as foederati with the purpose of def ending the frontier. As a result the use of money was never officially implemented, nor was taxation, although it was mentioned as one of the chief reasons for the defection of the Lazi in 541. Consequently, the role of Byzantine coins in barbaricum cannot be fully

PAGE 436

436 equated with the function of money in "primitive societies" integrated in European colonial empires. However, much can be learned from the initial responses to the introduction of European currency in the earlier stages of contact predating the d irect annexation or even in the early colonial period. At the very least the comparison can be reduced to the shared denominator of cultural contact between monetized states and stateless non monetizied societies. There are three major topics able to prov ide important comparative insights into the possible functions performed by Byzantine coins in barbaricum : alternative "monetary instruments" used by traditional societies, the motivation behind the use of foreign curencies, and the morality of exchange me diated by money. The fact that traditional societies did not use money with the physical resemblance and the general purpose function of western currencies does not mean that they did not employ other instruments of exchange. 316 What makes them different fro m the European all purpose money is their dual function as economic instruments and symbolic objects used in rituals and ceremonies. Their power was not so much derived from their instrinsic properties, although that was also important, as from their socia lly embedded value. This duality reflects the dichotomy between economic and 316 For primitive money, see C. Haselgrove and S. Krmnicek, "The Archaeology of Money," Annual Review of Anthropology 41 (2012): 235 50; P. Sibilla, "A proposito di moneta. Lineamenti di un percorso di ricerca," in Antropologia dello scambio e della moneta ed. P. Sibilla (Torino: Lib reria Stampaori, 2006), 9 46. From the older literature, see especially P. Bessaignet, "Monnaie primitive et thorie montaire," RESS 21 (1970): 37 65; G. Tucci, Origine et dveloppement de la monnaie primitive," RESS 21 (1970): 17 36; K. Polanyi, "The Sem antics of Money Uses," in Primitive, Archaic, and Modern Economies: Essays of Karl Polanyi ed. G. Dalton (Garden City, NY: Doubleday, 1968), 175 203; G. Dalton, "Primitive Money," AA 67, no. 1 (1965): 44 65; P. Einzig, Primitive Money in Its Ethnological, Historical, and Economic Aspects 2 nd ed. (N ew York: Pergamon Press, 1966).

PAGE 437

437 non economic exchange which can be noticed in many cultures, the Melanesian case studied by Malinowski being the most famous. 317 In the early modern period and in some places unti l the twentieth century, cowrie shells were used as the main medium of exchange. The cowrie ( cypraea moneta ) cannot be fully defined as "international currency" despite the fact that it was widely used from Africa to Oceania because its value and exchange rate were locally determined based on idiosyncratic criteria seldom dictated by economic principles. Nevertheless, the use of such standardized objects as medium of exchange was facilitated by the development of long distance trade where a mutually accepte d instrument was required. Cowries and beads introduced by Europeans played this role for the greater part of the modern period. Moreover, cowries should not be treated as an eminently primitive monetary instrument. The cowrie system knew various lower den ominations, forgeries, agreed upon prices and exchange rates, as well as the dire consequences of inflation which could render large accumulations practically worthless. 318 The most revealing example for the "modern" nature of cowrie based transactions is th e economy of Kapauku Papuans, who had a primitive monetary 317 B. Malinowski, Argonauts of the Western Pacific: An Account of Native Enterprise and Adventure in the Archipelagoes of Melanesian New Guinea (London: Routledge, 1922). See more recently A. Gell, "Inter Tribal Commodity Barter and Reproductive Gift Exchange in Old Melanesia," in Barter, Exchange and Val ue: An Anthropological Approach ed. C. Humphrey and S. Hugh Jones (New York: Cambridge University Press, 1992), 142 68. See also the essays in D. Akin and J. Robbins, ed., Money and Modernity: State and Local Currencies in Melanesia (Pittsburgh, PA: Unive rsity of Pittsburgh Press, 1999). For the duality, see K. Hart, "Heads or Tails? Two Sides of the Coin," Man 21, no. 4 (1986): 637 56; M. Godelier, Perspectives in Marxist Anthropology (New York: Cambridge University Press, 1977) 128 29. 318 C. A. Gregory, S avage Money: The Anthropology and Politics of Commodity Exchange (Amsterdam: H arwood Academic, 1997), 236 42.

PAGE 438

438 economy, including credit, the dominance of profit motivated transactions, and prices regulated by demand and supply. 319 Other monetary instruments included coins made of clay of different shape s, usually pierced, used by the Sao peoples in Camerun, Chad and Nigeria. 320 Most of them were star shaped and their value depended on the number of corners and whether they were pierced or not. Much like other primitive forms of currency such clay coins wer e also used as symbolic offerings. 321 The Buduma in the region of Lake Chad, the Lamba from Congo and the Baruya from New Guinea used units of salt as their main medium of exchange. 322 In Madagascar before the colonial period cattle was the main medium of exch ange and means of storing wealth. 323 The Kotoko from Chad used stone coins in the precolonial period, mainly sone axes and smaller pieces of stone as lower denominations. 324 Rolled strips of cotton and a variety of iron objects were other acceptable monetary i nstruments in Central Africa. 325 Cloth was the most important type of currency in Guin the Cape Verde islands. Various agicultural products and cotton 319 L. J. Pospisil, The Kapauku Papuans of West New Guinea (New York: Holt, Rinehart and Winston, 1963), 300 09. 320 For a comprehensive list of pr imitive money across the ages, see especially Einzig, Primitive Money 29 308. 321 J. P. Lebeuf, "Monnaies archaques africaines de terre cuite," RESS 21 (1970): 67. 322 A. Dorsinfang RESS 2 1 des merchandises chez les Baruya de Nouvelle Guine," RESS 21 (1970): 134. 323 J. Dez, "Monnaie et structures traditionelles Madagascar," RESS 21 (19 70): 185. 324 Lebeuf "Monnaies archaques africaines," 81 83; The circulation of polished stone axes was an old practice of Neolithic Europe, for which see E. Thirault, "The Politics of Supply: The Neolithic Axe Industry in Alpine Europe," Antiquity 79 (2005 ): 34 50. 325 Lebeuf "Monnaies archaques africaines ", 72 80; Dorsinfang Smets, "Les moyens d'change," 102 03

PAGE 439

439 were used in the Andes as monetary instruments in the precolumbian period, while in Mesoamerica cocoa bean s and large white cotton cloaks ( quachtli ) served this purpose. 326 Even in the early colonial period cocoa beans and Spanish coins circulated side by side with regulated exchange rates between the two accepted monetary instruments. The main motivation be hind the use of foreign currencies was the demand for cash revenue, which forced cultivators to sell a portion of their crop as well as manufactured goods to raise money. Much like the case of Byzantine villages, in well established colonial settings money was needed to pay taxes but in many other cases foreign money was the only form of payment one could use to purchase foreign goods needed in the household or to increase the social prestige of the owner. 327 European coins introduced in Congo in the 1920s we re used primarily to procure Western goods. 328 The Azande saw money as a special type of instrument needed to get access to goods they could not produce themselves and almost nothing was spent on subsistence goods. 329 The Hausa people in Nigeria first encounte red western coins when traders from the Mediterranean coast of North Africa extended their business to the south. Payment had to be made in coin and the local elite had to raise cash in order 326 C. Lazo Garca, Economia colonial y regimen monetario. Peru: siglos XVI XIX. Tomo I: Nascimiento e instauracin del sistema econmico monetario colonial (siglo XVI) (Lima: Banco Central de Reserva del Per, 1992), 55 57; J. L. de Rojas, La moneda indgena y sus usos en la Nueva Espaa en el siglo XVI (Tlalpan, D.F.: Centro de Investigaciones y Estudios Superiores en Antr opologa Social, 1998), 1 83 87. 327 C. J. Fuller, "Misconceiving the Grain Heap: A Critique of the Concept of the Indian Jajmani System," in Money and the Morality of Exchange ed. J. Parry and M. Bloch (New York: Cambrid ge University Press, 1989), 45. 328 Dorsinfang Smets, "Les moye ns d'change," 108 10. 329 C. R. Reining, "The Role of Money in the Zande Economy," AA 61, no. 1 (1959): 40.

PAGE 440

440 to buy western products such as weapons. 330 To the Lhomi from Nepa l money was useful for purchasing cooking oil, lamps, and medicine from the bazaar. 331 In most cases however, coinage was not needed or was even banned from the purchase of food or other locally produced subsistence goods and the general impression is that n othing indispensable was bought with coins. In this sense the western goods procured by primitive societies resemble the luxury Roman goods aqcuired by barbarian elites to bolster their social standing. One of the keys to understanding whether communities in barbaricum accepted Byzantine coins as a form of payment is to determine the level of trust needed for such transactions to take place. The value of the copper follis fluctuated a great deal in the sixth century, usually downwards, being constantly depr eciated in relation to the gold solidus. Ethnographic parallels suggest that non monetized village communities were reluctant to use money and prefered barter. It has been shown that for many Africans money was something eminently unsecure, something to ge t rid of before it becomes worthless. 332 People as diverse as the Nepalese, the Nigerians, the Kwaio from Solomon Islands, and the Greek villagers from Ambli on the island of Euboea distrusted money because of its unstable value and fiduciary nature. 333 330 G. Nicolas, "Circulations des biens et changes montaires en pays haoussa (Niger)," RESS 21 (1970): 114. 331 C. Humphrey, "Barter and Economic Disin tegration," Man 20, no. 1 (1985): 57. 332 P. M. Shipton, Bitter Money: Cultural Economy and Some African Meanings of Forbidden Commodities ( Washington, D.C.: American Anthropological Association, 1989), 6. 333 Humphrey, "Barter and Economic Disintegration," 63 ; J. Du Boulay, Portrait of a Greek Mountain Village (Oxford: Clarendon Press, 1974), 37; S. Heap, Money in Africa ed. C. Eagleton, H. Fuller, and J. Perkins (London: British Museum, 2009), 35; D. Akin, "Cash and Shell Money in Kwaio, Solomon Islands," in Money and Modernity 109 13.

PAGE 441

441 More over, no issuing state wants to see its coinage drain outside the frontier, especially when it is not token coinage but essentially commodity money. The Byzantine legislation prohibiting the export of precious metal coins finds an interesting equivalent in modern concerns with the same issue. The natives in Madagascar got so fond of European silver that they accepted only monetary payments in exchange for their goods. Fearing that large quantities of silver will be lost European powers were desperately tryi ng to discourage this practice complaining that "the silver 5 Franc pieces are diligently sought by the natives from Madagascar in e x change for their products which they would not sell, most often, unless they are paid in piasters which they bury with thei r dead or convert into ornaments and jewelry" as testified by a document from 1867. 334 Such practices were sometimes a form of resistance to the ambitions of the colonial state, like the case of British West Africa where the Africans continued to melt down E uropen money and use indigenous currencies instead. 335 In the initial phase of contact silver coins were acquired like any other European object and used as ornaments in necklaces or bracelets. The most common silver coins were piasters, reali ("pieces of ei ght"), and especially Maria Theresa thallers. 336 Again the economic vs non economic duality is present. The case of the Swahili pre colonial 334 J. Chauvicourt and S. Chauvicourt, Numismatique malgache. (Antananarivo: Trano Printy Loterana, 1968), 24 25 : "Quant aux pices de 5 francs en argent elles sont recherches avec avidit par les naturels de Madagascar en change de leur enfouies chez eux ou conve rties en ornements ou parures." 335 H. Fuller, "From Cowries to Coins: Money and Colonialism in the Gold Coast and British West Africa in the early 20 th Century," in Money in Africa 60. African nickel coins, which had a hole in the center, were used by carpenters and smiths as washers for nails and screws, see A. A. Lawal, "The Currency Revolution in British West Africa: An Analysis of Some Early Pro blems," in Money in Africa 63. 336 Eastern Economic Journal 27, no. 4 (2001): 443 62.

PAGE 442

442 Kilwa Islamic coinage is a case in point. The town of Kilwa Kisiwani on the Swahili coast minted its own coinage in t he name of various sultans from the eleventh through the fifteenth century. Most of the coins were copper issues destined for local circulation but their function was not exc l usively economic. Many were used as commemorative objects and some five percent o f the coins were pierced indicating that they were used as pendants. 337 This is strikingly similar to the figure of ca seven percent calculated for the number of pierced folles of Justinian, probably used for the same purpose. 338 On the other hand, monetary d ealings with foreigners were done only in precious metal coinage. Nicolas Mayeur, an eighteenth century trader and traveler in Madagascar drew attention to the fact that the Merina are willing to accept "any coins provided they are of silver and have the p roper weight," adding that "they may accept scrap or crafted silver but prefer it monetized and are even willing to lower their prices if they know that payment will be done in coin." 339 The adoption of European coins as an accepted form of currency in Madag ascar led to the practice of cutting the coins in halves and quarters and as small as 1/720 of the original size, in order to be used as lower denominations. 340 337 S. Wynne Jones and J. Fleisher, "Coins in Context: Local Economy, Value and Practice on the East African Swahili Coast," Cambridge Archaeological Journal 22, no. 1 (2012): 32. 338 zantine Coin Circulation," 177. 339 N. Mayeur, "Voyage dans le sud et dans l'intrieur des terres et plus particulirement au pays d'Hancove," Bulletin de l'Acadmie malgache 12 (1913): 163: "Toutes espces monnoyes quelconques, pourvu qu ou travaill mais ils donnent la prfrence a celui qui est monnoy, et mme font une diminution assez s eront pays de cette manire." 340 S. Kus and V. Raharijaona, "Small Change in Madagascar: Sacred Coins and Profane Coinage," in The Archaeology of Politics: The Materiality of Political Practice and Action in the Past ed. P. G. Johansen an A. M. Bauer ( Newcastle upon Tyne: Cambridge Scholars Publishing, 2011), 39.

PAGE 443

443 Indeed, one of the most important functions of foreign coins in primitive societies on four diffe rent continents was that of jewelry or bullion for the production of bracelets, necklaces and earrings and sometimes jewelry played the role of monetary instrument. 341 This practice is cl early reflected in Madagascar: In this non monetized country coins were originally integrated in the local system according to their intrinsic value not based on their monetary function. Coins were nothing else but objects to be bartered. They could be assembled to create dress sets or melted d own to produce silver jewelry or silver blades to be deposited in royal coffins. 342 The use of metal in different forms and shapes as currency was well known in Africa. Copper bars were used as currency in Niger, a fact recognized by Ibn Battuta who visited the region in 1352. 343 Small wire like objects (3 4cm long) called fils double tte were used in political centers of the empires of Ghana and Mali between the eleventh and the fourteenth century and various copper ingots were used in the Copperbelt regi on of Central Africa. 344 The Mongo from Congo used red copper cylinders or copper jewelry as media of exchange. In addition, the so called Katanga crosses made of copper (weighing ca 750g or more) were valued for the metal and 341 Chauvicourt and Chauvicourt, Numismatique malgache 4. 342 connaissait pas la monnaie, sera galem de troc, et les pices de monnaie se retrouveront runies en parure, ou fondues pour fair e de bijoux It was said that king Radama, who died in 1828, was buried in a silver coffin made from fourteen thousand Spanish silver coins, weighing almost 375 kg and several o ther examples are known from the nineteenth century. See Chauvicourt and Chauvicourt, Numismatique malgache 7. 343 N. Levtzion and J. F. P Hopkins, ed., Corpus of Early Arabic Sources for West African History trans. J. F. P. Hopkins, ( Princeton, N.J.: Mark us Wiener Publishers, 2000), 302. 344 L. Garenne double tte based Ingots : Copper Money objects at the Time of the Sahelian Empires of Ancient Ghana and Mali," in Money in Africa 1 montaire du Shaba Central entre le 7 e et le 18 e sicle," African Economic History 10 (1981): 125 28.

PAGE 444

444 melted down to produce copper jewelry. The smaller versions are often found in burial assemblages dating from the Early Modern period. 345 Similarly, the Tiv people in Nigeria used three foot brass rods as money, which they often melted down to cast jewelry. 346 The Lhomi from Nepal melted d own silver coins to produce bracelets or betls or used 347 The Fali from Cameroon used pierced French and English silver coins to adorn leather bands worn around their heads. 348 The same social prestige was sought by villagers from Ambli in Euboea who displayed their wealth by having a gold coin sewn on to the front of shoes to be worn on special occasions. 349 Iconography is also important as testified by the case of the silver 5 Franc piece displaying Hercules and two muses (Trinity allegory) often used as pendant on silver necklaces worn by young men from Madagascar. 350 Finally, the morality of monetary exchanges goes back to Aristotle and is one of the main topics studied by anth ropologists in relation to the use of money in primitive societies. 351 It is undeniable that money carried a negative symbolism in many 345 Dorsinfang on montaire du Shaba," 128 43. 346 P. Bohannan, "The Impact of Money on an African Subsistence Economy," The Journal of Economic History 19, no. 4 (1959): 498. 347 Humphrey, "Barter and Economic Disintegration," 63. 348 J. P. Lebeuf, Vtements et parures du Cameroun franais (Paris: ditions Arc en Ciel, 1946), 37 38 and pl. II/D, III/A, and VI/B C. 349 Du Boulay Portrait of a Greek Mountain Village 244. 350 Kus and Raharijaona, "Small Change in Madagascar ," 47 48. 351 M. Hnaff, The Price of Truth: Gift, Money, and Philosophy (Stanford, CA: Stanford University Press, 2010).

PAGE 445

445 societies. 352 The traditional system of exchange in most village societies was some form of barter. In many cases this remai ned the preferred form of exchange even after the introduction of European money. Whether we are talking about the Lhomi of Nepal, the Merina of Madagascar or the east African farmers in Kenya barter was dominated by the idea of equal exchange, a morally a cceptable and socially fulfilling means of acquiring subsistence goods. 353 European money was seen as a potential threat to the social structures in place. Money served to depersonalize social relations, to simplify them and reduce their meaning. The case of Madagascar is perhaps the best documented. The Merina in Madagascar drew a clear distinction between unpaid work performed as community service ( miasa ) and wages received from foreigners ( mikarama ), the latter gaining a clearly negative connotation when they interfered with ancestral laws. 354 For the peoples of the eastern coast "money is presented as having a power eminently corrosive, it is the main reason for the destruction of kinship, the binding solidarities of the village." 355 As a consequence, those who were employed and received money wages from foreigners were excluded from the village community, becoming themselves strangers through their economic position. 356 352 J. Parry and M. Bloch, "Introduction: Money and the Morality of Exchange," in Money and the Morality of Exchange 21; N. Morley, Trade in Classical Antiquity (New Yor k: Cambridge University Press, 2007), 79 89. 353 Humphrey, "Barter and Economic Disintegration," 64; Dez, "Monnaie et structures traditionelles," 180. 354 M. Bloch, "The Symbolism of Money in Imerina," in Money and the Morality of Exchange 17 5. 355 G. Althabe "Circulation montaire et communauts villageoises malgaches," RESS 21 (1970): 149: grand destructeur des lignages, des solidarits constit utives du villag e." 356 Ibid., 164.

PAGE 446

446 The same threat to the authority of the elders can be perceived in the function of money i n Dande communities from Zimbabwe, who were slightly more liberal with the acceptance of coins so long as it did not destroy social traditions. 357 The Kwaio from Solomon Islands are more conservative and refuse to allow the use of cash in prestige economy, d rawing a clear distinction between Eurpean and traditional Kwaio practices. 358 The means by which money was acquired was also important. In Kenya and Gambia, money obtained through theft, usury or by accident is considered "bitter money," boding ill for its owner. 359 Money was often "purified" to gain a neutral or even positive symbolism. In Madagascar money was often obtained as payment for agricultural products but it was spent on prestige gaining activities such as building houses or honoring the ancestors b y building funerary monuments. 360 Another form of purification was the royal ceremony in which the king received tokens of deference ( hasina ) in the form of coins. 361 In Fiji the traditional ritual of fundraising presupposed the purification of money through t he ceremony of "drinking cash". 362 Coins also had a magic symbolism, which was usually connected to the symbolism of precious metals, or to their apotropaic force, known from 357 D. Lan, "Resistance to the Present by the Past: Mediums and Money in Zimbabwe," in Money and the Morality of Exchange 203. 358 Akin, "Cash and Shell Money," 113 26. The fear that money posed a threat to social structure was not univers al. See for instance the case of the Kwanga from Papua New Guinea who did not draw an opposition between money and local media of exchange, K. Brison, "Money and the Morality of Exchange Among the Kwanga, East Sepik Province, Papua New Guinea" in Money and Modernity 155 61. 359 Shipton, Bitter Money 28, 72 73. 360 Althabe "Circulation montaire," 156. 361 D. Graeber, Toward an Anthropological Theory of Value: The False Coin of Our Own Dreams (New York : Palgrave, 2001), 232 37. 362 C. Toren, "Drinking Cash: The P urification of Money Through Ceremonial Exchange in Fiji," in Money and the Morality of Exchange 151.

PAGE 447

447 ancient times. 363 In Madagascar coins were used in rituals to insure immunity before appearing in court and were also used in purification rituals by being thrown into water before the ceremony of washing. 364 In India silver rupee coins could symbolize deities and were kept in clay pots storing the spiritual wealth of the clan. 365 In the Andes foreign coins are hoarded and buried into the ground constituting the treasure of the community, dug up and counted every year during a rain calling ceremony ( Chaopincha ) 366 In Mexico, Central America and Columbia money was "baptized" to become officially acceptable. 367 Common monetary instruments or specially designed ones were used as part of the traditional bride price in many primitive societies, such as Musanga shell disc money from Congo, but also on the Greek island of Euboea where gold coins were stor ed in chests to be given away as dowry. 368 What was the function of sixth seventh century Byzantine coins in barbaricum ? Gold and silver coins clearly had no monetary role. They were stored for their intrinsic value and constituted objects of prestige, much like Roman coins in barbaricum in the previous centuries. 369 Hoards like the ones found at Chibati and Nokalakevi in Georgia, 363 H. Maguire, "Magic and Money in the Early Middle Ages," Speculum 72, no. 4 (1997): 1037 54. 364 Dez, "Monnaie et structures tradit ionelles," 182 83. 365 Gr egory, Savage Money 71 72. 366 Ethnographic research and footage taken in 1996 by Axel Nielsen, Consejo Nacional de Investigaciones Cientficas y Tcnicas, Buenos Aires, Argentina, in the province of Nor Lipez, Department Potos, Bolivia. 367 Shipton, Bitter Money 75. 368 Dorsinfang Smets, "Les moyens d'change," 107; Du Boulay, Portrait of Greek Mountain Village 244. 369 A. Bursche, Later Roman Barbarian Contacts in Central Europe: Numismatic Evidence (Berlin: Mann, 1996). See also A. Bursche, "Function of Roma n Coins in Barbaricum of Later Antiquity. An Anthropological Essay," in Roman Coins Outside the Empire: Ways and Phases, Contexts and Functions 416.

PAGE 448

448 Malo Pereschepyne in Ukraine and Zemiansk Vrbovok in Slovakia indicate imperial gifts or political payments sent by Constantinople The large number of gold coins found in burials and the frequency of solidi and tremisses pierced or modified to be worn as jewelry indicates that Byzantine coins played an important role in displaying social status in societies from barbaricum Gold pla ted imitations of Byzantine solidi found in the Carpathian Basin and in the Ciscaucasus and produced for those who could not afford genuine pieces emphasizes the high demand for coins and underscores their non economic function. 370 Less than 200 gold coins h ave been found in the Carpathian Basin dominated by the Avars although perhaps other unproveniened coins from Central Europen museums can be added to increase the total. In any case, the number of surviving specimens is just a tiny fraction of the huge qua ntity of solidi sent to Avaria and reflects not only the usual survival rate of ancient coins but also the fact that many were melted down to produce jewelry. The use of coins to produce jewelry in barbaricum has already been tested by finding correlation s between the weight of coins and that of small pieces of jewelry, usually earrings with pendants and bracelets. 371 However, such calculations are not always relevant or even required to demonstrate that coins were used as raw material. Goldsmiths cut coins as necessary to produce jewelry and ornaments as testified by the eighth 370 I. Bna, ""Barbarische" Nachahmungen von byzantinischen Goldmnzen im Awarenreich," RIN 95 (1993): 529 38; Prokopenko, "Byzantine," 545 50; S. V. Gusev, Severo Vostochnyi Kav kaz v epokhu t (Moskow: Institut etnologii i a ntropologii RAN, 1995), 19 24. 371 I. Bna, "Byzantium and the Avars: The Archaeology of he First 70 Years of the Avar Era," in From the Baltic to the Black Sea: Studies in Medieval Archaeology ed. D. Austin and L. Alcock (London/Boston: Unwin Hyman, 1990), 117; J. Werner, "Die frhgeschichtlichen Grabfunde vom Spielberg bei Erlbach, Ldkr. Nrdlingen, und von Frst, Ldkr. Laufen a.d. Salzach," Bayerische Vorgeschichtsbltte r 25 (1960): 171 72.

PAGE 449

449 were found together with cut gold coins. 372 The use of gold coins for such purposes is also confirmed by contemporary accounts. When discussing the external politics of weakness who were compelled to send 40,000 gold solidi every year to the Turks to keep them quiet. Lavished by s uch gifts "this particular nation had turned to great extravagance; for they hammered out gold couches, tables, goblets, thrones, pedestals, horse trappings, suits of armor, and everything which has been devised by the inebriation of wealth." 373 Official pay ments from Constantinople were not the only means of distributing coins to barbaricum. Plunder, ransom for prisoners, and enslavement of captives taken from the Byzantine provinces are the most important alternatives recoded by written sources. Additional possibilities, such as casual loss by mercenaries returning home, travelers or traders are other potential means by which coins arrived in barbaricum. A clear distinction should be made between gold and copper coins when discussing such possibilities. Roma n POWs and especially VIPs fallen in captivity were ransomed for gold coins without exception, when a monetary ransom is involved. In the fifth century the Huns required 9 gold pieces for each Roman POW 374 By 600 the Avars had much more modest demands when they requested one solidus for each prisoner. The khagan was ready to bargain when Maurice refused the initial offer and lowered the sum to solidus but according to Theophanes the emperor would not even agree to pay 4 keratia (1 solidus = 24 keratia ) so the khagan slew the prisoners angered at the 372 I owe this information to Gabriela Bijovsky, Israel Antiquities Authority, Jerusalem 373 Theophylact Simocatta, Historia III, 6.11, ed. de Boor and Wirth, 124; trans. Whitby and Whitby 81. 374 Priscus of Panium, History 1, ed. an d tr ans. Blockley, 226 27.

PAGE 450

450 375 An unspecified number of prisoners had been ransomed from the Goths in 517 against 1 000 pounds of coined gold ( mille librarum auri denarios ), the equivalent of 72 000 solidi 376 Important pri soners were worth much more. The "Bulgars" managed to capture Constantine, the commander of the Roman army. According to Theophanes, the emperor paid 1,000 gold coins to ransom h is careless general, while Malalas, closer to the time of the event and perhaps better informed advanced the sum of 10,000 solidi. 377 Many ransoms of less conspicuous captives remained unrecorded and brought additional gold to barbaricum. 378 That much can be inferred from the large number of Romans enslaved by the barbarians during their frequent inroads south of the Danube. These episodes became almost a topos in the writings of Procopius who always mentioned the combination of booty and prisoners taken by th e "Bulgars", Kutrigurs, Antes, Slavs, and Gepids who invaded the Balkans during the reign of Justinian. Numbers are often exaggerated, such as the 120,000 prisoners taken by the "Bulgars" after one of their raids. 379 Procopius mentions large numbers of priso ners, without providing figures, for the invasion of the Antes in the early 540s before they became 375 Theophanes, Chronographia a. 6092, ed. de Boor, 280. 376 Marcellinus, Chronicon a. 517, ed. T. Mommsen, trans. B. Croke (Sydney: Australian Association for B yzantine Studies, 1995), 38 39. 377 Theophanes, Chronographia a. 6031, ed. de Boor, 217; Malalas, Chronographia 21, ed. Dindorf, 438. 378 On the topic of ransomed prisoners we still rely on E. Levy, "Captivus Redemptus," Classical Philology 38, no. 3 (1943): 159 76. 379 Procopius, Bella VI, 4.6, ed. Haury and Wirth, vol. 2, 165.

PAGE 451

451 allies of the Empire. 380 Several Slavic invasions around 550 led to additional displacements of population north of the Danube. The Slavs took booty and ensla ved "the young and the old alike" and later sacked Toperos on the Black Sea killing 15,000 men and enslaving tens of thousands of women and children. 381 In the embassy sent by the Utrigurs to Justinian, thousands of Roman prisoners taken by the Kutrigurs are mentioned as an argument to convince the Emperor to give them the subsidies due to the Kutrigurs who by then were "wearing gold and had no lack of fine clothes embroidered and overlaid with gold." 382 fruit when ma ny Roman captives, tens of thousand taken by the Kutrigurs during their incursions, managed to take advantage of the war with the Utrigurs and returned home unharmed. 383 Procopius is not the only author to mention the multitude of Romans taken into captivit y by barbarians. John of Ephesus informs us that the Slavs ravaged the Balkans for four years (581 584) took captives and got rich with the gold and silver plundered from the Romans. 384 After the successful retaliation staged by Tiberius II in cooperation wi th the Avars, tens of thousands of Roman captives north of the Danube were set free by the Khagan. 385 380 Procopius, Bella VII, 14.11, ed Haury and Wirth, vol. 2, 355. 381 Procopius, Bella VII, 29.1 and VII, 38.18 23, ed. Haury and Wirth, vol. 2, 4 23; trans. Dewing, vol. 4, 399. 382 Procopius, Bella VIII, 19.17, ed. Haury and Wirth, vol. 2, 588 89; trans. Dewi ng, vol. 5, 251. 383 Procopius, Bella VIII, 19.2, ed Haury and Wirth, vol. 2, 584. 384 J ohn of Ephesus, Historia Ecclesiastica VI, 25, ed. Brooks, 249 ( Et divites facti sunt et aurum et ) 385 Menander, Historia fr. 25.1, ed. Blockley, 219.

PAGE 452

452 Strategikon who tells us that prisoners were free to return home for a small ransom ( mist h os ). 386 Some of these prisoners who remained in "Sklavinia" of their own accord might be the "refugees" against whom the author of the Strategikon warns as they "have given in to the times, forget their own people, and prefer to gain the good will of the enemy." 387 In the 680s we learn about a group of descendants of the Roman prisoners taken to the Avars who participated as warriors in Avar campaigns They wanted to return to their homeland s outh of the Danube but they were perceived as half Romans (Sermesianoi Bulgars) because they had long mixed with barbarians. 388 Ransom paid for POWs and enslaved Romans taken to barbaricum may indeed account for some of the gold found beyond the frontier. A lot of wealth was concentrated north of the Danube if we are to believe sixth century writers. Aside from these isolated cases patched together to create the picture of a wealthy society we can also rely on of Baian, the khagan of the Avars, as Baian was extremely upset that the Slavs had dared to refuse his suzerainty and he was anxious to put them in their place while h oping that he would find the land full of 386 Strategikon XI, 4.4, ed. G. T. Dennis (Vienna: Verlag der sterreichischen Akademie der Wissenschaften, 1981), 372; trans. G. T. Dennis (Philadelphia: University of Pennsylvania Press, 1984), 120, translates "to r eturn to their own homes with a small recompense," but paying a ransom rather than receiving recompense is more likely. See also F. Curta, "Invasion or Inflation? Sixth to Seventh Century Byzantine Coin Hoards in Eastern and Southeastern Europe," Annali de ll'Istituto Italiano di Numismatica 43 (1996): 107. 387 Strategikon XI, 4.31, ed. Dennis, 380. For this account, see A. Madgearu, "About Maurikios, Strategikon, XI.4.31," Revue des tudes sud est europennes 35, no. 1 2 (1997): 119 21. 388 Miracula Sancti De metrii 284 287, ed. and trans. P. Lemerle, vol. 1 (Paris: ditions du Centre National de la Recherche Scientifique, 1979), 227 28. For the episode, see also S. Brezeanu, Romains et barbares dans les Balkans au VII e sicle la lumire des Miracles de Saint Dmetrius. Comment on autre," Revue des tudes sud est europennes 24, no. 3 (1986): 130 31.

