INTRODUCTION

             Authenticity and chronology  

Over the time Roman bronzes from Dacia have been analysed especially from the points of view of iconography and functionality. We think that it is high time they were regarded from another perspective, beginning with their authenticity being checked [1] , continuing by making a more accurate typological framing, and by trying a more narrow dating than the 2^nd-3^rd centuries, as we have done until now in most cases.

The first difficulty faced by the researchers of Roman bronze statuettes from Dacia is that for very few of them we know the conditions in which they were discovered, and that until now no lararium or closed complex with bronze statuettes has been attested for certain. That entails the authenticity of certain items belonging to the heritage in the museum of our country, having an unknown or uncertain place of discovery, originating in the antiquities trade, to be questioned. For instance, it is the case of the statuettes depicting Hercules of the Italic type, probably originating in Italy, but very likely not from the antiquity, but from the last century. Another unconquerable difficulty is that many statuettes discovered in Dacia can no longer be found in our museums [2] .  

Iconography  

In the present paper we have corrected iconographic interpretation errors, as, even if from that point of view the first issue has been generally accurate, a series of details allow a more refined iconographic interpretation, and, implicitly, a more accurate typological framing. Thus, we could recognize variants of Mercury in the hypostasis of Pantheus or Thot, resulting in the hypothesis that a statuette in the Museum in Cluj, originating in the camp of Ilişua, does not depict Bacchus, as stated in a monograph dedicated to the god [3] , but Libera or a Maenad, as set forth initially. A narrow examination of the pieces made it possible to find close analogies, and in certain cases entailed not only their being attributed to some workshops or craftsmen originating in various parts of the empire, but also a more accurate chronology.  

The iconographic repertoire of the bronze statuettes in Dacia corresponds to the heterogeneous nature of the colonization, the colonists brought “ex toto orbe Romano” as Eutropius put it (VIII 6.2.), to the wide range of deities and beliefs in the Roman world in the 2^nd and 3^rd centuries AD, and the ethnic diversity of the military deployed in Dacia [4] . It should be mentioned that the bronze images correspond not only to official cults, but especially to individual beliefs, hence the frequent depictions of Venus or Mercury. Moreover, at that time, the old Italic deities had been almost forgotten, as new or  synchretic deities were preferred. The deities of Celtic origin worshipped by interpretatio Romana are rare, and in Dacia there can be only one linked to the old autochtonous cults, maybe except for the knight-hero. However, the latter appears possibly, just once among the bronze statuettes, (no.140) and, frequently, on votive reliefs [5] .  

The range of representations from Dacia resembles from an iconographic and typological point of view that from Italy, other western and central provinces of the empire (Gallia, Germany, Raetia, Noricum) or contiguous with Dacia (the two Moesias, Pannonia). It is worth mentioning the wide range of representations, from Graeco-Roman deities to Egyptian and Oriental deities, from mythological characters to people. In Dacia the following deities and mythological characters are represented in bronze: Jupiter, Mars, Apollo, Mercury, Bacchus, Pan, Silvanus, Satyr, Priapus, Amor, Hercules, Lares, Genius Populi Romani?, Genius, Minerva, Diana, Venus, Fortuna, Abundantia, Victora, Libera?, Jupiter Dolichenus, Serapis, Men, Osiris. The representations of  Neptune, Vulcanus, Sucellus, and Epona lack, as well as the grotesque figures [6] .  

       The repertoire of bronzes from Dacia corresponds only partially to that of the votive reliefs or of the stone statuary, dominated by representations of originally non-Roman deities such as: Mithras, the Thracian knight or the Danubian knights [7] . There are deities represented in stone, like Pluto and Proserpine or Hecate, but that are not represented also in bronze. Instead, the images of Mercury and Venus, most frequently met among bronzes, are rarely represented in stone [8] . Only the bronze representations of Diana seem to correspond numerically and iconographically to the stone ones [9] .  

The repertoire of bronzes is not in accordance with that of the terra-cottas, much scarcer, where the image of Venus prevails as well as that of a pair sitting on a throne, either a man or a woman, or two women [10] . The clay representations of Venus, usually half naked, making a gesture of decency, resemble the bronze ones, and it is not out of question for them to have had a common source of inspiration.  

Provenience  

The bronze statuettes from Dacia come from towns, camps or rural settlements. As in other provinces of the Roman Empire also in Dacia bronze statuettes  must have been deposited in temples and places of worship, in household lararia or vestibules with various functions, as items, first of all for worship, but in some cases, also as objects with a decorative role [11] . Until now in Dacia no lararium has been discovered [12] , which is certainly a research lacuna. Roman cities, with residential districts, are investigated archaeologically very little. Often there are no data on the archaeological context  in which the statuettes were discovered, more often than not found by chance, so that it is almost impossible to establish whether they were lararia, small deposits of a bronze craftsman, who gathered bronze residue in order to reuse them or artifacts temporarily concealed in a certain place [13] . As already mentioned, in Dacia household lararia must have existed, as well as statuettes deposited in places of worship, in smaller numbers than in Italy and the western provinces of the empire, and probably heterogeneous from the point of view of the structure, stone or terra-cotta items next to bronze ones [14] .  

