Michèle Hayeur Smith (Textile Analysis); Greg Rebis (Illustration of the Lady in Blue.)
In 2014 an anomalous jar in the collections of the National Museum of Iceland was found to contain formaldyhyde and the soft tissue and skeletal remains of a partial face and cheek of an unidentified female. After a review of the museum’s documentation it was demonstrated that these were the remains of a woman from the Viking Age Ketilsstaðir burial in North Eastern Iceland initially discovered in 1938 by road builders. Immediately questions about preservation and the burial itself brought the whole matter to life once again and a full-scale project was undertaken between 2014 and 2015, looking at all of the preserved remains to reconstruct the life, death, and identity of this early settler. This culminated into an exhibit called Bláklædda Konan, Ný Rannsókn á Fornu Kumli (Bundled up in Blue, the re-investigation of a Viking grave) at the National Museum of Iceland in May 2015 and involved researchers from the University of Iceland, the National Museum of Iceland, Brown University, Durham University,the National Museum of Denmark, and De Code Genetics.
As with many Viking Age burial sites in Iceland, the grave was located quite far from any modern farmhouses, but potentially near the boundaries of earlier centres of occupation. In this case, the site was located 300 m north of the abandoned farm of Litlu-Ketilsstaðir, and 2.5 km north of Ketilsstaðir (Friðriksson, 2013, 500). Shortly after road workers came upon the burial, they contacted Mattiás Þórðarsson, the head of the National Museum at the time, who carried out excavations at the site between 1938 and 1942 (Eldjárn, 1958, 180; Friðriksson, 2013, 500; Einarsdóttir, 2015, 17). The burial was thought to have been surrounded by a small circular wall of about 18.5 cm wide (Friðriksson, 2013, 500). Within this enclosure was the grave of a female, resting on her left side in a flexed position. Her knees and hips were bent and the left side of her face touched one the two oval brooches that would originally have been located on her breasts. Other grave goods included a trefoil brooch, 42 beads, textile fragments, two whetstones, bone fragments of either a comb or knife handle, a spindle whorl, and a stone of unusual shape which proved to be a fragment of chalcedony (Eldjarn, 1958; Friðriksson, 2013, 500).
A version of the following text by Michèle Hayeur Smith, was published in the exhibit’s catalogue, (2015) Bláklædda Konan, Ný Rannsókn á Fornu Kumli (Bundled up in Blue, the re-investigation of a Viking grave), Publications of the National Museum of Iceland; pp.25-43.
Dress, Jewellery and Textiles
In the early 10th century, a young woman was buried close to the farm of Ketilsstaðir in Northeastern Iceland. All that remained of her body when found were her lower jaw, parts of the maxilla, her cheek, and a few fragmented bones. Her grave goods, a little more abundant, were those of a relatively well-to-do Viking Age woman. It was thanks to her oval brooches that we have as much information about her as we do today.
Oval brooches were made of copper alloy, gilded and embellished with silver wire and bosses (Hayeur Smith, 2004).Corroding copper releases metal ions that can contribute significantly to the preservation of organic materials, such as textiles, skin, or hair. This is what happened in the grave at Ketilsstaðir, as it would seem that her cheek lay upon the left brooch, while the other had slipped and was resting on her right upper arm. In areas where they are concentrated, these corrosion products become toxic enough to inhibit micro-organisms from breaking down fibres or other organic material, allowing actual tissues, in rare instances, to be preserved for centuries (Chen et al, 1998) .
As copper or iron oxides and other corrosion products form on the surfaces of objects such as brooches, they can also preserve details of organic materials that have actually decomposed entirely. In these cases, the organic components are replaced by mineralised copper or iron compounds. These are called pseudomorphs, and can provide additional details about organic materials that were intentionally placed, or happened to fall, against the metal objects within the burial. In this grave this mineralisation processes preserved textiles, skin, and flesh; some of the textiles on the underside of the oval brooches were mineralised to such an extent that the physical shape of the cloth has been replaced completely by metal corrosion products.
