The 25 m2 wool sail for one of the Oselver boats is the preliminary culmination of the Viking Ship Museum’s research into wool sails, which started in 1977. Since then the following vessels have been equipped with wool sails: Ottar (Skuldelev 1); Roar Ege (Skuldelev 3); Sara-Kjerstine, Hitra, Norway (Åfjord longboat); Embla, (Faroese boat); Sif Ege, Frederikssund (Skuldelev 3); Bialy Kon, Gross Raden, Germany (Ralswiek 2); Ottar (Skuldelev 1). Here follows a a short summary of the complex of problems concerning the various types of sail and sailcloth:
Different types of woollen cloth from the Viking Age and recent times
The picture which seems to be emerging is that, depending on the local area, resources and traditions, three types of weave or intersection were used to produce wool sail cloth. What are referred to today as two-shaft or plain weave (one thread over and under one thread), three-shaft or 2/1 twill (two threads over and under one thread) and four-shaft or 2/2 twill (two threads over and under two threads) – was in the past called einskept, tuskept and priskept respectively. Both the vertical warp and the horizontal weft appear in general to have been one-ply with the yarn spun clockwise (z) in the warp, and with yarn spun either clockwise (z) or anti-clockwise (s) in the weft, depending on the weave.
If we go back to the mid- and late 11th century, the time of the Skuldelev ships, all three weaves were probably used to make wool sails. On the basis of the available archaeological material (the Trondenes fragment) the museum has, however, concentrated on 2/1 twill (tuskept) for the manufacture of the wool sail for reconstructed Viking ships. This type of material is also documented in the ethnological material from the Nordic countries.
Subsequent to the Viking period and the Middle Ages, wool sails continued in use right up into the 19th century over a large area extending from Iceland and the Faroe Islands down over the Hebrides, Shetland Islands, Norway, Sweden and Finland. In Norway, there are still preserved wool sails or remains of these available for research, whereas there is probably a large amount of unrecognised archaeological and ethnological material in the other areas.
In Iceland and on the Faroe Islands, thread for the manufacture of wool sails was, until quite recently, spun on drop spindles, just as the cloth was woven on a warp-weighted loom. On the Faroe Islands it was, at least, possible to weave cloth lengths of about 5 m, which was sufficient to make the Faroese square sails. On Iceland, in the Middle Ages and later, people were able to weave unbroken lengths of almost 11 m on a warp-weighted loom. The fact that these lengths were not unusual in the Middle Ages is hinted at by the excavations at ‘the Manor Below the Sand’ in Greenland, where remains of a warp-weighted loom were found with a clothbeam which in size and form is identical with the type of beam used on Norwegian warp-weighted looms in the 18th and 19th centuries.
During the production of the great quantity of 2/1 twill wool cloth for Ottar’s sail, the threads were spun on a spinning wheel and the cloth woven on a horizontal loom. At the same time several large swatches were produced on a warp-weighted loom with thread spun using drop spindles. This parallel experiment showed that the quality of the product from the warp-weighted loom was absolutely on a par with that from the horizontal loom. The warp-weighted loom is, however, considerably slower to work with – something which is also apparent from the Icelandic material from the 19th century. Based on the experience gained in the parallel experiment five people could spin and weave the cloth for Ottar’s sail in a year. A sail would probably last about 30-50 years.
The square sail – construction and type
We have chosen to work with a type of square sail, which consists of unbroken vertical lengths of cloth sewn together. This type can both be traced back in time and also right up into recent times when sails of wool, linen and cotton were constructed in the same way.
The reason for using cloth in unbroken lengths is due partly to an assumption that it was possible to produce the lengths used (about 11 m on the Skuldelev 1 ship’s sail) in the 11th century, partly to an evaluation that the strength, form and efficiency of the sail is not sufficiently different in a sail with unbroken lengths of cloth and a sail where each individual length is composed of two or three pieces.
