Viking women: Shawls and cloaks

By Hilde Thunem (hilde@thunem.priv.no)
(Created August 20th 2022) (PDF)

This article represents my attempt to collect archaeological facts and interpretations of the sleeveless unshaped overgarments that were worn by Norse women during the Viking Age.

Old Norse have several terms for such garments; feldr, skikkja and mǫttull. Feldr refers to a simple cloak or blanket with corners (feldr-skaut), occasionally lined with fur or made from animal skin. Unlike the skikkja and mǫttull, that could be worn by both genders, feldr is not used to describe a woman's cloak (Cleasby and Vigfusson, 1874). Skikkja and mǫttull were sometimes used interchangeably. However, the loan-word mǫttull (from the Latin 'mantile') was usually used about cloaks of foreign fabric or foreign cut (Cleasby and Vigfusson, 1874). Both garments have corners (skaut), although the existence of such a word does not specify how many corners there were.

In the article, I will be focusing on the cloaks and shawls made from woven fabric and worn by women, sometimes with a closing brooch - the skikkja or mǫttull.

NB! This is a work in progress. Interpretations of the evidence will follow at a later date...

Contents

Facts: Archaeological evidence

Given the fragmentary evidence of textiles in Viking Age graves, the identification of possible overgarments is usually based on the stratigraphy. However, as overgarments appear in the same layers in the grave as blankets or duvets that were used to cover the deceased, it can be difficult to identify whether the tiny fragments that remain comes from a garment or a cover.

One indication of an overgarment is finding a brooch that may have fastened it. In many graves this is the only surviving evidence of a cloak or other wrap-type garment. However, not all overgarments were fastened with a brooch. In a handful of the graves from Birka, fabric from beneath the body, that is, from the back of the garments have been preserved. In nine graves there were fragments of the same type of fabric on top of the oval brooches and underneath the remains of the body, indicating an overgarment wrapped around the body. However, only five of the graves contained an extra brooch in addition to the oval brooches. Thus, some of these overgarments were worn without a brooch (Hägg 1974, p. 84).

Ylle som legat under spännbucklorna och kroppens förmultningsrester och alltså härrör från dräktens ryggsida, finns i 14 av gravarna. 11 gånger, dvs i 9 av de 14 gravarna, finns samma slags ylle också på spännbucklornas skal (...). De måste då härröra från ett plagg, som legat runt kroppen utanpå både särk och kjol. I samma gravar finns ett tredje spänne 5 gånger, hvilket innebär, att åtminstone 6 av de sammanlagt 11 plaggen bär ha varit arrangerade utan spännen.
Hägg 1974, p. 84

Even for the prosperous farmer women that wore oval brooches, having a third brooch was not that common, as shown by an examination of 833 female graves from the Merovingian and Viking Age period in Norway. Solberg (1985, p. 67-68) divides the graves into three categories; a lower group (34% of the graves) with beads and/or textile utensils, an intermediate group (51% of the graves) with at least one oval brooch, and an upper group (15% of the graves) with oval brooches and a third brooch.

The lower group is likely to be underrepresented in the recorded finds, as their simple graves with few items are much easier to overlook in the field than the richer graves (Solberg 1985, p. 71). However, this does not affect the comparison between the two other groups, where only about one in five of the women buried with oval brooches also had a third brooch.

Altogether 833 women's graves were recorded. They were divided into three groups. Group 1 has five or more beads and/or textile utensils, group 2 at least one conical or oval brooch, and group 3 conical or oval brooches and a third brooch. Group 1 constituted 44%, 24%, and 33% of the women's graves in western, central and eastern Norway, respectively. <...> Group 2 represents 40%, 60% and 54% of the women's graves in western, central and eastern Norway, respectively. Keys, beads and agricultural and textile utensils are often found in these graves. Group 3 represents the richest graves. In addition to the oval brooches and the third brooch, beads keys and agricultural and textile utensils are usually present. This group constitutes 16% in western and central Norway and 13% in eastern Norway.
Solberg 1985, p. 67-68

A further challenge is identifying what kind of overgarment a certain set of fragments belong to. Hägg (1974, p 95) notes that it is easier to identify tailored jacket- and tunic-type overgarments (by the presence of seams, and the fabric being cut at an angle), than to unequivocally identify an unshaped garment like the shawl, cloak or mantle. Unlike in Iron Age Finland and the Baltics, where shawls have been preserved by the metal decorations woven around the edges, there are no unique criteria to definitively identify the shawls or cloaks worn by Norse women during this period. However, in the following overview, some of the most likely instances of archaeological evidence for such garments are described.

Vad först tolkningen av materialten med utgångspunkt från fyndläget beträffar, tilkommer svårigheten, att lämningar av dräktens yttre livplagg i regel bevarats på samma ställen i graven, som lämningarna av dynor, hyenden, etc. Kring den döda, alltså över spännbucklornas skal och under kroppens förmulningsrester.
Hägg 1974, p. 60

Som det har visat sig, kan yllefragment från spännbucklornas skal inte a priori tolkas som lämningar av en mantel, och de textilavsnitt, som finns bevarade från de yttre livplaggen, inkluderar knappast heller några karaktäristiska delar av plaggen, som ger säkra utgångspunkter för en tolkning. Det är på grund av sömmar, fogar och skärningar lättare att bland fragmenten känna igjen rester av ett tilskuret och sytt plagg än av ett otillformat stycke som mantel eller sjal. Eftersom man inte i Birka som på så många håll i Finland och Baltikum vävt in metalldelar runt mantelns kanter och på så sätt har dess grundform bevarad, finns det inga säkra kriterier att gå efter för en definitiv identifikation av birkamanteln.
Hägg 1974, p. 95

Birka

The main excavation of Birka was conducted in the 1870s by Hjalmar Stolpe, yielding a large amount of textile material from the 9th and 10th century. The textile fragments were stored, and later analysed by Agnes Geijer (in 1938) and Inga Hägg (1974, 1986).

Of the 128 graves containing oval brooches analysed by Hägg (1974), 55 contain woollen fragments from other garments in addition to the dress suspended from the oval brooches (the smokkr). The majority (40) of these graves have fragments of one type of fabric - and is thus interpreted as having contained a single garment in addition to the smokkr and underdress. In 15 of the graves, however, there is woollen fabric of at least two types, in addition to the remains of the smokkr (p. 83).

The fragments from the additional garments are most commonly either broken diamond twill or what Hägg (1974, p. 83) calls pattern-woven fabric. The last type of fabric is sometimes found used in duvets covering the deceased but is also used in bands stitched as decoration on other fragments and may have been used on its own in cloak-type garments as well.

Ylle från andra livplagg än kjolen finns i 55 (?) av kvinnogravarna. Oftast (ca 40 gånger) är de bevarade ylleresterna från en grav av en och samma kvalité och bör därför snarast kunna uppfattas som delar av ett enda livplagg. I de återstående ca 15 gravarna finns ylle i minst två olika kvalitéer (materialet i kjolen undantaget).
Hägg 1974, p. 83

De utan tvekan vanligaste textilslagen är diamantkypert och mönsterväv. Att diamantkyper användes som material till livplaggen är väl belagt. Det kan däremot inte sies om mönsterväven. I tre gravar är det ganske tydligt, att mönsterväv klätt dynan till det bolster, som den döda vilade på (597, 825 och 968), därför att mönsterväven här förekommer tillsammans med rester av dun. <...>

I samband med ett par av frägmenten har Geijer nämnt möjligheten, att det skulle kunna vara fråga om mantlar (660 och 739, jfr. Ovan s. 64 f. och 65). I båda fallen hade styckena rester av kantbård och påminner på så sätt mera om klädesplagg än om enkla hyenden i en bädd. <...> I ett fall, grav 838, är det tydligt, att mönsterväven kommer från band, vilka varit fastsydda som prydnadsbårder på ett annat material, nämligen treskaftat diamantkypert (W17). <...> Mönsterväv har uppenbarligen använts i flera sammanhang, och det är troligt, att en del av gravarna ifråga bara har rester av ytterligare ett livplagg, därför att mönstervävsfragmenten antligen enbart utgjort prydnadsbårder på ett annat plagg eller kommer från en bädd i graven.
Hägg 1974, p. 83-84

The use of brooches at Birka

Diagram of brooches Analysing the distribution of brooches in various Birka graves, Hägg (1971, p. 143-144) notes that when extra brooches were worn in addition to the oval ones, the usual set-up was either a single brooch (usually an equal-armed or trefoil brooch) or two round brooches (one large and one small).

Illustration: The distribution of brooches in the female skeleton graves in Birka. Hägg 1971 p. 143

The brooch types indicate that the use of three brooches came first, and continued side by side with the later custom of wearing four brooches. In the early graves, the single brooch is often placed high, beneath the chin. In the four-brooch graves, the large brooch is often placed at the same height as the oval brooches, while the smaller brooch is placed at the neck, often closing the underdress (Hägg 1971 p. 143-144).

En undersökning av fördelningen av dräktspännen i de kvinnliga skelettgravarna i Birka visar att det finns två huvudtyper av uppsättningar med mantelspännen, nämligen dels med tre spännen: två ovala spännbucklor och ett likarmat eller treflikigt spänne, dels med fyra spännen: två ovala spännbucklor, ett stort och ett litet runt spänne. Det likarmade spännet är avgjort vanligast i trespännegravar (ca 30 gänger), medan det ingår två ganger i uppsättningar med fyra spännen. Det stora runda spännet däremot förekommer sex gånger i trespännegravar men 18 gånger tillsammans med litet runt spänne (fig. 4).

Mantelspännet tycks först ha haft en hög placering, ungefär som på figuren från Aska. Typerna anger att uppsättningen med tre spännen är den ålderdomligare, medan den med fyra spännen visar sig senare. Dock har båda uppenbarligen förekommit sida vid sida i Birka och influerat varandra ifråga om mantelspännets placering. I fyrspännegravarna är detta lågt placerat (se t. ex. grav 963), ungefär i höjd med spännbucklorna, och ett litet runt spänne sitter på mantelspännets tidigare plats, där det håller ihop särkens sprund. Det arkeologiska materialet ger belägg för en högre och en lägre placering av mantelspännet (som i sin tur inverkar på hur mycket av den övriga smyckeuppsättningen som varit synligt).
Hägg 1971 p. 144

Fragments pierced by the brooch

Three of the graves analysed by Hägg (1974) has a brooch that show evidence of having fastened an overgarment by piercing its edges. Hägg interprets all these as belonging to garments of a cloak-type, although she notes that the piercing of an edge is not definitive proof of the garment being a shawl or cloak instead of a tailored garment.

Från grav 483 kommer ett likarmat spänne med ett stycke gåsögonmönstrat ylle över nålhällaren, vilket är genomstunget av nålen (fig. 5). På baksidan av ett annat likarmat spänne (grav 1014) sticker på samma sätt nålen igenom ett ganska ansenligt stycke yllekypert (Geijers W 34). Överraskande nog hör dessa exempel till ovanligheterna. Det hänger givetvis samman med bevaringsomständigheter men, visar det sig, måste framför allt bero på att också ett annat arrangemang [for å lukke klesplagg] var vanligt i Birka.
Hägg 1971 p. 145

I tre fall förekommer det att nålen till det tredje spännet sticker direkt igenom framkanterna till det plagg, som spännet suttit i (483, 843 A, och 1014 ...). Möjligen skulle man kunna sätta ett sådant arrangement i samband med manteln, men några särskilda belägg för detta finns inte i materialet från gravarna i fråga.
Hägg 1974, p. 95

Grave 483

On the back of an equal-armed brooch there is a fragment of fine woollen diamond twill fabric (Hägg 1971 p. 145) with a folded edge at the end facing towards the middle of the brooch. The brooch pin has pierced straight through the folded edge (Hägg 1974, p. 61-62). There were no other brooches in the grave.

