A "Triangular" Shawl Style for the Viking Age?

By Ásfríðr Úlfvíðardóttir/ Rebecca Lucas.

Shawls and mantles are a type of garment that seems to receive very little attention in feminine Late Iron - and Viking - Age dress. These untailored panels of fabric are presumed to be worn across the shoulders, covering the upper body and pinned centrally on the collarbone (Callmer 2008), or chest (Hägg 1971). While on the archaeological side, there is a focus on brooch positions and fabric fragments (Callmer 2008; Hägg 1971), the question of how a shawl could be worn to achieve the two major styles from contemporary artwork-- 'triangular' and 'rectangular' -- does not seem to be discussed. Publications and information aimed at re-enactors (atte Dragon, 2004; Regia Anglorum, 2007; Glæsel, 2010) have aimed to answer this question but, in my opinion, are unsatisfactory.

Drawing together archaeological evidence of textiles, fastenings, and artwork, from Scandinavia and neighbouring cultures around the 8th-10th centuries. It is hoped that we can build up a picture of what a Viking Age shawl may have looked like, particularly the ubiquitous 'triangular' shawl from art.

Jewellery and Artwork

In Scandinavia, rectangular woollen shawls date to at least the early Pre-Roman Early Iron Age (5th-1st c. BCE) (Nationalmuseet Danmark, 2010), through the Viking Age (Hägg 1984, 1986), and later in areas under Scandinavian influence in Finland (Lehtosalo-Hilander, 1984) and the Baltic (Rabinovich ed., 1986; Zariņa 1970, 1998, 1999, 2006). But it is in the brief period from the Late Iron Age (400-800 CE) to the Viking Age (800-1050 CE) (Mannering, 2004), that artwork of women wearing shawls appear.

This artwork generally has a standardised appearance; usually a woman in profile, her hair in a knot, wearing a long garment that is covered by a shawl or a cloak (Mannering, 2004; Holmqvist, 1960) (Figure 1). This outer layer may be long, with pointed corners at the sides of the body, or shorter, with a diagonal hem. The artwork from this period falls into three types: the tiny foil pieces called guldgubber; so-called 'Valkyrie' pendants; and artwork from Gotlandic picture stones (Bau, 1981). Mannering (2004) concluded that the imagery is an accurate representation of the textile remains from Viking Age graves, but it appears she was comparing pre-Viking Age artwork with Viking Age dress (ibid.; Holmqvist, 1960). Furthermore, concentrating on the extensively published Viking Age grave-fields of Birka, Sweden, the presence of shawls in these female graves are exceedingly rare, compared to earlier-period burials (Hägg 1986, 1983, 1971). This discrepancy between artwork and archaeology may be because these figures are wearing pre-Viking Age dress, or alternatively it is a highly symbolic and old-fashioned costume (Mannering, 2004).

There are further issues when relying on the many metal figures, including their small size – often less than an inch high – their highly stylised appearance and difficulties in accurate dating. The so-called 'Valkyrie' pendants, often found in Viking Age burials, are frequently comparatively dated by the other goods within the grave (Holmqvist, 1960), and the dating of gold foil Guldgubber plaques has similar issues. As they are not usually associated with datable contexts, such as post-holes, they can therefore only be dated under certain circumstances, so are usually only given a very general 5th to 9th c. date (Ratke, 2009).

In contrast to the relatively homogenous artwork from the Later Iron Age on, archaeology has revealed that there were some major changes in shawl fashion occurring, assuming that burial clothes were identical to those worn in life. Prior to the 9th century, the central 'shawl brooch' of a disc-on-bow brooch had been worn at the throat, or at the very least above the paired brooches (Callmer, 2008). But by the Viking Age the 'shawl brooch' had shifted to below the tortoise brooches (Hägg, 1971, 1986) and could be round, equal-armed, trefoil-shaped (Hägg, 1971), or in Scotland and the Hebrides a ring-headed pin (Fanning, 1992; Welander et al., 1987). Hägg (1971) believes this new position was developed, because pinning at the throat would have covered a smaller brooch used to fasten the tunic keyhole neckline, limiting the options available to display ones' wealth.

The textiles associated with these brooches changed too. While previously shawls had been pinned directly with a brooch (Hägg, 1971), it seems the vast majority of women in the 9th century passed their brooch-pins through silk or linen loops. These loops are believed to be associated with a coat-like garment with sleeves (Hägg, 1983) which replaced the shawl almost entirely, although by the 10th century some women instead favoured wearing a woollen tunic underneath their apron dresses (Hägg, 1986). Although three women from graves Bj 483, Bj 843A and Bj 1014 continued to pin their outmoded shawls through the fabric (Hägg, 1974; 1983), they also appear, according to grave plans (Historiska Museet, 2010), to have fastened it lower on the body too, implying that whatever shape their shawl, it must have been a versatile garment, able to be pinned both at the throat and the breast.

