Produkcja późnych typów bransolet wężowatych na przykładzie znalezisk z cmentarzyska w Weklicach, stan. 7, pow. elbląski
Instytut Archeologii i Etnologii, Polskiej Akademii Nauk, Al. Solidarności 105, PL 00-140 Warszawa
Muzeum Archeologiczne w Gdańsku, ul. Mariacka 25/26, PL 80-833 Gdańsk
Data nadesłania: 20-01-2020
Data ostatniej rewizji: 14-03-2020
Data akceptacji: 07-04-2020
Data publikacji: 31-12-2020
Wiadomości Archeologiczne 2020;LXXI(71):161–187
The subject of this study is the technology of manufacture of late forms of silver shield-headed bracelets. The analysis is based on the bracelets from the Wielbark Culture cemetery at Weklice, Elbląg County, in N Poland (Fig. 1–3). They correspond to Blume III or Wójcik IVB and V types, and appear in single- and double-spiral variants. They are dated to the beginning of the Late Roman Period. The majority of such bracelets come from cemeteries located along the shores of the former bay of the Vistula Lagoon, whose remnant is present-day Drużno Lake. In antiquity, richly ornamented snake-headed bracelets with regular, strap and multi-spiral bodies were a distinctive type of women’s accessories. They are known from the Hellenistic Period (Fig. 4). They were also manufactured in goldsmith’s workshops of the Roman Empire (Fig. 5–7). In Roman goldsmithing, they were in fashion in the 1st and 2nd centuries CE; interest declined at the beginning of the 3rd century. The technique used (forging), the similarity of shapes and the regularity of profiles indicate that matrices or dies (swages) were used in their manufacture. The best-known example of blacksmithing and goldsmithing tools used to make ornaments utilising this method is the deposit from Daorson (BIH), the former capital of Illyria (Z. Marić 1979). Similar technology was used to craft the Roman snake-bracelets and snake-rings from the jeweller’s hoard from Snettisham, Norfolk (GB), dating to the mid-2nd century CE (C. Johns 1997). It is assumed that barbarian goldsmith’s workshops used raw materials imported from the Roman Empire. So far, no traces of exploitation and processing of non-ferrous metal ores in the Roman period have been recorded in Poland, allowing a conclusion that local workshops melted down Roman imports. The share of silver in the denarii varied and generally decreased as a result of successive reforms introduced by ruling emperors. However, metallurgical analyses (Table 1) show that shield-headed bracelets were made from high-grade raw material containing about 92–97% Ag, which excludes the possibility that the alloys were created by melting coins with varied silver content, e.g. fourrées. No archaeological sources confirm that the ‘barbarians’ had the ability to refine precious metals. Therefore, the raw material probably came from scrap vessels made of alloys containing 92–97% Ag. Given the enormous practical knowledge of goldsmiths of that time, the metal they had available was probably selected with respect to alloy composition. Raw material could also have been obtained by importing bars containing 94–95% Ag; however, such finds (known mainly from the frontier areas of the Roman Empire) date only to the 3rd and 4th century (K. Painter 1981). The fragments of cups discovered at the cemetery of the Wielbark Culture in Czarnówko, Lębork County, are an indication that high-grade silver from Roman vessels was used in Pomerania in the Roman Period. Metallurgical analyses show that they were made of alloys containing 96–99% Ag (J. Schuster 2018). In recreating the technology of manufacture of the bracelets in question, we also used our own observations concerning the assessment of alloy quality. Raw material was forged into long strips (up to 25 cm in length in the case of single-spiral forms, and up to 50 cm in length in the case of double-spiral forms) on which delamination and chipping could occur. They were the result of both the heterogeneity of silver and errors made during forging and are often still visible on final products (Fig. 8). This was possible due to the reduced hardness and resulting ductility of high-grade silver alloys with only a few percent of copper added. A common way of making the basic form of metal objects, both in Roman and ‘barbarian’ craftsmanship, was forging. Dies were used to create ornaments of repetitive shapes. They were usually two-piece sets (Fig. 9), with a top and bottom swage. The technique involves placing a heated rod or strip between the parts of a die and forging while shifting it until a suitable profile is obtained. Dies were basic elements of a blacksmith’s shop (Fig. 10, 11); in goldsmith’s workshops, a simplified version consisting of only the bottom swage was used. The technological properties of the alloys required the ‘cold’ forging method, during which the material changed to a fine-crystalline structure and hardened. The workpiece was occasionally soaked to recrystallise and plasticise the alloy. The use of this technology in barbarian metalwork is confirmed by the find of an anvil with ‘nail headers’ from Vimose on the island of Funen (DK), with a negative impression of a profile for forging on its underside (Fig. 12, 13). The bows of the Weklice bracelets were also forged in the manner described. Based on precise measurements, it can even be assumed that almost identical forming swages, with a negative impression of the design of approx. 