Brooches on the move in Migration Period Europe Magnus, Bente Fornvännen 2004(99):4, s. [273]-283 : ill. http://kulturarvsdata.se/raa/fornvannen/html/2004_273 Ingår i: samla.raa.se

Brooches on the move in Migration Period Europé
By Bente Magnus

Magnus, B. 2004. Brooches on lhe move in Migration Period Europé. (Spenner pä vandring i folkevandringstidens Europa). Fornvännen 99. Stockholm. During the sugar-beet harvest of 1942 a fragment of a Baltic crossbow brooch (KLM 20732) was found inside Gräborg, the largest of the Migration Period ring forts of Öland. On the basis of fragments of a casling mould for a similar brooch found at the central place of Hdgö on Lake Mälaren, Joachim Werner laler postulated that this type of brooch was mosl probably produced al Helgo. Around 1980 a female grave was excavated at the Bavarian row grave cemetery of Altenerding in Germany, containing a similar crossbow brooch and other piéces ofjewellery foreign to that region. The finds were tentativdy interpreted as pointing towards Seandinavia, and the lady was ascribed an ethnic affilialion: Swedish. This paper examines the arguments for this condusion and offers alternatives. It also touches upon lhe difficult problem of interpreting foreign objects in female graves, particularly brooches, as ethnic markers.
Hen le Magnus, Saltmälargalan 6, SE-111 60 Slot kholm

bente.magnus@raa.se

Gråborg, the Grey Fort, is the largest of the Migration Period ring forts of Oland. 16 forts are known today and three have been postulated on account of finds and place names (Wegra-us 1976, p. 36). Gråborg is not quite circular and the area within the walls measures 210 by 162 m. The surrounding wall is built mainly of rectangular blocks of limestone and its crown measures 640 m in circumference. During the Middle Ages an outer ditch was added to the defence system. There are three gates, one in the north, one in the south and one in the north-west, two of which were rebuilt in the Middle Ages and furnished with a vault and a tower. The area within the walls has been tilled at least since the 171b century. Several linds particularly from the Viking Period and the early Middle Ages (after AD 1050) have come to light during seasonal agricultural work (Stenberger 1933, p. 214 ff). U.E. Hagberg has direc-

ted excavations on a small scale in Gråborg during 2001—2003. In the late summer of ig42, the sugar-beet harvest was in full swing on a plot within the walls of Gråborg when a young woman suddenly found a small decorated metal object. Bertil Almgren, then a young student spending the summer on the island, was suniinoned, and could inform the public that the object was made of gilded silver and was part of a fibula (fig. ui-c; KLM 20732). The find is the bow and foot in one piece of a crossbow brooch made of gilded silver and decorated in chip carving with spiral tendrils and one animal head at each end. The brooch lacks the cross bar and its terminal knobs as well as the are. The pin, which judging from traces on the fibula was made of iron, is also missing. The fibula measures 66 nun in length and 1.2—1.5 mm in width.

lornvännen gg (2004)

274 Bente Magnus

£1 I
^ v ".
«•'

Fig. i a-c. Fragment of a crossbow brooch . Gilded silver. Gråborg ring fort, Öland, Sweden. KLM 20732. Photograph by ti. Hildebrand, Raä. Scale 1:1.

A produet front Helgo? The brooch fragment was first published by Werner (1951) who conduded that it was Masurian, a condusion accepted by Äberg (1953, p. 93 f). fn a låter artide about the mobility of objects and persons, Werner (1970) again focused on the Gråborg brooch due to a find from H d g ö of two fragments of a mould for the casting of a similar crossbow brooch (fig. 2). Werner refers to works of Äberg and Oxenstierna where the type is described as East Prussian (Masurgermanic). The two mould fragments from H d g ö and the Gråborg brooch made him question the traditionally ascribed origin of this type of brooches (ig7<», p. 80) and instead suggest a Scandinavian origin. Werner's tentative condusion was strengthened by a female grave (grave 421) at the Bavarian row grave cemetery at Altenerding in Oberbayern, Germany with an unusual set of jewellery (Sage 1 g84, p. 1 20 f). It consists of two dress pins and a neck ring of bronze, a bronze crossbow brooch with ribbon-shaped bow and
Fornvännen g.t {2004)

foot, and a compact gilded silver crossbow brooch of the Gråborg type. This set of jewellery led Werner to c o n d u d e thal the buried woman was most probably of Scandinavian origin and that her silver brooch may have been produced in a workshop at H d g ö (Werner 1970, p. 78 ff). Sage on the other hand, who excavated the grave, says the furnishing characterises a lady buried in the coslume of the southern Baltic region "...das charakterisiert aber wohl am treffendsten Grab 4 2 1 , in welchem im friihen 6. Jahrhundert eine Dame in der fräindländischen Tracht des sudlichen Ostseeraumes ihre letzte Ruhestätte fand" (1975, p. 275 f), which is not quite the same as Werner's condusion. Menghin ( i g g o , p. 75, Abb. 62) niaintained that the jewellery of grave 42 1 has its counici parts in the South Baltic region and that the lady in question probably originated there. What the terms "siidliches Ostseeraum'" or "South Baltic region" ineans is not entirely clear, but obviously neither Sage nor Menghin

