STUDIES

COMMENTARIA ARCHAEOLOGICA ET HISTORICA (III). CHRONOLOGY OF THE DACIAN SILVER HOARDS

Aurel Rustoiu1

Abstract: These comments stem from the recent publication of a number of studies regarding the silver artefacts of pre-Roman , with important implications for the relative and absolute chronology of the late La Tène period in the region in question, and also for some cultural and historical transforma- tions that happened in the same area. Some of the conclusions presented in these studies are insufficiently supported by arguments. Their publication requires a detailed analysis, which is meant to clarify a number of controversial aspects. For example, D. Spânu has recently divided the evolution of the Dacian silver hoards into two phases dated to the La Tène D2a and D2b (the period between ca. 75–65 BC and the Augustan age). To support this chronology, he chose to ignore a series of hoards, or only certain artefacts from other hoards, which did not fit into the suggested model. These efforts to push the chronology of the Dacian silver hoards within a particular time frame stem from his aim to use these discoveries as arguments for a series of a priori historical interpretations for which archaeological evidence is scant. This includes the idea that the local silver ornaments were all made exclusively of melted Mediterranean coins which supposedly reached pre- massively only after the defeating of Mithridates VI Eupator and the fall of the Kingdom of Pontus. However, the analysis of the silver jewellery from pre-Roman Dacia is demonstrating that the artefacts in question were made using both locally-sourced silver and melted Mediterranean coins. Chronologically, these hoards can be divided into three phases: first group dated to the La Tène D1, 150/125–75/50 BC; second group belongs to the La Tène D2, 75/50–30/25 BC; third group dated to the Augustan – Tiberian period, 30/25 BC – AD 25/30. It can be therefore concluded that the scenario proposed by D. Spânu for the chronology of the end of the Late Iron Age in the lower Danube region and , based on an erroneous dating of the silver hoards, is not credible. This kind of analysis requires the incorporation of different categories of archaeological evidence belonging to the last three centuries before the Roman conquest. Their contextual interpretation could provide a more reliable and detailed chronology of the evolution of the communities from each of the two territories in question. Lastly, this evolution was not uniformly equal across wide areas, as suggested by the typo-chronological tables and schemes drawn in the office by some researchers, since each community had its own history and evolution governed by a multitude of different social, economic, demographic or ecological factors. Keywords: silver hoards; pre-Roman Dacia; coins; Transylvania; La Tène chronology

These comments2 stem from the recent publication of a number of studies regarding the silver artefacts of pre-Roman Dacia, with important implications for the relative and absolute chronology of the late La Tène period in the region in question, and also for some cultural and historical transformations that happened in the same area. Some of the conclusions presented in these studies are insufficiently supported by arguments or are using erroneous, “fabricated” or false arguments. Their publication requires a detailed analysis, which is meant to clarify a number of controversial aspects.

1 Academia Română, filiala Cluj-Napoca. Institutul de Arheologie și Istoria Artei. Str. M. Kogălniceanu 12–14, 400084 Cluj-Napoca, . E-mail: [email protected] 2 For previous comments, see RUSTOIU 2012; RUSTOIU 2016a. Ephemeris Napocensis, XXX, 2020, p. 11–38, https://doi.org/10.33993/ephnap.2020.30.11 12 Aurel Rustoiu

1. Introduction: the theory of Daniel Spânu Daniel Spânu has spent most of the last two or three decades studying the Dacian silver hoard and has published numerous studies focusing on a variety of aspects3. Recently, he brought again into discussion the chronology of these hoards, extending his conclusions to the issue of late La Tène chronology in the lower Danube basin and the territory inside the Carpathians range4. His approach regarding the chronology of the Dacian silver hoards is based on an assemblage of numismatic and historical arguments5. Thus, he is stating that despite the exis‑ tence of local silver sources in complex ores from the western Carpathians, Dacian communities lacked the technological knowledge needed to mine them and extract the silver during the last phase of the Late Iron Age. Only the alluvial gold was extracted from Transylvanian river beds during that period. Consequently, the silver jewelleries from pre-Roman Dacia were made by first melting coins originating from the Mediterranean area. Within this scenario, the date at which these foreign monetary issues arrived north of the Danube became important. Using a synthesis of the numismatic studies that discussed this problem, D. Spânu is proposing four chronological phases. According to this interpretative model, the first phase is defined by the arrival, in the second half or at the end of the 2nd century BC, of the silver coins issued by Macedonia Prima. The second phase is related to the arrival of tetradrachms issued by Thasos and drachms of Dyrrachium and Apollonia. These cities became allies of Rome and intensified their coin issuing after 88 BC and, respectively, 75 – 65 BC in the context of the wars against Mithridates VI, their silver coins being used to pay stipends to the barbarian allies. Some Republican denarii issued during the first two or three decades of the 1st century BC also arrived during this phase. The third phase is characterized by an increased arrival of the Republican denarii (some of the recovered assemblages also contain local copies) during the last three decades of the 1st century BC, combined with early imperial denarii from the Augustan period. The fourth phase covers the assemblages containing post-Augustan imperial denarii, the most numerous ones being dated to the reign of . Using this chronology as a starting point, D. Spânu has been forced to date the hoards containing silver jewellery, some associated with coins belonging to the second and the third phase, to the period between ca. 75 – 65 BC and the Augustan age. Regarding the chronology of the hoards containing silver jewellery, he has divided it into two phases6 (Fig. 1), departing from the model proposed first by K. Horedt7 and later by V.V. Zirra and D. Spânu8, which consisted of three chronological phases. The new first phase proposed by D. Spânu includes hoards containing middle La Tène brooches with knobs, while the second phase includes hoards containing brooches with a rhombic plate (“shield”) on the bow and spoon-shaped brooches, both types being dated to the late La Tène. In the first case, the hoards also contain tetradrachms of Thasos and drachms of the aforementioned Illyrian cities, and sometimes also Republican denarii, while the hoards of the second phase contain late Republican and early imperial denarii. Consequently, the first phase has been dated to the La Tène D2a (75 – 25 BC), and the second phase to the La Tène D2b (25 BC – AD 25).

3 SPÂNU 2012. 4 SPÂNU 2019a. 5 SPÂNU 2012, 88–91, 138–140; SPÂNU 2019a, 172–178; etc. 6 SPÂNU 2002; SPÂNU 2012; SPÂNU 2019a. 7 HOREDT 1973. 8 ZIRRA/SPÂNU 1992. Commentaria archaeologica et historica (III). Chronology of the Dacian silver hoards 13

Fig. 1. The evolution of the Dacian hoards after SPÂNU 2019a.

Furthermore, D. Spânu has used this chronology to “systematize” the archaeological evidence datable “before” and “after” these silver hoards, this rather chaotic attempt turning the discoveries specific to the La Tène C2 and D1, as well as those belonging to the last century before the Roman conquest, upside down. His chronology of the hoards containing silver jewellery and the supporting arguments are discussed in detail in the following paragraphs.

2. Provenance of silver The practice of melting silver coins to obtain ingots needed to manufacture jewellery is attested in pre-Roman Dacia by the discovery of some silver bars associated with coins9, and also by some comparative metallographic analyses involving silver coins and jewellery associ‑ ated with them10. Still, as already mentioned, the method was used in workshops which had no access to the silver sources from Transylvania due to various reasons11. On the other hand, different analyses of the lead isotopes from silver artefacts dated to the Early Iron Age from the wider Mediterranean basin, from Spain to Greece and the Levant, have shown that in some cases the artefacts were made of mixtures of silver having different origins12. This practice could have also been used in pre-Roman Dacia, objects made of local silver and Mediterranean coins being recycled to manufacture new jewellery having different metallographic characteristics. Therefore, the XRF analyses employed by the late B. Constantinescu and his collaborators to