10.5 mm in width, were used. Slight differences in shape may result from the finishing treatment of an already forged bracelet (Fig. 14). Creation of a shield-headed bracelet was time-consuming work, requiring a lot of knowledge and skill. First, a silver bar was cast, which was then forged into a long strip. Forging a semi-finished product in a swage required the involvement of two people and excellent work organisation. The use of a metal stamp, shaped in the outline of the profile on the swage, made it possible to obtain a deep relief (Fig. 15). Observation of the undersides of bases and heads of snake bracelets indicates that they were formed slightly differently. The underside of the heads shows traces of irregular impacts (Fig. 16:1–3), which indicates that these parts were made using the free forging technique. Such a bracelet creation process was applied in the reconstruction presented here, with the body forged on a swage, and the heads hammered on a wooden and lead pad (Fig. 17, 18). Forged heads of the original Weklice bracelets are irregular in shape, and even the subsequent application of engraved and punched ornaments on the face did not fully mask this asymmetry. Free forging and die forging were the initial techniques that made it possible create a certain section of a decoration. Bracelets forged in this manner have uneven face surfaces. The next step was to even and refine the body by smoothing and grinding, first with a file and then with grindstones. To smooth the surface of ornaments made of soft alloys, a flat iron burin or a small chisel with a wide, hardened blade could also be used. Traces of such treatments in the form of scratched, parallel lines are visible on the analysed examples of Weklice bracelets. The edge of a polygonal file was used to divide the heads and collars and make grooves accentuating raised ridges (Fig. 19:1.2). An ornament in the form of two main motifs made with punches, i.e. incised lines imitating a twisted or beaded wire and an alternately stippled snake-zigzag (Fig. 19, 20), was later applied on the face surfaces of the bracelets. During these operations, washers were used to prevent damage to the thin sheet metal. A tool with flat blade, a type of small chisel, was commonly used (Fig. 19:3.4). Chasers with a curved undercut in the blade and pronounced, lateral teeth, which gave a clear semi-circular imprint, were rarely used. Usually, such a punch would leave a distinct mark of fangs on the sides (Fig. 19:5). Oblique, parallel lines imitating twisted wires were made with similar punches in imitation of beaded wires. In the case of the former, a better effect was achieved using a chisel with a semi-circular notch in the blade and thickened teeth on the sides. The stamped pattern had the shape of an oblique, slightly S-shaped line (Fig. 19:6). Another variant of this ornamentation consisted of incised ridges separated with an undecorated band (Fig. 19:7). The decorative snake (zigzag) motif was made by punching regular points on alternate sides of a raised ridge (Fig. 19:8.9). The final step was polishing, giving the decoration a shine. In ancient times, gold and silver jewellery was commonly polished with semi-precious stones. Polishers made of iron were also used, providing decorations made of silver, gold and even tin alloys with a perfect shine (Fig. 21). Another method of finishing ornaments was patination. In antiquity, blackening of silver products was fashionable and was probably also used by barbarian communities. In the case of the described shield-headed bracelets with flatly displayed patterns, it was even advisable to leave the blackened depressions in the stamped ornaments, as it intensified – against the background of the polished smooth surface – the impression of the ornament’s three-dimensionality (Fig. 22). The appearance of shield-headed bracelets in the Wielbark Culture was undoubtedly the effect of contacts between the local communities and the Roman Empire. The result of these contacts was a huge transfer of technical knowledge, crafting skills and aesthetic concepts, among others. The ancient, naturalistic snake motif, fashionable and common in the 1st and 2nd century CE, was adapted and stylistically transformed into its own ‘barbarian’ design. This phenomenon intensified in the second half of the 2nd century and the early 3rd century. The bracelets from Weklice described here were probably made in a local blacksmith/goldsmith workshop to the order of elites living in the settlement clusters of the Wielbark Culture, which stretched around the shores of the then bay of the Vistula Lagoon. These workshops based their manufacturing on their own technological tradition, preferring blacksmithing techniques, including the use of dies with elaborate profiles. This phenomenon can be observed not only in the metalwork of the Wielbark Culture, but also in other Germanic societies living in the south-western regions of the Baltic Sea coast.