The document discusses the two main types of shields used by the Przeworsk culture in the late Pre-Roman period: 1) shields modeled after Celtic designs with vertical ribs and band-shaped umbones, and 2) flat sheet shields with circular umbones. It explores the origins and influences on these shield designs, including potential influences from Illyrian, Samnite, Greek, and local Central European traditions. Specifically, shields with ribs are thought to have Celtic influences, while circular umbones had a more widespread distribution in Europe during the Hallstatt period and may have had multiple, concurrent origins rather than direct cultural contact influences.
The document discusses the two main types of shields used by the Przeworsk culture in the late Pre-Roman period: 1) shields modeled after Celtic designs with vertical ribs and band-shaped umbones, and 2) flat sheet shields with circular umbones. It explores the origins and influences on these shield designs, including potential influences from Illyrian, Samnite, Greek, and local Central European traditions. Specifically, shields with ribs are thought to have Celtic influences, while circular umbones had a more widespread distribution in Europe during the Hallstatt period and may have had multiple, concurrent origins rather than direct cultural contact influences.
The document discusses the two main types of shields used by the Przeworsk culture in the late Pre-Roman period: 1) shields modeled after Celtic designs with vertical ribs and band-shaped umbones, and 2) flat sheet shields with circular umbones. It explores the origins and influences on these shield designs, including potential influences from Illyrian, Samnite, Greek, and local Central European traditions. Specifically, shields with ribs are thought to have Celtic influences, while circular umbones had a more widespread distribution in Europe during the Hallstatt period and may have had multiple, concurrent origins rather than direct cultural contact influences.
The document discusses the two main types of shields used by the Przeworsk culture in the late Pre-Roman period: 1) shields modeled after Celtic designs with vertical ribs and band-shaped umbones, and 2) flat sheet shields with circular umbones. It explores the origins and influences on these shield designs, including potential influences from Illyrian, Samnite, Greek, and local Central European traditions. Specifically, shields with ribs are thought to have Celtic influences, while circular umbones had a more widespread distribution in Europe during the Hallstatt period and may have had multiple, concurrent origins rather than direct cultural contact influences.
The text discusses the two basic types of shields used by the Przeworsk culture and the diversity and origins of early circular umbones. It also discusses Celtic influences on weaponry from groups like the Illyrians and Samnites.
The two basic types of shields are those modeled after Celtic weapons with a vertical rib and band-shaped umbones, and flat sheet shields with circular or rectangular edges and umbones with closed circular calottes.
The text suggests that contact with Illyrian tribes and peoples of the Apennine Peninsula, especially the Samnites, likely played a significant role in shaping Celtic defensive armor. Large Samnite shields may have been adopted by Celtic tribes and influenced the development of Roman legion shields.
A NA L E C TA A R C HA E OL OGI C A R E S S OV I E NS I A
TOM 1 RZESZW 2006
Tomasz Bochnak Early Circular Umbones of the Przeworsk Culture. The Role of Local Tradition and Celtic Inuences on the Diversity of Metal Parts of Shields at the Beginning of the Late Pre-Roman Period In the late Pre-Roman period, the people of the Przeworsk culture used two basic types of shields with metal ferrules 1 . Te frst type includes shields modelled upon Celtic weapons, reinforced with a vertical rib and ferruled with band-shaped umbones (with open calottes), whereas the second type includes fat-sheet shields, ferruled with umbones with closed, circular calottes and with circular or rectangular edges. Te described diversity is, in the opinion of the author of this paper, a fundamental criterion for typological analysis of shield ferrules. Tis paper is devoted to the genesis and diversity of older umbones with closed calottes; however, it is also necessary to present the basic statements referring to band-shaped umbones. Genesis of Shields with Ribs Te emergence of shields with ribs in the environment of the Prze- worsk culture is commonly believed to be of Celtic infuence. A large shield reinforced by a rib ferruled with a metal band-shaped umbo ap- pears in the Celtic environment approximately during the LT B phase (Hafner 1976, 27, 110; Domaradzki 1977, 62; Brunaux, Lambot 1987, 130), even though older presentations of similar pieces of armament are known from the Villanova culture (Stary 1981, 292, 303, 304). It is believed that the Celts could have adopted the manner of using large shields from the Mediterranean peoples (Brunaux, Lambot 1987, 97). A signifcant role in the shaping of the defensive armament of the Celts was probably played by contact with Illyrian tribes and the peoples in- 1 Te history of studies on shield ferrules used in Europe (with parti- cular attention given to Polish lands) in the period of Celtic domination was presented in a separate paper (Bochnak 2005, 103106); therefore, this paper omits this issue and concentrates on the subject determined in the title. Te paper referred to above also discusses fnds of band-shaped umbones from the area of the Przeworsk culture, which are lef out of this article. 162 | Tomasz Bochnak habiting the Apennine Peninsula, in particular the Samnites. On the basis of archaeological data and accounts of ancient authors, it is be- lieved that the large Samnite shields could have been borrowed by Cel- tic tribes, and most probably this was also a beginning of the shields used by Roman legions (Stary 1979, 104, 105; 1981, 297). Te sheet of Celtic shields remained fat, whereas Roman shields were convex. Tis diference is difcult to explain. A. Rapin supposed that a convex shield could have protected better from thrusts, whereas a fat one could have provided a more efcient protection from slashes. In his opinion, fat shields were ofen used in the antiquity by the cavalry (Brunaux, Rapin 1988, 15). It seems that apart from Illyrian and Samnite impacts, the in- fuence of Greek civilisation on the Hallstatt culture also had a certain share in the development of Celtic weaponry. Proof of this is provided by a scene presented on a sheath from Hallstatt, dated from the 5 th cen- tury B.C. Te warriors presented there are equipped with shields, hel- mets and sometimes with shin guards, which could indicate the impact of hoplite armament. An element typical of Greek civilisation is also a curve shaped decoration adorning the shields. It is very signifcant that the Celts did not take over from the Greeks the double-point set of grips (hoplon) 2 , which allowed for signifcant freedom of movement. Te hoplites had shields with two grips, one of which was attached to the forearm, and the second was held by the hand. In this manner, the shield was supported on the forearm, and if it was necessary, the lef hand could be used to a limited extent; i.e., to hold a javelin. In the case of the Celtic shield, its weight made any prolonged use tiring, and the lef hand could not fulfll any additional functions (Picture No. 1). Terefore, it is also necessary to notice that the use of shields with a one-point set of grips was a signifcant hindrance during horse-rid- ing. On the other hand, the Celtic shield made it possible to protect oneself by bending the outstretched hand, which defected thrusts. Iconographic sources (including representations on the Hallstatt sheath and on the ferrule of a belt from Vele Ledine) and numerous fnds of Roman art (Biekowski 1928) show that large, oval shields were used both by infantrymen and by cavalry. Tere are also a few exam- ples confrming the use of shields by warriors fghting from carts. One such proof is a coin showing Lucius Hostilius Saserna I (the so-called Saserna I type), whose reverse presents a Celtic warrior equipped with a shield standing on a cart (Ostrowski 1999, 156, picture No. 92b). Most 2 Presentations of Celtic warriors holding a hoplon-stylised shield are common in Romantic art, yet incorrect from an archaeological point of view (Brunaux 1986, 60, 106). Early Circular Umbones of the Przeworsk Culture. The Role of Local Tradition and Celtic Infuences... | 163 Picture No. 1. Diferences in using a single and double point set of grips, based on a model of a Celtic shield provided with a rib (1-3) and a Greek shield (4-6): 1 de- fecting a thrust by means of bending the outstreched arm; 2 pushing the shield further out increased the feld of protection; 3 movement of the wrist allowed for relatively quick maneuvers of the shield; 4 shield movements are within one level; 5, 6 bending the forearm makes quick attack possible. (according to Brunaux; Ra- pin 1988, 20; Picture No. 10). 