Penelope Walton Rogers on Dyed Fabric From Iron Age Norway and Denmark

About three years ago, I wrote a post about likely colors for Viking apron dresses that attracted a significant number of comments.  One of the commenters was Hilde Thunem, who has written a lot about Viking costume herself. Hilde mentioned that Penelope Walton (now Penelope Walton Rogers), a British archaeologist who has done a significant amount of chemical analysis of dyes on early European textiles, had said anything of relevance to our discussion in either of two articles she had published about her work on Viking age textiles. I have a copy of one of the articles, and offered to re-read it to see whether it had information that might be useful on the question of apron dress colors.

That was in December 2010. Oops.

Anyway, I was reminded of that discussion when someone else recently posted a new comment in it, and I decided that it was high time for me to re-read the article and summarize the most useful parts. The citation for the article is:

Walton, Penelope. Dyes and Wools in Iron Age Textiles from Norway and Denmark. Journal of Danish Archaeology, vol. 7, pp. 144-158 (1988).

More than half of the article consists of a detailed and technical analysis of surviving fabric fragments to ascertain what types of sheep, and what parts of the sheep, the fibers in the fabrics came from, but the last six pages or so discuss Walton's dye analysis of Viking age textile finds.  Because most of Walton's analysis is not linked to what types of garments the tested specimens may have come from, and many of the samples discussed by Walton pre-date the Viking period, her analysis is not obviously relevant to our apron dress question, but the information it does provide supplies some interesting support for further inference and deduction.

Walton begins her discussion of dyes by saying that indigotin, the blue pigment from woad and (in countries where it grows) indigo, was the most common dye substance found in the Viking age wool fragments and that it was found in both coarse and fine textiles. She notes that, though woad was not originally native to northern Europe, archaeological finds of woad seeds suggest that it had reached Scandinavia by the Roman period.

Walton points out that some of the Scandinavian wool samples contained enough indigotin to have been very dark shades of blue.  In some of the Evebo/Eide samples it formed a "rich, deep stripe" on a red or orange background; in some of the Veka textiles it was used only on warp threads, so those twills would show dark diagonal lines of twill on a white background, and the Birka diamond twills were so dark as to be nearly black.  Walton also noted that when some specimens from Sandanger, Skjervum, and Sandegaard were tested, the indigotin was easily removed by tests for mordant dyes, though not by tests for vat dyes (which woad is).  Walton concluded from this that the indigotin had been applied over a mordant dye, though no mordant dyes proper were detected.  Walton suggested that this mysterious mordant dye might have produced a yellow color, and notes that tests of other textiles clearly indicated that woad had been used together with a yellow dye (possibly to produce a green).

Then Walton discussed specimens that tested positive for dyes that produce red shades.  She found that textiles from three of the wealthier graves in Norway, namely, Snartemo V, Veiem, and Evebø/Eide, tested positive for the substance alizarin, indicating that those textiles had been dyed with Rubia tinctorum L., otherwise known as dyers' madder.  This is curious because these graves date to the Migration Period. Although madder has been known to have been cultivated in Paris (by the 9th century CE) and England (about the same time), there's no clear evidence that it was found in Gaul, or in Scandinavia, during the Migration Period.  Walton suggests that either the dye, or madder-dyed textiles, may have been imported, given the wealth of the graves where it was found.  On the other hand, in textiles from Sejlflod and Hejrhøj in Denmark, a similar red dye that did not test positive for alizarin was found, ruling out dyers' madder as a source but suggesting that a similar, native-grown dye plant was used; Walton suggests Galium verum L. and Galium odoratum (L) as possibilities.  Walton did not succeed in ascertaining what substance produced the red seen on the Lønne Hede textiles, but eliminated some possibilities, including kermes, fungus red dyes, Rumex crispus, and Rubia and Galium discussed above.

Two Norwegian samples in particular, Veiem C348 and Evebø/Eide B4590, tested positive for one of the red insect dyes.  An extraction test ruled out cochineal (highly unlikely this early in Europe anyway) and lac. Walton suggests that the results for the Veiem sample are consistent with Polish cochineal; the Evebø/Eide specimen gave results that could be consistent either with kermes or Polish cochineal.  Polish cochineal might, Walton notes, have been available to Scandinavians through trade with the Baltic region during the Migration Period, and kermes from trade with the eastern Mediterranean or Near East.

Finally, textiles from Thorsbjerg in northern Germany and Fløjstrup in Denmark were detected to contain a lichen-based purple dye.  Some of the plants that can be used to produce such a dye are native to Scandinavia, but their use of dye plants does not seem to have been part of the native repertoire.  Walton accordingly suggests that this dye, or the textiles on which it was found, came from Frisia.  The Fløjstrup textile was particularly likely to be an import because it was made in an "unusual weave" (Walton references Margrethe Hald's Ancient Danish Textiles from Bogs and Burials, at page 100, on this point), and opines that it will not be possible to identify the source of the purple dye without identifying the provenance of the weave used to make the fabric where it was found.

I hope Hilde Thunem, and anyone else with an interest in dyeing, textiles, or clothing of early northern Europe, finds this summary interesting.  Anyone who is interested in obtaining this article for research purposes, please contact me via Google + or leave a comment.

Apron Dresses--More of the Blues

At the suggestion of Hilde Thunem, I re-read an article about dye testing of Scandinavian archaeological textiles that I'd obtained and read years ago.  The article is:

Penelope Walton: "Dyes and Wools in Iron Age Textiles from Norway and Denmark," 7 Journal of Danish Archaeology pp. 144-158 (1988).

In the article, Walton was interested both in analyzing specimens for dyes used and for wool types, in order to try to draw conclusions about the possible origins of the fabrics involved.  She drew samples from Norway, Denmark, and Germany, ranging in period from the Roman era to the Viking Age.  Consequently, only about a third of the samples analyzed were from the Viking Age, and none of the samples were categorized as to the type of garment from which they might have originated.