Oval brooches are frequently found in women’s burials across Scandinavia and the Viking world. From archaeological data we know that Viking Age women in Scandinavia wore long garments, their basic outfits consisting of a long sleeved dress of pleated linen, fastened at the neck with a brooch (Hägg, 1974, 108). A pair of oval brooches was worn on the chest, attached to the straps of a sleeveless apron or pinafore – often referred to as a smokkr – that was worn on top of the longer chemise(ibid.). Opinions vary about whether the smokkr (smock or apron) was indeed a short apron, or simply another dress worn over the long undergarment. It is not clear if it was constructed in two panels with open sides, or made of one piece of cloth open to one side( Geijer, 1938) or if it was constructed as a tube (Hägg, 1974, 108). Alternatively, it may have been open in the front, revealing embellished undergarments (Bau, 1981). A string of beads or a pendant was frequently hung between the brooches, along with other useful implements: knives, scissors, and sometimes keys (Hägg, 1974, 29 ; Jesch, 1991, l7).
The oval brooches, worn in pairs, are so widespread in the burials of Scandinavia that they are generally considered the most typical item of female Viking dress, and are so standardised that they have been used as gender identifiers in Viking burials, even where skeletal observations of sex are impossible (Dommasnes, 1982, 73). Their designs are equally standardised; and identical brooch types have been found in areas as remote and distant from one another as Iceland or Russia, wherever the Viking presence was felt. The recovery of pairs of oval brooches on women’s chests in Viking Age graves confirms that this was their intended placement on the body, where they clearly resemble a stylised or accentuated pair of breasts. This visual statement was manifest in their placement directly on the breasts or slightly above them (Wobst, 1977, 328), and was reinforced by their decoration with multiple bosses that make allusion to nipples (frequently nine – a recurring number in Viking Age religious art and mythology). Freyja, Norse goddess of female reproduction and fertility, was frequently given the name of “sow” (or bitch by slanderous Christian missionaries in Iceland), and the similarity between the appearance of these brooches and the belly of a lactating pig or other animal is striking (Hayeur Smith, 2004, 72-73). They could thus be direct reflections of female sexuality or, better still, they may have served the purpose of expressing notions of femininity, fertility and lactation, and/or associations with female divinities such as Freyja, by the hyper-emphasis of stylised female sexual traits. These brooches may have been symbols of married status (not unlike the wedding ring today), as they are not found in all women’s graves but occur regularly with adult women from a restricted social stratum within society, potentially married women who ran independent households (Hayeur Smith, 2004, 72-74; 200, 230). Evidence of wear, repair, and maintenance of individual brooches over considerable periods of time is a clear indication that these objects were not made just to adorn the dead or to mark a woman’s status in the afterlife, but were elements of active adornment during life.
The brooches from Ketilsstaðir belong to the P52 variety, according to the typology established in the early 20th century by Jan Pedersen (1928), who sought to organize Scandinavian archaeological material from the Viking Age. Ingmar Jansson later refined this typology, using coins from associated burials to establish relative dates, and he argued that P52 brooches were a later development in a sequence of oval brooch modifications that spans a long period, from the latter half of the 9th century to the end of the 10th (Jansson, 1985). By AD 1000 oval brooches disappeared completely, perhaps due to the Scandinavians’ conversion to Christianity, in the context of which the pagan associations with these brooches’ iconography may have been suddenly discordant.
Fig.1. Brooches from Ketilsstaðir. (Photo M. Hayeur Smith).
The woman from Ketilsstaðir also had a trefoil brooch, a three-tongued brooch said to be of Scandinavian origin and classified as a P91 type trefoil brooch,(Petersen, 1928: 99; Eldjárn and Friðriksson, 2000, 364; Hayeur Smith, 2004, 29) although its acanthus motifs are continental Frankish in origin. This brooch type was adopted by the Vikings and further modified or embellished with indigenous Scandinavian designs. This may imply that it was made in Scandinavia replicating foreign design, or it could also suggest that it actually originated somewhere within the Frankish kingdom and made its way to Scandinavia, perhaps along with loot collected by Viking raiders. The woman from Ketilsstaðir may have had a family connection with continental Europe. She was actually adorned with approximately 50 beads that would have hung between her two brooches and some of these beads also suggest possible exchange with the southern Mediterranean.
Fig. 2 Ketilsstaðir, 12436 (Photo M. Hayeur Smith)
Amongst the most unusual finds from Ketilsstaðir are the textiles. Textiles, as organic artefacts, rarely survive time, but due to the mineralisation processes described above, several of her dress items have survived to this day.
The textiles worn by women in Scandinavia varied greatly according to social status (and presumably task) – from coarse woollens to fine linens and decorative oriental silks.