Brooches on lhe move in Migration Period Europé 275 tion for a small ring. The pin's stern and head are separated by a moulding. According to Werner the pins are Scandinavian: "Dieses Nadelpaar ist skandinavisch und besitzt von Norwegen iiber Schweden bis Finnland zahlreiche Parallellen" (1970, p. 80). In the Migration Period dress pins were in use in Seandinavia and Finland, but they disappear from the graves in the early Merovingian Period in Norway and mainland Sweden with a change in dress fashion. On Gotiand and Bornholm, in Finland and particularly in the Baltic countries, however, dress pins continued in use. Perforated dress pins with rings to fästen long chains of bronze spirals are well known from Lettgallia (Balodis ig4o, p. 128, figs. 4, 5, 7). Fig. 2. Two fragments of clay moulds for casting a crossbow brooch. Hdgö, Uppland, Sweden. Scale I have not been able to find exact parallels to the pins from Altenerding among the published Scandinavian material. In the fairly large material of dress pins and casting moulds for dress pins from H d g ö , this design does not occur (Waller iggö, p. 48). Dress pins with mushroom-shaped heads in round are well known from Gotland both in the Migration Period (period VL2; Nerman 1935, p. 70 f, Taf. 38, figs. 382~3gi) and particularly in the Early Merovingian /Vendel Period (period VII:i; Nerman 1969, p. 6g 1, Taf. 14—15). None, however, have a flat and perforated mushroom shaped head. Most of the pins from Gotland belong to non-grave contexts but they also occur in a number of graves at the cemetery of Barshalder (Rundkvist 21x13, p. 156, fig. 1 o: 13D; i'.174, fig. 10:1 gF; p. 205, lig. 10:228). On Bornholm, dress pins belong to the female jewellery of the Early Merovingian Period, c. AD 540—600 (AM II-III, Hoilund phase lAB; Hoilund Nielsen 1987, p. 6g, lig. ig). At the cemetery of N0rre Sandegård Vest the use of dress pins ceased completely around the transition of phase 1D2 to phase 2A, i.e. AD 700 (Jörgensen & Nörgård Jörgensen igg7, p. 45, fig. 15). In Orsnes' seminal study from 1966 there are only a few dress pins (p. 160 ff, figs 197—805). Most of them have been found in Bornholm and belong early in the period. Although dress pins were much used in Seandinavia during the Migration Period, the dress pins of the Altenerding grave may well bc
Itonvännen gg (2004)

agreed with Werner about the Scandinavian origin of the buried lady. According to Menghin, the large Bavarian row grave cemeteries Erding-Altenerding and Straubing-Bajuwarenstrasse show signs of a mixed population. This stems from the hislorically attested withdrawal of the Roman troops in lhe 51b century, which left this region free for Germanic and other groups of tribal seltlers to move in. The. jewelkry from Allenerding grave 4 2 1 Werner's idea of a Scandinavian origin for the Bavarian lady and also for the crossbow brooch from Gråborg has continued to circulate in the archaeological literature (cf. .Arrhenius 1992; Koch iggg; Bitner-Wröblewska 2000, 2001). It may therefore be of interest to re-evaluate the basis for Werner's condusion, particularly in the light of research done since ig70. Let us treat the "Scandinavian" piéces first (fig. 3). The pair ofdress pinsis foreign to the Bavarian female dress of this period. I presume ihal Werner was convinced of their Scandinavian origin after taking part in the H d g ö conference in i g 6 8 and seeing the casling moulds for dress pins found at this site (Waller 1996). The Altenerding dress pins are made of bronze wilh a flat mushroom-shaped head with a perfora-

276

Bente M a g n u s

Fig.3. Tbc major objects from grave 4 2 1 , Altenerding, Bavaria (after Sage 1984). Fornvännen gg (2004)

Brooches on the måne in Migration Period Europé 277 of South Baltic origin but made according to Gotlandic or Bornholmian fashion. Another possible source for the dress pins may be AngloSaxon England where such pins, but mostly used singly, are rather common (Timby iggö, p. 5g; Evison ig87, p. 82). Ghronologically the dress pins belong in the Late Migration Period or Early Merovingian Period. I c o n d u d e that the dress pins constitute a weak link in any chain of indications aiming to put an ethnic stamp on the Altenerding lady. is the best preserved and the most distinctly fashioned of the group with spring and coil and punched decoration. Then follows the Kentish brooch, the brooch from Öland and the small indistinct brooch from Muhlhausen (Werner 1955, p. 76, Abb. 1, 1—3). However, new specimens of this type of brooch have turned up in recent excavations in Kent. In the cemetery on Mill Hill, the well-furnished female grave 73 contained a cast imitation of such a brooch (Parfitt &: Brugmann 1 gg7, p. 45 f, fig. 13c!, fig. 16). Also, the eponymous Ozingell site has produced another pair of brooches of the same type as the original 1 gth century find (Parfitt & Brugmann igg7, p. 46). It seems that the case for a Scandinavian origin for the Ozingell type brooches is rather weak. It is in fact most p r o bably an Anglo-Saxon type. A further brooch of this type was published in 2002 (Droberjar 8003, p. 109). It was found in a sunken feature building at Jenstein 2 near Prague and seems to be the finest specimen of them all with a varied punch decoration.