Photo of brooch and textiles from Bj483
Pavel Voronin, SHM Creative Commons Erkännande 2.5 Sverige

483. På baksidan av ett likarmat spänne från graven sitter ett stycke fin blåsvart kypertväv med en vikkant in mot spännets mitt. Spännenålen sticker rakt igenom denna vikkant. Kyperten torde komma från en mantel, som hållits samman över bröstet med det likarmade spännet (avbildat i Hägg 1971 fig. 5).
Hägg 1974, p. 61-62

Grave 843 A

Underneath the pin of a trefoil brooch there were two fragments of a woollen fabric that had been held closed by the brooch.

Drawing of brooch The brooch pin had pierced both sides of the opening (Hägg 1974, p. 67). The fabric was a coarse rep-woven wool with a thread count of 16x6 per cm (Geijer 1938, p. 168, 36).

The grave plan (Arbman 1940, p. 317) show the trefoil brooch laying almost inside one of the oval brooches, indicating that it may originally have been worn in a position between the oval brooches.

843 A. Under nålen till ett treflikigt spänne rester av ylle, W 32, från en mantel (?), som hållits samman med det treflikiga spännet. Nålen har stuckit direkt genom tyget på båda sidor, 843 A:1.
Hägg 1974, p. 67, illustration p. 132

843 A. Reste eines dunklen Ripses, W 32, und feineren, zweibindigen Stoffes, vielleicht FH.
Geijer 1938, p. 168


Grave 1014

Photo of textile fragment Fragments of a coarse, woollen four-shaft twill (currently brown) was found over the oval brooches, and under the remains of the body. Remains of the same wool were found on the back of an equal-armed brooch. The pin of the equal-armed brooch appears to have pierced directly through the fabric. The fragment is assumed to have belonged to a shawl or cloak (Hägg 1974, p. 69, 89). The twill had a thread count of 11 x 8 per cm (Geijer 1938, p. 38).

Photo: Geijer 1938, Taf 8

According to Geijer (1938, p. 174) there was a rusted wool string as well, but it is unclear from her description whether it was attached to the twill fragments, and if so where. Hägg (1974, p. 69) does not mention the string. The equal-armed brooch was found below the oval brooches (Arbman 1940, p. 425-426).

1014. Taf. 8:4, Abb. 10. Vierbindiger Köper. 2 Stücke von sehr dunkler, rein brauner Farbe. Kette fest gesponnen, Schuss gröber und loser, jene links, dieser rechts. Fadendichte per cm II x8. Oberseite zeigt sehr deutliche Textur mit rechts aufsteigenden Diagonallinien, Unterseite unklar. Gewebe dicht und ebenmässig ohne Verfilzung. <...> Auf und teilweise unter den ovalen Spangen und unter der gleicharmigen Spange dunkler, dicker Wollstoff, W 34, sowie eine verrostete Wollschnur gleich D I.
Geijer 1938, p. 38, 174

1014. Över och under de ovala spännbucklorna låg stycken av ett kraftig, nu brunt, fyrskaftat yllekyperttyg, W 34 (textilbeskrivning Birka III, s. 38, Taf 8:4). Rester av samma ylle fanns på baksidan av ett likarmat spänne. Nålen, vars mittdel saknas, har troligen stuckit direkt genom tyget, varför fragmentet kan antas härröra från en mantel.
Hägg 1974, p. 69

Closed with loops of fabric or string

Many of the overgarments found at Birka had a different closing mechanism, where loops of fabric or string had been attached to the edges of the garment, and a brooch held the loops together. Remains of silk and linen loops are found on the back of several brooches. The loops were just long enough to allow the edge of the garment to be placed at the edge of the brooch (Hägg 1971 p. 145).

Kring nålen på ett stort antal spännen sitter nämligen kvar små öglor av linne eller siden (fig. 6), som visar att man i mantelns båda framkanter sytt fast två korta öglor mitt emot varandra, en i var kant. Mantelspännets nål har sedan trätts genom dessa öglor istället för att stickas genom tyget - på samma sätt som ifråga om hängselkjolen. Några gånger kan man bestämma öglans längd tack vare att små fragment av själva mantelkanten sitter kvar. Det visar sig att den gjorts så lång, att den precis doldes under spännets kant. Med detta arrangemang vann man två saker: dels skonade man tyget från stickhål, nötningsspår och annat slitage från nål och spänne, dels gav man manteln ett ledigare fall.
Hägg 1971 p. 145, 146

Hägg (1971 p. 145) first believes these loops to belong to shawls, but after later analysis of the graves (1974, p. 85-89) identifies most of them as tailored jacket-type garments. The exception is Bj 466 where she believes the garment may have been an unshaped wrap-type garment like a shawl or cloak.

Grave 466

Drawing of 
stratigraphy The grave contained several fragments of a blue woollen broken diamond twill, found both over one of the oval brooches and underneath the remains of the body. At one place the textiles had been pressed between the top of the wooden coffin-lid and the oval brooch, preserving the stratigraphy; there was a small band-like fragment of silk closest to the brooch (interpreted as part of a separate garment), on top of that there was a linen layer and then the outmost layer was the blue broken diamond twill (Hägg 1974, p. 61).

There were several wool fragments (one 7 x 3.5 cm). The fabric is even and woven with precision. The thread count was 50 x 17 threads per cm (Geijer 1938, p. 28).

Illustration: Hägg 1974, p. 122. W16 is the blue woollen twill, with a layer of linen (FH) underneath.

There is a selvedge along one of the fragments of wool twill. A loop of wool string has been fastened to the selvedge and a piece of a brooch pin remains in the loop. The pin fragment fits on the back of a trefoil brooch that was found in the grave, indicating that this brooch was used to close the garment (Hägg 1974, p. 61).

According to the grave plan (Stolpes gravplaner) the trefoil brooch was found close to one of the oval brooches, indicating that it was worn in a position between the oval brooches.

Photo of textiles on pin from Bj466
Photograph: Geijer 1938, taf 5.5.

Hägg (1974, p. 61) interprets this as a blue broken diamond twill wool garment with a linen lining, covering the front and back of the body. She notes that while the trefoil brooch closed the garment along the selvedge, which is often seen as an indication of an unshaped garment, it is possible that also shaped garments like a jacket would have used the selvedge in this manner. On the other hand, the presence of a loop does not mean that the garment was shaped. Finds from Finland demonstrate that shawls could use loops as a closing mechanism (Hägg 1974, p. 95).

466. Dreibindiger Rautenköper. Der Stoff ist noch weich und schmiegsam, ein wenig durchsichtig. Ausführung ausserordentlich ebenmässig und genau. Farbe dunkelblau, ins Grüne spielend. Mehrere Fragmente, eins 7 x 3.5 cm. Fadendichte per cm 50 x 17.
Geijer 1938, p. 28

466. I graven fanns spännbucklor och ett treflikit spänne <...> Det tyg, som Stolpe åsyftar [som hefter ved en av skålspennene] är en fin, blå, treskaftad diamantkypert, W16 (textilbeskrivning Birka III, s. 28), av vilket lösa fragment finns bevarade men även sådana, som häftar vid metalldelar. Bland de lösa fragmenten tilldrar sig ett (Birka III, taf. 5:5) särskild uppmärksamhet. Det är en kantbit försedd med en ögla av yllesnodd, genom vilken en rostig spännenål sitter trädd (Birka III, s. 159: "Ein stück des Wollstoffes mit angenähter Schnur, D6, in die die Spangenadel hineingesteckt ist"). - Den lösa nålen genom yllesnoddsöglan i grav 466 passar in på plats på baksidan av det treflikiga spännet, från vilket den torde komma, 466:2. Yllekyperten W16 kan därmed antas vara identisk med det av Stolpe omnämnda tyget, som låg över högra spännbucklan, men under det treflikiga spännet. - Ena spännbucklan har i graven pressats mot kistlockets trä. Mellan spännbuckla och trärester finns följande textilskikt bevarade från skalet räknat, 466:5, jfr 466:4. 1) i en smal bandliknande remsa, eventuellt med en kant, 2) halvt förmultnat linne samt 3) yllekypert av W16-typ. Linneresterna (lager 2) mot yllekyperttygets avigsida är förmodligen vad som återstår av ett foder till W16-stycket.

Lagerföljden kan följas åt andra hållet i rester från dräktens ryggsida, vilka bevarats under ena spännbucklan. Samma W16 skymtar där med avigsidan uppåt mellan revbenen, som pressats ned under brättet, 466:3. Ytan är för övrigt täckt av ett fint damm från andra, redan förmultnade textilier.

Fragment av mörkblå yllekypert, W16, fanns alltså dels på ryggsidan och dels på bröstsidan over spännbucklorna tillsammans med ett skikt linne. Ett stycke av yllekyperten hade en bevarad vävkant, i vilken en ögla av yllesnodd var fäst. Genom öglan, som var vänd in mot kroppsmitten, satt nålen till det treflikiga spännet. Med detta torde det stå klart, att yllekyperten var materialet i ett livplagg. Det hade foder av linne. Ett sidenkypertfragment, 466:6, som låg under ylletyget men fortfarande utanpå kjolen, torde inte höra till något av dessa plagg.
Hägg 1974, p. 61

Inte heller förekomsten av en vävkant är nogåt säkert belägg, eftersom man för de flesta plaggen utnyttjade den vävde ytan så långt det gick och undvek skärningar, som reducerar materialet (jfr. Nyléns primära plaggformer, Nylén 1947, s. 167). Stadkanter har m.a.o. förekommit även på de tilskurna plaggen. Det innebär att kantstycken som det från grav 466 även kan ha hört till en tröja. Å andra sidan är inte förekomsten av snoddöglan här ett skäl, som behöver tala för att det rör sig om en tröja. En relativt väl bevarad yllesjal från en kvinnograv i Raisio, Ihala, Finland, (Inv. Nr. 14275: 273) har t.ex. en liknande ullgarnsögla fäst invid kanten.
Hägg 1974, p. 95

Shawls of pattern-woven fabric

Pattern-woven fabric were used in bands decorating the edges of the tailored overgarments. However, there are a couple of graves where the fabric appears to have been used to create a garment. Hägg (1974, p. 95) believes that in these cases, the pattern-woven fabric may have belonged to a cloak-type garment, instead of a duvet or blanket.

Grave 480

The grave contained two fabrics that may have belonged to outer garments. Underneath the body was a large fragment of woollen broken diamond twill, and underneath that, there was a pattern woven fabric (Hägg 1974, p. 61).

Neither fabric had left traces on the top of the oval brooches. However, a ring brooch was found located at waist height in the grave. Hägg (1974, p. 61, 1986, p. 63) proposes that it may have been used to close a cloak of pattern woven fabric. If the cloak was closed at this height, it is likely that it would not have covered the oval brooches. The broken diamond twill may have belonged to another outer garment, possibly closed by a trefoil brooch found in the grave (Hägg 1974, p. 61).

480. Ett ansenligt stycke organiskt material har bevarats under den ena spännbucklan, 480:1 (Birka III, s. 159). Det innehåller följande underifrån räknat, 480:2. 1) trä från gravens botten (Ho), 2) mönsterväv (M), som skymtar genom trasigheter i nästa lager, som 3) utgörs av ett stort sammanhängande stycke diamantkypert (W10). Över diamantkyperten 4) rester av den förmultnade kroppen (K), numera endast på de ställen, som pressats ned runt kanterna under spännbucklans brätten. Diamantkyperten har alltså legat under kvinnans rygg. <...> Mönstervävsfragmenten kan komma från en matta eller möjligen från ryggsidan av en mantel under den döda. Diamantkypertens läge över mönsterväven men under kroppens förmultningsrester visar, att den (W10) hör till bakstycket på ett livplagg. Varken mönsterväv eller diamantkypert har avsatt några spår på spännbucklornas skal.