Like most Viking women in mainland Scandinavia, the woman from Ketilsstaðir wore a long undershirt or gown. This long undergarment could be represented by a partially mineralised textile that was found adhering to a woven woollen fragment from the grave; alternatively this could represent a lining sewn into the top apron. This mineralised piece (12438-2) (Fig. 3) was sampled and sent to McCrone Associates Inc. in Chicago for fibre analysis. While the fibres themselves had undergone considerable degradation, it was possible to determine that three of these were made of cellulose fibres. Similar mineralised pseudomorphs adhering to the underside of the oval brooches’ copper alloy shells and to the corroded iron pins that held the textiles within the brooch were also noted (See Fig.3 , 4).
Fig. 3 Mineralized linen fibres (12438-2 ), (Photo M. Hayeur Smith).
Fig 4. Mineralised woollen and linen fibres stuck to the underside of the oval brooches. (Photos M. Hayeur Smith).
These fibres suggest that the long undergarment was probably made of linen, as was common.The cloth was a plain weave with z-spun warp yarns and z-spun wefts (Ewing, 2006, 28). Over this long shift she wore an apron or the typical pinafore noted by Geijer (1938) and Hägg (1991) fastened in place by her oval brooches. The apron was made of wool, and decorated at the top by a tablet-woven band with a stripe of a lighter colour. The apron itself, represented by fragments 12438-4a/ and 12438-1, was a woollen diamond twill, z/z spun with an almost equal thread count of 11/10. The presence of z/z spun yarns, which were also noted in the linen undergarment (12438-2), suggests a Norwegian spinning tradition for these garments. Z/z spun textiles persisted in Norway and Gotland from the third century into the late Viking Age. In other parts of Scandinavia, during the Viking Age, textiles were woven with z/s spun yarns ( Bender Jørgensen, 1992). Textiles found in Viking age Norse settlements in the British Isles were woven with z/z spun yarns while the textiles found in the Celtic (non-Norse) settlements were woven with z/s spun yarns as seen in the majority of Europe and the British Isles at this time period.
Fibres from the apron (12438-4a/ and 12438-1) tested positive for indigotin, implying that it had once been dyed blue with woad (Istatis tinctoria). The presence of blue clothing at Ketilsstaðir replicates a common theme in female Viking burials of the time in both Scandinavia and Iceland. In Iceland alone, 65% of female burials included blue textiles, as opposed to only 31% of male burials.
Fig. 5. Top portion of apron including the lighter tablet-woven band and apron cloth that was originally dyed blue. (Photo M. Hayeur Smith).
Fig. 6 Percentages of blue garments identified in male and female burials in Iceland from the Viking Age. (Hayeur Smith, 2014.)
The tablet-woven band (identified as such by Elsa Guðjónsson in 1996) (Guðjónsson, 1996, 21) that decorated the apron had a light-cream-coloured central band flanked by two brown bands on either side (12438-4b). Tablet-woven bands were common features of Viking Age dress, and those added to the garments of wealthy men and women could be quite ornate. The tablet-woven band from Ketilsstaðir was not particularly decorative except for a faint wave pattern, and its presence may reflect instead a particularity in the weaving process. In Iceland, as in other parts of the North Atlantic, tablet-woven bands were frequently used as starting borders for fabric woven on the warp-weighted loom. The ubiquity of this approach as a way of starting woven panels has, in fact, been used in textile analysis as proof of the use of this type of loom, wherever such tablet- woven bands are found in this region (Hoffmann, 1957) .
Fig.7 Microscopic x200 view of the tablet-woven band (12438-4b)
As the tablet-woven band from Ketilsstaðir was a starting border, rather than an appliqué or sewn-on decoration, it means that the rest of the apron was woven and integrated into the band. Unlike the apron, however, the band did not test positive for indigotin, even though fibre analyses conducted by the McCrone group suggested that both were made of wool.
An additional textile fragment was identified among the remains from Ketilsstaðir (12438-3), also bearing the imprint of one of the brooches.
Fig. 8 12438-3 Possible strap attached to the blue apron.