164 | Tomasz Bochnak probably, various manners of fghting infuenced the diversity of struc- tures and sizes of shields, yet it seems that until the eve of the La Tne period, there was no special equipment intended exclusively for infan- trymen or cavalrymen 3 (Brunaux, Lambot 1987, 97). Genesis of Shields with Circular Umbones Shields with circular umbones in the central section were commonly used in Europe in the Hallstatt period and at the beginning of the La Tne period. Tis is testifed to by iconographic and archaeological data from the area of the British Isles, the Iberian Peninsula 4 , the Apennine Peninsula and present-day Austria (Couissin 1926, 144, 145; Stary 1981, 293, 294, 298, 300, 301, Abb. 1; 1981a, 164, Taf. 1; Lorrio 1997, 192194; Birkhan 1999, 1120). Some of the discussed shields were round, others were oval; for example, shields presented on a situla currently stored in Providence (Rhode Island, USA). Statues of warriors from Este (Ita-ly) dated from the Hallstatt C (Ha C) period and fgures of carts from Strettweg (Austria) also have oval shields with circular umbones (Stary 1981, 293, 304, Egg 1996). Similar shields were also used on Polish lands, which is confrmed by presentations on ceramic dishes of the peoples inhabiting these areas (Fogel 1982, 102). A vessel of the Lusatian culture from Sobiejuchy (grave No. 45 Kujawsko-pomorskie province, Poland), dated from the end of the Ha C period or the beginning of the Ha D period, bears engravings interpreted as presentations of metal shields of the Ognica type or leather shields (similar in appearance) (Gediga 1970, 112, Picture No. 35, Bukowski 1971, 177180, Picture No. 1; Sylwestrowicz 1979, 28). Similar drawings are also present on face urns of the Pomerania culture; e.g., a drawing of a shield with concentric circles visible on an urn from Strzelno (or Samostrzel) in the Kujawsko-pomorskie province (Bukowski 1971, 183, 184, Picture No. 5.5). Elements of f- gurative presentations on urns from Gdynia-Redowo, Glicz Nowy, Grabw Bobowski and Lubieszw Tczewski are considered to be ima- ges of shields (all these places are located in the Pomorskie province) 3 Tere are premises which allow us to assume that in the Roman army, various types of shields were quickly diferentiated. Tis phenomenon may be confrmed by Titus Livius, who mentions that Titus Manlius, preparing for a fght, took up an infantry shield (History, VII, 10). 4 Te tradition of using round shields by the peoples inhabiting the Iberian Peninsula is mentioned by Polibius (History III, 114) and Diodor of Sicily (Historical Library, V, 33). Early Circular Umbones of the Przeworsk Culture. The Role of Local Tradition and Celtic Infuences... | 165 (La Baume 1920, 1934, 1935, 1950, 159; uka, 1971, 47, Picture No. 14; Bukowski 1971; Sylwestrowicz 1979, 2931, Picture No. 23, 26; Fogel 1982, 102). It is necessary to underscore the fact that distribution of shields with circular umbones in the Hallstatt period was probably not a result of inter-cultural contacts, but constituted a typical example of the concurrent development of the form. Te genesis of metal circular umbones which appeared in Europe in the period of Celtic domination has not yet been clarifed (uczkiewicz 2004, 5). Some researchers consider them a local product of peoples inhabiting the Central European Barbaricum (Jahn 1916, 165, 166; Schnberger 1952, 41); however, in the opinion of others, umbones of this type were developed as a result of the evolution of the La Tne band forms, and the fnds of the Przeworsk culture are an imitation of Cel- tic specimens (Klindt-Jensen 1950, 4648; Woniak 1970, 160; Doma- radzki 1977, 70, 71; Brunaux, Rapin 1988, 66, 67, Picture No. 33; Rapin 1983/1984, 78). Te advocates of the evolutionary theory of the genesis of umbones with closed calottes point to the existence of a sequence of band-shaped forms in archaeological sources, in which the calotte is gradually closed, which, in their opinion, led to the emergence of circular umbones (Picture No. 2). Important intermediate links in this evolutionary sequence are umbones of the Bartodzieje type and forms si- milar to the specimen from grave No. 92 from the Belgrade-Karaburma necropolis (Brunaux, Rapin 1988, 66, 67, Picture No. 33). However, this theory is burdened with two serious inadvertences. Whilst analysing the chronology of forms constituting individual elements of the discussed sequence, it turns out that the discussed concept is not corroborated by correctly dated specimens from the area of the Przeworsk culture and the Celtic circle. Studies on the transformations of shield ferrule forms in the Celtic environment are signifcantly hindered due to the changes in burial ritual, which took place from the end of the latter section of the middle La Tne period (LT C1) and which led to the disappearance of burial forms discernible by archaeological methods (Domaradzki 1977, 53, 54). A small number of shield ferrules from the LT C2 and LT D phases in comparison with the previous periods makes it difcult to carry out a typological and chronological analysis. Studies on the constructional transformations have to be based on insufcient fnds from the area of the Balkan Peninsula, northern Italy and Gaul. Among the younger umbones, dated from the LT C2 phase and deriving from a sanctuary from Gournay-sur-Aronde, there are no forms with clearly closed calottes, even though there are specimens with fan-spanned edges (Type VII, according to A. Rapin) (Brunaux, Rapin 1988, 82). Finds from burial grounds in Arqua di Petrarca and Clusone, located in 166 | Tomasz Bochnak Picture No. 2. Evolution outline of umbones from spindle-shaped to circular forms, synchronised with transformations in buckle forms and circular clasps with projec- tions. (according to Brunaux, Rapin 1988, Picture No. 33). northern Italy, confrm the occurrence of band-shaped umbones with closed calottes in the Celtic environment (Picture No. 2.B) only in the LT D1 and D2 phases (Gamba 1987, 250252, 265, Picture No. 11:2, 20:1, Brunaux, Rapin 1988, 66, Picture No. 33). Analogous umbones (the so called Arqua-Mokronog type) are similarly dated in the area Early Circular Umbones of the Przeworsk Culture. The Role of Local Tradition and Celtic Infuences... | 167 of present-day Slovenia (Gutin 1991, 57, 58, Abb. 30). A bronze specimen from Karaburma (grave No. 92) provided as an example of the subsequent stage in the evolution of band-shaped forms (Picture No. 2.C) is dated from the last few decades before Christ (Todorovi 1972, 77; Gutin 1984, 341). Te dating of specimens, which documents the evolutionary cha- racter of development of circular umbones, shows that there are no specimens older than the oldest forms with closed calottes deriving from the area of the Przeworsk culture. In the areas occupied by the Przeworsk culture, the frst umbones with circular edges emerge in the A1 phase of the early Pre-Roman period (~end of LT C1 and LT C2 phase). Combining the distribution of umbones of this type in the area of Central Europe with Celtic infuences is, accor-ding to the author of this article, dubious, because the specimens with hemispherical calottes are also found in the basin of the Elbe River and on Bornholm Island; i.e., areas where La Tne infuences are less visible than in the area oc- cupied by the settlements of the Przeworsk culture. Apart from the reservations presented above, which refer to the chronology of umbones and which are believed to form a sequence leading to the emergence of forms with closed calottes, the second in- advertence is the omission of important elements related with the pro- duction technology of umbones and the structure of the shield. As men- tioned earlier, there are two basic types of shields those reinforced by a rib (ofen ferruled) or those devoid of this element and the existence of intermediate forms is almost impossible 5 . Advocates of the theory on the emergence of round umbones from band-shaped forms ofen con- centrate on a seeming similarity of metal ferrules, omitting the entirely fundamental, in the authors opinion, issue of shield structure, which a given umbo was ferruling. An outline describing the evolution of band-shaped umbones suggests that circular umbones emerged as a re- sult of modifcation of band-shaped forms. Tese modifcations were gradually performed by crafsmen producing metal shield ferrules, and their consequence was the formation of full umbones with closed ca- lottes. However, it is necessary to point out that the production techno- logy of band-shaped umbones signifcantly difers from the manner of producing a specimen with a closed calotte. Tese diferences go back to the very beginning of processing iron, and the subsequent stages of forming band-shaped umbones cannot lead to circular umbones. Te emergence of subsequent forms with closed calottes is the result of de- veloping the process of forging a band-shaped umbo; yet in order to ob- 5 Te problem of hybrid structures and combining certain elements of both types of shields is presented in a further section of this paper. 