Other authors have considered the garment fastened with small loops to be a type of cloak or shawl (Ewing, 2007; Lukešová, 2011), but if it is then it probably was shaped at the shoulders. Practical attempts to wear folded sheets of fabric, like those proposed to mimic a 'triangular' look, with permanent loops have so far been unsuccessful. Without any shaping at the shoulders, there is no way to ensure that the shawl is being worn at precisely the same place every time, and so it is highly likely that the loops are in the wrong position without constant adjusting. With that said, I welcome any feedback from people who have managed to wear a 'triangular'-looking shawl with brooch loops.

If the details from the small guldgubber and 'Valkyrie' artwork have been interpreted correctly (e.g. Arrhenius, 1962) then these figures often show the earlier, throat-height brooch placement. For example, the 10th c. Aska pendant (Holmqvist, 1960) (figure 1), wears a disc-on-bow brooch at her throat, a brooch style that vanished from Swedish graves in the 8th century (with the exception of Gotland where it continued to be worn into the 11th century) (Callmer, 2008).

Despite the inconsistencies between the textile finds and art, a very rough timeline of artwork still reveals that there is a possible trend towards the disappearance of the shawl in the 10th century and overall the preferred style is for a shawl that is pointed at the back, described by Hägg (1971) and re-enactors (atte Dragon, 2004; Regia Anglorum, 2007) as being a 'triangular' shawl. A 'rectangular' shawl style appears to gain prominence by the Viking Age, as seen for example in the Oseberg tapestry (figure 1). But there is at least a century between the majority of women being buried without a shawl, and the iconic garment vanishing from art. It is therefore highly likely that the artwork is depicting earlier fashions from the Late Iron Age, and so we should be looking at these earlier graves for clues.

Figure 1
Figure 1: A Timeline of Artwork. (Opens in new window.)

A closer look at the textiles

What sort of textiles were preferred for shawls, may have implications for wearing. A tightly woven tabby is much less stretchy and likely to drape, compared to a loosely woven twill. All of the mantle or shawl fabrics mentioned below, are twill fabrics of various types with the exception of the repp-woven tabby from Bj 843A. An 8th century disc-on-bow brooch, with preserved textiles attached from Sandegårda, Gotland is the best-described and preserved textile-find for reconstruction (Hägg, 1971). Woollen textiles had corroded to the brooch pin, indicating that two different fabrics had been fastened; a doubled layer of 'medium-coarse' (Geijer, 1942) herringbone twill (Hägg, 1971) on the outside, and a 'fine diamond twill' closest to the body (Geijer, 1942). Geijer described the mantles from the Viking Age cemetary of Birka, as being of a fine woollen twill, possibly lined in wool or linen, and decorated with seam edging or cords (Geijer, 1938). Hägg (1971, 1983) re-analysed the textiles, and noted that the equal-armed brooches from 10th century graves Bj 483 and Bj 1014 preserved, respectively, a diamond twill and a four-shaft twill (Geijer's type W 34). Geijer (1938) described the twill from Bj 1014 as being brown in colour, with a texture of diagonal lines with 11 x 8 threads per cm. Hägg (1986) mentions that the trefoil brooch from Bj 843A had preserved a mid-range-quality repped wool of Geijer type W 32. The linings and decorative trim observed by Geijer are now believed to be a misidentification of the apron-dress layer (Hägg, 1983).

Two fragments have been identified as mantles from 10th century Hedeby harbour (Hägg, 1984). Due to the nature of the find -- they were rags used as caulking on a ship -- it is impossible to say if they had originally belonged to a man or a woman. Fragment 44 B, now 66.5 x 19 cm and 0.2 cm thick, is a fine, soft, and loose, evenweave 2/2 twill with a preserved tablet woven starting border. Fragment 84 is a 2/2 evenweave, loose and woolly fabric in a dark and light check with yellow and red threads. From the graves excavated at Hedeby, Hägg notes that the textiles are too fragmentary to determine if women were buried wearing a 'shawl' or a 'caftan', however she does note that the predominant fabric from this layer is a diamond twill (1991). Fragments identified as rectangular mantles from the settlement were all evenweave 2/2 twills (Hägg, 1991). The 10th century female Danish burial at Hvilehøj, contained a border from a rectangular shawl or cloak (Madsen, 1990). Finally, from a 1000-1100 CE grave (Rundkvist, 2003ab), identified as Bh 1966:27e, from Barshalder on Gotland, woollen textiles textiles woven in a three-shaft twill that have been interpreted as a folded shawl (Pettersson, 1968) that was pinned together with two animal-head brooches (Rundkvist, 2003a).

