A find of Celtic coins from the area of Balina Glavica (Sinotion) in Dalmatia In June 1998, the seventy year old X. X. from the village of Umljanović near Drniš brought eight Celtic coins to the Municipal Museum in Drniš. The coins were said to have been found by his grandfather on the plateau on top of the hillfort at Balina Glavica near the village of Umljanović to the south of Drniš, and had been kept by the family since then. Along the southern foot of the elevation of Balina Glavica and towards the village of Umljanović, abundant remains of architecture are visible, the remnants of the Roman settlement of Magnum, which received the status of municipium under Marcus Aurelius. At the top of the rise, which in the last few years has become greatly overgrown with trees and bushes, terraces can be seen along the slopes, which completely surround this isolated peak on the eastern edge of Petrovo Plain. During the recent war in Croatia, large quantities of pottery were discovered here when military trenches were dug. This elevation should be considered the site of Sinotion, one of the Delmatian towns mentioned by Strabo. slika lokalitete: Balina Glavica.jpg View of the top of Balina Glavica (Sinotion). The remains of the Roman settlement of Magnum are located on the level ground to the left of the height (photo P. Kos) Finds of Celtic coins in the region that was later the Roman province of Dalmatia are exceptionally rare, thus the find under discussion is surprising and doubts might even perhaps arise about the correctness of the data concerning the site of discovery, as cited by the discoverer. The fact, however, that the discoverer was a local inhabitant and that the coins were long kept in the possession of the family in the village nearby the site indicates the credibility of the cited data despite any possible hesitations. Catalogue Eastern Celtic coinage Early imitations of silver coins of Philip II Developmental types in southwestern Hungary Kapos type 1 AE OTA 491 9.2 g 12 22 2 AE OTA 491 9, 15 g 12 21 3 AE OTA 491 9, 0 g 12 21 4 AE OTA 491 8.3 g 12 21 Noricum Tauriscan group 5/ Đurđevac type 5 Tetr TKN 70-137 9.4 g 9 22 6 Tetr TKN 70-137 9.3 g 9 21 7 Tetr TKN 70-137 8.1 g 9 21 8 Tetr TKN ?-138 9.1 g 9 22 Coins nos. 5, 6 and 7 were struck with the same die on the reverse, and coins nos. 6 and 7 also with the same die on the obverse. slike novaca: uvijek prvo av, slijedi rv. Composition of the find The find is composed of four Celtic tetradrachms of the Đurđevac type and four coins of the Kapos type (also called the Regöly type). The remains of patina on the silver coins of the Đurđevac type confirmed that the coins had lain together in the soil and thus represented a unit, the question still remaining as to whether there had been more coins in the original composition of the hoard. The coins of the Đurđevac type were particularly worn on both the obverse and reverse, on the basis of which it can be concluded that they had been in circulation for a long time prior to their deposition. Two of the coins were stamped with the same die on the obverse, and three coins with the same die on the reverse. The coins of the Kapos (Regöly) type from the find were minted from pure copper, and were thus more corroded, but were less worn as a result of circulation. Die connections cannot be established for this type of coin. The obverse of the Đurđevac type coins depicts the head of Apollo to the left with a three-row pearl diadem, while a characteristic feature of this group is the mouth, depicted in the form of a necklace with thickened terminals (torques). The reverse depicts a horse to the left with a typical figure-eight shaped head and crescent-shaped hooves. Coins of the Kapos (Regöly) type depict a bearded head of Zeus to the right on the obverse, and a rider to the left with some kind of fluttering plume on the reverse. An exceptional feature of this find is the mixture of two kinds of Celtic coinage, as this is the first documented hoard to contain both coins of the Đurđevac type, attributed to one of the tribes from the tribal alliance of the Taurisci, and coins of the Kapos (Regöly) type, minted in the region settled by the Celtic tribe of the Hercuniates, settled in Pannonia. This would have been the region south of Lake Balaton, between the Sia River and the Mecsek Range in present day Hungary. In both cases, these were coins from the region of the future Roman province of Pannonia, thus from the area of western Illyricum, where both types of coinage were simultaneously in circulation at certain moments. Coins of the Đurđevac type are classified to the group of Celtic coins that K. Pink assigned to what is known as the Croatian group, while R. Göbl first attributed