Early Byzantine coin circulation in the Eastern Provinces

AJN Second Series 21 (2009) pp. 151–226 © 2009 he American Numismatic Society Early Byzantine Coin Circulation in the Eastern Provinces: A Comparative Statistical Approach Andrei Gândilă* Number is the ruler of forms and ideas and the cause of gods and demons Pythagoras (Taylor 1818: 78) his paper addresses two major issues in the study of early Byzantine coinage. First, the statistical validity of large public collections of Byzantine copper coins is established as a reliable indicator of coin production. Second, based on the rhythm of coin output inferred from the evidence of the public collections, a comparison is attempted between coin inds in the three major geographical areas of the Eastern Empire: the Balkans, Anatolia, and SyriaPalestine. his comparative analysis reveals a great deal of regional variation, but also common patterns in coin circulation. As early as the irst half of the nineteenth century antiquarians and collectors became interested in developing means of organizing and systematizing Byzantine coin series. he chief concern was to create “suites monétaires,” an attempt to gather all the known coin types issued by the Byzantine state. he 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 1911.1 However, the standard work for more than ity years * andrei.gandila@ul.edu. University of Florida, History Department. 1. de Saulcy (1836); Soleirol (1853); Sabatier (1862); Tolstoi (1912–1914). Equally important are Mionnet’s second volume of De la rareté et prix des médailles romaines (1827), the second part of Christian homsen’s collection, Erslev (1873), as well as the contribution to the classiication of coins of Justinian I by Friedländer and Pinder (1843). 151 152 Andrei Gândilă 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 ill the gaps in the Byzantine coin series and to provide scientiic access to an entire collection, including duplicates.2 he breakthrough made ater 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.3 Alfred Bellinger and especially Philip Grierson embarked on the task of reassessing many of the old datings and attributions in what became a seminal work for our understanding of early Byzantine coinage (hereater EBC). Cécile Morrisson went a step further by cataloguing the old and important collection of the Bibliothèque Nationale in Paris, enriched with the donations of leading scholar-collectors such as Gustave Schlumberger. Equally signiicant have been the major private collections of Byzantine copper coins made available to the scientiic world, such as those of Rodolfo Ratto and George Bates.4 Creating a corpus of all known Byzantine coin inds, as the latter has pleaded for in the introduction to his catalogue, 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 past ity years due to extensive archaeological research, oten 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 universities in Western Europe contribute to the wealth of EBC available for study.5 he stupendous task of assembling all known Byzantine coin types was attempted by Wolfgang Hahn in his series Moneta Imperii Byzantini.6 However, few initiatives have been taken towards a statistical understanding of the monetary circulation at the scale of the entire empire,7 although the use of quantitative tools was promoted and employed for assessing local provincial patterns in coin circulation.8 he use of statistical tools in Byzantine numismatics is largely a post-war development. he growing interest in elaborate means of quantiication lies both 2. Wroth (1908). 3. Bellinger and Grierson (1966–1968). 4. Ratto (1930); Bates (1981). 5. Most important are the collection of the Hunter Coin Cabinet in Glasgow (Bateson and Campbell 1998), the collection Köhler-Osbahr from the Duisburg Museum (Althof 1998–1999), the collection of the University of Göttingen (Sommer 2003), and the collection of the Bottacin Museum in Padova (Callegher 2000). 6. Hahn (1973–1981). 7. Morrisson (2002). 8. Metcalf (1964); Pradwic-Golemberski and Metcalf (1963); Pottier (1983); Morrisson and Ivanišević (2006); Noeske (2000). Early Byzantine Coin Circulation in the Eastern Provinces 153 in the need for a more complex method of analyzing the increasing number of coins and in the introduction of computer-based programs which facilitated such an approach. Mathematical tools have always been employed in numismatics; by necessity, coins needed to be counted and classiied based on chronological and typological criteria, but no attempt was made to analyze them statistically.9 D. M. Metcalf has been a pioneer in this respect.10 His study of Byzantine coins in Sirmia and Slavonia represents the irst elaborate attempt to use statistics in order to understand the EBC circulation in that region. 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, Antioch, and Sardis.11 Starting from the early 1980s one can observe an explosion of studies employing more or less sophisticated statistical tools. he main impetus had been provided by the organization of a Round Table dedicated to the use of statistics in numismatics, in which reputed numismatists and professional statisticians collaborated for a better implementation of statistics in numismatic research.12 Overtime, the work diversiied 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 hessalonica, represent early attempts to determine Byzantine mint output.13 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 C. Morrisson for small issues such as the joint reign solidi of Justin I and Justinian I and gold issues of Carthage, respectively.14 he 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. Consequently, students in the ield of Byzantine numismatics have been less engaged in the lively debate of the last two decades centered on the question of 9. Bell (1916); Edwards (1933); Mosser (1935); Bellinger (1938); Waage (1952); hompson (1954). 10. For an early methodological essay see Metcalf (1958). 11. Metcalf (1960). 12. Carcassone and Hackens, eds., (1981). For the statistical method see also Carcassone and Guey (1978); Carcassone (1987). 13. Metcalf (1969); Metcalf (1976). 14. Metcalf (1988); Morrisson (1981); Morrisson (1988). Although removed from the chronological focus of this article, the comprehensive die-study done by Füeg (1991) on the eighth century gold issues remains seminal for a general understanding of Byzantine gold coinage even if extrapolations can be problematic. 154 Andrei Gândilă mint output.15 Steps have been taken, however, to understand the metrology of the multi-denominational system of Byzantine coinage. he 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 study of Byzantine coinage but also for the monetary circulation in Syria, based on comparative statistics.16 In the past two decades, a number of statistic methods in Byzantine numismatics have been used in studies dealing with a substantial sample of coins, either hoards or stray inds. 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 appendix of statistical results, the monumental trilogy of Hans-Christoph Noeske on 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 collaborative work coordinated by C. Morrisson, V. Popović, and V. Ivanišević on the coin hoards found in the Balkans and in Anatolia.17 At the same time, studies of stray and single inds from major archaeological sites have included statistical analyses of recent inds and also previously published material.18 I. T S R  M  he purpose of this study is twofold: irst, it attempts to identify general luctuations in the production of base-metal coinage, based on the evidence of the major public collections, and secondly, it draws a series of comparisons between coin inds in the major geographical units of the Eastern Empire. his study deals exclusively with copper coins issued in the major Eastern mints: Constantinople, hessalonica, Nicomedia, Cyzicus, and Antioch. he reasons behind this decision are both technical and practical; most of the EBC inds originate in the eastern part of the empire, from the Danube to the Eastern Mediterranean provinces, while the monetary system itself was not standardized throughout the Empire. Egypt was largely self-suicient, Italy and the African mint at Carthage had diferent rhythms of coin production than the Eastern mints, and the ratio between gold and copper seems to have had regional particularities.19 herefore, in order to insure the accuracy of statistical parameters and ultimately of the historical conclusions drawn 15. Esty (1986); Buttrey (1993, 1994, and 1997); Duncan-Jones (1994) with an extensive review by W. E. Metcalf (1995); Callataÿ (1995). 16. Pottier (1983). See more recently Pottier (2004). 17. Curta (1996); Noeske (2000); Morrisson et al. (2006). 18. Marot (1998); Sheedy (2001); Butcher (2003); Evans (2006); Gândilă (2008). 19. Hahn (1973–1981). Early Byzantine Coin Circulation in the Eastern Provinces 155 from this material, Western provinces, including North Africa were let out. In quantitative terms, this means working with more than 10,000 bronze coins located in ive major collections (hereater ‘museum collections’)20 and another c. 8,000 from the major urban centers of the Eastern Empire . he single inds resulting from systematic archaeological research are unevenly distributed over the geographical area under consideration. he material from the Balkans is by far the most abundant, followed by the Near East, where numerous Syro-Palestinian sites have been excavated. Christopher Lightfoot has sketched the current state of the Byzantine research in Anatolia and drew attention to the lack of interest for Byzantine archaeological layers in favor of the presumably more sophisticated classical period.21 Recent research by Zeliha Demirel Gökalp has shown that the Turkish archaeological museums preserve a wealth of EBC found in Anatolia, which awaits publication.22 Although a few tentative steps have been taken towards a broad understanding of the coin circulation in the Balkans and the Middle East, 23 little has been done with respect to gathering the numismatic material for an in-depth comparative analysis, partly because of the still insuicient evidence. he case of the border province of Scythia is a unique situation, given that Romanian numismatists have constantly published comprehensive catalogues of recent inds and museum collections,24 bringing the total number of EBC inds to a dazzling igure of more than 3,000 specimens. 20. he collections under consideration are Dumbarton Oaks and the Whittemore collection (DOC); Bibliothèque Nationale in Paris (BNP); the collection of the British Museum (BMC); the collection Köhler-Osbahr in Duisburg (KOD); and the vast collection of the American Numismatic Society (ANS), still unpublished. hey were chosen based on size and on the preponderance of copper issues, including numerous duplicates. he collections of Tolstoi and Ratto are not included in the statistical 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 signiicance for the understanding of mint output by looking at the coin types they were unable to ind in their desire to assemble the entire Byzantine coin series. 21. Lightfoot (2002). 22. I wish to express my gratitude to Zeliha Demirel Gökalp for allowing me to consult her unpublished PhD dissertation and two other unpublished catalogues of the Byzantine coins from the Malatya and Bolu museums. 23. Morrisson (1989a); Morrisson (1998); Walmsley (1999); Gândilă, (2008). 24. See the chronicles of the recent inds compiled by B. Mitrea and Gh.Poenaru Bordea from Dacia, “Découvertes de monnaies antiques et byzantines en Roumanie” along with the ones regularly published in Pontica by Gh. Papuc, R. Ocheșeanu, G. Custurea, A. Vertan and G. Talmațchi under the title “Cronica descoperirilor monetare din Dobrogea.” While the coins published in the Romanian journals were mainly single inds resulting from archaeological research, in Bulgaria such articles are devoted exclusively to hoards, see T. Gerasimov’s series “Kolektivni nahodki na moneti” from Izvestiia na Arkheologicheskiia Institut. 156 Andrei Gândilă he relevance of site inds has been a debated issue. Philip Grierson has argued that the structure of site inds tends to favor the smaller coins because they were easily lost and not retrieved and thus they cannot ofer a completely reliable image of their circulation.25 What seemed acceptable at the time when Grierson was suggesting such an interpretation of site inds is no longer tenable in light of the material coming from the Balkans, Anatolia and the Near East. His conclusions were chiely based on the major centers of the early Byzantine Empire, Corinth, Athens, Sardis, Antioch, and Constantinople where excavations had yielded a large number of small denominations. he idea that small coins were more easily lost due to their size seemed perfectly reasonable both because of the structure of inds and a sort of natural logic suggesting that the smaller the coin (and the lesser the value) the higher the probability of it being lost.26 his, 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 ½-folles.27 he argument that excavators missed the smaller coins cannot be a valid explanation given the wealth of fourth–ithth century material recovered from the same archaeological sites, many coins being half the size of a Byzantine pentanummium. 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 assessing 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 for the circulating base metal coinage. Even if the method was proitable it could only point to the potential number of issues. he real number is efectively 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 inlow of fresh currency, to pay the army and the administrative apparatus, and to insure the collection of taxes. he 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 deinitive tool in assessing absolute coin production volumes.28 he devastating criticism of T. V. Buttrey in a period when such applications were lourishing has precisely the merit of pointing to problems with this model. Although the discussion was centered on Greek and 25. Grierson (1986, 42). But see Metcalf (1969, 94), who argued that “folles […] stand the same chance of being lost.” 26. his interpretation held as a general applicable rule is still advocated, especially for sites in the Middle East. See, for example, Sheedy (2001, 5). 27. Gândilă (2008, 318); Sheedy (2001, 44); Bellinger (1938, 95–119); Marot (1998, 322); Demirel Gökalp (2007). 28. See n. 15. Early Byzantine Coin Circulation in the Eastern Provinces 157 Roman coinage, his arguments were generally applicable to any ancient coin series. he skepticism regarding igures drawn for gold and silver series turns into total despair in the case of copper issues characterized by large die-populations and high wear factors due to intensive circulation.29 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 surviving dies, we would still be nowhere near having a great understanding of the phenomenon of coin production. What is needed, and is unfortunately illusory, is the absolute numbers for the varieties of denominations, mints, oicinae and dates in a given period. he only approach capable of producing relevant statistical igures, insofar as they can be determined with our current body of knowledge, is one based on the coin sample at our disposal. he statistical representativity of museum collections irst came into discussion in the 1950s when scholars were trying to make sense of the transformations that brought the once prosperous Byzantine Empire into a “Dark Age.” Using numismatic material from the British Museum, Alexander Kazhdan argued 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.30 George Ostrogorsky, on the other hand, using the same material from the British Museum, showed that the gold coinage, which he took to be more important than the base currency, in fact witnessed an important numerical increase during the same period. Furthermore, he introduced new data in the equation by analyzing two other major collections, those in Leningrad and Washington.31 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 ater that, Grierson, perhaps the highest authority in Byzantine numismatics at the time, completely refuted the concept that such collections could ever project a realistic picture of the money in circulation at any given time.32 His position, reiterated in the following decades, was based 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.33 he 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 29. See also Hendy (1985, 7–8). 30. Kazhdan (1954). 31. Ostrogorsky (1959) 32. Grierson (1961, 445–446); Grierson (1967a, 323–324). 33. Grierson (1986, 38–39). 158 Andrei Gândilă statistical representativity of public collections: “Les grandes collections, malgré le nombre considérable des pièces qu’elles possèdent, ne constituent donc pas un échantillon représentatif de la masse monétaire à un moment donné.”34 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 of duplicates. Such coins elude the parameters set by the reputed Byzantinist for the gold series. 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 greatly 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.”35 he major collections included in the present analysis have the advantage of being heterogeneous with respect to geographical sources of origin. Each of them in fact reunites smaller collections gathered at diferent times and in diferent places, and it is reasonable to suppose that they cover the entire Eastern Empire, albeit perhaps unequally so. he museums have usually kept records of their purchases, visible in the catalogues’ footnotes or more extensively in introductions revealing the historical background of the collection. Nevertheless, it is hard to trace back the mechanisms of gathering a particular collection.36 It is rather a “detective’s quest” and one is oten faced with having to sleuth back in time as far as the age of Enlightenment. Famous collectors have been recently discussed by scholars and more information on their life and collections has been brought to light.37 However, irst hand accounts of ind spots are hard to obtain even for current acquisitions given the discretion of many collectors and their providers, let alone for those almost a century old.38 Another methodological issue involves so-called “collector behavior” and is related to a more signiicant issue, namely that of establishing whether the sample of coins in various collections is representative of the total mass of coins produced in ancient times. To what extent can we safely trust the statistical results based on specimens from private collections? How much is the collectors’ preference for certain specimens going to color the big picture? A few arguments presented below point to the fact that a collector’s/ curator’s choice, although inherently present, does not have a dramatic efect on the type of material selected for this analysis. 