Ett ringspänne av brons låg i midjehöjd i graven. Det kan ha hört till en mantel, vilket i så fall skulle förklara, varför det inte fanns några sådana spår ovanpå spännbucklorna. Om manteln hållits samman över bröstet med ringspännet på en punkt belägen nedanför spännbucklorna, kan de inte heller ha varit täckta av mantelns framkanter. Eventuellt kommer alltså mönsterväven från en mantel, till vilken även ringspännet hörde, medan diamantkyperten måste härröra från ett annat livplagg, som kan sättas i samband med ett treflikigt spänne, som också fanns i graven.
Hägg 1974, p. 61

Grave 660

Fragments of textiles have been preserved both over and under the oval brooches, allowing the determination of the stratigraphy in the grave.

The outmost layer (M4) is a pattern-woven fabric with a decorative motif in red yarn, that is found on top of the brooches and underneath the body. The thread count is 30 x 6 threads per cm (Geijer 1938, p. 56). There are remains of a 1-1.2 cm wide sprang band that appears to be decorating the edge of the fabric (Geijer 1938, p. 130). Drawing of 
stratigraphy

Underneath the pattern-woven fabric there are remains of another textile; fragments of woollen broken diamond twill (W 20) with a possible linen lining are found on top of the brooches and underneath the body (Hägg 1974, p. 64-65).

Illustration Hägg 1974, p. 127

Geijer (1938, p. 56, 140) interprets the pattern-woven fabric as a possible cloak-type garment. Hägg (1974, p. 64 - 65) acknowledges the possibility, while proposing that the broken diamond twill may also have been an overgarment. There are no additional brooches in the grave, which may indicate that the garment(s) were draped freely around the body, or that they were not draped garments at all.

660. Både över och under spännbucklorna har textilier bevarats, lagerföljden kan rekonstrueras som 660:6 visar. Osäkerhet råder dock beträffande det översta lagret med mönsterväv, M4 (...) "wahrescheinlich über der einen Spange". Detta är ett av de större mönstervävfragmenten från Birka med dekorationsmotivet broscherat i rött yllegarn på en grundväv av hårdspunnen rips och med rester av sprang, D8, från en kantbård (?).

Mot ovansidan av ena spännbucklan har uppenbarligen ett stycke fin treskaftad diamantkypert, W20 (...) legat pressat med avigsidan in mot spännbucklan och rätsidan ut 660:3. <...> Från insidan av ena spännbucklan kommer ytterligare rester, som bildar en fortsättning på lagerföljden, 660:1a. Överst ett förmultningsskikt, (K) och därunder återigen den treskaftade diamantkyperten, W20, med avigsidan vänd uppåt, dvs in mot kroppen. I samband med detta textilskikt har också hittas siden, Birka III, s. 56: "Ganz in der Nähe der Spange lag ein zarter Wollstoff (...) mit Seidenstreifen und warscheinlich direkt darunter das... Ripsgewebe" alltså grundväven i den broscherade mönsterväven. Genom trasigheter i diamantkyperten, W20, från spännbucklans insida skymtar nämligen det understa bevarade lagret som utgörs av samma mönsterväv, M4, som i det större fragmentet.

Det är med utgångspunkt från denna grav, som Geijer (s. 140) snuddat vid tanken, att mönstervävnaderna i birkagravarna även skulle kunna ha använts som material i manteln: "Im Grab 660 diente vielleicht ein mehrfarbiges Gewebe (M4) als Mantel, aber sonst scheint es, als ob dieser Typus eher eine lose Decke dargestellt hätte". Vad som lockade Geijer till detta antagande var, att den med rött ullgarn broscherade ripsvaran, M4, tycks ha legat både över och under den döda så att det är sannolikt, att kvinnan haft den runt kroppen. Detsamma gäller i och för även för diamantkyperten som alltså förekom dels under ryggen dels över ena spännbucklan. Den ligger så som man kan vänta sig att finna resterna av ett yttre livplagg, nämligen i ett sammanhängande lager närmast förmultningsskiktet under spännbucklan samt tätt över skalet, möjligen med ett mellanliggande lager (linne?), som kan tolkas som ett foder. Ett tredje spänne saknas emellertid, vilket kan betyda antingen att kvinnan haft tyget fritt draperat kring rygg och axlar eller att det inte alls är fråga om ett draperat plagg.
Hägg 1974, p. 64 - 65

660. Taf. 37: 2. Fragment von Sprang, vermutlich Teil eines Bandes. Das Garn ist mit sehr fester Rechtsdrehung aus gelbbrauner Wolle gezwirnt und zarter als im vorigen Fragment. Auch hier liegen die Fäden gerade gezogen und dicht beisammen, sodass die im Bild schematisch gezeichneten Fadenbuchten kaum erkennbar sind. Nach Befeuchtung kann das Band jedoch vorsichtig der Breite nach ausgedehnt werden, wodurch die Technik deutlich hervortritt. Die Fadenflechtung ist regelmässig ohne irgend eine Art von Musterung. - Breite 10 bis 12 mm, Länge 3 cm. Das Band lag am Rand des Gewebes M 4. festgeklebt.
Geijer 1938, p. 130

Grave 739

Photo of pattern-woven textiles from Bj739 Unfortunately, there is very little information about the locations of the textiles in this grave. One of the fragments is of a corner of a woollen pattern woven fabric. According to Geijer (1938, p. 54) the fabric was red (possibly the underlying colour of the piece), blue-black, blue-green and white, with a thread density of 7-8 x 3 threads per cm. There was a fragment of a woollen tablet-woven band along the edge. Geijer interprets it as the corner of a rectangular cloak or blanket edged by tablet-woven bands. Hägg (1974, p. 65, 95) agrees.

Photo Gabriel Hildebrand, SHM Creative Commons Erkännande 2.5 Sverige

In addition to oval brooches, there was a small ring-pin and a brooch made from silver sheet metal in the grave (Arbman 1940, p. 263-264). It is unknown whether either of them was used to fasten the pattern-woven fabric.

Från graven kommer flera intressanta textilrester, tyvärr samtliga med mycket otillräckliga lägesuppgifter. Till dessa hör hörnet av en mönsterväv, M 2 (Birka III, s. 166; textilbeskrivning s. 54 f, Taf. 9:3), som efter Geijers beskrivning var röd (eventuellt grundfärg), blåsvart, blågrön och vit, kantad med ett brickvävt band av ylle. Ytterligare ett fragment av samma typ finns bevarat (Birka III. Taf. 9:1). Vi en nyckel i graven låg ännu ett stycke mönsterväv, dock av annan typ, M 5. Om väven av M 2 skriver Geijer följande: "Sicherlich stellt es die Ecke eines rechteckigen Mantels oder einer mit Bändern eingefassten Decke dar" (s. 54). Geijer har uppenbarligen inte i den sammanfattande dräkthistoriska beskrivningen av manteln (Birka III, s 140) byggt sin slutgiltiga bild på intrycken från denna grav eller grav 660, där det också antogs, att mönsterväven skulle kunna härröra från en mantel.
Hägg 1974, p. 65

Möjligen kan det mönstervävda hörnstycket från grav 739 uppfattas som en del av ett icke tillskuret plagg. I vart fall finns det inte några tecken på at mönstervävar (banden undantagna) använts som material i de tillformade plaggen.
Hägg 1974, p. 95

Veka and Hyrt

Several graves from Western Norway were excavated in the late 19th and early 20th century. The textile fragments were later analysed by Hana Lukešová (2011, 2015). Most of the textiles has been stored separately from the brooches, but by using old photographs and marks made from the brooches on the textiles, Lukešová were able to reconstruct the position of the textiles inside each brooch.

When examining 23 different finds in this manner, Lukešová identified 16 instances of outer garments that was laying over the brooches. The majority of these garments, thirteen out of sixteen, were made from rather coarse 2/2 twill. The remaining three were made from broken diamond twill. Many of the outer garments showed visible traces of blue (Lukešová 2015, p. 149-151).

Grave B 6228 at Veka

Grave B 4864 at Veka in Voss, Hordaland contained two oval brooches and a third brooch, that had been placed between them.

Veka fragments A large fragment of 2/2 woollen twill had been preserved, and as it bore stains from both one of the oval brooches and the third brooch, it was possible to determine the exact position of the third brooch.

There was also a small stain (from a round object with a diameter of roughly 5 cm) to the left of the oval brooch, indicating that there may originally have been a fourth brooch in the grave (Lukešová 2011 p. 156-159).

Illustration: Lukešová 2011 p. 160, Modified by adding English text.

The third brooch appears to have fastened a garment made of the 2/2 twill. At one end of the brooch there are fragments of samitum fabric (probably in silk). It appears to have been a band that has been stitched along the selvedge of the 2/2 twill (the stitches are still visible when using a microscope). The selvedge runs perpendicular to the brooch pin. At the other end of the brooch pin, there is remains of a fabric loop (Lukešová 2011 p. 156). The stains indicate that the oval brooches had been placed on top of the 2/2 twill, with their needles piercing the fabric, before fastening the loops of the suspended dress (smokkr) below.

Lukešová interprets the 2/2 twill fragments as the remains of a woollen cloak, decorated with a narrow band of silk samitum sewn to its selvedge. The cloak had two fabric loops, one on each side of the opening, and was fastened using the third brooch (and possibly the missing fourth brooch as well). She suggests that the oval brooches were placed on top of the cloak to display the jewellery of the deceased, and thus do not reflect the way they would have been worn along with the cloak in everyday life (Lukešová 2011 p. 157-159). Similar grave arrangements are known from two graves in Haithabu, where the oval brooches were piercing the underdress (serk), probably to keep the brooches stable when the dead women were carried to their graves, fully dressed and on display (Hägg 1991, p. 278).

Funnet fra Veka inneholdt to ovale spenner samt en komplett tredjespenne (midtspenne). Midtspennen har en oval fasong og et lite hull helt i kanten på den ene langsiden. Jeg antar at hullet har pekt nedover når draktspennen var i bruk (Fig. 5). Nålen på midtspennen er bevart og på denne fins det rester av en stropp på den ene siden og rester av samitum (Bender-Jørgensen 1986:264) på den andre. Samitum er en avansert type mønstret tekstil med to renningssystemer. Dette svært fine fragmentet (sannsynligvis silke) tilhører et bånd som har vært sydd på en jarekant på en kappe i kypert 2/2 (jf. Fig. 6). Sømmen er fremdeles synlig i mikroskop. <...>

I funnet fra Veka finnes det også fragmenter av en kappe i kypert 2/2. Denne har mest trolig vært båret over en selekjole. Kappen var festet med en midtspenne som gikk gjennom to stropper fra hver side av kappen. Sidene på kappen har hatt en jarekant som har vært pyntet med et smalt bånd. Dette båndet er vevd i et mønstret stoff som må ha vært kostbart, sannsynligvis samitum i silke. Stoff av en slik kvalitet som dette er ganske sikkert importert. Spenneparet i dette funnet lå direkte oppå kappen (der er svært tydelige avtrykk og flekker på den). Dette har ført til at spennenålen har måttet gå gjennom kappestoffet for å nå frem til stroppene i kjolen. Denne måten å feste plaggene på kan ikke ha vært særlig praktisk. En kvinne ville i så fall først ha måttet ta på seg kappen og så festet den til kjolen som var under den. Det er derfor rimelig å anta at denne måten å feste spennene på kun har vært benyttet ved selve gravleggingen. På den måten kunne alle som fulgte avdøde til graven se hvordan den døde har vært pyntet med draktsmykker uten at tekstiler har dekket til spennene.