In the neighbouring Baltic lands, Latgallian women's mantles were decorated with bronze spirals, which helped preserve both fabrics and shawl dimensions. Until the 13th century, these shawls were woven in a 2/2 twill and dyed dark blue (Zariņa, 1998). Ignoring the details of how the bronze spirals and rings were attached, these shawls also had tablet woven borders, followed by the warp threads making a 5-10 cm long fringe (Zariņa, 1998), the threads often plied together (Zariņa, 1999). The mantles of the Finno-Ugrian Livonians, similarly, were also woven in a 2/2 twill and fringed (Zariņa, 2006).The rectangular mantles of Finnish women may also have been twills. The early 11th century grave no. 56 at Luistari, Eura, contained twill fragments believe to come from a shawl (Lehtosalo-Hilander, 1984), and the 12-13th century mantle from grave 26 at Tuukkala was also fringed (ibid.).

Despite numerous reconstructions using square sheets of fabric, discussed below, it appears that the predominant shape of extant shawls and mantles is rectangular. Cloaks from ca. 406-219 CE (Mannering et al., 2010) bog finds of Borremose I and II were both approximately 1.00 x 1.50 m (Nationalmuseet Danmark, 2010), although I have been unable to find mention of them in Hald (1980). Hägg (1984) considered the woollen fragments from Hedeby harbour to have come from a rectangular mantle. 7th-13th c. mantles from Latgallian graves in Latvia, varied between 0.55-0.90 x 1.00-1.30 m (Zariņa, 1998), with a tendency to increase in size over the centuries, while Daugava Liv mantles between the 10th - 13th c. were between 0.70-0.90 x 1.40-1.60 m (Zariņa, 2006). A late 11th/early 12th century mantle from grave 404, Luistari, Finland was 'at least 140 cm and probably 150-155 cm' long, and 'at least 80 cm' wide (Lehtosalo-Hilander, 1982). The 12th century mantle from grave 6, Perniö, Finland was 0.94 x 1.47 m including the fringes (Lehtosalo-Hilander, 1984), while a mantle from the same cemetary in grave 7 was 0.90 x 1.50 m (Hägg, 1971).

Male cloaks were also rectangular, but appear to be larger, and closer to the 'double square' dimensions of the Frankish mantles mentioned in De Carolo Magno (Owen-Crocker; 2004). From the Icelandic Grey Goose Laws (Grágás) the shaggy, rectangular trade cloak (vararfeldr) was woven to a standard size and quality for pricing. Their length was two thumb-ells (þumalálnir) wide and four thumb-ells long, or, 1.024 x 2.048 m (Guðjónsson, 1962). This is echoed in a 9th century (Žeiere, 2011) cloak find from Tīra bog, Latvia, that was 210 cm long and 110 centimetres wide with starting and finishing tablet-woven borders (Ģinters, 1981). Cleasby and Vigfusson (1874) note that the word feldr is not used to describe female garments.

While fringing seems rare in Norse areas during the Viking Age (an exception are the fringed scarves from Dublin, see Wincott Heckett(2003)), there may still be the chance for embellishment with tablet-woven borders as per the Hedeby harbour textile. Hemmed edges on the mantle remains S8A-H from the Hedeby settlement were originally sewn with vegetable fibre, and there is a possibility these stitches contrasted with the originally red fabric (Hägg, 1991). A lack of fringing may look plain to our modern eyes, but it does lend credence to the idea, proposed here, that these shawls may have been folded so the edges would not have been seen.

Achieving a 'triangular' shawl look

As seen in the timeline from figure 1, above, the commonly depicted shawl has a triangular profile, with a diagonal hem lengthening towards the rear, and an apparently horizontal front. The best shape that mimics this look, when worn, is an inverted pentagram draped across the shoulder blades. A 1st - 4th century square, fringed, woollen mantle from Clongownagh Bog, Ireland achieved this shape by sewing down a single corner (Heckett, 2001), which is then worn at the neck. Adjusting how deep the fold is in turn allows for the shawl to be pinned at the neck or the chest.

Figure 2
Figure 2: Diagram of folding a square piece of fabric to make a 'triangular'-style shawl. Orange lines indicate fold lines.

This reconstruction differs on many points from the assumed dimensions and folding of 'triangular' shawls that have been proposed by Viking Age re-enactors. From discussions with other Viking Age enthusiasts, both online and off, and publications aimed at re-enactors (Regia Anglorum, 2007), by far the most common way to mimic the artwork is to fold a 150 x 150 cm square (Glæsel, 2010) of woollen fabric diagonally in half. However this style of shawl produces an unsightly curve in the hem of the garment that does not correspond with the artwork (figure 3).

Figure 3
Figure 3: Comparison of the differently folded shawls in side and in full. The traditional interpretation is on the left, and the authors interpretation is on the right.