34. Grierson (1986, 39). 35. Katsari (2003, 52). 36. Grierson in DOC I, xiii–xviii. 37. Bendall (2002); Grierson (1998); Morrisson (2001). 38. Grierson (1965, vi). Early Byzantine Coin Circulation in the Eastern Provinces 159 his study is devoted exclusively to copper issues, which sotens the efect of selectivity. By eliminating gold coins which are always more carefully selected and more rigorously arranged in a collection, we are let with a base metal series subject to a more random selection.39 One wonders, for example, how much selectivity there could have been in the creation of the Swiss collection purchased for Dumbarton Oaks which amounted to over 10,000 coins, mostly copper. he 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 structure is problematic.40 he oten huge number of copper coins in such collections suggests quantity and randomness as a major characteristic besides the basic desire to gather as many diferent types as possible.41 Furthermore, some collectors did not even specialized in Byzantine coinage.42 Perhaps the best example in this regard is the Köhler-Osbahr collection from the Duisburg Museum, in which the entire Byzantine series represents less than 5 percent of the 70.000 coins collected by Dr. Köhler, which included ancient, medieval, and modern coins. he collection is particularly strong in Greek, German, and Asian coinage. Numismatics itself was just one focus of his collecting interests, as Dr. Köhler assembled a very diverse collection of jewelry and minor arts covering a huge time period, from 3000 B.C. to the modern age. Ralf Athof, who published the catalogue of Byzantine coins, conirmed the fact that Dr. Köhler 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.43 Each large collection contains an important number of duplicates. Doubtless some selection occurs on the part of museum curators. he large collection at Dumbarton Oaks, for example, was subjected to the removal of the poorly pre39. 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 category are the ones gathered by William Herbert Hunt (Sotheby’s 1990 and 1991); Nadia Kapamadji (Boutin 1983); P. J. Donald (Baldwin’s 1995); Hugh Goodacre (Christie’s 1986); Anton C. R. Dreesmann (Spink 2000); and the collection sold by Bonham’s in December 1980. 40. Cf. Grierson (DOC I, xvi). 41. O’Hara in Bonham’s sale catalogue of Dec. 1980, 5, discussing an important collection of Byzantine gold and silver coins: “he 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 days of ‘investment portofolios’ amassing rows of identical coins of somewhat dubious quality” 42. Many of the collectors who donated their Byzantine coins, such as E. T. Newell, de Salis, Köhler, and H. C. Lindgren had only a marginal interest in Byzantine coins. 43. I am grateful to Ralf Althof from the Kultur- und Stadthistorischen Museum Duisburg for the valuable information provided on this important collector. 160 Andrei Gândilă served duplicated when the collection was published, but the state of preservation itself is oten governed by statistical principles. How curators deine a duplicate is also important. For the purpose of this study a duplicate signiies a coin of the same denomination, date, mint, oicina regardless of other variations pertaining to the use of diferent dies. Especially when large collections are involved curators may choose to deine a duplicate as a coin struck with the same die(s), but this rarely occurs in the case of common coppers. he unpublished collection of the American Numismatic Society (hereater ANS) meets the criteria for a statistically acceptable sample. he collection numbers more than 5000 early Byzantine coppers from the Eastern mints dated between 498 and 616 and is primarily the result of donations en masse in the last decades.44 In this area duplicates were never cleaned, removed, sold, or exchanged regardless of their condition, as long as they were legible. he major donations of bronze coins of this period are Lindgren (1984),45 Milrod (1984, ex George Bates),46 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 quality and workmanship as the main criteria, but not necessarily rarity.47 herefore, most coins are in mediocre condition at best and include numerous duplicates, which suggests a high degree of randomness. Most of the few purchases made by the ANS date to the early 1970s and the acquisition records point to a price range of $1–2.50 per coin. Only Eastern mints are included in the study, leaving out the more desirable and sought-ater Western mints. Sixth-to-seventh century Byzantine coppers from mints such as Constantinople, Nicomedia, hessalonica, Cyzicus, and Antioch have been the most common and cheapest coins on the market since the nineteenth century.48 Mass purchases of such cheap types are typical for major collectors. 44. he ANS collection is by far the largest; by comparison, the collection at Dumbarton Oaks, which is the second largest, has only c. 2800 pieces. 45. 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 EBC donated to the ANS have an identical green patina and similar dirt incrustations which suggest that they were part of a large hoard. he 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 irst decades of the seventh century. his group of coins was excluded from the statistical analysis of the ANS collection. 46. Bates (1981). 47. As a collector E. T. Newell is, of course, best known for his splendid collection of Greek coins, now at the ANS. 48. As early as the mid-nineteenth 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. he criterion is not so much style but degree of rarity and this considerable diference in Early Byzantine Coin Circulation in the Eastern Provinces 161 Quality is not always an issue; all collections under scrutiny have poorly preserved specimens even from the most common types. hey include specimens on which details such as the regnal year, the oicina, or the mint mark are no longer legible. A large number of smaller denominations, oten less appealing to the collector’s eye, represent an important percentage of the group, which suggests quantity not quality as a criterion. he hundreds of duplicates themselves point to the largely random nature of these collections. Admittedly, as Philip Grierson has argued on several occasions,49 some collectors intended to gather all the known (and hopefully unknown) types that they could ind. However, the mere fact that none of them was able to achieve this ultimate goal, coupled with the fact that all of them seem to have been very successful at gathering the same particular issues (certain regnal years, mints, oicinae, etc.) points to the fact that some types were more readily available than others. he fact that diferent collectors had the same success with some types and shared a similar failure in inding others indicates a luctuation in coin production which inevitably translates into the varying numbers of coins available today. Furthermore, there is a striking resemblance between the ive major collections in terms of structure and consequently, of statistical results (Figure 1). What counts in the end is the observable similarity of these collections, even when they are tested at the detailed level of annual luctuations. We do not have suicient information on each of the major individual 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 conirms the general pattern of annual luctuation. Unfortunately, the only archaeological samples comparable in size with the large museum collections are the single inds from the province of Scythia and a number of large circulation hoards from the Near East.50 By necessity, the analysis has to follow the nature of the evidence. he dating structure of the Near Eastern hoards makes them suitable for an analysis of the second half of the sixth century, for which the material is abundant (Figures B–D). On the other hand, the coins from Scythia are less useful valuation proves that collectors of Byzantine coins were already having a rough quantitative image of the Byzantine coin series and were therefore able to determine the degree of rarity for each type. Tolstoi’s estimations, Ratto’s sale catalogue with prices realized, and David Sear’s Byzantine Coins and their Values (1987) conirm this diference of appeal up to this day. 49. Grierson 1967,323–324; 1986, 39. 50. he most important hoards for this purpose are Chyrrus, Tell Bissé, Amman, Baalbek, Khirbet Fandaqumya, Quazrin, and a number of hoards with uncertain provenance in the Near East. See Todd (1987); Mansield (1995a, 1995b); Ariel (1996); Bates and Kovacs (1996); Pottier (1983); Noeske (2000, II); Naismith (2004). 162 Andrei Gândilă 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 ofers a good sample for the preceding decades (Figure A). he comparison between the collections and the inds from Scythia and the Near East ofer a high degree of correlation and is perhaps key to demonstrating the randomness of museum collections.51 Even more, Figure D clearly shows that those who assembled the major collections did not limit their collecting behavior to “one coin of each type.” he irst years of the reign of Heraclius shown here by way of example suggest that the museum collections follow the pattern of the single inds and hoards and not the variety of types. Obviously, both the single inds from Scythia and the hoards from the Near East represent types of evidence that were never subjected to selection at the hands of collectors and museum curators. Such statistical similarities indirectly undescore acceptable degree of randomness in the large collections under consideration. II. A C A  E B C C. M , S F,  H. he present study is not an attempt to determine the absolute number of coins produced by a certain mint or in a certain year.52 he graphs highlight the luctuations 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 moment. he evidence 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.53 Private and public collections, single inds 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 producing an accurate relection of the coin production 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 shown oten that site 51. he spikes observable in Figure B on coins from Scythia dated from 568 to 571 are explained by the signiicant number of coins issued at hessalonica, a phenomenon best illustrated by Figure 4. 52. Unpublished catalogues of the Byzantine coins from the Isparta and Bolu museums and lists of coin inds from Pisidian Antioch and Melitene, referred to in the following section, were kindly provided by Zeliha Demirel Gökalp from Anadolu University. 53. With these caveats in mind, Figure 3 is most useful for observing annual luctuations in coin production within shorter chronological units (e.g., reigns, monetary reforms, decades) and less for comparing, say, the coin output in 538/9 and 612/3. Early Byzantine Coin Circulation in the Eastern Provinces 163 inds, if they amount to a statistically relevant sample, are useful for observing the evolution of coin circulation in time in a circumscribed geographical area. Comparative analyses of site inds in a broader region provide a better understanding of the monetary economy in a larger unit of analysis, such as an administrative province, as has been shown for Scythia, Pisidia, and Arabia.54 Finally, the evidence of circulation hoards, which has been privileged by prominent Byzantinists and numismatists, usually informs us about the 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 East55 and the team of scholars coordinated by Cécile Morrisson for the Balkans,56 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 the general to detailed: nummia / year of reign (Figure 1), quantity of nummia / year of reform (Figure 2a; 5a–c),57 quantity of nummia year-by-year (for the period 538–616) (Figure 3), mints (Figures 6–19), and denominations (Figures 20–32). Obviously, the collections difer 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 number 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 from Justin II’s reign, in each collection. he chosen time span, 498–616, opens with the reform of Anastasius and ends with the abrupt decline in coin circulation ater 616 in several major centers of the Eastern Empire.58 he province of Scythia provides us with a 54. Marot (1998); Gândilă (2008); Demirel Gökalp (2007). 55. Pottier (1983); Noeske (2000). 56. Morrisson et al. (2006). 57. A conversion into solidi based on the ratios proposed by Hahn (see n. 75) is provided in Figure 2b, which gives a more accurate picture of the purchasing power of the base coinage in the early Byzantine period. 58. Wastage rates are sometimes included in analyses when long periods of circulation are involved. For the methodology and applications to Roman coinage see especially DuncanJones (1994, chapter 14). However, the extrapolation of modern wastage rates to ancient coinages remains of somewhat dubious value. Moreover, Figure A does not reveal any clear signs of wastage for the dated series of Justinian from Constantinople (27 years). Stray inds 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 relect the efects of wastage at the time when the hoards were concealed (i.e., fewer coins from the early regnal years). 164 Andrei Gândilă Map 1. Early Byzantine Coin Finds: Major Sites and Local Museums 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 inds 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 EBC available for study (Map 1). II.1. he reform of Anastasius and the pre-538 coinage In 498 Anastasius introduced a new system for the base-metal currency, one that would put an end to the crisis of the ith century, which rendered the low val- Early Byzantine Coin Circulation in the Eastern Provinces 165 ue currency almost worthless.59 However, as shown by Figure 2a, the number of small-module coins struck ater the reform does not seem to be very high, if compared with the quantity issued ater a second reform in 512. In geographical terms, a larger number of small-module issues can be found in the Danube area60 and, to an even larger extent, in a number of urban centers in Syria-Palestine—Jerusalem, Pella, Gerasa, Beth She’an (Scythopolis)—and especially Berytus where a unique situation can be noted.61 In Anatolia, urban centers such as Sardis, Sagalassos, Side, Pisidian Antioch, and Melitene point to a rather reduced impact of the reform in the irst period.62 To return to the special case of Berytus, almost 70 percent of the total number of EBC is represented by small-module issues. Berytus might well have represented an idiosyncratic circulating micro-medium, a semi-closed monetary environment which might have encompassed a larger area of Phoenicia,63 but it does, nevertheless, raise an important question regarding the withdrawal of these coins from circulation, once a new reform in 512 doubled the weight of the copper coin. As Kevin Butcher has shown 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 59. For the reform of Anastasius and its impact see in particular Blake (1942); Grierson (1967b); Metcalf (1969). 60. Especially at Tomis (9 out of 26 coins of Anastasius) and Dinogetia (3/8): Gândilă (2003–2005, 129–144); Shumen (4/16): Zhekova (2006, 65–66). he highest concentration has been recorded in Constantinople, at Saraçhane and Kalenderhane (19/26), Hendy (1986, 285–287); Hendy (2007, 196–198). hese types are scarcer among inds from the Western Balkans, in Serbia (5/39): Radić and Ivanišević (2006, 92–96), and Albania (3/21): Spahiu (1979–1980, 366–368), while in Greece the reform had little immediate impact: Hohlfelder (1978, 63); Bellinger (1930, 45); Edwards (1933, 121); Edwards (1937, 249); Mac Isaac (1987, 135); hompson (1954, 66–67). 61. Metcalf and Payne (1965, 130–131); Sheedy (2001, 128–130); Marot (1998, 461–464); Bellinger (1938, 95–97); Bijovsky (2002, 511–512); Butcher (2003, 257–264). One can add a few more sites or regions with a smaller quantity of EBC, of which a good proportion is made up of small-module coins of Anastasius. Such cases are Capernaum: Spijkerman (1975, 29, 31), Fayran: Noeske (2001, 708), and Mesopotamia: Prawdzic-Golemberski and Metcalf (1963, 90–92). A special case is the synagogue at ‘En Nashut where half the small batch of EBC are small module issues, Ariel (1987, 151, table 1). 62. Bell (1916); Bates (1971); Buttrey (1981, 212); Scheers (1993, 254; 1995, 314; 1997, 332; 2000, 525); Atlan (1976, 77). No such coins can be found in the local museums in Amasra (ancient Amastris): Ireland and Atesogullari (1996, 132); Bolvadin (vicinity of ancient Polybotos): Ashton, Lightfoot, and Özme (2000, 183). However, see a few specimens at Amasya (ancient Amaseia): Ireland (2000, 101), and Pessinus: de Wilde (1997, 107); Devreker (1984, 211). A few specimens found in Anatolia are now kept in the archaeological museum in Zagreb: Mirnik and Šemrov (1997–1998, 143–145, n. 11, 36). 63. See the hoard of small module folles found at Sarafand south of Beirut, Taylor (1977, 87). 166 Andrei Gândilă circulation at that time.64 It is hard to determine with any 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 ½-folles, K on ¼-folles) indicating that the new coins were worth twice as much as the old, smaller ones.65 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 diferent face-value.66 To be sure, these small and ephemeral issues remained in circulation throughout the sixth century as testiied by hoards found in the Eastern Empire.67 Despite their small size, such coins remain outside the danger of “Gresham’s Law,” as the state did not have to fear that the circulation of the newly introduced heavier specimens might be disrupted by the existence of those lighter issues, as long as their face-value was halved. he reuse of late Roman, early Roman and even Greek coins, based on a similar size and weight is not uncommon in the large centers of the empire and is also testiied by hoards containing such specimens. he museum collections conirm 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 (Figures 7a–b). here is a fairly balanced proportion of folles and ½-folles both in the collections and in the samples found during archaeological excavations, while the ¼-folles are themselves well represented (Figures 21a–b). his phenomenon shows that the divisionary system was functional and 64. he evidence of hoards seems to point in the same direction. Several hoards containing a large number of small module folles were found in collapsed buildings associated with the earthquake of 551. Butcher (2003, 283–286); Beliën (2005); Abou Diwan (2008). 