The bronze brooch from grave 421 is a crossbow brooch of a rare type named Ozingell after an early grave find from the Isle of Thanet, Kent, England. Here, two similar brooches were found (Shetelig 1 g 1 o, p. 73, fig. 44; Äberg 1983, p. 128, fig. 220; Werner 1955, p. 75; SchulzeDörrlamm 1 g86, p. 618 ff). Olher brooches of this type are one from Sättra, Gärdslösa parish, Öland; one from grave 130 at Hoogebeintuni near Leewarden on the coast of Holland (Werner 1955) and a fragmented piece from a female grave in the row grave cemetery at Gömersche Landstrasse, Mtihlhausen, Thuringia The bronze neckkl consists of a rod with cir(Behm-Blancke 1959, p. 240 ff, Taf. 3g; Schulzecular section and a flattened front part with Döriiannn 1986, p. 61g 0- The brooch from punched decoration. Its distinguishing feature Öland is according to Äberg a unique find in is the hook-and-eye fastening on the rightEast Seandinavia. It probably belongs with three hand side, which is unique. This feature is unother simple crossbow brooches, a bronze finger known in Seandinavia. Kivikoski considered it ring and some glass beads deformed by fire, a special Estonian feature without citing refefound in a cremation grave in a small cairn rences, cf. Orsnes 1966. Neck rings were p o (SHM inv. no. g585; Stenberger 1933, p. 63; pillar during the Migration and Early MeroBeskow Sjöberg ig87, p. 2go, no. 20, 340). vingian Periods on Golland and Bornholm, in Finland and in the East Baltic region. A few have T h e Miihlhausen grave contained several also been found on the Swedish mainland and small brooches of presumed Anglo-Saxon oriin AngloSaxon contexts (Adams & Jackson gin in addition to an imitation of an Anglo1990, p. 59, 156). Orsnes (1966, p. 168 1) assigSaxon great square-headed relief brooch with ned the Altenerding necklet to his group Q8 a tongue-like terminal lobe, no distinci side loand generally considered the south Scanbes on the loot plate and riidimentary decoradinavian necklets of the early Merovingian tion (Behm-Blancke 1959, fig. 3g). SchulzePeriod to be local produets inspired by the faDörrlamm (1986, p. 320) considered the small shion cn vogne east of the Baltic Sea. Wamers Ozingell brooch to be of Scandinavian origin (2000, p. 54 i), referring to Werner 1970, sugand the "relief brooch" to be an imitation of a gests a south Swedish origin for the lady of Scandinavian one. Haseloff (1981, p. 497 ff), Altenerding grave 421 and maintains that her on the other hand, considered the relief brooch necklet may have parallels from Öland. No to be a Gentral German copy of a Scandinavian necklet of this type is, however, known from model made by a craftsman who showed little Öland according to the publications Olands understanding of Germanic Animal Style I. järnåldersgravfält I-IV (The Iron Age cemeteThe type Ozingell brooch from Altenerding
Fornvännen gg (2004)

278 Bente Magnus ries of Öland). The same goes for Bornholm and Gotland. On Gotland necklets of similar shape but with the fastening contraption worn at the back of the neck are not known from grave finds, only from votive deposits. Nerman considers necklets an East Baltic influence (Nerman 1935, p. 71), which is as close as we can get at the moment. Consequently, the proposed Scandinavian origin för the Altenerding necklet is dubious. The crossbow brooch from the Altenerding grave is of a type well-known in the south Baltic area. It is made of gilded bronze cast in several piéces and measures 60 mm. Much research has been done on the origin and chronology of this brooch type (cf. Bitner-Wröblewska 2000, 2001; Bliujiené ms.). I will not enter that discussion, but present some views on the scant Swedish material related to the Altenerding grave, namely the Gråborg piece and the two fragments of a mould for casting a brooch of that type found at H d g ö . The brooch from Gråborg belongs to a type called Daumen/Tumijany (Bitner-Wröblewska 2000, p. 183; 2001, p. 83 ff). Only one part of it remains, the bow and foot in one piece (fig. 1). Originally it had a number of additional parts: a cross bar, two terminal knobs, two spirals, an are and a pin. The pin catch is rectangular with cross-hatching on both sides. The decoration of the brooch is executed in deep chip carving with animal motifs as well as scrolls. The decoration is divided in two halves by way of a ribbon-like moulding running the entire length of the brooch. It shows traces of niello with silver döts. The motifs are the same on both halves and may thus be understood seen from above as well as from the sides. The moulding branches off in four places framing two terminal animal heads and one at each end of the bow. The heads have open mouths displaying teeth or the tongue. On the foot of the brooch above the terminal head and on both sides of the midrib there is a panel with a pattern which may be interpreted as a couple of animal limbs. The decorative style looks at first glance like Scandinavian Style I. A doser look, however, reveals that it is not, but soniething with dose affinities. Compared with Scandinavian relief b r o o
l-toovänni-n t)g 121104)

