Swedish medieval loop braids

Suomenkielinen yhteenveto (short Finnish summary):
Aivan äskettäin ilmestyi englanninkielinen kirja Vadstenan luostarin tekstiilityöpajasta peräisin olevista palmikoiduista nauhoista. Onnistuin toistamaan näistä nauhamalleista lähes kaikki, niissä viimeisissä tarvitaan enemmän tekijöitä. Alkuperäiset nauhat on tehty pääsääntöisesti arvokkailla silkki- ja metallilangoilla, yksi on pellavainen. Minulla olisi aivan “oikeitakin” lankoja, mutta käsieni iho on tällä hetkellä niin karkeata ja vaurioherkkää, että käytin mallinauhoihin paksumpia puuvilla- ja lurex-lankoja. Värit ja metallilangan tyyppi (ytimen ympärille kiedottua litteää metallinauhaa) mukailevat kuitenkin alkuperäisissä nauhoissa käytettyjä materiaaleja, jolloin nauhojen värikkyys ja rakenne tulevat oikein esille.


 The book  Loop Braiding in Swedish Bridgettine Tradition

Published in 2011. ISBN 978-0-9562620-2-8.

Mid-March 20102 I became a happy owner of “European Loop Braiding. Investigations and results” by Noémi Speiser, Joy Boutrup and others. Parts III and IV are new, the earlier parts just haven’t been available for a while, but now I got the whole package. However, I started with Part III, Loop Braiding in Swedish Bridgettine Tradition as I was very keen to find out what had been done in Sweden during the Middle Ages – not least because there was a Bridgettine convent in Naantali in Finland as well. The book itself is about the first and biggest Bridgettine convent, Vadstena, but I’d think that at least some of their traditions would have travelled to the other convents founded during the Middle Ages. I’m not sure if there are any textiles left that can be traced back to Naantali, but I probably should start digging now that I know what to look for.

First a majorly interesting quote by Noémi Speiser in the Introdution on page 6:

Still, our conviction has grown gradually ever stronger that loop-braiding was the exclusive method of braiding in the middle-ages. the longer we research, the less we suspect that any other techniques co-existed.

Personally, I’m not sure this applies to braids for made for purely practical purposes, but it may well be true for the decorative braids. Not only do I find fingerloop braiding speedier than similar open-end braiding, it has more variety because of the double-layer loop structure. Just like with tablet-weaving, there always seems to be something new that I haven’t explored – and I haven’t even tried to implement my knowledge to modern materials yet, nor tried creating something really original.

The braids

My samples include all the fingerloop braid types in the book except two multi-person braids: a flat 10-loop twill and an openwork lattice.  Braids 1-12 are double-width single-layer braids (all twills with varying number of loops), braid 13 is twined.

As materials I used mercericed cotton and cheap “metal” knitting yarn, so my samples are a lot chunkier than the original braids. I do have silk and even some metal thread that could be appropriate, but my hands are currently so chapped and sensitive that thinner yarn eats its way through the skin where the loops are held  and silk keeps catching on the rough skin, so I have to stick to softer and cheaper materials for practice and sampling. My first choice is the mercerised and loosely spun and easily available Novita Tennessee that is a reasonable substitute for filament silk and gentle on my hands, but some braids are made with thinner and shinier pearl cotton. My so-called metal thread, Katia Luxury, is correctly composed, a thin strip of metal (in this case metallised plastic) wrapped around a fiber core (cotton in my case, silk in the originals). It is probably more pliable than the real metal thread, but stills stiffer than the cotton and creates some friction, so I think I got an idea of what it is like to add metallic thread to fingerloop braids (for more discussion on the matter, see the book p. 7).

I assume those nuns who worked at the textile workshops at Vadstena and other monasteries had to have access to hand lotions and such “frills” and didn’t need to do manual labour, otherwise it would have been very difficult for them to handle the valuable silks, especially in winter. On the other hand, fingerloop braiding doesn’t need as good lighting and eye sight as e.g. fine sewing or embroidery, and with practice one feels on the loops when things are right (just like knitters can use their fingers to “see” the knitting). After I’ve learned a pattern well enough to go into an “automatic”, I usually stop when a mistake is looming, because the sequence of movements or the loops just do not feel right to me.


