Kaupang, men and women

Ellen Høigård Hofseth (1999) points to the relatively high number of female graves at the Viking Age trading-site at Kaupang in Vestfold. Indeed at first sight, the relative number of female graves looks extraordinary, when compared to most regions in Norway. However, there are some chronological and source critical aspects worth exploring.

In Vestfold as a whole, female graves account for a quarter of the gendered graves – similar to many other coastal districts (cf Dommasnes 1982; Hofseth 1999). But female graves are relatively much more common in the region in the 9th century than in the 10th – 34 % of the total number of gendered graves against 13 %. This pattern is repeated in a number of other coastal districts (Dommasnes 1982:81-82).

The numbers for Kaupang exceed these figures, in the 9th as well as in the 10th century. Of the 41 datable female graves, 22 can be dated to the 9th century, and only 12 to the 10th. As for the 62 datable male graves from the Kaupang cemeteries, there are 16 from the 9th century and 38 from the 10th. Thus, the female graves comprise 58 % of the gendered graves in the 9th century, against 24 % in the 10th (the graves dated to the period 850-950 cannot account for the difference; of the 15 graves in question, eight are male and seven are female). Even if these are middle values for all the Kaupang cemeteries, the only cemetery with a significant number of datable, gendered graves seem to conform to this pattern. Thus, of 53 gendered, datable graves at Bikjholberget, female graves comprise 50 % of the graves in the 9th century, and 25 % in the 10th century. It is clear from this, that the Kaupang cemeteries seem to have a relative number of female burials that is substantially higher both in the 9th and the 10th century than in the rest of Vestfold.

How can one account for these differences? There are really three different questions to be answered. For starters, why is it that male burials seemingly outnumber female ones in the Vestfold material – and even more so in the Norwegian material as a whole? – even though Vestfold and the coastal districts of Østfold, just across the Oslo fjord, have a higher proportion of female graves than the rest of the country throughout the Viking Age? Where are the women? Secondly, why is it that the number of female graves relative to male graves decreases from the 9th century to the 10th? And, finally, why are there relatively more female graves at Kaupang than in the rest of Vestfold?

The gender distributions from Kaupang and Vestfold are at odds with results from other areas of Scandinavia. Starting with Birka, Gräslund (1980:82) finds that 58 % of the inhumations are female, against 61% of the cremations, although she suggests that the real distribution might be closer to 50-50. Of 113 10th and 11th century graves at Barshalder in Gotland, 37 % were female, 49 % male and 14 % gender neutral (Rundkvist 2003:79). A study of a sample of 76 sexed skeletons from Hedeby concluded that 62 % were male (Sellevold et al. 1984:214). Other South Scandinavian cemeteries also show a predominance of men: Stengade II (53 %), Kaagaarden (63 %) and Bogøvej (61 %), while there are relatively more women at Lejre (61 %) and Hesselbjerg (58 %) (Sellevold et al. 1984:214-215; Bennike 1994:169). With the exception of Birka and Barshalder, these studies are based on skeletal material – what we have here, is sexed burials, not gendered ones. Not only Kaupang and the rest of the material from Vestfold, but the Norwegian material as such – based as it is on the presence of gender specific artefacts in the graves – is biased relative to the South Scandinavian one, and the two cannot really be compared on equal terms. It is a telling fact that Per Holck’s anthropological analysis of Late Iron Age cremations in Southeast Norway, although only 42 burials from the period could be ascribed to either sex, concluded that 62 % of the cremation burials actually were female (Holck 1986, catalogue).

At Kaupang, the gendered burials account for only a part of the dated burials, while, more importantly, the dated burials themselves only account for a minor part of the total number of burials. Thus, the gendered graves belong to a rather exclusive category, and one must be careful when assuming that gendered graves equal sexed graves. Herein lies the answer to our first as well as to our third question. In principle, one would expect a more or less equal distribution of male and female sexed graves. The reason why we do not find an equal distribution in Vestfold, probably has to do with the difference betwen sexed and gendered graves, i.e. more males than females were buried with gender-specific artefacts, or, more likely, more males than females were buried with gender-specific artefacts that are preserved and can be recovered by archaeologists. There is a clear correlation between areas with a substantial number of professionally excavated graves and those with a relatively high proportion of female burials. This is evident in those parts of Vestfold where only few professional excavations have taken place, and the proportion of male graves is very high: After all, a corroded sword is easier to notice when ploughing than the remains of an oval bronze brooch or a few beads.