Den opprinnelige plasseringen av den ovale midtspennen kunne bestemmes svært nøyaktig takket være en tydelig flekk og et hull på et stort fragment av kappen. Flekken passer nøyaktig til formen på spennen. Dessuten ble det også funnet rester av kypert fra kappens hull på baksiden av spennen. Det finnes også et tydelig avtrykk i det øvre venstre hjørnet av det store kypertfragmentet. Formen på flekken tyder på at den har vært laget av et rundt metallobjekt som har hatt en diameter på ca. 5 cm. Etter min mening kan dette muligens tolkes som avtrykk etter en mulig fjerde spenne som var typisk for festing av kapper funnet i Birka (Hägg 1971:143-145).
Lukešová 2011 p. 156-159

Grave B 4864 at Hyrt

Hyrt fragments Grave B 4864 at Hyrt in Voss, Hordaland, in Norway, contained two oval brooches and the remains of a third brooch.

Illustration: Lukešová 2011 p 158. Modified by adding English text.

The textile find contains among other things very fine dark blue broken diamond twill, blue tabby, coarse 2/2 twill, and a small piece of yarn (Lukešová 2015, p. 149).

It was possible to identify a sequence of three types of weaves: broken lozenge twill, blue tabby, coarse 2/2 twill (Lukešová 2011 158). A photograph of the stage before the separation helped to identify an oval fragment of coarse 2/2 twill. It is clear that the fragment was under the brooch and a pin did not pierce it. It completely covered the whole area of the brooch inside, which means that this fragment does not represent a suspended dress. The upper hem of a suspended dress usually ends approximately within the first third of an oval brooch. The fragment of 2/2 twill might thus be some kind of tunic. The blue tabby made from flax or nettle probably belonged to a shift. Two upper and two lower loops on brooch II indicate that there were two strapped gowns. The top layer of fine diamond twill, which is to be observed on the old photograph as well, suggests some type of a cloak - an outer garment.
Lukešová 2015, p 149

Hyrt fragments There is a clear impression left by the third brooch on a fragment of broken diamond twill (this fragment also has a selvedge). The fragment was part of the top layer - worn over the oval brooches (Lukešová 2011 p. 156). Based on the evidence, Lukešová believes the third brooch fastened the edges of a woollen cloak made of broken diamond twill, similarly to what was the case in the grave from Veka.

Photograph: Lukešová 2015, p 146.

Funnet fra Hyrt inneholder to ovale spenner og rester av en tredjespenne (Fig. 3). Et tydelig avtrykk fra deler av denne spennen finnes på et fragment av diamantkypert med jarekant. Jeg antar at denne tredjespennen har vært benyttet som en midtspenne som har hatt som funksjon å knytte sammen jarekanter på en kappe. Denne tolkningen støttes også av lagrekkefølgen i tekstillagene som viser en diamantkypert som øvre lag. Sammenføying av to jarekanter med en midtspenne er også påvist i funnet fra Veka (jf. Fig. 6).
Lukešová 2011 p. 156

Lukešová does not discuss exactly how the third brooch would have closed the cloak or shawl. Given that there is no indications that the selvedge has been pierced at the site of the brooch, it is possible that this garment was fastened with fabric loops similar to the overgarment in B 6228 at Veka, but that the loop has deteriorated. There is no information about the position of the third brooch in the grave. However, if the brown stain on the right of the fragment is from one of the oval brooches, it would place the third brooch at the height of the centre of the oval brooches.


Haithabu

Various excavations of the former Viking settlement of Haithabu, near Schleswig Holstein in Germany, have yielded a large amount of 10th century textile material. Hägg (1991, p. 277) notes that there are fragments of overgarments in several of the graves. However, it is difficult to identify whether they are from cloaks or tunics/jackets. Most are made of diamond twill (in the female graves), although other weaves are found as well. Two graves with fragments that may be cloaks or shawls are described below.

Från ett av de yttre plaggen - mantel eller övertunika - kommer några jämförelsevis väl bevarade stycken. Materialet är huvudsakligen diamantkypert (enbart i kvinnogravar; Abb. 67,2; 75,2 etc.), men andra kypertbindningar förekommer också. I ett exempel, diamantkyperten i grav 159/1960, är ullfibrerna tydligt rödfärgade.
Hägg 1991, p. 277

Grave 159/1960

The grave contained remains of an overgarment that had been fastened by a rectangular brooch. The brooch was found in a position between the oval brooches, and the remains of its needle was piercing through the opening of a fabric loop. Drawing of textile 
fragment and brooches

While currently little fabric remains on the brooch itself, Schlabow (referred in Hägg 1991, p. 144) reports that there were fragments of woollen diamond twill found on its underside. Hägg confirms that this is likely, given the rusted discoloration on some of the fragments of the preserved diamond twill, and the tiny fragments of twill that remains on the brooch (Hägg 1991, p. 144).

Illustration Hägg 1991, p. 137

The stratigraphy in the grave is complex, and consists of several layers of textiles. According to Hägg, the rectangular brooch was used to close a garment that was worn over a smokkr and serk. This garment, made of red-brown diamond twill (12 threads/3mm x 7 threads/3mm), would have had fabric loops sewn to its edges (Hägg 1991, p 144-145, 277).

Die Rechteckfibel lag nach dem Röntgenbild zwischen den beiden Schalenfibeln, etwas schräg zur rechten Fibel hin. Aus den erhaltenen Fragmenten ergibt sich, daß die Nadel an der Unterseite der Rechteckfibel, von oben aus gerechnet, zuerst durch eine Stofföse (159:9) gestochen war. Die Öse muß deshalb an dem Gewand befestigt gewesen sein, das mit der Rechteckfibel zusammengehalten wurde. <...>

Der Rautenköper (159:13; Abb. 67.2) dagegen läßt sich nur an Hand der sonstigen Fragmente in die Schichtenfolge einordnen. Die Angabe Schlabows, daß dieser Stoff von der Unterseite der Rechteckfibel stammen soll, wird durch Verfärbungen der Stofffläche mit Grünspan und Rost der Fibel und der eisernen Fibelnadel sowie durch die eben besprochene Übereinstimmung in der Qualität dieses Tuchrestes (Fragment 159:12) mit jenen Resten an der Nadelrast (Fragment 159:10) bestätigt. Da die bereits rekonstruierte Schichtenfolge (vgl. S. 142) auf ein Obergewand aus Rautenköper mit z-Kette und s-Schuß schließen läßt, darf angenommen werden, daß das von der Rechteckfibel gelöste Rautenköperfragment (159:13; Abb. 67.2) vom gleichen Kleidungsstück stammt, und zwar entweder von seiner Rückseite (Abb. Schicht 6) oder, eher, im Hinblick auf seinen guten Erhaltungszustand, von seiner Vorderseite, wie alle übrigen Gewebereste an der Rechteckfibel (Schicht 10). Da die Rechteckfibel zu einem Gewand gehört hat, das über Hemd und Trägerrock getragen wurde und da dieses Gewand mit Bandösen versehen war (159:9), durch die die Fibelnadel stach, dürften ursprünglich Rautenköperstoff und Bandöse miteinander vernäht gewe sen sein und beide zu Schicht 10 gehört haben.
Hägg 1991, p. 144, 146

Grave 900/1960

A bronze brooch (with a thin iron pin) was found on top of a layer of diamond twill, with the pin piercing the twill. Underneath the diamond twill, was a layer of tabby, that Hägg believes belonged to a dress or smokkr. The grave also contained a knife in a leather sheath, that had almost the entirety of one side covered with the diamond twill. The position of the knife in the grave is unfortunately not known (Hägg 1991, p. 158-159).

Stray finds of possible cloak fragments

The lack of distinctive features makes it almost impossible to identify fragments of unshaped wrap-type garments like shawls and cloaks outside of the context of graves and their stratigraphy. However, Hägg (2015, p. 55) notes that under certain circumstances, the selvedges can be viewed as a identifying criterion, namely when they have been worked particularly carefully. The quality of the fabric may also be relevant - loose and woolly weaves would not be suitable for cutting into smaller pattern pieces or making tailored garments, but it fulfils all the requirements for cloak material.

Based on these criteria Hägg (1984, 1991, 2015) identifies some fragments among the stray finds from the harbour and settlement at Haithabu as possible cloak fragments. As these are stray finds, we have no information about the gender of the wearer.

Fragment H 44B

A fragment (66.5 cm x 19 cm) of fine woollen 2/2 twill (0.2 cm thick) was found in the harbour. The fabric is of high quality, remarkably soft and loose and made of fine wool yarn. It is covered on both sides with fibre pile. It may have been dyed, and is currently reddish-brown. Three of the sides of the fragment has been cut off or torn off almost straight. There is a tablet woven selvedge preserved on the fourth side (Hägg 1984, p. 63-64, 180-181).

Cloaks with edges trimmed by the use of tablet weaving have been well known since Roman times. Hägg (1984, p. 180) states that these selvedges should be regarded as a fairly reliable feature of the open cloak, and thus identifies H 44B as part of such a garment.

Fragment H 84

Another fragment from the harour is a piece of 2/2 twill (0.3 cm thick) with dark and light checks. The light checks are currently in yellowish and red shades. The fragment is poorly preserved, and it is not possible to discern its original shape. The weave is very loose, almost transparent. Hägg (1984, p. 63-64, 181-182) interprets the fragment as belonging to an unshaped garment, like a shawl or cloak, and argues that with such garments the emphasis is placed on the aesthetic effect of the patterned fabric surface.

Fragment S 8A-H

S 8A-H is a collection of eight differently sized fragments of a fabric in a fine quality 2/2 twill that was found at the settlement. The fabric is densely woven, but still thin and flexible, and Hägg (1991, p. 45) believes it originally was dyed red. Currently, the fragments are badly damaged, worn and with unevenly torn edges. The fabric has a badly worn fibre pile.

Two of the fragments, S 8B and S 8C, have remnants of a carefully worked selvedge (Schlauchkante - "tube selvedge"), which would have been created along the side edges of the fabric during the weaving, probably on a warp-weighted loom.

A third fragment, S 8A, is folded over and hemmed along one edge. The seam is currently almost deteriorated, as the thread was made of vegetable fibres (possibly linen). It runs across the warp threads of the woven fabric. The hem has the same width as the carefully created selvedges along the side of the fabric, and would have been perpendicular to the selvedge (Hägg 1991, p. 45).

Based on the focus on creating strong edges - by the carefully worked selvedges, and the perpendicular hem - and the straight shape indicated by the fragments, Hägg (1991, p. 45, 48) interprets them as the remains of a rectangular cloak.

Fragment S 8A-H (...). Acht unterschiedlich große Teile eines Gewebes in feinem, ursprünglich wahrscheinlich rotem Gleichgratköper 2/2 (...). Alle Fragmente sind stark beschädigt, ungleichmäßig abgerissen und abgenutzt, stellenweise auch verfilzt und eingelaufen. Der Stoff ist dicht, aber dünn und biegsam und besitzt einen stark abgenutzten Faserflor.

Zwei der Fragmente, B und C, haben Reste von einer Schlauchkante (...), die beim Weben an den Seitenkanten der Stoffbahn ausgeführt wurde. Ein drittes Fragment, A, ist schmal umgeschlagen und an einem Rand dicht abgesäumt. Die Naht mit Fäden aus jetzt fast völlig vergangenen pflanzlichen Fasern läuft quer zur Richtung der Kettfäden der gewebten Stoffbahn. Die Schlauchkante und der Randsaum sind gleich breit. Sie müssen ursprünglich rechtwinklig aneinander gestoßen haben.
Hägg 1991, p. 45

Fragment S49A

Fragment S 49A

A four-sided piece of medium-fine 2/2 twill (40 cm x 12 cm). The fabric is strong but still flexible and has a slight fulling and roughening. Along one of the sides, a simple selvedge has been preserved. Here, the fabric has been folded over once, creating a fold 4-4.5 cm wide, and stitched in place with a seam ca. 1 cm inside the fold. The other edge of the fabric is torn off, and there is a hole inside the edge Hägg (1991, p. 47). It is a simpler, and less fine fabric than S 8, and was not dyed.

