D bracteate designs on the back side of a relief brooch from Hällan, Jättendal parish, Hälsingland Rundkvist, Martin Fornvännen 2004(99):3, s. [177]-182 http://kulturarvsdata.se/raa/fornvannen/html/2004_177 Ingår i: samla.raa.se

D bracteate designs on the back side of a relief brooch from Hällan, Jättendal parish, Hälsingland
By Martin Rundkvist

Rundkvist, M. 2004. D bracteate designs 011 lhe back side of a relief brooch from Hällan, jättendal parish, Hälsingland. (D-braktealfigurer på baksidan av ett reliefspänne från Hällan i Jättendal sn, Hälsingland). Fornvännen gg. Stockholm. In 1841 and 1851 two Late Migration Period relief brooches (SHM 934 & 1774) were found at the farmstead of Hällan in Jättendal parish on the coast of Hälsingland in central Sweden. These brooches have been discussed, illustrated, or at least mentioned in most major studies of Scandinavian animal art. Onh in 1971 did Birgit Arrhenius note that there is animal art 011 the back side of the larger brooch too. She identified lhe motif as similar to that of certain gold bracteates of Monlelius's type D. This paper aims lo publish and discuss this link between late relief brooches and late gold bracteates.
Marlin Rundkvisl. Lakegatan 12. .S7-.-/J j 41 Saltsjöbaden, Sweden

arador@algonet.se

The farmstead of Hällan is situated in Jättendal parish on the coast of Hälsingland in central Sweden, near the 62nd paralld. Hällan is located at the foot of Hällansberget Hill, 011 the northern shore ofT.akeJättendalssjön (Brink 1990, pp. 292, 293, 298). This lake was a sheltered inlet of the Baltic Sea until the mid-lst millennium AD, but post-glacial shore displacement bas since lifted its surface to 15 m above sea level. The name Hällan is bdieved to incorporate the same linguistic root as e.g. H d g ö , meaning »holy» or indicating a place where violence is forbidden (Calissendorf 1964; Brink 1984, p. 68). In the mid-i 9th century two relief brooches (SHM 934 & 1774) were found at Hällan. In 1841 the farmstead was moved to a new site 400 ells (c. 240 m) east of the former one. The larger brooch was found at the new site in 1841 by nämndeman]ohA\\ Persson, an ell (c. 60 cm) beneath the ground surface. The spöt (registered ancient monument Raä Jättendal 221) was

bdieved to have been a »churchyard». The smaller brooch was found in 1851 at the abandoned site (Raä Jättendal 33) by farmer Erik Johansson, eight inches (c. 20 cm) beneath the ground surface.Judging from numerous other finds in the immediate vicinity, both sites appear to be part of a si ring of prehistoric cemeteries and settlements along the southern flanks of Hällansberget Hill (Hildebrand 1869, pp. 284-286). Both brooches have cast decoration in Salitfs (1935, pp. 233-234, 242) Style I, Haseloffs (1981, pp. 204—205) phase D, and are gilded. In terms of their outline shape, both belong to Sj0vold's (1993, pp. 32-35) type A-2a. Both have discs on their bows. The larger brooch (fig. 1, SHM 934) is made of silver wilh fine garnet cloisonné and gold filigree decoration. It has löst its head plate and pin and is visibly worn. Arrhenius (1971, p. 110) has suggested that its cloisonné discs may bave been taken from Frankish
Fornvännen gg (2004)

17 8 Martin Rundkvist

-». •

"

_tl

v>;*$ \*>
' V,

*

SV 'v4V
\

•

*
J^

.

M^^M

t^;
Fig. i. The larger relief brooch from Hällan, Jättendal parish, Hälsingland, central Sweden. Front side. Gilded silver. Preserved length 148.1 mm. Early (ith cenlury AD. Museum of National Antiquities, Stockholm, inv. no. 934. Photograph by Gabriel Hildebrand. Rg. 2. The larger relief brooch Ironi Hällan. Back side. Photograph by Gabriel Hildebrand.