9 See the hoard from Stăncuța (Buzău County): PREDA 1957. 10 See, for example, CHIRILĂ/MATEI 1986. 11 MEDELEȚ 1992, 228–229; RUSTOIU 1996a; RUSTOIU 1996b, 49–53. 12 WOOD ET AL. 2019; WOOD/MONTERO-RUIZ 2019. 14 Aurel Rustoiu study the metallographic composition of the silver jewellery and coins from pre-Roman Dacia are inconclusive13. In these cases, the analysis of lead isotopes would be more useful to identify the origin of the silver used in the manufacturing of the Dacian jewellery14. In ancient times, silver was extracted from galena (a lead sulphide mineral), being separated from the lead through cupellation. This technology was far less complicated15 than what D. Spânu has suggested. The ore was mined using open pits from the area of the western Carpathians16. The mining galleries from the same region only appeared during the Roman provincial period17. During this period, one pottery workshop from Ampelum reused the waste litharge resulted from cupellation to produce lead-glazed wares18. At the same time, during the survey conducted by the Austrian authorities at Grădiștea de Munte at the beginning of the 19th century, a large quantity of galena (around 1.5 t) was discovered, having a high silver content19, which suggests that the technology required to separate silver from lead was used in the period preceding the Roman conquest. Numerous locally-issued coins from Transylvania20, as well as the silver and gold jew‑ ellery21 of the “Celtic horizon” (the 3rd – 2nd centuries BC) indicate that some silver resources from the inner Carpathians region were at least partially used. The coin die for manufacturing scyphate coins, which was preserved as a “relic” among the inventory of the workshop from Pecica (Arad County)22, also indicates that silver coins were made locally before the arrival of the Mediterranean ones. A number of older chemical analyses, unfortunately involving a quite small number of samples, have indicated that locally-issued coins belonging to the “Celtic” horizon, as well as some silver jewellery of the “Dacian” horizon, were more likely made of Transylvanian silver23. Lastly, numismatic studies have shown that some monetary hoards from Dacia included variable quantities of Republican denarii which were copied in local workshops24. These copies were either mould cast or struck using coin dies25. The numismatic identification of the mould cast denarii in the local environment was also confirmed by the discovery of a faulty cast denarius of Marcus Antonius from 32/31 BC in a workshop from the Dacian settlement at Șimleu Silvaniei26. Regarding the local struck copies of the Republican denarii, aside from older discoveries of coin dies from Brașov, Ludești, Poiana-Galați, “Banat”, Tilișca and Grădiștea de Munte27, new finds also emerged, for example the die from the settlement at Ocnița28, while four die shafts have been illegally unearthed in the Orăștie Mountains by treasure hunters29. 13 This was also recognized in VASILESCU ET AL. 2015: “The accepted standard method to determine the provenance of metal using lead isotopes, which would give a more precise localization for the ore source, was unfortunately not accessible to us”. 14 See a description of this method in STOS-GALE/GALE 2009. 15 For the cupellation technology, see NRIAGU 1985. 16 Open pit mining is also documented in other European areas during the same period, for example in Limousin, France: CAUUET ET AL 2018; DOMERGUE ET AL. 2006 etc. 17 CAUUET ET AL. 2003. 18 LIPOVAN 1990. 19 PEȚAN 2018, 387–388. 20 PREDA 1973; PREDA 1998 etc. 21 RUSTOIU 2016b; RUSTOIU 2017 etc. 22 CRIȘAN 1978, 84–97; RUSTOIU 1996b, 56, 63. 23 STOICOVICI/STOICOVICI 1972; STOICOVICI/STOICOVICI 1973. 24 CHIȚESCU 1980, 125–128. 25 LOCKYEAR 2004; LOCKYEAR 2008; DAVIS 2006; WOYTEK ET AL. 2012 etc. 26 POP 2008, 99, Fig. 91. 27 See the list of coin dies from pre-Roman Dacia in MEDELEȚ 1994a, 203–204, annex no. 4. 28 IAROSLAVSCHI 2001. 29 DACII 2004, 62, no. 39–42. Commentaria archaeologica et historica (III). Chronology of the Dacian silver hoards 15

Their presence illustrates the amplitude of the Republican denarii copying in pre-Roman Dacia. In this context, two workshops provide the chronological limits of this activity. One is the workshop at Tilișca, where the coin dies have been found in a vessel belonging to the lower layer of habitation from one terrace of the settlement; the context of discovery cannot be dated later than the middle of the 1st century BC30. On the other hand, the copied denarii from Tilișca cover the period between 145/138 and 72 BC. Accordingly, the Republican denarii started to be copied as early as the second phase of the chronology proposed by D. Spânu for the arrival of the Mediterranean coinage north of the Danube, and not only during the third phase. The workshop from Grădiștea de Munte, which copied Republican denarii from 126 and 68 BC and imperial denarii issued under Tiberius (AD 14–37), was probably destroyed during the wars with the Romans at the beginning of the 2nd century AD31. Given the apparently widespread copying of the Roman coinage and the duration of this phenomenon, it is hard to believe that original denarii were brought in Dacia and melted to produce copies in the local workshops. The more likely explanation is that various local or foreign artisans working for the local rulers had access to the local sources of silver, which they used to produce these copies. The metallographic analyses of a series of locally copied Republican denarii have shown that sometimes the silver content was lower, though in general these coins had a high silver content32. Indirectly, these analyses indicate that the local workshops were familiarized with the technologies of extracting and processing the silver, including the alloying with other metals, for example copper. These technologies were transmitted from one generation to another, from the “Celtic” horizon to the “Dacian” one within various “clans” of artisans specialized in nonferrous metal processing. This was demonstrated long time ago through the analysis of the subaerate coins of the 3rd – 2nd centuries BC and of the jewelleries having a core made of copper or low quality silver alloy which was coated with a thin layer of annealed silver of high quality. The same technology was used for both coins and jewellery, and its use was more likely determined by the limited quantity of silver which was available for some workshops to produce these artefacts33. Summarising these observations, it can be said that if local silver was used alongside the one recycled from Greek and Roman coins, then the starting date of the manufacturing of jewellery included in the Dacian silver hoards is not conditioned by the arrival date of var‑ ious Mediterranean coinages. Moreover, only some of these hoards contain coins. The arrival of the silver coins issued by Macedonia Prima and Thasos, which followed an eastern route (through north-western Bulgaria, Wallachia, and Transylvania)34, and of those issued by the Illyrian cities, which followed a western route (north-western Bulgaria, the Scordiscian ter‑ ritory, Banat, Crișana, Transylvania)35, is only indicating the directions of some connections established by political entities from the eastern Mediterranean basin with some leaders of the indigenous communities from the north-western Bulgaria, the area north of the lower Danube and Transylvania since the second half of the 2nd century and the beginning of the 1st century

30 LUPU 1989, 54,100–102. See also RUSTOIU 1996a. 31 GLODARIU ET AL. 1992. Despite the objections raised by some researchers regarding the context of discovery and dating of the workshop from Grădiștea de Munte (see, for example, ZIRRA 2017, 65, n. 244), it is quite clear that at least before the middle of the 1st century AD, if not also later, copies of the Republican and early imperial denarii were produced in the area of the Dacian centre of power. 32 LOCKYEAR 2008; WOYTEK ET AL. 2012. 33 RUSTOIU 1996a. D. Spânu has suggested that the copper core was meant to increase the elasticity of some jewelleries, for example bracelets (SPÂNU/COJOCARU 2009, 102), an idea already put forth by N. Chidioșan and I. Ordentlich (CHIDIOȘAN/ORDENTLICH 1973), but this supposed characteristic made no sense in the case of coins, brooches, pendants etc, which were not experiencing this kind of mechanical stress normally. 34 CHIRILĂ/MIHĂESCU 1969; PROKOPOV 1993; PREDA 1998, 239–260, Fig. 18. 35 CONOVICI 1986; MEDELEȚ 1994b; PREDA 1998, 261–273, Fig. 19; UJES-MORGAN 2012. 16 Aurel Rustoiu

BC36. The northward distribution of these coins, as it appears on distribution maps37, seems to be related to the expanding area of the burials containing weaponry of the Padea-Panagjurski Kolonii from the lower Danube to the inner Carpathians region38. A number of ancient authors indicate that, at the end of the 2nd century and the beginning of the 1st century BC, the Roman state faced a series of plundering raids in Macedonia, conducted by members of these martial elites and their followers, which were only repelled by military interventions39. Within this war‑ like climate, some quantities of Hellenistic coins could have arrived to the north of the Danube as booty resulted from these incursions40. Therefore, the circulation of these coins in the Dacian environment was not related exclusively to the time of the Mithridatic wars, since they could have also arrived earlier.