65. Schindler (1955, 33–35). 66. See Mecalf (1969, 41–43), followed by Pottier (1983: 227–230) who suggested that the countermarks oten found on small module coins from the Middle East are a sign that the state was attempting to regulate the use of the pre-reform coins. he reduction of the face value was also accepted as a plausible hypothesis by Grierson (1982, 60) and Hahn and Metlich (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 (2008, 316–317). 67. In the Balkans and Anatolia out of 36 hoards containing coins of Anastasius, seven include small module types. he latest of these hoards, Caricin Grad 1952, ends in 595/6, testifying to the longevity of the small-module coins of Anastasius (Morrisson et al. 2006, 299). In the Near East small-module coins occur occasionally in large hoards ending in the seventh century such as Tel Bissé, Baalbek, Khirbet Fandaqumya and “Northern Syria,” Noeske (2000, II); “Lebanon” Kruszynski (1999). his evidence clearly contradicts Noeske’s supposition that the small-module series was immediately withdrawn in 512 (Noeske 2000, I, 150–151). Early Byzantine Coin Circulation in the Eastern Provinces 167 smaller denominations were used frequently in minor transactions. he museum collections, however, hardly contain any specimens of the smallest denomination, the nummus, found especially in Greece (Athens, Corinth, Kenchreai),68 Anatolia (Sardis, Sagalassos),69 and Palestine (Caesarea Maritima, Beth She’an, Ramat Hanadiv, Hammat Gader)70 and to a much lesser extent in the Balkans and at the Danube border.71 he retrieval of large numbers 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,72 might signal the fact that we are largely underestimating the sheer quantity of low value coins still in circulation deep into the sixth century.73 he contrasting image ofered by Sardis awaits more information from other centers in Western Anatolia in order to determine whether this is a particular case or a more general phenomenon. Philip Grierson attributed the paucity of minimi at Sardis to the negligence of the excavators, 74 but recent research in Anatolia, at Melitene and Pisidian Antioch shows that minimi are generally scarce. he period 512–538 is homogeneous in many respects, largely due to a stable ratio between the gold solidus and the copper follis, most probably 1:360.75 Figure 68. hompson (1954, 66); Edwards (1937, 249); Fisher (1984, 245); Hohlfelder (1978, 64). See also the case of Nemea where John Mac Isaac argued that minimi continued to circulate during the sixth century, Mac Isaac (2005, 185). 69. Bates (1971); Scheers (1993, 254; 1997, 332; 2000, 525). 70. Hamburger (1956, 115–138); Evans (2006, 180–203); Bijovsky (2002, 507–512); Barkay (2000, 413, table 4); Barkay (1997, 300). See also the hoard of minimi from Gush Halav, Bijovsky (1998, 77–106), and a hoard of minimi and ⅛-folles probably found in Lebabon, Phillips and Tyler Smith (1998); A signiicant number of one-nummus pieces have been found during the excavations on the Limes Arabicus, primarily at Lejjūn, Betlyon (1988, 171–172). 71. Gândilă (2008,318, table 5). 72. Gush Halav, Bijovsky (1998, 77–106), with a comparative discussion of the circulation of minimi in the irst half of the sixth century (Morrisson et al. 2006) Greece: several hoards in hebes, Athens, Corinth, and Kenchreai; hasos 1977, Argos 1892–1895, Hagios Nikolaos 1935, Kenchreai 1963, Kleitoria 1933, Megara before 1884, Patras 1938, Pellene 1937, Priolithos 1979, Spata 1982, Trype 1935, Zacha, Chersonissos. Dobrudja: Constantza 1929; Histria 1974. Several small hoards in Istanbul (Hendy 2007, 271–276). 73. See also the case of Gerasa where Marot has shown that late Roman coins are still present in sixth century archaeological contexts, Marot (1998, 304). 74. Grierson (1965, xi). 75. he ratio between solidus and follis has been taken from Hahn’s MIB I (1973, 27); MIB II (1975, 14–17), and MIB III (1981, 16). A consensus is yet to be reached regarding the calculation of this ratio and diferent propositions have been made in the past decades: Callu (1982), Pottier (1983, 252), Morrisson (1989b, 248); Morrisson and Ivanišević (2006, 51); Hendy (1985, 478). 168 Andrei Gândilă 2 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 inds than the irst decade of Justinian’s reign. In the larger framework of the “long sixth century,” however, the coins minted between 512 and 538 stand at a lower point than the post reform coinage of Justinian and the inlationary peak reached during the reign of Justin II. he larger quantity of coins from Justin I has already been noticed in the Near Eastern provinces,76 although it is hardly a general phenomenon and the evidence is still too scant to permit a conclusion in this respect.77 he phenomenon is conspicuous in the province of Scythia, where all the major sites without exception display a peak reached during the reign of Justin I.78 his is by no means characteristic for the Balkan area as a whole. he neighboring province of Moesia II ofers a contrasting image with a high occurrence of coins of Anastasius. 79 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 Justin I (Figure 5a).80 In Anatolia the evidence available from Sagalassos, Sardis, Side, Amaseia, Amastris, Pisidian Antioch, and Melitene ofers a mixed picture (Figure 5b)81 and so does the evidence from Cyprus, at Paphos, Salamis and Curium.82 he major Syro-Palestinian sites seem to be more correlated, with the notable exception of Berytus (Figure 5c). 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. he museum collections suggest a slight decrease in coin production during the reign of Justinian I, prior to his major reform in 538 (Figure 2). he archaeological evidence indicates that such a phenomenon is very clear in the Balkans83 and to a large degree in Anatolia,84 but seems to be somewhat irregular in the Near Eastern sites, where, without a clear distribution according to provinces, we 76. Grierson (1967b, 296); Walmsley (1999, 344). 77. Butcher (2003, 103, ig. 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. 78. Gândilă (2008, 322, table 3 and 4), where 10 major sites are compared. 79. Mihailov (2008, 281, table 4). 80. hompson (1954, 67); Edwards (1933, 121); Mac Isaac (1987, 135); Radić and Ivanišević (2006, 92–104), Ivanišević (1988, 90–94); Janković (1981, 72, table 6). 81. Scheers (2000, 525); Bates (1971, 19–26); Atlan (1976, 78); Ireland (2000, 102); Ireland and Atesogullari (1996, 132–123). 82. Nicolaou (1990, 192–204); Callot (2004, 41–43); Cox (1959, 77–78). 83. Gândilă (2008, 306, table 1); Radić and Ivanišević (2006, 110–125); Edwards (1933, 121–122). he situation is somewhat balanced in Albania: Spahiu (1979–1980, 368–377). 84. Bates (1971, 28–44); Ireland (2000, 102–105); Ireland and Atesogullari (1996, 133). Early Byzantine Coin Circulation in the Eastern Provinces 169 ind all three possible situations —the prevalence of coins from 518–527 (Pella),85 a balanced proportion (Gerasa, Nessana),86 and a larger number of coins from 527–538 (Caesarea Maritima, Hama, Antioch) (Figures 5a–c).87 Aside from these luctuations, the Near Eastern provinces yield the highest volume of inds dated to the pre-538 period. his characteristic is conirmed by the structure of the hoards found in the area, which contain a good number of pre-reform coins, even if most of these hoards were concealed ater 600.88 As a general observation the quantity of Justinianic pre-reform issues depends on the inluence of the mint of Antioch, which is rather insigniicant in the Balkans and most Anatolian sites (Pisidian Antioch and Amaseia being two major exceptions) (Figure 10b). According to the evidence from the collections, Constantinople is the most important mint during the period 512–538. Its inluence, however, gradually diminishes in favor of the Antioch mint, which greatly increases its output during the irst decade of Justinian’s reign, even surpassing Constantinople. he mint of Nicomedia retains a secondary role, while Cyzicus and hessalonica, re-opened by Justin I have only a modest output at this time (Figures 8a–10a). he mints issued especially folles and, somewhat surprisingly, a large number of ⅛-folles, particularly during the reigns of Anastasius and Justin I and to a lesser degree at the beginning of Justinian’s reign. Except for this latter period, the ½-follis is struck in smaller quantities. he role of the ¼-follis appears to be less signiicant during this period and it seems that, in most cases, it was the large number of ⅛-folles which fulilled the role of small change in the market (Figures 22a–24a). his phenomenon is less visible in the Balkans, where, with the exception of Ahtopol (Agathopolis),89 on the Black Sea coast, and of Constantinople,90 the urban centers and border fortresses yielded a very small number of eight-folles. Especially in Scythia, the balanced proportion between folles and ½-folles indicates that the latter was the only fraction required in a market whose intensity of small transactions was relatively low.91 In the western Balkans the proportion of folles is overwhelming, which could indicate that the severe disruption of urban life in the ith century had long-term consequences.92 In Anatolia the most substantial 85. Sheedy (2001, 130–131). 86. Bellinger (1938, 98–102); Marot (1998, 465–471); Bellinger (1962, 71–72). 87. Evans (2006, 183–188); homsen (1986, 62); Waage (1952, 149–155). 88. Noeske (2000 II) (Baalbek, Khirbet Fandaqumya, Syria 1974, Khirbet Dubel, Khirbet Deir Dassawi, Rafah, Amman), Pottier (1983); Metcalf (1975, 110–112); Mansield (1995a, 348–350); Naismith (2004, 296–297). 89. Iordanov, Koicev, and Mutafov (1998, 71, table I). 90. Hendy (1986, 287–295); Hendy (2007, 197–206). 91. Gândilă (2008, 318, table 5). 92. Spahiu (1979–1980, 366–377); Radić and Ivanišević (2006, 92–125); see also at the Iron Gates of the Danube (Janković 1981, 66, table 3). 170 Andrei Gândilă evidence comes from Sardis, as usual, where the ⅛-follis represents the main denomination in the period following the reform of 512, but its volume gradually decreases in the following decades prior to 538 in favor of the follis. At Amasya, Amastris, Side, Melitene and Pisidian Antioch the pattern of denominations resembles the situation in the Balkans where the main role is played by the follis, followed by the ½-follis. In the Near East we ind once again a mixed picture. It can be argued that the smaller denominations are more present in the Near Eastern provinces, especially in Antioch where the ⅛-follis is prevalent in this period, but also in other major sites like Caesarea Maritima, Nessana, and Berytus.93 As already mentioned, another characteristic is the high presence of minimi in Beth She’an, 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 relecting the real structure of denominations in circulation and more the habit of throwing small coins into the water as a symbolic ofering.94 In Syria II - at Hama (Epiphania), in Palaestina II - at Pella, and in Arabia - at Gerasa the structure resembles the one seen in the Balkans and in Anatolia, with very few small denominations (Figures 22b–24b). A constant feature of the Near Eastern provinces is the larger role played by the Antioch mint than in Anatolia or the Balkans (Figures 8b–10b). However, in centers like Pella, and especially Nessana, very far from Antioch, in Palaestina III, the mint of Antioch is less inluent.95 II.2. he post-reform coinage of Justinian I he four-year period following the reform of 538 is one of the most intriguing. It is also the only point in which the ive major collections under scrutiny present higher quantitative variation. Although there certainly was a dramatic increase in output immediately ater the reform, we can also accept that a certain bias existed in favor of collecting the eye-catching, impressively large folles of Justinian. his is highly visible in the cases of DOC and BNP (Figure 3). Interestingly, the coin inds from Scythia, where no collector’s choice is involved, share this high peak reached in the period 538–542. In all cases, including Scythia, the numbers point to a continuous decrease in mint output during the next two reform periods in the reign of Justinian, 542–550 and 550–565 (Figure 2a).96 he economy was not able to sustain a constant high output of heavy folles, whose introduction in the irst place must have relied on both economic and propagandistic agendas. It is signiicant in this respect that the majority of coins both in the collections and in Scythia 93. Evans (2006, 180–188); Bellinger (1962, 72); Butcher (2003, 263). 94. See the discussion by Barkay (2000, 415–417). 95. Walmsley (1999, 337, table 4); Bellinger (1962, 71–72); at Nessana the inluence of Alexandria, geographically much closer than Antioch, is more visible among the EBC. 96. For a possible explanation of this phenomenon see Pottier (1983, 241–242). Early Byzantine Coin Circulation in the Eastern Provinces 171 are comprised of folles, in a proportion usually higher than 70 percent (Figures 25a–b). he situation changed dramatically in the second half of the 550s when a huge number of ¼-folles looded the market (Figures 27a–b). hey 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 shit. It seems to correspond to a wider set of measures taken by Justinian in the last years of his reign. According to the current information, Nicomedia and Cyzicus stopped minting folles and ½-folles ater 561 and concentrated almost exclusively on striking ¼-folles, while hessalonica abandoned its idiosyncratic denominational system and began issuing ½-folles in 562.97 hese measures might have been caused by a need for small denominations ater the market had been overwhelmed by a high number of folles for two decades. Furthermore, the increased production of ¼-folles can be seen in all the regions of the Eastern Empire. At Noviodunum, on the Danube, 57 percent of the coins from 550 to 565 are ¼-folles; at Tomis, on the Black Sea, they represent 75 percent; at Corinth, 73 percent; at Sardis in Lydia, almost 70 percent of the inds, and at Antioch, 55 percent (Figure 27b). Even when very few coins are reported for this time interval we ind ¼-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.98 he mint of Constantinople gradually reduced its output in favor of Nicomedia, Cyzicus, and especially Antioch during the last reform period, 550–565 (Figure 13a). hessalonica still had a secondary role largely restricted to supplying the area of the western Balkans (Figure 13b).99 A geographic anomaly can be noted in the case of the Antioch mint: for reasons that are not clear, Antioch is very well represented in the collections of the museums in Amasra and Amasya, while closer to Antioch, at Side it is less well represented. 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 come from the central mint in Constantinople.100 he mint of Antioch appears to serve primarily the needs of the city but its inluence can be far-reaching as shown by the cases of Amasra and Amaseia. he higher presence 97. Hahn and Metlich (2000, 56–62). 98. Gândilă (2006–2007, 114–116); Scheers (2000, 525); Atlan (1976, 79–81); Callot (2004, 44); Cox (1959, 78); Butcher (2003, 268–269); Ariel (1986, 142); Lampinen (1992, 172); Evans (2006, 186–187); Barkay (2000, 413, table 4); Tushingham (1972, 199). 99. Radić and Ivanišević (2006, 122); Ivanišević (1988, 92); Spahiu (1979–1980, 376–377). he mint is rather under-represented in Greece proper and D. M. Metcalf (1976, 8) has explained its geographical distribution by restricting its role to military expenditure at the Balkan border. 100. Ariel (1982, 326); Morrisson (1995, 79); Walmsley (1999, 337, table 4–5); Evans (2006, 48, ig. 17); Butcher (2003, 257–269). 172 Andrei Gândilă 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. he 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 circulating scheme of the “long sixth century” is the role of such heavy specimens ater the weight-standard of the copper coin began to slide until it was inally established at half the weight of the Justinianic large follis.101 “Bad money drives out good” was an economic principle well understood in early Byzantium. he reform of Constantine IV, briely 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. he complexity of the early Byzantine monetary economy should be neither under- nor over-estimated by viewing it from the perspective of modern economic policies. Both single inds 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. he intensive excavations at Saraçhane and Kalenderhane in Istanbul have yielded close to 500 coins dated 491–616 and not a single one of them was a heavy follis or a ½-follis of Justinian. In Antioch, out of more than 2300 EBC, only two folles and four ½-folles are dated to 538–542. Large cities where imperial mints were located, as was the case at Constantinople and Antioch, certainly had tighter control over the circulating mass of coins in their urban areas. In the Balkans, both hoards and single inds point to an abundance of such heavy coins and, more signiicantly, their persistence until the last decade of the sixth century. In Scythia, coins from 538–542 represent more than 10 percent of the entire group of EBC, while the proportion is much higher in Moesia II and in the north-western Balkans, in Serbia.102 It is interesting that the major urban centers of Scythia, namely Tomis, Histria, and Noviodunum, yielded a smaller number of large folles, while none of the four hoards found at Histria contains such coins.103 In the fortresses defending the Danube frontier the situation is different. At Durostorum, 40 percent of the coins of Justinian are heavy issues from 101. BNP I, 61. 102. Gândilă (2008, 306, table 2); Belgrade museum (18.42 percent), Radić and Ivanišević (2006); Shumen museum (22.18 percent), Zhekova (2006). 103. Gândilă (2008, 322, table 4). Morrisson et. al. (2006, 170–174). Early Byzantine Coin Circulation in the Eastern Provinces 173 538–42.104 A small hoard of 51 coppers recently found at Capidava contains coins up to Tiberius II, and yet one third are heavy folles of Justinian. he coins were kept in a small textile container and were found overlapped in a row on the loor of a room destroyed by ire. he lack of intentionality allows a glimpse of an ordinary purse of coins probably handled by a soldier on the Danube frontier in the early 580s.105 Such examples suggest that the process of withdrawing the heavy series was more readily applicable in the major centers where control was tighter. Nevertheless, the coin hoards from the Balkans, as a general characteristic, contain heavy specimens as late as the 580s, as testiied by such inds across the peninsula, in Greece, Serbia, Bulgaria, and Romania.106 It is signiicant that, with one exception (Veliki Gradac), no such coins seem to appear in any of the hoards concealed in the 590s, a possible sign that the big coins of Justinian had been almost completely removed from circulation by the end of the century.107 Another phenomenon might suggest that the 590s represented a time of intensive withdrawal of heavier issues, namely the overstriking of Maurice folles on previous Justinianic folles, ater the lan was trimmed to meet the demand of the new weight standard. Several public collections108 and catalogues of site inds and hoards109 contain such overstruck specimens. Most of them date from the early 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 diiculty encountered by Justin II and Tiberius II in collecting the taxes from the border provinces of the Balkans, which received particular mention in the legislation of 566 and 575.110 he 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 heophanes 104. Author’s ile cards. 