Illustration: Hägg 1991, p. 47, slightly modified. 1:seam, 2:selvedge.

Hägg (1991, p. 48) identifies the fragment as a possible piece of a rectangular cloak, where the edge has been reinforced by folding the fabric and stitching (as an alternative to the carefully worked selvedges found in S 8).

In summary, there are few similarities between these four sets of possible cloak fragments from the settlement and harbour, apart from the type of weave (2/2 twill). One fabric is checked, with fluffy long fibres like a mohair fabric (H 84), one has twisted warp threads and with its roughened surface looks almost like a very fine animal hide (H 44B). The fragments from the settlement (S 8 and S 29), show some fibre pile, but otherwise have quite rough surfaces due to their many outer hairs. However, the practical properties of the fabrics are similar: They are warm, highly moisture-repellent due to the fibre material and treatment, and flexible, which is an important feature for cloak fabrics with regard to the mobility of the wearer (Hägg 1991, p. 48).

Other minor fragments

The harbour also yielded several fragments of lesser quality, that Hägg (2015, p. 56, 267-268) believes may be the remains of rough, travelling cloaks or possibly blankets. One similar fragment was found in the settlement. The fragments are severely shredded and deteriorated, making the identification very uncertain. However, they are mentioned here for completeness.

The fragments are mainly of very loose and flexible 2/1 twill with a one-sided fibre pile, often with weaving yarns made from different natural-coloured sheep's wool, or sometimes a 2/2 twill that had been roughened on both sides.

Fragment Weave Roughened
(One or two sides)
Description
H 1 Cross twill I, II? Very coarse, undyed light and dark weft threads
H 4 Cross twill I Undyed light and dark warp threads
H 26 2/1 twill   Very coarse weft threads; undyed light and dark warp threads
H 27A 2/1 twill I Very coarse weft threads
H 27B Cross twill Very coarse weft threads; undyed light and dark warp threads
H 28B-D 2/1 twill I Very coarse weft threads; dyed
H 31A 2/2 twill Very coarse weft threads; undyed light and dark warp threads
H 31B-D 2/2 twill I, II Very coarse weft threads
H 32A-B 2/1 twill I Undyed light and dark warp threads
H 37 Cross twill Very coarse weft threads; undyed light and dark warp threads
H 44A 2/2 twill I Coarse weft treads
H 47 2/1 twill I? Very coarse weft treads
H 49A 2/2 twill I, II Undyed light and dark warp threads
H 51 2/2 twill Very coarse weft treads
H 54B 2/1 twill I, II Very coarse weft treads
H 56B 2/1 twill I Very coarse weft threads; undyed light and dark warp threads
H 58 2/2 twill I, II Undyed light and dark weft threads
H 63 2/2 twill I, II? Very coarse weft threads; undyed light and dark warp threads
H 69 Cross twill Very coarse weft threads
H 70A-B Cross twill Very coarse weft threads
H 71 2/2 twill Very coarse weft threads; dyed
H 74 Cross twill Coarse weft threads; undyed light and dark warp threads
H 79 2/2 twill I, II Very coarse weft threads; undyed light and dark warp threads
S 46 2/1 twill I Very coarse weft threads
Overview by Hägg (2015, p. 56).

Grave C 27997 C from Kaupang

The grave was excavated in 1949 in Kaupang in Vestfold and is dated to the period 800-850 AD (Skre 2007, p. 116-117). It is the richest of the female burials in the entire Kaupang complex, containing a 30-40 years old woman probably buried in a boat. In addition to a couple of oval brooches, she had an equal-armed brooch, a rectangular brooch, four arm-rings and a number of different chains and neck-rings (Ingstad 1979, p. 160).

This burial has yielded c. 41 textile fragments. Some of these still lie in cakes of several layers, others have clear imprints of the bronze objects with which they were entangled. As far as it is possible to tell, the fragments represent five different fabrics. Three of these are in broken lozenge twill, one is a 2/2 diagonal twill, and one is a tabby. Several of the fragments have a black crust on one side, and this may mean that the grave once held a fabric of vegetable fibre, which has now disappeared. This may well have been a linen fabric.
Ingstad 1999, p. 267

Two of the textiles, a diagonal twill and a finely woven broken diamond twill, are interpreted as belonging to two suspended dresses (smokkrs), worn together with the oval brooches. The rectangular brooch was pinned to the front of these garments (Ingstad 1979, p. 162).

The two remaining broken diamond twill textiles in the grave were found in direct contact with a neck-ring (or possibly arm-ring) in bronze and several small bronze spirals. The spirals were probably originally attached to the chains that hung between the oval brooches. Ingstad (1979, p. 162) interprets the fragments as a finely woven twill cloak or other overgarment with a rougher twill used as lining. The cloak would have laid in direct contact with the neck and chest of the woman, resulting in the fabric being preserved on the jewellery.

Photo of fragment of twill a
from grave C 27997 C The fine twill (potential cloak fabric) is woven in a rare pattern - combining broken diamond twill with chevron twill, with constantly changing repeats. The seven fragments that have been preserved are of a very high quality, with 40 - 44 /14 threads per cm. One piece has a preserved starting border (Ingstad 1979 p. 160, 1999 p. 267).

Photo: Ingstad 1979, p 160

The rougher twill interpreted as a lining is broken diamond twill with 24/12 threads per cm. The largest piece is 12 cm x 6 cm, and has part of a starting border. Small pieces of the same starting border is preserved on three other fragments (Ingstad 1979 p. 161, 1999 p. 267).

When the equal-armed brooch was found (in 1949) there were remains of tabby on the back. Unfortunately, when the find was re-examined in 1972 these textile fragments were gone (Blindheim et. al. 1981, p. 219). There is no record of the position of the brooch. However, considering that the other brooch in the grave (the rectangular brooch) was pinned to the smokkrs, I find it plausible that the equal-armed brooch fastened the cloak. If this was the case, the tabby that disappeared may have been remains of loops from the front of the cloak, similar to what is found at Birka and Veka.

Finally, there are two fragments of a fine, loosely woven, woollen tabby (20/12 threads per cm) that is interpreted as a possible scarf or shawl (Ingstad 1979 p. 162, 1999 p. 267-268).

Det opplyses videre at en grovere og en finere kvalitet av ringvend fantes i direkte kontakt med en kraftig arm- eller halsring av bronse, samt stykker av små bronsespiraler. Birgit Heyerdahl-Larsen tenker seg helst at disse bronsespiralene, som det fantes flere av i graven, har vært festet til kjeder som har vært opphengt mellom de ovale spennene. Disse to kvaliteter av ringvendtekstiler kan muligens skrive seg fra en kappe eller et annet ytterplagg. Vi kan tenke oss at kanskje det originale og fine stoffet (a) skriver seg fra dette plagget, og at det grovere ringvendstoffet (b) har vært for i den. Denne kappen må ha ligget i umiddelbar kontakt med kvinnens hals og brystparti ettersom halsringen og bronsespiralene heftet til dette stoffet. Det er mulig at den har ligget over den døde kvinnen i graven.
Ingstad 1979 p. 162

Likearmet spenne av bronse, brutt i to nå sammenlimte deler. <...> På baksiden nålefeste og nåleskjede, begge med rester av jernnålen i. På baksiden i 1949 rester av toskaftet stoff, i 1972 ikke lenger til stede.
Blindheim et.al. 1981, p. 219


Hørning

Sketch of grave at 
Hørning A chamber grave was found underneath a church during an excavation in 1960 in Hørning in Denmark. The deceased was placed in the body of a carriage, a burial custom that signals that she was a high-status woman, and possibly Christian or with Christian connections. The grave is dated to the 10th century based on the grave goods and the use of a carriage as coffin. Little of textiles remained, but there were fragments of several tablet-woven bands with in-woven silver threads (Hedeager Krag 1990, p. 352).

One band was 5 cm wide and 2,5 m long, reaching from the brow of the deceased woman and down to her knees. The position of the band in the grave shows that the garment it once decorated was hanging straight down from the head on the left side, but had been folded around the right arm (Voss 1991, p. 194). The band was sewn to a thin fabric that appears to have been dyed blue (Hedeager Krag 2003, p. 70).

Illustration: Hedeager Krag 1990, p. 353

Other woven bands with metal threads had been preserved at the wrists, shoulders and thigh, indicating that the woman was wearing a decorated tunic. The deceased wore no brooches, oval or otherwise. Hedeager Krag (1990) argues that the Hørning grave and a similar chamber grave from Hvilehøj indicate that high status clothing in Denmark at the time was continental in style and clearly influenced by Frankish fashion (Hedeager Krag 1990, p. 353-357).

I graven fantes flere textilfragmenter, blant annet sølvindvirkede og brikvævede bånd. Desværre var de i meget dårlig stand. De ganske tynde sølvtråde var helt opløste, men nogle af de udskilte metalsalte har virket konserverende, således at små stykker textil er bevaret. Det ca. 5 cm brede bånd, der gik fra den dødes pande og ned til knæene på hver side af kroppen, kan have været kantebånd på en kappe. Der kan også udskilles flere 2-2,5 cm brede bånd med sølvindvirkede mønstre. Disse, der lå ved skuldrene, ved håndleddet og udfor lårbenet, kan have kantet en tunika. Såvel gravformen som flere af gravgaverne tyder på en kvindegrav med en datering i det 10. årh.
Hedeager Krag 1990, p. 352-353.

Den døde har været iført en dragt kantet med sølvindvirkede brikvævede bånd. Et af disse var et ca. 5 cm bredt og 2,5 m langt bånd med rhombeformet mønster, der fra issen strakte sig ned på begge sider af kroppen. Dragten må have hængt løst ned på venstre side af kroppen, medens den på den andre side var foldet op om højre underarm.
Voss 1991, p. 194

Et bredere brikvævet bånd, omtrent 2,5 m langt, gik fra den gravlagtes pande og ned til knæene på hver side av kroppen. Båndet kan stamme fra en kantning til en palla, som i europæiske sammenhænge kendetegner den kristne kvindes dragt (...). Ydermere er disse bånd af stor interesse, da de ifølge Nationalmuseets protokoloplysninger er påsyet et tyndt stykke stof, som ser ut til å være farvet blåt.
Hedeager Krag 2003, p. 70

Gotland

The island of Gotland, perhaps due to its position between Sweden, Finland and the Baltic areas, appears to have had its own clothing style during the Viking Age. During the period 500-700 AD regional brooch-types are emerging, and by the Viking Age many of the brooches in use are unique for the island (Anon 2011, p. 123, 124).

Photo of grave from Vallstena in Gotland The usual combination of brooches found in the graves is a pair of garment pins, a pair of animal-head brooches and a third brooch.

The two pins are usually found at the shoulders, sometimes with remains of woollen fabric that has been pierced. They may have been fastening a peplos-type garment similar to the ones worn in Finland and Latvia at the time (Thunmark-Nylén 2006, p. 432-433). Alternatively, they may have fastened a shawl or scarf (Toplak 2016, p. 124-125).

The animal-head brooches sometimes held strings of beads, but interpretations differs in regard to what garment they were fastened to.

Photo: Thunmark-Nylén 2006, p. 434. Grave at Vallstena. Two garment pins close to the neck, two animal-head brooches on each side of the chest, and a third brooch (a box-brooch) underneath the chin.

