Marcin Woźniak

This paper discusses the use of “legacy data” in research on Roman Period iron smelting in the territory of the Przeworsk Culture in Magna Germania. The dataset includes results of 240 analyses of smelting slag and iron ores chemistry. A majority of these analyses were conducted in the 1950s-1980s. The quality of these data is far below present-day standards. Only major elements were reported, analytical methods were often not specified (although optical emission spectroscopy and wet chemical analyses can be assumed in such cases) and information on detection limits, precision and accuracy of the results is missing. In spite of this, a Principal Component Analysis-Agglomerative Hierarchical Clustering treatment successfully isolated observations from the three main iron smelting regions of the Przeworsk Culture (the Holy Cross Mountains, Masovia and Silesia). These results to a degree confirm a theory proposed in the 1960s by J. Piaskowski, according to whom it was possible to distinguish iron produced in the Holy Cross Mountains from the iron produced elsewhere in the territory of what is now Poland on the basis of metal characteristics. These findings will pave the way to the ongoing research project on the Przeworsk Culture metallurgy. It is also argued that, apart from a search for new methods in iron provenance studies, more attention should be paid to results of earlier analyses and to a compilation of legacy databases. The other result is an open and flexible Agglomerative Hierarchical Clustering R code to examine discrimination between production areas and to propose artefact provenance patterns in a convenient interactive way using the R development environment.

The paper analyzes burial 3 from the Łajski burial ground located in Eastern Mazovia in central-eastern Poland (Fig. 1). It is attributed to the Przeworsk culture and dated to phase .2/.1 of the Roman period, i. e. around the second half of the 2nd century. The dress of the buried woman includes an element alien to this cultural environment. It is a necklace consisting of glass beads, bronze spirals and a lunula decorated with red and green enamel (Fig. 2: 5–8). Its composition and the reconstructed layout design are typical for the style of the so called barbarian enamels (Fig. 3; 4). The necklace points to a link (or origin) of its wearer with a broadly interpreted community of Baltic cultures in the Eastern European forest zone. Most likely, the woman came from Mazovia inhabited by the Balts of the Bogaczow culture or adjacent regions of Eastern Podlasie and the Belorussian Bug River basin occupied by the population of the late Zarubintsy horizon. The necklace has an imported element, i.e. a chain made from chain mail rings (Fig. 2: 3). Along with distinctive features of the funerary rite and the composition of the funerary offerings, this element indicates that the necklace owner was fully integrated and adapted to the Przeworsk community, i.e. the eastern Germanic peoples identified with the historical union of the Vandalic populations.