disc brooches and re-used. All five inlay piéces on the brooch's foot are secondarily attached, with disused primary rivet holes visible on the back-side. The pin catch is broken but has rivet holes that indicate that it has once been repaired. The smaller brooch (SHM 1774) is made of bronze and lacks only its pin. Judging from their Style I:D decoration, both brooches were made during the Iinal phase of lhe Migration Period (as established already by Montelius 1912, p. 14; cf. Äberg 1953, pp. 4 5 - 4 6 ) , c. J \ D 500—540, and may be contemporaneous with the earliest finds of the Vendel Period in south-eastern Seandinavia (cf. Näsnian 1984, p. 62 wilh refs; Magnus 199g;
Fomvännen gg (2004)

Hoilund Nielsen 2002; Rundkvist 2003 pp. SO-SOThese brooches were found and published very early (Montelius 18(19, pl. 6 fig. 2; 1874, fig. 442; Hildebrand 1869, fig. 10) and have been discussed, illustrated, or at least mentioned in most major studies of Scandinavian animal art. That is, their outline shape and the decoration of their front sides have been discussed. Only in 1971 did Arrhenius (1971, pp. 101, 110) publish a photograph of part of the larger brooch's back side, noting that there is animal art to be found ihere too, but of a more unusual kind (fig. 2). She suggested that il bad been made with copper or brass hamme-

D bracteate designs 179 red into grooves cut in the brooch's surface. Arrhenius identified the motif as similar to that of certain gold bracteates of Montdius's (1869) type D. There is no decoration on the back-side of the smaller brooch from Hällan. Technique The larger brooch has three animals on its back side, one on each of the foot plate's disc-shaped protrusions. In November 2000 the goldsmith and metal conservalor Hubert Hydman of the Museum of National Antiquities in Stockholm kindly studied the brooch at my request and reported the following conclusions. The brooch consists of an alloy with 80—90% silver, most of the remainder being copper. The outlines of the three animals have been incised into the back surface of the brooch with a knife or similar edged tool. The surfaces inside these outlines have been roughened in a criss-cross pattern with a chisel of little more than a millimetre's width. The roughened surfaces have then been covered with an amalgam of powdered gold and mercury, and the brooch heated until the mercury evaporated, leaving a gilded surface. The artisan has not managed to apply the amalgam perfectly within the incised outlines. On one hand, the animals have not been completely gilded; on the other, gilding has been smudged outside the lines but has then been carefully removed. It is possible to apply gilding to an object repeatedly in this männer without damaging earlier work. Thus it cannot be ruled out that the designs on the back-side were added to the brooch by a second artisan. They were, however, added before the holes for the secondary garnet fittings were drilled. Thus the simplest interpretation is that both sides of the brooch were decorated by the same artisan within a short interval. Animal designs The designs of the three animals (figs 3—5) are, as Arrhenius observed, copied from Montelius (1869) type D gold bracteates. All three are similar to Axboe et al.'s (1989, pp. 29—31,53—55) basic design 5, »the bizarre version of the quadruped monster with head hanging forward» (fig. 6). Thirteen gold bracteates with this de

Fig. 3. The animal on the left-hand disc of the larger brooch from Hällan. Diameter 25.3 mm. Traced by the author after a photograph by Gabriel Hildebrand.

Fig. 4. The animal on the middle disc of the larger brooch from Hällan. Diameter 32.1 mm. Traced by the author after a photograph by Gabriel Hildebrand.

Fig. 5. The animal on lhe right-hand disc of the larger brooch from Hällan. Diameter 26.9 mm. Traced by lhe author after a photograph by Gabriel Hildebrand. Fornvännen gg (2004)

18o Martin Rundkvisl Discussion Haseloff (1970) and Bakka (1973, p. 73) have shown that early Style I relief brooches and early gold bracteates (third quarter of the Sth century AD, cf. Näsnian 1984, Axboe 1999) share stylistic traits and motifs. This indicates contemporaneity and that the same artisans made both classes of objects (cf. Arrhenius 1975, p. 102). The larger brooch from Hällan, perhaps unsurprisingly, shows the same for the late brooches and bracteates (first half of the 6th century AD). The combination of three slightly different D bracteate designs on the brooch give a unique insight into the repertoire of a bracteate master at one point in time. Such knowledge cannot be had from the bracteate hoards, whose contents may have been collected during long time spans, nor from punch linked bracteates, which may have been made several decades apart. The gcographic distribution of the closest bracteate parallels to the designs on the brooch underline the intimate connections between central Sweden and Norway in the Migration Period (cf. Ramqvist 1991, pp. 66—69 with refs; 1992, pp. 225—226). No gold bracteates have been found in Hälsingland or adjoining provinces (Gästrikland, Dalarna, Härjedalen, Medelpad). Norway is also one of the main find areas for phase D of Style I, probably because Style II and the general Vendel Period type assemblage was not accepted in the western part of that country until two or three decades after they had replaced Style I and the general Migration Period assemblage in south-eastern Seandinavia. Acknowkdgements Heartfell thanks to Hubert Hydman for the technical assessment and to Mörten Axboe for helpful suggestions.