168 | Tomasz Bochnak tain an umbo with a closed calotte, it is necessary to use diferent tech- niques, starting from the early stage of metal processing. A crafsman producing an umbo has to determine its type at the very start, and this will infuence his subsequent activities 6 . Where, therefore, should one look for the genesis of a circular umbo with a closed calotte? If it is assumed that the idea of an umbo derives from the tradition of the late Pre-Roman period, then it is possible to presume that its shape constitutes a certain reference to the forms of shield reinforcement used in that period, yet it is necessary to remem- ber the natural conditions related with attempts to imitate analogous forms using a diferent material. In other words, introduction of new technologies and materials is ofen linked with attempts at imitating older, well-known forms with the use of new resources, yet due to the natural properties of such resources and the processing possibilities re- lated with them, such a transfer is not always possible. Older fnds of shields (or their fragments) from the areas of Europe located beyond the sphere of Mediterranean culture are known from Kvrlv (Martens 2001, 135142) and Hjortspring (Rosenberg 1937; Randsborg 1995). Both the wooden umbo from Kvrlv, radiocarbon dated from 760410 B.C., and the specimens from Hjortspring, whose chronology was determined to be from the end of the 4 th / beginning of the 3 rd century B.C., have spindle-shape forms and slightly resemble the signifcantly reduced ribs (spinae) of La Tne shields. Tese umbones were attached to fat boards using wooden pins (Kvrlv) or resin (Hjortspring). If one was to look for prototypes of metal umbones among the above-mentioned Scandinavian fnds, it could be concluded that the older metal fnds should also be spindle-shaped. However, it is necessary to notice that the elongated shape of wooden umbones depends on the fbrous structure of wood. If an umbo was bored into wood, it was given a spindle shape in order to avoid the cracking of the more delicate parts; i.e., those which were shaped whilst cutting across wood rings. It is understandable that whilst forging an iron sheet, it is possible to omit these aspects, yet providing a metal umbo with a spindle shape, however doable, requires a set of proper, specialised tools. Terefore, it can be assumed that the older iron umbones were not an imitation of earlier wooden umbones. However, it cannot be ruled out that circular umbones were created as a result of adapting iron in place of other organic substances used before (e.g. leather). A shield hardened by means of wax and hot water excellently protects against 6 Te author would like to thank Mr. Wojciech Sawiski, a blacksmith from Grodzisk Mazowiecki, for technical explanations regarding the produc- tion process of band-shaped and circular umbones. Early Circular Umbones of the Przeworsk Culture. The Role of Local Tradition and Celtic Infuences... | 169 thrusts, which was proved during a series of experiments (Picture No. 3) (Ziemacki 1964, 133, 134; Coles 1977, 199206). A shield made entirely of leather dated from the Bronze Age is known from Clonbrin in Ireland (Coles 1962, 160; 1977, 204; Eogan 1964, Pl. XXVIII). Wooden items were discovered in the British Isles which are supposed to be forms of a last, used for giving a wooden shield its desired shape. Items of this type derive from Cloonlara, Co. Mayo, Churchfeld and Kilmahamogue, Co. Antrim in Ireland (Coles 1962, 160, 161, 181; Pl. XXVI; Eogan 1964, Pl. XXVIII). A shield made entirely of wood and leather also derives from Ireland, from Clonoura (Stead 1985, 45, Pl. XI; Brunaux, Rapin 1988, 19, 21, Picture No. 8; Rafery 1983, 107). Due to natural factors, fnds of this type are very rare, and their absence in the area of Poland does not mean that similar shields were not used here (Fogel 1982, 102). Te existence of similar traditions in the Polish lands could be indicated by a few fnds of the Ognica/Herzsprung Picture No. 3. Circular shields from Bronze Age and early Iron Age. 1 bronze shield from Ognica, V BA (according to Bukowski 1971, Picture No. 3); 2 modern replica of a leather shield (according to Coles 1977, Photo No. 16); 3 presentation of a shield on a face urn from Strzelno (or Samostrzel) (according to Bukowski 1971, Picture No. 5.5); 4 decoration detail of a bronze ferrule of a horn from Wismar (Germany), V BA (according to Bukowski 1971, Picture No. 4). 170 | Tomasz Bochnak type shields, specimens of which were found in West Pomerania, in Holstein and in southern Germany (Bukowski 1971, 81). Te manner of producing shields from thin bronze sheets makes it in principle impossible to use them efectively in a fght and seems to indicate their gala character, even though their shape corresponds closely to leather shields. In the central section of the Ognica/Herzsprung type shields, similarly to the leather specimens from the British Isles, a hemispherical protrusion was marked, under which a grip was placed. It is necessary to notice that the hemispherical shape of this umbo constitutes the only form with numerous symmetry axes which was possible to obtain. Te protrusion has a shape similar to a low cupola, and the transfer of this shape onto a metal object is relatively easy and does not require specialised tools. Te manner of processing leather and a sheet of metal is of course fundamentally diferent, yet these materials are similar to each other as far as the possibilities of shaping are concerned. Tis similarity could have been a cause for which a sheet of iron was processed in a manner emulating leather umbones and not wooden ones. Older Circular Umbones in the Przeworsk Culture Low hemispherical umbones type 1 according to D. Bohnsack (type 1 according to the recently proposed typology of T. Bochnak) constitute the oldest form of circular ferrules of shields in Polish lands, and they are the form which began the development cycle of umbones used by the people of the Przeworsk culture during the late Pre-Roman period and the period of Roman infuences (Picture No. 4). Tey are considered one of the determinants of phase A1 in the Przeworsk culture (Hachmann 1956/57, map No. 3; 1961, 52, 56, 57, 77, 240, 241, Abb. 18; Peschel 1977, 270, 271, Abb. 3; Dbrowska 1988a, 28, 56, Breakdown No. 13). So far, no fnds of umbones with hemispherical calottes are known in the context which would allow for the determination of their chronology at the very beginning of phase A1, which is connected with the occurrence of long brooches that are devoid of ornamentation (type A according to J. Kostrzewski) and long, ornamented fbulas type B (Dbrowska 1988a, 27, 28). Such umbones occur in complexes containing specimens which are younger from the determinants of the beginning of phase A1 of the late Pre-Roman period. A complex from Gobiewo (grave No. 4 7 ) could provide a good 7 In a paper devoted to the armament of the peoples of the Przeworsk cul- ture in the Polish lands during the Early Pre-Roman period, wrongly deter- mined as Gobiewo, grave No. 1 (Bochnak 2005, 228). Early Circular Umbones of the Przeworsk Culture. The Role of Local Tradition and Celtic Infuences... | 171 example here, where the hemispherical umbo was discovered with a type C fbula according to Kostrzewski (Picture No. 5) (Kostrzewski 1919, 258). It is not out of the question that the long C type brooch was also among the accessories found in grave No. 3 in Becze in the Lubuskie province (Dbrowska 1988a, Breakdown No. 13, Lewczuk 1997, 84). Type 1 umbones are also known from areas neighbouring the zone of closelyknit settlements of the Przeworsk culture. Specimens with hemispherical calottes are known from the areas occupied by people of the Oksywie culture, areas over the Elbe River and other cultural entities partially connected with German settlements; e.g., the Kobyly group (Bohnsack 1938, 56; Mhling 1944, 55, 99, Abb. 27; Hachmann 1956/57, Map No. 3; Wogiewiczowie 1963, 53, 54, Picture No. 8; Peschel 1977, 270, 271, 277279; Frey 1986, 46, 47, Abb. 1; Seidel 2000, 99; Bockius, uczkiewicz 2004, 77, 79, 189, 190, Abb. 12.3). An exceptional fnd of a type 1 umbo comes from Bornholm (Hachmann 1956/57, Map No. 3, 1961, 192; Frey 1986, 46, 47, Abb.1). Finds of various forms with Picture No. 4. Umbones type 1. 1: Lachmirowice 2 Bartodzieje (grave No.) (accor- ding to Bohnsack 3: Przyborowo (grave No.). 