Walton extracted dyes from the samples with solvents and examined with at U-V/visible spectrophotometer; paper and thin-layer chromatography were used to confirm positive results for dye. Unfortunately,  the  dye results for most of these do not unambiguously appear in the results table in the article.  However, Walton makes some interesting comments in the text about dye results for a few of the Viking age specimens, even though the main thrust of the article was how the differences in wool type could pinpoint whether particular  textiles had been made in Scandinavia. In particular, Walton's results indicate that the "Birka-type" diamond twills--the type of fabric that was associated with the apron dress layer in the Birka finds--were most often blue, and blue was the most common color in general:

The most common dye to be identified was the blue indigotin (fig 3) which at this time almost certainly derived from the woad plant, Isatis tinctoria L. (the other possible source being the sub-tropical Indigofera tinctoria L.) Although woad is not a native of northern Europe, archaeological finds of its seeds suggest that it had reached Scandinavia by the Roman Period.

Indigotin was identified in fine and coarse textiles of all periods. In several of the Norwegian finds it seems to have been used for particularly dark shades:  at Evebø/Eide it formed a rich, deep stripe on a red or orange ground; in some of the Veka-type textiles only the warp had been dyed, so that the diagonal lines of twill would have stood out as dark blue on white; and in the Birka-type diamond twills, the dye was so dense that it was almost black.  Since woad is an especially difficult dye with which to work and the deeper shades of blue require repeated dyeings, this is yet more evidence that the makers of the Birka-types possessed considerable skills.

Page 153-54 (emphasis supplied).

One Viking age specimen was found to have been dyed with lichen purple, but it's not possible to identify from the article the type of garment from which this specimen may have come, and lichen-purple appears to have been rarely used in Scandinavia according to Walton:

Another rare dye of some significance is the lichen-derived purple found in textiles from Thorsbjerg in northern Germany and Fløjstrup in Denmark. This dye may be obtained from a range of lichens including some which are native to Scandinavia. However, although the dye was available in the north, knowledge of its use seems to have been rather limited.

(Page 156) (emphasis mine).

We don't know what kind of garments or other textile items these Viking era specimens might have been.  However, at Birka itself, the fine diamond twills are believed to have been from the apron dress layer, judging by where they were found in the graves.  If that is also true of the "Birka-type" diamond twills Walton discusses in her article, it is interesting that these twills are also dark blue--in fact they are "so dense that it was almost black."  It is also striking that, of the various Scandinavian textiles that tested positive for dye (as opposed to natural wool pigment, which Walton distinguishes in her test results), most were found to contain indigotin--i.e., blue. 

Though there may have been things I missed because of my difficulties in interpreting the tables in which Walton reported her results, it appears that Walton's survey showed that blue was the most common dye in Scandinavian prehistory--including the Viking period.  That makes it a bit more likely that apron dresses were dyed dark blue--if they were dyed at all.

Apron Dresses--Getting The Blues...

Most people who make Viking era costumes are aware that wool is an easy fiber to dye with natural vegetable dyes.  They, therefore usually choose to wear wool dyed (either by vegetable or modern chemical means) in various vibrant shades that can be achieved with natural dyes:  reds, blues, yellows, greens, oranges, even purples.  I have done so myself. 

But proof that our forebears could have done something in a particular way is by no means proof that they did things in that way.  As  I learn more archaeological textile evidence comes in from Scandinavia, and from northern Europe in general, I wonder whether clothing colors weren't more stereotyped than most costumers and reenactors would like to believe.  

Consider:

• The two largest apron dress fragments known (i.e., the pleated fragment from Køstrup, and the folded one from Pskov) were both the same color--blue.  (The fragment appears lighter today; apparently the chemical process by which woad dyes fabric partially reverses itself when the fabric is buried under the right conditions, as noted in this post and one of the articles referenced in it.)
• The next largest fragment, from Hedeby (though not so definitively identified as part of an apron dress), was dyed brown.
• To the extent she could identify color in the Birka textiles, Inga Hägg found only two colors: brown and dark blue (as discussed in Hilde Thunem's article in progress about the evidence for Viking apron dresses).
• Viking age finds made in the Baltic areas (Latvia, Lithuania) show a pattern; white undergarments (shirts, women's shifts) over which clothes are worn in...dark blue and brown.  This is particularly true of Latvia, where testing has revealed that overdresses and tunics were  dark blue and other garments were either made from unbleached linen or from undyed wools. Colors other than white, dark blue, or the browns and grays of undyed wools (red, yellow, green) appear only as accents, e.g.,, as stripes and other motifs found in woven belts.  
• Most of the apron dress loops that survive are of undyed linen (whether or not any other surviving portions of the garment were also of linen). 

Not all of these textiles have been analyzed for the presence of dye, but some have, tending to confirm the trend.

There are certainly archaeological textiles that were dyed in other colors, such as the reds of the Evebo find.  But it is striking that all of the finds that are likely to have come from apron dresses or similar overdresses seem to have been either blue or brown.  This restriction may be due to the limitations of our samples and other information, but it might reflect genuine information about such garments.  Were apron dresses blue or brown because they were *supposed* to be--just as Victorian men's suits were typically black?  Maybe Viking women didn't wear red?  Maybe green was too difficult to get right, or yellow insufficiently high-status to use in such a garment?

There's certainly food for thought here, and I will be keeping this evidence in mind the next time I make an apron dress--or wear one of the bright, non-blue apron dresses I already have.

EDIT:  I am currently re-reading one of the articles Hilde Thunem mentioned in her comment on this post.  It has enough material of potential, if indirect, interest that I will be devoting another post to it.