Ways of draping this traditional, diagonally-folded shawl to better match the artwork have also been suggested (eg. Norsefolk_2, 2010), such as folding the triangular shawl ends over the arms, so that the points are hidden. However these styles are either impossible to pin closed, or the pinning is suggested in non-standard places other than the central axis of the chest. Furthermore, by using a folded rectangle, it can give a possible explanation for the change in the 9th century to an (implied) single layer of fabric, pinned on the breast (Hägg, 1971, 1983, 1986), that matches with some artwork from the Oseberg and Rolvsøy tapestries and the pendant from Sibble (figure 1). Namely that the shawl was worn unfolded as a rectangle (figure 4).

Figure 4
Figure 4: Unfolded shawl, draped over the shoulders and worn as a rectangle.

Reconstructing the Sandegårda shawl.

The layers of the herringbone fabric from the 8th century find, indicate that the shawl was not folded symmetrically (refer to Geijer, 1942 for the diagrams), although, if the few female figures depicted front-on and wearing shawls or mantles are indicative of pre-Viking Age fashion, it is likely to have followed their tendency for symmetry, and appeared evenly draped when fastened (figure 5). Of the four images below, three seem to be fastened at the throat, and are short in the front to display bead-strings and dress-fronts. The fourth, from Russia, seems to show a woman holding the edges of her shawl. It is important, however, to note that none of these women are wearing an obvious 'triangular' shawl.

gold figure from TrønningeThe silver Óðinn from LejreSilver Freya from AskaGnezdovo hoard, Russia
Figure 5: Front-on images of feminine figures in symmetrically-worn clothing, including shawls or mantles.
Left: The gold figure from Trønninge, Denmark.Undated, but assumed Viking Age. (After Hupfauf, 2006)
Centre Left:The silver "Óðinn" from Lejre, Denmark. Dated 900-950 CE (Christensen, 2010). Photo: Mogens Engelund. Creative Commons License
Centre Right: The silver "Freya" from Aska, Sweden. Dated to the tenth century (Holmqvist, 1960). Photo: Christer Åhlin SHM. Creative Commons License
Right: Silver figure of a woman from the Gnezdovo hoard, Russia. Dated to the tenth century (After Duczko, 2004).

Although a pentagram shaped shawl may give the correct profile (see above, figures 2 and 3), the single layer of fabric that is pinned in this style, does not match the Sandegårda evidence. To mimic the folding of the preserved fragments, a slightly more complicated process is required. Take a shawl that is one and a half times as long as it is wide, fold the edges inward to create a square, baste the ends in place, and then fold down a corner. If the rectangle is not folded perfectly square, when the shawl is finished, one edge will hang lower than the other (figures 4 and 5), just as the Sandegårda fragments sit (Geijer, 1942).

Figure 4
Figure 4: Diagram of folding a rectangular piece of fabric to make a pentagon shape, and in turn, a 'triangular'-style shawl. Orange lines indicate fold lines, red lines indicate edges to be basted in place.

Figure 5
Figure 5: Photograph of folded, rectangular shawl showing the fabric folds matches those of the Sandegårda textiles, with black lines added for emphasis. Please see figure 3 for how this shawl looks draped on a humanoid form.
Click on image to enlarge.

This is not to say that this reconstruction is without issues of its own. It was suggested to me that the same textile remains could have come from a tubular-woven garment, such as the Huldremose peplos (Hald, 1980). While it is indeed possible, it appears to be unlikely. The peplos gown, designated Huldremose II, has been recently dated to 210-30 BCE (Mannering et al., 2010). The tubular skirt found with the Huldremose Woman, designated Huldremose I, is now dated to 210-41 BCE (ibid.), and a third tubular garment from the National Museum of Denmark has been dated to 400-200 BCE (ibid.). They are all collectively dated to the pre-Roman Iron Age, well before the Vendel and Viking Ages, when loom technology differed. It is believed that these tubular textiles were woven on a two-beam loom (Hald, 1980; Jørgensen, 2003; Owen-Crocker, 2004), which did survive until after the Viking Age as a tapestry weaving loom (Munro, 2003). However by the Vendel and Viking periods, fabric for clothing was more likely to be woven on a warp-weighted loom (Munro, 2003), which does not easily allow for tubular weaving (Hald, 1980). It is much more likely that these shawls were woven as flat panels, that could then be sewn or pinned together.

It is also unpublished how large the Sandegårda textile fragments are. Neither Geijer (1942) nor Hägg (1971) give the dimensions of the Sandegårda brooch, so it is impossible to extrapolate the size of the preserved wool pieces, but for such a short shawl pinned at the front, the corresponding fabric must be significantly less than 1 m wide. Prior to, and after this period, however, female shawls and mantles appear to have been only roughly 1x1.5 m in diameter, or smaller.