In January 2014, a bronze crossbow tendril brooch with sheet catch Almgren 172 (Fig. 1), uncovered accidentally near Wąsosz, a village in Grajewo district in the northern part of Wysoczyzna Kolneńska (Kolno Hills). In the Roman Period the territory of Wysoczyzna Kolneńska and the area along lower Biebrza River probably constituted the northern boundaries of the Przeworsk and Wielbark cultures. About 10 km to the north-west, and the same distance away to the north-east, traces of settlements from the Roman Period were discovered, assigned rather to Bogaczewska culture, or generally to the West Balt circle. The nearest research sites which can potentially be linked to Wielbark culture (Drozdowo, possibly Sieburczyn) are located about 30 km south of Wąsosz (Fig. 2), hence the attribution of the brooch from Wąsosz cannot be determined with accuracy. It is therefore worth investigating this fibula through other, similar finds, both from the Wielbark area, and from that of Balt milieu.
Due to the large variety of Almgren 172 fibulae, an attempt was made to divide them in several groups. Of 186 brooches from the territories of Wielbark culture, Masłomęcz group and West Baltic cultural circle, it was possible to conduct a more thorough stylistic analysis of 155 items. The purpose of the proposed classification is to introduce some order to facilitate studies on this type of finds. The main criterion for this classification is the shape of the bow, with ornamental elements taken into consideration as secondary factors.
Group I (Fig. 4) includes slim fibulae with unprofiled bows, semi-circular or triangular in cross section. Group II (Fig. 5) make also slim fibulae but with profiled bows and feet, i.e. partly facetted edges, with rectangular metope areas at the end of the bow and on the foot below the sheet catch. The metopes are usually not decorated; only sometimes cuts appear on the sides (Fig. 5:5), in the shape of ‘X’ or carved lines (Fig. 5:4, 11, 15). Group III includes items similar to pattern ZG-28 according to A. Kokowski, which means specimens that are chunky, with wide bows and feet, facetted, and with rectangular metope areas. They are mostly cut on the sides, or decorated with a minted ‘X’. Decorations in the form of parallel engraved lines also appear (Fig. 7:15; 9:11; 10:1), as well as chequered patterns (Fig. 7:10, 11), minted dots, lines or bows (Fig. 7:12, 14; 8:4, 6; 9:8), herringbone (Fig. 8:11) or concentric circles (Fig. 9:9). The vast majority of fibulae from this group have round, elongated or shaped lumps at the ends of the springs. Within this group two types have been identified: IIIA – with no lump on the head (Fig. 7, 8) and IIIB – with a lump on the head (Fig. 9, 10). Group IV is represented by fibulae whose bows are made from a tape (Fig. 11). Some of them have ornamental metopes at the ends of bows and feet, separated by a carved ornament or lightly facetted edges. These brooches can be also distinguished into two types: IVA – with no lump on the head (Fig. 11:1–8) and IVB – with a lump on the head (Fig. 11:9–15). Among type IVA, fibulae known only from Lithuania stand out, decidedly larger from other specimens, with long, narrow and multiply coiled springs – distincted into Lithuanian variant (Fig. 12). The bows of these fibulae are made from a wide, rectangular tape; some of them have facetted edges (the bows’ cross section is slightly trapezoid-shaped). (Fig. 12:1, 2). Some of the fibulae with sheet catch have a false spring, made from a separate piece of wire on both sides – described as A.168/172 variant (Fig. 6; 10:2–9).
No correlation between the shape of the sheet catches and the form of the bow has been determined; fibulae from all groups have sheet catches of different shapes, some of which are narrow, while others are wide, some decorated, and others plain.
From all the fibulae included in the proposed groups, the most numerous set are Group III fibulae, which constitute 49.4% of all registered specimens. Group I fibulae constitute 17.5%; Group II – 16.2%; and Group IV – 12.3% (Fig. 13:2). The percentages within those groups are different for the materials from the Wielbark culture and the West Balt circle, which is illustrated in diagrams no. 3 and 4 in the Figure 13. The differences have also been noted among fibulae decorated with the lump on the head as they are clearly more popular (in terms of numbers and percentages) in the Balt environment (excluding the fibulae from the area now part of Lithuania) (cf. Fig. 14). However, on the Wielbark territory specimens with more ornamented bows (except metopes) are more common.
The vast majority of the crossbow tendril brooches with sheet catch from both of those areas are made of bronze. 5 specimens discovered on the Wielbark area are made of silver; a few specimens of the Lithuanian variant are also made of this material. Iron specimens are found in the materials from the Dollkeim-Kovrovo, Bogaczewo and Sudovian cultures.
80 of the Almgren 172 fibulae discussed here come from the Balt enviromen, with the largest group found on the Bogaczewo culture territory (26 specimens) and neighbouring sites to the north-east, the so called Gołdap group of the Sudovian culture (25 specimens). Only a few were found in the Dollkiem/Kovrovo culture area, at the cemeteries located along the western shores of the Sambia Peninsula and lower Pregola River (15 specimens). One fibula was found at a burial ground of the Central Lithuania group in Marvelė on Niemen River. The remaining 13 specimens were found at several sites of the East Lithuania Barrow culture (see Fig. 15). Among the Balt finds most belong to Group III (28 specimens), with type B represented by 19 specimens. 9 specimens of Group I were also found, as well as 11 of Group II and 19 of Group IV (see Fig. 13:4). In the West Balt circle, crossbow tendril brooches with sheet catch could be dated from the end of phase C2 until the early stage of phase D. The oldest among those, i.e. specimens from Group I, were mostly found together with items from Late Roman Period. More accurately dated specimens of the largest groups III and IV can be dated to the early stage of phase D. Nonetheless, it appears that their occurrence cannot be stretched beyond the first phase of the Sambian horizon – they weren’t found accompanied by e.g. fibulae types Schönwarling/Skowarcz and Dollkeim/Kovrovo, or specimens with star-shaped foot.
In the Wielbark culture and the Masłomęcz group crossbow tendril brooches with sheet catch were discovered on 46 sites. They concentrate in places where stable settlement structures existed in phase C3–D, i.e. mainly in northern Masovia, southern Lublin province, and lower Vistula (see Fig. 15). Among 104 specimens, the majority are Group III fibulae (49 specimens), with a clear dominance of items without the lump on the head (IIIA – 29 specimens). 17 specimens belong to Group I, 13 to Group II, and 6 – to Group IV (see Fig. 13:3). The oldest is probably the fibula from grave XXXIV in Kozłówko, which can be dated to the late stage of phase C2. The chronology of the majority of Almgren 172 fibulae in the Wielbark culture and the Masłomęcz group may be defined as the phase C3–D. Tree specimens are most probably dated to the phase D, what excludes this type of brooches from being an indicator of phase C3 in the Wielbark culture.
The above attempt to provide an overview of Almgren 172 fibulae from the territories of Wielbark culture and the West Balt circle shows that they constitute a relatively diverse collection, both in terms of morphology and chronology. The oldest group comprises specimens with slim and unprofiled bows, frequently with a false spring, which could have appeared in the late stage of phase C2, or at the turn of phases C2 and C3 (C3–D). The chronological situation of the fibulae from Group II is less clear. The few well-preserved sites indicate that they are younger than the fibulae from group I, and their chronology is closer to the fibulae from Group III. The fibulae from Group III, as well as similar specimens from Group IV, are dated to phase C3–D, with the possibility to the extension of their chronology up to the developed stage of the Early Migration Period.
The crossbow tendril brooch with sheet catch from Wąsosz belongs into group IVA. Based on the conclusions reached above, one can date it possibly to C3–D, and most probably even the beginning of the Early Migration Period. It is still difficult to establish its cultural attribution. In terms of morphology and location, the fibula is more similar to specimens found in the Balt enviroment rather than those from the Wielbark territory. A specimen with a nearly identical bow and spring was found at a West Baltic burial ground in Podliszewo, ca. 35 km north-east of Wąsosz. Location-wise, the closest fibula of type A.172, which can be associated with the Wielbark culture, was found near Turosnia Kościelna in Podlasie, ca. 75 km south of Wąsosz. It is possible that the fibula discussed here provides evidence of the movements of Masuria inhabitants south, to territories which were either completely abandoned or largely depopulated. The proposed interpretation of the fibula from Wąsosz is a hypothesis which requires verification through excavations and research.