3. Chronology of the Dacian silver hoards K. Horedt first analysed these hoards systematically more than four decades ago, identi‑ fying three distinct chronological horizons based on the typological evolution of certain catego‑ ries of locally-made jewellery and their association with Hellenistic and/or Roman silver coins. Within this model, the first horizon corresponds to the Central European La Tène D1 (ca. 125 – 75 BC), the second one to the La Tène D2 (ca. 75 – 25 BC), and the third horizon to the Augustan and early Tiberian period (ca. 25 BC – AD 25)41. As previously mentioned, D. Spânu has opted to divide the evolution of these hoards into two chronological phases (Fig. 1), including the last two horizons of the chronology pro‑ posed by K. Horedt in a single phase. However, in order to demonstrate the validity of this chronology, D. Spânu has chosen to ignore the hoards whose composition is not fitting neatly into his interpretative model, or certain artefacts from some hoards which are not corresponding to his argumentation. For example, the Sărăcsău hoard, which contains brooches with knobs specific to the first phase, but also a brooch with a rhombic plate on the bow which in his theory belongs to the second phase, is missing from the typo-chronological table in his recently published article42. Likewise, the rigid neck-rings from the Sărmășag and Cehețel hoards, which belong to the first phase, are intentionally omitted because in his model these ornaments belong to the second phase43. It is obvious that his efforts to “systematize” the chronology of the

36 See also EGRI 2019, 66–68. 37 SPÂNU 2019a, 173, Fig. 2. 38 RUSTOIU 2015, Fig. 27. 39 For example, Florus, I, 39, 1–6; Titus Livius, Periochae, LXIII, LXV; Velleius Paterculus, II, 8, 3; Frontinus, II, 4, 3; Apian, Illyr., 14, mention that after 114 BC the troops led by C. Porcius Cato were crushed by the mentioned armed groups. The local situation returned to normal in 111–110 BC under the command of M. Livius Drusus, who pushed the invaders across the Danube. M. Minucius Rufus was forced to intervene militarily in 109–106 BC, following some new raids in Thrace and the Macedonian border. Then in 101 or 100 BC, Titus Didius repelled again the raiding barbarians, “ambushing them while they were scattered, looking out for spoils” (Florus, I, 39, 5). These conflicts continued during the first quarter of the following century. In 74 BC, the governor of Macedonia C. Scribonius Curio again reached the lower Danube while following the raiding barbarians, but decided against crossing the river because he was “frightened by the darkness of the forests” (Florus, I, 39, 6). Later in the same century, resumed the plundering raids in Macedonia and Illyricum, according to Strabo. 40 CONOVICI 1986, 82, has also considered that the coins of the Illyrian cities and the Republican denarii arrived massively in Dacia as war booties following the plundering raids conducted by Burebista. 41 HOREDT 1973, 151. 42 SPÂNU 2019a, 181, Fig. 5. 43 Similar omissions can be observed in this article when the author is discussing the presence of La Tène and Roman brooches in the inventories of some Dacian settlements. For example, it is stated that Roman brooches are missing from Căpâlna, though one is included in the illustration of the article: SPÂNU 2019a, Fig. 9 (table of Commentaria archaeologica et historica (III). Chronology of the Dacian silver hoards 17

Dacian silver hoards within a certain period of time stems from a need to find arguments for his a priori interpretative theories: the Dacian silver jewelleries were exclusively made of melted Mediterranean coins (Hellenistic and mainly Republican denarii); these foreign coins arrived massively in pre-Roman Dacia in the context of Mithridatic wars and the fall of the Kingdom of Pontus. However, it has been shown that locally-sourced silver was also used to produce these jewelleries and some copies of the Republican denarii. At the same time, some of the aforemen‑ tioned silver coins most likely arrived in pre-Roman Dacia earlier than the period proposed by D. Spânu. Another aspect that is important for the chronology of these silver hoards containing jewellery is the social environment in which the practice of depositing them emerged. From this point of view, several key features have to be taken into consideration in order to understand the manner in which the practice of depositing silver hoards evolved through time. The great majority of these hoards contain silver artefacts belonging to only one female costume assemblage44. From the functional perspective, these assemblages consist of ornaments for the head (earrings or hair spirals), neck (neck-rings or chains with pendants), limbs (brace‑ lets and/or anklets) and body (belts and brooches). Their fairly standardized functional structure and the fact that they were meant to be used by a single person, being then buried together at the same time, indicate that the assemblages were not simple accumulations of different valuables45, but were meant to fulfil certain practical and symbolic functions which must have remained largely unchanged through time. The typological analysis indicates that some of the components, like the brooches, bracelets and anklets, went through several stylistic transformations over time, while others had seen little change, so the latter could be less relevant for dating. This is the case of the Mediterranean-type chains, the hair spirals, or the neck-rings, which have not been included in the present typo-chronological table (Table 1). The same typological analysis also indicates that the rhythm of change from one type to another was different in the case of each category of ornaments, some types being adopted faster than others. As a consequence, some costume assemblages included both slightly old-fashioned elements and others of a newer type. These gradual changes46 can be best compared in the mentioned typo-chronological table (Table 1). Some of these differences could be a result of the variable degrees of connectivity, including access to skilled craftsmanship, from one region or community to another, though others could simply stem from individualized stylistic preferences, including perhaps a certain degree of conservatism in what concerns the female bodily ornamentation. Another key feature that has chronological relevance is the frequent inclusion of cer‑ tain types of Hellenistic or Roman coins, and/or silver or bronze vessels originating from the Mediterranean area, which were dated to the late Republican period. On the basis of these observations, the hoards containing silver jewellery can be divided into three typo-chronological groups or horizons. The first group is defined by the presence of brooches with knobs of the middle La Tène type (Fig. 2/1). They are sometimes associated with late La Tène type brooches having a bilateral spring, a plain or decorated bow and a rectangular, non-perforated catchplate. The phalera-type brooches, which was meant to identify different chronological phases in some Dacian settlements) and Fig. 10/15, the “Roman” brooch from Căpâlna. 44 There are a few exceptions: the Sărăcsău and Tilișca 2 hoards contain two costume assemblages each; due to the size of annular ornaments, one costume more likely belonged to an adult and the second to a child. 45 There are also a few exceptions, an example being the Remetea Mare hoard, but their rare occurrence is not affecting this observation. 46 This gradual transition was also noted in ZIRRA/SPÂNU 1992, though without further comments regarding its meanings. 18 Aurel Rustoiu

Table 1. Chronology of the silver hoards from Late Iron Age Dacia (EGRI/RUSTOIU forthcoming).

Hoards Coins Brooches (phalerae) Brooches w/ knobs Brooches D type brooches LT w/ rhombic plate Brooches brooches Spoon-shaped types of brooches Other Belts bracelets Twisted Bangles spiral bracelets Simple w/ palmettes Spirals w/ twisted ends Bracelets Earrings Mastoi Kantharoi vessels Late Republican groups Chronological București-H. Th 1 Lupu Sâncrăieni Dy, Th Agârbiciu Dy Cehei Dy Clipicești Th Tilișca 1 Tilișca 2 Ag Drăgești Dy, A, RD (75BC) Oradea 3 Cojocna Săcălăsău 1 ? Dy, A Săcălăsău 2 Cehețel Bistrița Sărmășag Mediaș Sărăcsău Coldău 2 Ghelința RD (67BC) Săliștea Bălănești Coada Malului Șaeș Cadea RD (41/38BC) Cerbăl RD (44/43BC) Rociu RD (81/76BC) Izvoru Frumos RD (81BC) Vedea Ag 3 Remetea Mare Th, RD (16/15BC) Cetățeni RD (8BC) Șeica Mică RD (28BC) Peteni RD (70/58BC) Hetiur Senereuș Commentaria archaeologica et historica (III). Chronology of the Dacian silver hoards 19