Zariņa (1998) considers the similarly-sized Latgallian shawls to be comparatively small and decorative, presumably rather than practical garments for warmth. Yet a 90 cm wide shawl seems best fits my frame, and keeps my upper chest warm, although the precise shawl dimensions are likely to vary from person to person. This is especially true when you think about the height of your average Norsewoman, and heights today. The average height of women in the Viking Age from the area of modern-day Denmark was 158.1 cm (Coupland, 2003) while the mean height of women over 20 years old, in the United States today (for example) is 162.2 cm (McDowell et al., 2008). While less than 4 centimetres difference of our hypothetical Norsewoman-of-perfectly-average-height does not sound like much, it can be enough for hems to rise, and for garments to appear too short. This is worth keeping in mind if you find the reconstructed shawl dimensions are too small.

The Issue of Warmth: A Second Shawl or a Neck-kerchief?

As generally re-enactors wear a cloak, shawl or caftan when they are cold, instead of as part of a possibly symbolic or ceremonial outfit, this is a concern. This reconstruction still remains an impractical garment, as when it is worn across the shoulder blades, instead of around the neck, it leaves the top of ones shoulders bare. The solution may lie in the idea of a ceremonial train, back-cloth or neck-kerchief, described here as an under-shawl.

Although Geijer (1942) proposed that the fine diamond twill found corroded to the brooch pin from Sandegårda may have been a veil, I have been unable to find any other finds from the 8th century that mention headwear. The earliest evidence for veils in Scandinavia I am aware of, are the linen and wool veils worn by the 9th century Oseberg 'queen' and her 'attendant' (Owen-Crocker, 2004), a century after this Gotlandic find.

It is, however, plausible that this fabric is the remains of a second, under-shawl. It can be reproduced with a narrow, rectangular piece of lightweight fabric pinned high upon the chest, using the same brooch for the over-shawl. This lighter and narrower piece closely covers the shoulders and neck, acting as a scarf.

Flemming Bau (1981) is better known for suggesting the addition of panels attached to the so-called apron dress to better match the contemporary artwork, a theory discussed in greater detail below, but he also briefly considered the idea of a second shawl pinned at the chest. From playing with rectangles of fabric in front of the mirror, what Bau interprets as two styles of train – the first pinned at the chest, and the second suspended from the back of an apron dress – are much more likely to be narrower or wider rectangles pinned at the throat.

Narrower rectangles give an almost-vertical train similar to the Tuna figure, while wider pieces give a near-diagonal train like the picture stone from Tjängvide (figure 1). This idea of a second shawl is bolstered by the textile-related grave finds of near-contemporary Latgallian and south-western Estonian women. Latgallian women could also buried in two mantles, the outer of bronze-decorated wool, and the inner of blue and white-checked linen (Zariņa, 1970), while a grave from Siksali cemetery, Estonia, "included a body with two shawls" (Riikonen, 2005).

Bau (1981) believed that Viking Age women wore a pleated, or lightly folded, train that was worn with loops that were hooked onto the paired tortoise brooch pins of the apron dress. With “no direct evidence for a train”, this interpretation was based on the presence of 'extra' strap-loops in the top half of the tortoise brooches at Birka (Bau, 1981). Thor Ewing appears to assume that the train was a half-circle cloak pinned to the tortoise brooches (Ewing, 2007). However Bau and Ewing's theories are dependent on the contemporary artwork showing women wearing an apron-dress in the first place, something which has not yet been proven conclusively with the available imagery (Hägg, 2009). The artwork of Norsewomen may have been from the highest class of society who did not wear the middle-class apron-dress and tortoise brooches (Madsen, 1990), thereby providing no attachment points for Bau's train, but strengthening his idea of a centrally-pinned under-shawl.

The dress of the highest class, with a trailing garment and no apron, is mentioned in the Rígsþula, a poem possibly dating to the 10th century (Amory, 2001). The work describes the origins of the classes of Norse society, and their contrasts, including differences in dress. The noblewoman Móðir (lit. Mother) is therefore compared to the free-woman Amma (lit. Grandmother). Amma wore “smokkr var á bringu, dúkr var á halsi, dvergar á öxlum” [a smock on her bosom, a kerchief at her neck, dwarf-pins at her shoulders] (Owen-Crocker, 2004), which has been interpreted as describing a "typical" apron dress and paired brooch set (Ewing, 2007). The aristocratic Móðir, however wore “kinga var á bringu, siðar slæður” [a coin-brooch at her bosom, a trailing slæður she wore] (Owen-Crocker, 2004), with no mention of the 'dwarf-pins'.

The slæður is a poorly-defined garment that is also mentioned in later Icelandic texts, such as Brennu-Njáls Saga, and has been interpreted as a long, trailing, feminine, scarf (DaSent, 1861), train (Cleasby and Vigfusson, 1874), robe (Cook, 2001) or gown (Cleasby and Vigfusson, 1874), depending on the translator. At the very least, it seems to be related to the verb slæða, meaning to trail or drag along the ground (Cleasby and Vigfusson, 1874). Using the descriptions from Rígsþula, it is tempting to conclude that the trailing hems on artwork, created with an over - and under - shawl, are part of the ceremonial dress of a high-born woman, and therefore may not need to be the most practical garment for warmth, nor require paired tortoise brooches to be worn.