My braid samples, braids 1, 4b, 6a, 5, 7, 9 and 13



Braid 1: Part of the netting on a  sudarium at Uppsala Cathedral, green silk, width about 4 mm.
This braid is a standard five-loop twill (“a thin lace of 5 bows” in Tak V Bowes Departed). Both the original braid and my sample include the same mistake, a spot were a loop was crossed when it was supposed to be open, and because of this the braid won’t fold flat at that point – I somehow lost concentration and made couple of these at the very end of the braid, at which point I decided that enough was enough. I can see that I still need to work on my braid tension, because even on a warp this short my beginning is too loose and the last bits are a bit too tight.


Braids 2-3 are similar to Braid 1, 2a and 3a are red, 2b and 3b are green.


Braid 4b: On a mitre in Västerås Cathedral, red silk with gold metal wrapped on brownish-red core, width about 3 mm. 4a otherwise the same, but green silk and gold. Dated either 1442 or 1454.

A standard five-loop twill with bicolour (departed, linked) loops. I.e. each loop has a silk shank and a metallic shank, resulting in an asymmetrical pattern. A special thing with this braid is that four of the five loops have a single silk thread and a single gold thread, but the fifth has a paired silk thread and a paired gold thread. According to Speiser this must be intentional, because the feature appears in both 4a and 4b (and repeats in some of the other braids, too). My sample, made with pearl cotton #8, repeats this feature, which is why the gold areas seem to bulge at a point. In cotton the doubling of thickness doesn’t show as much as in the gold that is less pliable.

NB. There’s also some delightful wide brocaded tabletweaving on the same mitre where the braids are used. And just like the different sides of the mitre have fingerloop braid edging in different colours, the brocade is different on different sides. I have to check whether these bands are charted in Nancy Spies’ Ecclesiastic Pomp and Aristocratic Circumstance”, as there’s and interesting spot motif I can’t recall seeing.


Braid 5: On the same sudarium as Braid 1, but used to cover a seam. Original green silk and gold metal wrapped on a red core. 5-6 mm wide.

Standard twill braid, but with six loops, three green and three gold. My braid is made with pearl cotton #8 and the sample is actually exactly the same width as the original.

The six-loop version is an unusual variant of the twill and one that gives “protruding central ridges over which the braid readily folds itself in grasping and connecting the two layers and protect the seam.” (p. 24)  This is true also for my sample braid, it really is like molded to fit over a seam with a bit of space under the middle ridges. That is, the braid was made for this purpose and the nuns really knew what they were doing.


Braid 6a: On the edge of the lid on a 15th century reliquary at Vadstena Convent. Fine red silk (5b metal wrapped over greenish core). Width about 2 mm.

This is a 3-loop twill! The braid can easily be reducted from the common 5-loop twill, but it does not appear in any of the period recipe books, except as a component of a multi-worker braid where one person has 5 loops. I find it interesting that the nuns would opt for a 3-loop twill braid instead of a simple 3-element open end braid (i.e. the very standard hair braid), I think that supports the claim that fingerloop braiding was The Braiding Technique, at least in this context.


Braid 7: A reliquary in Linköping cathedral, braids couched on the surface of the fabric as decorative knots. Strongly twisted undyed linen.

These braids are another oddity: 4-loop twill, so the cross section is trapezoidal, the braid isn’t as flat as the 5-loop variant. The choice of material is also unusual, but everything does make sense, as Speiser writes: “The fact is: these braids were functional rather than aesthetic. The large and soft central ridges were meant to prepare a soft bed for couching pearls, whilst the selvedges should be unobtrusive.” (p. 25). Most of the pearls are gone, but a few remain and there are other examples from the same workshop and same time with pearls intact, so the argument is strong. And it makes sense to do the knot base with fingerloop braid because the oblique twill structure is very pliable (like bias tape), so forming it into the knot formations (True Loves Knots and Spenser Knots) is easy, probably easier than trying to couch e.g. a bunch of unbraided threads.


Braid 8 is a 10-loop braid, i.e. made by four hands/two people, so it has to wait a bit.


Braid 9: Braid on an apparel of a lost sudarium, first half of the 15th century, green silk with gold metal wrapped on a red or yellow core. 7-8 mm wide.

The braid is a standard (?) 7-loop twill with three green and four gold loops. Speiser notes that in the convent this was probably worked on four hands, i.e. by two people, partly for the social aspect (it might also be faster, I don’t know yet). But as she writes, “I can work this braid easily on my own to hands.”  – and I agree. I don’t even need to resort to the complicated period recipe of Tak V Bowes Departed as my little finger is nowadays obedient enough, so I can get my forefinger through all of the other three loops to pick up a loop on the other side, so my method of working leads straight back to the 5-loop twill. Note though, that I’m not 100% sure if I got the braid structure correct, I’ll have to do some more reading and braiding before I’ll know for sure. (And Speiser mentions in Old English Pattern Books for Loop Braiding that she finds the two-loops-on-one-finger method faster, so maybe I just haven’t practiced enough yet….)