PAGE 453

453 gold, since the Roman Empire had long been plundered by the Slavs whose own land had never been raided by any other people at all." 389 These descriptions are in sharp contrast with the archaeological evidence. Finds of silver and gold objects are extremely rare north of the Danube, while gold coins themselves are not very frequent. There is nothing to match the rich burial assemblages of the Avars. At least two episodes suggest that Slavs did handle Byzantine gold co ins. Procopius informs us that the Gepids who controlled the crossing of the Danube taxed one solidus ( stater ) per head to ferry the Slavs across the river with the booty and prisoners taken from the Balkans. 390 In addition, the "false Chilbudios" from the f amous episode told by Procopius was bought "with gold" by the Antes from the Slavs. 391 Where is the wealth taken by the Slavs from Thrace and Illyricum? On one hand we may argue that the Avars did a thorough job and plundered "Sklavinia" taking all its rich es. However, despite the fact that the Slavs entered the political orbit of the Avars and probably paid tribute they were also known to make inroads south of the Danube on their own. On the other hand, the successful campaigns led by the Byzantine army nor th of the Danube in the last decade of the sixth century led to an inversation of the roles. Now the Roman commander was the one proud to send to Constantinople booty and prisoners taken from "Sklavinia". 392 In the western Balkans the campaign against the Av ars occasioned a confrontation of the Gepids, in fact the slaughter of some 30,000 Gepids caught unawares as they were celebrating a feast. According to 389 Menander, Historia fr. 25.1, ed. and trans. Blockley, 195. 390 Procopius, Bella VIII, 25.5, ed. Haury and Wirth, vo l. 2, 624. 391 Proco pius, Bella VII, 14.19, ed Haury and Wirth, vol. 2, 356. 392 Theophylact Simocatta, Historia VI, 8.7, ed. de Boor and Wirth, 237.

PAGE 454

454 Theophylact at the end of the campaign 3000 Avars, 8000 Slavs and 6200 other barbarians were taken pris oners, among them Gepids if we believe Theophanes, who gives different figures. 393 Moreover, in the eve of the mutiny which led to the deposition of Maurice in 602, the soldiers vociferated against the idea to camp north of the Danube because they were anxio us to bring home the booty taken from the Slavs. It is hard to determine whether such events ruined the Slavs so as to explain an archaeological record lacking any luster. It may well be that the Slavic social structure did not rely so heavily on the posse ssion of gold which they gave away more easily than the Avars. It seems that the Avars kept the gold in monetary form mostly to be included in burial assemblages, much like the natives of Madagascar observed by the Marquis de Mondevergue in 1868: "If they possessed any coins, they would bury them with their dead." 394 Although the picture might be distorted by the fact that Avar archaeology has been almost exclusively devoted to the study of Avar age cemeteries, the abundance of Byzantine coins and their imita tions in graves cannot be ignored. Although many were used as oboli almost 20 percent of the gold coins were pierced or otherwise modified to be worn as pendants, perhaps reflecting a double role as jewelry and obolus. Interestingly none of the copper coi ns found in Avar graves were pierced suggesting that only gold coins served this function in "Avaria." The same can be said about burial assemblages from the Caucasus, where genuine Byzantine solidi mounted to be work as pendants are surpassed in number by imitations and bracteates made of thin gold plated foil, which clearly had a ritual function only. Interestingly, both in the Carpathian 393 Theophylact, Historia VIII, 3.15, ed. de Boor and Wirth, 289; Theophanes, Chronographia a. 6093, ed. de Boor, 282: 3000 Avars, 800 Slavs, 200 Gepids, and 2000 other barbarians. 394 Chauvicourt and Chauvicourt, Numismatique malgache les enterrent avec leurs morts").

PAGE 455

455 basin and in the Caucasus this practice dates to the seventh century, at a time when the were constantly shrinking. Gold coinage might not have been the only currency brought north of the Danube by "barbarians." Moreover, copper coins may have been among the personal possessions of the captives taken from the Balkans. The fact that many copper coins are stray finds along major river valleys suggests a lot of movement. Displaced Romans dropping low value coins from their pockets are a good alternative to the multitude of traders envisaged in previous scholarship, for which we lack any informatio n in the politics of recruitment. We have already seen that Heruli, Gepids, and Cutrigurs were allies of the Empire after Justinian gave them the fortress of Turris in 545 and acted as foederates defending the frontier on the Lower Danube and the Delta. 395 The non Roman pottery found in frontier fortresses from Scythia may belong to Antes serving in R oman garrisons. Slavs also served in the Byzantine army as far as Italy and some attained a high rank in the Balkans, like Chilbudios whose campaigns north of the Danube in the late 520s and the early 530s were extremely successful and possibly Tatimer who served under Maurice. 396 It is very likely that the garrisons of the Danube frontiers were ethnically mixed, as suggested by the archaeological evidence. Their identity as "Romans" was defined by their function in the Byzantine defense system and 395 Procopius, Bella VII, 14.32, ed. Haury and Wirth, vol. 2, 359. For the location of Turris, see A. Madgearu, The Placement of the Fortress Turris," Balkan Studies 33, no. 2 (1992): 203 08. 396 H. Ditten, "Slawen im byzantinischen Heer von Justinian I. bis Justinian II," in Studien zum 7. Jahrhundert in Byzanz. Probleme der Herausbildung des Feudalismus eds H. Kpstein and F. Winkelmann (Berlin: Akademie Verlag, 1976), 59 72.

PAGE 456

456 of course by the regular payment received for their services. 397 It cannot be doubted that many of the bronze coins handled in the hilltop sites placed along the Lower Danube belonged to "barbarians" settled on the frontier, who also contributed to the distribution of such coins in their homeland. 398 What was the function of Byzantine copper coins in barbaricum ? The theory that they primarily reflect trade between the Empire and communities beyond the frontier or even transactions between communities in barbaricum is unt enable. To be sure, there is no reason to believe that trade was nonexistent. Despite the fact that written sources have nothing to say on this topic, the Strategikon in particular gives us precious clues as to what may have interested Roman military offic ials or even Byzantine traders who hoped to make a profit by supplying the army with provisions. The main preoccupation was to insure sources of supply while campaigning north or even south of the Danube. The author of the Strategikon warns against the des truction of agricultural fields and argues for the protection of the peasants who can supply the army with food. 399 Although this advice probably referred to the Danubian provinces, the priorities remained the same when campaigning in barbaricum. Even more s o, the author emphasizes the care which should be given to the food provisions secured from raids of plunder in enemy territory. 400 Moreover, it was strongly advised to gather all the 397 On Roman identity in the sixth century, see G. Greatrex, "Roman Identity in the Sixth Century," in Ethnicity and Culture in Late Antiquity ed. S. Mitchell and G. Greatrex (London: Duckworth and the Classical Press of Wales 2000), 267 92. 398 Almost a century has passed since Ernst Stein first drew attention to the demographic potential of tribes north of the Danube, for which see E. Stein, Stu dien zur Geschichte des byzantinischen Reiches, vornehmlich unter den Kaisern Justinus II u. Tiberius Constantinus (Stuttgart : J. B. Metzler, 1919), 119 20. 399 Strategikon I, 9.9, ed. Dennis, 104. 400 Strategikon IX, 3.14, ed. Dennis, 314.

PAGE 457

457 available food and bring it back to the Empire on pack animals and boats a long the rivers flowing into the Danube. 401 administrative reform and the creation of the qu a estura exercitus did not solve. On the other hand, the Romans were not alone in seeking t he agricultural products of the Danubian plain. The Avars were equally interested in insuring adequate supplies for the army. According to Michael the Syrian the khagan was ready to make a good offer to the people from "two Roman towns and other fortresses urging them to "sow and harvest, we shall claim only half of your due," referring to the tax paid to the Byzantine state. 402 The Avars were clearly interested in receiving tribute in kind, as suggested by the conditions imposed by the Avars on the Longobar ds in return for their help against the Gepids in the late 560s: "that the Avars received immediately one tenth of all the livestock that the Lombards possessed." 403 The Avars also wanted control over the lands of the Gepids, Transylvania with its salt resou rces being especially attractive to pastoralists. 404 It has been suggested that Byzantine merchants paid with copper coin for the agricultural products acquired from Barbaricum while "barbarians" used the same coins 401 Strategikon XI 4.32, ed. Dennis, 380. 402 Michael the Syrian, Chronicon X, 21, ed. J. B. Chabot, vol. 2 (Paris: Ernest Leroux, 1901), 361. For the Avar subsistence economy, see W. Pohl, "Herrschaft und Subsistenz. Zum Wandel der byzantinischen Randkulturen an der Donau vom 6. 8. Jahrhundert," in Awarenforschungen ed. Falko Daim, vol. 1 (Vienna: Institut fr Ur und Frhgeschichte der Universitt Wien, 1992), 13 24. 403 Menander, Historia fr. 12.2, e d. and trans. Blockley, 131 31. 404 K. Horedt, "The Gepidae, the Avars, and the Romanic Population in Transylvania," in Relations Between the Autochthonous Population 111 22.

PAGE 458

458 to pay for manufactured products from the Empire. 405 This view is widely shared but it suffers from preconceived ideas about the the nature of exchange and the role of money. Ethnographic parallels have taught us that "primitives" were only inclined to accept coin if it was essentially commodity mo ney, which the Byzantine follis, with its fluctuating weight and exchange rate to the solidus was certainly not. There is every reason to believe that barter alone would have sufficed to facilitate exchange north of the Danube. Anthropological studies indi cate that barter remained the most important form of exchange in primitive societies, even during the colonial period, and the nature of the sixth century society north of the Danube as reflected in the archaeological record suggests that a market economy did not exist. 406 Alternative media of exchange may have existed but none was preserved and it is possible that peoples in barbaricum used perishable materials for this purpose, if they used any monetary instruments at all. To be sure, Byzantine merchants ma y have used coins to pay for various products bought from the region close to the Danube, but much like the Merina from Madagascar, communities in barbaricum treated coins like any other Byzantine commodity. As in so many primitive societies coins were kep t and melted down never to return to Byzantium. Much like the French authorities who complained that the silver 5 Franc pieces "disappeared" in Madagascar, the Byzantine government probably discouraged the private export of coin, especially when inflation struck and the supply of metal needed to issue fresh coins became limited. 405 Teodor, Romanitatea nord 28. 406 Market systems themselves can exist even without the use of money. See recently, B. L. Stark and C. P. Garraty, "Detecting Marketplace Exchange in Archaeology: A Methodological Review," in Archaeological Approaches to Mark et Exchange in Ancient Societies ed. B. L. Stark and C. P. Garraty (Boulder, CO: University Press of Colorado, 2010), 35.

PAGE 459

459 Copper coins were not a good investment for communities in barbaricum unless they were located very close to the Danube and could be in permanent contact with the Byzantine fortress es on the other bank of the river. The only conceivable monetary transactions probably took place in Oltenia where the presence of a few Byzantine bridge heads on the left bank made the dangerous crossing of the Danube unnecessary. Nonetheless, bronze coin s have been found over a much wider area, sometimes very far from the Danube. Many were probably brought home by the soldiers who served in the Roman army or in the garrisons defending the Danube frontier. Their function is not hard to guess. "Big men" or simply village chiefs, some of whom may have served in the Roman army or led plunder expeditions in the Empire, could only retain their prestige if the way in which they manipulated wealth served the needs of the community. 407 In societies where prestige dep ended largely on access to Roman goods, coins were yet another type of import, to be worn as jewelry, to be given away as gifts or to be used in ceremonies and rituals. What is the evidence for the use of coins as raw material for the production of jewelry ? We have already seen that gold clearly served this purpose, not only in Barbaricum, but also in the Empire, where coins could be mounted or melted down. 408 Pierced coins, gold or copper, are also known from the early Byzantine provinces proving that the di stinction between money and ornament is not as clear cut as some 407 Curta, The Making of the Slavs 325 32. 408 See especially A. E. Jones, ""Lord, Protect the Wearer:" Late Antique Numismatic Jewelry and the Image of the Emperor as Talismanic Device," PhD Dissertation, Yale University, 2011. See also, A. Oddy, "La monnaie d'or dans la bijouterie travers les ages," Aurum 15 (1983): 10 16; J. A. Bruhn, Coin and Costume in Late Antiquity (Washington, D .C.: Dumbarton Oaks, Research Library and Collection, 1993).

PAGE 460

460 may think. 409 Unsurprisingly, such coins are also found in barbaricum and their number is probably higher for the copper coins, which are most often published without illustration and with insu fficient details regarding their state of preservation. On the other hand we have already seen that metalwork was a widespread activity north of the Danube as testified by the large number of tools (crucibles and metallurgical ladles) and stone molds used for the production of copper jewelry. There is some archaeological evidence suggesting that coins were a constant and reliable source of copper. Very few coins have been found in a clear archaeological context. The old finds from the sixth century settleme nts excavated in Bucharest have not been properly published and their association with other objects is unknown. 410 The same can be said about the folles found in the early medieval settlements from Alcedar Odaia and Lopatna in Moldova where they were found together with Luka Raikovetskaia ceramics, but no stratigraphical information is available. 411 We are best informed about the two early folles Moldavia, very far from the Danube. The sunken featured building no. 20 produced a follis dated 527 537 in association with a crucible and a ladle used to pour the metal. 409 C. Morrisson, "Monnaies et pseudo monnaies byzantines motifs chrtiens: cr oyance ou magie?," forthcoming. 410 D. V. Rosetti, "Siedlungen der Kaiserzeit und der Vlkerwanderungszeit bei Bukarest ," Germania 18 (Bucharest: Editura 1959), 33 34 ; M. CAB 2 (1965): 189, fig. 93/1 3. 411 I. A. Rafalovich, Slaviane VI IX vekov v Moldavii (Kishinew: Shtiinca, 1972), 40, fig. 9/2; Kropotkin, Klady vizantiiskikh monet 38, no. 293, fig. 17/26; A. Rikman and I. A. Rafal ovich, "K voprosu o Kratkie Soobshcheniia Instituta Arkheologii 105 (1965): 49, fig. 3/4.

PAGE 461

461 The coin was not cut or damaged but this was not a bsolutely necessary. 412 Recent research at Nicopolis in the northern Balkans has revealed unusually large accumulations of Late Roman copper coins together with scrap metal collected for recycling and although none of the coins, which are smaller than the Ea rly Byzantine pieces, were folded, cut or damaged, the association with metalwork is quite evident. 413 No damage can be seen on coins from the hoard found at Horge coins were hidden in a copper pitcher ( ca 1.3 kg) together with a copper chain ( ca 250g) and small scrap pieces of copper. 414 It is quite obvious that we are dealing with an accumulation of metal rather than a collection of coins. Olde r coins were also collected to be recycled, such as the twelve Roman imperial bronze coins from Marcus Aurelius to Philip I found in the sixth or the bronze coin of Nerva found together with scrap pieces of bronze in the settlement 415 The mixed composition of copper alloys of various 412 Cut or damaged Byzantine bronze coins have been found at Osijek (Croa tia) and Pavlivka (Ukraine) but no archaeological context is available. A hexagram of Heraclius cut in half was found in a grave from Linz and a cut fragment of a solidus struck for the same emperor was found in an Avar grave from Bkscsaba Repltr (Hun gary). See Gricke 28 (Osijek); P. A. Karyshkovski, Nahodki pozdnerimskikh i vizantiiskikh monet v Odesskoi oblasti," 7 (1971): 81, no. 10 (Pavlivk a); Hahn, "Die Fundmnzen des 5. 9. Jahrhunderts," 459 (Linz); Somogyi, Byzantinische Fundmnzen 28 29 no. 8 (Bkscsaba Repltr). 413 A. Poulter, "Interpreting Finds in Context: Nicopolis and Dichin Revisited," in Objects in Context, Objects in Use: Mat erial Spatiality in Late Antiquity ed. L. Lavan, E. Swift, and T. Putzeys (Leiden/Boston: Brill, 2 007), 694 98. 414 V. " Carpica 4 (1971): 253 269; G. Buzdugan, Carpica 6 (1974): 47 ri privind vasul de metal din tezaurul monetar ed. S. 127. 415 P. Roman and S. Dolinescu materiale autohtone din sec. al VI lea n Muntenia)," SCIVA 29, no. 1 (1978): 73 93. Some doubts have been raised regarding the dating of these complexes to the sixth century, for which see Curta, The Making of the Slavs 231 32.

PAGE 462

462 Byzantine mints and denominations makes them unsuited for analyses of metal composition and comparison with copper alloy artifacts. 416 However, Krzyszt of barbaricum and concluded that they were produced by melting down early Roman coins. 417 Let me pull the strands of my argument together. Whether used as souvenirs, apotropaic amulets, jewelry, objec ts of prestige or simply raw material for the production of copper alloy jewelry, Early Byzantine coins rarely returned to the Empire. If any transactions required the mediation of coins they were undertaken in Byzantine fortresses on the Lower Danube wher e peasants could come to sell their surplus. If Byzantine merchants venturing into the dangerous world of "barbarians" paid with coin for the products they acquired it was a unidirectional monetary transaction. The communities living in the shadow of the E mpire bought coins like any Byzantine commodity because they needed the metal or simply because they were looking to use them as jewelry to adorn their clothes or to hang on necklaces. Other non monetary and non economic means by which Byzantine coins trav eled across the frontier were related to the movement of people, "barbarian" soldiers returning to their village and hoping to achieve a higher status by bringing Roman exotica and souvenirs or Roman captives taken by barbarians together with their possess ions. 416 For coins, see T. Padfield, "Analysis of Byzantine Copper Coins by X Ray Methods (With a Numismatic Commentary by P. Grierson)," in Methods of Chemical and Metallurgical Investigation of Ancient Coinage, ed. T. Hall and D. M. Metca lf (London: Royal Numismatic Society, 1972), 219 36. For objects, see H. K. Cooper, "Analysis of Late Roman Byzantine Copper Alloy Art ifacts from Northern Jordan," MA thesis, University of Arkansas 2000. 417 "Nouvelles donnes concernant l'orfvrerie sur le territoire de la vovodie d'Olsztyn (Pologne)," Archaeologia Polona 19 (1980): 238 39.

PAGE 463

463 Fi gure 6 19. First to third century mummy portraits wearing coin necklaces and coin pendants (after Parlasca 1969, pl. 19/78, 34/142 and 56/226; Parlasca 1977, pl. 98/397).

PAGE 464

464 Figure 6 20. Late nineteenth century Algerian woman and her daughter we aring coin necklaces (1) and late twentieth century Andeans dugging up the community treasure made of old foreign coins and banknotes (2).

PAGE 465

465 Figure 6 21. Pierced Spanish Reales from the New World used in Madagascar (after C hauvicourt and Chauvicourt 1968 )

PAGE 466

466 Figure 6 22. Bracelet and necklace from Madagascar made of early modern European coins (after Chauvicourt and Chauvicourt 1968, 9)

PAGE 467

467 Figure 6 23. Byzantine solidi of Heraclius mounted into rings found in Austria at Emling and in Germany at Hausen and Gromehring (1 3) and necklace with a mounted solidus from the joint reign of Justin I and Justinian I from Dzhiginka, Russia (4) (after Wamser and Gebhard 2001, no. 183a; Steffgen et alii 1993, pl. 17; Somogyi 2011, 111, fig. 7; Kropotkin 1962, pl. 1 4)

PAGE 468

468 Figure 6 24. Head jewelry from Madagascar made of early modern European coins (1), dress jewelry made of seventh century Byzantine solidi from the Malo Pereschepyne hoard (2) and cut folles of Justinian from Haskovo (Bulgaria) (3) (after Chauvico urt and Chauvicourt 1968, 9; Kropotkin 1962, pl. 16; Tenchova 2011, pl. 47/845 and pl. 69/1135)

PAGE 469

469 Figure 6 25. Early Byzantine gold (1 6), silver (11), and bronze (7 10, 12 13) coins from Hungary, Serbia, Romania, and Georgia, pierced, cut and mounte d to be worn as pendants (after Somogyi 1997, no. 3, 8, 17, 52 and 72; Abramishvili 1965, pl. V/74, pl. VI/81, pl. VIII/111 and pl. XI/175; Poenaru et alii, no. 753; 46, fig. 1/1 )

PAGE 470

470 CHAPTER 7 CONCLUSIONS Two interpretations have shaped our understanding of the role played by the Danube frontier in Late Antiquity. One relied faithfully on the evidence of contemporary writers who epitomized the river as the ideal boundary between civilized Romans and "barbarians," while the other clu ng on to the legacy of the former Transdanubian province of Dacia to claim that the Empire never completely lost its political influence north of the river during Late Antiquity. Both are corre ct, albeit not always for the reasons mentioned by their proponents. The external policy of an empire is rarely straightforward and unequivocal and Early Byzantium makes no exception. On one hand, Roman emperors were interested in securing the Balkans by c losely guarding the river, on the other communication and contact with barbaricum was part and parcel of the frontier policy which included exchange, acculturation, and the recruitment of barbarians in the Roman army. There was no single and uniform fron tier policy on the Danube but a larger repertoire of solutions depending on the balance of power in barbaricum Undoubtedly, the Danube met the necessary preconditions to become a defensible frontier and was t he most strategically advantageous line of separation in the ancient Balkans. This fact was recognized not only by the Roman Empire but also by its bellicose neighbors, whether Dacians, Goths, Huns or Avars. Roman emperors took the necessary precautions an d complemented the natural virtues of the river with a strong fleet and a chain of fortifications running along its banks, but the functioning of this internal stability. Ma naging the cascade of crises between the fifth and the seventh

PAGE 471

471 century often entailed the serious neglect of one frontier in favor of another and as a words, it was not the systemic inefficiency of the Danube as a frontier that led to its ultimate failure but the way in which it was consciously managed by Late Roman emperors who had limited resources at their disposal. When speaking about the Danube frontier we must dist inguish between two different layers. The upper is high politics, the strategic priorities of the Empire which included the defense of Constantinople and the prosperity of Thrace and Illyricum, as well as political dealings with barbarian groups living n orth of the Danube. The lower layer is the individual experience of people living on the frontier and whose actions, which might be labeled as private initiative, gave shape to cultural change in the frontier region and beyond. The former is shrouded in the rhetoric of contemporary writers while the latter can only be discerned by turning to the archaeological evidence. Often the archaeological record seems to be at variance with the written sources. The difference between the Romans who defend the fron tier behind the thick walls of fortresses placed on the Danube and the barbarians bent on destroying the civilized world is purely situational. Events from the long sixth century have taught us that barbarians are often the Romans of tomorrow who may well begin by plundering the provinces of the Balkans only to end up defending them later as foederati or symmachoi. The northern Balkans in Late Antiquity was clearly a multiethnic environment. Traffic and communication between the two sides of t he river occurred naturally and the archaeological evidence shows that it was not prevented so long it was done peacefully and without upsetting the upper layer, that of high politics.

PAGE 472

472 The Byzantine artifacts most frequently found in barbaricum are precise ly those related to the militarized environment of the northern Balkans: fragments of LR1 and LR2 amphorae, the containers in which soldiers received their allowances of oil and wine, a variety of jewelry and dress accessories fashionable in the frontier f ortresses, and Byzantine coins. Cultural influences were not unidirectional and were not spread uniformly in the frontier region and beyond. The stylistic evolution and geographic distribution of the so called Slavic bow fibulae best illustrates cultural influences coming from outside the Empire. In fact, there are very few artifacts produced in the northern Balkans which can be described as reflecting just one cultural identity. Fashions developing on the sixth century Danube frontier reflect the multiet hnic background of their carriers and combine several cultural traditions relying on the Roman technology of producing jewelry and combining imagery derived from the cultures of the steppe and from Germanic styles. Analyzing the distribution of Byzantin e artifacts found in barbaricum is helpful in coming to grips with the widely different social and cultural complexities they encompass. Communities living outside the Empire emulated the Roman way of life by seeking access to the objects and goods most of ten encountered in the frontier region. Expressing ethnic and social identity in barbaricum was a process whose mechanics depended not only on relations with Byzantium but also on cultural contact and relations of power between groups within barbaricum. Di stribution maps clearly reveal the fact that different Byzantine artifacts were preferred, although most of them reflect the annona militaris concentrate in Oltenia and Moldavia pointing t o important regions for the recruitment of

PAGE 473

473 soldiers. Oil lamps including bronze liturgical lamps are found almost exclusively in the territory of the former province of Dacia, while pectoral crosses are encountered predominantly south and east of the Carpa thians, in Wallachia and Moldavia. Metallurgical activity and the production of bronze jewelry and dress accessories of Byzantine inspiration was most intense in the Subcarpathian hills of Wallachia and Moldavia, which are also important regions for the di stribution of Byzantine buckles and brooches. All these convey an image of communities living in the shadow of the Empire but adopting only certain cultural practices. Numismatic evidence, or the lack thereof, has been seldom used to understand the complex nature of early Byzantine frontiers. Most often than not, coins played their designated role in the scenario emphasizing separation, in which coins were seen as the result of plunder, or in the one illustrating contact, with coins being the instrument of a local monetary economy. The first step, however, has been to establish a connection with the monetary circulation in the northern Balkans especially since the influx of newly minted currency into the circulating medium of the frontier provinces bears upo n the age structure of finds from barbaricum. The monetary economy of Early Byzantium knew different levels of development depending on the type of settlement, which can be a large coastal town, a landlocked administrative center, a frontier fortress, or a small village. The Empire wide analysis of coin finds has revealed the fact that outside the confines of large urban markets monetary transactions were just one of the ways in which exchange took place. In addition, hill top military sites from the Balkan s display a transactions which involved the state as the most important agent.

PAGE 474

474 What exactly communities outside the Empire had to offer in exchange for Byzantine coins remains a matter for conjecture. In any case, trade was just one, certainly not the dominant, manner in which Byzantine coins arrived in barbaricum. In Transcaucasia, the Carpathian Basin, and the Lower Danube region gold and ceremonial silver issues are often the reflection of political and diplomatic payments made by Byzantine emperors to buy peace or to gain an ally in strategically sensitive regions. Many of these coins, redistributed to the local elite, found their way into burials often after having been turn ed into jewelry. Byzantine gold and silver coins were not used but worn and played a role in the display of social status. T he Avars in particular chose to melt down Byzantine gold coins in order to produce spectacular jewelry which would play a similarly important social function. The analysis of Byzantine coins from barbaricum in their historical and archaeological context has revealed the fact that we are not dealing with a "grand strategy" of the Byzantine Empire on the frontier. In fact we are dealing with many frontiers and many types of responses. Adaptation rather than rigid planning characterizes imperial action at the periphery. The distribution of early Byzantine coins from Central Europe to the Caspian Sea, in a wide geographic area on the north ern and eastern frontier of the Empire, reinforces this conclusion. At least three major patterns of distribution can be distinguished: Transcaucasia, dominated by regular and ceremonial silver coinage, the Lower Danube region where copper coin finds are o verwhelming, and the Carpathian Basin where gold issues predominate. The explanation lies in the particular nature of each of these frontier regions, the strategic priorities of the Empire, and the strengths of its enemies in barbaricum. Lazica

PAGE 475

475 and Iberia guarded the strategic passes in the Caucasus Mountains through which poured not only spices and exotica from the East but also nomadic raiders. The danger was deemed serious enough to require the joint action of the two great empires, the Byzantine and th e Persian, in one of the rare instances of collaboration. Failure to control movement in the Caucasus exposed the Byzantine heartland and the Black Sea coast. Procopius was perhaps exaggerating when suggesting that the loss of Lazica could lead to a major invasion by sea towards Constantinople, but the danger itself was very real. Sixth to seventh century Byzantine emperors took every precaution to secure this region. Gifts of silver and gold coins were sent to Lazi kings and to smaller Caucasian tribes in order to gain their loyalty and huge quantities of silver hexagrams were spent by emperors of the Heraclian dynasty to keep Transcaucasia in the Byzantine orbit. Very little of this policy can be traced in the Lower Danube region. Unlike Transcaucasia wher e the frontier was quite volatile and depended primarily on people, on the Danube the frontier was quite material and relied on the heavily fortified southern bank of the river. Transcaucasia had many tribes but no coherent defense system, while the Lower Danube had many fortifications but few Romans to defend it, at least if we believe authors such as Agathias who lamented the state of the garrisons around 559. The high density of hill top sites in the Danube valley indeed required many defenders, most of whom were recruited from barbaricum People, rather than objects or goods were the greatest asset of the regions north of the river. The energy and resources invested in Transcaucasia insured that eastern Anatolia would not be invaded and depopulated by "b arbarians." In the Balkans, however, such events were commonplace

PAGE 476

476 in the sixth century. The Danube was frequently crossed back and forth usually by small bands of raiders but sometimes by larger invading parties. Booty and thousands of enslaved Romans were taken north of the Danube. This continuous movement of people explains the large quantity of bronze coins available in the northern Balkans and by extension in barbaricum. Such coins did not have any monetary function outside the Empire except for isolate d transactions in the region close to the Danube. Much like the gold coins, they were used as raw material for the production of bronze jewelry as testified by single finds, hoards, and numerous stone molds for small pieces of jewelry. Others were kept as souvenirs or amulets pierced to be worn as pendants. The Carpathian Basin became once again a frontier of the Roman Empire after solidi were sent to the Gepids, the Longo bards, and the Avars to name only the most powerful confederations. These allies and adversaries at the same time were of a different caliber than the Slavs, the Antes, and the Kutrigurs in the Danube and the western Black Sea region. For all intents and p urposes the Middle Danube, the Drava, and the Sava rivers flowing from the Alps were lost to the Empire and no system of fortifications can be found there to help distribute Byzantine copper coins in barbaricum. Such coins are rare in the Carpathian Basin and in Central Europe and are mostly brought from Italy where prospects of enrichment through warfare attracted ambitious men from barbaricum Warfare and service in the Roman army are the common denominator of all three regions. Different circumstances, however, required different responses and Byzantine emperors of the sixth and seventh century had limited resources at their disposal to

PAGE 477

477 manage frontiers stretching on three continents. Adaptation and, indeed, at times improvisation and quick reaction time were crucial survival skills at the periphery. Coins are an excellent testimony of such developments and together with other categories of artifacts can open a window to the world of barbaricum where individual agency and culturally constructed responses were often shaped in relation to Byzantium. W ritten accounts as well as the available archaeological evidence point to a complex society developing in the frontier region, drawing its cultural identity from multiple sources and blending Roman and barbari an influences. Communities living in the shadow of the Empire forged their own identity in relation with Byzantium but targeted only certain Byzantine practices. Identities were constantly renegotiated at the periphery in the ever changing world of confli ct and cooperation between Romans and barbarians, whose very definition came to be blurred as the Empire grew weaker.