Die Charakteristika der gotländischen Frauentracht bestehen, in Hinblick auf den Trachtschmuck, aus einem Paar Trachtnadeln, einem Tierkopffibelpaar und einer Bügelscheiben- oder Dosenfibel als Drittfibel (III:45:2). Hinzu kommt die Gerätefibel mit Bändern oder Ketten als die Träger des Scheiden-Messers, der Schlüssel und eventuell anderer Gerätschaften. <...>

Die Trachtnadeln findet man im Normalfalle paarweise, beiderseits des Halses liegend. Einsame Trachtnadeln auf einer Seite des Halses kommen ab und zu vor und können möglicherweise um eine Holznadel ergänzt worden sein, die in der Erde verging. Ausnahmsweise hat man eine einsame Trachtnadel in der Halsgrube angetroffen, und man kann sich dann fragen, ob jene ein Umhängetuch oder den Halsschlitz eines Hemdes zusammenhielt.

Die Nadelspitzen zeigen am häufigsten nach hinten, und auch die halsnahe Lage zeigt an, daß die Nadeln hoch oben auf den Schultern saßen. Die wenigen Beispiele für Trachtnadeln in situ in Stoffstücken weisen darauf, daß man sie durch Wollstoff hindurch steckte. Ein auf die angegebene Weise zusammengehaltenes wollenes Kleidungsstück kann vom faltenreichen Peplos-Typ gewesen sein. Jenes Modell ist sowohl aus Finnland als auch aus dem livischen Gebiet bekannt. Thunmark-Nylén 2006, p. 432-433

The potential differences in clothing style means that archaeological evidence from Viking Age Gotland is of limited use when interpreting garments from mainland Scandinavia. On the other hand, the overgarment depicted on the woman on the Tjängvide picture stone in Gotland is almost identical to e.g. the pendant with a female figure found in Hjorthammar in Sweden, an indication that unshaped overgarments like shawls and cloaks may have been similar in shape and function on Gotland and on the mainland, even if the garments they covered were not.

Photo of picture stone from Tjängvida in Gotland Photo of pendant from Hjorthammar in Sweden
Picture stone from Tjängvida.
Photo: Wihlborg 2019, p. 80
Pendant from Hjorthammar.
Photo: By Berig, CC BY-SA 4.0

The Gotlandic overgarments were usually fastened in front with a box-brooch or bow-brooch. Sometimes a single animal-head brooch or a ringed brooch was used instead, and very occasionally a mainland-style round or equal-armed brooch is found. The box-brooches are almost always found in the same position as the earlier disc-on-bow brooches, high up on the chest underneath the chin. Most of the other brooches fastening the overgarments are found in the same position, but there are a few found lower on the chest (Thunmark-Nylén 2006, p. 435).

Die Drittfibel - Dosen- oder Bügelscheibenfibel (vgl. S. 677 f.) - findet man so gut wie immer unter dem Kinn, in der Halsgrube, hoch oben auf der Brust etc. Nicht nur die Lage der Dosenfibeln in den Skelettgräbern (III:45:4:a-b), sondern auch die charakteristische Abnutzung der Fibeln zeugen davon, daß jene dicht unter dem Kinn getragen wurden (III:45:5; vgl. III:1:1). <...>

Auch eine Tierkopffibel kann in der Halsgrube auftreten oder aber mitten auf der Brust, sowohl in Gräbern mit Tierkopffibeln als Paarfibeln als auch in Gräbern ohne solche Paarfibeln. <...> Ringfibeln, die ja normalerweise in die Männertracht gehören, können als Paarfibeln und als Drittfibel auftreten, und eine einzelne Tierkopffibel kann, wie gesagt, als Drittfibel dienen. In derselben Funktion treten gelegentlich festländische runde und gleicharmige Fibeln auf. Von den wenigen Kleeblattfibeln wurde jedoch keine einzige bei einer fachmännischen Grabuntersuchung angetroffen (S. 91).
Thunmark-Nylén 2006, p. 435

Grave 141/1964 at Kopparvik

The grave is dated to the period 980-1020, and contained two animal-head brooches and a box-brooch.

Two fragments of a tablet-woven band has been preserved on the back of the box-brooch. The bands are 1 cm wide, in dark blue wool, and is woven in a simple geometric pattern, using twelve tablets. The fragments (currently 16 and 21 cm long) was probably decorating the edge of a shawl or cloak that had been fastened with the box-brooch (Toplak 2016, p. 137).

Die einzigen großflächig erhaltenen Textilfragmente sind die Reste von zwei Bändern aus Brettchenborte aus dunkelblauem ("blåsvart") Wollgarn, die in der Dosenfibel aus Gr141 festkorrodiert waren, <...>
Die beiden Bänder aus Grab 141 sind 1 cm breit und 16 bzw. 21 cm lang und haben vermutlich als Verschluss eines Schals oder Mantels gedient, der mit der Dosenfibel auf der Brust geschlossen wurde. Die Borte wurde mit zwölf Brettchen gewebt und zeigt ein einfaches geometrisches Muster.
Toplak 2016, p. 137

A Merovingian cloak from Sandegårda

Fragments of an overgarment has been preserved on a disc-on-bow brooch from an early 8th century grave in Sandegårda in Gotland.

Photo of brooch with textiles

While the grave is from an earlier period, its well-preserved fragments may provide valuable insights in the overgarments that preceded the Viking Age garments.

Photograph: Geijer 1942, p. 373

The pin of the disc-on-bow brooch passes through two types of fabric. Closest to the body is a diamond twill fabric, held together in the front by the brooch. Both fragments of the diamond twill had a selvedge at the lower edge. Geijer (1942, p. 374) interprets this as a possible headscarf.

Drawing of layers of textiles A coarser herringbone twill fabric was laying on top of the diamond twill (Hägg 1971, p. 143). This fabric had been folded, with one fold running perpendicular to the brooch on one side, and another fold running parallel to the brooch on the other side (Geijer 1942, p. 374).

Illustration: Geijer 1942, p. 374. Diamond twill = a, Herringbone twill = b

Both Geijer (1942, p. 374) and Hägg (1971, p. 143) interprets the herringbone twill as belonging to a possible cloak.

Spännet är ett s. k. ryggknappspänne av vanlig gotländsk typ och kan dateras till cirka 700. Ylletyget är av två slag: dels det nyssnämnda fina gåsögontyget, dels en ordinär, medelgrov kypert som kan ha utförts ungefär var som helst. <...> Nålen tar igenom de båda tygsorterna vardera två gånger på ett sätt som visar att spännet sammanhållit de båda ändarna av tvenne på varandra lagda plagg - huvudduk och mantel? Närmast kroppen (överst på fig. 2) låg gåsögontyget, a; de båda hörnen ha bevarade slädkanter, som tyda på att hopfästningen varit intill dukens ena kant. Ovanpå detta satt "manteln" av det kraftigare tyget, b, som är dubbelvikt; de båda snibbarna möta varandra på olika ledd. Av bilden framgår att de sannolikt ha varit hopfästa hörn mot rak kant, vilket ju ger ett smidigare fall.
Geijer 1942, p. 373-374

Ett ryggknappsspänne sitter tätt under hakan på den lilla kvinnofiguren av silver från Aska, Hagebyhöga sn, Östergötland (...). Det kan jämföras med rygg-knappsspännet från en kvinnograv i Sandegårda, Sanda sn, Gotland, vars nål sticker genom mantelns båda framkanter (dvs. det fiskbensmönstrade tyget under diamantkyperten, fig. 2-3, som härrör från ett annat plagg).
Hägg 1971, p. 143


Peripheral finds

The archaeological evidence for cloaks and shawls in Scandinavia during the Viking Age is limited to fragments. However, similar garments from nearby areas, or from earlier and later periods, may provide information about the function, use and possible shapes of such garments.

Cloaks made of textiles are known throughout the entire Scandinavian prehistoric period from many different contexts, especially bog finds, burials and weapon deposits dated to the Early Iron Age (...). The vast majority of these finds are largely intact, while similar finds dated to the Late Iron Age, are highly fragmented.
Mannering 2017, p. 152

Oval cloaks

Some of the earliest textile cloaks known from Scandinavia are the well-preserved oval woollen cloaks found in four Bronze Age graves from Borum Eshøj, Trindhøj and Muldbjerg, all in Jutland in Denmark. The deceased were men, and the graves are dated to the period 1345-1365 BC (Mannering 2017b, p. 12).

Oval cloak from Muldbjerg
Photo: A wool cloak from Muldbjerg (1365 BC). Mannering 2017, p. 154

The use of oval shaped cloaks continued into the Pre-Roman Iron Age (500 - 1 BC), as shown by the find of such a cloak in the Gerum bog in Sweden. The Gerum cloak is dated to 360-100 BC and measures 250 x 199 cm (Mannering 2017, 153-154).

Photo of cloak from Gerum, folded over the shoulders Photo of oval cloak from Gerum

Photos: A wool cloak from Gerum by Swedish History Museum

The Gerum cloak was originally woven as a rectangular cloth, that had been cut into an oval shape and its edges had been overcast. It appears to have been folded in half and worn over the shoulders. The interpretation of how it was worn is based on matching up several places where the cloak has been penetrated by a sharp object. The cuts are consistent with someone stabbing the wearer several times (Sundström 2021, p. 24).

Rectangular cloaks and shawls

The Gerum cloak demonstrates that the oval shape was in use in the Pre-Roman Iron Age. However, most of the preserved cloaks and shawls from Scandinavia during this and later periods are rectangular.

Pre-Roman Iron Age (500 - 1 BC)

Rectangular cloak-type garments from this period can roughly be divided into three sizes; shawls or scarfs (150 cm long and 50 cm wide) that are worn on the upper body around the shoulders or neck, cloaks or wrap-around garments (150 x 100 cm), and blankets (200-250 x 150 cm) that are sometimes also draped around the body. The shawl or scarfs usually have fringes at both ends. All three sizes appear to have been worn by both genders (Mannering 2017b, p. 20).

De oftest forekommende dragter i moserne er rektangulære stykker tekstil, som har lukkede vævekanter på alle fire sider. De kan deles op i tre forskellige størrelser, kaldet tørklæder, kapper og tæpper. Fundkonteksterne viser, at disse forskellige typer slå-om klædninger blev draperet omkring kroppen og har tilsyneladende været brugt af både kvinder og mænd. Tørklæder har generelt en størrelse på cirka 150 x 50 cm og i de fleste tilfælde frynser på begge kortsider. Tørklæder blev båret på overkroppen omkring skuldrene eller svøbt omkring halsen som i Huldremosekvindens tilfælde. Kapper er større rektangulære stykker stof på cirka 150 x 100 cm, og disse tekstiler kunne svøbes omkring overkrop og fungere som en slå-om-kjole (...). Tæpper er den største type af de aflange stykker tekstil fra denne periode.
Mannering 2017b, p. 20

Rectangular scarf or shawl
Photo: A wool shawl or scarf from Huldremose bog (350-41 BC). Mannering et al. 2017, p. 122

Roman Iron Age (1 - 400 AD)

The rectangular cloak shape remain the most commonly used in this period.

Several finds from Scandinavia and northern Germany indicate that cloaks were woven in one piece, sometimes with in-woven elaborate tablet-woven borders and tassels at the corners of the cloak (Nockert 1991, p. 122). Not all of them were produced locally. The Obenaltendorf cloak have Roman origins (Möller-Wiering 2011 p. 119). This may be the case for some of the other cloaks as well.

Roman Iron Age Cloaks
Illustration: Cloaks found in Germany. After Wild 1988. Mannering 2017, p. 153

The decorations of these cloaks vary in complexity, from broad tablet-woven borders (in what was clearly high-status garments) to simpler starting cords at the border. However, there is no structural differences separating the magnificent high-status cloaks from the simpler versions (Wild 1988, p. 83).