ches, the decoration of the Gråborg brooch shows affinity with brooches of the GotlandÖland type (Näsman ig84; Sjovold igg4, figs 28—2g) in the following features: the silver döts in niello on the moulding, the spirals executed in chip carving on the bow, and the "buffer" which protrudes below the animal head terminal on the foot. The animal heads of the Gråborg brooch and other brooches of the same type are also similar to the animal-shaped Vendel Period sword pommel from Kirmukarmu, Vesilahti parish, Finland (Kivikoski 1973, Abb. 508). The general outline of the brooches is also obviously related to Early Vendel Period crab brooches like the very fine specimen from Hallbjäns, Lau parish, Gotland (fig. 4). The two piéces of a casting mould for a crossbow brooch of the Tumiany type found at H d g ö (fig. 2) may be interpreted in a number of ways. The mould must have been used to make a brooch of the compact type with a simplified construction in three parts, i.e. the whole brooch cast in one piece, plus a spiral and the pin on the back. No other fragments of the mould were retrieved at H d g ö , but we must assume that they once existed and were used for casting at least one compact crossbow brooch. Unfortunatdy, the mould fragments were found very early in the 25 years' campaign at H d g ö and no information about the exact find spöt within building complex no. 3 is recorded (Årsrapport 1968, p. 1 1). Tbousands of fragments of moulds for casling more than 210 Migration Period relief brooches were found in tbc workshop areas at Helgo, but lhe majority of the brooches produced have never been found (Lundström ig72,p. 137). They were most likely melted down and the metal re-used when the relief brooches went out of use. Most of lhe objects produced at H d g ö belong to the period c. AD 300—550. However, casting moulds for brooches of the låter 6th century, the Early Vendel Period (such a.s small equal-armed brooches. Husby brooches, crab brooches and bird brooches), demonstrate that at least one workshop continued to operate after the main production had come to an end. Despite the two mould fragments, my opinion is that there is little reason to believe that

Brooches on the move in Migration Period Europé 27g Fig. 4. Grab brooch. Hallbjäns, Lau parish, Gotland, Sweden. Drawing by C, Bonnevier. Scale
1:1.

the compact type of the Baltic crossbow b r o o ches originated and was generally produced in Seandinavia. There is a much stronger case for such brooches to have been produced in Masuria in present north-east Poland, as both Äberg and Werner once maintained. The crossbow brooches of Bitner-Wröblewska's two main types are foreign to the costume of Migration Period Seandinavia, female as well as male. Bitner-Wröblewska (2000, p. 183; 2001, p. 7g) has suggested a division of the Baltic and Masinian crossbow brooches according not to shape, but according to presence or absence of relief decoration. This original solution to a difficult problem gives two main groups of crossbow brooches, which she has named the Sensbnrg/Mragowo type and the Daumen/Tumiany type, respectivdy. To the Sensburg/Mragowo type (crossbow brooches of slender construction without relief decoration but with a animal head terminal on the foot) she reckons three specimens found at Öland: at Hönstorp, Algutsrum parish; at Skogsby, Torslunda parish; plus one unprovenanced brooch. The

brooches are stray finds, but the Hönstorp brooch was found not too far north of Gråborg (Åberg ig23, p. 131, fig. 234, 235). The Daunien/Tumiany type contains both slender and compact crossbow brooches with relatively large knobs and with most of the brooch body d e c o rated with either spirals or animal motifs in relief. The shape of the knobs demonstrates their relation to Late Roman brooches with onion knobs and to the låter Baltic poppy capsule crossbow brooches. The Daunien/Tinniany type, to which the Gråborg brooch belongs, was made in two main varieties, one composite and one compact. They have a wide distribution from Bavaria över Masuria to Lithuania, Latvia and Oland (see map in Bitner-Wröblewska 2001) and confirm the widespread net of contacts among people on the move that is characteristic of the mid-first millennium AD. As for the chronology of tbc brooches, opinions differ very little. Werner ( i g 7 o , p. 78) dated Altenerding grave 421 to the first half of the sixth century AD, which has become a guideline for the dating of this type of crossbow
Fornvännt-n t)t) 12004)