them to the eastern Norican group, and later to the tribal alliance of the Taurisci. The coin type received its name from the largest hoard of coins of this type, noted in the literature as Đurđevac, although the hoard was actually discovered in the vicinity of Šemovec near Đurđevac in the Drava River basin. Some 400 tetradrachms of this type were discovered in Šemovec in 1887, but of the original hoard only somewhat more than a quarter of the coins was preserved in the Numismatic Department of the Archaeological Museum in Zagreb. The mint for coins of this type must be sought somewhere in the vicinity of Đurđevac, and the otherwise rare tetradrachms were primarily in circulation in the general settlement area of the Celtic tribe of the Taurisci. It is characteristic that this is the only coin type among those attributed to the Taurisci whose circulation extended to the far easternmost edge of the region of circulation of the Celtic coinage of the tribal alliance of the Taurisci, and is not rarely also documented beyond their area of settlement. The other coinage group in the find consists of four coins of the Kapos (Regöly) type, which received its name from the valley of the Kapos River north of the town of Pécs in southwestern Hungary, where most coins of this type have been found. Tetradrachms and drachms are known from this group, and the discovery of coin dies in the ruins of a foundry and mint in the settlement of Szalacska nearby Kaposvár indicate that at least silver drachms of this type were minted in Szalacska in the 1st century BC. E. Gohl assigned the foundry or mint in absolute chronological terms to the middle of the 1st century BC on the basis of archaeological (tools, jewellery) and numismatic material. Later coins of the Kapos (Regöly) type in the size of tetradrachms and drachms, which were otherwise not discovered in the area of the mint in Szalacska, were minted exclusively from copper. The silver and copper coinage of the Kapos (Regöly) type were limited in monetary circulation primarily to the central Danubian region, north and east of the town of Pécs, and belonged to what has been termed the eastern Celtic minting group. D. F. Allen was inclined to the opinion that the minting of such coins could be attributed to the tribe of the Andizetes, who he incorrectly assumed had been settled in the region under consideration, but he later attributed the minting of this coinage to the Hercuniates. The chronological classification of the coins in the find The find in question does not offer any new elements for solving the open chronological problems related to the beginnings of minting among the Celtic tribes in the region of southwestern Pannonia. K. Pink placed the beginning of minting the Đurđevac coin type in the period around 90 BC. R. Göbl placed their minting somewhat later, in the period between 55 BC up to the conclusion of the Taurisci minting their own coinage in 21 BC. The minting of coinage of this type certainly had a long extent, as in the latest phases they were minted from silver of exceptionally poor quality. Göbl first conjectured that the Illyrian Wars of Octavian in 35-33 BC also encompassed the area where the Taurisci minted their coinage, although he was inclined to the opinion that this did not mark the end of minting their own coinage, which would have continued to some point before 16/15 BC, when the Romans annexed Noricum. It should be mentioned that today it is possible to establish that the minting of typologically related silver coinage both among the Noricans as well as the Taurisci started considerably earlier than Göbl had thought. D. F. Allen claimed that the Norican Celts had begun to mint their own coinage considerably before the end of the 2nd century BC, although his hypothesis remained unnoticed. G. Dembski on the basis of fairly unfounded arguments briefly supported the hypothesis that the Noricans would have begun to mint their own coinage ca. 113 BC during preparations for the war with the Cimbri and Teutones, and somewhat later the Taurisci would also have started minting their own coinage. Direct material proof for the claim to a fairly early beginning of minting among the tribe of the Norici has finally been gained only with the discovery of the Enemonzo find near the Zuglio region in Friuli, where early tetradrachms of the Kugelreiter type were found together with Roman victoriati, which were ceased to be minted in 170 BC. On the basis of later detailed analyses, G. Gorini justifiably suggested placing the minting of the earliest emissions of Norican coinage between 140 and 130 BC. The beginning of the second phase of Norican minting, characterized by Latin coin legends, was placed by Gorini (and similarly