Fig. 6. Type D gold bracteate, Axboe et al. 198g basic design 5. Unknown provenance. National Museum of Denmark, inv. no. 8661. IK 551. Diameter 26.6 mm. Drawing by M. Lange (Axboe et al. 198g, p. 29, fig. 21).

sign are known: eight from southernmost Norway, one from northern Jutland, and four without provenance. All three animals on the brooch possess the same anatomical parts: upper and lower jaw, a curved neck, a curved body, a front leg with a simple paw and a small spur, and a hind leg with a large composite foot. Two of the three hind feet consist of a simple spurred paw and a protruding heel, while the right-hand animal lacks the heel. None of the animals have a visible front haunch. Only the middle animal has a clearly visible hind haunch. The middle animal is the largest and drawn in the greatest detail. The body and hind parts are placed in the same männer on all three. The front leg is placed in the same männer on the left and right animals, but differentfy on the middle one. The jaws ofall three animals interlace differentfy with the front leg. Only the middle animal has its upper jaw bent under the lowerjaw like the design D5 gold bracteates.

Fornvännen gg (2004)

D bracteate designs References Äberg, N. 1953. Den historiska relationen mellan folkvandrings! id ocb vendeltid. KVHAA:s handlingar 82. Stockholm. Arrhenius, B. 1971. Granatschmuck und Gemmen aus nordischen Funden des friihen Mittelalters. Acta Universitatis Stockholmiensis. Studies in North-European Archaeology Series B. Department of Archaeology, University of Stockholm. ig75- Die technischen Voraussetzungen fiir die Entwicklung der germanischen Tierornamentik. Fruhmittelalterliche Studien 9. Munster.

181

chronology. Cardiff Studies in Archaeology. Oxford. H0ihmd Nielsen, K. 2002. Den anden germanske dyrestil - og dens vej til Norge. Era haug ok heiöni. Tidsskrift for Rogalands arkeologiske fmening 2002:2. Arkeologisk museum i Stavanger. Magnus, B. 1999. T h e assemblage from Hade in Gästrikland and its relevance for the chronology of the late Migration Period in eastern Sweden. In Hines et al. iggg. Montelius, O. 186g. Från jernåldern. Stockholm. 1874. Svenska fornsaker. Sveriges forntid. Försök lill framställning af den svenska fornforskningens resultat. Vol. 2, Jernåldern. Stockholm. i g i 2 . Mästerstycken i Statens historiska niiiseiiin. Stockholm. Näsman, U. 1984. Zwei Relieffibeln von der Insel Oland. Praehistorische Zeitschrift 59:1. Berlin. Ramqvist, P.H. i g g i . Uber ökonomische und soziopolitische Beziehungen der Gesellschaften der Nordischen Völkerwanderungszeit Fruhmittelalterliche Studien 25. Munster. 1998. Högom. The excavations 1 g4g-ig84. Archaeology and environment 13. Umeå, Stockholm, »Kiel. Rundkvist, M. 2003. Barshalder 2. Studies of Eate Iron Age Golland. University of Stockholm. Sali n, B. 1935. Die altgerma n ische Thierorna men tik. Typologische Studie iilier germanische Metallgegenslände aus dem IV. bis IX. fahrhundert, nebst einer Studie uber irisch» Ornamentik. Neue Auflage. Stockholm. Sj0vold, Thorleif. 1993. The Scandinavian relief Irrooches of Ihe Migration Period. An attempt at a new classificalion. Norske oldfunn 15. Institutt for arkeologi, kunsthistorie og numisinatikk. Oldsaksamlingen. Olso.