172 | Tomasz Bochnak hemispherical calottes in the lands neighbouring the area occupied by the settlements of the Przeworsk culture are considered to be proof for inter-cultural impacts and contacts with the zone of the Przeworsk culture (Hachmann 1957; Peschel 1977; Frey 1986, 46). Fragments of an umbo similar to type 1 come from the grave of the Poienesi-Lukaevka culture from Rctu in Romania (Babe 1993, 117, Table No. 44.10). It is necessary to stress that in the areas which were not inhabited by the people of the Przeworsk culture, type 1 umbones appear most ofen in chronological stages synchronised with the A2 phase (~LT D1) of the late Pre-Roman period. Typological and chronological analysis indicates that separation of diferent types of umbones (resembling the common model with a hemispherical calotte and narrow edge) took place very quickly afer the application of technological assumptions to the production technology of type 1 umbones. Tese are umbones with hemispherical calottes and rectangular edges (determined as the Bartodzieje type, Picture No. 5. Gobiewo (grave No. 4) (according to Gaerte 1929). Early Circular Umbones of the Przeworsk Culture. The Role of Local Tradition and Celtic Infuences... | 173 type 2 according to T. Bochnak typology), specimens with hemispherical calottes supported on a trimmed (type 3 according to T. Bochnak) or cylindrical collar (type 2 according to D. Bohnsack, type 4 according to T. Bochnak) and ferrules with hemispherical calottes and wide edges (type 10 according to D. Bohnsack, type 5 according to T. Bochnak). Te Bartodzieje type umbones deserve a more comprehensive description. Sometimes they are considered, as was noted above, an intermediate stage in the supposed evolution from the band-shaped forms to the circular forms (Adler 1993, 231). Te issues of their emergence and origin are widely discussed in modern literature (Bockius 1996; Kleemann 2002; Bockius, uczkiewicz 2004, 7679; uczkiewicz 2004, 131; Bochnak 2005, cf. previous literature). It is necessary to underscore the fact that the discussed ferrules are in fact umbones with closed calottes and rectangular edges or edges resembling a rectangle (Dbrowska 1988a, 55, 56), and due to this, they were sometimes considered a type 1 variant according to Bohnsack (Hachmann 1961, 53, 54, Abb. 13). Te rectangular edge could have acted as an additional reinforcement for the shield structure. Only two specimens of umbones of the described type are known, and both of them derive from the Polish lands from grave No. 7 in Bartodzieje in the Dolnolskie province (Jahn 1931, 56, 57, Abb. 48, 50; 1937; Pescheck 1939, 201, 202) and from Wymysowo in the Wielkopolskie province (grave No. 290) (Jasnosz 1951, 195, 195, Picture No. 295:3, 5) (Picture No. 6). the fragmentary status of both fnds signifcantly hinders reconstruction of their original shape. Tis remark refers to, frst of all, the specimen from Wymysowo, due to the fact that the picture presenting the umbo before conservation does not allow for an explicit determination whether its calotte was closed. It seems that the reconstruction of the umbo from Bartodzieje, published numerous times, is correct (inaccuracies may refer to the hypothetical shape of the edge, yet the presence of the closed calotte does not seem to be subject to any doubts, on the condition that its edge was correctly interpreted). M. Jahn considered the fnd from Bartodzieje as proof of Rhetic infuences, and his opinion was quoted in papers regarding cultural diversity on Silesian lands in the last ages before Christ (Jahn 1931, 5659; 1937, 108110; Pescheck 1939, 201, 202; Woniak 1970, 161). However, it is necessary to stress that M. Jahn formulated the thesis on the Rhetic reference of the specimen from Bartodzieje on the basis of the similarity of this fnd to an umbo from grave No. 100 from the Persona burial ground in Ornavasso (Picture No. 7.1) (Bianchetti 1895, Picture No. VII: 12; Graue 1974, 65, 66, Abb. 31:58). Te author of this paper is convinced that this analogy is easy to undermine, because it is limited to the similarity in horizontal 174 | Tomasz Bochnak projections of these fnds and does not take into account the most important feature, which is the type of shield ferruled by the umbo. Te fnd from Bartodzieje represents a group of umbones with closed calottes, and, without doubt, it ferruled a shield devoid of ribs; on the Picture No. 6. Umbones with rectangular edges. 1 Wymysowo (grave No. 290); 1a Wymysowo (grave No. 290) (reconstruction proposal) (according to Jasnosz 1951); 2 Bartodzieje (grave No. 7) one of the preserved umbo fragments; 2a Bartodzieje (grave No. 7) reconstruction proposal (according to Jahn 1931); 3 Jeszkowice (ac- cording to Jahn 1931). Early Circular Umbones of the Przeworsk Culture. The Role of Local Tradition and Celtic Infuences... | 175 other hand, the specimen from Ornavasso needs to be included in the group of so-called butterfy-shaped umbones 8 , which are classifed among shields with ribs (Picture No. 7.2). Moreover, the specimen from Ornavasso is much younger than the Silesian one. On the basis of dating a situla E.22 (variant Bargfeld) which derives from the same grave, J. Graue determined the age of grave No. 100 from the Persona burial ground as phase III, synchronised with the twilight of the late La Tne period (Graue 1974, 65). During the last few years, attempts have been made to incorpo- rate the Bartodzieje type umbones into a wider horizon of fnds and comprising materials from the area of Austria, Hungary, Slovenia and Northern Italy (Bockius 1996, 144150, Abb. 13, Bockius, uczkiewicz 2004, 7679). J. Kleemann attempted to polemicise with the hypothesis of R. Bockius, justly indicating the weak preservation status of the fnds which were supposed to be southern European analogies for the fnds of the Przeworsk culture. J. Kleemann thought that Bartodzieje type umbones should be treated as derivatives of hemispherical umbones (Kleemann 2002, 221223). Just as mentioned above, the author of this article believes that the concept according to which umbones with closed calottes and rectangu- lar edges are supposed to be proof of contacts with Southern Europe is 8 Te term butterfy-shaped umbo was used here in the context applied by P. Biekowski (Biekowski 1928, 139, 140) and not A. Rapin, who used the same name for band-shaped bosses with wings spreading in a fan-like manner (Brunaux, Rapin 1988, 82), similar to type II 2A, according to M. Domaradz- ki (Domaradzki 1977, 61) or the Skorba type, according to M. Gutin (1991, 5658, Abb. 29, 30). Picture No. 7. 1 butterfy-shape umbo from Ornavasso (grave No. 100) (according to Jahn 1937), not to scale; 2 presentation of a shield ferruled with such an umbo (according to Biekowski 1928), not to scale. 176 | Tomasz Bochnak based on very weak premises. Te majority of the seven points marked on the map in the paper of R. Bockius (1996, 151, Abb. 3), which are supposed to depict the spreading of Bartodzieje type umbones, mark specimens of various forms. Apart from marks denoting sites in Bar- todzieje and Wymysowo, according to R. Bockius, Bartodzieje type umbones were also found in Piotrkw Kujawski and Jeszkowice 9 . In re- ality, the umbo from Piotrkw Kujawski needs to be included among the band-shaped forms (Bochnak 2005, 106, 181, Table No. XXXI:1), whereas the umbo from Jeszkowice is known from a sketch and short references in literature (Jahn 1931, 58, 60; Abb. 60), yet very ofen it is wrongly connected with other specimens deriving from the same place, which confuse it with the cone-shaped umbo type J.5 (Jahn 1924, 35). It is necessary to underscore that it does not follow from the sketch in- cluded in the paper of M. Jahn that the fnd from Jeszkowice has a cir- cular, closed calotte (Picture No. 6.3). Te remaining umbones quoted by R. Bockius are specimens in very bad shape. Tere is no data al- lowing for a clear statement that the specimen from Este had a closed calotte, and in the opinion of the author of this paper, the preservation status of this fnd does make such a statement possible (Picture No. 8.2) (Bockius 1996, 148, 149, Abb. 2:2). Te umbones from Puszta Gc, Ngrd and Este counties are preserved in fragments, and on the basis of other well preserved specimens from Hungary and northern Italy, it is possible to propose diferent reconstructions of these fnds, cor- responding to band-shaped umbones of Mokronog-Arqua and Skor- ba types, which were common in this area (Picture No. 8.1, 3, 3a, 4, 9.3, 4) (Gutin 1991, 5658, Abb. 29, 30). Finds from Teurnia (Lippert 1992) and Wenns (Egg 2002), quoted in the paper of R. Bockius and P. uczkiewicz (2004, 7679), are specimens of band-shaped umbones with extended calottes (Picture No. 9.1, 2). Tese umbones have open calottes; therefore, they must have ferruled shields with ribs and not fat shields, as in the case of specimens from Bartodzieje (grave No. 7) and potentially from Wymysowo (grave No. 290). With respect to this diferentiation, the issue of similar arrangement of rivets on umbones from Wenns and Wymysowo, brought to attention by R. Bockius and P. uczkiewicz (Bockius, uczkiewicz 2004, 76), has marginal signif- 9 Comparing umbones from the area of Poland, R. Bockius made use of a map published by T. Dbrowska (Dbrowska 1988, 62; Abb. 8), where band- shaped umbones along with Bartodzieje forms are marked with one sign de- scribed as Bartodzieje type umbones. A description of the same sign on an analogous map in a monographic paper devoted to the early stages of the Prze- worsk culture (Dbrowska 1988a, 72, map 5) reads rectangular umbones, which is not an incorrect term. Early Circular Umbones of the Przeworsk Culture. The Role of Local Tradition and Celtic Infuences... | 177 Picture No. 8. Umbones indicated as analogies for Bartodzieje type umbones. 