280 Bente Magnus brooches (cf. Kazakevicius 1983, p. 104; BitnerWröblewska 2000, p. ig; 2001, p. 87). I find it difficult to suggest a chronological framework for this type of brooches based on the available material. Generally they seem to belong to the final phase of the Migration Period or even the earliest phase of" the Merovingian/Vendel Period. A n exa mple of exoga my ? Werner c o n d u d e d that the lady of the Altenerding grave 421 could be an obvious example of exogamy. To my knowledge no grave find with a set of jewellery similar to that of the Altenerding grave is known from Seandinavia. It comprises an unusual inixlure of Baltic and Anglo-Saxon items, and the only items that may tentativdy be labelled Scandinavian are the two dress pins. Ursula Koch (1 ggg) bas listed all known male and female grave finds containing objects that do not conform with the "normal" pattern of South German row grave cemeteries on the right-hand side of the Rhine in the second quarter of lhe sixth century. Her underlying idea is that female dress accessories disdose the ethnic affiliation of the deceased, i.e. that North Germanic/Scandinavian women who were married off to men further south in Europé were buried in their original regional costume. This was also the idea behind Werner's postulate that the woman of Altenerding grave 421 was an example of exogamy from South Seandinavia. Koch, who calls the lady in question "Swedish", considers the Scandinavian influx into South Germany to have come with the Thuringians, allowing links to the written history of this people. She may be right, but there are major problems in assigning graves to ethnic groups and postulate exogamy. Before suggesting an ethnic interpretation there are certain questions to be answered or at least discussed. The first is whether women given in marriage far from their home region were likely to be buried in their "native" costume. Secondly, was exogamy a common political way of securing bonds between families other tban among the Germanic societal d i t e in the Migration Period?
Fornvännen gg (2004)

To answer the first question: as far as we know, post-marital residence was virilocal, i.e. the woman went to live with her husband's family. If there were differences in costume between her home region and her new home, one would presume that she would be given the appropriate dress of her luisband's home region when moving into her new household (Straume i g g s , p. g7 ff)- But more to the point: is it likely that she would be dressed in her "native" costume or decorated with her "native" jewellery when buried or cremated? The second question is whether women were used as political pawns outside of the very d i t e among the Germanic societies. As far as we know, this was not common, and the lady of the Altenerding grave 421 did not, judging from the furnishings of her grave, belong to the top level of society. She may of course have been the victim of abduetion (Vierck 1970, p. 394) and have followed Thuringians 011 their way to South Germany, as proposed by Koch. Vierck's condusion that, judging from her jewellery, the lady of Altenerding was most probably Anglo-Saxon, seems better founded than any suggestion of a Scandinavian origin. Arrhenius (1993) has discussed exogamy in relation to early Germanic marriage traditions, låter c o dified in laws, about morning gift and dowry. She maintains that gold jewellery like bracteates and gilded silver great relief brooches may well have been part of a morning gift within the Germanic d i t e . When such piéces appear "foreign" in relation to the rest of the funerary outfit, exogamy may well be the explanation. Settlers and raiders My intention with this artide is not to question whether cases of exogamy are possible to deduce from an archaeological material. Rather, it is to focus on Werner's artide from 1970 which has had such a profound influence both 011 the chronology of the Baltic Middle Iron Age and 011 the conception of the importance of the Scandinavian influence south of lhe Baltic Sea. The widespread distribution of the D a u m e n / Tumyani brooches of the variant found in Gråborg may be due to itinerant craftsmen who adopted Scandinavian ornamental details when