PAGE 478

478 APPENDIX A CORPUS OF EARLY BYZANTINE COIN FINDS IN BARBARICUM A.1 Context Creating an inventory of early Byzantine coin finds in bar baricum is far from being a straightforward task. A mere compilation from previous publications, while a worthwhile endeavor in other contexts, would only add to the confusion and misrepresentation of early Byzantine coin finds so pervasive in older litera ture. Undoubtedly, the proper course of action is to undertake the task of creating an updated inventory by critically assessing the merits of previous publications. Most of the time this involves verifying the attribution of each single find and hoard aga inst the illustration by using the most recent reference catalogues. Most often than not such an exercise reveals important errors, which accumulated would distort not only the statistical analysis of coin finds in Barbaricum, but also the overall historic al interpretation of the numismatic material. Collecting information about the circumstances of each find is yet another difficult and often frustrating task. Being one of the most popular collectibles since the antiquarian age, precious pieces of informat ion such as the archaeological context or even the finding place are now lost or uncertain. In most cases, the presence of unprovenienced coins in smaller local museums is usually a good indicator that the coins were found in that region. Larger collection s housed in national museums can be more problematic. Many acquisitions were made from important collectors who built their collections while traveling in various parts of the former Byzantine Empire or by relying on auction houses, which rarely disclose t heir sources. The ideal case of a reliable archaeological context is recorded less frequently than one would expect, but

PAGE 479

479 even a general geographic placement of the finds can still provide an overview of of cultural contact in the frontier region and beyond especially if used in conjunction with other archaeological sources. The inventory covers a wide geographic area, from the Ore Mountains (separating Saxony from Bohemia) to the west, to the Caucasus Mountains to the east, and is meant both as a tool for future research and as a means of contextualizing the finds from the Lower Danube area. By necessity, the decision to include certain regions, while leaving out others, cannot be fully substantiated by historical processes taking place fifteen centuries ag o. It often has to do with the state of research in each country and the material available for study, either published catalogues or st ill unpublished finds in East European museums. From a historical perspective, the decision to cover this particular are a takes into account the major transformations of the sixth and seventh centuries, namely the rise of the Avar power in the Carpathian Basin, with its constel l ation of clients in the Lower Danube region and the steppe north of the Black Sea, and later the rise of the early Bulgar state in the Balkans. Tracing such developments from a historical and archaeological perspective requires a broad understanding of the political, military, and economic relations established between Byzantium and the outside powers through the lens of numismatic evidence, hoards and single finds. On the other hand, the complex nature of the early Byzantine frontier can be clearly seen in the inventory of coin finds which also includes sites that may well have been under the full con trol of the early Byzantine army and administration at certain times during the sixth and seventh centuries. Such cases include the strategic

PAGE 480

480 Byzantine bridge heads located on the Lower Danube's left bank, such as Drobeta and Sucidava, as well as important settlements on the eastern Black Sea coast, such as Pitiunt. In all cases, the large number of early Byzantine coins found on site strongly suggests an enduring imperial presence. While the Danube provided a convenient separation line in the northern Balk ans, the political status quo is less clear in the western part of the peninsula, Dalmatia in particular. Justinian's Gothic wars restored the imperial control over the eastern Adriatic coast, but it is difficult to approximate how far it extended inland. Unlike the case of the Lower Danube, drawing a line or identifying a natural barrier, for instance the Sava river, is far less convincing. To be sure, there is ample evidence for the presence of Byzantine goods in Bosnia & Herzegovina, Croatia, and Sloveni a, even far from the Adriatic coast, but for the purposes of this inventory all the finds from the "grey area" of the frontier have been included to provide a more nuanced picture of the frontier region and its cultural complexity. Indeed, the level of pub lication differs widely from one region to another. Fortunately, a recent volume has brought together papers dealing with the presence of Byzantine coins in Central Europe, which offer an updated picture of the Byzantine influence in barbaricum. 1 In additi on, important catalogues have been published in the past decades bringing to light an important number of Byzantine coin finds from the Czech Republic, Slovakia, Poland, Slovenia, Austria, and Hungary. 2 The most time 1 Byzantine Coins in Central Europe Between the 5th and 10th Century (Cracow: Polish A cademy of Sciences, 2009). 2 Czech Republic: J. Militk, Finds of Greek Roman and Early Byzantine Coins in the Territory of the Czech Republic. I. Bohemia (Wereren: Moneta, 2010); Slovakia: A. Fiala, "Byzantsk mince na Slovensku (6. Slove nsk Numizmatika 10 (1989): 57 Byzantine Coins from the 6 th and 7 th c. From Poland and their East Central European Context," in Roman Coins Outside the Empire: Ways and Phases, Contexts and Functions ed. A. Bursche et al. (Wetere n: Moneta, 2008), 195 224; Slovenia: P. Kos, "The Monetary Circulation in the Southeastern Alpine Region ca 300

PAGE 481

481 honored preoccupation with early Byzant ine coin finds can be safely attributed to Romanian numismatists and historians, no doubt because of the superior number of finds from the Danube region, but also as a direct result of a historiographic al inclination toward the study of Late Roman and Earl y Byzantine cultural influence north of the Danube. This particular interest insured a steady publication of finds followed by more comprehensive catalogues and attempts to interpret the numismatic material in its historical context. 3 A large body of liter ature is indeed a mixed blessing, as the scholar trying to put together an inventory of finds has to go back to the original publication of a certain coin and then trace it in subsequent works in order to determine its full attribution. This detective work often reveals a trail of errors, sometimes caused by the state of the discipline at the time of the initial publication, other times by carelessness or the lack of specialized knowledge of the Byzantine series. Most of the time this was due to the fact th at general catalogues of recent finds of Greek, Roman, and Byzantine coins, usually published on an annual basis, included only brief descriptions and were rarely supported by illustrations. 4 B.C. A.D. 1000," Situla 24 (1984 1985): 225 238, Austria: H. Winter, "Die byzantinischen Fundmnzen aus dem sterreichischen Bereich der Avari a," in Die Awaren am Rand der byzantinischen Welt: Studien zu Diplomatie, Handel und Technologietransfer im Frhmittelalter ed. F. Daim (Innsbruck: Universittsverlag Wagner, 2000), 45 66; W. Hahn, "Die Fundmnzen des 5. 9. Jahrhunderts in sterreich und den unmittelbar angrenzenden Gebieten," in Die Geburt Mitteleuropas. Geschichte sterreichs vor seiner Entstehung 378 907 ed. H. Wolfram (Vienna: Kremayr & Scheriau, 1987), 453 464. Hungary: P. Somogyi, Byzantinische Fundmnzen Der Awarenzeit (Innsbruck: Universittsverlag Wagner, 1997); P. Somogyi, "Byzantinische Fundmnzen der Awarenzeit. Einem Bestandaufnahme, 1998 2007," AAC 42 43 (2007 2008): 231 298 3 SCIV 23 (1972): 375 415; V. VII n teritoriile carpato BSNR 131 133 (1983 1985): 199 235; E. Oberlnder Trn oveanu, "La monnaie byzantine des VI e VIII e sicles au del de la frontire du Bas Danube. Entre politique, economie et diffusion culturelle," Histoire et Mesure 17, n. 3 (2002): 155 96. 4 Most significant is the chronicle of recent finds held periodically in Dacia by Bucur Mitrea and Gheorghe Poenaru Bordea between 1958 and 1996 under the generic title "Dcouvertes montaires en Roumanie,"

PAGE 482

482 Many of the finds before the Second World War have been attribut ed by using the standard reference catalogues of the day, now largely obsolete, while in the greater in regional museum journals well into the 1970s. The researcher trying to compile a catalogue of finds will often find the same coin republished three or four times, sometimes with full technical details, unfortunately rarely illustrated, or briefly mentioned in inventories of finds form a certain region or province. In some cases, the lack of dialogue between numismatists and archaeologists, amply addressed in chapter two, can be seen at work when the archaeological context is insufficiently explored by numismatists or when coins are misrepresented in archaeological reports and illustrations (e.g. Bucharest the most recent or the most reliable publications will be cited for each find. On occasion, reference will be given to previous literature which misrepresented th e respective find or ascribed it a different finding place, lest the reader is left with the impression that we are early Byzantine coin finds from present day Romania have received special attention in the past decades, although the publications often cited previous titles without attempting a critical analysis. 5 The review of early Byzantine coin finds from the former Soviet Union is a far more challenging exercise. The f inds from southern Ukraine have been thouroughly following a tradition introduced by Constantin Moisil in BSNR in 1913 ( Romnia mare ") 5 A. M. Velter, Transilvania n secolele V XII (Bucharest: Paideia, 2002); E. Oberlnder, XIV) (Cluj Napoca: Nereamia Napocae, 2003); 235; D. G. Teodor,

PAGE 483

483 published or republished by Elena Stoliarik although in many cases it proved to be a mere compilation based on previous works of P. O. Karyshkovskii, A. A. Nudel'man, and V. V. Kropotkin. 6 To be sure, most attempts to address the question of Byzantine coin is now half a century old, although it undoubtedly was an impressive achievement at that time. Unfortunately, a mu ch awaited recent edition of his book turned out to be nothing but a French translation filled with mistakes. No attempt was made to revise and update the catalogue. 7 The state of publication in the Transcaucasian republics of Georgia and Armenia is compl etely different. Paradoxically, the large number of finds and the overall potential of the region which acted as a buffer zone disputed by the Sasanian and the Byzantine empires, did not insure a proper visibility in Western literature. Tamara Abramishvili catalogue of Byzantine coins from the National Museum of Georgia (1965) was the first major publication of a large national museum collection in Europe dedicated exclusively to the Byzantine series. The occasional attribution errors were vindicated by a thorough illustration of the coins, which allows us to perform the necessary corrections based on current reference catalogues Many coins had a clear provena nce while some were the result of archaeological excavations. Later publications of coin finds re sulted from 6 E. S. Stoliarik, Essays on Monetary Circulation in the North W estern B lack Sea Region in the Late Roman and Early Byzantine Periods: Late 3rd century Early 13th Century A.D. ( Odessa: Polis Press, 1993 Topografiia kladov i nakhodok edinichnykh monet O. Karyshkovskii, Nahodki pozdnerimskih i vizantijskih monet v Odesskoj oblasti 7 (1971): 78 86; V. V. Kropotkin, Klady vizantiiskikh monet na territorii SSSR (Moskow: Iz d vo Akademii nauk SSSR, 1962). 7 V. V. Kropotkin, Les trou vailles de monnaies byzantines en U.R.S.S. ed. G. Depeyrot (Weteren: Moneta, 2006).

PAGE 484

484 systematic excavations at Pitiunt and Nokalakevi contributed to a better understanding of Egrisi during the early Byzantine period. 8 The Georgian language, rarely accessible to Western scholars contributed to the obscurity surrounding these finds, which constitute a goldmine for the study of the Caucasus in Late Antiquity. Unfortunately, the meritorious intention of bringing the region to the attention of scholars through the publication of monographs in the Moneta series did not always yiel d the desired scientific results. 9 Most of the publications constitute abridged English translations of works produced decades ago followed by uncritical and incomplete catalogue compilations of the previous literature, mainly o the publication of Byzantine single finds and hoards from the 1950s through the late 1980s. The reader studying the inventory included in this chapter will notice many instances of confusion regarding the composition of hoards (e.g. Magraneti) and the fi nding circumstances of both hoards and single finds (e.g. Ochamchire). The similar initiative regarding the Byzantine coin finds from Armenia was slightly more succe s with ancient and medieva l coin finds in Armenia. 10 Although not free of mistakes, the series provides precious numismatic material for the study of the Byzantine influence in Armenia in the sixth and seventh centuries. 8 T. Abramishvili, Sakartvelos sakhelmts'ipo muzeumis bizant'iuri monet'ebi (Tbilisi: Metsniereba, 1965); T. Abramishvili, Sakartvelos sakhelmts'ipo muzeumis bizant'iuri m onet'ebi (1966 1984) (Tbilisi: Metsniereba, 1989). 9 I. Tsukhishvili and G. Depeyrot, History and Coin Finds in Georgia: Late R oman and Byzantine Hoards (4th 13th c.) (Wetteren: Moneta, 2003) ; M. Tsotselia, Coin Finds in Georgia: (6th Century BC 15th C entury AD) (Wetteren: Moneta, 2010). 10 Mousheghian et al., History and Coin Finds in Armenia: Coins from Duin, Capital of Armenia, 6 7th c.: Inventory of Byzantine and Sasanian Coins in Armenia, 6 7th c. (Wetteren: Moneta, 2000); Musheghian et al., History and Coin Finds in Armenia: Coins from Ani, Capital of Armenia, 4th c. BC 19th c. AD (Wetteren: Moneta, 2000).

PAGE 485

485 The following inventory is divided into sections for single fi nd and hoards and is organized alphabetically by mentioning the finding place followed in parentheses by the main administrative unit and the country. Special mention was made for coins found in ar c haeological context or in the proximity of ancient sites. In addition, a number of standard abbreviations were used for gold (AV), silver (AR), and the most common copper (AE) denominations of the early Byzantine monetary system, M (40 nummia / follis ), K (20 nummia 1/2 follis ), I (10 nummia 1/4 follis nu mmia / 1/8 follis ) and IS (16 nummia ), as well as the for the early Byzantine mints most influential in the area under discussion, CON (Constantinople), NIK (Nicomedia), KYZ (Cyzicus), TES (Thessalonica), and ANT (Antioch). A.2 Inventory 11 A.2.1 Single Find s 1. A IUD (Alba, Romania). Justinian I: M, 527 565; A. Lakatos, "Monede vestul Romniei." EN 12 (2002): 247, no. 1B. Tiberius II: M, 581 zona Aiudului, Apulum 37 (2000): 443 44. Phocas: solidus ; Lakato s, "Monede bizantine ," 247, no. 1B. 2. A KHALI A TONI (Abkhazia, Georgia). Justinian I: heavy miliarensis CON, 527 537; 33 nummia Alexandria, 527 565; S. M. Shamba, Monetnoe obrashchenie na territorii Abkhazii: (V v. do n.e. XIII v. n.e.) (Tbilisi: Metsni ereba, 1987), 78 79, no. 114 and 81, no. 123. 12 3. A KHALTSIKHE (Samtskhe Javakheti, Georgia). Justin II: K, Rome (Moneta Militaris Imitiativa), 565 572, found on the left bank of Mtkvari river, not far from Atskuri fortress, on the right bank; Abramishvili Sakartvelos sakhelmts'ipo muzeumis (1966 1984), 21, no. 73 with n. 114. 4. A K URA (Kakheti, Georgia). 11 Numbers on the maps (Figures 6 6, 6 7, 6 12, 6 16, 6 17, and A 1) refer to numbers in the inventory. Finds whose authenticity is under seriou s doubt ha ve been excluded from the maps. 12 Same coins as Kropotkin, Klady vizantiiskikh monet, 45, no. 429 (Novi Afon); Abramishvili, Sakartvelos sakhelmts'ipo muzeumis, 124, no. 39 40 (Psirtskha).

PAGE 486

486 Heraclius: hexagrammon CON, 625 629(?); Abramishvili, Sakartvelos sakhelmts'ipo muzeumis (1966 1984), 27, no. 92. 5. A LBA I ULIA (Alba, Romania). 13 Ti berius II: tremissis CON, 578 582; Apulum 15 (1977), 665 666. Phocas: solidus CON, 602 610; Ibid., 666. 6. A LCEDAR Heraclius: M, CON, 611 612, found in 1961 during archaeological excavations Topografija 81, no. 1; I. A. Rafalovich, Slaviane VI IX vekov v Moldavii (Kishinew: Shtiinca, 1972), 40, fig. 9/2. 7. A LESHKINSKYI K HUTOR (K herson, Ukraine). Justinian I: AV solidus CON, 541 542; Stoliarik, Essays on Monetary Circulation, 138, no. 45. 8. A LEXANDRIA (Teleorman, Romania). Heraclius: solidus CON, 632 636; Butnariu, " 216, no. 1. 9. A (Olt, Romania). Justin I: M, CON, 518 527; Butnariu, " 216, no. 2. 10. A LKOVEN E MLING (Obersterreich, Austria). Heraclius: solidus CON, 625 629, found in a gra ve; Hahn, "Die Fundmnzen des 5. 9. Jahrhunderts 458. 11. A (Dolj, Romania). Heraclius: M, NIK, 612 3. 11a. A Justinian I: solidus CON, 527 537, pierced, loop a ttached; A. Vlcu, T. Isvoranu, and E. Nicolae, (Wetteren: Moneta, 2006), 150, no. 441 and pl. 15/441. 12. A MGATA (Karachayevo Cherkesiya, Russia). Constantine IV: solidus CON, 669 681; Kropotkin Klady vizantiiskikh monet 23, no. 35. 13. A NI (Kars, Turkey). 14 13 The coins are kept in the local history museum but ther e is no certainty that they were found in Alba hundred kilometers away from Alba Iulia (below, no. 230). 14 All coins have been found during archaeologica l excavations conducted by Joseph Orbeli and Nicholas Marr before the First World War. Many coins had to be redated because of the frequent discrepancies in the catalogue between the description and the reference to BMC.

PAGE 487

487 Anastasius I: M, CON, 498 512; 3 M, CON(?), 512 518; K, ANT, 512 518; Mousheghian et al., History (Ani) 72, no. 19 23. Justin I: M, CON, 518 522; M, CON, 522 527; CON, 522 527; K; Ibid., 72 73, no. 24 27. Justinian I: 3 M, CON, 527 537; M, NIK, 527 538; M, 527 538; M, CON, 539 540; M, CON, 542 543; M, ANT, 551 552; M, CON, 558 559; M, ANT, 559 560; 2 M, 538 565; K, 538 565; Ibid., 73 74, no. 29 41. Justin II: M, NIK, 567 568; 2 M, NIK, 568 569; K, NIK, 573 574; 2 M, CON, 576 577; M, NIK, 577 578; M; 2K; Ibid., 74 75, no. 42 51. Tiberius II: M, CON, 581 582; Ibid., no. 52. Maurice: M, CON, 586 587; 3 K, CON, 590 591; M, ANT, 591 592; K, ANT, 594 595; M, ANT, 601 60 2; K; Ibid., 75 76, no. 53 60. Phocas: K, KYZ, 603 604; M, ANT, 603 604; M, CON, 607 608; M, ANT; M; Ibid., 76, no. 61 65. Heraclius: 2 hexagrammata CON, 625 629; M, CON, 612 613; 3 M, 613 616; 12 nummia Alexandria; Ibid., 77, no. 66 73. Constans II: M, CON, 642 643; Ibid., 77, no. 74. Sixth century: I; Ibid. 73, no. 28. 14. A PALINA Justinian I: solidus CON, 537 542; Velter, Transilvania n secolele V XII 287, no. 3 (dated 527 538); 15 I. Winkler, Michael Pap Szatmri (1747 1812), SCN 3 (1960): 440. 15. A PARHANT (Tolna, Hungary). Justin II: M, CON, 572 573; Somogyi, "Byzantinische Fundmnzen," 268 269, no. 1/1. Maurice: M, CON, 589 590; Ibid. 269 270, no. 2 (later dated 590 591, cf. Somogyi, New Remarks on the Coin Flow 101, n. 39). Phocas: K, CON, 603 610; Somogyi, "Byzantinische Fundmnzen," 268 269, no. 1/2. 16. A RAD (Arad, Romania). Justinian I: 3 AE; Butnariu, " 217, no. 5. Justin II: M, NIK, 570 571; Ibid., 217, no. 6. Heraclius: hexagrammon CON, 615 625; E. Oberlnder r ale un punct de vedere numismatic," CN 8 (2002): 140, n. 142. 15 In the original publication, Iudi which corresponds to MIBE 7 (542 565). The description, however, clearly points to a different type ( MIBE 6, 537 542), which Sabatier did not illustrate in his 1862 reference catalo gue. Winkler, whose specialty catalogue, which had been otherwise long obsolete by 1960. Surprisingly, Viorel Butnariu wrongly identified the coin as type DOC 3b (= MIBE 5), which is an issue of 527 537, the wrong dating being adopted in subsequent publications as well. See Winkler, "Despre," 440; monedelor bizantine," 217, no. 4. It is also worth mentioning that Szatmri, an eighteenth century collector, noted that the coin had been found "together with many others," which might imply the existence of a dispersed hoard.

PAGE 488

488 17. A RADAC M ( Central Banat Serbia). Maurice: M, NIK, 583 584, found in a grave; Somogyi, Byzantinische Fundmnzen, 23, no. 1. 18. A RADUL N OU (Arad, Romania). Justin II: 1 AE; Lakatos, "Monede bizantine ," 251, no. 31. 19. A REVSHAT (Ararat, Armenia). Justin II: M, NIK; Mousheghian et al., History 168. 20. A RGETOAIA (Dolj, Romania). Phocas: M, CON, 602 610, found during archaeological excavations conducted by Grigore Tocilescu in 1905; B. Mitrea, Dcouvertes rcentes et plus anciennes de monnaies antiques et byzantines sur le territoire de la Rpublique Populaire Roumaine, Dacia 10 (1966): 412. 21. A T OTS I (Shida Kartli, Georgia). Constans II: hexagrammon CON, 654 659; Abr amishvili, Sakartvelos sakhelmts'ipo muzeumis, 71, no. 229. 22. B Justin I: M, CON, 518 522; B. Mitrea SCIV 18, 1 (1967): 201, no. 71. Curtea 217, no. 9. I, Rome, 547 549; E. Oberlnder Trnoveanu, XIV n lumina descoperirilor monetare," Suceava 26 28 (1999 2001): 317, n. 31. Justin II: M, CON, 574 575; Butnariu, " 217, no. 10. Tiberius II: M, CON 580 581; Oberlnder Trnoveanu, " 319, n. 40. Maurice: M, CON, 599 600; Butnariu, " 217, no. 11. Maurice Heraclius: M, CON; Ibid., 217, no. 12. 23. B P ALANKA He raclius: solidus CON, 629 631; solidus CON, 637 638; Somogyi, Byzantinische Fundmnzen, 24 26, no. 2 3. 1 AV; Prohszka, "Altneue byzantinische Mnzen der Awarenzeit (Ergnzungen zum Buch von Pter Somogyi: Byzantinische Fundmnzen der Awarenzeit Innsbru ck, 1997)," AAASH 55 (2004): 106, no. 10. 24. B P ETROVO S ELO Heraclius: solidus CON, 616 625; Somogyi, Byzantinische Fundmnzen, 26 27, no. 5. 25. B AD D EUTSCH A LTENBURG /P ETRONELL (=Carnuntum) (Niederosterreich, Austria). 16 16 The high number of gold coins is unusual and given the fact that many were acquired from private collectors it may be possible that at least some of them were initially part of larger, now dispersed, hoards. Their finding place is also under question, for whi ch see Winter, "Die byzantinischen und karolingischen," 47, n. 24; Somogyi, "Byzantinische Fundmnzen," 240, n. 8.

PAGE 489

489 Justin I: tremissis CON, 518 527; Winter, Di e byzantinischen und karolingischen Fundmnzen aus dem sterreichishen Bereich der Avaria eine Neubearbeitung, in Byzantine Coins in Central Europe Between the 5th and 10th Century ed. (Cracow: Polish Academy of Sciences, 2009), 336, no. 1/2 Justin I/ Justinian I: CON/NIK, 522 527 or 527 537; Ibid. 336, no. 1/1 and 347, fig. 6/1. Justinian I: solidus CON, 527 537; solidus 537 542; 6 solidi CON, 542 565; 3 tremisses 527 565; ANT, 527 529; M, CON, 538 539; K, CON, 544 545; M, CON, 5 47 548; IS, TES, 538 552; 2 I, 527 565; 2 527 565; Ibid. 336 38, no. 1/3 17 and 347 49, fig. 6/2 10. Justin II: 2 solidi 565 578; M, CON, 565 566; K, NIK, 570 571; M, CON, 575 576; K, TES, 575 576; M, NIK, 565 578; K, military mint, type 568 569. Ibid 338, no. 17 23 and 349, fig. 6/11 12. Tiberius II: solidus CON, 578 582; Ibid., 339, no. 1/23a. Maurice: 2 solidi CON, 583 602; I, Ravenna, 586 602; Ibid. 339, no. 23b c, 24 and 349, fig. 6/13. Phocas: solidus CON, 607 609; I, Carthage, 606 608; 12 nummia Alexandria, 602 610; Ibid. 339, no. 24a, 25 26 and 349, fig. 6/14. Heraclius: solidus CON, 616 625; solidus CON, 629 632; semissis CON, 613 641; solidus (gold plated imitation), after CON, 613 616; semissis (gold plated imitation), after Ravenn a, 610 613; I, Catania, 614 615; 12 nummia Alexandria, 610 641; I, Carthage, 610 641; Ibid. 339, no. 27 31 and 345, no. 17/1 2 with 351 52, fig. 6/23 24. Constans II: solidus CON, 662 667 (?); Ibid., 339, no. 32 and 349, fig. 6/15. 26. B i, Romania). Justinian I: K, CON, 540 541, found in the sixth century settlement; Butnariu, " 217, no. 14; D. G. Teodor, Teritoriul est carpatic n veacurile V rii poporului romn 26a. B AJA (Bcs Kiskun, Hungary). Heraclius: solidus CON, 616 625; I. Bna, "Review of P. Somogyi, Byzantinische Fundmnzen der Awarenzeit," AAASH 54, no. 1 2 (2003): 295. 26b. B ALOTASZLLS (B cs Kiskun, Hungary). Anastasius I: solidus 492 507; A. Kiss, "Rgszeti s numizmatikai adatok a Duna Tisza teleplstrtnethez," Mra Ferenc Mzeum Evknyve. Studia Archaeologica 4 (1998): 191 27. B ALTA T Justinian I: M, CON, 542 543; E. Oberlnder Trnoveanu and E. M. Constantinescu, Mousaios 4 (1994): 317, no. 22. 28. B ALTA V ERDE mania). Justin II: M, KYZ, 576 577; Butnariu, " 217, no. 13. 29. B AKONYTAMSI H ATHALOM PUZSTA (Veszprm, Hungary).

PAGE 490

490 Maurice/Phocas: solidus (imitation), found in a grave; Somogyi, Byzantinische Fundmnzen 23 24, no. 1a. 3 0. B ANATSKA P ALANKA S APAJA (South Banat, Serbia). 17 Justin I: M, CON, 518 527; E. Oberlnder Trnoveanu, 565), Muzeul 16 (2004):62, n. 65. 31. B ANATSKI B RESTOVAC (South Banat, Serbia). Justin II: 1 AE; Somogyi, Byzantinische Fundmnzen 31, no. 12. 32. B ANATSKI K ARLOVAC (South Banat, Serbia). Justinian I: l ight weight solidus (20 siliquae ), CON, 538 542; Oberlnder Trnoveanu, " 69 70, n. 72. 33. B ANATSKO N OVO S ELO (South Banat, Serbia). Phocas: M, NIK, 609 610; Oberlnder n. 109. 34. B ARLAD (Vaslui, Romania). Anastasius I: M, CON, 498 507; E. Oberlnder Carpica 13, n. 2 (1992): 228, no. 1. M, CON, 498 518. S. Langu, C. Onel, and C Giurcanu, "Note asupra unor monede dintr Elanul 93, November (2009) : 4. Justinian I: 4 AE; D. Gh. Teodor, rom 47, no. 55. I, CON, 562 563; Oberlnder bizantine ," 229, no. 2. Justin II: M, NIK, 570 571; Ibid., 229, no. 3. Maurice: K, CON, 583 584; I, CON, 582 560; Ibi d., 229, no. 4 5. Phocas: K, CON, 602 610; Ibid., 229, no. 6. Constans II: M, CON, 642 643; M, CON, 656 657; Ibid., 230, no. 7 8. 35. B RNBACH M ITTERNDORF (Styria, Austria). Justin I: M, TES, 519 522; Hahn, "Die Fundmnzen des 5. 9. Jahrhunderts 460. 36 B ATINA ( Osijek Baranja, Croatia). Justinian I: M, CON, 527 538; H. Gricke Baranje," in Radovi XIII. 1.10.1994) vol. 2, ed. N. Cambi and E. Marin (Vatican/Split: Pontificio Istituto di Archeologia Cristiana 37. B ATUMI (Ajaria, Georgia). Anastasius I: K, CON, 512 518; Abramishvili, Sakartvelos sakhelmts'ipo muzeumis (1966 1984), 11, no. 29. 38. B EBA V ECHE 17 Byzantine bridge head on the Danube's left bank.

PAGE 491

491 Constans II: solidus CON, 642 647; Somogyi, By zantinische Fundmnzen 27 28, no. 7. 39. B ECLEAN Justinian I: 1 AV; Lakatos, "Monede bizantine ," 247, no. 3. 40. B KSCSABA R EPLTR (Bks, Hungary). Heraclius: solidus CON, 616 625, found in a grave; Somogyi, Byzantinisch e Fundmnzen, 28 29, no. 8. 40a. B KSGYULA (Bks, Hungary). Justinian I: light weight solidus (21 siliquae ), CON, 537 Piese de in Imperiu. Studii n onoarea lui Radu Harhoiu ed. Renaissance, 2010), 147, fig. 2/5a b. Maurice: solidus CON, 583; Prohszka, Altneue byzantinische Mnzen 103, no. 3. 41. B EREZENI (Vaslui, Romania). Tiberius III: M, Ravenna, 698 70 5; Stoliarik Essays on Monetary Circulation, 142, no. 79. 42. B ESKO (Sanok, Poland). Justinian I: M, NIK, 544 545; 7. Jh. In Polen," in Byzantine Coins in Central Europe Between the 5th and 10th Century 497, no. 1, and 507, fig. 1 (dated 543 544 or 547). 43. B ICH VINTA P IT YUS (Abkhazia, Georgia). Anastasius I: 3 M, CON, 512 517; M, 512 517; 2 K, CON, 512 517; I. Tsukhishvili, " in vol. 2, ed. A. Apakidze (Tbilisi: Metsnierieba, 1977), 314 15, no. 11 16 and 605, fig. 11 15. Justin I: 9 M, CON, 518 527; K, CON, 518 527; M, NIK, 518 527; Ibid. 315 18, no. 17 27 and 606 607, fig. 19 24, 26 28. Justinian I: 7 M, CON, 527 538; K, CON, 527 538; I, CON, 527 537; M, Carthage, 533 538; 18 K, ANT, 529 531/2; K, CON, 540 541; I, CON, 538 542(?); Ibid. 319 22, no. 35 49 and 608 09, fig. 36 43, 46. M, CON, 527 537 (imitation?); 19 2 M, CON, 527 537; M, NIK, 527 537; K, CON, 527 537; K, 550 565(?); 5 527 538; Abramishvili, Sakartvelos sakhelmts'ipo muzeumis (1966 1984), 16 19, no. 49 50, 52, 55, 60, 66 67 with n 102. 20 18 MIBE 18 5a var. (CART instead of KART). 19 The follis is underweight (7.85g) and displays the mintmark NOC inst ead of CON. 20 Abramishvili does not provide any details regading these finds from Pityus included in her 1989 catalogue of Byzantine coins in the Georgian National Museum, but they most probably date from archaeological excavations con ducted after 1977, since they could not be identified (by weight or description) with any of the coins published by Izolda Tsukhishvili in 1977 (most of which also show up in

PAGE 492

492 Justin II: M, CON, 565 566; K, NIK, 569 570; M, NIK, 574 575; M, CON, 576 577; Tsukhishvili, "Bich vint is bizant iuri monet ebi," 323 24, no. 50 53 and 609 10, fig. 50 53. Tiberius II/ Maurice: K, Rome (military mint); Abramishvili, Sakartvelos sa khelmts'ipo muzeumis (1966 1984), 19, no 63. 21 Maurice: M, NIK, 596 597; 22 M, ANT, 596 597; M, CON, 600 601; 23 K, TES, 585 586; K, TES, 602; K, TES; Tsukhishvili, "Bich vint is bizant iuri monet ebi," 324 26, no. 54 57 and 610 11, fig. 54 57; Abramishvili, Sa kartvelos sakhelmts'ipo muzeumis (1966 1984), 23 24, no. 81, 83. Heraclius: M, CON, 613 616; M, CON, 615 624; 30 nummia CON, 630 631(?); Tsukhishvili, "Bich vint is bizant iuri monet ebi," 326 27, no. 58 60 and 611, fig. 58 60. Tiberius III: solidus CON; Abramishvili, Sakartvelos sakhelmts'ipo muzeumis, 127, no. 84. 44. B IECZ (Gorlice, Poland). Justinian I: M, 527 "Byzantinische Fundmnzen," 497 98, no. 2 and 507, fig. 2 (possibly Constantinople mint, based on the illustration). 45. B ILHOROD D NISTROVS KYI (Odesa, Ukraine). Justinian I: M, CON, 542 543, found in the area of the med ieval fortress; Stoliarik Essays on Monetary Circulation, 137, no. 38. 46. B JELOVAR (Bjelovar Bilogora, Croatia). Justinian I: M, CON, 539 "Byzantine Coins in the Zagreb Archaeological Museum Numismatic Collection. Anastasi us I (A.D. 497 518) Anastasius II (A.D. 713 715)," 30 31: (1997 1998): 211, no. 3. 47. B LEIBURG (Carinthia, Austria). Anastasius I: M, CON, 507 512; Hahn, "Die Fundmnzen des 5. 9. Jahrhunderts, 454. 48. B OBNICE (Nym burk, Czech Republic). Justin I: 1 AV; J. Militk, "Finds of the Early Byzantine Coins of the Early Byzantine Coins of the 6 th and 7 th Century in the Territory of the Czech Republic," in Byzantine Coins in Central Europe 369, no. C8. 49. B OBOKVATI (Ajari a, Georgia). 24 Early Byzantine: 72 coins; Tsukhishvili and Depeyrot, History and Coin Finds in Georgia 22. 21 Abramishvi li attributes it to Justinian I, but the description clearly corresponds to the imitative series type MIBEC 73 (Tiberius II) or MIBEC 155 (Maurice). 22 Same coin as Abramishvili, Sakartvelos sakhelmts'ipo muzeumis (1966 1984), 23, no. 82, without mentioning the finding place. 23 Same coin as Ibid., 22, no. 78 79 and pl. V/78, withou t mentioning the finding place. 24 In 1983 archaeologist Nino Inaishvili acquired 72 early Byzantine coins from inhabitants of Bobokvati. A number of 36 coins were found together i n a nursery garden, but there is no specific information about this potential hoard. The coins date from Anastasius to Maurice.