105. Author’s ile cards. 106. Morrisson et. al (2006). Most signiicant hoards are Koprivec, Zhalad, Adamclisi 1908, Athens 1908, and Eleusis 1893. 107. Morrisson et al. (2006). Hoards ending in the 590s: Reselec, Rakita, Soia, Histria (5 hoards), Caričin Grad 1952, Bosman, and Horgești, Movileni, Unirea, north of the Danube, in “barbaricum.” 108. Sommer (2003, 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). he 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 revaluation exercise. 109. Viminacium: Ivanišević (1988, 94, n. 56); Sardis: Bates (1971, 68, n. 562); Caesarea Maritima: Ariel (1986, 143, n. 67); Evans (2006, 193, n. 2472) (half-follis); Tell Bissé hoard, Leuthold (1952–1953, 39); Quazrin hoard, Ariel (1996, 75, n. 6). 110. Popescu (2005, 379). 174 Andrei Gândilă Confessor who argued that the imperial treasury could no longer sustain the regular payment of the troops, so the state was forced to cut a quarter of the salaries in 587.111 he decision to resize and overstrike larger issues, thus gaining additional metal and insuring the payment of troops in “new” coin, can be ascribed to the diicult inancial situation mentioned in the written sources. In Anatolia the big coins are less well represented than in the Balkans but still represent an important proportion of the total number of EBC. Excavations at Side, Pergamum, and Sagalassos have yielded a number of specimens while the local museums in Bolvadin, Amasra, and Amasya also contain heavy folles dated 538–542. here 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 590s lack any large coins of Justinian.112 here 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 scarcity to their withdrawal from circulation, while Henri Pottier and Cécile Morrisson have pointed to the fall in circulation between 538 and 565 and suggested that wars and natural disasters are important factors explaining this situation.113 More recently P. J. Casey attempted a closer analysis of the post-reform coinage by looking at the evidence from site inds and hoards across the Eastern Empire. His point of departure was a written source, Procopius’ Secret History, especially a passage where the Byzantine historian claims 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 from Syria-Palestine, Casey concluded that such circumstances may indeed explain the virtual absence of post-reform coins from Palestine, but admitted that such an argument is less compelling in the case of Syria.114 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.115 111. See the discussion by Yannopoulos (1987, 129). 112. Unfortunately, the information comes from a single major source, Sardis, where at least 4 hoards (found 1913, 1958, 1961, and 1968) ending ater 590 are relevant for this discussion. Another hoard, from Anemurium in Isauria, ends in 602 and has no coins prior to 578. Morrisson et al. (2006). 113. Grierson (1967b, 296); Pottier (1983, 55); Morrisson (1989, 192); Morrisson and Ivanišević (2006, 52). 114. Casey (1996, 220). 115. Noeske (2000, I, 152–153). Early Byzantine Coin Circulation in the Eastern Provinces 175 Although coins dated 538–542 are indeed conspicuously hard to ind, some are still reported at Jerusalem, Caesarea Maritima, Antioch, Berytus, Pella, and Nessana in six diferent provinces of the Near East.116 Coins issued during the remainder of Justinian’s reign, 542–565 are more common and they are found in almost all excavations conducted in the region, and in a number of hoards.117 It is thus fair to conclude that the post reform coinage did penetrate into the Near Eastern provinces, perhaps in smaller quantities than in the Balkans. his contrast should not be exaggerated, however, if we take into consideration the level of urbanization in the two regions. As noted, the major towns in Scythia yielded fewer heavy coins and a tighter control of the coins in circulation can certainly be envisaged in a highly urbanized region like Palestine, for instance. Antioch, the mint whose chief purpose was to serve the major Syrian city and its vicinity, issued coins in this period in especially high numbers starting from the late 540s (Figure 13a), of which only a small percentage reached more distant parts of the Empire such as the Balkan provinces. Coins minted in Constantinople and Nicomedia are extremely common among inds in the Near East and, judging by their increased output immediately ater the reform in 538, it is hard to imagine that the coins were artiicially 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 afected the circulation, but a long term disruption of the inlux of new coinage seems rather improbable.118 he 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 fact, although Casey’s central argument concerns Palestine, his comparative table includes only one, remote, Palestinian center, Nessana, notwithstanding his discussion of the hoards, concealed late in the sixth or early in the seventh century and consequently less relevant for the discussion.119 he urban record is still decidedly thin, but in the light of the 116. Fitzgerald (1929, 117); Ariel (1986, 142); Waage (1952, 152–154); Butcher (2003, 266–268); Sheedy (2001, 132); Bellinger (1962, 72). 117. From the major site inds discussed throughout this article only Nessana and Hama failed to produce any inds from 542–565. he relevant hoards are “Northern Syria” (Pottier), Syria 1974, Khirbet Dubel, Tell Bissé, Baalbek, Khirbet Deir Dassawi, cf. Noeske (2000, II); Qazrin (Ariel 1996, 70, table 1); “Northern Syria” (Todd 1987,178–179); “Near East 1993” (Mansield 1995b, 355); “Near East 2003” (Naismith 2004, 297), and a Near Eastern hoard which includes Arab-Byzantine issues as well (Bates and Kovacs 1996, 166). 118. he mint of Antioch ceased minting coins in years 14–15 of Justinian’s reign when the city was sacked by the Persians and in Justinian’s regnal years 17, 18, and 19 because of the Great Plague. See DOC, 143. For a list of the major earthquakes in Palestine see Russell (1985). However, once the crises were overcome, the mint was reopened. 119. Casey (1996, 217). 176 Andrei Gândilă new evidence, mostly but not completely inaccessible to Casey, it is more plausible to suggest that the heavy specimens did circulate in the area, but were more eficiently withdrawn from circulation at a later period. he large coins disappear 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 Near Eastern provinces followed the same pattern, largely because the major coin hoards from this region, with the exception of Rafah and a “North Syrian hoard,”120 have a closing date ater 595. he hoard of Rafah has a “closing coin” dated 573/4, but despite the early date of closing it contains no post-reform coins of Justinian. he “North Syria Hoard,” however, ends in 584/5 and has 16 post reform coins out of a total of 60 pieces, which means more than 25 percent of the entire hoard. Almost half the coins 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. he 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.121 he hoard contains 55 base-metal coins of Justinian of which 16 are post-reform issues, meaning almost 30 percent of the total. Six of the post-reform coins belong to the heavier standard.122 Because of its special nature, as an “open” savings hoard, fresh coins were constantly fed into this treasury and many were never taken out. his is probably the reason why these heavy folles escaped the process of withdrawal. It also conirms once again the presence of the post reform coinage of Justinian in Palestine, possibly in much greater numbers than we are inclined to believe, based on the surviving specimens from site inds 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 eiciently compared to the Balkans and even Anatolia. his would explain the pronounced scarcity of the big coins among inds in SyroPalestinian sites, given the fact that they circulated for a shorter period. he fact that the large coins were withdrawn from circulation can also be indirectly reinforced by the unusual number of pierced coins.123 Without attempting 120. Noeske (2000, 634–639); Todd (1987, 176–182). 121. Kindler (1986). 122. 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. 123. Pierced coins from the following decades, 542–616, are seldom found in public collections or among site inds. 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 (Delougaz 1960, 51 and plate 46). Early Byzantine Coin Circulation in the Eastern Provinces 177 to be comprehensive, I assembled the most signiicant instances where pierced folles of Justinian have been noted, both in public collections124 and among site inds from the major geographical areas of the Eastern Empire, the Balkans,125 Anatolia,126 and the Near East.127 Signiicantly, most of the coins are pierced at 12 o’clock, above the emperor’s head, which suggests that the coins were worn as pendants. One specimen from Pella is holed six times and was probably sewn 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. he 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 are pierced, so this is hardly an isolated phenomenon from a later period. It is unlikely that such coins, once demonetized, 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 ater the entire series was oicially withdrawn from circulation. Although it is not 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 owner of such a coin would have acknowledged its special nature only ater Justin II had introduced a follis half its weight. he big coins of Justinian were therefore highly regarded by the common people and perhaps reminded them of an ambitious age of military achievements and building programs, both lacking in the decades when such coins were probably being pierced. II.3. Inlationary tendencies and decline (565–616) he reign of Justin II witnessed a dramatic increase in coin output with a peak reached in the interval 570–575, ater a general tendency of accretion during the irst ive years. he prominent peak from 574/5 might be artiicially produced by the inclusion of numerous types described by Hahn as Moneta Militaris Imitativa, 124. MIB I (112, plate 22); Sommer (2003, 38, n. 106, plate 2); several specimens in BNP, DOC and in the ANS collection, the latter including a gold plated piece; Ratto (26 , n. 495 and 30, n. 583); KOD (56, n. 333, plate XV); Bateson and Campbell (1998, 13, n. 13, plate 2); Arslan (2000, 38, n. 14, plate III). 125. Radić and Ivanišević (2006, 117, n. 318, plate 20); Mirnic and Šemrov (1997–1998, 150, n. 102, plate 8); Poenaru, Ocheșeanu, and Popeea (2004, 35, n. 206); Hohlfelder (1978, 65, n. 1017, plate IV). 126. Bell (1916, 77, n. 639); Morrisson (1993, 55, n. 765, plate 8); Callot (2004, 191, ig. 18, n. 274). 127. Sheedy (2001, 132, n. 021 and 023, plate 10); Ariel (1986, 142, n. 54, plate I); Metcalf and Payne (1965, 185, n. 48). 178 Andrei Gândilă which bear the regnal year 10 (type MIB 89–93). he last years of the reign mark a sharp downfall in coin production, which coincide with the adoption of Tiberius as co-regent (Figure 3). hese luctuations in coin output derived from the study of the museum collections are paralleled by the numerous inds for Scythia, where the massive contribution of the mint of hessalonica forces a more dramatic increase until 570 (see also Figure 4). Numismatists have long drawn attention to the inlationist tendencies of the reign of Justin II, in direct relation with the devaluation of the follis, which went down from 216 folles / solidus to 525 and then 720. he huge volume of coins issued during this period is sometimes interpreted as a sign of crisis not of economic prosperity or increasing commercial activities.128 his is undoubtedly the relection of Justinian’s prodigal policy of expenditure on warfare and buildings, as well as of demographic decline caused by natural disasters such as large epidemics and intensiied seismic activity. However, this is by no means a crisis of catastrophic proportions. he 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 standard. Moreover, the diference in mint output between the previous reform period (550–565) and the reign of Justin II as a whole is higher than the diference in purchasing power (Figure 2b). his means that the volume of coins produced supersedes the theoretical level of inlation triggered by the devaluation of the follis. his can be interpreted either in economic terms suggesting that a certain level of prosperity still existed, or in relation to the military situation of the Empire and the need to pay the army. he high level of coin output might also be related to the policy of withdrawing the heavy coins of Justinian, which was a more or less successful 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. he mint of Constantinople is the most active in the irst reform period, 565– 570, covering approximately 40 percent of the total coin output. An important development is the importance gained by hessalonica and its ½-folles issued in great numbers during these years (Figure 14a). A few major changes occur in the second reform period, 570–578, when Constantinople, while still the major supplier, is closely followed by Antioch, which increased its output probably due to the conlict with Persia. Cyzicus became more important ater a period of low activity, while hessalonica drastically reduced its output for reasons discussed in the following section (Figure 15a). More than half the coins issued during the reign of Justin II are folles. he ¼-follis is less present, a sign that the monetary policy sustained by Justinian in his last years of reign was discontinued. he mint of 128. Poenaru (1981, 374–375). Early Byzantine Coin Circulation in the Eastern Provinces 179 Antioch alone continued to issue ¼-folles in signiicant quantities.129 he decline in the production of smaller denomination is considered a general characteristic of the second half of the sixth century,130 but this is not entirely accurate. Although ¼-folles are indeed rather scarce, the production of ⅛-folles maintains and even surpasses the levels of the preceding decades (Figure 28a). heir presence in urban settings, as it will be shown below, is an indication of a still vibrant monetary economy (Figure 28b). here is a sudden inlux of coins in the Danubian provinces ater Justin II’s decision to abandon the policy of regular payments sent to the northern barbarians in order to secure the border provinces. It is very probable that such a shit implied the arrival of additional troops to be stationed in the border fortresses along the Danube. he 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.131 his is not only a frontier-related phenomenon. Numerous inds from this period are also reported in major urban centers, such as Corinth (and Kenchreai), Athens, and Tomis, and to a lesser degree in rural areas,132 which gradually become isolated from the urban monetary economy. he extensive mint output at hessalonica explains the large number of coins in the provinces of the Balkans. hessalonica is especially inluential in the western half of the peninsula, in Greece, Albania, and Serbia133 and to a lesser degree in the east, at Odartsi, Agathopolis and in the region of Shumen.134 Although the evidence for Anatolia is still insuicient for broad generalizations, there is strong indication of a general increase in the volume of coins during the reign of Justin II. Apparently surprising, hessalonica is a major supplier of coins at Sardis, where, at least in the irst stage, 565–570, the coins struck by the Macedonian mint cover almost 35 percent of the inds (Figure 14b).135 his seems to be a general characteristic of towns from western Anatolia, close to the coast, 129. See also the observations of Pottier (1983, 186). 130. Pottier (1983, 150). 131. Gândilă (2008, 322, table 4); Janković (1981, 66, annex 3); Ivanišević (1988, 93–94); Popović (1978a, 181–185). 132. Bellinger (1930, 46–47); Edwards (1933, 125–127); Mac Isaac (1987, 135–136); Hohlfelder (1978, 68–71); hompson (1954, 68–69); Gândilă (2008, 322–323, table 4); Oberländer-Târnoveanu (2005, 383–384). 133. Edwards (1933, 125–127); Hohlfelder (1978, 68–71); hompson (1954, 68–69); Spahiu (1979–1980, 378–381); Radić and Ivanišević (2006, 132–135). See also the composition of hoards found in these areas: Morrisson et al. (2006). 134. Gândilă (2008, 323, table 5); Iordanov, Koicev, and Mutafov (1998, 72, table 4); Zhekova (2006, 79). 135. Bates (1971, 54–55). 180 Andrei Gândilă judging by the similar inds from Pergamum, Ephesus, and Side, 136 whereas further to the west the proportion dwindles, 15 percent at Pisidian Antioch, while no coins of hessalonica are so far recorded at Amaseia, Pessinus, and Melitene.137 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 second 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 inlux of coins of Justin II has been oten noted, especially because of the contrast with the post reform period of Justinian I, which is less prominent among inds. he number of inds is conspicuously high at Gerasa in Arabia and in Palaestine at Pella, Hammat Gader, Caesarea Maritima and to a lesser extent at Jerusalem and Nessana.138 Although they are fairly well represented at Antioch, Hama, and Apamea,139 no coins of Justin II have been reported among the admittedly small group of inds from Bālis,140 and they have a generally weaker presence in the rural settlements from Syria, such as Çatal Hüyük and Déhès,141 which parallels the situation observed by Ernest Oberländer-Târnoveanu in the case of the eastern Balkans.142 It also accords with the observations made by Clive Foss for rural settlements in Syria where the archeological evidence suggests a period of decline ater 550.143 he large number of coins of Justin II at Gerasa has been described by Alfred Bellinger as the most salient feature of EBC inds in this important city of the Decapolis.144 he situation was rightly ascribed to the high presence of coins from Nicomedia, partially conirmed by the subsequent inds from the “Macellum.” No clear explanation is given for this peculiar development.145 Dealing with a similar situation at Pella, Kenneth Sheedy has suggested that it might relect a new deployment of troops in the East for another episode of the war with Persia in the early 570s (Figure 5c).146 here is no such parallel at Antioch, but indeed at Apamea, 136. Morrisson (1993, 55); Milne (1925, 390); Atlan (1976, 81–82). 