Technique  

The issues regarding the manufacturing technique, those linked to the series, and, implicitly, to the workshops where the bronze statuettes were cast, from our point of view have been cleared up by A. Kaufmann-Heinimann after the autopsy of a huge amount of pieces, and after their long, profound, thorough research. She We agree with all her findings: all the statuettes were cast by cire perdue  by direct or indirect modelling, by using dies for the various parts of the piece, respectively (Hilfsnegative); the bronze models could be anywhere and any time copied in wax, in their turn functioning as “prototypes”. The statuettes could be carried easily to remote distances from the place of manufacturing, and there they became the pattern for others, hence the difficulty to circumscribe or locate certain workshops that must have been small or medium-sized [15] . An interesting detail as regards the manufacturing technique of  bronzes is provided by the medium-sized statuette of Apollo from Romula (no.14) where one can see clearly that the drapery was made separately.  

As regards the sizes, most statuettes are small. Among medium-sized bronzes only the statuettes of Apollo can be included (nos.14 and 15).  

Workshops  

In Dacia until now there have been no attested workshops where they used to manufacture statuettes exclusively. The latter used to be brought from other provinces, cast on the spot by pedlars or manufactured in special workshops for other categories of objects.  

The researchers of Roman bronzes have been trying to define and establish the places of some workshops that functioned in the empire. Thus, R.Fleischer [16] finds a produciton of bronzes in central Italy at the end of the Republic and the beginning of the 1^st century AD. The products used to be exported through the town of Aquileia, then to northern Italy, south and central Gallia, on the Rhine, and, finally, to Noricum and Pannonia, as there were small workshops, and to Carnuntum, as we may speak about a local production later than in Gallia and Germany, as late as the end of the 2^nd century.. On the basis of certain casting flaws, scrap, unfinished pieces or with features typical of provincial art he mentions workshops at Avenches, Strasbourg, Brigetio, in southern Pannonia and northern Italy [17] . The same R.Fleischer [18] , remarking the somehow common style of many statuettes, proves the existence of a Raetian workshop that functioned during the second half of the 2^nd century or the first half of the 3^rd century, located at Regensburg or between Augsburg and Enns, but that failed to export its products too far away. In his turn L.Beschi [19] talks about workshops at Brescia, Este, Aquileia, Concordia, Industria, Treviso, Aosta and notices in the Veneto zone faithfulness towards the paleo-Veneto traditions, openness towards imports and Graeco-Oriental influences, in Emilia – links with central Italy, and in Lombardy, Piedmont, Liguria – affinities with the Graeco-Roman world, and preferences for products of Hellenistic traditions. Given the high number of items discovered, the issue of the workshops in Gallia has been tackled by researchers, first of all by S.Boucher, who thinks that in the valleys of the rivers Rhône and Saône there were workshops from the end of the 1^st century or at least from the beginning of the 2^nd century AD, that the central-east region of Gallia from then onwards played the role of mediator between Italy and the Gallic world [20] , that in certain cases one can notice a direct influence of Greece over the workshops in Gallia [21] . In Milan they discovered eight figurines worked by the same hand, metal scrap, raw cast artifacts, crucibles and a stone hand mill [22] , and in Strasbourg they discovered casting residue, a loop, a flawed statuette, at Königshoffen, in a vicus they discovered the traces of a workshop in a craftsmen’s district, at Hortburg, near Colmar, a handicraftsman’s workshop, at Brumath, in an assembly, they discovered fragments from clay dies, bronze residue associated to a statuette deposit and silver votive plates, a lot of objects for recasting [23] . In Germany they mention workshops at Aventicum [24] , Augusta Raurica [25] and Colonia Agrippina, a workshop that functioned during the second half of the 2^nd century AD [26] . A series of workshops are mentioned also in Noricum, at Lauriacum [27] , in Pannonia at Carnuntum [28] , Mursa, Siscia and Sirmium where they discovered unfinished artifacts, flawed, fragments of large sizes [29] . At the same time, it seems that a workshop functioned also in northern  Macedonia at Stobi [30] or in Moesia at Novae [31] , as well as on the Dalmatian coast [32] .  

A few of the statuettes discovered in Dacia can be attributed to some workshops or strolling craftsmen coming from the areas in question. From central Italy probably come Jupiter from Potaissa (no.1), Mars from Potaissa (no. 9), Apollo from Apulum (no.15), Mercury from Romula (no. 23), the lar from Sucidava (no. 68), the lar from Deva (no.73), from northern Italy: Apollo from Gherla (no.12), Mercury-Thot from Sucidava (no. 33), Venus from Apulum (no. 106); the lar from Orlea (no. 70), the lar from Micia (no. 72) [33] , and from Gallia: Mars from Porolissum (no. 7); Mercury from Napoca (no. 16); Mercury from Transylvania, with unknown place of discovery (no. 36) and Mercury Pantheus from Ilişua (no. 38). From the East must have come a single piece – Venus from Potaissa (no. 96), and from an Alexandria workshop, possibly, boxing Eros (no. 63) [34] .  