PAGE 493

493 50. B O (North Banat, Serbia). Justinian: solidus ; found in grave 4; Epoque prhistorique et protohistorique en Yougoslavie Recherches et rsultats ed. G. Novak et al. (Belgrade: Socit archologique de Yougoslavie, 1971), 191. 51. B OHUSLAVICE (Hodonn, Czech Republic). Heraclius: solidus (imitation), 616 625; Militk, "Finds of the Early Byzantine Coins ," 361 and 381 82, no. M37. 52. B OLHRAD (Odesa, Ukraine). Justin II: M, KYZ, 576 577; Stoliarik Essays on Monetary Circulation, 138, no. 51. 53. B OL SHAIA O RLOVKA (Rostov na Donu, Russia). Tiberius III: solidus pierced, found in a grave; M. Kosianenko, "Pogrebenie u slobody Bol'shaia Orl ovka rannii pamiatnik saltovo maiackoi kul'tury," in Problemy khronologii arkheologicheskikh pamiatnikov stepnoi zony Severnogo Kavkaza ed. V. Ia. Kiiashko (Rostov na Donu: Izdatel'stvo Rostovskogo universiteta, 1983), 115. 54. B (Palilula, Serbia) Justin II: K; Somogyi, Byzantinische Fundmnzen 30, no. 10. 55. B OROLEA Justinian I: 1 AV, CON, 527 565; Butnariu, " 217, no. 15. 56. B (Sevnica, Slovenia). Heraclius: solidus ; Kos, "The Mon etary Circulation," 227, no. 23. 57. B (Suceava, Romania). Justinian I: M, CON, 527 538, found in the sixth century settlement, in sunken featured building 13; M, CON, 527 538, found in the sixth century settlement, in sunken featured building 20; D. G. Teodor, Suceava) (Bucharest: Editura Academiei, 1984), 61, fig. 30/1 2 and fig. 31/6,8. 58. B (Gorj, Romania). Justinian I: 1 AE; Oberlnder Trnoveanu, "Tra 59. B Phocas: M; Velter, Transilvania m secolele V XII 288, no. 14. 60. B RATEI (Sibiu, Romania). Justinian I: IS, TES, 538 552, found in the sixth century cemetery, M 190; L. Brzu, Ein gepi disches Denkmal aus Siebenbrgen: das Grberfeld Nr. 3 von Bratei (Cluj Napoca: Accent, 2010), 135; B. Mitrea, "Dcouvertes rcentes et plus anciennes de monnaies antiques et byzantines en Roumanie," Dacia 13 (1969): 550, no. 64. 61. B RATISLAVA (Bratislav a, Slovakia). Justin I: M, CON, 518 527; J. Hunka, "Finds of Byzantine Coins from the 5 th 10 th Century From the Northern Part of the Carpathian Basin," in Byzantine Coins in Central Europe 397, no. 4.

PAGE 494

494 Constantine IV: solidus CON, 674 681; Somogyi, Byzan tinische Fundmnzen, 31, no. 11. 62. B UCHAREST (Bucharest, Romania). Justin I: M, CON, 518 527, found on Turda st.; E. Oberlnder Trnoveanu, "Barbaricum apropiat X) Ia 4 (2004): 333, n. 12. K, CON, 518 527, found on Gen. 217, no. 19. Justinian I: M, CON, 527 537, found during archaeological excavations close to St. Nicolae 25 Butnariu, ndirea monedelor bizantine, 217, no. 21; 1 AE, 527 Kaiserzeit und der Vlkerwanderungszeit bei Bukarest," Germania 18 (1934): 210. 26 M, KYZ, 539 540, found in the area of St. Dumit ru church; Butnariu, monedelor bizantine, 217, no. 22; 1 AE, found during excavations on the Bl ; (Bucharest: Editura 1959), 33 34. Justin II: M, CON, 566 567, found during works at "23 August" stadium; M, CON, 571 bizantine, 218, no. 23 24. Justinian II: M, CON, 705 711, found in 2004 during archaeological excavations under the former building of the National Theater destroyed in WW2; E. Oberlnder olului al VIII in ed. G. (Bucharest: 75. 63. B UCHAREST M RELE (Bucharest, Romania). Anastasius I: solidus CON, 492 507; B. Mitrea, Dcouvertes montaires en Roumanie: 1981, 1982 et 1983, Dacia 28 (1984): 188, no. 131. 64. B UCHAREST O TOPENI (Bucharest, Romania). Justinian I: M, CON, 539 ndirea monedelor bizantine," 221, no. 105. 65. B UCHAREST S TRAULESTI M AICANESTI ( Bucharest, Romania ). 27 25 Butnariu mentions another find at the intersection of the -a coin belonging to the same type. But this is, in fact, the same coin found during excavations at Sf. Nicolae bizantine," 397, who mistaking ly reported two distinct finds. 26 Probably the same coin mentioned by Constantiniu, "Elemente romano bizantine," 668, n. 18. 27 Constantiniu, "Elemente romano bizantine," 668, n. 18 and 676, fig. 8/1 3 with the reverse of a follis from Constantinople (527 538), fig. 8/1, the obverse of a post reform issue (538 565), fig. 8/2, and the reverse of a half follis from Nicomedia (557 558), fig. 8/3. Fig. 8/1 and 8/2, due to the similar size of the coins, were mistakenly taken to represent the obverse and the reverse of the same coin, when in fact there are two different coins. None of the authors who included the coins from Bucharest expertize for an accurate attribution of the coins, wh ich otherwise constitute a crucial find, one of the rare

PAGE 495

495 Justinian I: M, CON, 527 537; K, KYZ, 540 541; K, NIK, 557 558, found during archaeological excavations in the sixth century settlement; M. Constantiniu "Elemente romano Munteniei n secolele VI VII e.n.," SCIV 17, n. 4 (1966): 676, fig. 8/1 3; M. Constantiniu, CAB 2 (1965): 189, fig. 93/1 3. 66. B (Taraclia, Moldova). Anastasius I: M, CON, 512 Topografiia kladov i nakhodok 86, no. 36. 67. B UDAKALSZ (Pest, Hungary). Heraclius: light weight solidus (20 siliquae ), CON, 616 625; Somogyi, Byzantinis che Fundmnzen, 31 32, no. 13. 67a. B UDAPEST (Budapest, Hungary). Heraclius: 1 AE, 612 613; Bna, Review of Somogyi 295. 68. B UDUREASCA (Prahova, Romania). Justin II: AE, 574 575; V. Teodorescu, "Les anciens Roumains," Roumanie. Pages d'histoire 5 (1980), no. 2: 76 77. 69. B URL CENI (Cahul, Moldova). Justinian I: 1 AE; Stoliarik, Essays on Monetary Circulation, 135, no. 29. 70. B UZ U Justinian I: CON, 541/542 551/552; Oberlnder Trnoveanu and Constantinescu, "Monede romane trzii ," 317, n. 23. Justin II: CON, 565 578; K, TES or KYZ, 567 568; K, TES, 572 573; Ibid., 319, n. 26, 29 and 30. Tiberius II: M, CON, 581 582; Ibid., 319, n. 28. Maurice: K, TES, 587 588 ; K, ANT, 592 593; Ibid., 317 19, n. 29, 31. Phocas: M, NIK, 605 606; Ibid. 319, 320, n. 28 a nd n. 31. Heraclius: solidus CON, 613 616; Ibid., 311, no. 32. Constantine IV: hexagrammon CON, 674 681; Ibid., 17, no. 33. 71. C ALAFAT (Dolj, Romania). Justinian I: I, Rome, 547 549; Oberlnder Trnoveanu, " 68, n. 70. Ju stin II: M, CON, 568 569; Butnariu, " 224, no. 176. 72. C MPENI (Cahul, Moldova). Justinian I: M, NIK, 538 539; K, NIK, 538 539; Stoliarik Essays on Monetary Circulation, 135, no. 27 28. occasions when early Byzantine coins have been found during systematic excavations in sixth century settlements. In fact, two of the coins had already been published and illustrated in an earlier art icle by the b, 2a b and 3. A brief comparison of the illustrations from the two articles reveals the fact that the obverse from fig. 8/2 in the 1966 article finds its rev erse at fig. 93/2b of the 1965 article.

PAGE 496

496 73. C MPULUNG M USCEL ia). Justin II: M; Butnariu, " 218, no. 32. 74. C ARACAL (Olt, Romania). Justinian I: K, CON, 541 542; Oberlnder Trnoveanu, " 68, n. 70. Maurice: M, CON, 590 591; Oberlnder Trnoveanu, La r 133, n. 74. 75. C AREI (Satu Mare, Romania). 28 Justin II: 565 578; Lakatos, "Monede bizantine ," 251, no. 33. 76. C RJA (Vaslui, Romania). Justin I: M, CON, 522 527, found together with sixth century pottery; Butnariu, onedelor bizantine, 218, no. 33. For the archaeological context, see Teodor, Descoperiri 64, no. 156. 77. C Topografiia kladov i nakhodok 82, no. 11. 78. EL KOVICE (Prague, Czech Republic). Justin II: K, NIK, 573 574; Militk, "Finds of the Early Byzantine Coins ," 374, no. C18. 79. C ELJE (Celje, Slovenia). 29 Justinian I: I, CON, 552 553; I, KYZ, 562 563; Kos, "The Monetary Circulation, 226, no. 16 17. Justi n II: K, TES, 574 575; Ibid., 226, no. 21. Constans II: M, CON, 642 645; 30 12 nummia Alexandria, 645 646; Ibid., 227, no. 29 30. 80. C EPTURA DE J OS (Prahova, Romania). Justin II: M, NIK, 569 570; Butnariu, 80a. Heraclius: 1 AV, CON; Prohszka, Altneue, 104, no. 5. 81. C ERNAT (Covasna, Romania). Justin II: K, 576 577; Butnariu, " 218, no 30. 82. ESTEREG (Central Banat, Serbia). Maurice: solidus CON, 584 602; Somogyi, Byzantinische Fundmnzen 32 33, no. 14. 28 Eighteen coins from the fourteenth century together with a fourth century Late Roman coin and the pentanummium of Justin II were brought to the museum in Oradea (Oradea, Romania) as a hoard found at Carei, but it i s highly unlikely that the ancient coins w ere part of the medieval hoard. 29 Also published in P. Kos, Die Fundmnzen der rmischen Zeit in Slowenien vol. 2 ( Berlin: Mann, 1988 ). 30 The coin appears as a solidus in Kos, "The Monetary Circulation," 227, no. 29, but is in fact a follis cf. Kos, Die Fundmnzen vol. 2, 58, no. 1547, with reference to MIB 162.

PAGE 497

497 83. C ETEA (Alba, Romania). Justinian I: solidus ; Velter, Transilvania n secolele V XII 289, no. 23. 84. C H ANDREBI (Shida Kart li, Georgia). Heraclius: hexagrammon CON, 625 629; Abramishvili, Sakartvelos sakhelmts'ipo muzeumis 126, no. 71. 85. C HECEA Constans II: semissis CON, 642 668; Oberlnder Trnoveanu, vremuri, 143, n. 154. 86. C ( "Byzantinische Fundmnzen," 500, no. 4. 87. C HEMBURKA (Krasnodar, Russia). Constans II: light weight solidus (20 siliquae ), CON, 662 668 (after MIB III ); Kropotkin, Klady vizantiiskikh monet 23, no. 33. 88. C HEREPIN (Cherkasy, Ukraine). Maurice: M; Kropotkin, Klady vizantiiskikh monet 37, no. 282. 89. C HERNOMORKA (Odesa, Ukraine). Maurice: M, Cherson, 582 602; Stoliarik, Essays on Monetary Circulation 139, no. 55 (Justin II). 90. C H IATURA (Imereti, Georgi a). Phocas: light siliqua CON, 602 607; Abramishvili, Sakartvelos sakhelmts'ipo muzeumis (1966 1984) 24, no. 86 and pl. VI/86. 31 91. C H IORA (Racha Lechkhumi, Georgia). Heraclius: solidus CON, 632 635; Abramishvili, Sakartvelos sakhelmts'ipo muzeumis 64, no. 187. 92. C H KHALT A (Abkhazia, Georgia). Justinian I: 1 AV, found in the fortress; Kropotkin, Klady vizantiiskikh monet 46, no. 477. 93. C HKHOROTSQU (Samegrelo Zemo Svaneti, Georgia). 32 Justinian I: 2 solidi ; Kropotkin, Klady vizantiiskikh monet 46, no. 478. 94. C HMI (North Ossetia Alania, Russia). Maurice: tremissis CON, 582 584(?), pierced twice, found in 1882 in a grave, during archaeological excavations conducted by D. Ia. Samokvasov; Kropotkin, Klady vizantiiskikh monet 31, no. 142/1. Pho cas: 2 solidi CON, 602 610, one of the coin s is pierced twice; Ibid. 31, no. 142/2 3. 95. C H ORVILA (Imereti, Georgia) 31 Same coin as Abramishvili, Sakartvelos sakhelmts'ipo muzeumis, 61, no. 171 and pl. XI/171, but withou t mentioning the finding place. 32 Possibly same co ins as below no. 522.

PAGE 498

498 Constans II: 1 AR; Tsotselia, Coin Finds in Georgia 149, no. 790. 96. C HOTUSICE (Kutn Hora, Czech Republic). Anastasius I: tremiss is CON, 492 518, with suspension loop; Militk, "Finds of the Early Byzantine Coins 367, no. C2. 97. C IOROIU N OU (Dolj, Romania). Justinian I: I, CON, 550 565; Oberlnder Trnoveanu, " 68, n. 70. Heraclius: K, CON, 610 616(?); Butnariu, " 218, no. 31. 98. C (UTAG, Moldova). Justin II: M, CON; Stoliarik Essays on Monetary Circulation 138, no. 50. 99. C OADA I ZVORULU I Tiberius II: solidus CON, 578 582, possibly part of a hoard found in the nineteenth century; Gh. Poenaru Bordea and B. Mitrea, Dcouvertes montai res en Roumanie 1992 (XXXVI), Dacia 37 (1993): 316, no. 71. 100. C OCIOC (Ilfov, Rom ania). Anastasius I: M, 512 518; 33 Butnariu, " 218, no. 34. 101. C (Cahul, Moldova). Justinian I: M, CON, 527 538; M, CON, 539 540; M, CON, 542 543; K, NIK, 538 539; Stoliarik Essays on Monetary Circulation 135, n o. 23 Topografiia kladov i nakhodok 87, no. 45a b. 34 102. C Heraclius: M, KYZ, 612 613; M, CON, 619 620, found together with sixth century 36; Teodo r, Teritoriul est carpatic 23, fig. 16/6 7. 103. C (Teleorman, Romania). Justinian I: M, NIK, 555 556; Butnariu, " 218, no. 37. Justin II: M, CON, 571 T eleorman," CN 3 (1980), 139. 104. C ORNI Justin II: M, ANT, 570 571; Butnariu, " 218, no. 38. 105. C OROTNA (Slobozia, Moldova). 33 In the original publication Bucur Mitrea was less certain about the attribution to Anastasius preferring a wider dating between 512 and 538, for which see B. Mitrea, "Dcouvertes rcentes de monnaies anciennes sur le territoire de l a Rpublique Populaire Roumaine," Dacia 4 (1960): 591, no. 14. 34 man in 1976 were accidental finds during works at a rural stadium by the Prut river during which a number of early medieval graves were destroyed. Among the f inds collected from the site there was also a gold coin of Michael VII (1071 1078).

PAGE 499

499 Anastasius I: M, CON, 512 518; M, CON, 512 518; A I Mussorov and L V N osova, "Nahodki vizantiiskih monet V VI vv. n Nizhnem Dnestre Stratum+ Znameniia civilizacii 6 (2001 2002): 304, no. 1 2, fig. 1 2. Justin I: K, KYZ or NIK, 518 527; Ibid., 304, no. 3, fig. 3. Justinian I: M, ANT, 554 Topografiia klad ov i nakhodok 85, no. 29. K, TES, 564 565, no. 160, pl. VI.10; Mussorov and Nosova, "Nahodki vizantiiskih monet ," 304 306, no. 1, fig. 4. Justin II: K, TES, 573 574; Ibid. 306, no. 5, fig. 5. 106. C (Vrancea, Romania). Heraclius: tremissis ; Ober lnder Trnoveanu, " 320, n. 46. 107. C OZLA Severin, Romania). Justinian I: K, CON; Velter, Transilvania n secolele V XII 289, no. 30. 108. C RAIOVA (Dolj, Romania). Anastasius I: 1 AV; Butnariu, delor bizantine, 218, no. 39. Justinian I: M, KYZ, 558 559; Oberlnder 68, n. 70. Maurice: M, CON, 587 588; K, TES, 588 589; Ibid., 133, n. 76. Heraclius: M, CON, 612 613; Butnariu, ne, 218, no. 40. 109. C Severin, Romania). Phocas: solidus CON, 607 609; Somogyi, Byzantinische Fundmnzen 33 34, no. 16. 110. C SRDASZLLS B ARTHALOM (Bks, Hungary). Heraclius: tremissis CON, 610 613; Somogyi, Byzantinische Fundmn zen 34 35, no. 17. 111. C Justin II: K, TES, 574 575, accidental find in a sixth century settlement; Teodor, Teritoriul est carpatic 23 and fig. 16/5. 112. C URCANI 35 Justinian I: solidus CON; Oberlnder Trnoveanu, Barbaricum apropiat, 337, n. 15. Constans II: solidus CON, 662 667; Oberlnder Trnoveanu, "Barbaricum apropiat," 357, n. 48 (date 659 668); A. Vlcu, II. Monnaies Byzantines (Wett eren: Moneta, 2009), 70, no. 193. 36 113. D ALJ (Osijek Baranja, Croatia). 35 According to Oberlnder Trnoveanu both coins have the same registry number (nr. 1657) so we might be dealing with just one coin. Alternatively, they may be part of a di spersed hoard registered under a single entry in museum records. 36 Vlcu expressed doubts that the coin was found in Curcani based on the fact that the coin belonged to a collector from Curcani who purchaesd coins from various other places, see Vlcu, Les monnaies d'or 13.

PAGE 500

500 Justinian I: 2 M, CON, 527 538; M, NIK, 541 542, M, Carthage, 534 539; 2 12 nummia Alexandria; Gric ke "Justinijanov novac," 1150 1154, no. 6, 8, 26, 41, 43, 45. 114. D Anastasius I: solidus ; H. Machajewski, "Skandynawskie elementy kulturowe na Pr 40 (1992): 88. 115. D (Anenii Noi, Moldova). Justinian I: M; Stoliarik Essays on Monetary Circulation 136, no. 34. 116. D ESZK (Csongrd, Hungary). Justinian I: 2 M, CON, 527 537; both coins found in the vicinity of an Avar age cemetery; Somogyi, Byzantinische Fundmnzen 35 36, no. 18. 117. D IDI C H Q ONI (Samegrelo Zemo Svaneti, Georgia). Heraclius: solidus CON, 625 629; Abramishvili, Sakartvelos sakhelmts'ipo muzeumis 64, no. 186. 118. D MANISI (Kvemo Kartli, Georgia). J ustin I: M, CON, 518 522; M, CON, 518 527, both found during archaeological excavations; Abramishvili, Sakartvelos sakhelmts'ipo muzeumis 42, no. 63; Abramishvili, Sakartvelos sakhelmts'ipo muzeumis (1966 1984) 13, no. 36. Justinian I: M, CON, 527 537, f ound in 1969 during archaeological excavations; Ibid. 17, no. 54 and pl. III/54. Justin II: M, NIK, 570 571; Ibid. 20, no. 71 and pl. IV/71. Phocas: M, CON, 604 605; Abramishvili, Sakartvelos sakhelmts'ipo muzeumis 61, no. 174. 119. D OBRENI nia). Justin I: 1 AE; Oberlnder Trnoveanu, " 316. 120. D (Dunajsk Streda, Slovakia). Justin I: 1 coin; Fiala, "Byzantsk mince ," 57, no. 1. 121. D OBRUN (Olt, Romania). Justinian I: M, NIK, 543 544; Butnariu, " 218, n. 42. 122. D (Vlcea, Romania). Justin II: M, NIK, 569 570; " 219, no. 43. 123. D OLN B OUSOV (Mlad Boleslav, Czech Republic). Anastasius I: 1 coin, possibly part of a hoard; Militk, "Finds of the Early Byzantine Coins ," 367 68, no. C5. 124. D Anastasius I: CON, 512 517; G. Poenaru Bordea and P. I. Dicu, "Monede romane SCN 9 (1989): 78, no. 85. 125. D OMNESTI (Ilfov, Romania).

PAGE 501

501 Maurice: solidus CON, 583 602; B. Mitrea, Dcouvertes rcentes et plus anciennes de monnaies antiques et byzantines sur le territoire de la Rpublique Populaire Roumaine, Dacia 7 (1963): 597, n o. 52. 126. D (Arad, Romania). Anastasius I: K, CON, 512 518, found in an inhumation grave, in a Gepid ceramic Archologische Denkmler der Gepiden im Mitteldonaubecken (4 54 568 u. Z.) (Budapest: Verlag der Ungarischen Akademie der Wissenschaften, 1961), 145, 292 and pl. CCLXXII/2 (for the pot). 37 127. D OROHOI Justinian I/ Heraclius: several AE; Butnariu, " 219, no. 45. 128. D Justinian I: solidus CON, 542 565; 38 Vlcu et al. Les monnaies 153, no. 460 and pl. pl. 16/460. 39 129. D (Suceava, Romania). Justin I: M, CON, 518 527; Butnariu, " 219, no. 47. 1 30. D RANDA (Abkhazia, Georgia). Phocas: solidus ; Kropotkin, Klady vizantiiskikh monet 43, no. 408. 131. D ROBETA Anastasius I: 2 M, CON, 507 512; Oberlnder Trnoveanu, antichitate, 54. 6 M, CON, 512 517; Ibid., 5 5, n. 50. Justin I: 6 M, 518 527; Ibid., 60 61, n. 63. Justinian I: 5 M, CON, 527 538; K, CON, 527 538; M, NIK, 527 538; M, CON, 542 543; M, CON, 543 544; M, CON, 544 545; M, CON, 546 547; M, KYZ, 546 547; IS, TES, 542 547; K, Salona, 540 542; M, CON, 553 554; IS, TES; K, TES, 563 564; Ibid., 68, n. 70. BSNR 88 89 (1994 1995): 273, no. 35. (art. pp. 269 274) Justin II: K, TES, 564 565; K, TES, 567 568; 2 M, CON, 568 569; K, CON, 568 569; M, CON, 569 570; M, NIK, 569 570; K, TES, 569 57 0; M, CON, 570 571; M, NIK, 570 571; K, TES, 570 571; M, military mint, 570 571; M, NIK, 571 572; K, TES, 574 575; M, CON, 575 576; M, NIK, 575 576; K, TES, 575 576; K, TES, 568 578; 2 CON, 565 578; NIK; 565 578; 2 K, TES, 565 578; Oberlnder Trnove anu, vremuri, 124, n. 17. Tiberius II: M, NIK, 580 581; CON, 578 582; Ibid., 129, n. 56. 37 Csallny provided a confusing attribution to Justinian I and then to Justin I, which was later corrected by B. Mitrea, Romna," SCIV 13, no. 1 (1962): 223. 38 Dated 527 edelor bizantine," 219, no. 46. 39 Misidentified as a light weight solidus (20 siliquae ) in Oberlnder Trnoveanu, "Barbaricum apropiat," 336 and 337, n. 15, with reference to DOC 3i.2 (full weight solidus ).

PAGE 502

502 Maurice: light weight solidus (20 siliquae ), CON, 583 602; M, CON, 582 583; K, TES, 582 583; M, CON, 588 589; M, ANT, 590 591; M, CON, 591 592; M, NIK, 592 593(?); K, TES, 594 595; K, TES, 595 596; M, ANT, 596 597; K, TES, 598 599; K, TES, 583 584. Ibid., 132, n. 70. Heraclius: M, CON, 612 613; M, CON, 613 614; Ibid., 138, n. 111. Tiberius III: M, CON, 700 701; Ibid., 147, n. 194. 132. D UKLA (Krosno, Poland). Justin 133. D G Justinian I: M, CON, 541 542; Butnariu, " 219, no. 48. 134. D (Baranya, Hungary). Heraclius: light weight solidus (20 siliquae ), CON, 616 625; Somogyi, Byzantinische Fundmnzen 37 38, no. 20. 135. D VIN (Ararat, Armenia). 40 Anastasius I: M, CON, 517 518; Mousheghian et al., History 62, no. 16. Anastasius /Justin I/ Justinian I: K; Ibid., 62 64, no. 17. Justin ian I: M, NIK, 527 538; Ibid. 62, no. 18. Justin II: M, NIK, 566 567; 41 M, CON, 570 571; M, CON, 572 573; 2M; Ibid., 62, no. 19 24. Phocas: M, ANT, 603 604; Ibid., 62 63, no. 25. Heraclius: 3 hexagrammata CON, 615 625; 17 hexagrammata CON, 625 629; 2 hex agrammata CON; 2 M, CON, 631 632; 2 M, CON, 635 636; Ibid., 63, no. 26 51. Tiberius III: solidus CON, 698 705; Ibid., 64, no. 52. 135a. D ZHAGA (Karachayevo Cherkesiya, Russia). Heraclius: 2 silver foil imitations (bracteates) after solidus dated 632 639 found in grave 40 and 48, respectively; E. V. Rtveladze and A. P. Runich, "Nahodki indikatsii vizantiiskikh monet vblizi Kislovodska," VV 32 (1971): 220 and fig. 1/3. Constans II: 2 thin gold foil imitations (bracteate s ) after solidi dated 654 667, found in grave 42; Ibid. 220 21 and fig. 1/5 6. Constantine IV: thin gold foil imitation (bracteate) after solidus dated 669 674; Ibid. 221. 136. D ZHIGINKA (Krasnodar, Russia). Justin I: solidus CON, 527 (joint reign of Justin I and Justinian I), attached t o a gold necklace with precious stone pendants; Kropotkin, Klady vizantiiskikh monet 21, no. 9 and fig. 14. 137. E (Borsod Abaj Zempln, Hungary). 40 All coins have been found during archaeological excavations in the ancient city of Duin 41 Two coins struck at Nicomedia in 566 567 are listed in the catalogue (no. 21 22) but the illustration on pl. I/21 22 (rubbing drawing and an actual photograph) shows that we are in fact dealing with one and the same coin.

PAGE 503

503 Justin II: M, NIK, 572 573, found in a Gepid cemetery, grave 31; Somogyi, Byzantinische Fundmnzen 38 39, no. 21. 138. E ISENSTADT (Burgenland, Austria). Justinian I: tremissis CON, 527 565; Winter, "Die by zantinischen un karolingischen," 344, no. 13 and 351, fig. 6/21. 139. E (Bks, Hungary). Constans II: miliarensis (imitation), 651 654; Somogyi, Byzantinische Fundmnzen 39, no. 22. 140. E NNS (Obersterreich, Austria). Justinian I: solidus CON, 5 42 565; Hahn, "Die Fundmnzen, 458. 141. E PURENI (Vaslui, Romania). Justinian I: 1 AE; Oberlnder Trnoveanu, " 317. 142. F (Hunedoara, Romania). Tiberius II: 1 AE; Lakatos, "Monede bizantine ," 247, no. 8. 143. F CIU (Vaslui, Romania). Justin II: M, CON, 576 577, together with sixth century pottery; Butnariu, numismatice 88, no. 288. 144. F (Suceava, Romania). Justinian I Heraclius: AE; Butnariu, " 219, no. 50. 145. F Maurice: M, KYZ, 589 Pontica 28 29, (1995 1996), 282, no. 10. 145a. F REUNDORF (Niedersterreich, Austria). Justinian I: solidus CON, 525 537, with suspension loop, found in grave 422; C. Blesl, Grber des 6. Jahrhunderts zwischen der Traisen und dem Wienerwald in Niedersterreich, in Kulturwandel in Mitteleuropa: Langobarden, Awaren, Slawen: Akten der Internationalen Tagung in Bonn vom 25. bis 28. Februar 2008 ed. J. Bemmann and M. Schmauder (Bonn: R. Habelt, 2008), 327, fig. 17. 146. F URSY (Kiev, Ukraine). Justinian I: M, CON, 527 537; Kropotkin, Klady vizantiiskikh monet 33, no. 192. 147. F (Rezina, Moldova). Justinian I: M, CON, 539 540; Butnariu, izantine, 225, no. 200. 148. G 42 Justinian I: solidus CON, 527 537; 3 M, CON, 527 537; M, NIK, 527 537; Butnariu, " 219, no. 51 55. 42 Possibly part of a hoard.

PAGE 504

504 149. G ALIAT (North Ossetia Alania, Russia). Heraclius: soli dus CON, 613 625(?), pierced twice; found in a funerary chamber during excavations conducted by E. I. Krupnov; also in the chamber, an Ummayad dirham (700 701); Kropotkin, Klady vizantiiskikh monet 30, no. 136. John Smedley suggested this might be a 23 s iliquae light solidus for which see J. Smedley, Seventh Century Byzantine Coins in Southern Russia and the Problem of Light Weight Solidi," i n Studies in Byzantine Gold Coinage eds W. Hahn and W. E. Metcalf ( New York: American Numismatic Society, 1988 ), 116, no. 22. 150. G ARNI (Kotayk, Armenia). Constans II: hexagrammon CON, 648 652; M; Mousheghian et al., History 178, no. 3 4. 151. G HEORGHE D OJA (Mure Romania). Justinian I: 1 AR; 43 XXVI.1. 152. G HERLA (Cluj, Romania). Anastasius I: K, CON, 512 518; Velter, Transilvania n secolele V XII 291, no. 53. Maurice: K, CON, 588 589; Lakatos, "Monede bizantine ," 248, no. 10. 153. G Justin II: M, NIK, 569 nord Memoria Antiquitatis 18 (1992): 72, no. 17, fig. 40/2. 154. G HINDENI (Dolj, Romania). Justin I: M, CON, 518 527; K, CON, 527; Butnariu, bizantine, 219, no. 58 59. Justinian I: M, CON, 541 542; M, NIK, 556 557; Ibid., 219, no. 60 61. Justin II: K, NIK, 566 567; M, CON, 574 575; Ibid., 219, no. 62 63. 155. G IARMATA Justin II: M, NIK, 574 575; Lakatos, "Monede bizantine ," 251, no. 35. 156. G (Dolj, Romania). Justin II: M, NIK, 568 569; K, TES, 576 577; Butnariu, bizantine, 219, no. 65 66. Tiberius II: K, TES, 578 579; Ibid., 219, no. 67. 157. G OG Maurice: M, CON, 597 598; Oberlnder Trnoveanu, " 133, n. 72. 158. G OICEA (Dolj, Romania). Anastasius I: M, CON, 512 517; Teodor, Romanitatea nord 106, fig. 14/1. Justin II: K, NIK, 566 567; I bid. 106, fig. 14/7. Maurice: M, CON, 583 584; Ibid. 106, fig. 14/6. 43 probably from the sixth century," but does not provide any description or illustration so the attribution must be taken with caution.

PAGE 505

505 159. G ORNET (Prahova, Romania). Justin II: M, NIK, 573 574; C. Preda, bizantine, BSNR 80 85 (1986 1991), 290, no. 7. 160. G ORYACHI K LYUCH (Kr asnodar, Russia). Justin I: solidus ; Kropotkin, Klady vizantiiskikh monet 21, no. 7. 161. G OVORA (Vlcea, Romania). Justinian I: 1 AE; D. Tudor, SCIV 16, n. 1 (1965): 184, n. 20. 162. G RODZISKO D OLNE Herac lius: M, NIK, 613 "Byzantinische Fundmnzen," 500 01, no. 6 and 511, fig. 6/12. 163. G ROPENI Phocas: 1 AE; Preda, monedelor bizantine 402 with n. 59. 164. G RUMEZOAIA (Va slui, Romania). Phocas: K, KYZ, 603 604; K, CON, 603 604, both f ound together with sixth century Carpica 11 (1979): 206, fig. 1 1/1 2. 165. G URJAANI (Kakheti, Georgia). 44 Constans II: solidus CON, 642 644(?); Abramishvili, Sakartvelos sakhelmts'ipo muzeumis 127, no. 76. Constantine IV: solidus ; Tsotselia, Coin Finds in Georgia 149, no. 793. 166. G VANK ITI (Imereti, Georgia). Co nstantine IV: solidus CON, 681 685; Abramishvili, Sakartvelos sakhelmts'ipo muzeumis 73, no. 242. 167. G VESO (Racha Lechkhumi and Kvemo Svaneti, Georgia). Justin I: 1 AV; Tsotselia, Coin Finds in Georgia 130, no. 692. 168. G YENESDIS (Zala, Hungary). Constans II: solidus CON, 654 659, found in an Avar cemetery, grave 64; Somogyi, Byzantinische Fundmnzen 43, no. 26. 169. H AGIMUS Justinian I: M, CON, 527 Topografiia kladov i nakhodok 85, no. 30. 170. H AJDDOROG V ROSKERT (Hajd Bihar, Hungary). Heraclius: light weight solidus (20 siliquae ), CON, 610 613, with suspension loop found in the Avar age cemetery, grave 1; Somogyi, Byzantinische Fundmnzen 43 44, no. 27. 171. H AJDNNS F RJHALOM JRS (Hajd Bihar, Hun gary). Constans II: solidus Rome, 662 663, found in the Avar cemetery; Somogyi, "Byzantinische Fundmnzen," 272 273, no. 5. 44 Possibly same as Kropotkin, Klady vizantiiskikh monet 43, no. 399 (Constantin e IV). Medea Tsotselia mentions two solidi of Constans II and Constantine IV, res pectively, both found in 1948.