1 Puszta Gc; 2 Este; 3 Reka (grave No. 7) (13 according to Bockius 1996, Abb. 2); 3a Reka (grave No. 7) (according to M. Gutin 1991, Abb. 29), 4 Roje pri Moravach (according to Knez 1977; Table No. 8.3). 178 | Tomasz Bochnak Picture No. 9. Umbones indicated as analogies for Bartodzieje type umbones. 1 Enns (according to Bockis, uczkiewicz 2004, Abb. 12.1); 2 Teurnia (according to Lippert 1992, Abb. 3 KLM 1999); 3 Vevey (grave No. 26) (according to Brunaux, Ra- pin 1988), 4 Odak, (according to Todorovi 1965, Table No. 1); 5 Mihovo (accor- ding to Bockius, uczkiewicz 2004, Abb. 12.2). Early Circular Umbones of the Przeworsk Culture. The Role of Local Tradition and Celtic Infuences... | 179 cance in the opinion of the author of this article. Te fnds from Teur- nia and Wenns cannot play an important role in deliberations about the chronology of shield ferrules, because they do not derive from sacri- fcial sites the so-called Brandopferpltze used over a long period of time (cf. Egg, 2002; Gleirscher 2002, further literature). Te chrono- logy of the Brandopferpltze from Teurnia comprises the LT C2 LT D period (Lippert 1992, 294299). In principle, the only umbo outside the area of the Przeworsk cul- ture, which could constitute an analogy for the fnd from Bartodzie- je (and possibly from Wymysowo), is a specimen from Mihovo, deri- ving from a grave where fbula, A.236 10 was found (Picture No. 5) (Bockius, uczkiewicz 2004, 77, 78, Abb. 12:2). However, it is necessary to pay attention to the fact that the published drawing of this umbo is very schematic and the openings for rivets were not marked on it. It is also not known whether the drawing refects the actual status of preser- vation or whether it is a reconstruction. In fact, a horizontal projection of the umbo from Mihovo seems to suggest that the calotte was open (the convexity of the calotte is not circular), whereas the profle clearly indicates that it was open. In relation to the above, it is possible to state that in the light of the available archaeological sources, the umbo from Bartodzieje (grave No. 7) does not have any close analogies in the Celtic world and probably is a local product. It is quite possible that the umbo from Wymysowo (grave No. 290) is an analogy for the Silesian specimen, yet its status of preservation makes unambiguous determination impossible in the opinion of the author of this article. However, it cannot be ruled out that in the future the above determinations will have to be adjusted if similar umbones with closed calottes and rectangular edges are disco- vered in territories occupied by the Celtic culture 11 . It is possible that the signifcant sizes of the edge constituted an additional element reinforcing the structure of the shield board. In the 10 R. Bockius and P. uczkiewicz, referring to the Ph.D. dissertation (un- published) of H. Windl and devoted to the materials from Mihovo, empha- sise that the site was examined at the end of the 19 th / beginning of the 20 th
century and the contemporary methods of explorations do not give certainty as far as the homogeneity of the separated complexes is concerned (Bockius, uczkiewicz 2004, 78). 11 Te potential foreign provenance of the specimen from Wymysowo may be confrmed by bronze rivets which, according to S. Jasnosz, were sup- posed to attach the umbo to the shield (Jasnosz 1951, 194, 195, Picture 295.4). Te people of the Przeworsk culture used copper alloys in a very limited ex- tent, which allows one to presume that items made of bronze might have been 180 | Tomasz Bochnak opinion of the author of this paper, the appearance of the umbo from Bartodzieje indicates references to band-shaped La Tne umbones; yet these infuences did not lead to the use of shields with ribs, but were only refected in attempts of making an umbo with a closed calotte similar to Celtic specimens. Te discussed specimen (and possibly also an umbo from Wymysowo) most probably proves the reception and transformation of Celtic infuences according to the shield making technologies prevailing in the Przeworsk culture. Despite the fact that Bartodzieje type umbones constitute a form which is typologically younger than hemispherical umbones with round edges, they are considered one of the determinants of phase A1 of the late Pre-Roman period (Dbrowska 1988a, 56). Determining a more precise chronology of Bartodzieje umbones is problematic. In the opinion of T. Dbrowska, the discussed type of ferrules should be dated from the older section of phase A1, which can be synchronised with the end of LT C1 phase, even though this researcher allows for the determination of their origin from the phase corresponding to the LT C2 stage (Dbrowska 1988a, 56). In the opinion of the author of this paper, the second possibility is more plausible due to the fact that so far there is no date allowing for the da- ting of specimens with hemispherical calottes at the very beginning of A1 stage, and Bartodzieje type umbones probably constitute a form typologically younger than umbones with a circular edge. Tis dating is not contradicted by the presence of a long fbula type C discovered in grave No. 7 in Bartodzieje. Reception of Metal Shield Ferrules in the Cultures of Central Europe. Subject to La Tne Inuences and Celtic Culture As it has been emphasised a number of times, the impact of the La Tne culture has resulted in the taking over of various elements of Celtic armament by the people of the Przeworsk culture, and one of the basic elements connected with the Celtic impact is the popularisation of iron. Without doubt, double-edged swords with metal sheaths and spurs have to be included in the Celtic heritage. Te heads of pole arms used by the people of the Przeworsk culture are also considered a testimony to the impact of the La Tne culture, yet this remark refers mainly to heads whose edges were cut in the form of leaves. In the imported. On the other hand, the author of this paper is not familiar with ex- amples of attaching the iron umbo with bronze rivets from other cultural cir- cles from the period of Celtic domination. Early Circular Umbones of the Przeworsk Culture. The Role of Local Tradition and Celtic Infuences... | 181 armament of people inhabiting Polish lands during the late Pre-Roman period, there are categories of weapons produced locally. Popularisation of single-edge swords, typical for the Baltic Sea basin though ofen encountered in areas occupied by the Przeworsk culture, points to the development of local arms production which omits Celtic traditions. Te shields used by the people of the Przeworsk culture in the late Pre- Roman period probably derived from two separate traditions. On the basis of archaeological sources, it is possible to diferentiate a small group of shields reinforced by ribs ferruled with band-shaped umbones and a more numerous group of shields ferruled with circular umbones. Shields with band-shaped umbones, based on Celtic models, occur at the beginning of the Przeworsk culture, yet there is no basis on which to believe that in the Polish lands they are older than shields without ribs. It is difcult to point out reasons for such a limited reception of the Celtic shield model. Probably shields ferruled with circular umbones, which are from local tradition, were better adapted to the local manner of fghting than the bigger and heavier Celtic shields reinforced with a vertical rib. It is probable that the reduction of shield size is related with conditions of the natural environment and the dominant manner of fghting. A change in the conditions of the natural environment is, according to J. Harmand, a cause for a reduction of the size of the shield of the Roman soldier during the Gallic war (Harmand 1967, 6769). A large shield of a Roman legionary, presumably deriving from the Samnite scutum, fulflled its task very well during fghting in northern Italy and in the de-forested areas of Gaul. Te signifcant shield size made it possible to lock the shields together over the warriors heads 12