Brooches on the move in Migration Period 281 producing typical Baltic brooches (Bitner-Wröblewska 2001, p. 87 f). Brooches like the one from Gråborg are considered to be jewellery for high status men (cf. Bliujiené ms.), but could obviously be worn also by women outside of the core distribution area of the brooches. The Gråborg brooch may thus have belonged either to a man or a woman. If we choose to see it as a male attribute, the next question is, what a high-ranking Baltic or Masurian man was doing within the walls of the largest Migration Period fortress of Öland. This question is eqnally relevant for the other single finds of Baltic brooches known from Öland (cf. Äberg 1923, p. 123 ff.) The position of the island of Öland in the Baltic Sea opens it to peaceful as well as martial contact both with the Swedish mainland and areas east and south of the Baltic Sea. The 16 ring forts testify to the population's need for defence, and the numerous hoards of Roman solidi, ring gold and gold jewellery from this period give an indication of the island's wealth (I lerschend 1980, p. 252). The gold finds are not spread evenly över the island, bul gather in the coastal areas on either side of Gråborg, the largest ring fort of the island, in the region located between the third and fourth largest ring forts Ismanstorp and Mossberga as well as south of the second-largest ring fort Löt (Näsman ig84, p. 127). The distribution of Baltic brooches on Öland seems to be connected with settlements, as only one has as yet been found in a grave. Settlers and raiders from the east and south Baltic regions were probably common on Öland in the Migration Period, despite the fact that a number of single brooches is all the proof we have. There is, however, a folk tradition connected with the ring fort of Gråborg saying that the fort was once owned by a king Bugislev. He may tentativdy be identified with the historical person Burislef, son of the i2th century Swedish king Sverker the old in his second marriage to Rikissa, daughter of king Boleslev III of Poland (Stenberger 1933, p. 232, note 2). Even if this tradition may have appeared as a result of a Medieval historical situation, it is not unlikely that a similar situation may have occurred during the Migration Period at Gråborg or one of the other forts of Öland. The island's wealth must have worked like a magnet on warlords of the east and south shores of the Baltic, and the threat of seaborne attack must have been p e r m a n e n t Moreover, from Öland the straits of Kalmar gave easy access to mainland Sweden. The Baltic interest indicated by the single finds of Baltic Migration Period brooches on Öland does not seem as clear in the Vendel Period and appears very limited in the Viking Period. There is a låter type of Baltic crossbow brooch with poppy capsule knobs, related both to the compact and to the slender type of crossbow brooches from the Migration Period a n d / o r Early Vendel Period. It occurs in Finland and in the east Baltic area in the Viking Period. A fragment of this type of brooch came to light in the ring fort of Eketorp on Öland and another at Hilleshög in Uppland (J.P. Lamm ig87, p. 70 ff). The Hilleshög fragment has interlace d e c o ration which is otherwise unknown on brooches of this type. This caused J.P. Lamm to postulate a possible Scandinavian origin for this type of brooch as he associated it with the two much earlier mould fragments for a crossbow brooch from Helgo. Hilleshög is not very far from Helgo and the låter urban settlement of Birka in Lake Mälaren, and both sites have yielded workshop debris from jewellery casting (Ambrosiani 1994, igg5; K. Lamm 1977). Like Werner, J.P. Lamm indicated that the Swedish workshops may have contributed to the production of Baltic crossbow brooches. Only further finds of workshop material may strengthen or weaken this suggestion. Origin myths current in the Migration Period among a number of Germanic gentes taught that their mythical forebears had once emigrated from the island of Scandza. This idea seems to have a modern archaeological counterpart. PostscrifH Several of the Baltic/Masurian Migration Period crossbow brooches and the Viking Period ones have very distinct knobs in the shape of large poppy capsules, a feature which may originally have been inspired by provincial RoFomvännen gg (2004)

282

Bente M a g n u s Haseloff, G. 1981. Die germanische Tierornamentik der \'ölkerwanderungszeil. Berlin. Herschend, F. 1980. Myntat och omynlat guld. 'Ivå studier i öländska guldfynd. Uppsala. Hoilund Nielsen, K. 1987. '/.ur Chronologie der jungeren germanischen Eisenzeit auf Bomholm. Acla Archaeologica 57. Copenhagen. 1999. Female gravegoods of southern and eastern Seandinavia from the late Germanic Iron Age or Vendel Period. Hines, J. et al. (eds). The Pace of Change. Studies in Early Medieval Chronology. Cardiff Studies in Archaeology. Oxford. - 2000. The political Geography of Sixth- and Seventh Century Southern and Eastern Seandinavia on the Basis of Material Culture. Kazakevicius, V. et al. (eds). Archaeologica Hallien \. Vilnius, Jörgensen, L.& Nörgård Jörgensen, A. 1997. N0rre Sandegård Vest. A Cemetery from lhe 6—Sth Centuries on Bornholm. Nordiske Fortidsminder Serie B, vol. 14. Copenhagen. Kazakevicius, V. 1983. A rare animal-headed crossbow brooch from Plinkaigalis in Lithuania. Fornvännen 78. Kivikoski, E. 1973. Die Eisenzeit Einnkmds. Neuausgabe. Finnische Altertumsgesdlschaft Helsinki. Koch, U. 1999. Nordeuropäisches Fundmaterial in Gräbern Suddeulschlands rechts dei Rhein. von Freeden et al. (eds). Völker an Nord- und Oslesee und die Franken. Römisch-Germanische Kommission. Reiss-Museum Mannheim. Bonn. Kriiger B. (ed.). 1986. Die Germanen. Geschichte und Kullur der germanischen Slämme in Mitteleuropa II. Berlin. Lamm, J.P. 1987. Konsthantverk och kommunikation över Östersjön under järnåldern. Skepp nrb Smide. Artiklar tillägnade Ver Lundström. Statens Sjöhistoriska Museum. Stockholm. Lamm, K. 1977. Early Medieval Metalworking on H d g ö in Central Sweden. Oddy, W.A.(ed.). Asjrects of early metallurgy. British Museum. London. Lundström, A. 1972. Relief Brooches. Introduction to Form element and Variation. Holmqvist, W. et al. (eds). Excavations al Helgii IV. Workshop Earl I. KVHAA. Stockholm. Menghin, W. 1990. Eruhgesehichle Bayerns. Stuttgart. Nerman, B. 1935. Die Völkerwanderungszeit Gotlands. KVHAA. Stockholm. 196g. Die Vendelzeit Gotlands. Tafeln. KVHAA. Stockholm. Nowakowski, W. 1998. Die Funde der römischen Kaiserzeit nnd der Viilki-rivanderungszeil aus Masuren. Bestandskataloge / Museum fiir Vor- und Fruhgeschichte, Band (i. Berlin. Näsman, U. 1984. Zwei Relieffibeln von der Insel Öland. Ihaehislorische Zeitschrift 1. Berlin 8c New York.