Just like the traditional square sail of recent times, the lengths of cloth are sewn together using round seam giving a completely smooth joint on the front side. With a square sail, the front surface is always the same when sailing close-hauled, where the low pressure gives a high wind speed, which ideally should pass through without hindrance. The sail has transverse rope reinforcements on the reverse. These are primarily to give the cloth strength during reefing and they are sewn on in such a way as not to 'lock' the movement of the cloth.
The choice of this type of square sail makes it possible to evaluate wool cloth of different types and quality. Without, for example, diagonal reinforcing bands the wool cloth must 'take care of itself' and is therefore easier to assess. At the same time, the results are directly comparable with sailcloth of linen, cotton and Duradon (to include a modern type of fabric).
Iconographic evidence and weave length
The iconographic record contains a great deal of variation with regard to types of square sail. Here, there are not only vertical panels but also chequered and lozenge-patterned sails as well as sails with diagonal bands. Even though the museum has chosen to work exclusively with sails made up solely of vertical panels, the work has given experience that can contribute to considerations of what it is that the iconographic material illustrates.
For example, how was the chequered sail – as seen on the Gotlandic picture stones – produced? The theory has been proposed that the earlier Viking Age warp-weighted looms were not able to weave lengths of cloth greater than the height of the loom. The sails had, therefore, to be sewn together of several smaller pieces of cloth. One can imagine perhaps that every household in the area produced a 'square' (just as has been documented in connection with the production of sails on the Shetland Islands in recent times). One can also imagine that the rectangular panels were of different colours. On a purely technical level, a sail composed of many small pieces sewn together with round seam probably functioned very well.
The Gotlandic picture stones also show sails bearing thin diagonal lines. This could represent sails sewn together from many small pieces of cloth – but this seems unlikely. The lines could also represent diagonal reinforcing bands intended to control cloth that is a little too loose and elastic (it should be added that it is not at all certain that the picture stones show wool sails). Or these lines could be a combination of reinforcing bands and decoration.
The diagonal patterns can later be traced as coloured textile bands in both written and iconographic sources without it being possible to establish whether these function as reinforcement of “living” sail cloth, decoration or a combination of the two. On some of the sails, the construction of the diagonal bands is accentuated at the point where they cross each other – a detail which is virtually identical with the decorations on, for example, the sledge runners from the Oseberg find. The pattern, shown both as bands and lines, was generally common and widespread at that time, both on land and at sea. And one should not ignore the significance of the decorative function that was maintained right up into the 14th century – as shown by the frescoes in Brejninge Church from the end of that century; these show coloured diagonal bands on both sides of the sail.
The iconographic material provides important evidence for research, but it must be used with great critical sense and with a basis in a well-founded understanding of the material. Accordingly, it is not possible to draw conclusions with regard to the material used on the basis of the depicted sail construction and, for example, maintain that sails made up of vertical panels – without diagonal lines or bands – were made of linen or hemp. As the museum’s research and surviving traditional sails show, they could also have been made up of wool. Conversely, linen sails could well have been equipped with reinforcements, coloured bands etc., running diagonally or vertically over the joints.
In conclusion, relative to the theoretical discussion concerning wool sails, caution is recommended with respect to developmental theories about a continuing improvement in wool sails from the Viking Age up into recent times. Advanced sailing-ship hulls such as Skuldelev 1 and 3, and advanced sailing- and rowing-boat hulls such as Skuldelev 2, 5 and 6, were reliant on sailcloth of a correspondingly high quality.
Andersen, Erik 1995: Woollen material for sails. In: Olsen, Olaf et al. (eds) Shipshape. Essays for Ole Crumlin-Pedersen on the occasion of his 60th anniversary February 24th 1995. Roskilde, 249-270.
Andersen, Bent & Erik Andersen 1989: Råsejlet - Dragens Vinge. Roskilde.
Andersen, Erik, Ole Crumlin-Pedersen, Søren Vadstrup & Max Vinner 1997: Roar Ege. Skuldelev 3 skibet som arkæologisk eksperiment. Roskilde.
Andersen, Erik, Jytte Milland & Eva Myhre 1989: Uldsejl i 1000 år. Roskilde.