and also provided the necessary protection against projectiles (Caesar, Te Gallic War, II, 6; V, 3849; VII, 6890). However, when the legions entered the forested areas of Gallia Belgica 13 , their tactics had to be changed. Te Gauls, plaguing the Roman army, made use of land forms and forests to prepare ambushes, with respect to which defensive techniques of the legionaries working out in the open turned out to be inefective (Deyber 1987, 148). Smaller shields performed their tasks better on forest paths; they were easier to maneuver and more useful in combat. Such shields were most probably used by the Germanic tribe led by Ariovist, even though it is necessary to remember that this tribe also used a formation described by Caesar as a phalanx (creating a wall which protected the warriors) (Caesar, Te Gallic War, I, 52). 12 Devising this tactic is ascribed to the Celts, and it was described by Livy (History of Rome from its foundation, V, 43). 13 Caesar mentions forests in the area of Gaul numerous times (Caesar, Te Gallic War, II, 18, 19; III, 28, 29, 38; VI, 5). 182 | Tomasz Bochnak Picture No. 10. Shield elements combining characteristics of circular specimens and specimens with ribs. 1 Mehrum type (according to Kaczanowski 1995); 2 spindle- shaped umbo (according to Kaczanowski 1995); 3 shield image on an altar from Nmes (according to Dechelette 1914). Early Circular Umbones of the Przeworsk Culture. The Role of Local Tradition and Celtic Infuences... | 183 It is possible that the natural conditions of the Polish lands was an infuence on meagre popularity of large shields reinforced by ribs in the Przeworsk culture. Presumably, smaller shields, similar to those used in this area for centuries, turned out to more useful. Indirect proof for the common use of small shields may be the diversity of weapons. It is well known that the type of defence strictly depends on the type of weapons used by the enemy. If the two fghting sides belong to a single cultural circle, the type of arms used is quite established, because changes are introduced most ofen as a consequence of clashes with enemy fghting in a diferent manner or as a result of technological progress (e.g. the popularisation of iron). Among the militaria of the Przeworsk culture, there prevail weapons used in fghting in clashes, and archaeological sources allow for the conclusion that the role of projectiles was slight (though probably underestimated). Tis diversity of weapons is better matched by shields of smaller sizes, which prove useful in hand-to- hand combat, whereas more massive and larger forms are more efcient when the enemy uses projectile weapons (Fogel 1982, 103). Terefore, it can be surmised that during armed conficts between groups of the Przeworsk people, the Celtic manner of fghting with swords was mainly applied; nevertheless, single-edged swords were also used for thrus- ting and slashing. Due to the fact that the use of projectiles was limited, the basic method of defence could have been shields whose size was smaller than the Celtic shield. Tis type of defense constitutes a logical supplement for weapons used by the Przeworsk culture. In light of the above determinations, the comparison of sets of weapons of Roman legionaries and Celtic warriors (with respect to whom there are nume- rous accounts from ancient historians and iconographic presentations) is quite interesting. Weapons used by the Roman legions were compiled in a manner that made it difcult to choose appropriate shields provi- ding protection during a fght. In hand-to-hand combat with an enemy using large shields and fghting in a close formation, usually short wea- pons are suitable and can be used for precise thrusts over the shield (Hazell 1981, 78, 79). Te legionaries used hurled weapons in a fght, the best protection against which was provided by large shields; yet during hand-to-hand fghting, they used short gladius swords, which were much more efective than the long Celtic swords. It was also more difcult to shield oneself from their thrusts. Modifcations of the Roman armament resulting from fghting in the Iberian Peninsula and the borrowing of weapons from the Celtic and Iberian environment combined with the organisation and discipline of the legions made the Gallic war relatively short in comparison with previous Celtic Roman conficts, even though the Gauls made organised resistance attempts. 184 | Tomasz Bochnak Te chronology and dynamics of the distribution of umbones with closed calottes in the La Tne culture indicates that these ferrules caught on in the Celtic environment relatively slowly (Stead 1985, 37). Te frst specimens of this type appear in the area of the La Tne culture at the end of the LT C / beginning of the LT D phase; i.e., later than in the Przeworsk culture, where the older umbones with closed calottes are dated from phase A1 of the late Pre-Roman period and are synchronised with the period comprising the end of LT C1 and LT C2. Among the oldest fnds of this type are the isolated fnds from Mokronog and Roje pri Moravah (Domaradzki 1977, 62; Stead 1985, 37). In Domaradzkis opinion, the majority of circular umbones from the area of the Balkans related with the La Tne culture should be dated from the LT D phase (Domaradzki 1977, 62). Te umbones from graves Nos. 169 and 187 from Nove Mesto (Knez 1987, 105, Taf. 23; 1992, 60, 63, T. 61.2, 67.3) are dated from the later section of this phase. Te concept of a shield devoid of a rib was slowly adopted by the Celtic environment. In the area of Bavaria, individual umbones with circular edges appear in the course of the LT D1 phase (Krmer 1952, 334; 1962, Abb. 1) and during the D2 phase in the middle Rhine basin (Decker 1968, 51; Polenz 1971, 49; 50; Joachim 1974, 163165). Among the oldest specimens with closed calottes (deriving from Gaul) is the umbo with a short spike discussed above and found in Alise-Ste-Reine and historically dated from 52 B.C. Tis umbo, having its closest analogies in the areas of Denmark, Germany and Poland, is believed to be an import in the Celtic environment and is connected with members of Germanic tribes led by Ariovist who stayed in the territory of Gaul (Duval 1987, 62; Sievers 1995, 151, 152, Abb. 38.1; Sievers 2001, 139, 153, 154; Bochnak, in printing). Also, a fgure of a warrior or a god, discovered in the area of the Saint-Maur sanctuary (Oise department, France), with a shield with a circular umbo is believed to be a product created under the impact of Germanic infuences due to the lack of features typical of the Celtic culture (Brunaux 1986, 78). Circular umbones in the area of Gaul do not appear until the LT D phase (Schnfelder 2002, 5456). Introduction of shields devoid of ribs and ferruled with umbones with closed calottes did not lead to an immediate disappearance of typical Celtic forms with band-shaped umbones, which can be testifed to by a fnd from Heimbach-Weiser dated from the end of the 1 st century B.C., where specimens of both types were included among rich cremation burial accessories (Joachim 1973, 3334, 41, Abb. 11). It cannot be ruled out that the shields ferruled by them had varied sizes and were used in slightly diferent situations. One should also remember that the presented picture of diversity with respect to chronology and distribution Early Circular Umbones of the Przeworsk Culture. The Role of Local Tradition and Celtic Infuences... | 185 of circular umbones is signifcantly hindered by a sudden decrease in the number of Celtic materials dating from the 2 nd and the 1 st century B.C. and resulting from changes in the burial ritual of the La Tne culture. It is difcult to determine the character of infuences which led to the re-introduction of umbones with closed calottes in the Celtic environment. Without doubt, the expansion of the Roman state in the Celtic area had fundamental signifcance for the cultural situation. Also, the contribution from Germanic tribes should not be disregarded, which began to play an important role in the politics of Rome starting from the end of the frst half of the 1 st century B.C. Te presence of Germanic tribes on the lef bank of the Rhine is confrmed by written and archaeological sources. Nevertheless, the tradition of using shields with circular umbones, which persisted in extensive areas of Europe at the beginning of the La Tne period (the Apennine and the Iberian Peninsula), should also be taken into consideration (Couissin 1926, 144, 145; Stary 1981, 293, 294, 298, 300, 301, Abb. 1; Lorrio 1997, 192 194). Tis tradition survived in the Iberian Peninsula until the 2 nd