PAGE 506

506 172. H ARKNY (Baranya, Hungary). Heraclius: solidus (imitation), CON, 616 625; Somogyi, Byzantinische Fundmnzen 44 45, no. 28. 173. H Justin II: M, CON, 572 "Byzantinische Fundmnzen," 501, no. 7 and 509, fig. 6/8. 174. H EREPEIA (Hunedoara, Romania). Justinian I: AE, CON, 538 539; Lakatos, "Monede bizantine ," 248, no. 11. 175. H RHELY (Csongrd, Hungary). Justin II: solidus CON, 567 578; Somogyi, Byzantinische Fundmnzen 45, no. 29. 176. H OGHIZ U NGRA Maurice: solidus (imitation), after CON, 584 602; 45 K, 597 598 46 ; Lakatos, "Monede bizantine," 247, no. 7 and 248 no. 12. 177. H (Prague, Czech Republic). Anastasius I: tremissis CON, 492 518; Militk, "Finds of the Early Byzantine Coins," 366, no. C1. 178. H ORGA (Vaslui, Romania). 47 Justinian I: M, ANT, 537 ine," 220, no. 72. Justin II: K, TES, 569 570; Ibid. no. 73. 179. H ORODENKA (Ivano Frankivsk, Ukraine). Justinian I: solidus ; Kropotkin, Klady vizantiiskikh monet 36, no. 254. 180. H T NEC (Chrudim, Czech Republic). Justin I: solidus CON, 522 5 27, found in the area of the Roman settlement; Militk, "Finds of the Early Byzantine Coins ," 368, no. C6. 181. H (Vaslui, Romania). Justinian I: M, CON, 541 542; Butnariu, " 220, no. 74. Justin II Heraclius: AE; Ibi d., 220, no. 75. 182. I Justinian I: 1 AV; Butnariu, " 220, no. 76. 45 Possibly the same coin as no. 344 below (Rupea). In both cases the finding place is somewhat vague, between Hoghiz and Ungra and the area of Rupea, respectively. The three settlements are within a 5 10km radius. Moreover, in one of the early publications Kurt Horedt mentioned that the solidus from the area of Hoghiz and Ungra, which was described as an imitation based on the shape of the edge, belonged to H. Mller from Rupea. This notice might constitute the initial source of the confusion and we may in fact be dealing with one and the same find. See K. Horedt, secolele IV XIII (Bucharest, Editura Aca demi ei RPR, 1958), 106, no. 2. 46 Same as Somogyi, Byzantinische Fundmnzen, 39 47 Both coins were found together with sixth century wheel and hand made pottery and a so called "Slavic" bow fibula. See Teodor, mismatice 101, no. 363.

PAGE 507

507 183. I DVOR (South Banat, Serbia). Heraclius: solidus CON, 625 629; Somogyi, Byzantinische Fundmnzen 103, no. 93; Somogyi, "New Rem arks on the Coin Flow ," 94, n. 33 (mislocated as Idvor, Romania). 184. I Heraclius(?): solidus CON, found in an Avar grave; Oberlnder Trnoveanu, "La 185. I UJNOE (Cahul, Moldova). Phocas: M; Nude Topografiia kladov i nakhodok 86, no. 39 (Trubaevka). 186. I ZMAIL (Odesa, Ukraine). Justin I: K, TES, 518 527; Stoliarik Essays on Monetary Circulation 134, no. 15. 187. J SZAPTI (Jsz Nagykun Szolnok, Hungary). Heraclius: tremissis (imitatio n), found in the Avar age cemetery, grave 264; Somogyi, Byzantinische Fundmnzen 47 48, no. 32. 188. J UTAS (Veszprm, Hungary). Phocas: M, NIK, 603 610, found in the Avar age cemetery, grave 116; Somogyi, Byzantinische Fundmnzen 48 49, no. 33. 189. K A MUNTA (North Ossetia Alania, Russia). 48 Anastasius I: solidus CON, 507 518; 2 silver brateates, one with suspension loop; Kropotkin, Klady vizantiiskikh monet, 30, no. 138/1; Y. A. Prokopenko, Byzantine Coins of the 5 th 9 th Century and their Imitations i n the Central and Eastern Ciscaucasus, in Byzantine Coins in Central Europe 545 46. Justinian I: solidus CON, 537 542; Kropotkin, Klady vizantiiskikh monet 30, no. 138/2. Justin II: solidus CON, 565 578; 1 AE; Ibid., 30, no. 138/3, 11. Tiberius II: so lidus CON, 578 582; Ibid., 30, no. 138/4. Maurice: solidus CON, 582 584(?);Ibid., 30, no. 138/5. Phocas: solidus CON, 602 610; Ibid., 30 no. 138/6. Heraclius: solidus CON, 613 632; 2 solidi CON, 632 641. Ibid. 30, no. 138/7 9. Constans II: solidus C ON, 654 659, pierced twice; found during archaeological excavations conducted by G. D. Filimonov; Ibid. 30, no. 139; 1 AR. Ibid. , 30, no. 138/10. Early Byzantine: 1 AE, found during archaeological excavations conducted by G. D. Filimonov; Ibid. 30, no. 140. 190. K APUSTYNTSI Justinian I/ Justinian II: solidus ; Kropotkin, Klady vizantiiskikh monet 36 37, no. 261. 190 a K ARCAG (Jsz Nagykun Szolnok, Hungary). 48 All coins, except the two specimens found during archaeological excavations, are late nineteenth century accidental finds from early medieval graves in Kamunta. Imitations of early Byzantine gold coins are also mentioned, with no further details, as well as gold bracteates of Anastasius and Phocas. The coins have been redated after MIB

PAGE 508

508 Constantine IV: solidus CON, 674 681; Prohszka, Altneue byzantinische F undmnzen 105, no. 7. 191. K HERSON (Kherson, Ukraine). Anastasius I: CON, 512 518; Stoliarik Essays on Monetary Circulation 133, no. 9. Justinian I: M, CON, 541 542; Stoliarik Essays on Monetary Circulation 137, n. 44. 192. K ICHKAS (Zaporizhia, Ukraine). Justin I: 1 AE; Stoliarik, Essays on Monetary Circulation 134 no. 19. Justinian I: 1 AV; Ibid. 138, no. 46. 193. K IEV (Kiev, Ukraine). Anastasius I: M, found during archaeological excavations on the Kiselivka Hill; Kropotkin, Klady vizantiiskikh monet 33, no. 184. Justinian I: M, found during archaeological exca vations on the Kiselivka Hill; Ibid. 33, no. 184. 194. K ISKUNDOROZSMA (Csongrd, Hungary). Phocas: light weight solidus (20 siliquae ), CON, 603 607; Somogyi, Byzantinische Fundmnzen 36 37, no. 19. Constans II: solidus CON, 667 668, found in the Avar c emetery Daruhalom II grave 21; Somogyi, "Byzantinische Fundmnzen," 273 274, no. 6. 194a. K ISLOVODSK (Stavropol, Russia). Tiberius III: 5 gold plated silver imitations after solidi of Tiberius III, found on a slope near a plundered grave at Kugul cl ose to Kislovodsk; Rtveladze and Runich, Nahodki indikatsii, 221 and fig. 1/8 10. 195. K ISZOMBOR (Csongrd, Hungary). Anastasius: solidus CON, 491 507, found in the Gepid cemetery, grave 40; Csallny, Archologische Denkmler 17 4; pl. CXV/11. Phocas: solidus CON, 603 607, found in the Avar cemetery O, grave 2; Somogyi, Byzantinische Fundmnzen 53 54, no. 36. Heraclius: light weight solidus (20 siliquae ), CON, 616 625, with suspension loop, found in the Avar cemetery at Tanyahal om "Byzantinische Fundmnzen," 274 275, no. 7. 196. K LRAFALVA (Csongrd, Hungary). Phocas: solidus (imitation), found in the Avar age cemetery, grave 30; Somogyi, Byzantinische Fundmnzen 54 55, no. 37. 19 6a K LIN Y AR (Stavropol Russia). Maurice: solidus found in the Alan age cemetery, together with Byzantine and Alan jewelry, grave 341; H. Hrke and A. Belinsky, Nouvelles fouilles de 1994 1996 dans la necropole de Klin Yar, in Les sites archologiques en Crime et au Caucase durant Age ed. M. Kazanski and V. Soupault (Leiden/ Boston: Brill, 2000), 201. Heraclius, solidus CON, 632 641 found in the Alan age cemetery, together with gold jewelry; Hrke and Belinsky, Nouvelles fouilles de 1 994 1996 202. 197. K LOS TERNEUBURG (Niedersterreich, Austria).

PAGE 509

509 Justinian I: M, KYZ, 541 542; Winter, "Die byzantinischen und karolingischen ," 340, no. 2 and 350, fig. 6/17. 198. K OLN (Koln, Czech Republic). 49 Justin I: 2 M, 518 527. Justin II: 1 AE. C onstans II (?): solidus (?) (gold plated), found near "the pagan graves." Early Byzantine: 3 coins. 199. K LKED F EKETEKAPU (Baranya, Hungary). Justinian I: IS, TES, 542 547, found in cemetery A, grave 354; Somogyi, Byzantinische Fundmnzen 56 57, no. 40. Justin II: solidus CON, 568 578, found in cemetery B, grave 119; K, KYZ, 569 570, found in cemetery A, grave 253; Ibid. 56, no. 39 and 57, no. 41/1. Maurice: solidus CON, 584 602, found in cemetery B, grave 119; Ibid. 57, no. 41/2. Heraclius: solidus CON, 613 616, found in cemetery A, grave 29; Ibid. 55 56, no. 38. 200. K ORETI (Imereti, Georgia). Heraclius: hexagrammon ; Abramishvili, Sakartvelos sakhelmts'ipo muzeumis 126, no. 70. 201. K OSZALIN (Koszalin, Poland). Anastasius I: solidus ; Machajewsk i, "Skandynawskie elementy ," 88. 202. K RANJ (Kranj, Slovenia). Justinian I: tremissis pierced; found in grave 43; V. Stare, preseljevanja ljudstev (Ljubljana: Narodni muzej, 1980), 107. 203. K RASNOPILKA (Che rkasy, Ukraine). Justin I: 1 coin; V. V. Kropotkin, "Novye nakhodki vizantiiskikh monet na territorii SSSR ," VV 26 (1965): 179, no. 80. 204. K RASNYI G ORODOK (Tambov, Russia). Maurice: M, Chersonesus; S. I. Andreev and N. V. Filimonova, "Ranneslavianskie kul'tury v Tambovskoi obl.," in Verkhnee Podon'e: arkheologiia, istoriia. Vypusk 3 k 65 letiiu so dnia rozhdeniia B.A. Folomeeva (1942 2001) i 25 letiiu so dnia nachala arkheologo geograficheskikh rabot na Kulikovom pole (1982 2007 ), ed. O. V. Burova, A.i N. Naumov and N. K. Fomin (Tula: Gosudarstvennyi muzei zapovednik "Kulikovo pole", 2009), 23; 20 fig. 2/5. 205. K 50 Anastasius I: M, CON, 507 512; Kos, "The Monetary Circulation, 226, no. 5. 206. K h Republic). Justinian I: solidus ; Militk, "Finds of the Early Byzantine Coins ," 380 81, no. M32. 49 The coins were found in the region of Koln, but some of them might belong to a dispersed hoard given the unusually high number of finds. See Mi litk, "Finds of the Early Byzantine Coins, 37 4 75, no. C19 and 378, no. C26. 50 Also published in Kos, Die Fundmnzen vol. 1.

PAGE 510

510 207. K RSICE (Psek, Czech Republic). Phocas: tremissis CON, 602 607; Militk, "Finds of the Early Byzantine Coins ," 375, no. C21. 208. RTS ANISI (Tbilis i, Georgia). Anastasius I: M, CON, 498 512; Abramishvili, Sakartvelos sakhelmts'ipo muzeumis 39, no. 49. 209. K RYMSKAYA (Krasnodar, Russia). Heraclius: solidus CON, 613 629; Kropotkin, Klady vizantiiskikh monet 21 22, no. 12. 210. K (Koln, Czech Republic). Heraclius: solidus (gold plated silver imitation), CON, 625 629, pierced; light weight solidus CON, 616 625, both coins found close to older hill fort period settlements; Militk, "Finds of the Early Byzantine Coins ," 360 and 376, no. C23 24. 211. K ULA Phocas: light weight solidus (20 siliquae ), CON, 603 607; Somogyi, Byzantinische Fundmnzen 58 59, no. 43. 212. K UNGOTA (Bks, Hungary). Justinian I: light weight solidus (20 siliquae ), CON, 542 562, found in an Avar grave; Somo gyi, Byzantinische Fundmnzen 59 60, no. 44. 213. K UNMADARAS (Jsz Nagykun Szolnok, Hungary). Heraclius: solidus ; Somogyi, Byzantinische Fundmnzen 60 61, no. 45. 214. K UNSZENTMRTON (Jsz Nagykun Szolnok, Hungary). 51 Constans II: solidus CON, 662 667; Somogyi, Byzantinische Fundmnzen 61, no. 46. 215. K UP YANS K (Kharkiv, Ukraine). Constans II: solidus CON, 654 659; Kropotkin, Klady vizantiiskikh monet 37, no. 265. 216. K UTAISI (Imereti, Georgia). Anastasius I: M, CON, 512 517; Abramishvili, Sakar tvelos sakhelmts'ipo muzeumis 123, no. 24; M, CON; K, CON; Tsukhishvili and Depeyrot, History and Coin Finds in Georgia 24, no. 1. Justin I: M, CON; Ibid., 24, no. 2. Justinian I: 4 M, CON; CON; Ibid., 24, no. 3. Justin II: M, NIK; Ibid., 24, no. 4. T iberius II: M, CON; Ibid., 24, no. 5. Maurice: M, NIK; 2 M, ANT; Ibid., 24, no. 6. Phocas: M, NIK; Ibid., 24, no. 7. 217. L (Teleorman, Romania). Justin II: M, NIK, 568 569; Butnariu, " 220, no. 78. 51 Same coin as Prohszka, "Altneue byzantinische Mnzen," 103, no. 2 with finding place Bks (Bks, Hungary).

PAGE 511

511 217a. L AK (Borsod Abaj Zempln, Hungary). Justinian I: solidus CON, 527 537; Prohszka, "Altneue byzantinische Mnzen," 105, no. 8. 218. L ASKI Anastasius I: AV; Machajewski, "Skandynawskie," 88. 219. L ASLOVO ( Osijek Baranja Croatia). Justinian I: M; Gricke "Justinijanov novac, 1155, no. 58. 220. L Justinian I: 1 AR; Lakatos, "Monede bizantine ," 248, no. 13. 221. Anastasius: solidus ; ski, Katalog stanowisk (Warsaw: Instytut Archeologii i Etnologii Polskiej Akademii Nauk, 1997), 33. 222. L ENDORF (Carinthia, Austria). Justinian I: solidus CON, 542 565; Ha hn, "Die Fundmnzen des 5. 9. Jahrhunderts ," 454. 223. L EU (Dolj, Romania). Justinian I: 1 AE; Butnariu, " 220, no. 79. 224. L EUNTEA Justin II: M, NIK, 570 571; Stoliarik Essays on Monetary Circulati on, 139, no. 53. 225. L IENZ (Tyrol, Austria). Justin I/Justin II: 1 AV; Hahn, "Die Fundmnzen des 5. 9. Jahrhunderts 461. 226. L INZ (Obersterreich, Austria). Heraclius: hexagrammon CON, ca. 630, found in a grave; cut in half; Hahn, "Die Fundmnzen de s 5. 9. Jahrhunderts," 459. 227. L IVEZENI Justinian I: M, NIK, 542 romane trzii ," 80, no. 116. 228. L JUBLJANA (Ljubljana, Slovenia). 52 Anastasius I: tremissis CON, 492 518; Kos, "The Monetary Circulation, 226, no. 3. 229. nd). Justin II: I, Carthage, 565 "Byzantinische Fundmnzen," 501 02, no. 9 and 509, fig. 6/9. Maurice: M, NIK, 583 584; Ibid., 502, no. 10 and 510, fig. 6/10. 230. L OKJANDARI (Kvemo Kartli, Georgia). Constans II: solidus CON, 642 647; Abr amishvili, Sakartvelos sakhelmts'ipo muzeum is (1966 1984) 35, no. 131 and pl. VII/11. 231. L OPATNA (Orhei, Moldova). 52 Also published in Kos, Die Fundmnzen vol. 1.

PAGE 512

512 Justinian I: M, CON, 539 540, found in association with Luka Rajkovetskaia ceramics; Kropotkin, Klady vizantiiskikh monet 38, no. 293, f ig. 17/26. 232. L OUKA (Most, Czech Republic). Maurice: M; Militk, "Finds of the Early Byzantine Coins ," 375, no. C20. 233. L Heraclius: light weight solidus (20 siliquae ), CON, 616 625, found in a grave; Somogyi, Byzantinische Fundmnzen 62 63, no. 48. 234. L OVOSICE Justinian I: I, Rome, 547 552; Militk, "Fin ds of the Early Byzantine Coins ," 371, no. C14. 235. L (Hodonn, Czech Republic). Anastasius I: solidus CON, 492 507, found in the "Migration period" cemetery, grave 113; Militk, "Finds of the Early Byzantine Coins ," 378 379, no. M28/1. Justinian I : tremissis CON, 527 565, with suspension loop, found in the "Migration period" cemetery, grave 55; Ibid., 378 379, no. M28/2. 236. M AGDALENSBERG (Carinthia, Austria). Justin II: M, NIK, 570 571; Hahn, "Die Fundmnzen des 5. 9. Jahrhunderts 454. 237. M (Vlcea, Romania). Justinian I: solidus CON, 527 538; 53 220, no. 80. 238. M AIAKI (Odesa, Ukraine). Justinian: M, NIK, 538 539; Stoliarik, Essays on Monetary Circulation 137, no. 37 239. M AJUR (Sisak Moslavina, Croatia). Justin II: M, CON, 571 240. M AKHARADZE (Guria, Georgia). Maurice: 1 AE; Abramishvili, Sakartvelos sakhelmts'ipo muzeumis 125, no. 53. 240a. M AKUKHIVKA (Poltava, Ukraine). Heraclius: solidus CON, 632 641, found together with gold objects most likely in a grave; Kropotkin, Novye nakhodki 178, no. 73. 241. M Justin II: solidus CON, 567 578; Butnariu, " 220, no. 81 (dated 565 568). 242. M ANTA (Cahul, Moldova). Justin I: M, CON, 518 Topografiia kladov i nakhodok 86 87, no 42. 243. M Justinian I: M, CON, 539 540; Militk, "Finds of the Early Byzantine Coins ," 380, no. M30. 53 very close at 4km distance. See G. I. Petre Govora, Continitatea daco IV VII e.n. n lumina noilor descoperiri arheolo ," Drobeta 2 (1976): 114.

PAGE 513

513 244. M ARAZLIIVKA (Odesa, Ukraine). Heraclius: hexagrammon CON, 625 629(?); Stoliarik Essays on Monetary Ci rculation 141, no. 67. 54 245. M ARIA S AAL (Carinthia, Austria). Maurice: 12 nummia Alexandria, 590 602; Hahn, "Die Fundmnzen des 5. 9. Jahrhunderts ," 454. Heraclius: 12 nummia Alexandria, 613 618; Ibid., 454. 246. M ARIBOR (Maribor, Slovenia). Heraclius : M; Kos, "Monetary," 227, no. 27. 247. M (Dolj, Romania). Justinian I: M, ANT, 554 83. 55 248. M ASIS (Ararat, Armenia). Heraclius: solidus CON, 641; Mousheghian et al., History 168 69. 249. M ATEU I (Rezina, Moldova). Early Topografiia kladov i nakhodok 81, no. 4. 250. M (Sibiu, Romania). Tiberius III: 1 AE; I. Dimian, "Cteva descoperiri monetare bizantine pe teritoriul R.P.R.," SCN 1 (1957): 197. 56 251. M EHADIA Severin, Romania). Heraclius: solidus CON, 616 625; Oberlnder vremuri," 139, n. 124. 252. M EL NYKY (Cherkasy, Ukraine). Justinian I: M, KYZ, 548 549; 1 AE; Kropotkin, Klady vizantiiskikh monet 37, no. 280. 253. M (Bks, Hungary). H eraclius: solidus (imitation), CON, 616 625, found in a grave; Somogyi, Byzantinische Fundmnzen 63 65, no. 49. 254. M (Vaslui, Romania). Phocas: K, TES, 602 255. M ISKOVICE (Kutn Hora Czech Republic). 54 s reference to Tolstoi (wrong plate, 51 instead of 48) and BMC is contradictory and points to several types of hexagrammata but the most frequent ones display a K on the reverse and Heraclius Const antine as adult, which corresponds to type MIB 140, dated 625 629. 55 Possibly the same coin as no. 341 below. 56 No illustration was provided and the confusing attribution to BMC pl. XL/8 ( solidus ) makes the identification irretrievably uncertain, despite the recent attempt of Pter Somogyi to speculate that the coin was an issue of Syracuse, for which see Somogyi, "Byzantinische Fundmnzen," 257. Most other Velter, Transilvania n secolele V XII 293, no. 83.

PAGE 514

514 Anastasius I: solidus CON, 507 518; Militk, "Finds of the Early Byzantine Coins 367, no. C3. 256. M ITTERNDORF (Niedersterreich, Austria). Justinian I: M, CON, 538 539; Winter, "Die byzantinischen und karolingischen," 340, no. 4 an d 350, fig. 18. 257. M OLDOVA V ECHE Severin, Romania). Justin II: M, CON, 570 571; Oberlnder n. 35. 257a. M OKRAYA B ALKA (Stavropol, Russia). 57 Justinian I: M, 538 656 (imitation), found in a grave. E. V. Rtveladze and A. P. Runich, Novye n ahodki vizantiiskikh monet i indikatsii v okrestnostiakh Kislovodska, VV 37 (1976): 153 and fig. 1/9. Justin II: light weight solidus (22 siliquae ), CON, 578 (joint reign Justin II and Tiberius II), found in grave 114, pierced twice; foure solidus found in a grave; Ibid., 151, 153 and fig. 1/1 2. Maurice: light siliqua CON, 583 602, found in grave 113, pierced twice; Ibid., 151 52 and fig. 1/3 4. Phocas: light siliqua CON, 607 610, found in grave 113 together with the siliqua of Maurice mentioned above and four Sasanian drachms dated between 531 and 628; copper imitation of a silver light miliarensis found in grave 74; gold plated silver imitation of an obverse of a solidus (bracteate), found in a grave; Ibid., 152 53 and fig. 1/5; Rtveladze and Runich Nahodki indikatsii, 219 and fig. 1/1. 58 Heraclius: gold imitation (bracteate), found in grave 117; Rtveladze and Runich, Novye nahodki, 153. 258. M OKRONOG (Mokronog Trebelno, Slovenia). Anastasius I: solidus ; 59 Kos, "The Monetary Circulation, 226, no 2. 259. M OSTOV (Galanta, Slovakia). 60 Heraclius: solidus (forgery), after CON, 625 629; Somogyi, New Remarks on the Flow 94, n. 33. 260. M TSKHETA (Mtskheta Mtianeti, Georgia). 57 Thirteen copper bracteates imitating Byzantine gold coins are described by Anna Ierusalimskaia in her monograph, A. Ierusalimskaia, frhmittelalterliche Funde an der nordkaukasischen Seidenstrasse (Munich: Editio Maris, 1996), 199 201, n o. 11 23 and pl. XLVI/fig. 109. 58 In addition, eleven imitations of solidi of Heraclius dated 613 625 have been found i n a grave near Kislovodsk. The imitations were pierced twice and were found in the area of the skull indicating that they were perhaps decorating a hat or a cap. See Rtveladze and Runich, "Nahodki ind ikatsii," 219 220 and fig. 1/2. 59 Possibly the same coin as no. 411 below. Both are late nineteenth century finds, see Kos, Die Fundmnzen v. 1, 3 90 91, no. 221.1 and 223/1.2. 60 Previously the finding place was thought to be Horn Saliby, for which see Fiala, "Byzantsk," 57 58, no. 5, fig. 1/4a b.

PAGE 515

515 Justin I: 1 coin, found in 1871 in grave during archaeological excavation s conducted by F. Baiern; Kropotkin, Klady vizantiiskikh monet 44, no. 424. Justinian I: heavy miliarensis CON, 527 537 (pierced), found in a stone grave during the archaeological excavations conducted by M. Berdzenishvili in 1961; Abramishvili, Sakartve los sakhelmts'ipo muzeumis 44, no. 74. Maurice: M, CON, 589 590, found in 1961 near Ninotsminda church during archaeological excavations conducted by M. Berdzenishvili; Ibid., 125, no. 58. Heraclius: hexagrammon found in 1937 during archaeological excava tions; Ibid., 126, no. 67. Constans II: M, CON, 642 643; Abramishvili, Sakartvelos sakhelmts'ipo muzeumis (1966 1984) 36 37, no. 138 and pl. VIII/138. Constantine IV: M, CON, 669 674; Ibid., 37, no. 139. 261. M YHIYA (Mykolayiv, Ukraine). Heraclius: M, NI K, 614 617; Stoliarik Essays on Monetary Circulation 141, no. 69. 262. N EAJLOVU (Giurgiu, Romania). Justin I: 1 AE, 518 527; Butnariu, " 220, no. 84. 263. N EKHVOROSCHA (Poltava, Ukraine). Early Byzantine: 1 AV (seventh century); Kropotkin, Klady vizantiiskikh monet 36, no. 251. 264. N EULENGBACH (Niedersterreich, Austria). Constans II: K, CON, 655 658; Winter, "Die byzantinischen," 340 41, no. 5 and 351, fig. 6/19. 265. N ICKELSDORF (Burgenland, Austria). Justinian I: solidus ; Winter, "Die byzantinischen," 344, no. 14. 266. N IZHNIY D ZHERAKH (North Ossetia Alania, Russia). Anastasius I: solidus CON, 507 518; Kropotkin, Klady vizantiiskikh monet 30, no. 141. 267. N OKALAKEVI (Samegrelo Zemo Svaneti, Georgia). Anastasiu s I: K, CON, 507 512; T. Abramishvili, Nokalakevisa da nojikhevis in Nokalakevi 1982 ed. P. Zakaraia (Tbilisi: Metsniereba, 1987), 277, no. 12. Justin I: 3 M, CON, 518 527; Ibid., 277 78, no. 13 15; M; Kropotkin, Klady vizantiiskikh monet, 44, no. 427. 61 Justinian I: M, CON, 527 538; Abramishvili, Nokalakevisa da nojikhevis 278, no. 16. 1 AE; Tsotselia, Coin 137, no. 727; M, CON, 546 547; T. Abramishvili, Nokalakevis arkeologiuri ekspeditsiis mier in Nokalakevi Arkeopolisi: III. arkeologiuri gatxrebi 1983 1989, ed. P. Zakaraia (Tbilisi: Metsniereba, 1993), 270, no. 1 and pl. L/5. Justin II: K, TES, 568 569; K; Abramishvili, Nokalakevisa da nojikhe vis 278, no. 17 18. 61 Kropotk not dated with the regnal year so this may in fact be an issue of Justinian I or his successors.

PAGE 516

516 Maurice: solidus CON, 583 602, Ibid., fig. 8/2. 62 268. N OST E (Shida Kartli, Georgia). Anastasius I: M; Abramishvili, Sakartvelos sakhelmts'ipo muzeumis 123, no. 23. 269. N OVACI (Giurgiu, Romania). Justinian I: M, CON, 540 541; But nariu, " 220, no. 85. Constans II: K, Carthage, 651 656; Ibid., 220, no. 86. 270. N OVOSIL S KE (Odesa, Ukraine). Anastasius I: K, CON, 512 518; Stoliarik, Essays on Monetary Circulation 132, no. 4. Maurice: M, 582 602; V. 15 note 2002), 120, 123, no. 1. 271. N YREGYHZA K ERTGAZDASG (Szabolcs Szatmr Bereg, Hungary). Maurice: light weight solidus (23 siliquae ), found in a grave; Somogyi, Byzantinische Fundmnzen 67 68, no. 52. 272. N YZHNYA D UVANKA (Luhansk, Ukraine). Constantine IV: M, with brothers Heraclius and Tiberius; Kropotkin, Klady vizantiiskikh monet 35, no. 233. 63 273. O ARDA D E J OS (Alba, Romania). Justinian I: solidus CON, 542 565; V. Suc iu, Noi descoperiri mone Alba, Apulum 38 (2001): 256 and fig. 2. 274. O okolie, Slovakia). Justinian I: several AV, possibly part of a hoard; Fiala, "Byzantsk," 57, no. 4. 275. O CHAKIV (Mykolaiv, Ukraine). Justin I: K, CON, 518 527; Stoliarik E ssays on Monetary Circulation, 134, no. 18. Justinian I: 2 K, CON, 538 565; Ibid., 137, no. 40 41. 276. O DAIA Heraclius: M, CON, 614/5, found in 1960 during archaeological excavations in the early medieval settlement; A. Rikman and I. A. Rafalovich, K voprosu o sootnoshenii Kratkie Soobshcheniia Instituta Arkheologii 105 (1965): 49, fig. 3/4. 64 62 Possibly same as the solidus mentioned by Tsotselia, Coin F inds in Georgia 1 43, no. 762, but found in 1988. 63 The date 668 669 was provided by Kropotkin, himself relying on late nineteenth early twentieth century e joint issues to 668 description of the coin it is impossible to re attribute the specimen from Nyzhnya Duvanka. 64 Same coin, but found at Lopatna instead of Odaia, in I.A. Rafalovich, Slaviane VI IX vekov v Moldavii

PAGE 517

517 277. O DORHEIU S ECUIESC (Harghita, Romania). Constantine IV : solidus CON, 674 681; Somogyi, Byzantinische Fundmnzen 68 69, no. 53. 278. FLDEK (Csongrd, Hungary). Phocas: solidus CON, 607 609; Somogyi, Byzantinische Fundmnzen 69 70, no. 54. 279. O HABA J IU (Gorj, Romania). Justinian I: solidus CON, 537 542 65 ; Vlcu, Isvoranu, and Nicolae, Les monnaies 150, no. 442 and pl. 15/442. 280. O ITUZ Maurice: 1 AE; Butnariu, " 221, no. 99. 281. O K AMI (Samtskhe Javakheti, Georgia). Constantine IV: solidus CON, 669 674; Abramishvili, Sakartvelos sakhelmts'ipo muzeumis 72, no. 239. 282. O Anastasius I: M, CON, 498 518; Stoliarik Essays on Monetary Circulation 133, no. 7. 283. O LARI (Olt, Romania). Justin I: solidus CON, 522 52 7; Oberlnder Trnoveanu, antichitate, 61, n. 63. 284. O Phocas: solidus CON, 602 610; Oberlnder Trnoveanu, Barbaricum apropiat, 356, n. 45. Heraclius: 1 AE; Butnariu, " 2 20, no. 88. 285. O L VIOPOL (Mykolayiv, Ukraine). Justinian I: 1 AE; Stoliarik, Essays on Monetary Circulation 137, no. 43. 285a. O PAVA Justin II: 1 AE, 576 577; Bna, Review, 295. 286. O RLEA (Olt, Romania). 66 Topografiia kladov i nakhodok," 81, no. 3, f edelor bizantine," 225, no. 206 65 Dated 527 7. 66 Several other sixth century coins found at Orlea were part of the coin collection of the local school teache r Constantin Iliescu: Justinian I (2 coins), Justin II (6 coins), Maurice (4 coins); unspecified sixth century ruler (1 coin), for which see B. Mitrea, istake, the coins from Orlea, a settlement close to s left bank, were later placed at Orlea, Alba county, in Transylvania, for which see A. M. Velter, XII n bazinul carp atic (cu ," SCIV 39, no. 3 (1988): 267, no. 72. As so often happens, the error gained further authority by getting cited in subsequent publications. In this particular case we are dealing with a double error as the coins were also misrepresented, gold coins instead of bronze, for which see Lakatos, "Monede bizantine," 249, no. 17.