Göbl) without any convincing arguments in the period after 60/50 BC, and by Dembski in 64/63 BC, just before the besieging of the Norici by the Taurisci and the Boii. On the basis of the firm arguments offered by the Enemonzo hoard, in my opinion it is necessary to examine the entire chronological classification of individual coin emissions of the tribal alliance of the Norici and Taurisci, as well as certain neighboring peoples, not to mention the beginnings of minting the Đurđevac in Kapos (Regöly) type coins. The minting of silver coins of the Kapos (Regöly) type belongs to the later phase of eastern Celtic minting, and in an absolute chronology, R. Göbl placed the beginning of minting of this coin group in the period after 53 BC, and K. Biró-Sey placed it as late as just before the Roman conquest of Pannonia. D. F. Allen supported a fairly early beginning of minting silver coins of this group, while M. Torbágy placed it in the last third of the 2nd century BC. At first coins of this type were minted from high quality silver, while later the quality of the metal declined, indicating that they had a lengthy existence. A. Kerényi also decided for a lengthy minting of coins of this type because of the minting in two metals. The early silver tetradrachmas and drachmas of this type were minted at the same time, with the early tetradrachmas weighing between 12 and 14 g. Their later phase was characterized by an exceptionally poor quality of the metal from which they were minted, in most cases almost entirely copper, and the weight of these bronze coins was on average between 8 and 9 g. The large bronze coins have the same diameter as the tetradrachmas (22-23 mm), and bronze coins were also minted in the size of a drachma. M. Torbágy assigned the minting of bronze coins of the Kapos (Regöly) type to the second half of the 1st century BC. The minting of stylistically coarse coins from poor quality metal was placed by D. F. Allen at the end of the 1st cent. BC, as A. Kerényi had also suggested previously. R. Göbl, to the contrary, set forth a thesis that the coins of the Kapos (Regöly) type had been minted yet further to ca. 21 BC. An entirely new absolute chronology of minting bronze coins of the Kapos (Regöly) type was suggested by K. Biró-Sey, who explained the transition from silver to bronze minting of the Kapos (Regöly) type coins with the fact that coins with the poorer quality metal only began to be minted after the Roman conquest of Pannonia and that they were used exclusively for local purposes. The fact that bronze coins were not documented in the framework of the mint or foundry activities at Szalacska would perhaps speak in favor of their very late minting. Nonetheless, the thesis about the minting of bronze coins of this type as late as the Roman conquest of Pannonia (which can be placed in 12-9 BC) does not seem likely. The analogous minting of their own coinage among the tribe of the Eravisci in the vicinity of Budapest (again primarily for local use), immediately prior to the Roman occupation shows that the coins then minted in this area were created under the decisive influence of Roman coinage and that they were also minted according to the standard of the Roman denarius. The minting of their own coinage by Celtic tribes according to a Greek monetary standard (modelled on the tetradrachm) after the Roman conquest of Pannonia would have undoubtedly required some kind of Roman agreement, and is thus highly unlikely. Also, the minting of coins of the Syrmian type in bronze (phase C) among the more easterly Scordisci, which is similar to the example of the Kapos (Regöly) type coins at least in terms of the continuation of minting coins of the same type as in silver (phases A and B), is assigned chronologically to the mid 1st century BC. Both kinds of Celtic coinage represented in the find could theoretically have circulated monetarily for decades after the Roman occupation of Pannonia. At least for the region of Noricum, such a possibility is indirectly indicated by several new hoard finds, in which Norican large and small silver coins are accompanied by Roman early Imperial denarii. The area of circulation of both kinds of coins The exceptionally scarce documented finds of Celtic finds in the area of the later province of Dalmatia indicate that they did not circulate here, or rather that in this area Celtic coinage was not a valid means of payment. Individual recorded Celtic silver coins from the region of the later province of Dalmatia nonetheless indicate that in this or some other way Celtic coins exceptionally arrived in this area. Both types of coins represented in the find were characteristic for the monetary circulation in the region of the later province of