Fig. 2. Typical sets of bodily ornaments from horizons 1, 2 and 3: 1. Cojocna; 2. Cadea (both after Fettich 1953); 3. Șeica Mică (after Floca 1956). brooches with anthropomorphic decoration from the Herăstrău hoard47 and the phalerae from the Lupu hoard can be included in the same group. Regarding the annular ornaments, the hoards of the first group contain either twisted wire bracelets or simple massive bangles (the two types are never found together in the same assemblage). Sometimes they are associated with large silver spirals which could have been worn either on the upper arm or on the leg below the knee48. Wire bracelets (or anklets?) having the ends twisted around each other already appeared during this horizon, though they became more popular during the following horizons. There was also a marked preference for certain types of composite belts. Regarding the metal vessels, the hoards of this group sometimes contain Hellenistic silver cups of the mastos type and kan‑ tharoi, as well as some types of late Republican bronze vessels, for example situlae and beakers of the Gallarate type. Lastly, some hoards also contain Hellenistic coins: tetradrachms of Thasos, or drachms of the Illyrian colonies at Dyrrachium and Apollonia. In only one case, the latter issues are associated with Republican denarii, the latest dated coin being from 75 BC. The Sărăcsău hoard, which contains two costume assemblages, one belonging to an adult and the second to a child49, whose composition is specific to the first horizon, is including one brooch with a rhombic plate on the bow, part of the child costume, which is specific to the second horizon. This particularity seems to suggest that the hoard was interred slightly later than other hoards from the same group, thus marking a transitional phase towards the second horizon in the evolution of this phenomenon. The second group of hoards is defined by the presence of brooches having a rhombic plate (“shield”) on the bow (Fig. 2/2). The brooches whose plate is decorated with anthropomorphic 47 Contra ZIRRA 2017, 74–75, 185. 48 For the latter possibility, see MEDELEȚ 1993. 49 MEDELEȚ 1994a, 217. 20 Aurel Rustoiu motifs, found in the Bălănești and Coada Malului hoards, can be included in the same group due to their morphological similarity. They are associated with spiral ornaments (hair spirals and large bracelets) having the ends decorated with punched palmettes and zoomorphic protomes, wire bracelets (or anklets?) having the ends twisted around each other, and earrings. The late La Tène type brooches and the twisted wire bracelets continued to also be used in some cases. Some hoards of this group contain late Republican denarii whose latest issuing date is around 40 BC. The hoards belonging to the third horizon sometimes still include brooches having a rhombic plate on the bow, though the group is primarily defined by the presence of spoon- shaped brooches (Fig. 2/3). These assemblages also include spiral ornaments having the ends decorated with punched motifs, and bracelets (or anklets?) having the ends twisted around each other. Among them, the Vedea hoard contains a silver beaker of the Gallarate type, which seems to have remained in use for a longer period. Some of the hoards also contain silver coins usually dated to the second half of the 1st century BC and the Augustan period. The Remetea Mare hoard stands out among these discoveries due to the presence of some earlier dated artefacts, like the silver bangles specific to the first group and the Thasos tetradrachms. At the same time, the hoard contains a series of later dated brooches, which are not appearing in other contemporaneous assemblages. Most specialists, including K. Horedt50 and D. Spânu51, ascribed these brooches to the late LT type which is sometimes encountered in hoards belonging to the first and second groups. However, they belong to other types, including the brooches having a large bilateral spring and the chord attached to the bow, which have been found in settlements from pre-Roman Dacia in contexts mainly dated to the 1st century AD52, or the unusual brooch having two knobs on the bow, which probably imitates similar brooches of the Norican-Pannonian type made of bronze53, dated to the 1st century AD. Aside from that, the Remetea Mare hoard has an unusual functional structure, pointing to a slightly different style of bodily ornamentation that is reflecting perhaps the manner in which this phenomenon has evolved at the periphery of the area controlled by the Dacian kingdom at the beginning or in the first half of the st1 century AD. Regarding the absolute chronology, a series of artefacts belonging to these hoards offer some relevant data for each of the aforementioned groups. Thus the middle La Tène type brooches with knobs, which are specific to the first group of hoards, represent the later regional “interpretation” of the iron or bronze brooches having large spheres on their foot, which are dated to the La Tène C1 – C254 in the eastern “Celtic” area of the Carpathian Basin55. A similar origin could be presumed for the silver or bronze brooches of the Jarak type, which are specific to the hoards and funerary inventories from the Scordiscian area in modern Serbia56; these are contemporaneous with the aforementioned brooches with knobs from pre-Roman Dacia. The Jarak type brooches are dated to the La Tène D1 (corresponding to the so-called Beograd 3 horizon)57. Similar brooches have been found in a hoard hidden under the floor of a house belonging to the first layer of habitation at Židovar (north-eastern Serbia), being dated to the first half of the st1 century BC58 (Fig. 4). On the other hand, the radiocarbon dating of some wood fragments recovered from the funerary pyre of a tumulus burial located near the Dacian

50 HOREDT 1973, 133, Fig. A2b-c. 51 SPÂNU 2012, 45–46, Type 2.1.1. 52 RUSTOIU 1997a, 43–44, Type 10. 53 GARBSCH 1965, Type A 236a-c. 54 See, for example, DIZDAR 2009, 295–297, Pl. 4–5. 55 RUSTOIU 1997b, 152–153, Fig. 2. 56 BOŽIČ 1981, 320, 328, Pl. 4/43; GUŠTIN 1984, Pl. 50/1. 57 BOŽIČ 1981; BOŽIČ 2008, 145–146; GUŠTIN 1984. 58 JEVTIĆ ET AL. 2006; RUSTOIU ET AL. 2017, Fig. 13. Commentaria archaeologica et historica (III). Chronology of the Dacian silver hoards 21 fortress at , which contained a panoply of weapons and a ceremonial chariot specific to the La Tène D1, as well as a bronze situla of the Eggers 20 type, indicates a date range between 160 – 50 and 195 – 95 BC59. Likewise, the presence of Hellenistic coins and rarely of the late Republican ones, as well as that of the metal vessels of Mediterranean origin, provide supple‑ mentary arguments for the dating of the first group of hoards to the 150/125 – 75/50 BC60. The second group of hoards is characterized by the presence of the brooches having a rhombic plate on the bow. The type originates from the Orlea-Maglavit type brooches which were specific to the area occupied by the so-called Padea-Panagjurski Kolonii group, being dated to the La Tène D161. Accordingly, the brooches with a rhombic plate have to be dated later, to the La Tène D262. The same group of hoards also contains silver spiral ornaments (bracelets and hair spirals) having the ends decorated with punched palmettes and zoomorphic protomes. Similar ornaments made of gold have recently been discovered in the Grădiștea de Munte area ( County), in south-western Transylvania (Fig. 3), the location of the Dacian king‑ dom’s centre of power mentioned in ancient literature as Ζαρμισεγεθουσα τò βασίλειον63. The archaeological site at Grădiștea de Munte cannot be dated earlier than the middle or the second half of the 1st century BC64. At the same time, one spiral ring has been found in a context belonging to the second layer of habitation of the aforementioned settlement at Židovar, dated to the 50 BC – AD 10/2565 (Fig. 4). Some of the hoards also contain late Republican denarii that are dated not later than just before the Augustan period. Accordingly, the hoards belonging to the second group could be dated to the La Tène D2, corresponding to the 75/50 – 30/25 BC. It has to be noted that the period during which these hoards were interred was significantly shorter than that of the first group’s hoards. This pattern could also explain, at least partially, the smaller number of discoveries. The third group of hoards still contains brooches with a rhombic plate on the bow, though the spoon-shaped ones are also included. The latter brooches represent the local “inter‑ pretation” of the brooches having the spring protected by a concave cover (Schüsselfibeln), which are encountered in Central Europe and northern Italy. Unlike the western examples, the brooches from pre-Roman Dacia were made of two separate elements: the body, and the spring with the pin, which were then soldered together. The spoon-shaped brooches appeared at the end of the 1st century BC and became popular in some areas of pre-Roman Dacia until the middle of the 1st century AD66. At Židovar, such artefacts are associated with Jezerine type brooches, while at Liubcova (Caraș-Severin County), on the left bank of the Danube’s Iron Gates, the spoon- shaped brooches are present in the first phase of the settlement which ended at the beginning of the 1st century AD67 (Fig. 4). Lastly, a fragmentary brooch of the same type was found in grave 9 from the cemetery at Belgrade-Karaburma, which was dated to the second half of the 1st cen‑ tury AD68. Regarding the end date of the third group of hoards, an important clue is provided by the absence of the strongly profiled brooches of Roman origin, which arrived in pre-Roman Dacia in large numbers only after the organization and consolidation of the province of , 59 TELEAGĂ 2014. 60 See the absolute dating of the LT D1 in RIECKHOFF 1995 and RUSTOIU 1996b. 61 POPOVIĆ 1991; RUSTOIU 1997a, 37–38; ZIRRA 2017, 62 suggested a dating to the end of the 2nd century and the first half of the 1st century BC. 62 RUSTOIU 1997a, 44–46; ZIRRA 2017, 70 dated them to the end of the 1st century BC. 63 Ptolemy, Geogr. 8.11.4; see also SPÂNU 2012, 167–178; RUSTOIU 2016a, 73–81. 64 GLODARIU 1995, 124–130; RUSTOIU 2016a, 76. 65 RUSTOIU ET AL. 2017, Fig. 13. 66 See a wider discussion in RUSTOIU 1997a, 48–50. The dating proposed in 1997 was later confirmed by recently published discoveries, for example those from Poiana-Movila Vârtop: SPÂNU 2019b. 67 RUSTOIU ET AL. 2017, Fig. 13. 68 EGRI 2016, 346, Fig. 1. 22 Aurel Rustoiu

Fig. 3. Gold bracelets from Sarmizegetusa Regia (photo A. Rustoiu). mainly in the second half of the 1st century AD69. Lastly, these hoards sometimes include late Republican and early Imperial denarii. As a consequence, the third group of hoards could be dated to the Augustan – Tiberian period, corresponding to 30/25 BC – AD 25/30. Summarising the above observations, the chronology of the three groups of hoards con‑ taining silver jewellery from pre-Roman Dacia covers the following periods: – First group: La Tène D1, 150/125 – 75/50 BC; – Second group: La Tène D2, 75/50 – 30/25 BC; – Third group: the Augustan – Tiberian period, 30/25 BC – AD 25/30.