A preliminary analysis of this fragment, which differed considerably from the apron in terms of weave, indicated that it may have been a part of a cloak or shawl of some sort. However, the imprint of the oval brooch on this piece of textile suggests instead that it was found under one of the brooches, and therefore, that it was a part of the apron. It is a woven 2/2 twill, and was confirmed by microscopic analysis to be made of wool, woven more in the style in which Icelandic vaðmal, common a century or two later, was woven, with z/s-spun yarns and a thread count of 9/7 (Hayeur Smith, 2014) This piece also tested positive for indigotin, which supports its identification as one of the straps holding up the apron.
Thus, the textiles, in their initial position during burial, would have been displayed according to the illustration above. The tablet woven band and the diamond twill apron, as well as the linen undershirt, were pinned to the inside of one of the brooches. The strap made of a 2/2 twill was attached to the apron and endured the weight of the brooch as it left its imprint upon the cloth.
Textile samples from both the strap and the apron were sent to the University of Copenhagen for strontium isotope analysis by Dr. Karin Frei (2015) in order to determine the provenance of the wool. The results confirmed that their strontium ratios are comparable to local Icelandic baselines identified in geology, implying that the wool used to make her textiles came from Iceland (Frei, personal communication). The textiles, therefore, reflect the cultural complexity of Iceland at the time of the settlement. The weaving and the sourcing of the wool indicate that Iceland’s settlers spun Icelandic wool with a mixture of z/z and z/s spinning, combining the textile traditions of both Norwegian and Celtic British Isles women, respectively, to produce the cloth found in the Ketilsstaðir grave.
Dating the Ketilstaðir grave, Kevin P. Smith, (Deputy Director and Chief Curator, Haffenreffer Museum of Anthropology, Brown University, Rhode Island) Haffenreffer Museum of Anthropology,Brown University.
Two samples from the woman’s clothing and one from her tooth were sent to Beta Analytic laboratories in Miami, Florida, for AMS radiocarbon dating. AMS (Accelerator Mass Spectrometer) dating is a precise form of radiocarbon dating appropriate for dating textiles and small samples, since it does not require removing more than a couple fragments of thread to get a good sample. Wool is an excellent material for radiocarbon dating because it only grows for a single year before it is shed or plucked. However, it is important to realize that dating the cloth she wore does not tell us when she lived, died, or was buried. It tells us when the wool was gathered from which her garments were made. On the other hand, the dating of the tooth can provide information about when the woman died.
It is not enough to measure the carbon to get the final dating; the basic dates received from the lab must be calibrated according to different environmental factors. For example, the sheep, whose wool was used, may have been fed seaweed as fodder or inland grass and both have different effects on the carbon ratios from which dates are calculated. Marine animals and plants don’t absorb carbon from the atmosphere but from the ocean, which has a lot of fossil carbon floating around. As a result, seaweed collected today off the rocks in Reykjavík could give a radiocarbon age of 250-400 years old, simply because it has absorbed ancient “marine” carbon rather than “new” carbon from the atmosphere. However, our results clearly indicate that the sheep whose wool was used for these garments ate an entirely terrestrial diet uncontaminated with marine carbon. The same issues affect samples taken from human bones or teeth, which need to be calibrated, too. According to isotopic studies reported elsewhere in this publication, this woman’s diet shifted significantly from a primarily terrestrial protein diet to a combined marine and terrestrial diet between the age of five and ten years old (Walser, 2015). The tooth that was analyzed formed when she was around 2-3 years old, and this was considered while calibrating the date.
To date the woman’s clothing, one sample was taken from her apron and another from its strap. The results of these analyses suggest that her clothes were made some time between AD 765 and AD 895. AMS dating of the woman’s tooth implies that she was born before AD 900, suggesting that she came to Iceland as a child between AD 870 and AD 900 (The date obtained on collagen from her tooth (Beta-407926; 1190±30 bp) provided a calibrated age range for this date at two-standard deviations (95.4% probability) of calAD 722-945, with an 89% internal probability that the best estimate for this sample’s age lies within the interval calAD 766-899, with only a 3.5% probability that the true age range for the formation of this tooth collagen lies between calAD 900 and calAD 940). If she arrived as an infant, towards the end of that period, she would have died near the end of the first quarter of the 10th century, ca. AD 915-925.
Fig.9. Calibrated AMS dates on textile samples (12438-3/12438-4a) and tooth collagen from the Ketilsstaðir burial. The top three figures show the raw AMS date in red and the age range determined by calibration in gray. The lower figure plots all three calibrated, statistically identical, AMS dates together, with the red line marking the position of the 871±2 AD “Landnám tephra”.