137. Ireland 2000, (105–106); de Wilde (1997, 107); Devreker (1984, 195–196). 138. Bellinger (1938, 103–113); Marot (1998, 472–480); Sheedy (2001, 134–136); Barkay (2000, 299, table II); Evans (2006, 188–190); Ariel (1982, 326); Bellinger (1962, 72–73). 139. Waage (1952, 155–157); Balty (1984, 240–244); homsen (1986, 62). 140. Hennequin and Abū-l-Faraj (1978, 7–8). 141. Vorderstrasse (2005, 498); Morrisson (1980, 279). 142. Oberländer-Târnoveanu (2004, 348). 143. Foss (1997). 144. Bellinger (1938, 13). 145. his characteristic can be also noticed in most Syro-Palestinian hoards: Baramki (1939, 83–84); Leuthold (1971, 15); Mansield (1995a, 350–351); Ariel (1996, 70); Bates and Kovacs (1996, 166); Naismith (2004, 298). 146. Sheedy (2002, 49). Early Byzantine Coin Circulation in the Eastern Provinces 181 the number of coins from Nicomedia is overwhelming.147 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 has an unusually signiicant presence (Figure 14b).148 he collections point to a high output of pentanummia during the reign of Justin II (Figure 28a). his is indeed conirmed by inds in Constantinople and Pessinus where more than 75 percent of the inds are pentanummia pieces;149 at Tomis, Sardis, and Antioch they represent approximately one third of the total, while in Greece they cover less than 20 percent (Figure 28b).150 he cluster of small change in large urban centers points to a necessity of the market, which seems to be less felt in small towns and fortresses and even less in rural contexts. Less than 15 percent of the coins found in Scythia are pentanummia, while in Moesia Secunda they amount to a mere 4 percent.151 No such coins are recorded among the published inds from Albania, Amaseia, Melitene, Caesarea Maritima, Pella, and Gerasa to name only the most important.152 In 579 Tiberius II, now sole ruler, attempted an ambitious reform designed to celebrate his consulship. he weight of the follis was lited to a Justinianic standard and the collections as well as the numerous inds from Scythia indicate a high mint output for this special series (Figure 3). 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 occasion of his consulship testify to the irregular nature of these special issues. As usual, the mint of Constantinople issued more than half the coins put in circulation, followed by Antioch and Nicomedia. he mint of hessalonica is ranked higher than Cyzicus, probably because of the increasing military activity at the Danube border (Figure 16a). he production of the peculiar 30-nummia introduced by Tiberius II was perhaps less impressive than the collections would let us believe. his scarcer denomination was likely to attract the collectors’ attention which explains their heavy presence in the collections, on average amounting to 10 percent of the entire number of coins attributed to Tiberius II (Figure 29a). Scythia ofers a more 147. Balty (1984, 240–244). 148. Ireland (2000, 105–106). 149. Hendy (1986, 297–300); Hendy (2007, 208–211); de Wilde (1997, 107); Devreker (1984, 211). 150. Isvoranu and Poenaru Bordea (2003, 153, table 3); Bates (1971, 49–61); Waage (1952, 156). 151. Gândilă (2008, 318, table 5); Mihailov (2008, 281, table 4). 152. Spahiu (1979–1980, 378–381); Ireland (2000, 105–106); Ariel (1986, 142); Lampinen (1992, 172); Evans (2006, 188–190); Sheedy (2001, 134–136); Bellinger (1938, 98–102); Marot (1998, 465–471). 182 Andrei Gândilă realistic proportion, with the 30-nummia accounting for less than 4 percent.153 In the three major regions of the Eastern Empire the inlux of coins issued by Tiberius II is 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 inds from Dinogetia, Capidava, Aquis, Viminacium, and the Belgrade museum.154 On the western sector of the limes the coins from hessalonica played an important role, and indeed in Greece, at Corinth and Athens, where close to 40 percent of the coin inds are issued by the Macedonian mint.155 he mint of Antioch becomes more important in the Balkans, and it is possibly a sign that some troops were brought from the eastern front, despite Tiberius’ tendency to concentrate on the war with Persia.156 here is a generally higher number of coins of Tiberius found on the Black Sea coast, Histria, Callatis, and Accres Castellum being a few major examples.157 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 following decades, a sign of the gradual disintegration of urban life in the area. Very few coins have been found in Constantinople, at Saraçhane and Kalenderhane,158 and a similar situation can be seen in the northern part of Anatolia judging by the coins from the museums in Amasya and Bolu, as well as in Pisidia at Antioch, Sagalassos and in the area of modern Isparta.159 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 (Figure 5b).160 Jumping to the island of Cyprus one notices a contrasting image: at Salamis and Curium the number of coins of Tiberius II is conspicuously high given his short reign.161 Considering that Caesarea Maritima in Palestine has yielded a similarly high number of coins from this period,162 we can advance a provisional hypothesis that the coins of Tiberius II circulated more intensively on sea routes. he observation regarding the circulation of smaller denominations made for the reign of Justin II remains valid for the short reign of Tiberius II (Figure 29a). 153. Gândilă (2008, 318, table 5). 154. Gândilă (2008, 322–323, table 4); Janković (1981, 66, annex 3); Ivanišević (1988, 94); Radić and Ivanišević (2006, 136). 155. Edwards (1933, 128–129); Mac Isaac (1987, 136); hompson (1954, 69). 156. Whitby (1988, 87). 157. Gândilă (2008, 322–323, table 3). 158. Hendy (1986, 300–301); Hendy (2007, 212). 159. No coins of Tiberius can be found among the four excavation reports from Sagalassos mentioned in n. 62. 160. Bates (1971, 63–66); Atlan (1976, 84). 161. Callot (2004, 46–47); Cox (1959, 79). 162. Evans (2006, 190–191). Early Byzantine Coin Circulation in the Eastern Provinces 183 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 29b). 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 at 600 folles per solidus. he 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 3). he irst peak is partly due 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. he other peak, in 602, coincides 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 the usual military cuirass. A large number of coins were issued in a very short time interval, which explains why many specimens are overstruck on previous issues. It also points to a crisis of raw material for striking fresh coins, a typical phenomenon in the irst two decades of the seventh century. he outstanding feature in mint activity is the high output of Antioch throughout the period, sometimes surpassing the production of Constantinople (Figure 17a). he intense military activity which characterizes the reign of Maurice is an important factor in explaining this phenomenon. he 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. he major role of Antioch even ater the eastern front was closed, coupled with the unusually low output of Constantinople in the last years of the sixth century when the Empire was waging war against the Slavs and Avars are somewhat perplexing. Between 7 and 22 percent of the coins of Maurice in the northern Balkans, including the Danube fortresses were issued in Antioch. Especially in the western sector of the Lower Danube the coins from Antioch are 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 been found.163 his is also true for the coins found in Constantinople, at Kalenderhane and Saraçhane (Figure 17b).164 It is unlikely that the mint of Antioch was commissioned to insure the payment of troops stationed in the Balkans. It is more probable that the coins were brought by the large number of troops transferred by Maurice ater the war with Persia was brought to an end. his does not help to explain the low activity of the mint of Constantinople especially between 597 and 602 (excepting the consular 163. Gândilă (2008, 322–323, table 4); Iordanov, Koicev, and Mutafov (1998, 73, table 6). 164. Hendy (1986, 301–305); Hendy (2007, 213–215). 184 Andrei Gândilă type in 602) when the war against the Avars was in full motion. Michael Whitby considers that the emperor’s decision to leave the troops stationed north of the Danube for the winter of 602 had speciic military purposes, contra heophylact, who suggested that inancial considerations were behind this decision.165 Although multiple factors might have been at play, the low mint output of Constantinople and also Nicomedia, Cyzicus, and hessalonica point to a serious inancial crisis which must not be underestimated, despite the emperor’s expenditure with the occasion of his consulate. he coin inds from the Balkans concentrate on the eastern part, in the provinces adjacent to the Black Sea. he payment of the troops stationed in the Danubian fortresses seems to have been a serious problem. he numerous inds 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.166 Its absence is due to the low activity of the mints of Constantinople, Nicomedia, Cyzicus and hessalonica, already mentioned, and coincides with some serious military setbacks at the Danube frontier, menaced by the Avars who reached as far as Tomis, the capital of Scythia, besieged in 597–598.167 he low ebb in coin production during the last ive years of the century indicates that military payments in the Balkans were delayed; discontent was certainly building up in the frontier garrisons and would eventually turn into rebellion in 602. heophanes Confessor, though a later source, informs us about the deep inancial crisis of the state, forced to pay the soldiers’ salaries only one third in coin and the rest in commodities.168 hessalonica, 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. he 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 hessalonica. 169 Figure 4 represents a comparison between the coins from the collections (336 coins) and the inds from Scythia (248 coins). he 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 National Museum in Belgrade provides forty-six specimens from Justin II of which eighteen are dated 569/70 and thirteen, 574/5.170 165. Whitby (1988, 165–169). 166. Gândilă (2008, 311). 167. Madgearu (1996, 50) 168. See the discussion by Yannopoulos (1987, 129). 169. Popović (1975, 459–464) ; Popović (1978b, 622); Metcalf (1991, 142). 170. Radić and Ivanišević (2006, 132–135). Early Byzantine Coin Circulation in the Eastern Provinces 185 he year-by-year luctuations bring forth even more interesting facts about the hessalonican mint output ater 578. D. M. Metcalf maintained that 580 was a critical moment when the mint activity was virtually paralyzed.171 Figure 4 partly conirms Metcalf ’s assertion with respect to the mint’s inluence in the Balkans around year 580, although a few specimens are available in the Belgrade collection as well as in Scythia. While this might be true for the Balkans (579–582 provides a striking diference in the comparison chart), the major collections provide six specimens from 580/1 and less for the next three years. herefore, one can conclude that Metcalf ’s statement is true as a local feature and not necessarily a problem of mint output. he same goes with the theory introduced by Vladislav Popović who supposed that the mint activity at hessalonica virtually ceased in 585/6, and yet we ind four specimens in the collections and no less than ive (only single inds included) in Scythia.172 he most critical period in the mint’s activity occurred toward the turn of the century, 597–600, and is a more general phenomenon of the monetary economy. he statistical value of coin inds of Maurice in Anatolia is usually double the value established for the Balkans (Figures 5a–b). he frequent invasions and the general devastation of the Danube provinces are certainly among the major reasons for this striking diference. Much as in the Balkans, however, the numerous coin inds from urban sites like Sardis, Amaseia, Pisidian Antioch, Side, and Malatya usually form a continuous sequence until 595.173 he scarcity of coins from the later years and the similarity with the Balkans force us to conclude that economic reasons afecting the mint output should be held responsible for this situation and not military activity, pervasive in the Balkans, but almost nonexistent in most of Anatolia. Another distinctive characteristic is the large inluence 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 suicient information is more than 20 percent. Two prominent exceptions are Sardis and Amasra where the mint of Antioch is less signiicant or not present at all. In both towns, however, the coins from hessalonica represent an important proportion of the group, suggesting diferent channels of coin distribution in Anatolia, not necessarily based on geographic location (Figure 17b).174 he collection of the archaeology museum in Bolu (ancient Claudiopolis), in the ancient province of Bithynia, far from Antioch, has no coins of Maurice from hessalonica, but many from Antioch, amounting to 30 percent of the total. his is a general characteristic of the collection for the entire sixth 171. Metcalf (1991, 142). 172. Popović (1978b, 622); Gândilă (2008, 320). 173. Bates (1971, 67–78); Ireland (2000, 106–107); Atlan (1976, 84–86). 174. Bates (1971, 72–73); Ireland and Atesogullari (1996, 124). 186 Andrei Gândilă century. Jumping to the island of Cyprus we notice a combined inluence of Antioch and hessalonica at Salamis and Curium, no doubt because of the maritime dimension of both the towns and the mints.175 In Syria-Palestine the proportion of coins issued by Maurice is slightly higher than in Anatolia. Antioch itself provides a high number, due to the presence of the mint in the city and its inluence is also felt in the vicinity, if less overwhelming, at Hama, Déhès, Apamea, and Çatal Hüyük.176 Southward, on the coast, at Berytus and Caesarea, a large number of coins from Antioch are recorded among inds, perhaps a sign of commercial activities and also in towns from the Palestinian inland, such as Jerusalem and Nessana.177 he 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 the Balkans and Anatolia.178 he decline in coin circulation is now felt in some previously prosperous towns, most importantly at Pella and Gerasa. Kenneth Sheedy has explained this situation by the impoverishment of the two centers,179 while Alan Walmsley has suggested that state consignments ended in this period.180 It is hard to make any generalizations at the scale of the entire province of Arabia or Palaestina 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 inds from Maurice Tiberius are present. he lower denominations continued to be struck in limited numbers and their use was generally restricted to the urban economy (Figure 30b). Both the collections and site inds point to a sharp decline in the production of ⅛-folles, which remain abundant only among the inds from Constantinople and to a lesser degree at Sardis and Ephesus.181 he ¼-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 Déhès, in Syria.182 A new increase in coin output can be noticed during the reign of Phocas, which is another reason to reconsider the merits of his reign (Figure 2a). Historians in the past have relied perhaps too heavily on written sources biased against Phocas to describe his reign in overly negative terms. he collections prove that his coin175. Callot (2004, 48–51); Cox (1959, 80). 176. Waage (1952, 159–160); homsen (1986, 62); Morrisson (1980, 279); Balty (1984, 240–244); Vorderstrasse (2005, 498–499). 177. Butcher (2003, 271–273); Ariel (1986, 142–143); Evans (2006, 193–194); Ariel (1982, 326); Bellinger (1962, 73). 178. Noeske (2000, II); Morrisson et al. (2006). 179. Sheedy (2001, 52). 180. Walmsley (1999, 345). 181. Hendy (1986, 301–305; 2007, 213–215); Bates (1971, 67–78); Milne (1925, 390). 182. Isvoranu and Poenaru Bordea (2003, 153, table 3); Callot (2004, 51); Waage (1952, 159–160); Morrisson (1980, 279). Early Byzantine Coin Circulation in the Eastern Provinces 187 age was abundant, although it must be noted that a good proportion of his copper coinage is made of overstrikes, usually on coins of Maurice. his is not so much a case of damnatio memoriae as a direct result of a shortage of copper building up in the course of the sixth century as a consequence of inlation, hoarding, and casual loss. Chiely folles were overstruck and mainly in Constantinople, Nicomedia, and Cyzicus. Based on the evidence of the collections, the overstruck group represents on average c. 25 percent of the total, while the carefully published inds from Sardis reveal that c. 22 percent of the folles of Phocas are overstruck.183 he mint of Antioch is excluded from these calculations because it rarely overstruck its issues. Except for the collection in Paris,184 there is a strong indication of decline in mint output in the last regnal years, probably because of the turmoil created by the Heraclian revolt. he mint of Antioch continues to be the second most important ater Constantinople and its output can be related to the new ofensive initiated by the Persian king ater the deposition of Maurice. Surprisingly, the two major mints are closely followed by Cyzicus, which became extremely active ater having a minor role throughout the sixth century (Figure 18a). he military conditions in the Balkans become aggravated ater the rebellion of 602. Although the thesis of the collapse of the Danube limes in 602 is no longer tenable,185 numerous fortiications, particularly on the western sector, were severely afected. here is a marked diference in coin circulation between the settlements on the border, such as Capidava, Novae, Aquis, and Viminacium,186 where few or no coins of Phocas were found, and towns further away from the military operations, such as Tomis, Accres Castellum, Corinth, Athens, and Constantinople itself, where the volume of inds marks a visible increase compared to the reign of Maurice (Figure 5a).187 A very similar tendency can be observed in Anatolia, where most of the urban settlements have yielded a large number of inds from Phocas (Figure 5b). One major exception is Melitene situated close to the front line ater the Persians had occupied the main strategic towns in Upper Mesopotamia. he mint of Antioch, less visible in the Balkans, covers more than 30 percent of the 183. Bates (1971, 67–78). 