4. Conclusions. Cultural-historical groups and the late La Tène chronology in the lower Danube region and Transylvania Starting from the chronology of the Dacian silver hoards, which in the opinion of D. Spânu covers the La Tène D2a and D2b sub-phases from the Central European chronology, the same author has attempted to fill the periods “before the (sic!) hoards” and “after the hoards”,

69 RUSTOIU 2017, 52–53. Commentaria archaeologica et historica (III). Chronology of the Dacian silver hoards 23 Fig. 4. Chronological evolution of the settlements from the Iron Gates region (after RUSTOIU ET AL. 2017). (after RUSTOIU region Gates the Iron of the settlements from evolution Chronological Fig. 4. 24 Aurel Rustoiu more precisely the La Tène C2 and D1 and Eggers B270. In order to achieve this, D. Spânu has appealed to a few artificial cultural-chronological constructs. For example, the Padea-Panagjurski Kolonii group, characterized by burials containing sets of weapons and / or harness fittings (and rarely chariot fittings) in their inventories, is only localized in south-western Romania, north-western Bulgaria and the Danube’s Iron Gates region, on both banks71 (Fig. 5). In reality, the distribution area of these burials is much wider, and this has been noted quite often in the specialist literature of the last few years72 (Fig. 6/1). These burials reflect a particular manner of expressing the martial identity of the elites during the “Dacian horizon” of the La Tène C2 – D2 phases73. Regarding the late La Tène phase that characterized the “Dacian horizon”, D. Spânu has suggested to identify it as the “Grădiștea de Munte – Poiana group” (Fig. 5). The two epony‑ mous sites, which are supposed to be representative for the entire “Dacian horizon” during the La Tène D2 and Eggers B2 phases, had almost nothing in common. Archaeological evi‑

Fig. 5. Cultural groups after SPÂNU 2019a. dence from Grădiștea de Munte indicates the presence of a settlement having urban charac‑ teristics, a “central place” that has no other parallel in pre-Roman Dacia74. On the other hand,

70 SPÂNU 2019a, 182–195, Fig. 6 (chronological scheme). 71 SPÂNU 2019a, Fig. 1. 72 RUSTOIU 2005; RUSTOIU 2012; ŁUCZKIEWICZ/SCHÖNFELDER 2008; POPA 2018 etc. 73 RUSTOIU 2008; RUSTOIU 2018 etc. 74 Unfortunately, the results of the archaeological investigations made in the Orăștie Mountains during the last century are still limited to a number of quite brief archaeological reports published over the years and some Commentaria archaeologica et historica (III). Chronology of the Dacian silver hoards 25

Fig. 6. 1. Distribution of graves containing weapons of the Padea-Panagjurski Kolonii (after ŁUCZKIEWICZ, SCHÖNFELDER 2008 and RUSTOIU 2012); 2. Dacian kingdom from Burebista to : dotted line – territories controlled by Burebista; dashed line – extent of the Dacian kingdom after Burebista and until the Roman conquest; black squares – Greek cities on the western Black Sea coast; black dots – Dacian settlements on the Siret valley; black stars – fortified Dacian settlements with archaeological layers dated to the 1st century BC – 1st century AD. (after RUSTOIU/FERENCZ 2018). 26 Aurel Rustoiu the settlement at Poiana (Galați County) is located at the periphery of the “Dacian” world, belonging to a particular socio-political and cultural environment specific to the Siret valley (its evolution being similar to that of the settlements at Răcătău and Brad)75. Chronologically, this settlement was surely established in the La Tène D1 phase76. Likewise, in order to define the horizon preceding the hoards, D. Spânu has invented the “Cetățeni – Vadu Vacilor group”. This is supposedly characterizing the La Tène C2 and D1 phases, the dating being based on a series of amphorae. This kind of dating is not exactly helpful, since these amphorae have only been discovered outside the Carpathians range and are absent from Transylvania during the period in question77. Regarding the discoveries from the east of the Carpathians, the Poienești – Lukašeuca group, well defined and analysed by M. Babeș78 and more recently by V. Iarmulschi79, is not raising particular chronological questions. Still, this cultural phenomenon is rather marginal for the subject of this article (Fig. 5). On the basis of these observations, it can be said that the cultural and archaeolog‑ ical constructs used by D. Spânu are not supported by archaeological evidence. These are not reflecting the true evolution of the communities from the lower Danube region, and even less of those from the territories inside the Carpathians range at the end of the Late Iron Age and the beginning of the imperial period. It would have been more appropriate to use the existing archaeological evidence from the lower Danube region and Transylvania as a starting point for this kind of analysis. In the later region, the “Dacian cultural and chronological horizon” succeeded the earlier “Celtic horizon”80, while in the lower Danube region the “Dacian horizon” followed the one that characterized the period of the 5th – 3rd centuries BC81. Returning to the situation from Transylvania, it is quite clear that the “Celtic horizon” finished at the end of the La Tène C1, as attested by the evolution of the Celtic cemeteries which suddenly ceased to be used at the end of this period. A similar fate has been noted in the case of the Celtic settlements from this region. On the middle Mureş valley, in south-western Transylvania, Celtic rural settlements that ended in the La Tène C1 were followed on the same spots by new settlements having inventories specific to the “Dacian horizon”82 (Fig. 7). This succession is clearly visible stratigraphically, and the settlement at Şeuşa, near Alba Iulia, provides one of the best documented examples83 (Fig. 8). In the first horizon of the settle‑ ment were discovered bronze brooches specific to the La Tène C1 phase, which have good analogies, for example, in cemeteries from Slovakia84, and also in the nearby regions85. They were accompanied by glass bracelets of the Haevernick 6b, 7b and 8b types, dated to the La catalogues of finds which were more recently published. This is contributing to the emergence of many doubts regarding the discoveries from the area of the main Dacian centre of power. Among the contributions which attempt to synthesize the results of these investigations, see FLOREA 2011 and FLOREA 2014, and also PEȚAN 2018. 75 See URSACHI 1986–1987; VULPE/TEODOR 2003; RUSTOIU 2002b etc. 76 RUSTOIU 2005b. See a recent detailed analysis of the brooches discovered at Poiana in CIUBOTARU 2020. 77 See EGRI 2014; EGRI 2019, 128–130, Fig. 81–82. 78 BABEȘ 1993. 79 IARMULSCHI 2016. See also BABEȘ/IARMULSCHI forthcoming. I am grateful to V. Iarmulschi for allowing me to read the book before publication. 80 RUSTOIU 2018. 81 MĂNDESCU 2010. 82 RUSTOIU 2005. 83 RUSTOIU/FERENCZ forthcoming. 84 See BUJNA 2003, Type H2-A, Fig. 51. 85 See the brooch from Telești: RUSTOIU 2004–2005, Fig. 1/5. Commentaria archaeologica et historica (III). Chronology of the Dacian silver hoards 27

Fig. 7. Transition from Celtic to Dacian horizon in SW Transylvania (after RUSTOIU 2005). 28 Aurel Rustoiu

Fig. 8. Stratigraphic succession of the settlements at Şeuşa: 1 Celtic horizon; 2 Dacian horizon (after RUSTOIU/FERENCZ forthcoming).

Tène C186. Lastly, a horse bit that is typical to the Celtic horizon belongs to the same archaeo‑ logical layer87. Regarding the second horizon of the settlement, belonging to the Dacian period, it included, among other things, iron brooches of the middle La Tène type having small knobs on the foot, which in the Taurisci environment are dated to the La Tène C1b/C288. They were accompanied by a cross-shaped pendant made of bronze, which in the Dacian environment appeared in the La Tène C289, as well as a series of artefacts specific to the settlements and funerary contexts dated to the La Tène D1, for example a ceramic relief-decorated hemispherical cup90 and a horse bit of the “Thracian” type91. Regarding the funerary discoveries belonging to the “Dacian horizon”, the majority of the burials containing weapons of the Padea-Panagjurski Kolonii type have been dated to the La Tène D1. The single cemetery containing archaeological finds belonging to the La Tène C2 was recently unearthed at Sâncrai, on the middle Mureş valley. In the cemetery, the cremated remains were placed in rectangular pits. Six burials were unearthed so far, four of them con‑ taining weapons, military equipment and costume accessories which allow a precise dating92. The latest dated burials belong to the La Tène D2 (the Augustan period), as one grave from Piatra Craivii (Alba County) is demonstrating93.