century B.C. which is testifed to by written sources (Polybius, History, III, 114, Diodor of Sicily, Historical Library, V, 33). Sporadic application of round shields with circular umbones in the Celtic environment is corroborated by stele presentations from Clunia, dating from the period of the reign of Octavian Augustus or an earlier period (Blazquez 1963, 412414, Picture No. 8, 9). It cannot be ruled out that the relatively quick reception or reintroduction of shields with circular calottes in the Celtic environment is a result of the continued existence of local production traditions and usage of this type of armament. It is necessary to emphasise that at the beginning of the period of Roman infuence, a group of hybrid shields appeared, combining certain features of the two separate structural concepts which were used in the La Tne period (i.e. shields ferruled with umbones with closed calottes and with open calottes reinforcing the vertical rib). Certain specimens of shields of one type have features characteristic of the second type. An example here is provided by a shield with a rib, yet ferruled with a circular umbo, which is presented on an early Roman altar from Nmes (Picture No. 10.3) (Dechelette 1914, Picture 496). Without doubt, in this case the umbo divides the rib into two parts, weakening the shield structure. An important example depicting how the two concepts of shield structure overlapped may be early Roman umbones of the Mehrum type (Picture No. 10.1) (Gechter, Kunow 1983, 466; Abb. 16, 13; Zieling 1989, 132; Kaczanowski 1992, 61, 62, Picture No. 15.1; Waurick 1994, 8, 11, Abb. 7.3) and the so-called spindle-shaped umbones (Picture No. 10.2) (Bockius 1989; Kaczanowski 1992a, 62, Picture No. 15.2; Waurick 1994, 186 | Tomasz Bochnak 8, 22, 23, Abb. 7.7; 14.4). Te Mehrum type umbones, known only from three locations, are umbones with circular edges and circular calottes with a marked rib (Picture No. 10.1). In the opinion of P. Kaczanowski, the described umbones did not comprise the legionaries armament, yet they were used by the auxiliaries; i.e., units in which representatives of individual ethnic groups inhabiting the Roman provinces served (Kaczanowski 1992, 62; Waurick 1994, passim). Tese could have been Celts or members of the Tracian tribes (Kaczanowski 1992, 62). Te spindle-shaped umbones refer to the ferrules of Celtic shields, yet the rib which they reinforced was so delicate that most probably it did not have any major functional signifcance. Dating of these specimens, younger both with respect to typical specimens with open and closed calottes, indicates that they cannot be considered intermediate forms between the two basic types of shield structure. Tese relics can be compared to the Bartodzieje type umbones, which are two centuries older and which also combined certain features typical for circular and rectangular umbones. Recapitulation Circular and band-shaped umbones represent diferent traditions, and there are no relations of genetic character between them. Te characteristics of the older types of umbones used by the people of the Przeworsk culture in the late Pre-Roman period, which have been presented above, clearly indicates that the fnds of band-shaped umbones, deriving from the Celtic environment, remain in a clear minority with respect to umbones with closed calottes. Terefore, it can be assumed that the people inhabiting Polish lands at the beginning of the late Pre-Roman period more ofen used shields without the rib reinforcing the structure. Umbones with closed calottes ferruling them were known in the Polish lands since phase A1. Terefore, these fnds demarcate the older horizon of ferrules from the times of Celtic domination in Europe and, at the same time, constitute a beginning stage of evolution of metal umbones, which were used in the area of the Barbaricum in the late Pre-Roman period and in the period of Roman infuence. Te emergence of hemispherical umbones may be considered a result of the fusion of local traditions related with the application of hypothetical leather umbones and impacts from the Celtic circle manifested in the popularisation of iron processing. Early Circular Umbones of the Przeworsk Culture. The Role of Local Tradition and Celtic Infuences... | 187 Bibliography Adler W. 1993 Studien zur germanischen Bewafnung. Wafenmitgabe und Kampfweise im Niederelbegebiet und im brigen Freien Germanien um Christi Ge- burt, Bonn. Babe M. 1993 Die Poieneti-Lukaevka-Kultur. Ein Beitrag zur Kulturgeschichte im Raum stlich der Karpaten in den letzten Jahrhunderten vor Christi Ge- burt, Bonn. Bianchetti E. 1895 I Sepolcreti di Ornavasso, Atti di Societa Archeologiae e Belli Arti di Tori- no 6, Torino. Biekowski P. 1928 Les Celtes dans les arts mineurs grco-romains, Krakw. Birkhan H. 1999 Kelten. Versuch einer Gesamtdarstellung ihrer Kultur, Wien. Blazquez J.-M. 1963 Lhrosation questre dans la Pninsule Ibrique, [in:] Celticum 6, Actes du Troisime Colloque International dEtudes Gauloises, Celtiques et Protoceltiques, Chateaumeillant-Bourges (Cher), 2730 Juillet 1962, Supplment Ogam-Tradition Celtique 86, 405423. Bochnak T. 2005 Uzbrojenie ludnoci kultury przeworskiej w modszym okresie przedrzym- skim, Rzeszw. Prepared for printing: Lumbo lappendice central dAlise-Saint-Reine et les trouvailles les plus proches provenant de lEurope centrale et septen- trionale, Antiquits Nationales. Bockius R. 1996 Zu einigen Schildbeschlgen der jngeren vorrmischen Eisenzeit aus Fundkomplexen der Przeworsk-Kultur, [in:] Z. Woniak (ed.), Kontakte lngs die Bernsteinstrae (zwischen Caput Adriae und den Ostseegebieten) in der Zeit um Christi Geburt, Materialien des Symposiums Krakw 26.29 April 1995, Krakw, 143154. Bockius R., uczkiewicz P. 2004 Kelten und Germanen im 2.1. Jahrhundert vor Christus. Archologische Bausteine zu einer historischen Frage, Monogr. RGZM 58, Mainz. Bohnsack D. 1938 Die Burgunden in Ostdeutschland und Polen, Leipzig. Brunaux J.-L. 1986 Les Gaulois. Sanctuaires et rites, Paris Brunaux J.-L., Lambot B. 1987 Guerre et armement chez les Gaulois (450-52 av. J.-C.), Paris. Brunaux J. L., Rapin A. 1988 Gournay II. Boucliers et lances, dpots et trophes, Paris. 188 | Tomasz Bochnak Bukowski Z. 1971 Tarcze oraz ich wyobraenia w kulturach uyckiej i wschodniopomor- skiej, Archeologia Polski 16, 177196. Coles J. M. 1962 European Bronze Shields, Proceedings of the Prehistory Society, N. S. 1962, 28, 156190. 1977 Archeologia dowiadczalna, Warszawa. Couissin P. 1926 Les armes romaines, Paris. Dbrowska T. 1988 Bemerkungen zur Entstehung der Przeworsk-Kultur, Praehistorische Zeitschrif 63, 5380. 1988a Wczesne fazy kultury przeworskiej. Chronologia zasig powizania, Warszawa. Dchelette J. 1914 Manuel darchologie prhistorique celtique et gallo-romaine t. II: Archol- ogie celtique ou protohistorique, Troisime partie, Second Age du Fer ou poque de La Tne, Paris. Decker K. V. 1968 Die jngere Latnezeit im Neuwieder Becken, Jahrbuch fr Geschichte und Kunst des Mittelrheins 1, 7208. Deyber A. 1987 La gurilla gauloise pendant la guerre des Gaules (5850 avant J.-C.), tudes Celtiques 24, 145183. Domaradzki M. 1977 Tarcze z okuciami metalowymi na terenie Celtyki, Przegld Archeolo- giczny 25, 5395. Duval A. 1987 Les armes dAlsia au Muse des Antiquites Nationales, Revue historique des Armes 1987/2, 5662. Egg M. 1996 Das hallstattzeitliche Frstengrab von Strettweg bei Judenburg in der Obersteiermark, Monogr. RGZM, 37, Mainz 1996. 2002 Eisenzeitliche Wafenweihungen im mittleren Alpenraum, [in:] Kult der Vorzeit in den Alpen, Opfergaben-Opferpltze-Opferbrauchtum, Bozen, 961984. Eogan G. 1964 Te Later Bronze Age in Ireland in the light of recent research, Proceed- ings of the Prehistoric Society 30, 268351. Fogel J. 1982 Uzbrojenie ludnoci kultury pomorskiej, Przegld Archeologiczny 27, 87123. Frey O. H. 1986 Einige berlegungen zu den Beziehungen zwischen Kelten und Ger- manen in der Sptlatnezeit, [in:] Gedenkschrif fr Gero von Merhart zum Early Circular Umbones of the Przeworsk Culture. The Role of Local Tradition and Celtic Infuences... | 189 100. Geburstag (= Marburger Studien zur Vor- und Frhgeschichte 7), 4579. Gaerte W. 1929 Urgeschichtliche Ostpreuens, Knigsberg. Gamba M. 1987 Analisi preliminare della necropoli di Arqua Petrarca (Padova), [in:] D. Vitali (ed.) Celti ed Etruschi nellItalia Centro-Septentrionale dal V sec. a. C. alla romanizzazione, Imola, 237270. Gechter M., Kunow J. 1983 Der frhkaiserzeitliche Grabfund von Mehrum. Ein Beitrag zur Frage von Germanen in rmischen Diensten, Bonner Jahrbucher 183, 449 468. Gediga B. 1970 Motywy fguralne w sztuce ludnoci kultury uyckiej, Wrocaw. Gleirscher P. 2002 Alpine Brandopferpltze, [in:] Kult der Vorzeit in den Alpen, Opferga- ben-Opferpltze-Opferbrauchtum, Bozen, 591634. Graue J. 1974 Die Grberfelden von Ornavasso, Hamburger Beitrge zur Archologie, Beihef 1, Hamburg. Gutin M. 1984 Die Kelten in Jugoslavien. bersicht ber das archologische Fundgut, Jahrbuch des Rmisch-Germanischen Zentralmuseums Mainz 34, 305 365. 1991 Posoje. Posoje in der jngeren Eisenzeit, Ljubljana. Hachmann R. 1956/57 Zur Gesellschafsordnung der Germanen in der Zeit um Christi Ge- burt, Archaeologia Geographica 5/6, 724. 1957 Ostgermanische Funde der Sptlatnezeit in Mittel- und Westdeutsch- land.Ein Beitrag zum Problem des Nachweises von Bevlkerungsbewe- gungen auf Grund des urgeschichtlichen Grundstofs, Archaeologia Ge- ographica 6, 5568. 1961 Die Chronologie der jngeren vorrmischen Eisenzeit. Studien zum Stand der Forschung im nrdlichen Mitteleuropa und in Skandinavien, 41. Bericht der Rmisch-Germanischen Kommission 1960, 1276. Hafner A. 1976 Die westliche Hunsrck-Eifel Kultur, Berlin. Harmand J. 1967 Larme et le soldat Rome de 107 50 avant notre re, Paris. Hazell P. J. 1981 Te Pedite Gladius, Antiquaries Journal 61/1, 7382. Jahn M. 1916 Die Bewafnung der Germanen in der lteren Eisenzeit, Wrzburg. 