184. he collection of the Bibliothèque Nationale shows a prominent peak in 609/10, apparently inexplicable (Figure 3). A closer examination of the coins’ provenance reveals the fact that all coins belonged to the Schlumberger collection; most of them were minted in Antioch and were purchased in Aleppo by the French scholar and might be part of a hoard. 185. Barnea (1990). 186. Gândilă (2008, 322–323, table 4); Dimitrov (1998, 111); Janković (1981, 66, annex 3); Ivanišević (1988, 94). 187. Gândilă (2008, 322–323, table 4); Bellinger (1925, 46–47); Edwards (1933, 130–131); Fisher (1984, 245); Mac Isaac (1987, 136); hompson (1954, 69–70); Hendy (1986, 305– 308); Hendy (2007, 216–218). 188 Andrei Gândilă coins found in Anatolia, an important increase compared to the reign of Maurice. As expected, the inds from the Near Eastern provinces show an even more pronounced inluence of Antioch, especially at Gerasa and Caesarea, probably itself a sign of the threat posed by the Persian armies (Figure 18b).188 However, the inluence of Antioch stops being so pervasive in the Near East. At Hama, Jerusalem, and Nessana, where Antioch had always been an important supplier of fresh coins, no such inds have been reported.189 his 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 Anatolia, the volume of fresh coins stagnates or decreases during the reign of Phocas (Figure 5c). he extreme scarcity of coins from hessalonica, which had been a minor but steady supplier of coins since the reign of Justin II, is possibly another sign of decline. he low number of smaller denominations adds to this picture of fall in coin circulation (Figure 31b). Besides Antioch, no other urban center in Syria-Palestine provides such issues.190 he rare hoard of small change found in Aleppo could originate from the circulating medium of the great Syrian metropolis.191 By comparison, Tomis and Odartsi in the Balkans, 192 Sagalassos, Pisidian Antioch, and Sardis in Anatolia,193 and Salamis in Cyprus194 continued to receive lower denominations, ¼-folles and even ⅛-folles. It is oten considered that the long reign of Heraclius marks the end of Antiquity. he empire lost its eastern provinces to the hands of the Arabs and its inluence in the Balkans to the repeated attacks of Slavs and Avars. By necessity, this comparative analysis has to end with the irst decade of Heraclius’ reign, when the coin circulation drops in Scythia, our main element of comparison with the museum collections. he irst years of reign are characterized by an abundant coinage with a high peak reached in 612–614 (Figure 3). It coincides with an important change in iconography, the frontal bust of the emperor being replaced with the standing igures of Heraclius and his son Heraclius Constantine. One could argue that the successive changes in iconography are the reason why the collections possess so many coins from this time interval. his is not the case, since the abundant material from Anatolia and Syria shows a very similar peak in the same years and is, of course, un188. Bellinger (1938, 114–116); Marot (1998, 481–483); Ariel (1986, 143); Evans (2006, 195–196). 189. homsen (1986, 62); Ariel (1982, 326); Bellinger (1962, 73–74). See, however, the group of coins purchased in Jerusalem in 1963 in which 30 percent of the coins of Phocas come from Antioch (Metcalf and Payne 1965, 209). 190. Waage (1952, 161–162). 191. Mansield (2003, 354–355). 192. Isvoranu and Poenaru Bordea (2003, 153, table 3). 193. Scheers (1995, 314); Bates (1971, 88–89). 194. Callot (2004, 52–53). Early Byzantine Coin Circulation in the Eastern Provinces 189 afected by selection in the hands of collectors or museum curators (Figure D).195 Even from these early years of reign, the complexity of the early Byzantine monetary system began to slowly break down. he mint of Antioch was closed, never to be reopened for regular issues, while the mints in Cyzicus (615/6) and Nicomedia (618/9) were temporarily shut-down. he multi-denominational system ceased to be functional, ater the mints stopped issuing ⅛-folles and concentrated c. 90 percent of their activity on the production of folles (Figure 32a). Over 90 percent of the coins from the irst six years of reign are overstruck, especially the ones belonging to the second type of Grierson’s classiication.196 he “conversion” of large quantities of coins issued by Maurice and Phocas afects 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 his reign. he Balkans witness a severe downfall in coin circulation in the second decade of the seventh century. Isolated spots of Byzantine control usually located around urban 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 anything but the last payments sent to the small garrisons still holding the Empire’s position at the Danube.197 Urban life continued its course to some extent at Tomis and Mesembria and especially at Corinth and Athens, to name the most important centers (Figure 5a),198 but later in the century the eastern Balkans would be menaced by a new and long lasting enemy of the Byzantine state, the Bulgars. Most of the coins found in the Balkans were issued either in Constantinople and Nicomedia, while Cyzicus and hessalonica have a negligible presence (Figure 19b). As expected, most of them are folles, over 80 percent on average, and ½-folles; ¼-folles are lacking, even among inds from Constantinople (Figure 32b). In Anatolia we encounter a totally diferent situation. he coins of Heraclius are among the most common EBC found during excavations or in local archaeology museums. In statistical terms they usually represent between 20 and 40 percent of the total number of inds, which is in sharp contrast to the picture ofered by the Balkans. Excavations at Sardis and Side yielded a particularly high number of inds (Figure 5b), while the coins preserved in the Bolu and Isparta museums 195. Bell (1916, 82–95); Bates (1971, 95–109). 196. DOC II/1, 226. 197. Oberländer-Târnoveanu (1996, 97–127); Gândilă (2006–2007, 118); Gândilă (2003– 2005, 140); Dimitrov (1998, 111); Ivanišević (1988, 94). 198. Gândilă (2008, 322–323, table 4); Edwards (1933, 131–132); Mac Isaac (1987, 136); hompson (1954, 70). 190 Andrei Gândilă point to a massive inlux of coins in the respective areas.199 It should be noted however that the abundance of coins in these early years predates the Persian invasion, which initially afected the south-eastern part of Anatolia but soon got deeper into the Byzantine heartland where the Persians sacked Caesarea, Ancyra, and Sardis and took the island of Rhodes.200 he abundance of early Heraclian issues is most striking in the islands, in Cyprus,201 at Salamis, Curium, and Paphos and on Samos202 in the Aegean, where an impressive number of coins have been recovered from the Tunnel of Eupalinos, among which are some rare ¼-folles. he developments in Cyprus have been ascribed to an increased strategic importance of the island ater Antioch was occupied by the Persians, which might also explain the ephemeral presence of an oicial mint on the island.203 he irst decades of the seventh century brought an unprecedented series of invasions led by the Persians and later Arabs which sealed the fate of the Byzantine provinces in Syria-Palestine. he increased number of hoards testiies to the growing insecurity in the area ater 602.204 Site inds, however, provide us with a mixed picture. Relatively few early coins of Heraclius found their way into Antioch and Apamea,205 the two major cities of the region, the inds being two or even three times fewer than in Anatolia, somewhat resembling the situation encountered in the northern Balkans. Surprisingly, at Hama206 the inds are much more numerous and correspond with an unexpected period of reconstruction late in the sixth century.207 In Palestine we notice a sensible increase in coin inds, at Caesarea, Jerusalem, and Pella,208 but they are by no means characteristic of the region as a whole (Figure 5c).209 199. Bates (1971, 95–109); Bell (1916, 82–95); Atlan (1976, 88–92). 200. Foss (1975, 721–747). 201. Callot (2004, 54–75); Cox (1959, 80–81); Nicolaou (1990, 194–199). 202. Jantzen (2004, 156, 160). 203. Metcalf (2001, 135). 204. he relevant hoards are Khirbet Dubel, Tell Bissé, Baalbek, Khirbet Fandaqumya, Khirbet Deir Dassawi, Cyrrhus, “Syria,” and Deir Dassawi Noeske (2000, II); Quazrin Ariel (1996); “Northern Syria” Mansield (2003); “Lebanon” Kruszynski (1999), and probably also three hoards with uncertain Near Eastern provenance buried ater 602, Mansield (1995); Naismith (2004); For a recent catalogue of hoards from Palestine see Waner and Safray (2001). 205. Waage (1952, 162–164); Balty (1984, 240–245). 206. homsen (1986, 62–63). 207. Foss (1997, 259). 208. Evans (2006, 197–198); Ariel (1982, 326); Sheedy (2001, 139–141). 209. Few or no early Heraclian coppers have been reported among the fairly large number of inds from Hammat Gader, Chorazin, Tel Jezreel, and Samaria, see Barkay (1997, 279– 300); Kloetzli (1970, 367–369); Moorhead (1997, 162–163); Fulco and Zayadine (1981, 221–223). Early Byzantine Coin Circulation in the Eastern Provinces 191 C he statistical validity of large collections of EBC can no longer be overlooked. he comparison of ive major collections with the numerous inds from the province of Scythia and the hoards from the Near East has revealed a number of quantitative similarities, which need to be addressed. he purpose of analyzing the collections from a statistical perspective is not to provide us with absolute igures. Unfortunately the statistical tools oten used in numismatics are far more sophisticated and precise than the sampled evidence, which is most of the time fragmentary. here are too many lacunae in our knowledge of the monetary policies conducted by Early Byzantium to attempt any deinitive propositions. he nature of the evidence and the inherent methodological limitations are an invitation to caution.210 Any statistical results will need to be conirmed and re-conirmed by future evidence before attempting any conclusive remarks. Relative luctuations can, however, be discerned at this point and the large number of copper coins in the major museum collections ofer 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 more diicult and much more needs to be done in the realm of interpretation to explain such luctuations. Such an understanding cannot be accomplished only by studying the de-contextualized coins from the public collections. Archaeological excavations in the Balkans, Anatolia, and the Near East ofer the most promising perspectives for understanding regional patterns in a comparative fashion. Nonetheless, the high correlation of the major collections of EBC remains instrumental for a better understanding of annual coin production. Where signiicant site inds 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, including some of my own conclusions regarding the coin circulation in Dobrudja, 211 need to be reassessed, because “abnormal” levels in coin circulation can not be properly detected and understood without basic knowledge of the “normal” pattern. It will soon become apparent that low points in the statistical curve of a region are very oten relections of coin production and distribution at the center and less the result of provincial developments alone.212 210. See Robertson (1989) for a methodological discussion and an invitation to caution in the case of Roman coins, which could be easily extrapolated to the Byzantine period as well. 211. Gândilă (2003–2005). 212. he activity of provincial mints should be part of the general explanation regarding local particularities. Indeed the mint of hessalonica will heavily inluence the coin distribution in the western Balkans, while the mint of Antioch will have a similar efect in Syria 192 Andrei Gândilă I have shown in the previous sections that, in spite of the still insuicient evidence, an 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 diferent levels of monetization and economic integration. Studying coin circulation in a single province without reference to the circulating medium in other corners of the Empire can lead to false generalizations and unreliable interpretations of the numismatic material. he coin inds from a major urban site will not inform us suiciently about coin circulation in the whole province, while the inds from a province will not be necessarily relevant for an entire region. herefore, I have tried to paint, perhaps in overly broad strokes, a comparative tryptich of circulation, with one panel devoted to the Balkans and two others for Anatolia and Syria-Palestine in the hope that future studies will soon correct and improve this provisional efort. he broad outlines drawn for Anatolia, and to a certain extent for the Near East, are subject to change as new evidence surfaces. At least in the northern Balkans the current body of evidence is large enough to insure the stability of present analyses although, of course, at more detailed level the overall scheme will certainly require minor adjustments. he common features observed in many urban centers of the Near Eastern provinces are likely to endure, although new data is expected to color the grey areas in the big picture and perhaps to bring more homogeneity in what seems like a very diverse landscape. Anatolia is by far the most sensitive region to future developments and constitutes the most promising avenue to test both the uniting and the distinctive features of coin circulation in the major geographical units of the Eastern Empire. A: I wish to express my gratitude to Dr. Florin Curta for reading and critiquing several drats; Ralf Althof, Gabriela Bijovsky, Dr. Zeliha Demirel Gökalp, Dr. Ernest Oberländer-Târnoveanu, and Dr. Alan Stahl for commenting on the inal drat and for providing valuable information and access to unpublished material from the Balkans, Anatolia, and Palestine, and to the curatorial staf of the American Numismatic Society for allowing me to study their rich collection of Byzantine coins. Sections of this paper were presented in 2009 at the International Congress on Medieval Studies, Kalamazoo, at the Summer Seminar of the American Numismatic Society, New York, and at the International Numismatic Congress, Glasgow. I thank all these audiences for their questions and comments. (Figures 6–19). Given the diferent channels of distribution, towns from Syria and Macedonia will certainly difer to some extent in their circulation patterns. Early Byzantine Coin Circulation in the Eastern Provinces 193 A ANS = American Numismatic Society BMC = Wroth, W. Catalogue of the imperial Byzantine coins in the British Museum, vol. 1. London, 1908. BNP = Morrisson, C. Catalogue des monnaies byzantines de la Bibliothèque Nationale (491–1204). Tome prémier: D’Anastase à Justinien II (491–711). Paris: Bibliothèque Nationale, 1970. DOC = Bellinger, A. R., and P. Grierson, eds. 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. KOD = Althof, R. Sammlung Köhler-Osbahr. Band V/1–2. Byzantinische Münzen und ihr Umfeld. Duisburg, Kultur und Stadthistorisches Museum, 1998– 1999. MIB = Hahn, W. Moneta Imperii Byzantini, 3 vol. Vienna: Verlag der Österreichischen Akademie der Wissenschaten, 1973–1981. Ratto = Ratto, R. Monnaies byzantines et d’autres pays contemporains à l’époque Byzantine. Lugano, 1930; reprint by J. Schulman. Amsterdam, 1959. R Ariel, D. T. 1982. A survey of coin inds in Jerusalem. Liber Annuus 32: 273–326. ———. 1986. he coins. In Excavations at Caesarea Maritima, 1975, 1976, 1979: Final Report, L. I. Levine and E. Netzer, eds., pp. 137–148. Jerusalem: Hebrew University of Jerusalem. ———. 1987. Coins from the synagogue at ‘En Nashut. Israel Exploration Journal 37: 147–156. ———. 1996. A hoard of Byzantine folles from Qazrin. ‘Atiqot 29: 69–76. Arslan, E. A. 2000. Catalogo delle monete bizantine del Museo Provinciale di Catanzaro. Catanzaro: Amministrazione Provinciale di Catanzaro. Ashton, R., C. Lightfoot, and A. Özme. 2000. Ancient and Medieval coins in Bolvadin (Turkey). Anatolia Antiqua 8: 171–195. Atlan, S. 1976. 1947–1967 Yılları Side kazıları sırasında elde edilen sikkeler. Ankara: Türk Tarih Kurumu Basımevi. Baldwin’s. 1995. Byzantine gold coins from the P. J. Donald collection, October 11. London. Balty, J. 1984. Monnaies Byzantines des maisons d’Apamée: étude comparative. In Apamée de Syrie: bilan des recherches archéologiques 1973–1979: aspects de l’architecture domestique d’Apamée: actes du colloque tenu à Bruxelles les 29, 30 et 31 mai 1980, J. Balty, ed., pp. 239–248. Bruxelles: Centre belge de recherches archéologiques à Apamée de Syrie. 194 Andrei Gândilă Baramki, J. 1939. A hoard of Byzantine coins. Quarterly of the Department of Antiquities in Palestine 8: 81–85. Barkay, R. 1997. Roman and Byzantine coins. In he Roman baths of Hammat Gader. Final Report, Y. Hirschfeld et al., pp. 279–300. Jerusalem: he Israel Exploration Society. ———. 2000. he coins of Horvat ‘Eleq. In Ramat Hanadiv excavations. Final report of the 1984–1998 seasons, Y. Hirschfeld et al., pp. 377–419. Jerusalem: he Israel Exploration Society. Barnea, Al. 1990. Einige Bemerkungen zur Chronologie des Limes an der unteren Donau in spätrömischer Zeit. Dacia 34: 283–290. Bates, G. E. 1971. Byzantine coins. Cambridge MA: Harvard University Press. ———. A Byzantine Coin Collection, Boston: Privately Printed, 1981. Bates, M. L., and F. L. Kovacs. A hoard of large Byzantine and Arab-Byzantine coppers. Numismatic Chronicle 156: 165–173. Bateson, J. D, and I.G. Campbell. Byzantine and Early Medieval Western European Coins in the Hunter Coin Cabinet. London: Spink, 1998. Beliën, P. 2005. A hoard of Byzantine folles from Beirut. Numismatic Chronicle 165: 314–322. Bell, H. W. 1916. Sardis, vol. XI, part I, 1910–1914: Coins, 76–95. Leiden: Brill. Bellinger, Alfred R. 1930. Catalogue of the coins found at Corinth, 1925. New Haven: Yale University Press. Bellinger, A. R. 1962. Coins. In Excavations at Nessana, H. D. Colt, ed., pp. 70–75. London: British School of Archaeology in Jerusalem. ———. 1938. Coins from Jerash, 1928–1934. New York: American Numismatic Society. Bendall, S. 2002. A neglected nineteenth century numismatist. Numismatic Circular 110: 261–264. Betlyon, J. W. 1988. Coins from the 1985 Season of the Limes Arabicus Project. In Preliminary Report on the 1985 Season of the Limes Arabicus Project, S. T. Parker, ed., pp. 162–174. Bulletin of the American Schools of Oriental Research. Supplemental Studies 25. Bijovsky, G. 1998. he Gush Halav hoard reconsidered. ‘Atiqot 35: 77–106. ———. 