86 HAEVERNICK 1960; GEBHARD 1989, 128–134. 87 ZIRRA 1981; WERNER 1988, Type XIVA. 88 DIZDAR 2013, 232–235. 89 RUSTOIU 1996b, 122, Type 1. 90 EGRI 2014, with the bibliography. 91 RUSTOIU 2002a, 16, with bibliography. 92 RUSTOIU ET AL. 2017; RUSTOIU/FERENCZ forthcoming. 93 RUSTOIU/GHEORGHIU 2010. Commentaria archaeologica et historica (III). Chronology of the Dacian silver hoards 29

The La Tène D1, when the burials containing sets of weapons became more numerous and the first silver hoards appeared is also the period in which the first hilltop fortresses and fortified settlements were established on the main waterways from Transylvania. These types of settlements continued to exist until the Roman conquest of Dacia (Fig. 6/2). A detailed contex‑ tual analysis of each site could provide a more precise internal chronological evolution, but aside from a few exceptions, this kind of information is missing. However, the available archaeological evidence clearly indicates that the hilltop fortresses, the burials with sets of weapons and the silver hoards are interconnected as parts of a wider social and cultural phenomenon (see Table 2).

Table 2. Chronological evolution of the social and cultural manifestations at the end of the Iron Age in Transylvania.

Concomitantly with the emergence of the hilltop fortresses and fortified settlements, a number of new ritual practices also emerged. For example, pits containing human remains, often dismembered or bearing traces of violence, have been discovered within the settlements (fortified or rural) and in their surroundings. These rituals could be explained from different perspectives, according to their contexts. They reflect the different funerary treatment of some members of the communities, or perhaps of the foreigners94. It can be therefore concluded that the scenario proposed by D. Spânu for the chro‑ nology of the end of the Late Iron Age in the lower Danube region and Transylvania, based on an erroneous dating of the silver hoards, is not credible. This kind of analysis requires the incor‑ poration of different categories of archaeological evidence belonging to the last three centuries before the Roman conquest. Their contextual interpretation could provide a more reliable and detailed chronology of the evolution of the communities from each of the two territories in question. Lastly, this evolution was not uniformly equal across wide areas, as suggested by the typo-chronological tables and schemes drawn in the office by some researchers, since each com‑ munity had its own history and evolution governed by a multitude of different social, economic, demographic or ecological factors. Some of these communities went through a number of rapid or radical transformations, whereas others largely maintained their lifestyle over longer periods of time, with only some gradual changes that are often difficult to observe archaeologically. For this reason, the observations made some years ago remain valid. “Individuals and communities from the past created and used material culture according to their own needs, irrespective of the modern researchers’ agendas! Accordingly, certain goods could have remained in use for a longer period of time, even if archaeologists assumed that they ceased to exist after the end of the corresponding chronological phase or sub-phase... In other cases, the chronological gaps appearing between regions and communities could be explained by the fact that particular social, economic or cultural changes happened at different moments in time due to various

94 RUSTOIU 2016c. 30 Aurel Rustoiu localized reasons. The beginning of the Early Iron Age and the spread of iron metallurgy, the so-called Latènization of temperate Europe, or the Romanization are all examples of complex cultural phenomena that went through different stages and embraced different forms from one region to another. As a consequence, archaeological evidence related to these cultural transfor‑ mations could have taken different shapes through time and across space. Aside from regional variations in material culture, these processes also led to changes in the local social, political, economic or behavioural norms and practices, which had an impact on both the indigenous and the incoming populations. The question is whether any chronological scheme, elaborate or not, is able to incorporate the diversity of these cultural transformations. This is a theme of reflection especially for archaeologists who are devoted to sophisticated chronologies”95. These observa‑ tions can also apply to the case of the chronology of the Dacian silver hoards.

BIBLIOGRAPHY

BABEȘ 1993 M. BABEŞ, Die Poieneşti-Lukaševka-Kultur. Ein Beitrag zur Kulturgeschichte im Raum östlich der Karpaten in den Jahrhunderten vor Christi Geburt (Bonn 1993).

BABEȘ/IARMULSCHI forthcoming M. BABEȘ/V. IARMULSCHI, Așezarea și necropola de tip Poienești-Lucașeuca de la Borosești (jud. Iași). Forthcoming.

BOŽIČ 1981 D. BOŽIČ, Relativna kronologija mlajše železne dobe v jugoslovanskem Podonavju. AVes 31, 1981, 315–336.

BOŽIČ 2008 D. BOŽIČ, Late La Tène-Roman cemetery in Novo Mesto. Studies on fibulae and on the relative chronology of the Late La Tène period (Ljubljana 2008).

BUJNA 2003 J. BUJNA, Spony z keltských hrobov bez výzbroje z územia Slovenska (Typovo- chronologické triedenie LT B- a C1 spôn). SlovA 51, 1, 2003, 39–108.

CAUUET ET AL. 2003 B. CAUUET/B. ANCEL/C. RICO/C. TĂMAȘ, Rețelele miniere antice. Misiunile arheologice franceze (1999–2001). In: P. Damian (ed.), Alburnus Maior I (București 2003), 471–530.

CAUUET ET AL. 2018 B. CAUUET/C. G. TĂMAȘ/M. BOUSSICAULT/M. MUNOZ, Quantites et controle de l’or produit a l’age du fer en Gaule du Centre-Ouest. In: C. Rico/A.Orejas (eds.), Los metales preciosos: de la extracción a la acuñación (Antigüedad – Edad Media). Dossier des Mélanges de la Casa de Velázquez. Nouvelle série 48, 1, 2018, 13–42.

CHIDIOȘAN/ORDENTLICH 1973 N. CHIDIOȘAN/I. ORDENTLICH, Un tezaur de argint de la Oradea. Crisia 3, 1973, 97- 108.

CHIRILĂ/MATEI 1986 E. CHIRILĂ/AL. V. MATEI, Tezaurul dacic de la Cehei. ActaMusNapoca 10, 1986, 97- 117.

95 RUSTOIU 2016d, 235–236. Commentaria archaeologica et historica (III). Chronology of the Dacian silver hoards 31

CHIRILĂ/MIHĂESCU 1969 E. CHIRILĂ/G. MIHĂESCU, Tezaurul monetar de la Căprioru. Contribuții la studiul cronol‑ ogiei și circulației tetradrahmelor Macedoniei Prima și ale Insulei Thasos (Târgoviște 1969).

CHIȚESCU 1980 M. CHIŢESCU, Les monnaies géto-daces de type romain républicain et imperial. Dialogues d’Histoire ancienne 6, 1980, 123–138.

CIUBOTARU 2020 N. CIUBOTARU, Rolul fibulelor în construirea identităților individuale și de grup. Studiu de caz: așezarea getică de la Poiana (sec. I a. Chr. – I p. Chr.). Teză de licență (Cluj-Napoca 2020). Manuscript.

CONOVICI 1986 N. CONOVICI, Aspecte ale circulației drahmelor din Dyrrhachium și Apollonia în Peninsula Balcanică și în Dacia. BSocNumRom 77–79, 1983–1985 (1986), 68–88.

CRIȘAN 1978 I. H. CRIȘAN, (Arad 1978).

DACII 2004 DACII 2004, A. Pescaru et al. (eds.), DACII. Catalogul expoziției (Deva 2004).

DAVIS 2006 P. DAVIS, Dacian imitations of Roman Republican denarii. Apulum 43, 321–356.

DIZDAR 2009 M. DIZDAR, Željezni pjasi tipa Zvonimirovo LT 6 kao dio ženske srednjolatenske nošnje u zapadnoj Panoniji i na Jugoistočnoalpskom prostoru. VjesAMuzZagreb 42, 2009, 271–304.

DIZDAR 2013 M. Dizdar, Zvonimirovo/Veliko polje. A cemetery of the La Tène culture (Zagreb 2013).

DOMERGUE ET AL. 2006 C. DOMERGUE/V. SERNEELS/B. CAUUET/J.-M. PAILLER/S. ORZECHOWSKI, Mines et métallurgies en Gaule à la fin de l’âge du Fer et à l’époque romaine. In: D. Paunier (ed.), Celtes et Gaulois, l’Archéologie face à l’Histoire, 5: la romanisation et la question de l’héritage celtique. Actes de la table ronde de Lausanne, 17–18 juin 2005 (Glux-en-Glenne 2006), 131–162.

EGRI 2014 M. EGRI, Mediterranean wine and Dacian conviviality. Ancient and modern myths and archae‑ ological evidence. In: C. N. Popa/S. Stoddart (eds.), Fingerprinting the Iron Age: Approaches to Identity in the European Iron Age. Integrating South-eastern Europe into the Debate (Oxford 2014), 48–62.