PAGE 518

518 Anastasius I: M, CON, 512 517; Butnariu, " 220, no. 89. Justin I: M, NIK, 518 527; Ibid., 220, no. 90. Justinian I: M, CON, 527 538; M, CON, 538 539; K, KYZ, 555 556; Ibid., 220 21, no. 91 93. Justin II: K, TES, 568 569; Ibid., 221, no. 94. Maurice: K, CON, 582 583; Ibid., 221, no. 95. 287. O ROLIK (Vukovar Syrmia, Croatia). Tiberius II: K, TES, 581 288. O Anastasius I: 1 AE; Velter, Transilvania n secolele V XII 296, no. 107. Justinian I: 1 AE; Butnariu, " 221, no. 96. Justin II: M, NIK, 567 568; K, CON, 571 572; Ibid., 221, no. 97 98. Maurice: 1 AE; Velter, Transilvania n secolele V XII 296, no. 112. 289. O ania). Heraclius: solidus CON, 613 616; Oberlnder Trnoveanu, vremuri, 140, n. 136. 67 290. O SIJEK (Osijek Baranja, Croatia). Anastasius I: M, CON, 512 517; Coins ," 144, no. 21. Justinian: solidus CON, 537 542; tremissis CON, 527 556; M, CON, 527 538; M, NIK, 543 544, broken; CON, 542 562; 68 12 nummia Alexandria; Gricke "Justinijanov novac, 1149 54, no. 1 2, 7, 20, 28, 44. Maurice: K, CON, 592 593; tine Coins ," 187, no. 622. Heraclius: solidus CON, 616 625; Ibid., 194, no. 703. 291. O STROVO (Central Banat, Serbia). Anastasius I: M; Oberlnder 292. O STROVU B ANULUI Tiberius I monedelor bizantine," 221, 102. 293. O STROVU C ORBULUI Justin II: E; Nicolae, "Descoperiri," 272, no. 29. 294. O STROVU M ARE Justinian I : M, CON, 541 542; Butnariu, " 221, n. 103. Maurice: M, CON, 596 597; Ibid., 221, n. 104. 67 Initially published at the beginning of the twentieth century, the coin was identified as an issue of Constans II and Constantine I V, but after inspecting the coin, now in the collection of the Banat Museum Oberlnder Trnoveanu has recently reattributed it to Heraclius and Heraclius Constantine. For the older literature, see Somogyi, Byzantinische Fundmnzen 70 71. In his recent synopsis of finds the coin still appears as an issue of Constans II, for which see Somogyi, "Byzantinische Fundmnzen," 236. 68

PAGE 519

5 19 295. O STR"WEK (Otwock, Poland). "Byzantinische Fundmnze n," 502 03, no. 11. 296. O 69 Brliba and C. Mihai, "Descoperiri monetare la AM 19 (1996): 259 no. 2 and 257, fig. 3/2. monedelor biz antine ," 405. Justin II: K, TES, 569 Brliba and Mihai, "Descoperiri monetare ," 259, no. 3 and 258, fig. 4/3 (wrong date, 575 576). Phocas: K, CON, 603 610; Ibid., 259, no 4 and 258, fig. 4/4. 297. O ZORA T "TIPUSZTA (Fejr, Hungary). Const antine IV: solidus CON, 669 674, found in an Avar grave; Somogyi, Byzantinische Fundmnzen 71 72, no. 56. 298. P ANCIU ( Vrancea, Romania). Justin I: M, CON, 522 527; Butnariu, " 221, no. 109. 299. P ARUTYNE (Mykolayiv, U kraine). 565; Stoliarik Essays on Monetary Circulation 137, no. 42. Justin II: 1 AE, found in the ancient city of Olbia; Stoliarik Essays on Monetary Circulation 139, no. 56. 300. P AVLIVKA (Odesa, Ukraine). Maurice: K, CO N, broken in half; Karyshkovski, Nahodki pozdnerimskikh 81, no. 10. 301. P ECHANKA (Kabardino Balkaria, Russia). Justin II: solidus CON, 565 578, with suspension loop; found in a grave; Kropotkin, Klady vizantiiskikh monet 30, no. 132. 302. P ECHENAYA (Kirovohrad, Ukraine). Heraclius: solidus CON, 625 629 ( MIB Klady vizantiiskikh monet 33, no. 196. 303. P ECICA (Arad, Romania). Justinian I: solidus CON, 542 552; Oberlnder Trnoveanu, antichitate, 69, n. 72. 304. P CS (Baranya, Hungary). Justinian I: tremissis (imitation); Winter "Die byzantinischen und karolingischen," 346, no. 18 and 352, fig. 6/25. 69 Three coins were donated by a student from O place. Some scholars have speculated that the coins were in fact found in the neighboring Byzantine Monede bizantine descope AM 23 24 (2000 2001): 351.

PAGE 520

520 Maurice: tremissi s CON, 583 602, found in the Avar cemetery at Alsmakr grave 1; Somogyi, "Byzantinische Fundmnzen," 279 80, no. 10. Phocas: tremissis (gold plated copper imitation), Ravenna, 602 607, probably from a grave from Gyrvros; Somogyi, Byzantinische Fu ndmnzen, 72, no. 57. 305. P EISCHING (Niedersterreich, Austria). Heraclius: light weight solidus (20 siliquae ), CON, 616 625; Winter, "Die byzantinischen und karolingischen ," 341, no. 6. 306. P ERIENI (Vaslui, Romania). Justin II: 1 AE, NIK, 568 569 nucu bizantine," 357, no. 1, n. 67 and n. 68. 307. P (Dolj, Romania). Justinian I: M, KYZ, 542 543. Butnariu, " 221, no. 110. 308. P ETEA Maurice: 1 coin; Lakatos, "Monede bizantine ," 249, no. 18. 309. P IATRA N Justinian I: 1 AE; Butnariu, " 221, no. 111. Phocas: I, CON; Oberlnder Trnoveanu, " 320, nota 45. 310. P INSK (Brest, Belarus). Constans II: M, Syracuse, 662 668 ( MIB BMC ); Kropotkin, Klady vizantiiskikh monet 38, no. 299. 311. P Justin I: M, CON, 518 527; Poenaru Bordea and Dicu, "Monede romane trzii ," 81, no. 117. Justin II: M, NIK, 570 571; Ibid., 81, no. 119. Phocas: K, KYZ, 603 604; Ibid., 81, nr. 120. Heraclius: M, CON, 613 614; Ibid., 81, no. 121. 312. P IUA P ETRII Justin I: M, CON, 518 527; Oberlnder Trnoveanu, Barbaricum apro piat, 333, n. 12. Justinian I: 4 AE; Butnariu, " 221, no. 112. 313. P (Prahova, Romania). Justinian I: K, ANT, 544 545; Oberlnder Trnoveanu, Barbaricum apropiat, 337, n. 15. 314. P OIANA D ). Justinian I: IS, TES, 538 552 (after MIBE ), found in the area of the early medieval settlement (8 th 9 th c.); Butnariu, " 221, no. 116. 315. P OJEJENA Severin, Romania). Justin I: I, CON, 518 527, found in the ar ea of the Roman castrum ; Butnariu, " 221, no. 117.

PAGE 521

521 316. P TI Justinian I: M, CON; Butnariu, " 225, no. 208. 317. P Justinian I: 1 AE; V. Chir ica and M. Tanasachi, 318. P (Giurgiu, Romania). Maurice: K, CON, 582 583; Butnariu, " 221, no. 118. 319. P OTP ORANJ (South Banat, Serbia) Justin II: M, CON, 571 572; Somogyi, Byzantinische Fundmnzen 73, no. 57a. 320. P (Liberec, Czech Republic). Anastasius I: M, CON, 512 517; Militk, "Finds of the Early Byzantine Coins," 367, no. C4. 321. P Justinian I: solidus CON, 542 565; Militk, "Finds of the Early Byzantine Coins ," 371, no. C13. 322. P RIGOR Severin, Romania). Justinian I: M, CON, 527 537; Lakatos, "Monede bizantine," 252, no. 39. 323. P RIGREVICA (West Ba Heraclius: solidus CON, 632 641; Somogyi, Byzantinische Fundmnzen 73 74, no. 58. 324. P RUNDU (Giurgiu, Romania). Maurice: M, KYZ, 593 594; Butnariu, " 222, no. 121. 325. P Just inian I: M, CON, 540 541; I, after 546; I, Rome, 547 548 or 548 549; 1 AE; "Byzantinische Fundmnzen," 503, no. 12 and 507, fig. 6/3; 503 04, no. 13 and 508, fig. 6/4; 504, no. 14 and 508, fig. 6/5; 504, no. 15. 326. P TUJ (Ptuj, Slovenia). 70 Anas tasius I: 1 AE; Kos, "The Monetary Circulation, 226, no. 7. Justinian I: I, Rome, 547 Salona, 552 565; Kos, Die Fundmnzen vol. 3 497, no. 915. Maurice: tremissis Ravenna, 583 602; Kos, "The Monetar y Circulation, 226, no. 22 Heraclius: M, Sicily, 616 629(?); Ibid., 227, no. 28. Constans II: tremissis Syracuse, 654 662; Kos, Die Fundmnzen vol. 3, 497, no. 916. 327. P ULST (Carinthia, Austria). 70 The coins of Anastasius, Maurice, and Heraclius have also been published in Kos, Die Fundmnzen vol. 2. However, the descriptions were taken from a nineteenth century inventory of finds and must be tr eated with caution. See F. Pichler, Repertorium der steirischen Mnzkunde. II. Die Mnzen der rmischen und byzantinischen Kaiser in der Steiermark (Graz: Leuschner & Lubensky, 1867 ).

PAGE 522

522 Justinian I: tremissis CON, 527 565; Hahn, "Die Fundm nzen des 5. 9. Jahrhunderts 455. 328. P UTNA (Vrancea, Romania Justin II: M, KYZ, 574 575; Butnariu, " 222, no. 122. 329. R J OS (Dolj, Romania). Justin I: M, CON, 518 527; K, KYZ, 518 527; Butna riu, " 222, no. 123 125. Justinian I: 2 AE; Ibid., 222, no. 126 127. Phocas (?): 1 AE; Ibid., 222, no. 128. 330. R (Suceava, Romania). Justin I: 1 AE, Butnariu, " 222, no. 129. 331. R MNICU V LCEA (Vlcea, Romania). Anastasius I: M, 512 517; Oberlnder Trnoveanu, " 55 56, n. 51. Justin I: 1 AE, 518 527; Ibid., 61, n. 63. Justinian I: M, CON, 539 540; Ibid., 68, n. 70. Justin II: 2 AE; Oberlnder Trnovea Maurice: K, ANT, 594 595; Ibid., 133, n. 77. 332. R Heraclius: K, TES, 610 611; Poenaru Bordea and Dicu, "Monede," 79, no. 95. 333. R URENI (Vlcea, Romania). Justin II: M, NIK, 569 570 ; Butnariu, " 222, no. 133. 334. R AVAZD Moson Sopron, Hungary). Heraclius: M, NIK, 612 616; Winter, "Byzantinischen," 346, no. 19 and 352, fig. 6/26. 335. R AZAS (Ogre, Latvia). Justinian I: K, ANT, 539 540; Kropotki n, Klady vizantiiskikh monet, 39, no. 308. 336. R EISBERG (Carinthia, Austria). Anastasius I: K(?), CON, 512 517; Hahn, "Die Fundmnzen des 5. 9. Jahrhunderts ," 455. 337. R (Olt, Romania). Justinian I: solidus CON, 542 edelor bizantine," 222, no. 131 (dated 538 545). Justin II: solidus (Gepid imitation); Oberlnder vremuri," 143, n. 147. M, NIK, 568 222, no. 132. Constans II: K, CON, 647 655; O berlnder 143, n. 150. 338. R ESKO Anastasius: 1 AV; Machajewski, "Skandynawskie," 88.

PAGE 523

523 339. R IFNIK Justinian I: tremissis Rome, 538 542, found in grave 100; L. Bolta, Rifnik pri (Ljubljana: Narodni muzei, 1981), 36 and pl. 17/10; M, CON, 540 541, found behind the apse during archaeological excavations in the early Medieval settlement; Kos, "The Monetary Circulation 226, no. 15. 340. R IVNE (Kirovohrad, Ukraine). Heraclius: solidus CON, 629 631 (dating based on the illustration); Stoliarik Essays on Monetary Circulation, 141, no. 70 and fig. 14/3. 341. R (Dolj, Romania). Justinian I: M, ANT, 554 555; SCIV 23 (1972), no. 1: 146. 342. R OMANOVSKAIA (Rostov na Donu, Russia). Constans II: solidus CON, 661 663, found in a grave dug into barrow 3; A. I. Semenov, kochevnikov Vostochnoi Evropy, Arkheologicheskii sbornik Gosudarstvennogo Ermitazha 29 (1988): 109. Constantine IV: solidus CON, 681 685, found in a grave; Ibid., 90 91. Leontius: solidus CON, 695 698; Ibid., 90 91. 343. R UGINOASA Maurice: 1 AE; Chirica and Tanasachi, Repertoriul arheologic 354. 344. R UPEA 71 Maurice: solidus (imitation); Velter, Transilvania n secolele V XII 298, no. 144. 345. R USTAVI (Kvemo Kartli, Georgia). Justin I: 1 AE found in 1950 during archaeological excavations; Abramishvili, Sakartvelos sakhelmts'ipo muzeumis 123, no. 33. Constans II: 1 AR, found in 1949 during archaeological excavations; Ibid., 127, no. 77. 72 346. S 73 71 See above, no. 176 (Hoghiz Ungra). 72 Kropotkin, Klady vizantiiskikh mo net 45, no. 439 (Constantine III). 73 Cmpie, a larg er neighboring village, which explains why later publications mention two different coins from two different finding spots, when in fact we are dealing with one and the same find. In addition, V. iption (sic), found at which he tentatively attributed to Justinian I, a decidedly dubious attribution which cannot be taken for Repertoriul arheologic 91, XVIII.1. Moreover, the coin was attributed to Justinian I an d later to Phocas, which increased the confusion, some recent inventories mentioning two solidi (one Lakatos, "Monede bizantine bizantine," 247, no. 6A. and 6B; Velt er, Transilvania n secolele V XII 289, no. 24 and 298, no. 146. To make this even more perplexing, A. Zrinyi suggested that the coin was actually found at

PAGE 524

524 Phocas: light weight solidus (20 siliquae ), CON; Oberlnder Trnoveanu, "La 347. S ADON (North Ossetia Alania, Russia). Phocas: solidus CON, 602 610, found in 1929 during archaeological excavations conducted by E. G. Pchelina; Kropotkin, Kla dy vizantiiskikh monet, 31, no. 141a. 348. Heraclius: solidus CON, 616 625; Somogyi, "Byzantinische Fundmnzen," 288, no. 18. 349. S ALCIA ( Phocas: 1 AE; Butnariu, " 222, no. 134. 350. S Justin II: K, KYZ, 576 Topografiia kladov i nakhodok, 85, no. 31. 351. S ANGEORGIU DE C AMPIE Justinian I: solidus ; Lakatos, "Monede bizantine ," 248, no. 20. 352. S NNICOLAUL M ARE Romania). Justinian I: 1 AE; Butnariu, " 222, no. 142. Heraclius: hexagrammon CON, 615 625; Oberlnder Trnoveanu, vremuri, 140, n. 142. 353. S NNICOLAUL R OMN (Bihor, Romania). Justinian I: 1 AE; Lakat os, "Monede bizantine ," 252, no. 40. 74 354. S NPETRU G ERMAN (Arad, Romania). Heraclius: light weight solidus (20 siliquae ), CON, 616 625, found in the grave of an Avar mounted warrior; Somogyi, Byzantinische Fundmnzen, 77, no. 65. 355. S (Vaslui, R omania). Justin I: M, CO N, 518 522, together with sixth century pottery; Butnariu, " 222, no. 136. 75 For the archaeological context, see Teodor, Descoperiri, 144, no. 619. Hu ngarian names of the two settlements (Szabad and Szabd). His argument seems valid since Kurt Horedt had already placed this find at Voiniceni in an early study of early medieval Transylvania after descoperiri arheologice din secolele IV XIII e. n.," Marisia 6 (1976): 144, no. 38; K. Horedt, istoria Transilvaniei n secolele IV XIII (Bucharest: Editura Academiei RPR, 1958), 107, no. 8; K. Horedt, "Avarii n Transilvania," SCIV 7, n. 3 4 (1956): 398, no. 22. 74 Possibly the same coin as no 352 above (Snnicolaul Mare). 75 Reference to previous literature containing illustration sends to an eleventh century follis of Roman III ni. There i s no mention of a sixth century coin in the original archaeological report either, for which see G. Coman, XI n sudul Moldovei (stepa Elan Prut) ," AM 6 (1969): 288 90. In a later article Coman briefly mentioned the find from Justin I without providing any details, for which see G. Coman, "Noi la secolele V XI n partea de sud a Moldovei ," Acta Moldaviae Meridionalis 1 (1979): 93 and 94, fig. 13/1 ( drawing); Teodor, 144, no. 619, attributes the coin to Justinian I instead of Justin I.

PAGE 525

525 356. S ASKHARI (Mtskheta Mtianeti, Georgia). Mau rice: solidus CON, 582 602; Abramishvili, Sakartvelos sakhelmts'ipo muzeumis, 55, no. 131. 357. S AVCI Heraclius: solidus ; Kos, "Monetary Circulation," 227, no. 24. 358. S CHTZEN AM G EBIRGE (Burgenland, Austria). Maurice: K, TES, 591 592; Winter, "Die byzantinischen und karolingischen," 344 45, no. 15. 359. S CURTA Constantin IV: hexagrammon 668 669; Butnariu, bizantine, 222, no. 137. 360. S (Alba, Romania). 76 Justinian I: M, KYZ, 554 555, found in a garden; Butnariu, bizantine, 222, no. 138; Velter, Transilvania n secolele V XII 299, no. 158. 77 Justin II: M, CON, 569 bizantine, 222, no. 139. M, CON, 567 56 8; 78 I, TES, 567 568(?); 79 Popa, " Apulum 12 (1974): 296 298, no. 7 and fig. 1/6. Tiberius II: M, NIK, 580 581; Ibid., 296, no. 8 and 298, fig. 1/6. 80 361. EITIN (Arad, Romania). Heraclius: l ight weight solidus (20 siliquae ), CON, 616 625(?); Oberlnder Trnoveanu, " 139, n. 127. 81 362. EMPETER V S AVINJSKI D OLINI 82 Anastasius I: M, CON, 517 518; Kos, "The Monetary Circulation, 226, no. 6. 76 Coins donated by local collectors, with no precise information about provenance, unless noted. 77 Both Velter and Butnariu took the dating from the description provided in the original publication (regnal year 30=556 557). However, the illustration provided in that publication clearly shows a reverse displaying regnal year 28 (554 555, similar to DOC 51b). The mint is not clear an d it appears both as CON and KYZ in different publications; the style points to the former. Apulum 6 (1967): 627, fig. 2. no. 4 (dated reg nal year 37!). 78 Velter used the original publication without correlating the description with the illustration. The description indicated date 565 566 although it mentioned regnal year 3 (567 568), clearly visible on the illustration as well, for which se 296, no. 5 and 298, fig. 1/5. 79 Attribution was made after Wroth, pl. XI/15 (I, TES, 567 corresponds to 569 570. The state of preservation and the poor quality of the illus tra tion make the dating uncertain. 80 Same coin as Velter, Transilvania n secolele V XII 299, no. 162 and monedelor bizantine," 222, no. 140, with a wrong attributio n to Maurice (M, NIK, 601 602). 81 ea mon edelor bizantine," 223, n. 149. 82 Also published in Kos, Die Fundmnzen vol. 2.

PAGE 526

526 Justin I: M CON, 518 527; Ibid., 226, no. 8. Justinian I: M, NIK, 538 552; Ibid., 226, no. 14. Justin II: M, KYZ, 574 575; Ibid., 226, no. 20. Heracius: M, NIK, 611 612; Ibid., 227, no. 25 26. 363. S ENNAYA (Krasnodar, Russia). Heraclius: M, CON, 610 612; Kropotkin, Klady vizantiiskikh monet 22, no. 19a. 83 364. (Suceava, Romania). Justinian I: 1 AE; Butnariu, " 223, no. 150. 365. ERPENI (Anenii Noi, Moldova). Justin I: K, CON, 518 Topografiia kladov i nakhodok 84, no. 25. 366. S ERPOVOE (Tambov, Russia). Constantine IV: solidus CON, 669 674; solidus CON, 674 681; both pierced at the center; Kropotkin, Klady vizantiiskikh monet 29, no. 126. 367. S FNTU G HEORGHE 84 Justinian I: 1 AV; Velter, Unele 268, no. 163. P hocas: 1 AV; Ibid., 268, no. 90. 368. S HABO (Odesa, Ukraine). Justinian I: M, NIK, 527 538; Stoliarik, Essays on Monetary Circulation 137, no. 39. 369. S HEMOKMEDI (Guria, Georgia). Justinian II: solidus CON, 705 711; Abramishvili, Sakartvelos sakhelmt s'ipo muzeumis (1966 1984) 37, no. 140 and pl. IX/140. 370. S HEVCHENKIVKA (Odesa, Ukraine). Maurice: K, TES, 584 585; Stoliarik Essays on Monetary Circulation 139, no. 58. 371. S HILDA (Kakheti, Georgia) Heraclius: solidus CON, 641; Abramishvili, Saka rtvelos sakhelmts'ipo muzeumis (1966 1984) 25, no. 87 and pl. VI/87. 372. S IBIU (Sibiu, Romania). Constans II: solidus CON, 662 667; P. Prohszka, Beitrge zurm sptrmischen und byzantinischen Goldmnzverkehr zwischen dem 4. und 8. Jahrhundert in Siebenbrgen," CN 12 13 (2006 2007): 91. 83 Kropotkin described the coin as a follis catalogue, plate 46/56, which is in fact an early follis of Her aclius from Constantinople 84 These may in fact be the coins from Voiniceni and Apalina as the original publication by Alexandru Popa cited by Velter does not explicitly mention Sf. Gheorghe in relation with the coins, but most likely refers to coins found archaeological evidence from the upper Mure dating to the early Middle Ages. See A. Popa, ," Marisia 6 (1976): 21. length treatment of coin finds from ted, without verifying the original publication. See Velter, Transilvania n secolele V XII 299, no. 163 164; Repertoriul arheologic 171.

PAGE 527

527 373. S ISAK (Sisak Moslavina, Croatia). Anastasius I: tremissis CON, 492 no. 2. Justinian I: tremissis CON, 527 565; I, Rome, 542 547; Ibid., 149, no. 79 and 166, no. 329. Phocas: K, CON, 603 610; K, Carthage, 606 607; Ibid., 191, no. 670 and 194 no. 699. 374. S LOBOZIA M ARE (Cahul, Moldova). Anastasius I: M, CON, 498 Topografiia kladov i nakhodok, 87, no. 46. 375. S OCHI (Krasnodar Russia). Justinian I: light miliarensis CON, 537 565 (after MIB ); Kropotkin, Klady vizantiiskikh monet 22, no. 22. 376. S OLT T TELHEGY (Bcs Kiskun, Hungary). Heraclius: M, CON, 613 614; Somogyi, "Byzantinische Fundmnzen," 280 81, no. 11. 376a. S OMB OR (Vojvodina, Serbia). Constans II: 1 AE, 652 653; Bna, Review, 295. 377. OMCUTA M ARE Justinian I: so lidus CON, 542 565; IX/10 and pl. XI/2 (enlarged). 85 378. (Alba, Romania). Justin II: solidus found in the Avar cemetery, grave 10; Somogyi, Byzantinische Fundmnzen 78, no. 66. 379. S PRNCENATA (Olt, Romania). Justin II: K; C. Preda, Monede antice descoperite n raza comunei Sprncenata, jud. Olt," SCN 14 (2011): 168, no. 22. 380. S RPSKI K RSTUR (North Banat, Serbia). Heraclius: solidus found in a grave; Somogyi, Byzantinische Fundmnzen 58, no. 42. 381. S T P AUL IM L AVANTTAL (Carinthia, Austria). Justinian I: tremissis CON, 527 565; Hahn, "Die Fundmnzen des 5. 9. Jah rhunderts 455. 382. S TAASDORF (Niedersterreich, Austria). 85 Although the enlarged illustration clearly shows a full weight regular constantinopolitan issue (type MIB 7 ), the authors described it as an imitation after Ravenna (type BMC 37), an attribution accepted uncritically in subsequent publications, for which see Lakatos, "Monede bizantine," 252, no. 43; Butnariu, ter, Transilvania 300, no. 172 (copper imitation !); I. V vestic al Romniei," EN 12 (2002): 222, no. 29.

PAGE 528

528 Anastasius I: M, CON, 517 518; Hahn, "Die Fundmnzen des 5. 9. Jahrhunderts 458. Justin I: M, CON, 522 527; Ibid., 458. 383. S Justinian I: M, NIK, 547 548; Butna riu, " 223, no. 145. 384. S TAPAR Constantine IV: 1 AV; Somogyi, Byzantinische Fundmnzen 78, no. 67. 385. S TRCI Anastasius I: K, CON, 512 517, possibly from a hoard; Poenaru Bord ea and Dicu, "Monede romane trzii," 78, no. 85. 386. S TARODZHERELIYEVSKAYA (Krasnodar, Russia). Phocas: solidus CON, 607 610 (after MIB ), found in a grave; Kropotkin, Klady vizantiiskikh monet 22, no. 25. 387. S TEINBRUNN (Burgenland, Austria). Justini an I: solidus CON, 537 542; Winter, "Die byzantinischen und karolingischen," 345, no. 16 and 351, fig. 6/22. 387a. S TEJANOVCI (Vojvodina, Serbia). Constans II: miliarensis CON, 659 668, found in a grave; Somogyi, Byzantinische Fundmnzen 78 79, no. 68. 388. S TILLFRIED (Niedersterreich, Austria). Justinian I: 12 nummia Alexandria, 527 565; Winter, "Die byzantinischen und karolingischen," 341, no. 7 and 351, fig. 6/20. 389. S TOLNICENI (Vlcea, Romania). 86 Justinian I: 1 AV, found in the area of ancient Buridava; Butnariu, monedelor bizantine, 223, no. 147. 390. S (Brno, Czech Republic). Phocas: M; Militk, "Finds of the Early Byzantine Coins ," 381, no. M36. 391. S TUDENO (Postojna, Slovenia). 87 Anastasius I: 1 AV, CON/TES, 491 518; Kos, "The Monetary Cir culation, 226, no. 4. 392. S TUDINA (Olt, Romania). Justin II: 1 AE; Oberlnder Trnoveanu, " 125, n. 23. 393. (Cazin, Bosnia & Herzegovina). Anastasius I: solidus CON, 507 394. TROVO (Nov Zmky, Slovakia). mince ," 57, no. 3. Justin II : M, during excavations in the N eolithic settlement; Ibid., 57, no. 3. 86 87 Also published in Kos, Die Fu ndmnzen vol. 1.

PAGE 529

529 Heraclius: K, Alexandretta, 608 610; Hunka, "Finds of Byzantine Co ins ," 397, table 1, no. 8. 395. S UCIDAVA (Olt, Romania). Anastasius I: M, CON, 512 518; Butnariu, " 226, no. 1. Justin I: M, CON, 518 522; 3 M, CON, 518 527; K, CON, 518 527; M, CON, 518 527; M, 518 527; K, NIK, 518 527; I bid., 226, no. 2 6; A. Vlcu and E. Nicolae, Monede bizantine descoprite la Sucidava, in Arheologia Mileniului I p. Chr. privind istoria ed. Oscar Print, 2010), 300, no 1 3. Justinian I: solidus CON, 527 538; tremissis CON, 527 565; M, CON, 527 537; M, NIK, 527 538; M, CON, 538 539; K, CON, 538 539; M, NIK, 538 539; K, NIK, 538 539; 2 IS, TES, 538 562; M, CON, 539 540; K, CON, 541 542; M, NIK, 541 542; M, KYZ, 548 549 ; I, CON, 552 553; M, CON, 558 559; I, CON, 562 563; I, CON, 550 565; I, NIK, 559 560; I, CON; M, NIK, 538 565; M; Butnariu, " 226 227, no. 7 23; Vlcu and Nicolae, Monede bizantine 300 01, no. 4 11. Justin II: 2 M, CON, 566 567; K, NIK, 566 567; K, TES, 566 567; K, CON, 568 569; 2 M, NIK, 568 569; 2 K, TES, 568 569; 3 K, TES, 569 570; 2 M, CON 570 571; M, NIK, 570 571; K, KYZ, 570 571; K, ANT, 571 572; M, CON, 572 573; M, NIK, 573 574; K, TES, 573 574; M, CON, 574 575; K, KYZ, 574 575; K, TES, 574 575; M, CON, 575 576; M, ANT, 575 576; K, TES, 575 576; M, CON, 576 577; 2 K, TES, 576 577; CO N, 565 578; K, TES, 565 578; Butnariu, " 227, no. 24 47; Vlcu and Nicolae, Monede bizantine 302 04, no. 12 29. Tiberius II: M, ANT, 578 579; 2 K, CON, 579 582; 2 K, TES, 578 579; K, TES, 579 580; M, CON, 580 581; Butnar iu, " 227, no. 48;.; Butnariu, " 227, no. 48 50. Butnariu, monedelor bizantine, 227; Vlcu and Nicolae, Monede bizantine 304 05, no. 30 34; Oberlnder Trnoveanu, ce de vremuri, 129, n. 57. Maurice: tremissis CON, 578 582; M, CON, 583 584; 2 K, CON, 583 584; K, TES, 583 584; M, CON, 585 586; K, CON, 585 586; K, TES, 585 586; M, ANT, 585 586; 2 M, CON, 586 587; K, CON, 587 588; M, NIK, 587 588; K, TES, 588 589; K, NIK, 598 590; 2 M, NIK, 590 591; M, KYZ, 590 591; K, CON, 591 592; K, CON, 592 593; K, CON, 596 597; M, KYZ, 598 599; 2 K, 582 602; Butnariu, " 227, no. 51 62; Vlcu and Nicolae, Monede bizantine 305 07, no. 35 49; Vlcu, Les monnaies d'or 60, no. 148. 396. S UHULUCENI Heraclius: tremissis CON, 610 613; 88 E. Stoliarik, "Novye nachodki poz dnerimskich i vizantiiskikh monet v Pruto Dnestro Dunaiskoe ., ed. A T Smilenko A V Gudkova A A Kozlovskij (Kiev: Naukova Dumka, 1987), 94, no. 14. 397. S UKHUMI (Abkhazia, Georgia). 88 Stoliarik dated the coin 629 630 by making reference to Tolstoi 28 29 and BMC 93 94, which in fact corresponds to MIB 73, dated 610 613.

PAGE 530

530 An astasius I: 2 M, CON, 512 517; Shamba, Monetnoe obrashtenie 77 78, no. 111 112. Justin I: M, ANT, 522 527; Ibid., 78, no. 113. Justinian I: 1 AE; Kropotkin, Klady vizantiiskikh monet, 45, no. 446; M, NIK, 543 544; 33 nummia Alexandria, 538 565, found dur ing archaeological excavations conducted 544; M, 548 549(?); Shamba, Monetnoe obrashtenie 80 81, no. 121 122, 124 125. Constans II: solidus ; Abramishvili, Sakartvelos sakhelmts'ipo muzeumis 126 27, n o. 75. Tiberius III: solidus CON, 698 705; Ibid., 74, no. 247. 398. S ULEJ"W Justin II: M, NIK, 574 "Byzantinische Fundmnzen," 505, no. 17. 399. ULETEA (Vaslui, Romania). Anastasius I: M, CON, 512 517; Oberlnder Trnovea nu, " 316, n. 22. 400. S ULTANA Sixth century: 1 AE; Butnariu, " 223, no. 148. 401. Justin I: M, CON, 518 527; I. Cndea, "Descoperiri moneta XIX, Istros 1 (1980): 375, no. 2. 402. S VETE G ORE (Bistrica ob Sotli, Slovenia). 89 Justinian I: solidus CON, 538 542; Kos, "The Monetary Circulation, 226, no. 10. 403. S ZADZKO (Stargard, Poland). Heracliu s: hexagrammon CON, 625 "Byzantinische Fundmnzen," 505 06, no. 18 and 511, fig. 6/13. 404. S ZEGED (Csongrd, Hungary). Constans II: solidus CON, 654 659, found in grave 24 from ; miliarensis (imitation), CON; Somogyi, Byzantin ische Fundmnzen 79 82, no. 69 and 71. Constantine IV: solidus (gold plated copper imitation), CON, 674 681 Ibid., 80 81, no. 70. 405. S ZEGVR (Csongrd, Hungary). Maurice: solidus (imitation), after CON, 584 602, found in a grave in Spoldal; Somogyi, B yzantinische Fundmnzen 84 85, no. 75. Heraclius: solidus CON, 616 625, found the Avar grave 761; solidus (imitation), after CON, 616 625, found in grave 855; solidus CON, 616 625, found in grave 873; solidus TES, 616 617, found at Spoldal; Ibid., 82 87, no. 72 74 and 76. 406. S ZEKSZRD T "SZEGI (Tolna, Hungary) Tiberius II: solidus (imitation), after CON, 578 582, found in the Avar cemetery; Somogyi, "Byzantinische Fundmnzen," 282 283, no. 13/1. 89 Also published in Kos, Die Fundmnzen vol. 1.