The most obvious links of Dacia are with Moesia Superior. Venus (no.116) must have come from the same workshop with an item from Singidunum, Venus (no.120) from the same workshop with one from the limes of Moesiei Superior, while Hercules (no.64) must have been cast in the same die with one from Ratiaria. Most items must have been cast on the spot. The high number of statuettes discovered at Drobeta could point to the existence of a workshop, but we should take into account that in the city situated on the Danube there was a busy trade, probably also with bronze items. A high number of statuettes come also from Romula. Strange is the case of the two cities at Apulum, where the number of statuettes is insignificant. The only plausible explanation would be that many of them have disappeared during the last century in the antiquities trade. The common stylistic features of many items are not conclusive and do not allow thie being included in series [35] . It is undoubtful that, although cast by cire perdue, for manufacturing the statuettes also bivalve dies were used [36] , otherwise we could not explain the striking resemblance between Hercules from Romula (no. 64) and Hercules from Ratiaria, that of Minerva from Porolissum (no. 80) and of Minerva from Oderzzo or that of Venus from Potaissa (no. 100) and Venus from Este. If there were workshops in Dacia, and future researches will confirm or refute the hypothesis, they must have functioned during the second half of the 2^nd century and the first decades of the 3^rd century.  Judging from some comon stylistic features of some statuettes one may set forth the existence of some workshops at Romula and at Potaissa. Two trends should be remarked: the female silouettes have long waists, the hips a little prominent like Venus from Gherla (no.102) or Venus from Potaissa (no. 96), and the male bodies are somehow heavy built, with excessively big hands, like those of the satyr from Porolissum (no. 45). Could some of the bronze statuettes announce the artistic features of late antiquity?  

From the point of view of the style and execution one may remark a few exquisite statuettes, such as: Jupiter from Potaissa (no. 1); Mars from Potaissa (no. 9); Apollo from Apulum (no. 15) or Venus from Potaissa (no. 96). Interesting from the point of view of the iconography are the statuettes of Diana in armour (no. 84) and of the dadophor  (no. 134), broad-shouldered, that could have belonged to a Mithriac sunsign, such as that from Angleur [37] .  

Chronology  

The chronology of the statuettes remains a difficult issue [38] . For as accurate a chronology as possible it is necessary to corroborate all the available data, from the conditions of the discovery, to the examination of style, from the hair style to the dated analogies. Thus, we could date back to the second half of the 1^st century AD : Jupiter from Potaissa (no. 1), the Lar from Sucidava (no. 68), the Lar from Transylvania (no. 69), the child-Lar from Deva (no. 73); during Augustus’ Age: Apollo (no. 15); by the end of the 1^st century – beginning of the second century: Mercury from Romula (no. 23), Mercury from Sarmizegetusa Regia (no  35), Mercury, with unspecified location from Transylvania (no. 36), Amor from Micia (no. 63); during Trajan’s Age: Venus from Apulum (no. 90); during Hadrian’s Age: Minerva from Ohaba (no. 79); during Trajan-Hadrian’s Age: Mars from Potaissa (no. 9); during the first half of the 2^nd century: Mercury from Drobeta (no. 22), Mercury-Thot from Drobeta (no. 32), the Lar from Micia (72); during Antoninus’ Age: Jupiter from Drobeta (no. 2), Silvanus (no. 46), Hercules from Romula (no. 66), Fortuna from Romula (no. 121), during Severus’ Age: Mercury from Ulpia Traiana (no. 27), Bacchus from Drobeta (no. 40), Pan from Gârla Mare (no. 42), the satyr from Gherla (no. 44), Venus from Hinova (no. 95), Venus from Gherla (no. 103), Venus from Potaissa (no. 110), and from the 3^rd century: the satyr from Porolissum (no. 45), the dadophor from Drobeta (no. 134).  

Perhaps better than other works of art the bronze statuettes express the beliefs, tastes, individual aspirations and even trade links [39] . As far as the bronzes from Dacia and those from contiguous provinces are concerned, it is obvious that they do not imitate works of major art [40] , but that some of them, the most achieved artisitically, preserve a vague influence from them, which served as a source of inspiration [41] . Most bronze statuettes are eclectic works [42] . The way to finding archetypes is sinuous, and there is no chance to discover vanished works of major art starting from bronze statuettes. Bronze artists, free of canons, let their imaginaiton wander and made the best of their talent and skilfulness. The bronze statuettes discovered north of the Danube may provide an interesting material for comparison, given that Dacia was a Roman province for two centuries only. As in the case of other creation of Roman art, also the bronze statuetes are relevant for the cultural and artisitc links with northern Italy, the contiguous provinces and the western ones of the empire, while standing proof to the uniformity in diversity of Roman civilization.