Acknowledgements: Funding for analysis of the textiles, dye analysis and AMS dating and strontium analysis, comes from the “Weaving Islands of Cloth, Gender, Textiles, and Trade across the North Atlantic, from the Viking Age to the Early Modern Period” (NSF no. 1303898) National Science Foundation, Arctic Social Sciences/Polar Program. Analyses were conducted by the author; Dr Karin Frei, from the Univeristy of Copenhagen; McCrone Associates Inc.; Nathanial Crockett, Department of Textiles, Fashion Merchandizing and Design, University of Rhode Island. AMS dating was carried out by Beta Analytic, Miami Florida.
Bau, F, (1981). “Seler og slæb i vikingetid: Birkas kvindedragt i nyt lys”. Kuml.
Bender Jørgensen. L. (1992). North European Textiles until AD 1000. Århus. Disputats:Aarhus University Press.
Chen, H.L. (1998). “Preservation of Archaeological Textiles Through Fibre Mineralization”. Journal of Archaeological Science (1998) 25, 105-1021.
Dommasnes, L.H. (1982). “Late Iron Age in Western Norway. Female Roles and Ranks as Deduced from an Analysis of Burial Customs.” In Norwegian Archaeological Review , 14 (1-2), pp. 70-84.
Einarsdóttir, S. S. “The Ketilstaðir Burial. From the Archaeological Perspective”. In Bláklædda Konan, Ný Rannsókn á Fornu Kumli (Bundled up in Blue, the re-investigation of a Viking grave), Publications of the National Museum of Iceland, 17-25.
Eldjárn K. (2000). Kuml og Haugfe úr heiðnum sið á íslandi. Republished by A. Friðriksson. Reykjavik: Mál og Menning.
Ewing, T. (2006). Viking Clothing. Tempus Publishing.
Friðriksson, A. (2003). La place du mord, les tombes vikings dans le paysage culturel islandais. Thèse pour obtenir le grade de Docteur de l’Université Paris, Sorbonne, Études germaniques.
Geijer, A. (1938). Di Textilfunde aus den Gräbern, (Birka III).
Guðjónsson, E.,(1996), “Fágæti úr fylgsnum jarðar: fornleifar í þágu textíl -og búningarannsókna”, in Skírnir,166 (vor 1992), pp. 7-29.
Hayeur Smith, M. (2004) Draupnir’s Sweat and Mardöll’s tears: An Archaeology of Jewelry, Gender and Identity in Viking Age Iceland. British Archaeological Reports, John and Erica Hedges Ltd & Archaeopress, Oxford.
(2014).”Thorir’s Bargain: Gender, vaðmal, and the law”. In World Archaeology – Special Issue: The Archaeology of Legal Culture, edited by A. Reynolds and K.P. Smith, Vol. 45(5), pp.730-746, Routledge.
Hoffman, M. (1957). The Warp-weighted Loom: Studies in the History and Technology of an Ancient Implement. The Norwegian Research Council for Science and the Humanities, Oslo.
Hägg I., (1991). Textilfunde aus der Sieldung und aus den Gräbern von Haithabu. Beschreibung und Glierderung, Berichte über die Ausgrabungen in Haithabu, herausgegeben von. K Shietzel, 29, Neumunster.
Hägg I. (1974). Kvinnodräkten i Birka. Uppsala, Aun 2.
Jansson, I. (1985). Ovala Spännbucklor, En studie av vikingatiden standardsmycken med utgångspunkt från Björkö-fynden. Aun 7, Uppsala.
Jesch, J. (1991). Women in the Viking Age. Boydell Press, Woodbridge.
Pedersen, J. (1928). Vikingatidens Smukker, Stavanger Museum.
Walser J. W. III (2015). “Reading the Bones. From the Osteologist’s Perspective”. In Bláklædda Konan, Ný Rannsókn á Fornu Kumli (Bundled up in Blue, the re-investigation of a Viking grave), Publications of the National Museum of Iceland, 47-57.
Wobst, H.M.(1977). “Stylistic Behavior and Information Exchange” in Anthropological Papers, University of Michigan, Museum of Anthropology 61(1977), 328-332.
To read about the reconstruction of the “Lady in Blue”‘s garment see our blog by Marianne Guckelsberger and Marled Mader.