PAGE 531

531 Heraclius: solidus CON, 616 625, found in the Avar cemetery; Ibid., 282 283, no. 13/2. 407. S (Borsod Abaj Zempln, Hungary). Heraclius: light weight solidus (20 siliquae ), CON, 616 625, pierced; Somogyi, "Byzantinische Fundmnzen," 283 284, no. 14. 408. S ZENTENDRE (Pest, Hungary). Justin II: tremissis CON, 565 578, found in grave 1 or 2; Somogyi, Byzantinis che Fundmnzen 87 88, no. 77. Phocas: solidus CON, 609 610, found in grave 3; Ibid., 88 89, no. 78. 409. S ZENTES (Csongrd, Hungary). Justin II: 1 AE, 572 573; Bna, Review of Somogyi 295 (Szentes Ts). Heraclius: light weight solidus (20 siliquae ), CON, 616 625, found in a grave; Somogyi, Byzantinische Fundmnzen 89, no. 79 (Szentes Jaksor). 410. S Z (Csongrd, Hungary). Anastasius I: miliarensis found in the Tglagyr cemetery, grave XII; J. Cseh et al., Gepidische Grber felder im Theissgebiet II (Budapest: Magyar Nemzeti Mzeum, 2005), 123. 411. T C (Fejr, Hungary). Tiberius II/Maurice: solidus (forgery), obverse: Tiberius II; reverse: Maurice, CON, 582 583, found in grave 7, pierced, in the cemetery at Gorsium; Somogyi Byzantinische Fundmnzen 89 90, no. 80. Heraclius: solidus found at ; Bna, Review of Somogy 295. 412. (Gorj, Romania). Justin I: M, CON, 518 522; Butnariu, " 223, no. 157. Phocas: M, CON, 602 610; Ibid., 223, no 158. 413. PALA (Ialoveni, Moldova). Justinian I: M, CON, 539 540; Stoliarik Essays on Monetary Circulation 136, no. 33. 414. T PLNY Moson Sopron, Hungary). Phocas: light weight solidus (20 siliquae ), CON, 603 607; Somogyi, Byzantinische Fundmnzen 90 91, no. 81. 415. T RGU O CNA ia). Justinian I: 1 AE, KYZ; Butnariu, " 223, no. 154. 416. T T (Komrom Esztergom, Hungary). Heraclius: tremissis (imitation), after CON, 610 625; Somogyi, Byzantinische Fundmnzen 91 92, no. 82. 417. T J OS (Teleorman, Romania). Justinian I: 1 AE; Preda, Descoperiri, 293. 418. T BILISI (Tbilisi, Georgia). Justin I: M, CON, 518 522 found on Gori street; Abramishvili, Sakartvelos sakhelmts'ipo muzeumis (1966 1984) 13, no. 35.

PAGE 532

532 Maurice: M, CON, 585 586; M, A NT, 594 595, found on Gori street; Ibid., 21, no. 74 (Tiberius II) and 23, no. 79. 419. T ECUCI Justinian I: K, CON, 538 542(?); 90 I, NIK, 556 XI," SCIV 21, n. 1 (1970): 113, n. 82 and 121 fig. 9/2 and 9/4. Justin II: M, CON, 574 575; Ibid., 113, n. 82 and 121 fig. 9/3. 91 420. T (Dolj, Romania). Anastasius I: M, 512 517; Oberlnder Trnoveanu, " 55 56, n. 51 Justin II: 1 AE; Oberlnder Trnoveanu, " 125, n. 28. 421. IBUCANII DE S US ania). Justinian I: M, CON, 541 542; Butnariu, " 223, no. 155. 422. (Teleorman, Romania). Justinian I: M, KYZ, 540 541; Butnariu, " 223, no. 156. 423. T Justinian I: M, CON, 543 544; Lakatos, "Monede bizantine ," 252, no. 44A. Justinian I: 12 nummia Alexandria; Ibid., 252, no. 44B. 424. T ISZAGYENDA G "I T" (Jsz Nagykun Szolnok, Hungary). Maurice: solidus CON, 582 583, found in the Gepid cemetery; Somo gyi, "Byzantinische Fundmnzen," 284 285, no. 15. 425. T ISZAKESZI (Borsod Abaj Zempln, Hungary). Justin II: K, TES, 568 569, Somogyi, "Byzantinische Fundmnzen," 285 286, no. 16. Maurice: light weight solidus (20 siliquae ), CON, 584 602; Somogyi, Byzant inische Fundmnzen 92 93, no. 83. 426. T ISZAVASVRI K ASHALOM (Szabolcs Szatmr Bereg, Hungary). Heraclius: 2 light weight solidi (20 siliquae ), CON, 616 625, found in the Avar cemetery, grave 34; Somogyi, "Byzantinische Fundmnzen," 286 287, no. 17. 427. T OLISA (Posavina, Bosnia & Herzegovina). Justinian I: tremissis CON, 537 428. T ONCIU Justinian I: 1 AE; Lakatos, "Monede bizantine ," 249, no. 24. 90 Dating based on the rubbing drawing provided by Teodor. The date cannot be read, but the coin belongs to the large module t ype issued between 538 and 542. 91 Teodor wrongly attributed the coin to Heraclius, although the rubbing drawing of the reverse, the only illustration provided, clearly shows the rev erse of a follis of Justin II, type MIB 43d, with Chi Rho above the large "M." In her monograph Elena Stoliarik imported both the illustration and the wrong attribution; see Stoliarik, Essays on Monetary Circulation, 140, no. 65 and fig. 16/4.

PAGE 533

533 428 a T RKSZENTMIKL"S (Jsz Nagy kun Szolnok, Hungary). Justinian I: 1 AV; Prohszka, Altneue byzantinische Mnzen 107, no. 14. 429. T RAISMAUER (Niedersterreich, Austria). Justin II: M, CON, 568 569; Winter, "Die byzantinischen und karolingischen ," 341, no. 8. 430. T RBINC (Mirna, Sl ovenia). 92 Anastasius I: solidus ; Kos, "The Monetary Circulation, 226, no. 1. 431. T REBUJENI (Orhei, Moldova). Justin I: M, CON, 518 527, found in 1963 during archaeological excavations Topografiia kladov i nakhodok 83 no. 15; Rafalovich, Slaviane 40, fig. 9/3. Jus tinian I: M, CON, found in the early m edieval settlement; Ibid., 83, no. 16. 432. T Justin II: siliqua Carthage, 565 567; Kos, Die Fundmnzen vol. 3, 131, no. 1. 433. T SEBELDA T SIBILIUM (Abkhazia, Georgia). Justinian I: solidus CON, 542 565; heavy miliarensis CON, 527 537; 2 light miliarenses 537 565; 2 M, CON, 527 537; M, ANT, 529 532, found during archaeological excavations at Tsibilium (settlement and cemetery); Iu. N. Vor onov and V. A. Iushin, " Kratkie Soobshcheniia 128 (1971): 103, fig. 43; Shamba, Monetnoe obrashtenie 79 80, no. 115 120; Iu. N. Voronov and O. Kh. Bgazhba, Materialy po arkheologii Tsebel issledovanii Tsibiliuma v 1978 1982 gg.) (Tbilisi: Metsniereba, 1985), fig. 18/2 3 and fig. 113/47. 434. T SIKHISDZIRI (Ajaria, Georgia). Anastasius I: M; K, both coins found during archaeological excavations in the early medieval church; Tsukhi shvili and Depeyrot, History and Coin Finds in Georgia 22. 435. T S INTS KARO (Kvemo Kartli, Georgia) Heraclius: hexagrammon CON, 615 637; Abramishvili, Sakartvelos sakhelmts'ipo muzeumis 126, no. 66. 436. T SKHUMARI (Samegrelo Zemo Svaneti, Georgia). A nastasius I: solidus CON, 491 507; Abramishvili, Sakartvelos sakhelmts'ipo muzeumis 38, no. 42. 437. T UDORA Heraclius: tremissis CON, 610 613; Stoliarik, Essays on Monetary Circulation 141, no. 68 and fig. 14/4. 93 438. T 92 Also publi shed in Kos, Die Fundmnzen vol. 1. 93 Stoliarik cited DOC 54 (dated 613 641) as reference but the illustration of the specimen from Tudora points to an earlier issue from 610 613, type MIB 73.

PAGE 534

534 Justin I: M, CON, 518 527; Gh. Poenaru Bordea and P. I. Dicu, "Monede romane trzii ," 79, no. 96. Justin II: K, CON, 570 571; Ibid., 79, no. 97. 439. T URDA (Cluj, Romania). Justinian I: solidus ; Velter, Transilvania n secolele V XII 302, no. 190. 1 AE; Lakatos, "Monede bizantine ," 249, no. 19. Constans II: 1 AE; Ibid., 249, no. 19. 440. T N N AD V LTAVOU Justinian I: M, CON, 542 543; Militk, "Finds of the Early Byzantine Coins ," 371, no. C12. 440a. U CHKEKEN (Karachayevo Cherkesiya, Russia). Tiberius II: silver imitation; Rtveladze and Runich, "Novye nahodki," 153 54. 441. U LMENI 94 Justin I: K, CON, 522 527; Butnariu, " 223, no. 159. Justinian I: 2 M, CON, 527 538; M, ANT, 536/7 539; I, CON, 556 557; Butnariu, " 223, no. 160 163. Just in II: M, CON, 570 571; M, NIK, 574 575; Ibid., 223, 164 165. 442. U LMI (Giurgiu, Romania). Justinian I: M, ANT, 537 443. U Brliba and V. Butnariu, "Desc operiri monetare din Moldova. I," AM 12 (1988): 319, no. 55. 444. Unknown location, Arad county, Romania. Justinian I: AR miliarensis CON, 527 537; Oberlnder antichitate," 69 70, n. 72. 445. Unknown locations, Armenia. Heracl ius: hexagrammon CON, 615 625; Mousheghian et al., History 199, no. 1. Constans II: hexagrammon CON, 648 652; Ibid., 199, no. 2. 446. Unknown locations, Banat, Romania/ Serbia. 95 Anastasius I: solidus CON, 507 518; 3 M, CON, 512 517; M, CON, 517 518; Ob erlnder Trnoveanu, " 56, n. 52. Justin I: M, CON, 518 527; K, CON, 518 527; M, CON, 527; Ibid., 62, n. 65. Justinian I: miliarensis CON, 527 537; K, CON, 527 537; K, CON, 541 542; M, CON, 544 545; M, KYZ, 544 545; M, CON, 557 558; Ibid., 69 70, n 72. 94 Possibly part of a hoard. 95 Coins from the collection of the 's collection belonged to Zs. Orms, who collected Byzantine coins from the area of Or Veche in the nineteenth century, see Oberlnder also Somogyi, "Byzantinische Fundmnzen," 247 54.

PAGE 535

535 Justin II: 2 solidi CON, 568 578; tremissis CON, 565 578; Somogyi, "Byzantinische Fundmnzen," 292 294, no. 23 25. M, CON, 567 568; M, CON, 570 571; M, CON, 573 574; M, NIK, 575 576; 2 K, TES, 567 568; 3 K, TES, 569 570; 2 K, TES, 574 575; Oberlnd er Trnoveanu, Maurice: M, CON, 586 587; M, KYZ, 597 598; Ibid., 133, n. 84. Heraclius: light weight solidus (20 siliquae ), CON, 616 625; Somogyi, Byzantinische Fundmnzen 27, no. 6; M, CON, 612 613; M, NIK, 613 614; M, CON 615 624; Oberlnder Constans II: M, CON, 643 655; Oberlnder 143, n. 151. 447. Unknown locations, Bosnia & Herzegovina/ Croatia/Serbia. 96 448. Unknown locations, Buco vina, Romania/ Ukraine. 97 Anastasius I: M, CON, 498 518; K, NIK, 498 518; ANT, 512 518; M. Gogu, Suceava," Suceava 26 28 (1999 2001), 293, no. 1 3. Anastasius I/ Just in I: M, CON, 498 527; Ibid., 293, no. 4. Justin I: M, CON, 518 527; K, CON, 518 527; Ibid., 293 94, no. 5 6. Justinian I: M, CON, 527 538; M, ANT, 557 558; Ibid., 294, no. 7 8. Justin II: M, CON, 575 576; K, CON, 574 575; M, NIK, 570 571; M, NIK, 575 576; M, ANT, 572 573; K, TES, 569 570; Ibid., 294 295, no. 9 14. Maurice: 1 solidus CON, 583/584 602; M, CON, 582/583?; M, CON, 591 592; K, CON 583 584; K, CON, 601 602; K, NIK, 584 585; M, ANT, 582 583; M, ANT, 595 596; Ibid., 295 296, no. 15 22. Heraclius: M, CON, 626 627; M, CON, 629 630(?); M, CON, 629 630; M, NIK, 614 615; M, TES, 614 615; 12 nummia Alexandria, 632 641; Ibid., 297, no. 26 31. 449. Unknown location, Carinthia, Austria. Justinian I: K, CON, 527 537; Hahn, "Die Fundmnzen des 5. 9. Jahrhund erts ," 455. 450. Unknown locations, Georgia. 98 96 The collections of the Zagreb Archaeological Museum and the National Museum in Belgrade include some 1350 Early Byzantine coins, most of them without proper information regarding the finding place. The b ulk of the collections is probably made of pieces found in the provinces of Illyricum including the Adriatic coast. Based on the few specimens with a recorded finding place north of the Sava river, it can be speculated that some of the unprovenienced coins may have been found outside the theoretical administrative borders of the Early Byzantine Empire. This is more likely the case of the unprovenienced pieces from the museum in Osijek. Only the coins of Justinian I have been fully published so far, 57 piece s, of which 16 have a known finding place (this inventory no. 36, 113, 219, 290, 483). See, Gricke 1155. For the collections in Zagreb and Belgrade, see Mirnik and Byzantine Coins 51 153. 97 Coins from the collection of the Bucovina Museum in Suceava. Some of the coins might have been found at Callatis in the Early Byzantine province of Scythia during archaeological excavations conducted by Teofil Sauciuc n the 1920s 1930s, see Gogu, "Mone dele bizantine," 285 with n. 8. 98 This entry is based almost exclusively on the two catalogues published by Tamara Abramishvili in 1965 and 1989, repectively. Some of the coins listed with no mention of the finding place in her 1965 catalogue were republished in the 1989 catalogue, this time with indication of provenience. The same might be true for other specimens, which were never republished with full details, although the finding place might be

PAGE 536

536 Anastasius I: I, CON, 498 507; M, CON, 507 512; K, CON, 498 512; 7 M, CON, 512 517; K, NIK, 512 517; Abramishvili, Sakartvelos sakhelmts'ipo muzeumis 38 40, no. 43 48, 50 52; Abramishvili, Sakartvelos sakhelmt s'ipo muzeumis (1966 1984) 10, no. 21 22. Justin I: solidus CON, 518 522; 2 solidi CON, 522 527 (one with a suspension loop); 8 M, CON, 518 522; 2 M, CON, 522 527; CON, 522 527; Abramishvili, Sakartvelos sakhelmts'ipo muzeumis 40 42, no. 53 55, 57 6 2, 65 68; Abramishvili, Sakartvelos sakhelmts'ipo muzeumis (1966 1984) 12, no. 30. Justinian I: 3 solidi CON, 538 552 (one is pierced); light miliarensis CON, 537 565 (pierced); 2 M, CON, 527 537; 2 K, CON, 527 537; 527 537; M, Carthage, 527 537; 2 M ANT, 531 537; 3 centenionales Cherson, 537 552; IS, TES, 538 552; M, CON, 539 540 (pierced); M, ANT, 539 540; I, Carthage, 540 541; M, NIK, 541 542 (pierced); M, ANT, 546 547; M, KYZ, 547 548; M, ANT, 550 551; M, ANT, 551 552; M, CON, 554 555; M, ANT, 5 54 555; M, NIK, 555 556; M, ANT, 557 558; M, NIK, post 538; 4E, CON, 542 565; Abramishvili, Sakartvelos sakhelmts'ipo muzeumis, 43 50, no. 69 71, 73, 75, 77 104. Justin II: solidus CON, 567 578; M, KYZ, 567 568; K, TES, 568 569; K, TES, 569 570; M, NIK, 5 70 571 (pierced); K, KYZ, 570 571; M, CON, 571 572; 99 M, CON, 572 573; M, ANT, 572 573; M, CON, 574 575; K, CON, 574 575; M, CON, 575 576; 2 K, TES, 575 576; M; Ibid., 50 52, no. 105 107, 109 120. Tiberius II: solidus CON, 578 582; M, NIK, 580 581; Ibid., 53, no. 122 123. Maurice: solidus CON, 583 602; Abramishvili, Sakartvelos sakhelmts'ipo muzeumis (1966 1984), 21, no. 75 and pl. IV/75. Solidus CON, 583 584; solidus CON, 583 602; light miliarensis CON, 583 602; M, KYZ, 582 583; M, CON, 583 584; K, CON 587 588; M, NIK; M, ANT, 589 590; M, ANT, 591 592; 2 M, ANT, 592 593; M, ANT, 593 594; M, KYZ, 593 594; M, ANT, 594 595; 2 M, ANT, 595 596; K, Cherson, 590 593; M, Cherson, 592 595; M, CON; M, NIK; Abramishvili, Sakartvelos sakhelmts'ipo muzeumis, 53 58, no. 124, 132 145, 148 151. Phocas: tremissis CON, 603 610 (pierced); K, KYZ, 603 604; M, ANT, 604 605; M, CON, 605 606; M, ANT, 605 606; M, KYZ, 607 608; M, ANT, 607 608; M, CON, 608 609; Ibid., 61 62, no. 170, 172 173, 176 180. Heraclius: 2 solidi ; Tsot selia, Coin Finds in Georgia 142, no. 757. 100 Solidus CON, 613 616; solidus CON, 639 641; 2 hexagrammata CON, 615 625; 3 hexagrammata CON, 625 629; 2 hexagrammata CON, 632 635; hexagrammon CON, 637 641; M, CON, 611 612; M, CON, 612 613; M, NIK, 612 61 3; M, CON, 615 616; M, CON, 613 616; M, KYZ, 617 618; M, CON, 629 630; M, CON, 631 632(?); K, CON, after 629; 3 M, CON; M; K (possibly imitations); Abramishvili, Sakartvelos sakhelmts'ipo muzeumis, 63 68, no. 185, 188 194, 196 197, 199 204, 206 211, 213 21 4; Abramishvili, Sakartvelos sakhelmts'ipo muzeumis (1966 1984), 33, no. 127. recorded in museum acces sion records. In addition, in her 1965 book Abramishvili used the reference catalogues of the day BMC Tolstoi and even Sabatier so the coins had to be redated and sometimes reattributed based on the more recent MIB corpus. 99 Found in Svaneti region. 100 The coins were found in an unknown location in Svaneti region and may belong to a dispersed hoard.

PAGE 537

537 Constans II: solidus CON, 662 667(?); solidus CON, 667 668(?); miliarensis CON, 648 652; 3 hexagrammata CON, 642 647; 3 hexagrammata CON, 648 652; 3 hexagrammata CON, 654 6 59; hexagrammon 659 668; K, Carthage, 641 642; K, Carthage, 642 647; M, CON, 643 644(?); M, CON, 644 645(?); M, Syracuse, 662 667; M, CON, 666 668; M; Abramishvili, Sakartvelos sakhelmts'ipo muzeumis, 68 72, no. 212, 215 221, 223 224, 226 228, 230, 232 23 7. Constantine IV: solidus CON, 669 674; 2 solidi CON, 681 685; hexagrammon CON, 674 681; Ibid., 73, no. 240 241, 243 244. Justinian II: solidus CON, 692 695; solidus, CON, 705 711; Ibid., 73 74, no. 245 246. Early Byzantine: K; Abramishvili, Ibid., 61 no. 175. 451. Unknown locations, Hungary. 101 Maurice: tremissis Rome, 583 602; Somogyi, "Byzantinische Fundmnzen," 291 292, no. 22. Heraclius: solidus CON, 610 613; 2 solidi CON, 613 616; 3 solidi CON, 616 625; light weight solidus (20 siliquae ), CON, 616 625; solidus Ravenna, 637 640; solidus CON, 641; solidus (imitation), after CON, 613 616; Somogyi, Byzantinische Fundmnzen 99 102, no. 89 91 and 103 107, no. 94 97. Solidus CON, 616 625; Somogyi, "Byzantinische Fundmnzen," 290 291, no. 21. Light weight solidus (20 siliquae ), CON, 616 625; Bna, Review of Somogyi 295. Constans II: solidus CON, 651 654; Somogyi, Byzantinische Fundmnzen 102 103, no. 92. Justinian II: 1 AV; Prohszka, Altneue byzantinische Mnzen 108, no. 16. 452. Unknown lo cation, Krasnodar, Russia. Constans II: 1 AR; Kropotkin, Klady vizantiiskikh monet 21, no. 2. 453. Unknown locations, Moldova. Anastasius I: 1 AE, 498 Topografiia kladov i nakhodok 87, no. 1. Justinian I: 1 AE; Ibid., 87, no. 1. Justin I I: M, CON, 575 576; Ibid., 88, no. 2. 454. Unknown locations, Moravia, Czech Republic. Justin II: K, NIK, 572 573; Militk, "Finds of the Early Byzantine Coins," 381, no. M33. Late Roman / Early Byzantine: solidus ; Ibid., 381, no. M35. 455. Unknown locatio 102 Justinian I: 2 solidi ; M; Lakatos, "Monede bizantine," 248, no. 15; M, CON, 541 542; Velter, Transilvania 294, no. 96. Justin II: K, TES, 575 576; Velter, Transilvania n secolele V XII 294, no. 98. Maurice: solidus CON, 584 602; Somogyi, Byzantinische Fundmnzen, 65 66, no. 50. 101 Unprovenienced coins from the colection of the National Museum in Budapest and other local museums in Hungary 102 Two solidi of Justinian and Maurice, respe ctively, are illustrated in Zrinyi, "Repertor iul arheologic," pl. LXXV/b c.

PAGE 538

538 Heraclius: M, NIK, 612 613; Velter, Transilvania n secolele V XII 295, no. 100. 456. Unknown locations, North Ossetia Alania, Russia. 103 Justinian I: solidus. Justin II: solidus. Tiberius II: 1 AE. 4 57. Unknown locations, Oltenia, Romania. Anastasius I: 6 M, 512 517; Oberlnder Trnoveanu, " 55 56, n. 51. Justin I: tremissis CON, 518 527; M, 518 527; K, 518 527; K, 527. Ibid., 61, n. 63 and n. 64. Justinian I: AR miliarensis TES, 527 BSNR 86 87 (1992 19 93): 123 24. M, CON, 527 538; M, CON, 545 546; M, NIK, 556 557; IS, TES, 552 557; Oberlnder 68, n. 70. Solidus CON, 545 565, pierced; M, CON, 539 monedelor bizantine," 224, no. 185 18 6. Justin II: 28 AE; Oberlnder n. 33. Tiberius II: M, NIK, 580 581; Ibid., 130, n. 62. Maurice: K, CON, 582 583; K, NIK, 582 583; M, CON, 583 584; K, CON, 586 587; K, CON, 588 589; M, ANT, 590 591; K, K YZ, 592 593; K, CON, 595 596; K, TES, 599 600; Oberlnder Phocas: M, NIK, 606 607; Ibid., 137, n. 109. Heraclius: 2 solidi CON, 613 638; M, NIK, 610 611; 2 M, CON, 612 613; M, CON, 614 615. Ibid., 138, n. 116, 117, 119 and 120; V. Petac, Descoperiri inedite de monede BSNR 86 87 (1992 1993): 319, no. 3 Constans II: hexagrammon CON, 659 668; Oberlnder Trnoveanu, de vremuri, 145, n. 174. 458. Unknown locations, Sloveni a. Anastasius I: solidus CON, 507 518; Kos, Die Fundmnzen vol. 3, 582, no. 2. Justinian I: solidus CON, 527 537; solidus CON, 542 565; siliqua Ravenna, 537 552; Ibid., 582, no. 3 4 and 583, no. 2. 459. Unknown locations, former Tekovsk stolica reg ion, Slovakia. Justin I: M; E; Fiala, "Byzantsk mince ," 57, no. 2. Justin II: M; K; Ibid., 57, no. 2. 460. Unknown locations, Transylvania, Romania. Justinian I: solidus ; Velter, Transilvania n secolele V XII 301, no. 184. Justin II: solidus CON, 567 5 78; Somogyi, Byzantinische Fundmnzen 76, no. 63. Constantine IV: semissis CON; Ibid., 75 76, no. 62. 461. Unknown locations, Wallachia Romania. Justin I: solidus CON, 522 527; Butnariu, " 224, no. 181. 103 Kropotkin, Klady vizantiiskikh monet 31, no. 143.

PAGE 539

539 Justinian I: M, CON, 527 537; Oberlnder Trnoveanu, Barbaricum apropiat, 337, n. 15. Tiberius II: M, ANT, 581 582; Ibid., 351, n. 37. Maurice: solidus CON, 583 602; Ibid., 353, n. 40. 462. U RBNISI (Shida Kartli). Justin I: K, found in 1959 during archaeological excav ations conducted by Nikoloz Berdzenishvili; Abramishvili, Sakartvelos sakhelmts'ipo muzeumis, 123, no. 33a. 463. U RICANI Justinian I: M, CON, 544 545; Butnariu, " 223, no. 166. 464. U ROI (Hunedoara, Romania). Ju stinian I: M, KYZ, 541 542, found in the local cemetery ; Butnariu, monedelor bizantine, 223, no. 167 465. U RZICA M ARE (Dolj, Romania). Justinian I: M, ANT, 559 560; Oberlnder Trnoveanu, " 68, n. 70. 466. V 104 Justinian I: solidus ; Lakatos, "Monede bizantine ," 250, no. 26. Heraclius: 2 solidi ; Ibid., 2 50, no. 26. 467. V (Olt, Romania). Justin I: solidus CON, 518 527; Butnariu, " 224, no. 171. 468. V ADUL LUI I SAC (Cahul, Moldova). Anastasius I: CON, 512 518; Stoliarik Essays on Monetary Circulation, 132, n o. 2. 469. V ALEA M ARE Justinian I: K, ANT, 564 565; C. Preda, Descoperiri, 294, no. 5. 470. V ALEA S TANCIULUI (Dolj, Romania). Justin II: 1 AE, CON; Butnariu, " 223, no. 168. 471. V ALEA V OIEVOZILOR Justinian I: solidus CON, 538 " 223, no. 169. 472. V Justin II: M, KYZ, 573 574; Butnariu, e, 223, no. 170. 473. V ANADZOR (Lori, Armenia). 104 In 1891 the National Hungarian Museum in Budapest acquired a "hoard" of Roman imperial denari and Late Roman, Early Byzanti ne and Byzantine gold coins spanning some one thousand years. Most probably the coins either belong to separate hoards or fragments of hoards and stray finds were mixed in this collection.

PAGE 540

540 Justi nian I: M, CON, 543 544; Mousheghian et al., History 180, no. 2. 474. V ARDENUT (Aragatsotn, Armenia). Justin II: M, ANT, 576 577; Mousheghian et al., History 167. 475. V ARN Jus tin I: tremissis 518 527; Zbojnk, "Antike Mnzen ," 413, no. 24. 105 476. V RPALOTA (Veszprm, Hungary). Heraclius: M, CON, 612 613, found in the Gymnasium cemetery in grave 229; Somogyi, Byzantinische Fundmnzen 93, no. 84. 477. V (Orhei, Moldov a). Maurice: M, CON, 589 Topografiia kladov i nakhodok 83, no. 14. 478. V Justinian I/ Tiberius II: 3 AE; Teodor, "Descoperiri ," 164, no. 733. 479. V ASYLIVKA (Odesa, Ukraine). Justinia n I: K, NIK, 538 539; Stoliarik Essays on Monetary Circulation 136, no. 30. Maurice: M, CON, 589 590; Ibid., 139, no. 57. 480. V ERBOVYI L OG (Rostov na Donu, Russia). Justinian II: solidus CON, 705 711, found in a grave; A. Naumen ko and S. I. Bezuglov, j biznci s irni importleletek a Don Mra Ferenc Mzeum vknyve. Studia archaeologica 2 (1996): 250. 480a. V ERKHNII C HIRYURT (Dagestan, Russia). Maurice: 2 solidi CON, one is pierced, both have loops attached, found in graves 6 an d 17, respectively; S. V. Gusev, Severo t (Moskow: Institut etnologii i antropologii RAN, 1995), 11 12, no. 1 2 and 46, fig. 1/1 2. Heraclius: solidus CON, 616 625, loop attached, found in gra ve 16; 3 solidi (imitations), one is pierced, two have loops/lugs, found in graves 20, 40a, and 79a, respectively; Gusev, Severo Vostochnyi Kavkaz 12 14, no. 3 5, and 7 and 46 47 fig. 1/ 3 and fig. 2/1 3. Constans II: solidus (imitation), CON, 654 659 wi th welded lugs, found in grave 14 ; Ibid., 14, no. 8 and 47, fig. 2/4. 481. V (Hunedoara, Romania). Justinian I: solidus found in the early medieval settlement, no archaeologic al context; Lakatos, "Monede bizantine ," 250, no. 27. 482. V IENNA (Wien, Austria). antinischen und karolingischen ," 343, no. 12b/1. Tiberius II: M, CON, 578 579, found in a grave during archaeological excavations in 1910; Ibid., 342, no. 11. 105 Same coin as Prohszka, "Altneue byzantinische Mnzen," 107, no. 15.

PAGE 541

541 Maurice: 2 12 nummia Alexandria, 590 602; CON, 582 602; 106 Ibid., 342 43, no. 11b/1, 12a, and 12b/2. Heraclius: M, NIK, 625 629; Ibid., 342 43, no. 11b/2. 483. V INKOVCI (Vukovar Syrmia, Croatia). Justinian: M, CON, 539 540; M; Gricke and 1155, no. 59. 484. V IRGEN O BERMAUERN (Tyrol, A ustria). Justin I: tremissis CON, 518 527; Hahn, "Die Fundmnzen des 5. 9. Jahrhunderts ," 462 485. V Justinian I: M, CON, 541 542; Papasima, "Monede bizantine inedite," 280, no. 4. 486. V OINICENI Phocas: l ight weight solidus (20 siliquae ), CON, 603 607; Somogyi, Byzantinische Fundmnzen 94, no. 85. 487. V OLGOGRAD (Volgograd, Russia). Justinian I: 1 coin; Kropotkin, Klady vizantiiskikh monet 29, no. 122. 488. V OLOS KE Maurice: M, Cherson, 582 602, found during archaeological excavations at the mouth of river Sura; Kropotkin, "Novye nakhodki ," 172, no. 28. 489. V OSKETAP (Ararat, Armenia). Justin II: M, ANT; Kropotkin, Klady vizantiiskikh monet 43, no. 386 (Shirazlu). 490. V C (South Banat, Serbia). Justin II: solidus CON, 568 578; Somogyi, Byzantinische Fundmnzen 94 95, no. 85a. 491. V YLKOVE (Odesa, Ukraine). Justinian I: 1 AE; Stoliarik Essays on Monetary Circulation 136, no. 31 (M, according to Kropotkin, Klady vizan tiiskikh monet, 35, no. 242). 492. V YNOHRADNE (Kherson, Ukraine). Anastasius I: 1 AE; Stoliarik, Essays on Monetary Circulation 133, no. 8. 493. V YSOK Justin II: K, ANT (?), 565 578; Militk, "Finds of the Early Byzantine Coins," 381, no. M34. 494. W IENER N EUSTADT (Niedersterreich, Austria). Constans II: M, Syracuse, 659/662 668; Winter, "Die byzantinischen," 341, no. 9. 495. Y EREVAN (Yerevan, Armenia). Heraclius: hexagrammon CON, 625 629; Mousheghian et al., History 197. 496. Y OSYPIVKA (Kirovohrad, Ukraine). 106 Might be the same coin found in the Botanic Garden and attributed to Justinian I.