Pannonia. The distribution in terms of circulation is known at least approximately for coins of the Đurđevac type. In addition to the most numerous find of coins of this type at Šemovec near Đurđevac, rare individual finds of such coins are documented primarily in the Sava River valley all the way to Gomolava in Vojvodina in the east, as well as individually in the area of southwestern Hungary and southeastern Slovenia. In general terms we also know the distribution range of coins of the Kapos (Regöly) type. Findspots of coins of this type are concentrated in the Kapos River valley, especially in the region of Somogy and Tolna in southern Hungary, and individual finds of these coins, particularly bronze, are also documented at great distances from the production centers. Individual examples of both of the coin groups represented in the find under discussion appear together at several sites in southwestern Hungary. The general characteristic of Celtic coinage in the Pannonian region is a relatively narrowly limited area of circulation, with a rare penetration into monetary circulation in those regions where coins of other groups were predominant. In such cases, these coins were quite often used as blanks for minting local coin types, and thus they were not received as a valid means of payment. As was shown above, it is characteristic for coins of the Đurđevac and Kapos (Regöly) types that they have been discovered far away from their production centers and their limited region of intensive circulation, and often also at the same site (e.g. Szalacska, Velem, Pest), which would indicate a contemporaneous and parallel, thus equivalently tolerant mutual circulation of both types of coins, considering the reciprocal trade between both regions of coin circulation. A corpus of the classical coin finds from the region of the later Roman province of Dalmatia has never been produced, and thus the characteristics of the monetary circulation in this area are not known, and it can merely be stated that from the beginning of the 1st century BC a major role was played by Roman Republican coins. Dating of the find For coins of the Đurđevac type in the discussed find, one side was characterized by evident wear on the die used to mint the coins, and on both signs of wear could be noted on the surface of the coins, as a result of circulation. The average weight of the four coins of the Đurđevac type in the find is approximate to the average weights of all documented coins of this type that were minted with later dies. An important factor in Celtic numismatics in determining the relative chronological order of minting individual coin emissions is a decline in the average weight of the coins, corresponding to a lengthier chronological distance from the start of minting individual types. In the course of time, the quality of the metal from which the later coins were minted usually also declines. If the average weight of the early coins of the Đurđevac type measured almost 10 g, the weight of the four coins in the find with 8.97 g, is convincing data in terms of the relative chronology. Also the metal that was used to mint the coins in the find has a smaller proportion of silver. The latest phase of minting for the coins of the Kapos (Regöly) type in the find, which otherwise do not exhibit such great wear from circulation, is indicated by the fact that they were minted from poor quality, almost entirely copper metal, while the average weight of the four coins of this type in the find was 8, 9 g, which corresponds to the average weight of the bronze coins of this type. Without considering the open chronological questions discussed above, related to the beginning of minting the coins of both types represented in the find, it is to some extent possible to establish a relatively late minting for the coins of both groups in the find, which could well place it in the late second half of the 1st century BC. Of both of the groups, the bronze coins of the Kapos (Regöly) type were minted later than the coins of the Đurđevac type, which thus exhibit greater traces of wear from circulation. All the coins together thus represent a sample of coinage that was in circulation in the central Pannonian region in the last decade of the 1st century BC, or perhaps even in the first decade of the1st century AD. Interpretation In the case of the find of Celtic coins from the area of Balina Glavica, this was coinage that was characteristic for monetary circulation in Pannonia in the second half or at the end of the 1st century BC, and it could certainly not be in circulation in the region of central Dalmatia. The coins were most probably brought by someone for whom their use was common and everyday. This would most likely