2002. he Coins. In S. Agady, M. Arazi, B. Arubas, S. Hadad, E. Khamis, and Y. Tsafrir, he Bet Shean Archaeological Project, 507–512. In What Athens has to do with Jerusalem. Essays on classical, Jewish, and Early Christian art and archaeology in honor of Gideon Foerster, Leonard V. Rutgers, ed., pp. 423–534. Leuven: Peeters. Blake, R. P. 1942. he monetary reform of Anastasius and its economic implications. In Studies in the history of culture, pp. 84–97. Freeport NY: Books for Libraries Press. Early Byzantine Coin Circulation in the Eastern Provinces 195 Bonham’s. 1980. A catalogue of standard Byzantine and Dark Age gold coins, December 3. London. Boutin, S. Collection N.K. Monnaies des Empires de Byzance. Wetteren: Cultura, 1983. Butcher, K. 2003. Archaeology of the Beirut Souks 1. Small change in ancient Beirut: the coin inds from BEY 006 and BEY 045: Iron Age, Hellenistic, Roman and Byzantine periods. Berytus 45–46, 2001–2002. Beirut: American University of Beirut. Buttrey, T. V. 1981. Byzantine, medieval and modern coins and tokens. In Greek, Roman, and Islamic coins from Sardis, T. V. Buttrey, A. Johnson, K. M. Mac Kenzie, and M. L. Bates, pp. 204–224. Cambridge: Harvard University Press. ———. 1993. Calculating ancient coin production: facts and fantasies. Numismatic Chronicle 153: 335–351. ———. 1994. Calculating ancient coin production II: why it cannot be done. Numismatic Chronicle 154: 341–352. ———, and S. E Buttrey. 1997. Calculating ancient coin production, again. American Journal of Numismatics 9: 113–135. de Callataÿ, F. 1995. Calculating ancient coin production: seeking a balance. Numismatic Chronicle 155: 289–311. Callegher, B. Catalogo delle monete bizantine, vandale, ostrogote e longobarde del Museo Bottacin. Vol. I. Padova: Comune di Padova, Musei e biblioteche, 2000. Callot, O. 2004. Salamine de Chypre. XVI Les monnaies. Fouilles de la ville 1964– 1974. Paris: Boccard. Callu, J.-P. 1982. Le tarif d’Abydos et la reforme monétaire d’Anastase. Proceedings of the 9th International Congress of Numismatics, Berne, September 1979, T. Hackens and R. Weiller, eds., pp. 731–740. Louvain-la-Neuve: Association internationale des numismates professionnels. Carcasonne, C. 1987. Methodes statistiques en numismatique. Louvain-la-Neuve: Séminaire de Numismatique Marcel Hoc. ———, and J. Guey. 1978. Valeur statistique des petits échantillons. Revue Belge de Numismatique 124: 5–21. ———, and T. Hackens. 1981. Statistique et numismatique: table ronde organisée par le Centre de mathématique sociale de l’Ecole des hautes études en sciences sociales de Paris et le Séminaire de Numismatique Marcel Hoc de l’Université Catholique de Louvain, Paris, 17–19 sept. 1979. Strasbourg: Conseil de l’Europe, Assemblée parlementaire. Casey, P. J. 1996. Justinian, the limitanei, and Arab-Byzantine relations in the 6th c. Journal of Roman Archaeology 9:214–222. Christie’s. 1986. he Goodacre collection of Byzantine coins, April 22. London. 196 Andrei Gândilă Cox, D. 1959. Coins from the excavations at Curium, 1932–1953. Numismatic Notes and Monographs 145. New York: American Numismatic Society. Curta, F. 1996. Invasion or inlation? Sixth to seventh century Byzantine coin hoards in Eastern and Southeastern Europe. Annali di Istituto Italiano di Numismatica 43: 65–224. Delougaz, P. 1960. Coins. In A Byzantine church at Khirbat Al-Karak, P. Delougaz and R. C. Haines, pp. 50–52. Chicago: University of Chicago Press. Demirel Gökalp, Z. 2007. Yalvaç ve Isparta arkeoloji müzelerinde bulunan Bizans sikkeleri. PhD dissertation. Anadolu Üniversitesi Sosyal Bilimler Enstitüsü. Devreker, J. 1984. Les monnaies de Pessinonte. In Les Fouilles de la Rijksuniversiteit te Gent a Pessinonte 1967–1973, J. Devreker and M. Waelkens, eds., vol. I, pp. 173–215. Brugge: De Tempel. Dimitrov, K. 1998. Poznorzymskie i wczesnobizantyjskie monety z odcinka iv w Novae z lat 294–612. Novensia 11: 99–112. Duncan-Jones, R. 1994. Money and government in the Roman Empire. New York: Cambridge University Press. Edwards, K. M. 1933. Corinth VI: Coins, 1896–1929. Cambridge, MA: Harvard University Press. ———. 1937. Report on the coins found in the excavations at Corinth during the years 1930–1935. Hesperia 6 (2): 241–256. Erslev, K. ed. 1873. Catalogue de la collection de monnaies de feu Christian Jürgensen homsen. Seconde partie: Les monnaies du moyen-age, tome I. Copenhagen: Imprimerie de hiele. Esty, W. 1986. Estimating the size of a coinage. Numismatic Chronicle 146: 185–215. Evans, DeRose J. 2006. he joint expedition to Caesarea Maritima. Excavation reports. Volume VI. he coins and the Hellenistic, Roman, and Byzantine economy of Palestine. Boston: American Schools of Oriental Research. Fisher, J. E. 1984. Corinth excavations, 1977, forum southwest. Hesperia 53 (2): 217–250. Fitzgerald, G. M. 1929. he coins. In Excavations in the Tyropoeon Valley, Jerusalem 1927, J. W. Crowfoot and G. M. Fitzgerald, pp. 103–132. London: Palestine Exploration Fund. Foss, C. 1975. he Persians in Asia Minor and the end of Antiquity. he English Historical Review 90: 721–747. ———. Syria in transition. A.D. 550–750: an archaeological approach. Dumbarton Oaks Papers 51: 189–269. Friedländer, J., and M. Pinder. Die Münzen Justinians. Berlin, 1843. Fulco, W. J., and F. Zayadine. 1981. Coins from Samaria-Sebaste. Annual of the Department of Antiquities Jordan 25: 197–225. Early Byzantine Coin Circulation in the Eastern Provinces 197 Füeg, F. 1991. Die Solidusausgaben 717–803 in Konstantinopel. Revue suisse de numismatique 70: 35–54. Gândilă, A. 2003–2005. Sixth-to-seventh century coin circulation in Dobrudja. Cercetări Numismatice 9–11: 109–166. ———. 2006–2007. Early Byzantine Capidava: the numismatic evidence. Cercetări Numismatice 12–13: 97–122. ———. 2008. 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, I. Lazarenko, ed., vol. 1, pp. 301–330. Varna, 2008. Grierson, P. 1961. Coinage and money in the Byzantine Empire 498–c. 1090. In Moneta e scambi nell’alto medioevo, pp. 411–453. Spoleto: Presso La Sede del Centro. ———. 1965. he interpretation of coin inds. Numismatic Chronicle 5: i–xiii. ———. 1966. he interpretation of coin inds (2). Numismatic Chronicle 6: i–xxi. ———. 1967a. Byzantine coinage as source material. In Proceedings of the XIII International Congress of Byzantine Studies, Oxford, 5–10 September 1966, pp. 317–333. London: Oxford University Press. ———. 1967b. he monetary reforms of Anastasius and their economic consequences. In International Numismatic Convention, Jerusalem 1963; he patterns of monetary development in Palestine and Phoenicia in Antiquity, A. Kindler, ed., pp. 283–302. Tel-Aviv/Jerusalem: Schocken. ———. 1982. Byzantine coins. Los Angeles: University of California Press. ———. 1986. Circolazione monetaria e tesaurizzazione. In La Cultura bizantina, oggetti e messaggio: moneta ed economia, A. Guillou, ed., pp. 37–57. Rome: L’erma di Bretschneider. ———. 1998. Memoir on the Coin Room, Dumbarton Oaks Center for Byzantine Studies. Hahn, W., and M. A. Metlich. 2000. Money of the incipient Byzantine Empire. Vienna: City Press. Hamburger, H. 1956. Minute coins from Caesarea. ‘Atiqot 1: 115–138. Hendy, M. F. 1985. Studies in the Byzantine monetary economy c. 300–1450. London: Cambridge University Press, 1985. ———. 1986. he coins. In Excavations at Saraçhane in Istanbul, R.M Harrison, M. V. Gill, M. Hendy, S. J. Hill and D. Brothwell, vol. I, pp. 278–313. Princeton: Princeton University Press. ———. 2007. Roman, Byzantine and Latin Coins, in Kalenderhane in Istanbul. he excavations, edited by Cecil L. Striker and Y. Doğan Kuban, pp. 175–276. Mainz: Verlag Philipp von Zabern. 198 Andrei Gândilă Hennequin, G, and A. Abū-l-Faraj. 1978. Les monnaies de Bālis. Damascus: Institut Français de Damas. Hohlfelder, R. L. 1978. Kenchreai, eastern port of Corinth. III. he coins. Leiden: Brill. Iordanov, I., A. Koicev, and V. Mutafov. 1998. Srednovekovijat Ahtopol VI–XIII v. spored dannite numizmatikata i sfragistika. Numizmatika i Sfragistika 5, n. 2: 67–89. Ireland, S. 2000. Greek, Roman and Byzantine coins in the museum at Amasya. London: Royal Numismatic Society. ———, and S. Atesogullari. 1996. he Ancient coins in Amasra Museum. In Studies in ancient coinage from Turkey, R. Ashton, ed., pp. 115–137. London: Royal Numismatic Society. Isvoranu, T., and Gh. Poenaru Bordea. 2003. Monede bizantine de la Tomis și împrejurimi în colecția Institutului de Arheologie Vasile Pârvan. In Simpozion de numismatică dedicat împlinirii a 125 de ani de la proclamarea independenţei României, Chişinău, 24–26 septembrie 2002, Comunicări, studii şi note, pp. 137– 161. Bucharest: Editura Enciclopedică. Ivanišević, V. 1988. Vizantijski novac (491–1092) iz zairke Narodnog Myzeja y Pojarevci. Numizmatičar 11: 87–99. Janković, Đ. 1981. Podunavski deo oblasti Akvisa u VI i pocetkom VII veka. Belgrade: Arheoloski institut. Jantzen, U. 2004. Die Wasserleitung des Eupalinos. Die Funde. Bonn: Rudolf Habelt GmbH. Katsari, C. 2003. he Statistical analysis of stray coins in museums: the Roman Provincial coinage. Nomismatika Chronika 22: 47–52. Khazdan, A.P. 1954. Vizantijskie goroda v VII–XI vv. Sovestkaja archeologija 21: 164–183. Kindler, A. 1989. he synagogue treasure of Meroth, Eastern Upper Galilee, Israel. In Proceedings of the 10th International Congress of Numismatics, London, September 1986, I. A. Carradice, ed., pp. 315–320. Wetteren: Cultura. Kloetzli, G. 1970. Coins from Chorazin. Liber Annuus 20: 359–369. Kruszynski, M. 1999. A group of Byzantine coins from Lebanon. Notae Numismaticae 3–4: 221–242. Lampinen, P. 1992. he coins, preliminary report, 1990. In Caesarea papers: Straton’s Tower, Herod’s Harbour, and Roman and Byzantine Caesarea, R. L. Vann, ed., pp. 169–172. Ann Arbor, MI: University of Michigan. Leuthold, E. 1952–1953. Monete byzantine rinvenute in Syria. Rivista Italiana di Numismatica i Scienze Aini 54–55: 31–49. ———. 1971. Monete byzantine rinvenute in Cirrestica. Rivista Italiana di Numismatica i Scienze Aini v. 73: 9–23. Early Byzantine Coin Circulation in the Eastern Provinces 199 Lightfoot, C. 2002. Byzantine Anatolia: reassessing the numismatic evidence. Revue Numismatique 158: 229–239. Mac Isaac, J. D. 1987. Corinth: coins, 1925–1926. he heater District and the Roman Villa. Hesperia 56 (2): 97–157. ———. 2005. Early Christian and later coin inds from Nemea. In Excavations at Nemea III. he coins, J. D. Mac Isaac and R. Knapp, pp. 183–237. Los Angeles: University of California Press. Madgearu, A. 1996. he province of Scythia and the Avaro-Slavic invasions (576– 626). Balkan Studies 37 (1): 35–61. Mansield, S. J. 1995a. Unknown (Near East), 1994 or before. Numismatic Chronicle 155: 348–354. ———. 1995b. Unknown (Near East), 1993 or before. Numismatic Chronicle 155: 354–358. ———. 2003. A hoard of twenty Byzantine copper coins. Numismatic Chronicle 163: 354–355. Marot, T. 1998. Las monedas del Macellum de Gerasa (Yaras, Jordania): aproximación a la circulación monetaria en la provincia de Arabia. Madrid: Museo Casa de la Moneda. Metcalf, D. M. 1958. Statistische Analyse bei der Auswertung von Münzfundmaterialen. Jahrbuch fur Numismatik und Geldgeschichte 9: 187–196. ———.1960. he currency of Byzantine Sirmia and Slavonia. Hamburger Beiträge zur Numismatik 14: 429–444. ———. 1964. Some Byzantine and Arab-Byzantine coins from Palaestina Prima. Israel Numismatic Journal 2 (3–4): 32–47. ———.1969. he origins of the Anastasian currency reform. Chicago: Argonaut. ———. 1991. Avar and Slav invasions into the Balkan peninsula (c. 575–625): the nature of the numismatic evidence. Journal of Roman Archaeology 4: 140–148. ———. 1976. he copper coinage of hessalonica under Justinian I. Wien: Verlag der Osterreichischen Akademie der Wissenschaten. ———.2001. Monetary recession in the Middle Byzantine period: the numismatic evidence. Numismatic Chronicle 161: 111–155. ———, and S. Payne. 1965. Some Byzantine and Arab-Byzantine coins obtained in Jerusalem. Numismatic Circular 73: 130–2, 185–6, 208–10; 257–58. Metcalf, W.E. 1975. A Heraclian hoard from Syria. Museum Notes 20: 109–137. ———. 1988. he joint reign gold of Justin I and Justinian I. In Studies in Early Byzantine Gold Coinage, Wolfgang Hahn and William E. Metcalf, eds., pp. 19–27. New York: American Numismatic Society. ———. 1995. Review of Duncan-Jones 1994. Revue Suisse de Numismatique 74: 145–159. Mihailov, S. 2008. Vidovete nominali v monetnoto obrashtenie na vizantiiskite provintsii Ckitiia i Vtora Miziia (498–681 g.). In Numismatic, sphragistic and 200 Andrei Gândilă epigraphic contributions to the history of the Black Sea coast, I. Lazarenko, ed., vol. 1, pp. 278–300. Varna. Milne, J. G. 1925. J. T. Wood’s coins from Ephesus. Numismatic Chronicle: 385–391. Mionnet, T. E. 1827. De la rareté et du prix des médailles romaines. Volume II, 2nd ed. Paris. Mirnik, I. and A. Šemrov. 1997–1998. Byzantine Coins in the Zagreb Archaeological Museum numismatic collection. Anastasius I (A.D. 497–518)-Anastasius II (A.D. 713–715). Vjesnik Arheološkog muzeja u Zagrebu 30–31: 129–258. Moorhead, T. S. N. 1997. he Late Roman, Byzantine and Umayyad periods at Tel Jezreel. Tel Aviv 24 (1): 129–166. Morrisson, C. 1980. Les monnaies. In Déhès (Syrie du nord) campagnes I–III (1976– 1978), recherches sur l’habitat rural, J.-P. Sodini, G. Tate, B. Bavant, S. Bavant, J.-L. Biscop, D. Orssaud, and C. Morrisson, pp. 267–287. Paris: Librairie Orientaliste Paul Geuthner. ———. 1981. Estimation du volume des emissions de Solidi de Tibère et Maurice à Carthage (578–602). In Statistique et numismatique: table ronde organisée par le Centre de mathématique sociale de l’Ecole des hautes études en sciences sociales de Paris et le Séminaire de Numismatique Marcel Hoc de l’Université Catholique de Louvain, Paris, 17–19 sept. 1979, C. Carcassone and T. Hackens, eds., pp. 267–284. Strasbourg: Conseil de l’Europe, Assemblée parlementaire. ———. 1988. he Moneta Auri under Justinian and Justin II, 537–578. In Studies in Early Byzantine gold coinage, Wolfgang Hahn and William E. Metcalf, eds., pp. 41–64. New York: American Numismatic Society. ———. 1989a. La monnaie en Syrie Byzantine. In Archéologie et histoire de la Syrie II. La Syrie de l’époque achéménide à l’avènement de l’Islam, J.-M. Dentzer and W. Orthmann, eds., pp. 187–204. Saarbrücken: Saarbrücker Druckerei und Verlag. ———. 1989b. Monnaie et prix a Byzance du Ve au VIIe siècle. In Hommes et richesses dans l’Empire byzantin, tome I, IVe-VIIe siècle, 239–260. Paris: Lethielleux. ———. 1993. Die byzantinischen Münzen. In H. Voegtli, Pergamenische Forschungen 8: Die Fundmünzen aus der Stadtgrabung von Pergamon, 8–13. Berlin: Walter de Gruyter. ———. 1995. La difusion de la monnaie de Constantinople: routes commerciales ou routes politiques? In Constantinople and its hinterland, C. Mango and G. Dagron, eds., pp. 77–89. Aldershot: Ashgate. ———. 1998. La circulation monétaire dans les Balkans à l’époque justinienne et post- justinienne. In Acta XIII Congressus internationalis archaeologiae christianae, N. Cambi, and E. Marin, eds., vol. II, pp. 919–930. Split, 1998. ———. 2001. La donation Schlumberger (1929). In Trois donations byzantines au Cabinet des Médailles: Froehner (1925), Schlumberger (1929), Zacos (1998), D. Feissel, C. Morrisson, J.-C., Cheynet and B. Pitarkis, eds., pp. 21–50. Paris: Bibliothèque Nationale. Early Byzantine Coin Circulation in the Eastern Provinces 201 ———. 2002. Byzantine money: its production and circulation. In Economic history of Byzantium, Angeliki Laiou, ed., pp. 909–966. Dumbarton Oaks. ———, and V. Ivanišević. 2006. Les emissions des VIe-VIIe siècles et leur circulation dans les Balkans. In Les Trésors monétaires byzantins des Balkans et d’Asie Mineure (491–713), C. Morrisson, V. Popović, and V. Ivanišević, eds., pp. 41–73. Paris: Lethellieux. ———, V. Popović, and V. Ivanišević, eds. 2006. Les Trésors monétaires byzantins des Balkans et d’Asie Mineure (491–713). Paris: Lethellieux. Mosser, S. McA. 1935. A bibliography of Byzantine coin hoards. Numismatic Notes and Monographs 67. New York: American Numismatic Society. Naismith, R. 2004. A hoard of Byzantine copper coins ending with the last year of Maurice. Numismatic Chronicle 164: 296–299. Nicolaou, I. 1990. Paphos II. he coins from the House of Dionysos. Nicosia: Cosmos Press. Noeske, H.-C. 2000. Münzfunde aus Ägypten I. Die Münzfunde des ägyptischen Pilgerzentrums Abu Mina und die Vergleichsfunde aus den Diocesen Aegyptus und Oriens vom 4.-8. Jh. n.Chr. 3 vols. Berlin: Mann. Oberländer-Târnoveanu, E. 1996. Monnaies byzantines des VIIe-Xe siècles découvertes a Silistra dans la collection de l’Académicien Péricle Papahagi consevées au Cabinet des Medailles du Musée National d’Histoire de Roumanie. Cercetări Numismatice 7: 97–127. ———. 2003. La monnaie dans l’espace rural byzantin des Balkans Orientaux—un essai de synthèse au commencement du XXIe siècle. Peuce 14: 341–412. Ostrogorsky, G. 1959. Byzantine cities in the early middle ages. Dumbarton Oaks Papers 13: 45–66. Phillips, M, and S. Tyler-Smith. 1998. A sixth-century hoard of nummi and ivenummi pieces. Numismatic Chronicle 158: 316–324. Poenaru Bordea, Gh. 1981. Problèmes historiques de la Dobroudja (VIe-VIIe siècles) a la lumiére des monnaies byzantines traitées par des méthodes statistiques. In Statistique et numismatique: table ronde organisée par le Centre de mathématique sociale de l’Ecole des hautes études en sciences sociales de Paris et le Séminaire de Numismatique Marcel Hoc de l’Université Catholique de Louvain, Paris, 17–19 sept. 1979, C. Carcassone and T. Hackens, eds., pp. 365–377. Strasbourg: Conseil de l’Europe, Assemblée parlementaire. ———, R. Ocheşeanu, and Al. Popeea, 2004. Monnaies Byzantines du Musée de Constanţa (Roumanie), Wetteren: Moneta. Popescu, E. 2005. Le village en Scythie Mineure (Dobroudja) à l’époque protobyzantine. In Les villages dans l’Empire byzantin (IVe-XVe siecle), J. Lefort, C. Morrisson, and J.-P. Sodini, eds., pp. 363–380. Paris: Lethielleux. Popović, V. 1975. Les témoins archéologiques des invasions avaro-slaves dans l’Illyricum byzantin. Melanges de l’école française de Rome 87 (1): 459–464. 202 Andrei Gândilă ———. 1978a. Catalogue des monnaies Byzantines du musée de Srem. In Sirmium VIII, Dj. Bošković, N. Duval, V. Popović, and G. Vallet, pp. 180–195. Rome/ Belgrade: École Française de Rome/Institut Archéologique de Belgrade. ———. 1978b. La descente des Koutrigours, des Slaves et des Avars vers la Mer Egée: Le témoinage de l’archéologie. Comptes Rendus de l’Academie des Inscriptions: 596–648. Pottier, H. 1983. Analyse d’un trésor de monnaies en bronze enfoui au VIe siècle en Syrie byzantine: contribution à la méthodologie numismatique. Wetteren: Cultura. ———. 2004. Nouvelle approche de la livre byzantine du Ve au VIIe siècle. Revue Belge de Numismatique et Sigillographie 150: 51–133. Prawdzic-Golemberski, E. J., and D. M. Metcalf. 1963. he circulation of Byzantine coins in the south-eastern frontiers of the Empire. Numismatic Chronicle 123: 83–92. Radić, V., and V. Ivanisević. 