1924 Die Gliederung der Wandalischen Kultur in Schlesien, Schlesiens Vorzeit in Bild und Schrif, Neue Folge 8, 2036. 190 | Tomasz Bochnak 1931 Die Kelten in Schlesien, Leipzig. 1937 Rtische Kultureinfsse in Bhmen und Schlesien des letzten vorchrist- lichen Jahrhunderts, Sudeta 13, 103111. Jasnosz S. 1951 Cmentarzysko z okresu pnolateskiego i rzymskiego w Wymysowie, pow. Gosty, Fontes Praehistorici 2, 1282. Joachim H. E. 1973 Ein reich ausgestattetes Wagengrab der Sptlatnezeit aus Neuwied, Stadtteil Heimbach-Weis, Bonner Jahrbucher 73, 144. Kaczanowski P. 1992 Importy broni rzymskiej na obszarze europejskiego Barbaricum, Krakw. Kleemann J. 2002 Nochmals zum Schildbuckel der Variante Bartodzieje (Zeippern), [in:] C. von Carnap-Bornheim, J. Ilkjr, A. Kokowski, P. uczkiewicz (eds.), Bewafnung der Germanen und ihrer Nachbarn in den letzten Jahrhun- derten vor Christi Geburt, Lublin, 119226. Klindt-Jensen O. 1950 Foreign Infuences in Denmarks Early Iron Age, Acta Archaeologica 20, 1229. Kostrzewski J. 1919 Die ostgermanische Kultur der Sptlatnezeit, Leipzig. Knez T. 1977 Keltski grobovi iz Roj pri Moravach, Keltske Studije 4, 105127 1987 Sptlatnezeit und frhe Kaiserzeit in Novo mesto, Archeologisches Kor- respondenzblatt 17, 103109. 1992 Novo mesto II. Keltsko-rimsko grobie, Novo Mesto. Krmer W. 1952 Das Ende der Mittellatnefriedhfe und die Grabfunde der Sptlatne- zeit in Sdbayern, Germania 30, 330339. La Baume W. 1920 Vorgechichte von Westpreussen, Danzig. 1934 Wiederherstellung ostgermanischer Wafen der Sptlatnezeit, Prhis- torische Zeitschrif 25, 160165. 1935 Zur Darstellung von Schilden auf ostgermanischen Urnen der frhen Eisenzeit, Altpreuen 1/3, 145151 1950 Zur Bedeutung der bildichen Darstellungen auf Gesichturnen der Frhen Eisenzeit, Praehistorische Zeitschrif 34/35, 1949/1950, 158178. Lewczuk J. 1997 Kultura przeworska na rodkowym Nadodrzu w okresie lateskim, Pozna. Lippert A. 1992 Ein latnezeitlicher Opferplatz in Teurnia Spittal an der Drau, [in:] A. Lippert, K. Spindler (eds.) Festschrif zum 50-jhrigen Bestehen Institues fr Ur- und Frhgeschichte der Leopold- Franzes-Universitt Innsbruck, Bonn, 285304. Early Circular Umbones of the Przeworsk Culture. The Role of Local Tradition and Celtic Infuences... | 191 Lorrio A. J. 1997 Los Celtiberos, Alicante. uczkiewicz P. 2004 Uzbrojenie ludnoci kultury przeworskiej w modszym okresie prze- drzymskim, [in:] J. Andrzejowski, A. Kokowski, Ch. Leiber (eds.), Wan- dalowie. Stranicy bursztynowego szlaku, LublinWarszawa, 127141. uka J. L. 1971 Obrzdek pogrzebowy i wierzenia u plemion kultury wschodniopomor- skiej na Pomorzu Gdaskim, cz II, Pomorania Anitqua 3, 21100. Martens J. 2001 A Wooden Shield-Umbo from Kvrlv, Scania. Some Remarks on the Weaponry of Early Pre-Roman Iron Age in Northern Europe and the Origin of Hjortspring Warriors, [in:] M. Meyer (ed.) ...trans Albim fu- vium, Festschif fr Achim Leube zum 65. Geburtstag, Forschungen zur vorrmischen, kaiserzeitlichen und mittelalterlichen Archologie (= Inter- nationale Archologie, Studia Honoraria 10), Berlin, 135159. Mhling W. 1944 Das sptlatnezeitliche Brandgrberfeld von Kobil, Prag. Ostrowski J. 1999 Staroytny Rzym. Polityka i sztuka, Krakw. Pescheck Ch. 1939 Die frhwandalische Kultur in Mittelschlesien, Leipzig. Peschel K. 1977 Frhe Wafengrber im Gebiet der sdlichen Elbgermanien, [in:] B. Chropovsk (ed.) Symposium Ausklang der Latne-Zivilisation und Anfnge der germanischen Besiedlung im mittleren Donaugebiet, Bratis- lava, 261281. Polenz H. 1971 Mittel- und sptlatnezeitliche Brandgrber aus Dietzenbach, Studien und Forschungen, 4. Rafery B. 1983 A catalogue of Irish Iron Age antiquities, Marburg. Randsborg K. 1995 Hjortspring. Warfare and Sacrifce in Early Europe, Aarhus. Rapin A. 1983/84 Larmement du guerrier celte au Deuxime Age du Fer, [in:] L art cel- tique en Gaule, Catalogue de lexposition, Marseille, Paris, Bordeaux, Di- jon, s. 6979. Rosenberg G. 1937 Hjortspringfundet, Nordiske Fortidsminder, III, Kbenhavn. Schnberger H. 1950 Die Sptlatnezeit in der Wetterau, Saalburg Jahrbuch 11, 21130. Schnfelder M. 2002 Das sptkeltische Wagengrab von Bo. Studien zu Wagen und Wagen- grber in der jngeren Latnezeit, Monogr. RGZM 54, Mainz. 192 | Tomasz Bochnak Seidel M. 2000 Die jngere Latnezeit und ltere Rmische Kaiserzeit in der Wetterau, Fundberichte aus Hessen, 34/35 (1995/95). Sievers S. 1995 Die Wafen, [in:] M. Redd, S. v. Schnurbein, P. Barral, J. Bnard, V. Bronquier-Redd, R. Goguey, M. Joly, H.-J. Khler, Ch. Petit (eds.), Neue Ausgrabungen und Forschungen zu der Belagerungswerken Caesars um Alesia (19911994) (= 76. Bericht RGK), 75158. 2001 Les armes Alsia, [w:] Alsia. Fouilles et recherches franco-allemande sur les travaux militaires romains autour du Mont-Auxois (19911997), 2 Le matriel, (red. M. Redd, S. v. Schnurbein), Paris, 121209. Stary P. F. 1979 Keltische Wafen auf der Apennin-Halbinsel, Germania 57, 90110. 1981 Ursprung und Ausbreitung der eisenzeitlichen Ovalschilde mit spindel- frmigem Schildbuckel, Germania 59, 287306. 1981a Zur Eisenzeitlichen Bewafnung und Kampfesweise in Mittelitalien, Marburger Studien zur Vor- und Frhgeschichte 3, Marburg. Stead I. M. 1985 Te Battersea Shield, London. Sylwestrowicz J. 1979 Interpretacja znaczenia motyww boru, jelenia i tarczy w przedstawie- niach fguralnych kultury wschodniopomorskiej, Pomorania Antiqua 9, 960. Todorovi J. 1965 Predlog hronoloke podele keltskog latena u junoj Panonii i severnom Balkanu, [in:] Materijali Simpozijum Praistoriske i Srednjevekovne sekcije Arheolokog drutva Jugoslavije 3, 2752. 1972 Praistorijska Karaburma I, Nekropola mlaeg gvozdenog doba, Musej Grada Beograda Monografja 3, Beograd. Waurick G. 1994 Zur Rstung von frhkaiserzeitlichen Hilfstruppen und Verbndeten der Rmer, [in:] C. v. Carnap-Bornheim (ed.), Beitrge zu rmischer und barbarischer Bewafnung in den ersten vier nachchristlichen Jahrhun- derten. Marburger Kolloquium, LublinMarburg, 125. Wogiewiczowie M. D. i R. 1963 Uzbrojenie ludnoci Pomorza Zachodniego w pocztkach naszej ery, Materiay Zachodniopomorskie 9, 9166. Woniak Z. 1970 Osadnictwo celtyckie w Polsce, WrocawWarszawaKrakw. Zieling N. 1989 Studien zu germanischen Schilden der Sptlatne- und der rmischen Kaiserzeit im freien Germanien, BAR. International Series 505, Oxford. Ziemacki W. 1965 Tarcze bojowe i tarcze okazjonalne, Z Otchani Wiekw 30, 133135. Early Circular Umbones of the Przeworsk Culture. The Role of Local Tradition and Celtic Infuences... | 193 Wczesne umba koliste kultury przeworskiej. Rola tradycji lokalnych i wpyww celtyckich na zrnicowanie metalowych czci tarczy w pocztkach modszego okresu przedrzymskiego Streszczenie Artyku omawia genez i zrnicowanie najstarszych umb koli- stych, przedstawiono rwnie skrtowo podstawowe dane dotyczce umb tamowatych. Omwiono zasadnicze rnice konstrukcyjne po- midzy tarczami z eberkiem okutym umbem tamowatym oraz tar- czami o paskiej pycie, wzmocnionymi umbem z zamknit kalot. Umba koliste byy powszechnie uywane w Europie w okresie halszta- ckim i w pocztkach okresu lateskiego i naley przypuszcza, e wy- ksztaciy si one niezalenie w rnych czciach Europy; na Pwyspie Iberyjskim, Apeniskim, na Wyspach Brytyjskich oraz w Europie rod- kowej. Problem genezy metalowego umba kolistego, ktre pojawio si w Europie rodkowej w okresie dominacji celtyckiej, nie zosta dotych- czas wyjaniony. Cz badaczy uwaa je za miejscowy wytwr ludnoci zamieszkujcej rodkowoeuropejskie Barbaricum, zdaniem innych na- tomiast, umba tego typu powstay jako wynik ewolucji lateskich form tamowatych. W artykule przedstawiono argumenty przemawiajce za pierwsz z opcji, oraz podwaono hipotez przeciwn, m.in. zwraca- jc uwag na fakt, e pierwsze metalowe umba koliste s starsze od umb uwaanych za formy porednie midzy tamowatymi a kolistymi. Zdaniem autora, najstarsze metalowe umba koliste powstay w wy- niku zaadaptowania nowego surowca elaza w miejsce innych uy- wanych wczeniej substancji organicznych, np. skry. Skrzane tarcze z kolistymi umbami znane s z Wysp Brytyjskich, a na stosowanie zbli- onych oson na ziemiach polskich wskazuj rda ikonografczne. Skrzane umba maj ksztat zbliony do niskiej kopuy, a przeniesienie tego ksztatu na wyrb metalowy jest stosunkowo proste i nie wymaga wyspecjalizowanych narzdzi. Sposb obrbki skry i blachy metalowej oczywicie rni si od siebie zasadniczo, jednak materiay te s zbli- one do siebie, jeeli chodzi o moliwoci modelowania wybranych ksztatw. Wspomniane podobiestwo mogo by przyczyn, dla kt- rej blach elazn obrabiano naladujc umba skrzane, a nie drewnia- ne. Naley bowiem zauway, e wyduony ksztat najstarszych umb drewnianych ze Skandynawii (Kvrlv, Hjortspring) jest uwarunkowany 194 | Tomasz Bochnak wknist struktur drewna. Gdy drono w drewnie umbo, nadawano mu ksztat wrzecionowaty, by zapobiec pkaniu najbardziej delikatnych partii, tj. tych, ktre ksztatowano tnc w poprzek sojw. W artyku- le przedstawiono rwnie zrnicowanie najstarszych umb z zamknit kalot, w tym tzw. umb typu Bartodzieje, ktre zdaniem autora naley uwaa za form lokaln, typow dla kultury przeworskiej.