However, we have no way of knowing just how long the garments from Sandegårda originally were. As the aforementioned Amma was described in the Rígsþula as working indoors, she may not have worn her mantle, but was still wearing a dúkr (covering or neck-kerchief) around her shoulders. It is possible that this is a description of the fine, diamond twill fabric from Sandegårda, that was never intended to drag along the ground like a slæður at all. From 10th - 11th century Dublin, narrow woollen and silk textiles identified as scarves (Wincott Heckett, 2003) may have been worn knotted the neck. The evidence could be argued either way, although using a 10th century poem and later archaeological finds to justify an 8th century clothing reconstruction is a stark reminder of what little evidence is available.

Is it a Viking Age Fashion, or Earlier?

It is tempting, but foolhardy to treat the Late Iron and Viking Ages, and the 8th to 10th centuries as being a homogeneous time period with identical fashions throughout the Viking world. It is likely that if 'triangular' shawls were worn, it would only be in particular places and eras.

While the 9th century Oseberg tapestry and 10th century Rolvsøy tapestry seem to show roughly contemporary dress with a rectangular mantle and brooch on the breast (figure 1), it appears to have been unfashionable in 9-10th century Birka. Meanwhile there is evidence for mantles being worn in the 10th century female Danish burial at Hvilehøj, from Hedeby harbour and 11-12th century Barshalder, implying that shawls may still have been worn by women outside of the trading centre of Birka.

Although the evidence for this mantle is slim, this particular interpretation does make for an adaptable garment that can be worn both as a 'triangular' and 'rectangular' shawl, that can be pinned at both the throat and chest, and worn folded and basted, or unfolded.

With that said, the single layers of wool preserved from some Birka graves may indicate that for much of Scandinavia in the Viking Age, no shawl, or, an unfolded rectangular shawl were worn, instead of this hypothetical 'Sandegårda style'. The lack of evidence of a trailling undershawl from the Viking Age (Hägg, 2009), that appears in Iron Age artwork, may explain the original purpose of the fine diamond twill from Sandegårda, but it also implies that this particular layer is unlikely to have been worn in this manner by the 9th century.

How popular this dress layer was in the Viking Age, and whether it was seen as old-fashioned at the time, is still unknown. From what little evidence remains, though, it is very unlikely that during the Viking Age there was a single shawl style universally worn by all Norsewomen, at all times.


Thank-you to Mistress Þóra Sharptooth for drawing my attention to the Clongownagh Bog find, and the many, many people who generously answered my incessant questions about their preferred shawl-wearing style.


All websites valid as of March 20, 2011.