2006. Byzantine coins from the National Museum in Belgrade. Belgrade. Robertson, A. S. 1989. he accidents of survival. In Proceedings of the 10th International Congress of Numismatics, London, September 1986, I. A. Carradice, ed., pp. 315–320. Wetteren: Cultura. Russell, K. W. 1985. he earthquake chronology of Palestine and northwest Arabia from the 2nd through the mid-8th century A.D. Bulletin of the American Schools of Oriental Research 260: 37–59. Sabatier, J. 1862. Description générale des monnaies Byzantines. Paris, Chez Rollin et Feuardent. de Saulcy, F. 1836. Essai de classiication des suites monétaires Byzantines. Metz: Lamort. Scheers, S. 1993. Catalogue of the coins found in 1992. In Sagalassos II: Report on the third Excavation Campaign of 1992, M. Waelkens and J. Poblome, eds., pp. 249–260. Leuven: Leuven University Press. ———. 1995. Catalogue of the coins found in 1993. In Sagalassos III: Report on the fourth excavation campaign of 1993, M. Waelkens and J. Poblome, eds., pp. 307–326. Leuven: Leuven University Press. ———, H. Vanhaverbeke, and J. Poblome. 1997. Coins found in 1994 and 1995. In Sagalassos IV: report on the survey and excavation campaigns of 1994 and 1995, M. Waelkens and J. Poblome, eds., pp. 315–350. Leuven: Leuven University Press. Scheers, S. 2000. Coins found in 1996 and 1997. In Sagalassos V: report on the survey and excavation campaigns of 1996–1997, M. Waelkens and L. Loots, eds., pp. 509–549. Leuven: Leuven University Press. Schindler, L. 1955. Die Reform des Kupfergeldes unter Konstantinos IV. Numismatische Zeitschrit 86: 33–35. Early Byzantine Coin Circulation in the Eastern Provinces 203 Sear, David. 1987. Byzantine coins and their values. London: Seaby. Sheedy, K. 2001. Byzantine period coins. In Pella in Jordan, 1979–1990: the coins, K. Sheedy, A. R. Carson, and A. Walmsley, pp. 43–55 and pp. 129–145. Sydney: Adapa. Spahiu, H. 1979–1980. Monedna bizantine të shekujve V-XIII të zbuluara në territorin e Shqipërisë/ Monnaies byzantines des Ve-XIIIe siècles découvertes sur le territoire de l’Albanie. Iliria 9–10: 353–422. Spijkerman, A. 1975. Cafarnao III. Catalogo delle monede de la città. Jerusalem: Pubblicazioni dello Studium Biblicum Franciscanum 19. Soleirol, M. Catalogue des monnaies byzantines qui composent la collection de M. Soleirol. Metz: Lamort, 1853. Sommer, A. S. Katalog der Byzantinischen Münzen. Göttingen: Universitätsverlag Göttingen, 2003. Sotheby’s. 1990. he William Herbert Hunt collection. Highly important Byzantine coins, I, December 5–6. New York. Sotheby’s. 1991. he William Herbert Hunt collection. Highly important Byzantine coins, II, June 21. New York. Spink. 2000. he Dr. Anton C. R. Dreesmann collection of ancient coins, Part II: Byzantine and early European gold coins, July 13, London. Taylor, T. 1818. Iamblichus. Life of Pythagoras. Translated by homas Taylor. London: J. M. Watkins. Taylor, G. 1977. A hoard of small module coins of Anastasius. Coin Hoards 3: 87. homsen, R. 1986. he Graeco-Roman coins. In Hama, fouilles et recherches, 1931– 1938. III, A. Papanicolaou Christensen, R. homsen, and G. Ploug, pp. 59–69. Copenhague: Nationalmuseet. hompson, M. 1954. he Athenian agora, v. 2, coins from the Roman through the Venetian period. Princeton: he American School of Classical Studies at Athens. Todd, R. 1987. A late sixth-century hoard from northern Syria. Numismatic Chronicle 147: 176–182. Tolstoi, J. 1912–1914. Monnaies byzantines. Saint-Petersburg. Torbatov, S. 2002. Monetnata circulaciia v gradishteto kraj Odartsi. Tărnovo: Faber. Tushingham, A. D. 1972. he excavations at Dibon (Dhībân) in Moab. he third campaign 1952–53. Cambridge: he American Schools of Oriental Research. Vorderstrasse, T. 2005. Coin circulation in some Syrian villages (5th–11th Centuries). In Les villages dans l’Empire byzantin (IVe-XVe siecle), J. Lefort, C. Morrisson, and J.-P. Sodini, eds., pp. 494–510. Paris: Lethielleux. Waage, D. 1952. Antioch-on-the-Orontes, vol. IV, part 2: Greek, Roman, Byzantine and Crusaders’ coins. Princeton: Department of Art and Archaeology of Princeton. Walmsley, A. 1999. Coin frequencies in sixth and seventh century Palestine and 204 Andrei Gândilă Arabia: Social and economic implications. Journal of the Economic and Social History of the Orient 42 (3): 326–350. Waner, M, and Z. Safrai. 2001. A catalogue of coin hoards and the shelf life of coins in Palestine hoards during the Roman and Byzantine periods. Liber Annuus 51: 305–336. Whitby, M. 1988. he Emperor Maurice and his historian: heophylact Simocatta on Persian and Balkan warfare. Oxford: Clarendon Press. de Wilde, G. 1997. Monnaies au Musée de Pessinonte. Epigraphica Anatolica 28: 101–114. Zhekova, Zh. 2006. Moneti i monetno obrashtenie v srednovekovniia Shumen. Soia: Iupi-Tp. Yannopoulos, P. 1987. Inlation, dévaluation et réévaluation à la transition des mondes romain et byzantin. In Histoire économique de l’antiquité: bilans et contributions de savants belges présentés dans une réunion interuniversitaire, T. Hackens and P. Marchetti, eds., pp. 123–133. Louvain-la-Neuve: Séminaire de numismatique Marcel Hoc, Collége Erasme. he Early Byzantine Coin Circulation in the Eastern Provinces 205 Fig. A. Justinian I—Constantinople Mint (% nummia/year of reign) COLLECTIONS SCYTHIA 563-564 HOARDS BALKANS Fig. B. Justin II (% nummia/year of reign) COLLECTIONS SCYTHIA HOARDS NEAR EAST HOARDS BALKANS 577-578 576-577 575-576 574-575 573-574 572-573 571-572 570-571 569-570 568-569 567-568 566-567 565-566 15.50 15.00 14.50 14.00 13.50 13.00 12.50 12.00 11.50 11.00 10.50 10.00 9.50 9.00 8.50 8.00 7.50 7.00 6.50 6.00 5.50 5.00 4.50 4.00 3.50 3.00 2.50 2.00 1.50 1.00 0.50 0.00 564-565 561-562 562-563 559-560 560-561 557-558 558-559 556-557 554-555 555-556 552-553 553-554 551-552 549-550 550-551 548-549 546-547 547-548 544-545 545-546 542-543 543-544 540-541 541-542 539-540 538-539 15 14.5 14 13.5 13 12.5 12 11.5 11 10.5 10 9.5 9 8.5 8 7.5 7 6.5 6 5.5 5 4.5 4 3.5 3 2.5 2 1.5 1 0.5 0 206 Andrei Gândilă Fig. C. Maurice Tiberius—Antioch Mint (% nummia/year of reign) COLLECTIONS Fig. D. Heraclius (% nummia/year of reign 610–617) 45.00 42.50 40.00 37.50 35.00 32.50 30.00 27.50 25.00 22.50 20.00 17.50 15.00 12.50 10.00 7.50 5.00 2.50 COLLECTIONS TYPES (cf.MIB) SARDIS TELL BISSÉ HOARD 616-617 615-616 614-615 613-614 612-613 611-612 610-611 0.00 601-602 599-600 HOARDS NEAR EAST+ANTIOCH (site) 600-601 598-599 596-597 597-598 594-595 595-596 592-593 593-594 590-591 591-592 588-589 589-590 587-588 585-586 586-587 583-584 584-585 582-583 11 10.5 10 9.5 9 8.5 8 7.5 7 6.5 6 5.5 5 4.5 4 3.5 3 2.5 2 1.5 1 0.5 0 he Early Byzantine Coin Circulation in the Eastern Provinces Fig. 1. Nummia/year of reform (%) 22 20 18 16 14 12 10 8 6 4 2 0 sI ta as An siu n ti us J I an ni I sti Ju n ti us ric sI iu er b J Ti as e I II 16 -6 oc Ph au M li ac er ) ( us H DOC BNP BMC ANS KOD Fig. 2a. Nummia/year of reform (%) DOC BNP BMC ANS KOD 610-616 602-610 582-602 580-582 578-580 570-578 565-570 550-565 542-550 538-542 527-538 518-527 512-518 498-512 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0 207 208 7 6.8 6.6 6.4 6.2 6 5.8 5.6 5.4 5.2 5 4.8 4.6 4.4 4.2 4 3.8 3.6 3.4 3.2 3 2.8 2.6 2.4 2.2 2 1.8 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 32 31 30 29 28 27 26 25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0 498-512 512-518 518-527 BMC ANS BNP BMC 542-550 550-565 565-570 ANS KOD KOD 570-578 578-580 SCYTHIA 580-582 582-602 602-610 610-616 Andrei Gândilă 538-542 Fig. 2b. Solidi/year of reform (%) 527-538 DOC BNP Fig. 3. Nummia/year 538–616 (%) DOC 538-539 539-540 540-541 541-542 542-543 543-544 544-545 545-546 546-547 547-548 548-549 549-550 550-551 551-552 552-553 553-554 554-555 555-556 556-557 557-558 558-559 559-560 560-561 561-562 562-563 563-564 564-565 565-566 566-567 567-568 568-569 569-570 570-571 571-572 572-573 573-574 574-575 575-576 576-577 577-578 578-579 579-580 580-581 581-582 582-583 583-584 584-585 585-586 586-587 587-588 588-589 589-590 590-591 591-592 592-593 593-594 594-595 595-596 596-597 597-598 598-599 599-600 600-601 601-602 602-603 603-604 604-605 605-606 606-607 607-608 608-609 609-610 610-611 611-612 612-613 613-614 614-615 615-616 16 14 12 10 he Early Byzantine Coin Circulation in the Eastern Provinces Fig. 4. Thessalonica Mint (565–602) (% coins/year) 610-616 8 602-610 6 BELGRADE CONSTANTINOPLE 582-602 Scythia 578-582 4 565-578 Collections ATHENS SHUMEN 527-565 Fig. 5a. Selected locations in the Balkans (% nummia/year of reform) TOMIS 518-527 2 0 40 35 30 25 20 15 10 5 0 CORINTH (&Kenchreai) 498-518 565-566 566-567 567-568 568-569 569-570 570-571 571-572 572-573 573-574 574-575 575-576 576-577 577-578 578-579 579-580 580-581 581-582 582-583 583-584 584-585 585-586 586-587 587-588 588-589 589-590 590-591 591-592 592-593 593-594 594-595 595-596 596-597 597-598 598-599 599-600 600-601 601-602 209 210 Andrei Gândilă Fig. 5b. Selected locations in Anatolia (% nummia/year of reform) 50 45 40 35 30 25 20 15 10 5 AMASY A SARDIS SIDE PISIDIAN ANTIOCH 610-616 602-610 582-602 578-582 565-578 527-565 518-527 498-518 0 MELITENE Fig. 5c. Selected locations in the Near East (% nummia/year of reform) 55 50 45 40 35 30 25 20 15 10 5 ANTIOCH BER YTUS CAESAREA MARITIMA PELLA GERASA 610-616 602-610 582-602 578-582 565-578 527-565 518-527 498-518 0 NESSANA he Early Byzantine Coin Circulation in the Eastern Provinces Fig. 6. Constantinople mint (% nummia/year) 32.50 30.00 27.50 25.00 22.50 20.00 17.50 15.00 12.50 10.00 7.50 5.00 2.50 DOC BNP BMC ANS KOD Fig. 6a. Mints 498–616 (%) 50 45 40 35 30 25 20 15 10 5 0 CON NIC DOC CYZ BNP BMC ANT ANS KOD TES 610-616 602-610 592-602 582-592 578-582 570-578 565-570 550-565 542-550 538-542 0.00 211 212 Andrei Gândilă Fig. 6b. Mints 498–616 (%) 70 60 50 40 30 20 10 0 CON NIC CYZ ANT TES TOMIS CORINTH (& Kenchreai) SARDIS PISIDIAN ANTIOCH CAESAREA MARITIMA GERASA Fig. 7b. Mints 498–512 (%) Fig. 7a. Mints 498–512 (%) 100 95 95 90 90 85 85 80 80 75 75 70 70 65 65 60 60 55 55 50 50 45 45 40 40 35 35 30 30 25 20 25 15 20 10 15 5 10 0 5 CON NIC 0 CON DOC NIC BNP BMC ANS KOD TOMIS CORINTH (& Kenchreai) CONSTANTINOPLE AMASY A SARDIS PISIDIAN ANTIOCH ANTIOCH CASEAREA MARITIMA GERASA he Early Byzantine Coin Circulation in the Eastern Provinces Fig. 8a. Mints 512–518 (%) 213 Fig. 8b. Mints 512–518 (%) 100 95 95 90 90 85 85 80 80 75 75 70 70 65 65 60 60 55 55 50 50 45 45 40 40 35 35 30 30 25 25 20 20 15 15 10 10 5 5 0 CON 0 CON NIC DOC BNP ANT BMC ANS KOD NIC ANT TOMIS CORINTH (& Kenchreai) CONSTANTINOPLE AMASY A SARDIS PISIDIAN ANTIOCH ANTIOCH CAESAREA MARITIMA GERASA Fig. 9a. Mints 518–527 (%) Fig. 9b. Mints 518–527 (%) 95 90 85 65 80 60 75 70 55 65 50 60 45 55 40 50 45 35 40 30 35 25 30 20 25 15 20 15 10 10 5 5 0 0 CON NIC DOC CYZ BNP BMC ANT ANS KOD TES CON NIC CYZ ANT TES TOMIS CORINTH (& Kenchreai) CONSTANTINOPLE AMASY A SARDIS PISIDIAN ANTIOCH ANTIOCH CAESAREA MARITIMA GERASA 214 Andrei Gândilă Fig. 10a. Mints 527–538 (%) Fig. 10b. Mints 527–538 (%) 60 100 95 55 90 85 50 80 75 45 70 40 65 60 35 55 50 30 45 40 25 35 30 20 25 20 15 15 10 10 5 5 0 CON 0 CON NIC DOC BNP BMC ANT ANS KOD NIC ANT TOMIS CORINTH (& Kenchreai) CONSTANTINOPLE AMASY A SARDIS PISIDIAN ANTIOCH ANTIOCH CAESAREA MARITIMA GERASA Fig. 11a. Mints 538–542 (%) Fig. 11b. Mints 538–542 (%) 100 70.00 95 90 65.00 85 60.00 80 55.00 75 70 50.00 65 45.00 60 40.00 55 35.00 50 45 30.00 40 25.00 35 20.00 30 25 15.00 20 10.00 15 5.00 10 5 0.00 0 CON NIC DOC BNP CYZ BMC ANS ANT KOD CON NIC CYZ ANT TOMIS CORINTH (& Kenchreai) CONSTANTINOPLE AMASY A SARDIS PISIDIAN ANTIOCH ANTIOCH CAESAREA MARITIMA GERASA he Early Byzantine Coin Circulation in the Eastern Provinces 215 Fig. 12b. Mints 542–550 (%) Fig. 12a. Mints 542–550 (%) 80.00 75.00 45.00 70.00 65.00 40.00 60.00 35.00 55.00 50.00 30.00 45.00 40.00 25.00 35.00 20.00 30.00 25.00 15.00 20.00 15.00 10.00 10.00 5.00 5.00 0.00 0.00 CON CON NIC DOC BNP CYZ BMC ANS NIC CYZ ANT ANT KOD TOMIS CORINTH (& Kenchreai) CONSTANTINOPLE AMASYA SARDIS PISIDIAN ANTIOCH ANTIOCH CAESAREA MARITIMA GERASA Fig. 13b. Mints 550–565 (%) Fig. 13a. Mints 550–565 (%) 95 90 60.00 85 80 55.00 75 50.00 70 45.00 65 60 40.00 55 35.00 50 45 30.00 40 25.00 35 30 20.00 25 15.00 20 10.00 15 10 5.00 5 0 0.00 CON NIC DOC CYZ BNP BMC ANT ANS KOD CON TES NIC CYZ ANT TES TOMIS CORINTH (& Kenchreai) CONSTANTINOPLE AMASY A SARDIS PISIDIAN ANTIOCH ANTIOCH CAESAREA MARITIMA GERASA 216 Andrei Gândilă Fig. 14a. Mints 565–570 (%) Fig. 14b. Mints 565–570 (%) 70 50 65 60 45 55 40 50 35 45 40 30 35 25 30 20 25 20 15 15 10 10 5 5 0 0 CON NIC DOC CYZ BNP BMC ANT ANS CON TES KOD NIC CYZ ANT TES TOMIS CORINTH (& Kenchreai) CONSTANTINOPLE AMASY A SARDIS PISIDIAN ANTIOCH ANTIOCH CAESAREA MARITIMA GERASA Fig. 15b. Mints 570–578 (%) Fig. 15a. Mints 570–578 (%) 75 35 70 65 30 60 55 25 50 45 20 40 35 15 30 25 10 20 15 10 5 5 0 0 CON NIC DOC CYZ BNP BMC ANT ANS KOD TES CON NIC CYZ ANT TES TOMIS CORINTH (& Kenchreai) CONSTANTINOPLE AMASY A SARDIS PISIDIAN ANTIOCH ANTIOCH CAESAREA MARITIMA GERASA he Early Byzantine Coin Circulation in the Eastern Provinces 217 Fig. 16b. Mints 578–582 (%) Fig. 16a. Mints 578–582 (%) 95 90 65.00 85 60.00 80 55.00 75 50.00 70 65 45.00 60 40.00 55 35.00 50 45 30.00 40 35 25.00 30 20.00 25 15.00 20 10.00 15 10 5.00 5 0.00 0 CON NIC DOC CYZ BNP BMC ANT ANS CON TES KOD NIC CYZ ANT TES TOMIS CORINTH (& Kenchreai) CONSTANTINOPLE AMASYA SARDIS PISIDIAN ANTIOCH ANTIOCH CAESAREA MARITIMA GERASA Fig. 17b. Mints 582–602 (%) Fig. 17a. Mints 582–602 (%) 75 55 70 65 50 60 45 55 40 50 35 45 40 30 35 25 30 20 25 15 20 15 10 10 5 5 0 0 CON NIC DOC CYZ BNP BMC ANT ANS KOD TES CON NIC CYZ ANT TES TOMIS CORINTH (& Kenchreai) CONSTANTINOPLE AMASY A SARDIS PISIDIAN ANTIOCH ANTIOCH CAESAREA MARITIMA GERASA 218 Andrei Gândilă Fig. 18a. Mints 602–610 (%) Fig. 18b. Mints 602–610 (%) 75 45 70 65 40 60 35 55 50 30 45 40 25 35 20 30 25 15 20 10 15 10 5 5 0 0 CON NIC DOC CYZ BNP BMC ANT ANS CON TES KOD NIC CYZ ANT TES TOMIS CORINTH (& Kenchreai) CONSTANTINOPLE AMASY A SARDIS PISIDIAN ANTIOCH ANTIOCH CAESAREA MARITIMA GERASA Fig. 19b. Mints 610–617 (%) Fig. 19a. Mints 610–616 (%) 95 70 90 65 85 80 60 75 55 70 50 65 60 45 55 40 50 35 45 40 30 35 25 30 25 20 20 15 15 10 10 5 5 0 0 CON NIC DOC BNP CON CYZ BMC ANS NIC TES KOD CYZ TES TOMIS CORINTH (& Kenchreai) CONSTANTINOPLE AMASY A SARDIS PISIDIAN ANTIOCH ANTIOCH CAESAREA MARITIMA GERASA he Early Byzantine Coin Circulation in the Eastern Provinces Fig. 20a. Denominations 498–616 (%) 219 Fig. 20b. Denominations 498–616 (%) 85 70 80 65 75 60 70 55 65 60 50 55 45 50 40 45 35 40 30 35 25 30 25 20 20 15 15 10 10 5 5 0 0 M XXX K I E M XXX TOMIS DOC BNP BMC ANS KOD Fig. 21a. Denominations 498–512 (%) K I E CORINTH (& Kenchreai) CONSTANTINOPLE SARDIS AMASY A PISIDIAN ANTIOCH ANTIOCH CAESAREA MARITIMA GERASA Fig. 21b. Denominations 498–512 (%) 100 60 95 90 55 85 50 80 75 45 70 65 40 60 55 35 50 30 45 40 25 35 30 20 25 20 15 15 10 10 5 5 0 M 0 M K DOC BNP BMC I ANS KOD K I TOMIS CORINTH (& Kenchreai) CONSTANTINOPLE AMASY A SARDIS PISIDIAN ANTIOCH ANTIOCH CAESAREA MARITIMA GERASA 220 Andrei Gândilă Fig. 22a. Denominations 512–518 (%) Fig. 22b. Denominations 512–518 (%) 100 95 70 90 65 85 60 80 55 75 50 70 65 45 60 40 55 35 50 30 45 25 40 35 20 30 15 25 10 20 5 15 10 0 M K I 5 E 0 M DOC BNP BMC ANS KOD K I E TOMIS CORINTH (& Kenchreai) CONSTANTINOPLE AMASY A SARDIS PISIDIAN ANTIOCH ANTIOCH CAESAREA MARITIMA GERASA Fig. 23b. Denominations 518–527 (%) Fig. 23a. Denominations 518–527 (%) 95 90 85 55 80 50 75 45 70 65 40 60 35 55 50 30 45 40 25 35 20 30 25 15 20 10 15 10 5 5 0 M 0 K DOC BNP I BMC ANS E KOD M K I E A TOMIS CORINTH (& Kenchreai) CONSTANTINOPLE AMASY A SARDIS PISIDIAN ANTIOCH ANTIOCH CAESAREA MARITIMA GERASA he Early Byzantine Coin Circulation in the Eastern Provinces 221 Fig. 24b. Denominations 527–538 (%) Fig. 24a. Denominations 527–538 (%) 95 90 85 80 80 75 75 70 70 65 65 60 60 55 55 50 50 45 45 40 40 35 35 30 30 25 25 20 20 15 15 10 10 5 5 0 0 M K DOC I BNP BMC ANS M E K TOMIS KOD I E CORINTH (& Kenchreai) CONSTANTINOPLE AMASY A SARDIS PISIDIAN ANTIOCH ANTIOCH CAESAREA MARITIMA GERASA Fig. 25b. Denominations 538–542 (%) Fig. 25a. Denominations 538–542 (%) 100 85 95 80 90 75 85 70 80 65 75 70 60 65 55 60 50 55 45 50 40 45 35 40 30 35 25 30 20 25 15 20 15 10 10 5 5 0 M 0 K I E M TOMIS DOC BNP BMC ANS KOD K I E CORINTH (& Kenchreai) CONSTANTINOPLE AMASY A SARDIS PISIDIAN ANTIOCH ANTIOCH CAESAREA MARITIMA GERASA 222 Andrei Gândilă Fig. 26a. Denominations 542–550 (%) Fig. 26b. Denominations 542–550 (%) 80 75 75 70 70 65 65 60 60 55 55 50 50 45 45 40 40 35 35 30 30 25 25 20 20 15 15 10 10 5 5 0 0 M K I E M K TOMIS DOC BNP BMC ANS KOD I E CORINTH (& Kenchreai) CONSTANTINOPLE AMASY A SARDIS PISIDIAN ANTIOCH ANTIOCH CAESAREA MARITIMA GERASA Fig. 27b. Denominations 550–565 (%) Fig. 27a. Denominations 550–565 (%) 100 60 95 55 90 85 50 80 75 45 70 40 65 60 35 55 30 50 45 25 40 35 20 30 15 25 20 10 15 5 10 5 0 0 M K DOC BNP I BMC ANS KOD E M K I E TOMIS CORINTH (& Kenchreai) CONSTANTINOPLE AMASY A SARDIS PISIDIAN ANTIOCH ANTIOCH CAESAREA MARITIMA GERASA he Early Byzantine Coin Circulation in the Eastern Provinces 223 Fig. 28b. Denominations 565–578 (%) Fig. 28a. Denominations 565–578 (%) 90 60 85 80 55 75 50 70 45 65 60 40 55 50 35 45 30 40 35 25 30 20 25 15 20 15 10 10 5 5 0 0 M M K DOC BNP I BMC ANS K I E A E KOD TOMIS CORINTH (& Kenchreai) CONSTANTINOPLE AMASY A SARDIS PISIDIAN ANTIOCH ANTIOCH CAESAREA MARITIMA GERASA Fig. 29b. Denominations 578–582 (%) Fig. 29a. Denominations 578–582 (%) 70 65 65 60 60 55 55 50 50 45 45 40 40 35 35 30 30 25 25 20 20 15 15 10 10 5 5 0 M 0 XXX K I E M TOMIS DOC BNP BMC ANS KOD XXX K I E CORINTH (& Kenchreai) CONSTANTINOPLE AMASY A SARDIS PISIDIAN ANTIOCH ANTIOCH CAESAREA MARITIMA GERASA 224 Andrei Gândilă Fig. 30a. Denominations 582–602 (%) Fig. 30b. Denominations 582–602 (%) 85 75 80 70 75 65 70 60 65 55 60 50 55 45 50 40 45 35 40 30 35 25 30 25 20 20 15 15 10 10 5 5 0 0 M K I M E TOMIS DOC BNP BMC ANS KOD K I E CORINTH (& Kenchreai) CONSTANTINOPLE AMASY A SARDIS PISIDIAN ANTIOCH ANTIOCH CAESAREA MARITIMA GERASA Fig. 31b. Denominations 602–610 (%) Fig. 31a. Denominations 602–610 (%) 85 70 80 65 75 60 70 55 65 60 50 55 45 50 40 45 35 40 30 35 25 30 20 25 20 15 15 10 10 5 5 0 0 M XXX DOC K BNP BMC I ANS KOD E M XXX K I E TOMIS CORINTH (& Kenchreai) CONSTANTINOPLE AMASY A SARDIS PISIDIAN ANTIOCH ANTIOCH CAESAREA MARITIMA GERASA he Early Byzantine Coin Circulation in the Eastern Provinces 225 Fig. 32b. Denominations 610–616 (%) Fig. 32a. Denominations 610–616 (%) 100 100 95 95 90 90 85 85 80 80 75 75 70 70 65 65 60 60 55 55 50 50 45 45 40 40 35 35 30 30 25 25 20 20 15 15 10 5 10 0 5 M 0 M K DOC BNP BMC I ANS KOD K I TOMIS CORINTH (& Kenchreai) CONSTANTINOPLE AMASY A SARDIS PISIDIAN ANTIOCH ANTIOCH CAESAREA MARITIMA GERASA