j\xboe, M. iggg. T h e chronology of the Scandinavian gold bracteates. In Hines et al. iggg. Axboe, M. et al. ig8g. Die Goldbrakteaten der Vöbkerwnnderungszeil y . i , Ikonographischer Katalog (IK3, Text). Miinstersche Mittdalter-Schriften, 2 4 / 3 , 1 . Miinchen. Bakka, E. ig73- Goldbrakteaten in norwegischen Grabfunden. Datierungsfragen. Eriihmillebtllerlicbe Studien 7. Munster. Brink, S. 1 g84. Ortnamn i Hälsingland. Stockholm. 1990. Sockenlnldning och sockennamn. Studier i äldre territoriell indelning i Norden. Acta Academiae Regiae Gustavi Adolphi 57. Studier till en svensk ortnamnsallas 14. Uppsala. Calissendorf, K igÖ4. H d g ö . Namn och bygdyi. Uppsala. Haseloff, G. 1970, Goldbrakteaten - Goldblattkreuze. Neue Ausgrabungen und Forschungen in Niedersachsen 5. Hildesheim. 1981. Die germanische Tierornamentik der Völkerwanderungszeit. Studien zu Satin's Stil I. Vorgeschichtliche Forschungen 1 7. Berlin.

Hildebrand, H. 186g. Den äldre jernåldern i Norrland. Antiqvarisk Tidskrift för Sverige 2. KVHAA. Stockholm. Hines, J.; H0ilund Nielsen, K & Siegmund, F. (eds). iggg. The pace of change. Studies in Early-Medieval

Fornvännen gg (2004)

182 Martin Rundkvist Sammanfattning 1841 och 1851 hittades två folkvandringstida reliefspännen vid Hällan i Jättendal sn, Hälsingland (SHM 934 & 1774)- Fyndplatserna ligger ca. 240 m isär i en rik fornlämningsmilj ö på norra stranden avjättendalssjön nedanför Hällansberget. Sjön var en skyddad vik av Östersjön till mitten av första årtusendet e.tKr. men har nu sin yta 15 m över havsnivån. Spännena från Hällan är dekorerade i sen Stil I, Haseloffs fas D, och har ända sedan de hittades blivit flitigt diskuterade inom djurstilsforskningen. De tillhör folkvandringstidens slutskede och kan vara samtida med den äldsta Stil II i sydöstra Skandinavien. Först 1971 konstaterade dock Birgit Arrhenius att det större spännet har djurornamentik även på baksidan, tre kompositioner som liknar vissa guldbrakteater. Föreliggande uppsats avser att publicera och kort diskutera dessa brakteatornament. Brakteatdjuren är placerade bakpå de tre runda plattorna på spännets fot och utförda med brännförgyllning inom graverade begränsningslinjer. Fastän inte identiska är de alla snarlika den grupp av D-brakteater som Axboe et al. kallar komposition 5, »den bisarra versionen av fyrfotemonstret med hängande huvud». Tretton sådana guldbrakteater är kända: åtta från sydligaste Norge, en från norraJylland och tyra utan känd fyndort. Det är känt sedan tidigare att tidiga guldbrakteater och tidiga reliefspännen i Stil I uppvisar stora likheter i ornamentiken. Det betyder förmodligen att de är samtida och att samma hantverkare tillverkade båda föremålstyperna. Det stora spännet från Hällan visar att samma sak gäller även för de sena reliefspännena och sena guldbrakteaterna. Att de tre djuren är lite olika illustrerar på ett unikt sätt hur stor samtida fonnrepertoar en brakteatmästare kunde ha. Varken brakteatskatter eller punskopplade brakteater uppvisar en samtidighet och individualitet med samma säkerhet. Ur geografisk synvinkel illustrerar spännena från Hällan folkvandringstidens täta kontakter mellan Norge och södra Norrland. Norge är ett av hemområdena både för D-brakteater och för sen Stil I i allmänhet, förmodligen beroende på att man i landets västra delar var sen att acceptera Stil II. Brakteatdjuren från Hällan står ensamma i Hälsingland, ett landskap utan fynd av guldbrakteater. Inga har heller påträffats i de angränsande landskapen Gästrikland, Dalarna, Härjedalen eller Medelpad.

Forrwänrwn gg (2004)