The article analyzes an intaglio made of blue glass paste and set in a small, bronze ring. Found during an archaeological survey in Kraśnicza Wola, near the district of Grodzisk Mazowiecki, Masovian Voivodeship, Poland, the object is now kept at the Museum of Ancient Masovian Metallurgy in Pruszków. It presents the bust of a woman to the left with long hair and dressed in a robe. In front of her, the letters CA•VX• are visible. There are two more letters behind the head, but only one is clearly visible: R(?)•I. The intaglio used to be considered an ancient work of art. However, after closer examination, it turns out that the gem is the work of the famous German engraver (Johann) Christoph Dorsch (1676–1732). The paper includes a discussion on the dubious antiquity of the object and the identification of the person portrayed. The ring may testify to an aristocratic presence in Kraśnicza Wola much earlier than was previously thought. However, it is more probable that the ring was simply accidentally lost.

The vast cemetery of the Przeworsk Culture at Kleszewo yielded over 100 graves dated to the Late Pre-Roman Period and ca. 150 graves dated to the Early Roman Period. What is striking is the  fact  that  only  four  burials  were  furnished
with weapons or other warrior gear (a sword’s belt elements). One of them contained only a single shield rivet and a sword’s scabbard fittings. None of the graves contained typical men’s tools, personal  items  or  attire,  while  findings  commonly associated with women are very frequent.
Moreover, all four graves with military objects are dated chiefly to phase A 2 (or A2/A3), thus representing the opening phase of the cemetery, exclusively.  In  this  respect  two  phenomena
concerning  Przeworsk  Culture  burial  rites  are subject  to  analysis:  the  presence  of  cemeteries lacking  weapon  graves  and  the  practice  of  furnishing graves with only ‘fragmentary’ military offerings  (sometimes  interpreted  as  the  pars pro  toto  principle).  The  first  one  is  represented mostly in what we distinguish as ‘children’ and ‘female oriented’ cemeteries (as regards the social roles of buried individuals rather than their biological  sex  or  age).  We  suggest  that  grave  site at Kleszewo became ‘female oriented’ in phase A3. As for the second phenomenon, it is hard to interpret; however, links with the exact pars pro toto rule are doubtful. Some of the graves of this kind  also  contained,  apart  from  military  items (esp.  shield  rivets),  typically  female  objects  or possible women’s remains, according to anthropological data.
The  appearance  of  both  phenomena  in  the Przeworsk  Culture  cannot  be  easily  explained. They  could  be  somehow  related  to  older  traditions  of  the  Early  Iron  Age.  However,  direct relations with Germanic societies inhabiting the Elbe basin are also to be taken into consideration (esp.  concerning  ‘gender  oriented’  cemeteries), as  many  other  finds  discussed  in  the  text  indicate  similar  connections.  After  all,  most  probably these were two-way influences, as the practice of furnishing graves with only ‘fragmentary’ military offerings, esp. shield fittings, seems to appear in the Przeworsk Culture milieu earlier than in grave sites located in Central Germany.