  • Amory, F. 2001. 'The Historical Worth of Rígsþula' alvíssmál 10; 3–20.
    Available online at: http://userpage.fu-berlin.de/~alvismal/alvinh.html
  • atte Dragon, B. 2004. Real Viking Project: Women's Clothing.
    Available online at: http://web.archive.org/web/*hh_/www.realviking.vikingagevessels.org/pdfs/C_+Viking+Womens+Costume.pdf
  • Arrhenius, B. 1962. “Det flammande smycket” [The incandescent jewellery] Fornvännen; 79-101.
    Available online at: http://fornvannen.se/pdf/1960talet/1962_079.pdf
  • Bau, F. 1981. "Seler og Slaeb i Vikingetid: Birka's Kvindedragt i nyt lys." [Straps and trains in the Viking Age: Birka's female costume in a new light] KUML; 13-47
  • Bender Jørgensen, L. 2003. "Northern Europe in the Roman Iron Age, 1 BC - AD 400" in The Cambridge History of Western Textiles. Volume I. (Cambridge University Press); 93-102.
  • Callmer, J. 2008. “The Meaning of Women's Ornaments and Ornamentation” Acta Archaeologica 79; 185-207.
  • Christensen, T. "Odin fra Lejre" [Odin from Lejre] Danefæ - Skatte fra den danske muld; 174-178.
    Available online via the Roskilde Museum.
  • Cleasby, R. and Vigfusson, G. 1874. An Icelandic-English Dictionary (Clarendon Press).
    Available online at: http://www.ling.upenn.edu/~kurisuto/germanic/oi_cleasbyvigfusson_about.html
  • Coupland, S. 2003 "The Vikings on the Continent in Myth and History." History 88(290); 186-203.
  • Cook, R. [trans.] 2001. Njal's Saga (Penguin Books).
  • DaSent, G.W. [trans.] 1861. The Story of Burnt Njal.
    Available online at: http://omacl.org/Njal/
  • Duczko, W. 2004. Viking Rus: Studies in the Presence of Scandinavians in Eastern Europe (Brill).
  • Ewing, T. 2007. Viking Clothing [Revised Edition] (Tempus Publishing).
  • Fanning, T. 1992. "Ringed Pins, the Hiberno-Norse and the Discovery of Vinland" Archaeology Ireland, 6(1); 24-26.
  • Geijer, A. 1938. Birka III. Die Textilfunde aus den Gräbern [Birka III. The textile finds from the graves] (Kungl. Vitterhets Historie och Antikvitets Akademien).
  • Geijer, A. 1942. "En dräktdetalj frän 700-talet." [A detail of costume from the 700s] Fornvännen 37; 373-374.
    Available online at: http://fornvannen.se/pdf/1940talet/1942_373.pdf
  • Ģinters, V. 1981. Tracht und Schmuck in Birka und im ostbaltischen Raum [Costume and Jewellery in Birka and the Eastern Baltic Space] (Kungl. Vitterhets Historie och Antikvitets Akademien).
  • Glæsel, N. 2010. Viking. Dress, Clothing, Klær, Garment (Nille Glæsel Forlag)
  • Guðjónsson, E.E. 1962. Forn röggvarvefnaður [Ancient pile weaving] Árbók Hins Íslenzka Fornleifafélags 59; 12-71.
    Available online via the National and University Library of Iceland
  • Hald, M. and Olsen, J. [trans.] 1980. Ancient Danish Textiles from Bogs and Burials: A Comparative Study of Costume and Iron Age Textiles (National Museum of Denmark).
  • Heckett, E.W. 2001. "Beyond the Empire: An Irish Mantle and Cloak" in The Roman Textile Industry and its Influence: A Birthday Tribute to John Peter Wild (Oxbow); 91-97.
  • Historiska Museet 2010. Birkagravar. [Birka graves]
    Available online at: http://kulturarvsdata.se/shm/context/html/480 http://kulturarvsdata.se/shm/context/html/1031
    and http://kulturarvsdata.se/shm/context/html/512
  • Holmqvist, W. 1960. “The Dancing Gods” Acta Archaeologica 31; 101-127.
  • Hupfauf, P.R.M. 2006. Signs and Symbols Represented in Germanic, Particularly Scandinavian Iconography Between the Migration Period and End of the Viking Age [PhD Thesis]. (University of Sydney)
    Available online at: http://hdl.handle.net/2123/662
  • Hägg, I. 2009. Populärvetenskap – en samvetsfråga? [Popular science – a question of conscience?]
    Available online at: http://ingahagg.cybersite.se/text_108109.html
  • Hägg, I. 1991. Textilfunde aus der Siedlung und aus den Gräbern von Haithabu [Textile finds from the settlement and graves of Hedeby] (Karl Wachholz Verlag)
  • Hägg, I. 1986. Die Tracht. [The Dress] In Birka II:2. Systematische Analysen der Gräberfunde (Almqvist and Wiksell International); 51-72.
  • Hägg, I. 1984. Die Textilfunde aus dem Hafen von Haithabu [The textile finds from the Harbour of Hedeby] (Karl Wachholz Verlag)
  • Hägg, I. 1983. “Viking Women's Dress at Birka: A Reconstruction by Archaeological Methods.” In Cloth and Clothing in Medieval Europe (Heinemann); 316-350.
  • Hägg, I. 1974. Kvinnodräkten i Birka: Livplaggens rekonstruktion på grundval av det arkeologiska materialet (Archaeological Institute, Uppsala)
  • Hägg, I. 1971. Mantel och kjortel i vikingatidens dräkt. [Mantle and coat in Viking-Age costume] Fornvännen 66; 141-153.
    Available online at: http://fornvannen.se/pdf/1970talet/1971_141.pdf
  • Krafft, S. 1956. Pictorial Weavings from the Viking Age (Dreyers Forlag)