have been a Celtic local from the region of central Pannonia, and not a Roman soldier or civilian, as would be indicated by the fact that the Romans in that period used late Republican or early Augustan coinage, which in the Celtic world of the eastern Alpine region also enjoyed the role of a supra-regional means of payment. It is unlikely, in fact, that coins such as are represented in the hoard under discussion, at least part of which at first glance were of poor quality, would represent interesting loot to anyone whatsoever. In the region of Dalmatia, this coinage could not have had any purchasing power, and therefore it had no value. The discoverer noted the broader region of Balina Glavica as the site of discovery of the find. This is a strategically located crest above Petrovo Plain opposite the village of Umljanović near Drniš, where we can place the Delmatean town of Sinotion, mentioned by Strabo. Sinotion lay on the important road Via Gabiniana, which was built in 16/17 AD and connected Salona through Andetrium with Promona. From this point, as part of the Dolabella road network, it continued onwards to Siscia, although Dolabella built it only as far as Strmica, and the Romans continued the construction towards Siscia only in 47/48 AD. Octavian's military conquests of the region of the later provinces of Pannonia and Dalmatia between 35 and 33 BC at least partly took place through Aquileia and present-day Slovenia in the direction of the Lika region, and further along the Kolpa/Kupa River valley towards Segesta (later the Roman Siscia, present-day Sisak). The Taurisci were one of the defeated tribes in this war, and Octavian conquered Segesta most probably with the aid of Norican allied tribes. Later, Octavian's army penetrated further inland against the Delmatae, although it is more likely that the army arrived in boats across the Adriatic and from there penetrated into the hinterland. An overland invasion from Segesta through the still hostile territory of the Iapodes is unlikely. In this invasion, Octavian conquered important Delmatean strongholds, among others Promona and Sinotion. Given the relatively late dating for the minting of the coins represented in the find, and the hardly likely direct overland route between Siscia and central Dalmatia along local roads through an area occupied by insubordinate local populations, the presence of Celtic coinage in this area cannot be connected with the mentioned events. The Pannonian War began in 14 BC, although the Pannonian tribes (probably the Breuci from the region between the Drava and Sava Rivers) surrendered to Agrippa in 13 BC. In the following year Tiberius took command, and the war against the Pannonians and the Delmatae extended from 11 to 9 BC, while details cannot be reconstructed exactly on the basis of the scarce sources. From that point onwards, the area of the later provinces of Pannonia and Dalmatia was under Roman authority. An uprising of Pannonian and Dalmatian tribes occurred in 6 AD, again stifled by Tiberius, who in 9 AD penetrated into central Dalmatia from Siscia along the Una or Vrbas valleys. The Celts from the interior, i.e. from Pannonia, penetrated into Dalmatia relatively late, at least there is no material proof of their presence in Dalmatia before the beginning of the 1st century AD. Only in the first decades AD, in central Dalmatia, primarily in the hinterland of Šibenik and Split, where both legionary camps were located at Burnum (Šuplja Crkva) and Tilurium (Gardun), was the presence noted of Celtic Pannonian and Norican soldiers, as best attested by onomastic studies of the epigraphic material. These soldiers served in the Roman army and came with it to Dalmatia, most probably after 9 AD, when they began to be recruited in Pannonia after the Pannonian-Dalmatian rebellion. Thus, for example, in the area of Magnum, located below Balina Glavica, a funerary monument was discovered from the mid 1st century AD, which mentions a Roman soldier with the Celtic name Vercaius, which otherwise appears primarily in the region of Noricum. The Celtic coins from Balina Glavica, which we have shown could have originally come only from the specific region of Pannonia, could have arrived in Dalmatia only along the land routes that connected the Pannonian region with that of Dalmatia, thus primarily along the valleys of the Una or Vrbas Rivers through present-day Bosnia, which was the easiest passage from Siscia towards Dalmatia. The deposition or loss of the group of Celtic coins in the area of Balina Glavica is most probably best dated to the late Augustan period, in the period of Tiberius' expedition against the rebellious Pannonians and Dalmatians, thus in the first decades of the 1st century AD.