So, when we do find that 58 % (9th century) and 24 % (10th century) respectively of the dated graves at Kaupang are female, these high figures are due mainly to the substantial number of graves being archaeologically investigated at Kaupang. As for the relative decrease in the gendered female graves in the 10th century, it would be tempting to attribute this to an influx of males at Kaupang in the late period. However, the decrease can be traced also in Vestfold as a whole, as in most of Norway (Dommasnes 1982:81-83). A more likely explanation would seem to be either a real decrease in gendered female burials compared to male, or a change in the way female graves were gendered; the latter could be caused for instance by the partial and gradual abandonment of oval brooches as part of the female dress in the 10th century, under influence from Frankish and/or Byzantine single-brooch dresses (Hedeager Krag 1994; cf Ingstad 1999:243-244).). It is, furthermore, a distinct possibility that female dress customs were rather more varied at any point through the Viking Age than often assumed (Martens 1969:88; see also Blindheim 1947:117-118). A survey of the Viking Age evidence from Denmark supports that the poorest and the richest women, as indicated by grave furnishings, did not wear oval brooches (Hedeager Madsen 1990:104).

Litterature
Bennike, Pia 1994. An anthropological study of the skeletal remains of vikings from Langeland. In Grøn, Ole, Anne Hedeager Krag & Pia Bennike, Vikingetidsgravpladser på Langeland, pp 166-197. Rudkøbing.
Blindheim, Charlotte 1947. Drakt og smykker. Studier i jernalderens drakthistorie i Norden. Viking 1947, pp 1-139. Oslo.
Dommasnes, Liv Helga 1982. Late Iron Age in Western Norway. Female Roles and Ranks as Deduced from an Analysis of Burial Customs. Norwegian Archaeological Review, vol 15, nos 1-2, pp 71-84. Oslo.
Gräslund, Anne-Sofie 1980. Birka IV. The Burial Customs. A study of the graves on Björkö. Stockholm.
Hedeager Madsen (Krag), Anne 1990. Women’s dress in the Viking Period in Denmark, based on the tortoise brooches and textile remains. In Walton, Penelope and John P. Wild (eds), Textiles in Northern Archaeology, pp 101-106. NESAT III Textile Symposium in York. London.
Hedeager Krag, Anne 1994. Dragtudviklingen fra 8.-10. årh. e.Kr. i Sydskandinavien – med udgangspunkt i skålformede spænder. Lag 5, pp 7-114. Højbjerg.
Hofseth, Ellen Høigård 1999. Historien bak handelskvinnen på Kaupang. Kvinnegraver fra vikingtid langs Vestfoldkysten. Viking 62, pp 101-129. Oslo.
Holck, Per 1986. Cremated bones. A medical-anthropological study of an archaeological material on cremation burials. Oslo.
Ingstad, Anne Stine 1999. Tekstilene. In Blindheim, Charlotte et al 1999, Kaupang-funnene II. Gravplassene i Bikjholbergene/Lamøya. Undersøkelsene 1950-1957. Del B. Oldsaksformer. Kulturhistorisk tilbakeblikk. Del C. Tekstilene, pp 217-272. Norske Oldfunn XIX. Oslo.
Martens, Irmelin 1969. Gravfeltet på By i Løten, Hedmark. Universitetets Oldsaksamling Årbok 1965-1966, pp 11-148. Oslo.
Rundkvist, Martin 2003. Barshalder 1. A cemetery in Grötlingbo and Fide parishes, Gotland, Sweden, c. AD 1-1100. Excavations and finds 1826-1971. Stockholm.
Sellevold, Berit J. et al. 1984. Iron Age Man in Denmark. Copenhagen.