m a n b r o o c h e s with o n i o n s h a p e d k n o b s (Nowakowski i g g 8 , p . 5 5 ) . T h i s m a k e s o n e w o n d e r w h e t h e r the m e n (or w o m e n ) who wore them w e r e t h o u g h t to possess e s o t e r i c a n d / o r m e d i cal k n o w l e d g e . P o p p y c o m e s in m a n y v a r i a n t s a n d t h e o p i u m p o p p y was well k n o w n in antiquity. Pliny, in his b o o k s o n N a t u r a l H i s t o r y ( b o o k i g , p . 167—16g; b o o k 2 0 , p . 198—206), discusses t h e d i f f e r e n t species of p o p p y a n d is r a t h e r explicit a.s r e g a r d s t h e p r o d u c t i o n of o p i u m a n d its m e d i c a l u s e . A cknowledgements 1 a m i n d e b t e d t o Dr. K a r e n H o i l u n d N i e l s e n , Ä r h u s a n d Dr. K e n t A n d e r s s o n , S t o c k h o l m for reading the manuscript and comnienting u p o n it, a n d to A u d r o n é Bliujiené, Vilnius, for letting m e read h e r manuscript on the Lithuanian Migration Period.

References Adams, B. 8cJackson, D. 1990. The Anglo-Saxon Cemetery at Wakerley, Norlhaniplonshire. Excavations by Mr. D. Jackson, 1968-9. NorthamptonshireArchaeology 22 (1988-89). Northampton. Ambrosiani, B. 8c Eriksson, B. 1994. Birka vikingastaden 4. Sveriges radios förlag. Stockholm. 1995. Birka vikingastaden 5. Sveriges radios förlag. Stockholm. Balodis, F. 1940. Det äldsta iMlkmd. Uppsala. Behm-Blancke, G. 1959. Angdsächsischc-r Import in Thuringen. Ausgrabungen nnd Funde. Nachrichtenblatt fur Vor- undEruhgesehichle4. Berlin. Beskow-Sjöberg, M. 1987. Olands järnäUersgravjäll I. Statens Historiska Museum & Riksantikvarieämbetet. Stockholm. Bitner-Wröblewska, A. 2000. Between Guronia and Bavaria. Animal-head brooches resulting from long-distance connections during Migration Period. Kazakevicius, V. et al. (eds) Archaeologia Hallit11. |. Lithuanian Institute of History. Vilnius. — 2001. From Samland to Rogaland. East-Wesl connections in Ihe Baltic basin during Ihe Early Mlgiiition Period. Warszawa. Bliujenc, A. Ms. Origin of the sixth century Baltic zoomorphic artefacts and the development of a Baltic animal style. Droberjar, E. 2002. Encyklojmlie Rimské a Germanske Archeologii- v Ceeliäili a na Moravf. Prague. Evison.V. 1987. Dover. The Buekland Anglo-Saxon Cemetery,aArchaeological Report 3. Historie Buildings and Monuments Gommission for England. London. fomvånntn gg (2004)

Brooches on the move in Migration Period 1984. Glas och handel i senromersk tid och folkvandringstid. En studie kring glas jrän Eketorji-ll, Öland, Sverige. Ann 5. Uppsala. Nörgård Jörgensen, A. 1999. A peaceful discussion of a martial topic: the chronology of the Scandinavian weapon graves. Hines,J. et al. (eds), lhe Pace of Change. Slndies in Early Medieval Chronology. Cardiff Studies tn Archaeology. Oxford. Parfitt, K. 8c Brugmann, B. 1997. The Anglo-Saxon cemetery on Mill Hill, Deal, Kent. Society for Medieval Archaeology, Monograph Series 14. London. Pliny. (1950). Natural History, vol. V. The Loeb Classical Library. London. (1951). Natural History, vol. VI. T h e Loeb Classical Library. London. Rundkvist, M. 2003. Barshalder 1. A cemelery in Grötlingbo and Fide parishes. Golland, Sweden, c. Al) r— 1100. Excavations and finds 1826— i g y 1. Stockholm Archaeological Reports 40. Dept of Archaeology, University of Stockholm. Sage, W. 1975. Das Reihengräberfeld von Altenerding. Ausgralmngen in Deutschland 2. Mainz. Schulze-Dörrlamm, M. 1986. Romanisch oder Germanisch? Untersuchungen zu den Armbrnstund Biigdknopffibdn des 5. und 6. Jahrhunderts 11. Chr. aus den Gebieten westlich des Rheins und siidlich der Donau. Jahrbuch des Römisch-Germanischen Zentralmuseums Mainz 33:2. Mainz. Shetelig, H. 1910. Smaa spaender fra folkevandringstiden. Oldtiden. Tidsskrift for norsk forhislorie 1. Stavanger. Stenberger, M. 1933. Öland under äldre järnåldern. KVHAA. Stockholm. Straume, E.M. 1995. On the question of exogamy as a basis for distribution in the Germanic Iron Age. Resi, H.G. (ed.) Produksjon og samfunn. Beretning fra det 2. Nordiske jernaldersymfmsium på Grånavolkn. Varia 30. Oslo. -