Schlabow lists at least 16 cloaks (some of them converted into shirts and trousers), and they can be arranged in a continuous spectrum from the most elaborate, Thorsberg Prachtmantel I, to the simplest, Neddenaverbergen, Obenalterdorf. <...> I submit that we cannot divide one group of cloaks from the rest on structural criteria.
Wild 1988, p. 83

There are also finds of smaller shawls or scarves. In two graves from Lønne Hede in Denmark the upper part of the body of the deceased was wrapped in a rectangular garment of woollen twill. Fragments of a tubular selvedge and remains of fringes found in one of the graves indicate that these were shawls with selvedges along two edges and fringes at the ends. Another grave from Lønne Hede contained a small (100 x 40 cm) scarf or shawl with fringes along all four sides, covering the body of what may have been a child (Demant 2007 p. 86-88).

Another example is the partially preserved cloak that was found wrapped around a girl in a Roman Iron Age grave at Hammerum in Denmark. This cloak was larger, with a width of at least 130 cm, and a length of 95 cm. It had long fringes (minimum 27 cm) along the lower edge of the cloak (Mannering and Ræder Knudsen 2019, p. 71-72).

Migration period (440 - 550 AD)

The find material from the Migration period is more fragmented than the earlier periods, but still indicate the use of rectangular cloaks. In high-status graves like the Högom find, or Evebø, the cloaks were decorated with complex tablet-woven bands that were sewn onto the cloak (Nockert 1991, p. 129). At the same time, the large woollen cloak from Vejen in Denmark, believed to be from the Migration Period (Hald 1950 p. 63, Nockert 1991, p. 122), have simpler tablet-woven bands woven as part of the cloak on all four sides.

The oval cloaks of the Bronze Age were cut from a square piece of cloth, while the cloaks of the Iron Age retain all four edges of the weave. The shape of the cloak probably changed during the pre-Roman Iron Age. The cloak of the Roman era is distinguished by being woven, usually, in one piece instead of being cut to length. Cloth from the warp-weighted loom is characterised by completed edges on at least three sides - a starting border and selvages - and sometimes on the fourth side as well the finishing border. Cloaks woven in this way did not have to be hemmed all round - they were ready as soon as they were taken off the loom. <...>

As the finds make clear, rectangular cloaks with decorative bands existed in Scandinavia during both the Roman Iron Age and the Migration Period. On the Roman Iron Age cloaks, bands and cloak fabric were woven simultaneously, while, judging by the finds, the bands of the Migration Period cloaks were sewn on - probably because the technique for making the bands had now grown so complicated that it was easier to weave them separately. But cloak fabrics and bands were still produced in close conjunction. From the cloak corners surviving from Evebø and Högom, we can tell that the weaving of the corners, i.e. the ends of the bands, was completed when the size of the cloak was clear and the bands, perhaps, had already been sewn on.
Nockert 1991, p. 122, 128-129

Wearing early period cloaks

The smaller cloaks and scarves were probably wrapped around the body, with or without a fastening mechanism. However, several of the rectangular cloaks from all three periods were very large, and may have been folded in order to be possible to wear. The Roman Iron Age cloak from Hunteburg bog has holes from a fibula showing that it was folded in the middle, and Nockert (1991, p. 122) suggests that the large, decorated cloak from Thorsberg bog was folded on the diagonal when worn.

Examples of large early period cloaks that are sufficiently preserved to calculate their dimensions

Date Place Dimensions Source
350 BC - 90 AD Skærsø, Denmark 206 x 150 cm Mannering 2017:153, 155-156
200 - 400 AD Thorsberg, Germany 236 x 168 cm Nockert 1991:122, Möller-Wiering 2011:41
245 - 415 AD Hunteburg, Germany 260 x 180 cm Nockert 1991:122, Möller-Wiering 2011:118
Migration Period Vejen, Denmark 202 x 153 cm Nockert 1991:122

Diagram of the wool 
cloak from Clongownagh bog A smaller cloak from 200-400 AD found in Clongownagh bog in Ireland was worn in a different manner. Two fragments with respectively a starting border and a selvedge was found, indicating a rectangular or square cloak that was at least 1.20 m x 1.26 m. One corner had been folded down in a triangle (12 x 9 cm) and stitched in place. The cloak would have been folded around the body in a cone, and fastened in the front (Heckett 2001, p. 92-93).

Illustration: Heckett 2001, p. 92
A-C = fringe of warp ends, c. 45 mm deep; B-D = fringe of weft threads, 70-80 mm deep; E = stitched fold; F = reconstructed areas of cloth.

This is a practical way to wear a square or rectangular textile as a cloak, since pinning the two contiguous sides together, and folding down the top corner make for a more generous wrap-around. The remaining corner hangs down at the back or side, giving considerable coverage. The cloth is worn on the bias which affords more stretch and comfort. The arms take up a goodly amount of the cloth so the long 'tail' is somewhat shortened. This shape is seen on a mounted horseman at the back of a Pictish stone at Meigle, Perthshire, and on female cloaked figures in soumak tapestries from the early ninth century Oseberg ship burial.
Heckett 2001, p. 93

Iron Age cloaks and shawls from Latvia

An 9th century hoard containing artefacts and textiles were found in Tīra Bog in Latvia. There was a Scandinavian settlement close to the find site, and the probable exchange of goods could explain why some of the textiles in the find originate from Northern Europe (Žeiere 2013, p. 191).

Most of the almost 200 textile fragments were too small to allow for identification of the garments they once belonged to, but there were some fragments of leg bindings, a possible hood fragment and several larger fragments belonging to a woollen cloak. Enough of the cloak had been preserved, including starting and finishing borders and selvedges, that the original dimensions could be determined: 110 x 210 cm (Žeiere 2013, p. 187-191).

drawing of man wearing cloak, 
tunic, trousers and leg bindings

The rectangular cape was fairly large, measuring 110 x 210 cm, woven from fine, good quality dark blue woollen thread. The starting and end borders of the cape are preserved, which were made of simple mono-colour tablet-woven bands only 6-7 mm wide, as were both selvedges, each being somewhat different - one of them ended in tabby weave, while the other had been folded back twice and sewn using a different kind of thread. The presence of tablet-woven starting and end borders, where the fringe has been woven, indicates the use of a vertical warp-weighted loom. At two corners of the cape the threads of the edge of the border and the threads of the tablet-woven trims have been twisted and tied into a knot, forming small decorative loops about 4 cm long.
Žeiere 2013, p. 187

Illustration: Davidson and Pīgozne 2010, p. 19

The cloak is believed to have belonged to a man before it was deposited in the bog, as it was found together with other artefacts interpreted as the equipment and dress of a Curonian warrior; two shields, a penannular brooch, a bracelet, a chisel, a drinking horn, a small leather bag and a ball of woollen thread (Žeiere 2013, p. 187, 191).

Iron Age (7th - 13th century) Latvian women wore woollen (usually twill) shawls fastened on the chest. These could be heavily decorated with bronze spirals, rings and other decoration (Davidson and Pīgozne 2010, p. 21-22). The decorations have aided in the preservation of the shawls in the graves, making it possible to accurately find their dimensions. However, they are also a reminder that Latvian clothing is markedly different from Scandinavian clothing during this period.

The decorated shawls are believed to have been reserved for festive occasions, as they are not practical in everyday work. However, some graves contain shawls with no bronze decorations, and in later graves (12th -13th century) undecorated shawls are sometimes found underneath the heavily decorated shawls. These are interpreted as possibly representing everyday wear (Davidson and Pīgozne 2010, p. 21-22).

The decorated shawls were rectangular, and varied between 0.55 - 0.90 x 1.00 - 1.30 cm. They became larger and more densely woven (with finer thread) later in the period (Zariņa 1998, p. 151). The undecorated versions appear to have been of the same size as the decorated ones (Davidson and Pīgozne 2010, p. 22).

Bei der Herstellung der Schultertücher in diesem Zeitraum wurden wollene, in Köper 2/2, in einzelnen Fällen im 13. Jahrhundert auch in Köper 1/2 gewebte, gewöhnlich dunkelblaue Gewebe verwendet. <...> Die Dichte der Gewebe erhöhte sich im Laufe der Zeit, die Garne wurden feiner und die Schultertücher (55-90 x 100-130 cm) zeigen die Tendenz größer zu werden. Im allgemeinen waren sie vergleichsmäßig klein und dekorativ.
Zariņa 1998, p. 151

Late Iron Age Shawls from Finland

Finnish finds are usually not a good source for the reconstruction of Viking clothing, due to the marked differences in Finnish and Norse dress of the period. In the case of an unshaped garment like a shawl however, there are probably sufficient similarities between the function of the garments that the Finnish finds may be of relevance when interpreting Norse garments.

Several shawl fragments have been found in graves from the Finnish Late Iron Age (AD 800 - 1055/1300). The shawls were decorated with spirals of copper alloy. These metal decorations preserved the edges of the shawls from the usual deterioration in the graves, sometimes to such a degree that the entire outer edge has been found, allowing for definitive determination of the dimensions of the shawl (Hägg 1971, p. 141-142).

Shawl fragments with 
fabric loops The preserved shawls are roughly 1.5 x 1.0 m. For example, three shawls from Yliskylä, Perniö, have the dimensions 1.5 x 0.9 m, 1.47 x 0.94 m, and 1.47 x 0.74 m respectively (Hägg 1971, p. 142).

Several of the shawls have preserved the remains of a closing mechanism, where there are loops of fabric or string attached to each edge of the shawl. A woollen shawl from a grave in Raisio, Ihala had a loop made from woollen yarn (Hägg 1974, p. 95), while there were two preserved loops found as part of the fragments of a shawl at Vilusenharju (Luoma 2003).

Photo: Shawl from Vilusenharju. Luoma 2003

The shawls with loops were presumably worn with a brooch in front. However, other variations exist, e.g. in grave 26 at Tuukkala where the cloak brooch was on the right shoulder, leaving the right arm uncovered (Lehtosalo-Hilander 1984, p. 32).

In grave 26 at Tuukkala the dress was fastened with the oval brooches high up on the shoulders. The shift brooch was under the chin and the mantle brooch was on the right shoulder. Thus the mantle had covered the left arm and it had been fastened on the right. <...> If we consider all the Finnish mantle finds as a whole, we can observe that the mantle may have been fastened also in the middle, and sometimes the brooch may have been on the left shoulder, too.
Lehtosalo-Hilander 1984 p. 32

Smyckeuppsättningarna med ett tredje spänne som mantelspänne har sina motsvarigheter i korstågstida gravar i Finland där också själva textilierna emellanåt bevarats så väl att mantelns dimensioner kunnat mätas. De ungefärliga måtten 1,50 x 1,00 m, som antagits för Birkagravarnas del, kan stämma rätt väl med de mantlar eller själar som avbildningarna visar.<...>

T.ex. mantlarna frän kvinnogravarna 1, 6 och 7 i Yliskylä, Perniö, Finland, med måtten 1,50 x 0,90 m, 1,47 x 0,94 m resp. bredd 0,75 m. Textilierna är här bevarade i korrosionen frän invävda bronsspiraler i hörn och kambårder. Se vidare H. Appelgren-Kivalo, Finnische Trachten aus der jüngeren Eisenzeit, Helsingfors 1907.
Hägg 1971, p. 141-142

En relativt väl bevarad yllesjal från en kvinnograv i Raisio, Ihala, Finland, (Inv. Nr. 14275: 273) har t.ex. en liknande ullgarnsögla fäst invid kanten.
Hägg 1974, p. 95

Semi-circular cloaks

There are several well-preserved semi-circular cloaks from central Europe from 650-1050 AD. They are high-status garments, made from costly materials and worn by kings, queens, emperors and church leaders (Kania 2010). One such cloak is the fringed silk cloak (dated to 680 AD) that was found in the grave of queen Bathilde in Chelles in France (Mannering 2017, p 161). It is possible that semi-circular cloaks were in use outside of a royal or ecclesiastical context during this time period, but if so, the evidence of it has not survived.