Again, I used pearl cotton #8 and Katia Luxury and the finished sample is of the same width as the original braid. My sample is actually narrower than the original, only 6 mm wide; pearl cotton (or silk!) #5 might give the correct result. Note that the metal thread used in the original is bundled, there’s one (!) bundle of two threads and three bundles of three metal threads.


Braid 10 is a wider variant of braid 9 and requires six hands, i.e. three workers. I don’t think I’ll get this far for a while, I’ll need some experience in multi-worker fingerloop braiding before I’ll be able to recreate this one (the description is quite clear, I’m just not yet equipped to understand it!).

Braids 11 is an openwork lattice braids that need several braiders (6-8 hands), although the number of active loops at any moment can be managed by one person. Thus I think I might be able to recreate these and other similar braids (like the Catherine Wheel in the 17th century manuscripts), but I have to do a bit of research and thinking first.
NB. These openwork braids are the ones that are often referred to as the earliest examples of bobbin lace in Scandinavia. However, Speiser’s argumentation, based on some mistakes on the braids and the number of elements, is quite solid and she’s supported by the 17th century braiding books. So I’m with her there and suspect that one day more early openwork lattice insertions that have been classified as bobbin lace may turn out to be fingerloop braids (although the early bobbin lace braids are very similar in structure).


Braid 12 is similar to Braid 13 (see below), but made with red silk and gold wrapped on a silk core.


Braid 13: On a reliquary in Linköping Cathedral, blue silk, silver metal wrapped over an off-white core, width about 6 mm.

This braid is structurally single-course oblique twining and the pattern is exactly the same as nr. 29, A Broad Lace Chevron, in Take V Bowes Departed. This is the only one of these Bridgettine braids that uses silver-coloured metal thread, all others have gold as metal. Interestingly, this braid is sewn next to a woven tape in red and gold!

Whilst making this braid I followed the instructions in Tak V Bowes and not the ones by Speiser, so my braid may have a tiny structural difference to the original (I’ll have to take a closer look at this). As the threads in the original braid are paired, I used the metal thread paired and then a single width of the thicker cotton as of some reason I don’t have any #8 cotton in blue (I’ve got all the other heraldic colours, but not blue – I wonder what I’ve been thinking about…). This combo made at least a pretty balanced braid; the original has lost most of the metal and the silk may have lost some softness, so the braid in the photo is very open compared with the usually quite dense structure of the twined braids, evident in my sample.



This is what Noémi Speiser writes on the use of the braids on pages 3-5 of Loop Braiding in Swedish Bridgettine Tradition:

The nuns of Vadstena apparently meant to convey something with each braid they added to an object. Braid was used both decoratively and functionally, and most often both simultaneously.
…the pliability of the braids very often decided the chose of braid type: a woven band is far stiffer [than a loop braid].

Most commonly the braids occur as an edge decoration, an outer finish.
Some braids were purely decorative.
Occasionally braids seem to have provided an emergency expedient when things perhaps did not turn out quite as the nuns intended.)

One thing that many people remark on at seeing my fingerloop braiding samples in general is their colourfulness and variety. The nuns at Vadstena both understood and utilised these properties in their work with ecclesiastic textiles and objects. As the embroideries and fabrics they worked with were of the highest standard, even their braids are made with silk, silver and gold. The braids blend in with the objects, I find that the high bling factor of these braids comes out better in my samples than in the photos – of course it helps that my gold and silver are bright and untarnished, these items must have been truly wonderful when they were new.

There’s one item I’m planning to recreate sometime this year and that’s a velvet covered box in Sund church. It has a simple patterned velvet covering and all the seams and edges are covered with braid 8 (10-loop twill), that is also used as a hinge. I know I have a wooden box stashed somewhere waiting for sanding and painting, but I could just as well cover it. If I mess up, I’ll lose my work, but otherwise the investment is small – and if I don’t mess up, I’ll have something very good looking and practical at my disposal to use e.g. as cash box cover at medieval markets. The photos in the book of the Sund box are pretty clear and as the covering is in such a bad state that the seams are falling apart, the construction is fairly easy to detect. The braids seem to play a major part in keeping the covering in place, stretchy as they are.
(Unfortunately I don’t seem to be able to find any photos online of that box or similar boxes/shrines, I’ll add them when I find something or get my own done. May just be that I can’t come up with the correct term in Swedish or something.)