283

Ti 1 1 1 by, J. 1996. The Anglo-Saxon Cemetery al Emping ham II, Rutland. Oxbow Monograph 70. Oxford. Vierck, H. 1970. Zum Fernverkehr iiber See im 6. J a h r h u n d e r t angesichts angdsächsischer Fibdsätze in Thiiringen. Eine Problemskizze. Hauck, K. et al. Goldbrakteaten aus Sievern. Miinsterische Mittdalterschriften 1. Miinchen. Waller, j . 1996. Dräktnålar oeh dräktskick i Osira Mälardalen. Aun 23. Uppsala. Wamers, E. 2000. Der Runenreij aus Aukn. Archaeologische Reihe 17. Museum fiir Vor- nnd Friihgeschichte - Archäologisches Museum. Frankfurt am Main. Wegrac-ns, E. 1976. The Oland ring-jorls. Borg, K et al.

Eketorp, Fortification and Setdement on Oland,
Sweden. The Monument. KVHAA. Stockholm. Werner, J. 1951. Eine ostpreussische Biigdfibd aus der Hennegau. Germania 29. Berlin. 1955. Biigdfibeln des 6. Jahrhunderts aus Domburg, Zeeland. Berichten van de Rijksdiensl voor hel oudheidkundig Bodemonderzoek 6. Amersfoort 1970. Zur Verbreitung fnihgeschichllicher Mctallarbeiten (Werkstatt - Wanderwerk - Handel - Familienbindung). Antikvariskt Arkiv 38. Early Medieval Studies 1. KVHAA. Stockholm. Äberg, N. 1919. Osljneussen in der Völkerwanderungszeit. Uppsala. 1923. Kalmar läns förhistoria. Siidrn Kalmar lån I. Kalmar. 1933. Den historiska relationen mellan folkvnnil ringslid och vendeltid. KVHAA. Stockholm. Arsrapjmrl 1968. Helgo och metallundeisökningarna. Statens historiska museum. Stockholm. Olands pirnåldersgravfält vols 1—4, 1987-2001. Eds Sjöberg, Hagberg 8c Rasch. Statens historiska museum X: Riksantikvarieämbetet Stockholm. Orsnes, M. 1961. Form og stil i Sydskandinaviens yngre jernabler. Nationalmuseet. Copenhagen.

Sammanfattning
A r t i k k e l e n tar sitt u t g å n g s p u n k t i e t f u n n av e n baltisk a r m b r o s t f i b u l a i n n e i G r å b o r g , Ö l a n d s st0rste b o r g fra f o l k e v a n d r i n g s t i d e n . Et p a r f r a g m e n t e r av e n s t o p e f o r m til slike s p e n n e r b l e f u n n e t veldig tidlig i u t g r a v n i n g e n e p å H d g ö , o g j o a c h i m W e r n e r k o p l e t disse to funn e n e s a m m e n o g h e v d e t i e n artikkel at slike baltiske fibler var p r o d u s e r t p å H d g ö i folkev a n d r i n g s t i d e n . D a e n l i k n e n d e s p e n n e ble f u n n e t i en kvinnegrav p å et rekkegravfelt i O b e r bayern, og graven dessuten inneholdt fremmed e g j e n s t a n d e r som m a n m e n t e p e k t e m o t Skandinavia (så s o m d r a k t n å l e r , e n d r a k t s p e n n e av typ e n Ozingell o g e n h a l s r i n g ) , var g r u n n e n lagt for å gi k v i n n e n e n svensk etnisk t i l h ö r i g h e t . A r t i k k e l e n g j e n n o m g å r a r g u m e n t a s j o n e n for d e n n e slutningen og awiser den, m e n påpeker p r o b l e m e n e m e d å tolke f r e m m e d e gjenstander, isaer d r a k t s p e n n e r i kvinnegraver, s o m etniskt- m a r k ö r e r . Fornvännen gg (2004)