EGRI 2016 M. EGRI, The Beograd 4 horizon in the Scordiscian environment. Chronological delimitation and interpretation. In: S. Berecki (ed.), Iron Age Chronology in the Carpathian Basin. Proceedings of the international colloquium from Târgu Mureş 8–10 October 2015 (Cluj-Napoca 2016), 339–356. 32 Aurel Rustoiu

EGRI 2019 M. EGRI, Connectivity and social dynamics in the Carpathian Basin (1st century BC–1st century AD). An archaeological investigation (Cluj-Napoca 2019).

EGRI/RUSTOIU forthcoming M. EGRI/A. RUSTOIU, Late Iron Age silver hoards from Dacia, forthcoming.

FLOREA 2011 G. FLOREA, Dava et oppidum. Débuts de la genèse urbaine en Europe au deuxième âge du Fer (Cluj-Napoca 2011).

FLOREA 2014 G. FLOREA, Sarmizegetusa Regia – the identity of a royal site?. In: C. N. Popa/S. Stoddart (eds.), Fingerprinting the Iron Age. Approaches to identity in the European Iron Age. Integrating South-Eastern Europe into the debate (Oxford 2014), 63–75.

GARBSCH 1965 J. GARBSCH, Die norisch-pannonische Frauentracht im 1. und 2. Jahrhundert (München 1965).

GEBHARD 1989 R. GEBHARD, Der Glasschmuck aus dem Oppidum von Manching. Die Ausgrabungen in Manching 11 (Stuttgart 1989).

GLODARIU 1995 I. GLODARIU, Addenda aux “Points de repere pour la chronologie des citadelles et des établisse‑ ments daciques des Monts d’Orăștie”. ActaMusNapoca 32/I, 1995, 119–134.

GLODARIU ET AL. 1992 I. GLODARIU/E. IAROSLAVSCHI/A. RUSU, Die Münzstatte von Sarmizegetusa Regia, EphemNapoc 2, 1992, 75–58.

GUŠTIN 1984 M. GUŠTIN, Die Kelten in Jugoslawien. Übersicht über das archäologische Fundgut. JbRGZM 31, 1984, 305–363.

HAEVERNICK 1960 TH. HAEVERNICK, Die Glassarmringe und Ringperlen der Mittel- und Spätlatènzeit auf dem europäischen Festland (Bonn 1960).

HOREDT 1973 K. HOREDT, Die dakischen Silberfunde. Dacia 17, 1973, 127–167.

IARMULSCHI 2016 V. IARMULSKI, A few considerations regarding the chronology of the Poenești- Lucașeuca culture. PZ 91, 2, 471–494.

IAROSLAVSCHI 2001 E. IAROSLAVSCHI, Die Münzstanze von Ocnita und Einige Bemerkungen zu den Dakischen Stanzen. ActaMusNapoca 38, 2001, 17–26. Commentaria archaeologica et historica (III). Chronology of the Dacian silver hoards 33

JEVTIĆ ET AL. 2006 M. JEVTIĆ/M. LAZIĆ/M. SLADIĆ, The Židovar treasure. Silver jewellery hoard from the settlement of Scordisci (Vršac – Beograd 2006).

LIPOVAN 1990 I. T. LIPOVAN, Cu privire la ceramica cu glazură plombiferă din Ampelum. StCercIstorV 41, 3–4, 1990, 273–291.

LOCKYEAR 2004 K. LOCKYEAR, The Late Iron Age Background to Roman Dacia. In: W. S. Hanson/I. P. Haynes (eds.), Roman Dacia. The Making of a Provincial Society, Journal of Roman Archaeology Suppl. Ser. 56 (Portsmouth, RI, 2004), 33–74.

LOCKYEAR 2008 K. LOCKYEAR, Aspects of Roman Republican coins found in Late Iron Age Dacia. In: V. Spinei/L. Munteanu (eds.), Miscellanea Numismatica Antiquitatis. In honorem septagenarii magistri Virgilii Mihailescu-Bîrliba oblata (Bucureşti 2008), 147–176.

ŁUCZKIEWICZ/SCHÖNFELDER 2008 P. ŁUCZKIEWICZ/M. SCHÖNFELDER, Untersuchungen zur Ausstattung eines späteisen‑ zeitlichen Reiterkriegers aus dem südlichen Karpaten- oder Balkanraum, JbRGZM 55, 2008, 159–210.

LUPU 1989 N. LUPU, Tilișca. Așezările arheologice de pe Cățănaș (București 1989).

MĂNDESCU 2010 D. MĂNDESCU, Cronologia perioadei timpurii a celei de a doua epoci a fierului (sec. -V III a. Chr.) între Carpați, Nistru și Balcani (Brăila 2010).

MEDELEȚ 1992 F. MEDELEȚ, A propos de l’exploitation et de l’utilisation des métaux précieuses dans la Dacie préromaine. Symposia thracologica 9, 1992, 226–235.

MEDELEȚ 1993 F. MEDELEŢ, Au sujet d’une grande spirale dacique en argent du Musée National de Belgrade. Caietele Banatica (Reşiţa 1993).

MEDELEȚ 1994a F. MEDELEŢ, În legătură cu o mare spirală dacică de argint aflată în muzeul Naţional din Belgrad. Addenda: unele precizări cu privire la observaţiile asupra tezaurelor de argint din Latène-ul târziu. Analele Banatului 3, 1994, 192–230.

MEDELEȚ 1994b F. MEDELEȚ, Contribuții la repertoriul numismatic al Banatului. Epoca Latène (sec. IV î. Chr. – 106. Analele Banatului 3, 1994, 239–308.

NRIAGU 1985 J. O. NRIAGU, Cupellation: The Oldest Quantitative Chemical Process. Journal of Chemical Education 62, 8, 1985, 668–674.

PEȚAN 2018 A. PEȚAN, Sarmizegetusa Regia, I. Redescoperirea cetății (Alun 2018). 34 Aurel Rustoiu

POP 2008 H. POP, Argintul dacic sălăjean (Cluj-Napoca 2008).

POPA 2018 C. N. POPA, Modeling identities. A case study from the Iron Age of south-east Europe (Cham 2018).

POPOVIĆ 1991 P. POPOVIĆ, Fibule tipa “Orlea-Maglavit”. ZborMuzBeograd 14, 1, 1991, 319–326.

PREDA 1957 C. PREDA, Contribuţii la problema provenienţei argintului din tezaurele geto-dacice în lumina descoperirii monetare de la Stăncuţa. StCercIstorV 8, 1957, 1–4, 113–124.

PREDA 1973 C. PREDA, Monedele geto-dacilor (București 1973).

PREDA 1998 C. PREDA, Istoria monedei în Dacia preromană (București 1998).

PROKOPOV 1993 I. PROKOPOV, Notes about the Character of Some Large-scale Coinages in the Thrace during IInd-Ist Centuries B.C. In: Proceedings of the XIth International Numismatic Congress (Louvain-la-Neuve 1993), 173–180.

RIECKHOFF 1995 S. RIECKHOFF, Süddeutschland im Spannungsfeld von Kelten, Germanen und Römern. Studien zur Chronologie der Spätlatènezeit im südlichen Mitteleuropa (Trier 1995).

RUSTOIU 1996a A. RUSTOIU, Podoabe dacice de argint cu miez din metal de calitate inferioară. EphemNapoc 6, 1996, 43–54.

RUSTOIU 1996b A. RUSTOIU, Metalurgia bronzului la daci (sec. II î. Chr. – sec. I d. Chr.). Tehnici, ateliere şi produse de bronz (Bucureşti 1996).

RUSTOIU 1997a A. RUSTOIU, Fibulele din Dacia preromană (Bucureşti 1997).

RUSTOIU 1997b A. RUSTOIU, Influences celtiques dans l’orfèvrerie dacique du Ier s. av. J.-C. In: P. Roman (ed.), The Thracian World at the Crossroads of Civilisations I (Bucharest 1997), 152–167.

RUSTOIU 2002a A. Rustoiu, Războinici şi artizani de prestigiu în Dacia preromană (Cluj-Napoca 2002).

RUSTOIU 2002b A. RUSTOIU, Die östliche Gruppe des dakischen Schmucks. Eine Untersuchung bezüglich der interregionalen Beziehungen im vorrömischen Dakien im 1. Jh.v.Chr. – 1 Jh.n.Chr. In: A. Rustoiu/A. Ursutiu (eds.), Interregionale und kulturelle Beziehungen im Karpatenraum (2. Jahrtausend v. Chr. – 1. Jahrtausend n. Chr.) (Cluj-Napoca 2002), 191–226. Commentaria archaeologica et historica (III). Chronology of the Dacian silver hoards 35

RUSTOIU 2004–2005 A. RUSTOIU, Celtic-indigenous connections in Oltenia during middle La Tène. Observations concerning a Celtic grave from Teleşti. EphemNapoc 14–15, 2004–2005, 53–71.