  • Krag, A.H. 2007. “Christian influences and symbols of power in textiles from Viking Age Denmark. Christian influence from the continent.” In Ancient Textiles: Production, Craft and Society (Oxbow Books); 237-243.
  • Lehtosalo-Hilander, P.-L. 1982. Luistari II: The Artefacts (Suomen Muinaistoyhdistyksen Aikakauskirja)
  • Lehtosalo-Hilander, P.-L. 1984. Ancient Finnish Costumes (Finnish Archaeological Society)
  • Lukešová, H. 2011. "Fragmenter av kvinnedrakter fra vikingtiden – Metode for identifikasjon av gamle tekstilfunn" [Fragments of women’s costumes from the Viking Age – A method for indentification of ancient textile finds] Viking 74; 149-164.
    Available online as a link from a news article at the University of Bergen.
  • Madsen, A.H. 1990. "Women's dress in the Viking period in Denmark, based on the tortoise brooches and textile remains" In Textiles in Northern Archaeology: NESAT III Textile Symposium in York 6-9 May 1987 (Archetype Publications); 106-111.
  • Mannering, U. 2004. “Dress in Scandinavia: Iconography of the 5th-10th centuries A.D.” In Priceless Invention of Humanity – Textiles. (Łódzkie Towarzystwo Naukowe); 67-74.
  • Mannering, U., Possnert, G., Heinemeier, J., and Gleba, M. 2010. "Dating Danish Textiles and Skins from Bog Finds by Means of 14C AMS." Journal of Archaeological Science 37; 261-268.
  • McDowell, M.A., Fryar, C.D., Ogden, C.L., and Flegal, K.M. 2008. "Anthropometric Reference Data for Children and Adults: United States, 2003–2006" National Health Statistics Reports 10
    Available online at: http://www.cdc.gov/nchs/data/nhsr/nhsr010.pdf
  • Munro, J.H. 2003. "Medieval Woollens: Textiles, Textile Technology and Industrial Organisation, c.800-1500." in The Cambridge History of Western Textiles. Volume I. (Cambridge University Press); 181-227
  • Nationalmuseet Danmark. 2010. Dragter i tekstil fra ældre jernalder. [Clothing of textiles from the Early Iron Age]
    Available online at: http://oldtiden.natmus.dk/livet_i_oldtiden/hvordan_gik_de_klaedt/aeldre_jernalders_dragter/dragter_i_tekstil_fra_aeldre_jernalder/language/u/
  • Norsefolk_2, 2010. Shawls.
    Available online at: http://groups.yahoo.com/group/Norsefolk_2/message/21384 (membership with Yahoo! account required to view archives.)
  • Owen-Crocker, G.R. 2004. Dress in Anglo-Saxon England (Boydell Press.)
  • Pettersson, K. 1968. "En gotländsk kvinnas dräkt. Kring ett textilfynd från vikingatiden" [A Gotlandic Woman's Dress. Surrounding a Textile Find from the Viking Age.] Tor 12; 174-200.
  • Rabinovich, M.G. [ed.] 1986. Drevnjaja odezhda narodov Vostochnoj Evropy. [Ancient Clothing of Peoples of Eastern Europe.] (Nauka.)
  • Ratke, S. 2009. Guldgubber - Einblicke in die Völkerwanderungszeit [Guldgubber - a glimpse into the younger iron age] [PhD Thesis] (Rheinische Friedrich-Willheims Universität Bonn).
    Available online via: urn:nbn:de:hbz:5-19093
  • Regia Anglorum. 2007. Members Handbook: Viking.
    Available online at: http://www.regia.org/members/handbook/viking.pdf
  • Riikonen, J. 2005. "Iron Age Aprons from Southwestern Finland" Rituals and Relations: Studies on Society and Material Culture of the Baltic Finns (Academia Scientiarum Fennica); 31-72.
  • Rundkvist, M. 2003a. Barshalder 1: A cemetery in Grötlingbo and Fide parishes, Gotland, Sweden, c. AD 1-1100. (University of Stockholm)
    Available online at: http://www.archive.org/details/Barshalder_1
  • Rundkvist, M. 2003b. Barshalder 2: Studies of Late Iron Age Gotland (University of Stockholm)
    Available online at: http://www.archive.org/details/Barshalder_2
  • Welander, R.D.E., Batey, C. and Cowie, T.C. 1987. "A Viking Burial from Kneep, Isle of Lewis" Proceedings of the Society of Antiquaries of Scotland 117; 149-174.
  • Wincott Heckett, E. 2003. Viking Age Headcoverings from Dublin (Royal Irish Academy)
  • Zariņa, A. 1970. Seno Latgaļu Apģērbs 7-13 gs. [Ancient Latgallian clothing, 7-13th centuries] (Zinātne)
  • Zariņa, A. 1998. "Neue Materialien für das Entwicklungsschema der Latgallischen Schultertücher aus dem 7.-13. Jahrhundert" [New materials for the developmental scheme of Latgallian shawls from the 7-13th centuries] In: Bender Jørgensen, L and Rinaldo, C [Eds.] Textiles in European Archaeology: Report from the 6th NESAT Symposium, 7-11th May 1996 in Borås (GOTARC);151-156
  • Zariņa, A. 1999. Apģērbs Latvijā 7.-17. gs. [Garments in Latvia in the 7-17th centuries] (Zinātne).
  • Zariņa, A. 2006. Salaspils Laukskolas Kapulauks 10.-13. gadsimts. [Salaspils Laukskola Cemetery, 10-13th centuries] (Latvijas vēstures institūta apgāds)
  • Žeiere, I. "Fragments of Male Clothing in the 9th Century Bog Hoard in Latvia" NESAT XI Program.
    Available online at: http://www.nesat.de/abstracts/lecture_zeiere.pdf