PAGE 542

542 Heraclius: solidus CON, 616 625, found in grave 6; O. S. Beliaev and I. O. Molodchikova, "Pokhovannia k ochyvnikov na r. Orel," Arkheolohiia 28 (1978): 86, fig. 2/4. 107 497. Z AGREB (Zagreb, Croatia). Justin II: M, CON, 570 571; K. Simoni, "Zagreb i okolica u ranom srednjem vijeku," in 56. 108 498. Z AIM Anastasius I: 1 AE, 498 518; Stoliarik Essays on Monetary Circulation 133, no. 6. 499. Z AMRD R T (Somogy Hungary). Heraclius: light weight solidus (20 siliquae ), CON, 616 625, found in grave 1392; Somogyi, Byzantinische Fundmnzen 95 96, no. 86. 500. Z ATOKA (Odesa, Ukraine) Justin II: K, KYZ; Stoliarik, Essays on Monetary Circulation 139, no. 54. 501. Z ELEMR (Hajd Bihar, Hungary). Phocas/Heraclius: tremissis (imitation), obverse: Heraclius, 610 613; reverse: Phocas, CON, 602 610; Somogyi, Byzantinische Fundmnzen 96, no. 87. 502. ELOVCE Early Byzantine: 1 AV, found in the Avar age cemetery (second half of the seventh century), grave 170; Zbojnk, "Antike Mnzen ," 413, no. 28. 502a. Z HABOTIN (Cherkasy, Ukraine). Heraclius: light weight solidus (20 si liquae), CON, 639 641, with suspension loop. Smedley, Seventh Century, 113, no. 2 (attribution and dating); Kropotkin, Klady vizantiiskikh monet 37, no. 274 (description and context). 503. Z HEBOT A (Mtskheta Mtianeti, Georgia). Heraclius: hexagrammon CON, 635 637, found in a grave, during archaeological excavations; Abramishvili, Sakartvelos sakhelmts'ipo muzeumis 126, no. 74. 504. Z IMNE (Volyn, Ukraine). Early Byzantine: 1 silvered imitation of a bronze issue, found during archaeological excavations in the early medieval settlement; Kropotkin, Klady vizantiiskikh monet 31, no. 147; V. V. Aulikh, vii St. n.e. v Zakhidnii Volyni ( Kiev, "Naukova dumka", 1972 ), 20. 505. Z IMNICEA (Teleorman, Romania). Maur ice: M, CON, 588 589; Butnariu, " 224, no. 175. 506. Z MIYNYY I SLAND (Ostriv Zmiyinyy, Ukraine). Anastasius I: 1 AE; Stoliarik Essays on Monetary Circulation 132, no. 5; 107 a good indication that it had been worn as a pendant. 108 Same c Byzantine Coins," 177, no. 488.

PAGE 543

543 527; Ibid., 134, no. 16 and 256, fig. 21/1. Justinian I: M, CON, 527 538; Ibid., 136, no. 32 and 256, fig. 21/2. Constans II: M, Cherson, 654 668; Ibid., 141, no. 66 and 256, fig. 21/3 (Heraclius). 109 507. Heraclius: light weight solidus (20 siliquae ), CON, 616 "Byzantinische Fu ndmnzen," 506, no. 19 and 511, fig. 6/14. 508. Z OLOTONOSHA (Cherkasy, Ukraine). Justinian I: M, ANT, 539 540; Kropotkin, Klady vizantiiskikh monet, 37, no. 275. Maurice: 1 coin; Kropotkin, Klady vizantiiskikh monet, 37, no. 276. 110 509. Z SANA C IPRUSSZK (B cs Kiskun, Hungary). Heraclius: solidus CON, 610 613; Somogyi, "Byzantinische Fundmnzen," 288 89, no. 19. 510. Z UGDIDI (Samegrelo Zemo Svaneti, Georgia). 41.85 42.5 8 0 3 13 Justin I: heavy miliarensis CON, 518 527; Abramishvili, Sakartvelos sakhelmts 'ipo muzeumis, 40 41, no. 56. 111 Constans II: hexagrammon CON, 648 652; 112 hexagrammon CON, 659 668; Abramishvili, Sakartvelos sakhelmts'ipo muzeumis (1966 1984), 35 36, no. 133. 511. YWIEC Justinian I: M, 541 "Byzantinisc he Fundmnzen," 506 07, no. 20. A.2.2 Hoards/ Collective Finds 512. A RNOLDSTEIN (Carinthia, Austria). Hoard of 15 solidi, of which only 6 survive, including an imitation after a solidus of Justinian. Last coin: a light weight solidus (22 siliquae ) of Just in II, Ravenna, 571 572. Hahn, "Die Fundmnzen des 5. 9. Jahrhunderts 454. 513. A RTSVABERD (Tavush, Armenia). A hoard of 102 Byzantine silver coins was found in 1967 during agricultural works. The coins were hidden in a ceramic container. All coins are hexagrammata of Heraclius. Last coin: 635 637. Mousheghian et al., History 193 94. 514. B M An unknown number of gold coins of Maurice, now lost; Somogyi, Byzantinische Fundmnzen 26, no. 4. 109 Stoliarik attributed the coin to Heraclius but the rubbing drawing of the reverse provided by the author clearly shows an issue of Constans II from Cherson, type MIB 227. 110 Five other coins found at Zolotonosha are mentioned by Kropotkin, but no precise details are provided. There is no indication that they might have belonged to a larger hoard. The Justinian follis was found accidentally by a schoolboy who was digging a hole, at 40cm into the gro und. 111 Same coin as Kropotkin, Klady vizantiiskikh monet 44, no. 408a, with attribution to Justinian I. 112 Same coin as Abramishvili, Sakartvelos sakhelmts'ipo muzeumis 70, no. 222 and pl. XIV/222.

PAGE 544

544 515. B ERNECEBARTI (Pest, Hun gary). Accidental find of 17 gold coins, possibly from a grave, information is available only for one coin, Phocas, solidus CON, 603 607 Somogyi, Byzantinische Fundmnzen 29 30, no. 10. 516. B ICH VINTA P ITSUNDA (Abkhazia, Georgia). A hoard of 55 penta nummia all from the reign of Justin I, was found in 1961 during archaeological excavations at Pitsunda, near the north wall XIII. Tsukhishvili and Depeyrot, History and Coin Finds in Georgia 72 73, no. 4. 517. B IE LSKO B (Bielsko Hoard found accidentally in 1921. The 23 coins spanning the reigns of Vespasian through Justinian II were hidden in a ceramic pot; the hoard is now lost and only a brief description survives Last coin: 685 695 or 705 7 11. "Byzantinische Fundmnzen," 498 99, no. 3. 518. B I (Koln, Czech Republic). Accidental find of a small hoard containing silver and bronze coins from Trajan Decius t o Justin II. Last coin: 573 574. Militk, "Finds of the Early Byza ntine Coins ," 374, no. C17. 518a. B UCHAREST (Bucharest, Romania). A small hoard of 12 copper coins from Justin II and Tiberius II found in 1929. No details are known and the hoard may be incomplete. Last coin: 580 581. Trsors 182, no. 83 (same as 414, no 350). 113 519. ERVEN H RDEK (Koln, Czech Republic). Accidental find of a small hoard of 20 bronze coins of which only seven survive, from Maximianus to Justin I. Last coin: 518 527. Militk, "Finds of the Early Byzantine Coins ," 368 69, no. C7. 520. C HEGEMSKY Di strict (Kabardino Balkar Republic, Russia). In 1898 eight gold coins from Anastasius to Heraclius were acquired from a local collector, along with two casts after a solidus of Constantine IV and an eighth century solidus respectively. The coins may have b een part of a larger hoard. Kropotkin, Klady vizantiiskikh monet 30, no. 133. 521. C HIBATI (Guria, Georgia). A large hoard of solidi was found by accident in September 1958. The coins were hidden in a ceramic pot. Kropotkin, Klady vizantiiskikh monet, 46 no. 475, citing the initial 1959 report by D. G. Kapanadze, mentions 2000 coins of which ca. 800g had been melted down. Only 124 coins were subsequently recovered by the Georgian museum in Tbilisi. The most recent and detailed publication mentions 121 co ins from Tiberius II to Heraclius. Many specimens are die linked. Last coin: 610 613. T. 113 The hoard was first mentioned by S. McA. Mosser, A Bib liography of Byzantine Coin Hoards (New York: American Numismatic Society, 1935) 15 after a note received from Constantin Moisil stating that the hoard had been found in Bucharest in 1929. Later, Moisil published the coins stating that the provenience was unknown. The hoard was donated by an important collector in Bucharest, Constantin 49 53 (1944): 112.

PAGE 545

545 Abramishvili, " in Akad. S. Janashias 25 B (1968): 159 76; Tsukhish vili and Depeyrot, History and Coin Finds in Georgia 75 79 and pl. 2 6. 522. C HKHOROTSQU (Samegrelo Zemo Svaneti, Georgia). Two solidi of Justinian, possibly belonging to a larger hoard, were found in the village of Chkhorotsku in 1952. Abramishvili, Sak artvelos sakhelmts'ipo muzeumis 124, no. 42. 523. C UDALBI Accidental find of 28 bronze coins from Anastasius I and Justin I. The coins had been deposited in a ceramic pot covered with a stone slab to avoid spilling. Last coin: 522 527. Trsors 414 15, no. 351. 524. D EDOPLISTS KARO T S ITELITS KARO (Kakheti, Georgia). A large hoard of Sasanian and Byzantine silver coins was found accidentaly during agricultural works in 1977. The hoard was initially dispersed but a large number of coins were subsequently retrieved, 1385 Sasanian drachms and te n hexagrammata of Heraclius. Last Byzantine coin: 637 641. M. Tsotselia, History and Coin Finds in Georgia: Sasanian and Byzantine Coins from Tsitelitskaro (AD 641) (Weteren: Moneta, 2002), 86 87. 114 525. D NIPROPETROVS K region (Ukraine). A necklace with 72 Byzantine coins was found accidentaly by a peasant during agricultural work which led to the destruction of a female barrow. Four coins have been described as imitations, while the necklace was described as being composed of 72 gold plated coins, which ra ises the posibility that all coins are imitations/forgeries. The prototype was described as a semissis of Constans II, in fact a light weight solidus of 20 siliquae of that emperor (dated 642 646 after MIB III ). Kropotkin, Klady vizantiiskikh monet, 38, 31 no. 149; K. V. Golenko, " Vizantiiskii Vremennik 11 (1956): 292 94. 526. D Accidental find of 20 Byzantine gold coins, of which some were attached to a gold necklace. The coins were found in 1902 at 1m into the ground among wide slabs of stone displayed horizontally, a possible indication of a grave. Teodor, Descoperiri 81, no. 255. 115 114 For an early book length publication of a portion of the hoard then available for study, see I. Dzhalagania, Monety klady Gruzii: klad sasanidskikh i vizantiiskikh monet iz Tsiteli Tskaro (Tbilisi: Metsniereba, 1980). Some coins were also mentioned in Abramishvili, Sakartvelos sakhelmts'ipo muze umis (1966 1984) 26, n. 142 and 26 33, no. 90, 93, 106, 120 126. 115 Dan G. Teodor alone dates the coins to the fifth and sixth centuries. In the original publication Constantin Moisil mentioned twenty Byzantine gold coins without any indication of their da ting. The current location of the coins is unknown and they were likely divided among finders back in 1902. Moisil himself received the information from the local priest. It is uncertain how Teodor suddenly advanced a dating to the fifth and sixth centurie s more than eighty years after the original publication. Moreover, in his 1972 inventory Constantin Preda was noting that the date of the hoard is not known, for which see Preda,

PAGE 546

546 527. D (Vlcea, Romania). Hoard of hexagrammata found in unknown circum stances, only three hexagrammata of Constans II were retrieved. Last coin: 659 66 bizantine," 230; Somogyi, Byzantinische Fundmnzen 131. n. 34 (for MIB attribution and dating). 528. D VIN (Ararat, Armenia). A) A mixed h oard of Sasanian and Byzantine silver coins was found accidentaly in a field close to Duin (1955). Most of the coins were subsequently acquired by the Yerevan Museum, in several stages which created some confusion regarding the exact number of coins in the hoard. The total seems to be around 300, approximately one third being hexagrammata of Heraclius, most of which are dated 625 629. Mousheghian et al., History 107 08 and pl. 9. B) A hoard of 115 hexagrammata of Heraclius and his co regents was found clos e to the south eastern wall of the cathedral (1947). Last coin: 638 641. Ibid., 131 33 and pl. 9. 529. E CHMIAZDIN (Armavir, Armenia). A hoard of four Byzantine hexagrammata from Heraclius and Constans II with no details of the date or the circumstances in which it was found. Last coin: 648 652. Mousheghian et al., History 170 and pl. 25. 116 530. E NGUR (Tukums, Latvia ). Accidental find of 43 coins, one of which (the latest) is a solidus struck for Anastasius I. 137. 117 531. F (Harghita, Romania). Acci dental find of a large number of gold coins ( ca 300), now dispersed. Fourteen specimens have been preserved in two different museums. Last coin: solidus of Heraclius, 616 625. Somogyi, Byzantinische Fundmnzen 40 42, no. 24. 532. G Accidental find of 12 hexagrammata from Heraclius to Constantine IV. The hoard was found on the Danube's bank together with ceramic fragments from the container where the coins had been deposited. Last coin: 674 681. Trsors 168, no. 68; Somogyi, Byzan tinische Fundmnzen 128. n. 21 (for MIB attribution and dating). 533. G RCHI (Tavush, Armenia). dac)," BSNR 20, July December (1913): 63, no. 24. 116 Another hoard of ca. 20 30 Byzantine silver coins of Heraclius and Heraclius Constantine was found in Echmiazdin in 1908. A. Pakhomov was able to study only ten specimens. Since no information exists about the circumstances in which the second hoard was found it is possible that we are in fact dealing with one and the same hoard. For the early hoard, see Kropotkin, Klady vizantiiskikh monet, 43, no. 387. 117 The coin of Anastasius was also mentioned by Kropotkin, Klady vizantiiskikh monet, 39, no. 313.

PAGE 547

547 A small hoard of 10 12 silver coins of Heraclius was found hidden in a ceramic pot (1942). Information was retrieved only for one specimen, a hexagrammon of Heraclius and Heraclius Constantine, CON, 632 635. Mousheghian et al., History 194 95. 534. G ROPENI Accidental find of 47 copper coins from Justinian I and Justin II deposited in a ceramic pot close to Danube's bank. Last coin: 577 578 Trsors 417 18, no. 354. 535. G YUMRI (Shirak, Armenia). Accidental find of a mixed hoard of 108 Sasanian and Byzantine silver coins. The Byzantine hexagrammata span the reigns of Heraclius and Constans II. Last Byzantine coin: 654 659. 118 Mousheghian et al., History 182 83; M. I. Kamera and K. V. Golenko, Leninakanskii klad sasanidskikh i vizantiiskikh monet (1956 g.), VV 19 (1961): 172 193; Mousheghian et al., History 182 83 and pl. 25. 536. H ELLMONSDT (Obersterreich, Austria). A small hoard of bronze coins, of which description exists for four specimens from Maurice to Justinian II. Last coin: 685 695. Hahn, "Die Fundmnzen des 5. 9. Jahrhunderts," 458 59. 537. H LINSKO (Chrudim, Czech Republic). Accidental find of a group of seven silver and br onze coins from Vespasianus to Justin II deposited in a small ceramic vessel. Last coin: 572 573; Militk, "Finds of the Early Byzantine Coins ," 373 74, no. C16. 538. H alloy pitcher of a type well spread throughout the Eastern Mediterranean containing 57 coins from Justin II to Maurice. The hoard also includes a bronze chain and 13 scrap pi eces of bronze sheet. Last coin: 597 598. Trsors 418 Unele 99 127. 539. H RADEC K RLOV (Hradec Krlov, Czech Republic). Accidental find of a group of coins, of which on ly two survive; the early Byzantine specimen is a dekanummium of Justinian I, CON, 540 541; Militk, "Finds of the Early Byzantine Coins ," 370 71, no. C11. 540. H ROZOV (Bruntl, Czech Republic). Small hoard of four bronze coins found during archaeologica l excavations on the left bank of the Hrozov River; all coins are from Carthage, the earliest being a Punic coin from the third century BC. Last coin: 662 667. Militk, "Finds of the Early Byzantine Coins ," 382, no. M38. 541. I Turkey). A hoard containing an unknown number of hexagrammata of Heraclius and 118 Mousheghian dated the latest coin of Constans II to 648 651/2 although reference is given to MIB 150, which corresponds to 654 659. The illustration provided on pl. 25/16 confirms that the coin indeed belongs to type MIB 150.

PAGE 548

548 17 specimens, hexagrammata dated between 615 and 629. Kropotkin, Kla dy vizantiiskikh monet 42, no. 370. 542. I L ICH (Krasnodar, Russia). peninsula. The coins were found among amphora remains, possibly the container in which the coins had been de posited. The hoard includes 140 gold staters of third and fourth century kings of Bosporus as well as five solidi of Justinian I. Last coin: 542 565. N. A. Frolova and E. Ia. Nikolaeva, "Il'ichevskii klad monet 1975 g.," VV 39 (1978), 173 79. 543. K ARLINO A hoard of 6 solidi Scandinavian bracteates, and fragments o f gold jewelry. The latest coin was struck for Anastasius. dziejw. VI wieku," in ed. H. 544. K ARSIB"R (Drawsko, Poland). A hoard of 25 solidi, the latest of which was struck for Anastasius. "Znaleziska monet rzymskich z terenw Pomorza przechowywanie w zbiorach Muzeum Kulturalno Historycznego w Stralsundzie," 43 (1999): 176. 545. K ELEGEIA (Kherson, Ukraine). Accidental find of seven solidi from Heraclius and Consta ns II. The hoard was found in 1927 on the left bank of the Dnieper and included coins and jewelry made of gold and silver, fragments of silver plate, and fragments of two glass containers. One solidus of Heraclius is an imitation. A light weight solidus (2 2 siliquae ) of Justinian I, found in the vecinity might have been part of the same hoard. Last coin: 641 646. A. I. Semenov, "Vizantiiskie monety Kelegeiskogo kompleksa," Arkheologicheskii sbornik Gosudarstvennog Ermitazha 31 (1991): 121 30; Kropotkin, Kl ady vizantiiskikh monet 37, n o 268 ( Kelegeyskiye Khutora ). 546. K HOTYN (Chernivtsi, Ukraine). Accidental find of three bronze coins from Anastasius I to Justinian I, which may have belonged to a larger hoard. I. Corman, pruto nistrian n epoca evului mediu timpuriu (sec. V VII d.Chr.) 2. 546a. K ISLOVODSK (Stavropol, Russia). A short note was published about "dozens of folles of Justinian I" found in Kislovodsk in the area of the furn iture factory. "Na karavannoi trope," (1973), no. 3: 76. 547. K LUK (Nymburk, Czech Republic). 119 119 Possibly part of the Pod brady hoard, below no. 569. and both hoards wer e described as having been found in the sand. Moreover, in both cases the coins existence of two different hoards at such a small distance given the high den sity of finds in the Central Bohemian Region, on both sides of the Elbe.

PAGE 549

549 Accidental find of a group of coins, now dispersed, in a large flooded sand quarry; only one co in has been retrieved, Phocas, K, Carthage, 606 607; Militk, "Finds of the Early Byzantine Coins ," 375, no. C22. 548. K OBULETI (Ajaria, Georgia). A hoard of gold coins was found accidentally in a kitchen garden; four solidi of Justinian I minted in Const antinope were retrieved. Kropotkin, Klady vizantiiskikh monet 45, no. 444 (Smekalovka). 549. K M (Cieszyn, Poland). An unknown number of coins were part of the hoard, although it is possible that we are dealing in fact with two separate hoards; among the coins, a follis of Justinian I, now lost. "Byzantinische Fundmnzen, 501, no. 8. 550. K OSH (Aragatsotn, Armenia). Accidental find of 52 or 54 hexagrammata of Heraclius and Constans II. Last coin: 648 652. Mousheghian et al., History 164 65 and pl. 23 24. 120 551. K TSCHACH L AAS (Carinthia, Austria). A hoard of 20 gold coi ns, of which only four are preserved, solidi of Justinian I, CON, 542 565. Hahn, "Die Fundmnzen des 5. 9. Jahrhunderts 454. 552. K UPUSINA A large hoard of solidi, from Zeno to Phocas, now lost. Trsors 420, no. 357. 55 3. L UHANS K Accidental find of an unknown number of gold coins of Justinian I on the premises of a factory in Luhansk. The find w as recorded ca. 1899 in a private collection, but no other details are known. Kropotkin, Klady vizantiiskikh monet 35, no. 232. 554. L YMARIVKA Accidental find of a Sassanian silver bucket probably containing a hoard of gold coins, of which only one was mentioned, a small gold coin of Justin I. Kropotkin, Klady vizantiiskikh monet 35, no. 231. 555. M AGRANETI (Mtskheta Mtianeti, Georgia). A mixed hoard of eleven Sasanian and four Byzantine silver coins of Heraclius was found in 1967 during archaeological excavations in a seventh century building destroyed by fire. Four more hexagrammata of Constans II were retrieved from the same building during the 1968 campaign and may well be part of the same hoard. 121 120 Mousheghian dates the latest coins of Constans II to 647 648 although reference is given to MIB 144, which corresponds to 648 652. Illustration on pl. 24 confirms that the coins ind eed belong to type MIB 144. The same holds true for the other finds of Constans II belonging to this type, such as the single find from Garni or the specimens from the Ech miadzin and Stepanavan hoards. 121 The coins of Constans II were not mentioned by Mede a Tstotselia who preferred to treat them as two different hoards, see Tsotselia, History and Coin Finds in Georgia: Sasanian Coin Finds and Hoards (Weteren: Moneta, 2003), 74 75, with year 1968 instead of 1967 for the initial find; Tsotselia, Coin Finds in Georgia 141, no. 749 and 148, no. 784, mentioning only three hexagrammata of Constans II with inaccurate reference to Abramishvili. The 1977 publication by Abramishvili, with full catalogue, illustration,

PAGE 550

550 They belong to types MIB 144, 145 and 150, dated 648 652 and 654 659, respectively. T. Abramishvili, Klad monet iz Magraneti, in Numizmaticheskii sbornik: posviashchaetsia pamiati D. G. Kapanadze ed. V. A. Lekvinadze (Tbilisi: Metsniereba, 1977), 73 82 and pl. VII. 556. M AISTROV ( Zaporizhia, Ukraine). Accidental find of a few hundred gold coins of the Heraclian dynasty, hidden in a ceramic pot on an island on the Dnieper River; only one coin was retrieved, a solidus CON, 632 639 ( MIB in, Klady vizantiiskikh monet 31 32, no. 159. 557. M ALO P ERESHCHEP YNE (Poltava, Ukraine). In 1912 a large hoard of Byzantine and Sassanian gold and silver plate and jewelry was found near the village, in the area of an early medieval cemetery. The hoard also contained some 70 early Byzantine gold coins from Maurice to Constans II. The bulk of the hoard is made of light weight solidi of 20 siliquae of Heraclius and Constans II, with a dating no later than 646. I. V. Sokolova, Monety Pereshchepinskogo kla da, in Sokrovishcha Khana Kubrata ed. O. Fedoseenko (St. Petersburg: AO Slaviia, 1997), 17 41. 558. M ALECHOWO epizod ," 57. 559. M OVILENI Accidental find of 26 bronze coins hidden in a ceramic pot on the bank of Siret river. Last coin, 613 614. Trsors 421 22, no. 358. 560. M RZEZINO (Puck, Poland). Hoard of 150 solidi, accidentally found in 1795. 130 solidi were struck for Anastasius monet z Mrzezina (Gm Numizmatyczne 42 (1998): 59 60. 561. M TSKHETA (Kartli, Georgia). Hoard comprised of nine silver coins of Heraclius ( hexagrammata ) found accidentaly in 1902. Three specimens are preserved in the Georgian National Museum, all dated 625 629. Kropotkin, Klady vizantiiskikh monet, 44, no. 425; Abramishvili, Sakartvelos sakhelmts'ipo muzeumis (1966 1984), 30, no. 111 113 and pl. VII/111 112. 122 562. N EKRESI (Kakheti, Georgia). and description of the archaeological context is perhaps the most reliable. The coins were also mentioned in her 1989 catalogue of Byzantine coins from the Georgian National Museum, for which see Abramishvili, Sakartvelos sakhelmts'ipo muzeumis (1966 1984) 35 36, no. 132, 135 137. 122 No. 111 is the same coin as the one published by Abramishvili, Sakartvelos sakhelmts'ipo muzeumis, 65, no. 195 and pl. XIII/195, who does not mention the finding place. The remainding two hexagrammata from the Mtskheta hoard could not be identified in Abramishvili s 1965 cata logue, as none of the weights or illustations provided match the ones in th e 1989 catalogue (no. 112 113).

PAGE 551

551 A small hoard of three silver coin s, two Sasanian drachms of Hormizd IV (579 590) and an issue of Maurice, was found during archaeological excavations conducted in Nekresi in 2003. Tsotselia, Coin Finds in Georgia 142, no. 755. 563. N OKALAKEVI (Samegrelo Zemo Svaneti, Georgia). A) A hoar d of 23 light weight solidi (23 siliquae ) of Maurice was found in 1930 during archaeological excavations at Nokalakevi. Most of the coins are die linked. Tsukhishvili and Depeyrot, History 74, no. 5 and pl. 1; T. Ia. Abramishvili, Nokalakevskii klad, VV 23 (1963): 158 165. B) A hoard of nine copper coins from Anastasius to Justin II was found during archaeological excavations in 1975. Tsotselia, Coin Finds in Georgia 135, no. 715. 123 564. O (Vaslui, Romania). Acidental find of 34 bronze coins from Phocas to Constans II. Last coin, 652 657. Trsors 422 23, no. 359. 565. O CHAMCHIRE (Abkhazia, Georgia). In 1958 the Georgian State Museum acquired a group of 58 Byzantine coins whi ch had been found hidden in a ceramic pot. The find itself, however, dated back to 1903 so the information is not entirely reliable. The eight early Byzantine coins date from the reigns of Justin I to Maurice. Last coin: 601 602. T. Abramishvili, Ochamchi reshi Akad. S. Janashias sachelobis 24 B (1963): 57 59 and pl. I/1 6. 124 566. O DISHI (Samegrelo, Georgia). A hoard of 28 silver coins was found accidentally in 1966 together with remains of a ceramic pot; only 13 coins have been recovered, two siliquae of Maurice and eleven hexagrammata of Heraclius. Last coin: 625 629. Tsukhishvili and Depeyrot, History and Coin Finds in Georgia 81, no. 8 and plate 6; Abramishvili, Sakartvelos sakhelmts'ipo muzeumis (1966 1984), 22, n. 120 and 22 31, no. 76 77, 88, 94 102, 119. 567. P IUA P ETRII (Ialomi An unknown number of hexagrammata were probably part of a hoard found before Byzantine fortress of Carsium, located on the right bank, in the Byzantine province of Scythia. Information has been retrieved for three coins of Constans II and Constantine IV. Last coin: 668 673 (according to Oberlnder Trnoveanu, Barbaricum apropiat, 357 58) or 674 681 (acording to Somogyi, New, 116). M. Pauker, 123 The coin finds resulted from archaeological excavations at Nokalakevi in the 1970s and early 1980s had been published already by Tamara Abramishvi li, with no mention of a hoard (see above, no. 267). Moreover, no such hoard can be identified in the catalogue of Byzantine coins from the Georgian National Museum published in 1989 by the same author, Abramishvili, Sakartvelos sakhelmts'ipo muzeumis (196 6 1984) 124 The 58 coins acquired by the museum most likely represented just a fraction of the initial hoard. In addition, the collection spans more than five centuries, from Justin I to Nicephorus III Botaniates and clearly does not resemble the tradition al structure of Byzantine copper coin hoards. There is a conspicuous gap of some 300 years between the last Early Byzantine coin (Maurice) and the next one in the chronological sequence of the group (Leo VI), so we are most likely dealing with at least two different, a nd probably incomplete, hoards.

PAGE 552

552 135 136 (July December 1945): 53. 125 568. P LUMBUITA Mixed hoard of some 35 copper coins from the second to the sixth century; the bulk of the hoar d is made of Late Roman issues. Only two early Byzantine coins are part of the hoard, the latest being a dekanummium of Justinian, NIK, 556 557. Trsors 423, no. 360. 126 569. P (Nymburk, Czech Republic). Accidental find of a small hoard of bronze c oins, now dispersed, seven coins have been preserved, from Justin I to Constans II. Last coin: 651/652 656/657; Militk, "Finds of the Early Byzantine Coins ," 376 77, no. C25. 570. P ODGORNENSKII (Rostov na Donu, Russia). Collective finds in cemetery IV, during archaeological excavations. A) In a grave dug into barrow 14: solidus of Constans II and solidus of Constantine IV (661 663). Naumenko and Bezuglov, j biznci," 247. B) In a grave dug into barrow 2: Three solidi of Constantine IV, four solidi of J ustinian II, and two solidi of Leontius (695 698). Ibid., 247 48. 571. P RAGUE (Prague, Czech Republic). Accidental find of 26 bronze coins in the Nebu residential area: Roman imperial, Late Roman, and early Byzantine coins, from Augustus to Justinian I. Last coin: 540 541. Militk, "Finds of the Early Byzantine Coins ," 369 70, no. C10. 572. P RISEACA (Olt, Romania). Accidental find of 141 hexagrammata from Constans II and Constantine IV and two silver earrings hidden in a ceramic pot. Last coin: 674 6 81. B. Mitrea, "Date noi cu privire la secolul VII. Tezaurul de hexagrame bizantine de la Priseaca (jud. Olt)," SCN 6 (1975): 113 25. 573. R ABKA Z DR"J (Nowy Targ, Poland). Two silver light miliarenses of Justinian I, CON, 527 537, possibly belonging to a larger hoard have been found accidentaly during repair works on the road linking Rabka and Mszana Dolna; both coins belong to the same type and were struck with the same pair of di es, judging by the illustration. "Byzantinische Fundmnzen," 504 05, no. 16 and 508, fig. 6 6 7. 574. R 125 The coins were originally mentioned by Mina Pauker, but no reference to a hoard was made. She had acquired a larger group of coins found at Piua Petrii, including a hoard of Roman denarii early Roman bronze c oins, as well as Byzantine issues. Among them there were three hexagrammata attributed after Sabatier and often confused in subsequent publications mentioning these coins. They might indeed be part of a hoard, but this is only a matter of conjecture relyin g on the unsual presence of three hexagrammata on the same site. 126 V. Butnariu expressed certain doubts regarding the hoard, arguing that the large number of ancient coins would be too unusual in a sixth century hoard. However, a significant number of sim ilar hoards found in the Balkans, as well as in barbaricum Plumbuita. See Butnariu, " 200, 204.

PAGE 553

553 Accidental find of some 200 silver and copper coins hidden in a ceramic pot. The hoard was dispersed and only nine coins have been studied, spanning some six hundred years, from Nero to Heraclius. Last coin: 610 611. Poenaru Bordea and Dicu, "Monede romane trzii ," 79. 574 a S ALOVO (Rostov na Donu, Russia). 42.4666 47.145 8 0 3 13 574a Collective find of three solidi two of Leontius (695 698) and one of Tiberius III (698 705), in a grave dug in barrow 2 of the cemetery IV at Salovo. All coins are pierced at vostochnoevropeiskikh stepei I puti slozheniia kul'tury Khazarii," in Stepi Evropy v epokhu srednevekov'ia. Khazarskoe vremia ed. A. V. Evgelevskii (Donetsk: Institut Arkheologii NAN Ukrainy/Donetskii Nacional'nyi Universitet, 2001), 172 174 with fig. 4. 575. S ARACHILO (Kvemo Kartli, Georgia). A hoard of ten hexagramma ta of Heraclius and Heraclius Constantine was found in the local church in 1979. All coins belong to the same type, dated 625 629. Tsukhishvili and Depeyrot, History 82, no. 9 and plate 7 (Bolnisi); Abramishvili, Sakartvelos sakhelmts'ipo muzeumis (1966 1 984), 26, n. 144 and 26 30, no. 91, 103 105, 107 110, 114 115. 576. S AUR M OGIL SKIY (Donets'k, Ukraine). he ground. All coins belong to the same type, light weight solidus of 20 siliquae CON, 542 562. Kropotkin, Klady vizantiiskikh monet 36, no. 253 (Beloyarovka). 577. EICA M (Sibiu, Romania). Hoard found accidentaly in 1853 on the bank of Trnava Mare river comprising ca 50 100 gold coins and gold jewelry (bow fibula and earing with polyhedral pendant). Only 36 coins have been studied, issues from Theodosius I I to Justin I. Last coin: 518 527. P. Somogyi, Der Fund von Kleinschelken (Siebenbrgen, 1856) im Lichte neuentdeckter Archivdaten, in Byzantine Coins in