RUSTOIU 2005 A. RUSTOIU, The Padea-Panagjurski Kolonii Group in south-western Transylvania (Romania). In: H. Dobrzańska/V. Megaw/P. Poleska (eds.) Celts on the margin. Studies in European Cultural Interaction (7th Century BC – 1st Century AD) Dedicated to Zenon Woźniak (Krakow 2005), 109–119.

RUSTOIU 2008 A. RUSTOIU, Războinici şi societate în aria celtică transilvăneană. Studii pe marginea mormân‑ tului cu coif de la Ciumeşti (Cluj-Napoca 2008).

RUSTOIU 2012 A. RUSTOIU, Commentaria Archaeologica et Historica (I). 1. The grave with a helmet from Ciumeşti – 50 years from its discovery. Comments on the greaves; 2. The Padea- Panagjurski kolonii group in Transylvania. Old and new discoveries. EphemNapoc 21, 2012, p. 159–183.

RUSTOIU 2015 A. RUSTOIU, The Celtic horizon in Transylvania. Archaeological and historical evidence. In: S. Berecki, Iron Age Settlement Patterns and Funerary Landscapes in Transylvania (4th – 2nd Centuries BC) (Târgu Mureş 2015), 9–29.

RUSTOIU 2016a A. RUSTOIU, Comentaria archaeologica et historica (II). Banatica 26, 2016, 73–100.

RUSTOIU 2016b A. RUSTOIU, Lords and ladies of the rings. Saddle-shaped finger-rings from the Carpathian Basin. ARozhl 68, Praga 2016, p. 333–362.

RUSTOIU 2016c A. RUSTOIU, Settlements of life and death: Human interments in late Iron Age sites from the Sighişoara region (Mureş county, Romania). In: S.-C. Ailincăi (ed.), Settlements of Life and Death. Studies from Prehistory to Middle Ages. Proceedings of an International Colloquium, Tulcea, 25th–28th of May 2016 (Cluj-Napoca 2016), 313–330.

RUSTOIU 2016d A. RUSTOIU, Some questions regarding the chronology of La Tène cemeteries from Transylvania. Social and demographic dynamics in the rural communities. In: S. Berecki (ed.), Iron Age Chronology in the Carpathian Basin. Proceedings of the international colloquium from Târgu Mureş 8–10 October 2015 (Cluj-Napoca 2016), 235–264.

RUSTOIU 2017 A. RUSTOIU, Silver jewellery in the early La Tène cemeteries from Banat. The hybridization of bodily ornaments. Dacia 61, 2017, 183–205.

RUSTOIU 2018 A. RUSTOIU, The Carpathian and Danubian area. In: C. Haselgrove, K. Rebay- Salisbury, P. S. Wells (eds.), The Oxford Handbook of the European Iron Age, Oxford Handbooks Online (Oxford 2018). 36 Aurel Rustoiu

RUSTOIU ET AL. 2017 A. RUSTOIU/I. V. FERENCZ/A. DRĂGAN, Dacian fortified settlements in the Iron Gates region during Late Iron Age (2nd century BC – 1st century AD). In: A. Zanoci et al. (eds.), Iron Age fortifications on the Tisa-Dniester space. Proceedings of Saharna Summer Colloquium July 14th–17th, 2016 (Brăila 2017), 187–208.

RUSTOIU/FERENCZ 2018 A. RUSTOIU/I. V. FERENCZ, Burebista at the Banat’s Danube. Archaeological and historical data regarding the evolution of settlements in the Iron Gates Region in the 1st century BC. Banatica 28, 2018, 123–149.

RUSTOIU/FERENCZ forthcoming. A. RUSTOIU/I. V. FERENCZ, The chronology of pre-Roman Iron Age in Transylvania. The archaeology of fluid ethnic and social identities. Forthcoming.

RUSTOIU/GHEORGHIU 2010 A. RUSTOIU/G. GHEORGHIU, „General” and “particular” in the dressing fashion and metalwork of pre-Roman Dacia (An iron variant of the Jezerine-type brooches from Piatra Craivii – Alba County). In: I. Cândea (ed.), The and their neighbours in Antiquity. Studia in Honorem Valerii Sîrbu (Brăila 2010), 447–457.

SPÂNU 2002 D. SPÂNU, Studien zum Silberschatzfund des 1. Jahrhunderts v.Chr. von Lupu, Rumänien. PZ 77, 2002, 84–136.

SPÂNU 2012 D. SPÂNU, Tezaurele dacice. Creaţia în metale preţioase din Dacia preromană (Bucureşti 2012).

SPÂNU 2019a D. SPÂNU, Core issues of Late La Tène periodization in Romania. Peuce 17, 2019, 167- 206.

SPÂNU 2019b D. SPÂNU, Fibulae with Canid Protome (Tierkopffibeln) in Romania. Plural 7, 2, 2019, 111–128.

SPÂNU/COJOCARU 2009 D. SPÂNU/V. COJOCARU, The Dacian hoard from Bucureşti-Herăstrău. Archaeological and archaeometallurgical approaches. MatCercA s.n. 5, 2009, 97–116.

STOICOVICI/STOICOVICI 1972 E. STOICOVICI/F. STOICOVICI, Monedele de argint dacice și specificul lor chimic și metalo‑ grafic. ActaMusNapoca 9, 1972, 375–382.

STOICOVICI/STOICOVICI 1973 E. STOICOVICI/F. STOICOVICI, Compoziția argintului din obiectele de podoabă dacice. ActaMusNapoca 10, 1973, 541–543.

STOS-GALE/GALE 2009 Z. A. STOS-GALE/N. H. GALE, Metal provenancing using isotopes and the Oxford archaeo‑ logical lead isotope database (OXALID). Archaeol. Anthropol. Sci. 1, 2009, 195– 213. Commentaria archaeologica et historica (III). Chronology of the Dacian silver hoards 37

TELEAGĂ 2014 E. TELEAGĂ, Die Scheiterhaufen aus Cugir und Tarinci: Ein Beitrag zu den Bestattungssitten der Balkanhalbinsel und des vorrömischen Dakiens in der Spätlatènezeit. PZ 89, 2014, 305–336.

UJES-MORGAN 2012 D. UJES-MORGAN, 1st Century B.C. Drachms of Apollonia and Dyrrhachium in the Territory of the Scordisci. A Prologue of the Roman Conquest of the Balkans. In: E. Paunov/S. Filipova (eds.), Studia in honorem Iliae Prokopov sexagenario ab amicis et discipulis dedicata (Veliko Tărnovo 2012), 367–388.

URSACHI 1986–1987 V. URSACHI, Fortificații dacice pe Valea Siretului. Carpica 18–19, 1986–1987, 31–52.

VASILESCU ET AL. 2015 A. VASILESCU/B. CONSTANTINESCU/D. STAN/M. RADTKE/U. REINHOLZ/G. BUZANICH/D. CECCATO, Studies on Ancient Silver Metallurgy using SR XRF and Micro-PIXE. Radiation Physics and Chemistry 117, 2015, 26–34.

VULPE/TEODOR 2003 R. VULPE/S. TEODOR, . Așezarea geto-dacică de la Poiana (București 2003).

WERNER 1988 M. WERNER, Eisenzeitliche Trensen an der unteren und mittleren Donau. PBF XVI/4 (München 1988).

WOOD/MONTERO-RUIZ 2019 J. R. WOOD/I. MONTERO-RUIZ, Semi-refined silver for the silversmiths of the Iron Age Mediterranean: A mechanism for the elusiveness of Iberian silver. TrabPrehist 76, 2, 2019, 272–285.

WOOD ET AL. 2019 J. R. WOOD/I. MONTERO-RUIZ/M. MARTINÓN‑TORRES, From Iberia to the Southern Levant: The Movement of Silver Across the Mediterranean in the Early Iron Age. Journal of World Prehistory 32, 2019, 1–31.

WOYTEK ET AL. 2012 B. E. WOYTEK/M. RODRIGUES/F. CAPPA/M. SCHREINER/M. RADTKE/U. REINHOLZ, Imitations of Roman Republican Denarii. New Metallurgical Data. American Journal of Numismatics 4, 2012, 133–162.

ZIRRA 1981 V. ZIRRA, Latènezeitliche Trensen in Rumänien. HambBeitrA 8, 1981, 115–171.

ZIRRA 2017 V. V. ZIRRA, Fibulele de schemă La Tène din România (Craiova 2017).

ZIRRA/SPÂNU 1992 V. V. ZIRRA/D. SPÂNU, Observaţii asupra tezaurelor de argint din La Tène-ul târziu. StCercIstorV 43, 1992, 401–423.