Semi-circular silk cloak
Photo: Cloak of Bathilde in Mannering 2017, p. 160

One of the earliest examples of more ordinary (or at least not explicitly royal) semi-circular cloaks from Scandinavia is the woollen cloak that was found in a grave at Leksand in Sweden. The cloak is dated to ca. 1210 AD +/- 120. The deceased was identified as a woman, based on the finds in the grave (Nockert 1983, p. 100).

Drawing of Leksand cloak The cloak appears to have been fastened by two loops of string sewn to the edgings of the cloak, and possibly held together by a separate string.

Illustration: Carlson, M. 1996

It was not preserved in its entirety, but the largest fragment is 70 cm wide, with a selvedge at one end, that had been sewn together with the selvedge of a separate piece of fabric. The latter fragment is only preserved in a width of 5 cm but is interpreted as likely being of the same width as the first fragment. This would result in a full-length cloak with the length of the back being roughly 150 cm (Nockert 1983, p. 102-103).

From the shoulders and c. 12 cm down over the chest lay a folded piece of diamond twill (broken 2/2 lozenge twill) which continued down along the two long sides of the grave. From the right shoulder down the right side lay a tablet-woven band of 4.5 cm width, which was sewn along the piece of diamond twill as a completed edging. From the left shoulder ran a similar band which crossed over the chest and continued down the right side of the grave. All the textile fragments originate from one and the same garment, that is, the cloak in which the woman's body was swathed. The front edgings of the cloak were decorated with a tablet-woven band with a richly varied geometrical pattern.
Nockert 1983 p. 100-101

Another example of a woollen semi-circular cloak is the one worn by a man found in the Bocksten bog in Sweden. The find is dated to 1300-1360 AD (Nockert 1985, p. 88). Drawing of Bocksten cloak

Illustration: Carlson, M. 1996

This cloak was constructed from two lengths of fabric stitched together, and several smaller pieces. A head-opening had been cut, but otherwise the long side of the semi-circle had its selvedge intact. The cloak was worn with the opening on the right shoulder and had a length from top to bottom of ca 110 cm (Nockert 1985, p. 41, 47).

Shawls and cloaks in costume iconography

An additional source of information regarding outer garments worn by Viking Age women is the pendants, tapestries and picture stones depicting female figures. Caution must be used when trying to translate iconography into functional garments, as we do not know whether these figures are wearing ordinary clothing, or if they are entities of power like Valkyries or goddesses that may be wearing mythical garments. The issue is further complicated by the fact that the art is not naturalistic. However, in the absence of complete garments, the pictorial evidence may provide a useful aid to interpretation.

Wihlborg (2019) provides an overview of Viking Age jewellery depicting females. Many of these have been considered by Mannering (2017) in her analysis of how clothing is depicted in late Iron Age and Viking Age costume iconography.

Long cloaks fastened in the front

Photo of pendant from Hagebyhögda in Sweden Cloaks of some kind appear relatively often among the female figures. While the majority of the cloaks are long, there are also figures showing shorter versions. Most of the cloaks are fastened in the front, at the centre of the chest (Mannering 2017, p. 118).

One of these figures have a clearly depicted fibula fastening her cloak. It is an over-sized disc-on-bow brooch, like the ones worn by women in the late Iron Age. The costume is highly stylistic, but Mannering (2017, 110-112) interprets it as comprising a flowing cloak, open in the front, and a dress.

The pendant was found at Hagebyhöga in Sweden, in a grave dated to 975 AD. However, the objects found in the grave range over a period of more than 150 years, with the pendant being among the oldest (Mannering 2017, 111).

Photo by Gabriel Hildebrand, SHM Creative Commons Erkännande 2.5 Sverige

Three small (ca 1.7 cm high) three-dimensional figures found in Denmark provide a more naturalistic depiction of long cloaks.

Photo of figure from Lejre in Denmark Photo of figure from Boeslunde in Denmark
Figure from Lejre in Denmark.
Photo: National museum of Denmark
Figure from Boeslund in Denmark
Photo: Museum West Zeeland, 2020.

The figure from Lejre is dated to around 950 (Christensen 2010, p. 143). It sits on an elaborately decorated chair that is interpreted as a high-seat, and the presence of two birds (possibly ravens) may indicate that it is Lidskjalv, the throne of Odin. The costume comprises a cloak, a long tunic or dress and four rings or strings of beads on the chest. Around the neck is a ring. Above it is a horizontal thickened line, which Christensen (2010, p 155) interprets as a moustache or beard. The figure is wearing a helmet or hat of some kind on their head. Christensen speculates that the figure may be Odin, dressed in female clothing.

A figure from Boeslund, dated to 8th - 9th century, is wearing similar clothing. A long cloak, a dress and three strands of jewellery on the chest. However, this figure has the characteristic hairstyle seen in other pendants depicting women, namely the hair is loose and tied in a knot at the back. While the costume of this figure have no masculine elements, the similarity between it and the figure from Lejre has led to some speculation whether this too could be interpreted as Odin wearing women's clothes (Museum West Zeeland, 2020).

Photo of figure from Trønning in Denmark Whatever the identity of the figures, from a clothing perspective the interesting part is that it depicts female clothing that include a long cloak, open in the front and closed at the neck. Both cloaks are decorated along the sides.

Finally, there is a small gold figure that was found in Trønning in Denmark in 1868 (Christensen 2010, p. 147-148). This figure also wears a decorated cloak, a long dress or tunic with bands in the front, and possibly jewellery or a broad band on the chest. The hair is tied in a knot at the back.

Photo: Christensen 2010, p. 148

Rectangular cloak with visible corners

The long cloaks worn by the three-dimensional figures could be either rectangular or semi-circular. They are not included in Mannering's analysis; however she identifies rectangular cloaks on a set of pendants (Mannering 2017, p. 112)

Photo of pendant from Grödinge in Sweden One such pendant was found in a cremation grave in Sibble, Grödinge in Sweden. The cloak is embellished with a dotted pattern and is clearly open in the front. It tapers down into a point at the side of the figure. From the chin to the thighs another marking is visible, which Mannering (2017, p. 112) proposes could be the other side of the cloak.

Photo by Gabriel Hildebrand, SHM Creative Commons Erkännande 2.5 Sverige

Photo of pendant from Uppsala in Sweden A somewhat similar pendant was excavated as part of the contents of boat grave 36 in Gamla Uppsala in Sweden. It is dated to the 9th century.

The figure in the pendant is blurred, but the costume comprises at least two layers. Mannering (2017, p. 112, 114) interprets the outermost layer as a cloak which tapers down into a point at the side. The cloak is plain, but it is possible the triangular striped area on the shoulder belongs to it.

Photo: Lindman, Upplandsmuseet

Photo of tapestry In her analysis of Scandinavian late iron age costume iconography, Mannering concludes that this type of cloak is also depicted on the women in the Oseberg tapestries (Mannering 2017, 118).

Photograph: Unimusportalen, part of Oseberg tapestry

Female outerwear is attested on the Oseberg and Rolvsøy tapestries. The former includes 30 figures and the latter two. It consists primarily of a floor-length cloak of Outerwear Group lc (OGlc) with a pointed end at the centre of the figure. As the patterning and colours of the outerwear recorded on the Oseberg tapestry cannot be determined with any certainty, they are all recorded as plain, while the cloaks from the Rolvsøy tapestry are recorded as dotted. In comparison, two female figures from the jewellery category dated to the Viking Age, too, are recorded as wearing OG1c cloaks.
Mannering 2017, p. 142-144


Triangular cloaks

Cloaks that are depicted as triangles with curved or straight edges are common in the iconography from late Iron Age Scandinavia. While not as common in the figures from the Viking Age, they are still clearly present.

Photo of pendant from Nygård in Denmark One pendant with such a cloak was uncovered as a stray find from Nygård in Denmark. The figure is wearing a long, triangular cloak with a pattern of bows or half-circles. In the curved front edge of the cloak, a diagonally-striped border is visible. According to (Mannering 2017, p. 114) the border may be suggesting that the edge of the cloak ended in a fringe. Underneath the cloak, a plain dress can be seen. The pendant is dated to the Viking Age (Wihlborg 2019, p. 96).

Photo by National museum of Denmark

Photo of pendant from Tuna in Sweden Another pendant was found in connection with boat grave III in Tuna/Alsike in Sweden. It is tentatively dated to the Viking age. However, there were graves in the area from the Vendel period, and it is possible that the pendant came from one of them (Wihlborg 2019, p. 84).

This figure is wearing a plain, triangular cloak that flares out at the back. Mannering (2017, p. 112, 116) suggests that the circles at the front of the figure may be a disc fibula (closing the cloak), combined with a necklace.

Photo by Gabriel Hildebrand, SHM Creative Commons Erkännande 2.5 Sverige


Photo of picture stone from Tjängvida in Gotland Photo of pendant from Hjorthammar in Sweden A possible shawl or triangular cloak is visible on the lower part of a silver pendant that was recovered in a burial near Hjorthammar, Blekinge in Sweden. It is dated to the 9th century (Holmqvist 1960, referred to in Wihlborg 2019, p. 80).

The figure is wearing a triangular cloak decorated with a dotted pattern. A dress ending in a train and marked with stripes along the back is visible beneath the cloak. The figure is interpreted as female based on the costume (Mannering 2017, 114).

A similarly shaped garment is worn by a woman on a picture stone from Tjängvide in Gotland. There is no pattern on the triangular cloak. The figure is interpreted as female based on the costume and the hair in a knot at the back of the head.

Photo (left): Wihlborg 2019, p. 80
Photo (right): By Berig, CC BY-SA 4.0

Photo of earspoon from Bj 507 Finally, there is an earspoon found at Birka, in grave Bj 507, that depicts another woman wearing a triangular cloak.

The cloak has a curved front edge, and tapers into a point at the back, almost giving the impression of wings. It is patterned with diagonal stripes.

Photo by Ulrik Skans, SHM Creative Commons Erkännande 2.5 Sverige


Open at the side

Lastly, there is a cloak type that appears to be open at the side. On the gold-foil figures (that mostly predate Viking Age) this type of cloak is only depicted worn by men (Mannering 2017, p. 118). However, all three figures in the pendants from the Viking age have the hair-tied-in-a-knot that is interpreted as a female characteristic.

Photo of pendant from Bj 826 Photo of pendant from Bj 968   Photo of pendant from Truso in Poland
Pendant from 9th century grave Bj 826 in Birka, Sweden.
Photo by Gabriel Hildebrand, SHM Creative Commons Erkännande 2.5 Sverige
Pendant from 9th century grave Bj 968 in Birka, Sweden.
Photo by Gabriel Hildebrand, SHM Creative Commons Erkännande 2.5 Sverige
  Pendant from Truso, Poland.
Photo by National museum of Denmark.

Mannering (2017, 112) describes the costume of both figures as an ankle-length outfit, divided into three sections. There is a foremost part with three vertical grooves whose top end is concealed by the arm. The rearmost part is similarly separated into three vertical sections. The arm sticks out from between the first and the second grooves and only the two rearmost grooves continue all the way up to the shoulder. Both the foremost and rearmost parts of the costume have irregular marked horizontal stripes, suggesting that these parts belong together.

The middlemost part comprises a large blank area with a border at the lower hem. The border is triple horizontal-striped in the case of the pendant from Bj 826 and double checked in the case of the pendant from Bj 968. Both pendants are dated to the 9th century (Mannering 2017, p. 108, 112).

 

A pendant was found in Truso, in Poland. It is dated to the 9th century (Wihlborg 2019, p. 125).

The figure appears to wear a similar outfit to the two Swedish examples. However, while the middlemost part has a large blank area, the lower hem is decorated similarly to the foremost and rearmost parts, suggesting that it belongs together with them.


Interpretation

To be continued....

Bibliography

Illustration sources