Weaving an 11th Century Irish Brat
By THL Cassandra of Glastonbury
I can document fake fur?! Once I wrapped my brain around that concept, the fun began.I found references to pile weaves in both Scandinavian and Irish sources. North European Symposium on Archaeological Textiles to the rescue! NESAT is my favorite textile reference source. My search led me to weaving techniques and hauling out the books on 1960's weaving. So, here is a timeline: Icelandic Viking, Ireland of Henry VIII, and Hippie. This was an intellectual exercise, honest...
My goal was to make a practical garment from the sheep out. That meant spinning, dyeing, weaving, finishing, sewing, and embellishing the garment. It also meant researching construction and methodology - what was done, and how it was done.
Wool was an integral part of medieval economy and was commonly used for cloth production since the Roman times. It was common to use S-Twist  singles for the weft  and Z-Twist  singles  for the warp.  This allowed the fibers to lock together better in the fulling  process. It also helped prevent the grain of the fabric from skewing or shifting, since spun singles have energy stored in the twist, unlike plied yarns. 
Brown is easily produced either from dyes or from the natural color of some sheep.
Dyed yarns were generally much preferred, since wealthy and aristocratic are regularly described as wearing brightly coloured clothes, while only the poor and the pious wore undyed wool cloth. [Walton]
Walnut hulls are available locally for only the effort of gathering them, another sign of economy, then and now.
Cloth of this nature could have been woven on a variety of looms, from a warp-weighted loom  to a frame loom to a horizontal loom  such as I used. Warp-weighted looms were used from pre-Viking times to the 12th century when they faded from common use. Horizontal looms in period would resemble a modern counterbalance loom where the shed  drops instead of rising. They arrived in Europe in the 11th century.
Straight and reverse twills have been woven since Neolithic times. A twill weave  makes a closer and stronger cloth than a tabby weave,  thus making the cloth thicker, warmer and more durable. The diagonal lines have the added benefit of making the cloth more pliable. The tufts or pile adds a factor of extra warmth and water resistance to the cloth by creating an added layer and insulation, making it very desirable for outdoor wear in a wet and cold climate.
Pile cloths can be found from Viking times to Ireland of Henry VIII, who forbade his subjects to wear such barbarous clothing. Two examples were found in the Cronk Moar dig. Seven piled weaves (tabby or twill) were found in Anglo-Saxon England (350-1050 AD) and catalogued under "indeterminables" by Lise Bender Jorgensen. A ninth century Icelandic fragment comprised of two pieces joined by a seam. The ground weave was a four-end twill.
Interestingly enough, the pile consists, not of spun yarn but of continuous locks or tufts of long hair, between 15 and 19 centimetres long, taken straight from the fleece. [Geiger]
An Anglo-Scandinavian archaeological find with a thread count of 5 threads per centimeter, Z x S-spun yarn, and dyed with madder was found at Copperate and detailed by Penelope Walton. Other fragments were found at Lund, Sweden; Wolin, Poland; York, England; Dublin, Ireland; Heynes, Iceland; and Birka Sweden. The most famous shaggy mantle is the Mantle of St. Brighid. It has a tabby weave and the nap is drawn and curled.
Standardization of measure was brought about by the weaving guilds and sometimes mandated by laws. Even then, it could vary depending on region For instance, the standard shaggy coat (Icelandic) is defined as being "four thumb ells long and two thumb ells wide with thirteen tufts across" [Jochens] Thread count of cloth found in the Middle Ages varied widely. Some coarse fabric has been found with an ends-per-inch (epi) between 8 and 24. Fine fabrics were found between 95 and 248 epi. Of the three Irish finds from the Roman Period, "the counts vary between 8/7 and 17/16 threads/cm" [Jorgensen]. This comes out to 31 to 67 epi and is a medium fabric.
The Irish Brat is a simple rectangular piece of cloth pinned at one shoulder. A variation of this basic design has an oval hem.
The brat was the most colorful, versatile and warmest garment in the early Irish wardrobe. It was four-cornered, roughly rectangular in shape and being of wool was probably treated or 'fulled,' to a dense finish. [Dunlevy]
Woven pile cloaks also formed part of the traditional costume of the Scottish highlander. They could be rectangular or semi-oval.
Irish Brat Modeled by Baron Thorfinn Grimkelsson, Baron of Andelcrag
They were also found in earlier Icelandic lore and law. The tufted coat (roggvarfeldr) or pile coat was a significant trade item. Given the scarcity of fur, this would have been a good substitute since it does not involve killing animals for their fur and would protect the wearer from bad weather.
Law specified that a standard coat must have thirteen tufts across the width. [Jochens]
The coat also doubled as a blanket. By 1200, the shaggy coats had gone out of fashion.
Curly fleece mantles were described in Irish folklore as chronicled by monks in the 11th and 12th centuries. Several characters were described as wearing them in "The Destruction of Da Derga's Hostel."
A possible early Irish example is the relic known as the Mantle of St. Brigid... This piece is now in Belgium but is believed to be of Irish origin dating to the 11th century rather than the sixth, the presumed lifetime of the saint. The textile is red in colour and is described as having a thick curly fleece woven in with either the warp (probably) or the weft. [Heckett]
The same fabric could also be used for a comforter or bedspread. The bedspreads were favored among monastic communities and among the less well to do.
I decided on Coopsworth sheep for the warp and weft, and Border Leister sheep for the locks. I was informed by a group of local spinners that both breeds have excellent felting properties and that they were good choices for a novice spinner. The Border Leister had long locks and luminosity that I found attractive. Not being allowed to raise sheep in the city limits, I bought my wool from sheep farmers.
Since the wool was raw and I had no wish to store grease and dirt, I scoured it. For sanitation reasons, I used soap to wash the wool instead of stale or fermented urine (as would be period). As the Border Leister was going to be used for the locks, I did not need to process it further.
I set to carding the Coopsworth into rolags using hand cards to make the wool easier to spin. I picked out as much of the second cuts and vegetable matter as I could before carding. However, some remained behind causing noils, or lumps, in the yarn and frustration to the spinner. I also managed to build up some impressive muscles in my back and shoulders.
The yarn was hand-spun on a wheel. For the warp, I tightly spun the wool into Z-twist singles. For the weft, I spun the wool into S-twist singles. I spun 1,300 yards for the weft and 1,800 yards for the warp.
By the time I was done, I could see a noticeable improvement in my spinning. Over the course of processing the wool into yarn, I went from 5 pounds of raw wool to 3 pounds of yarn. The entire operation from raw wool to yarn took an entire summer.
A single color fabric could be produced by either dyeing the fleece (in the wool) before spinning, dyeing the spun yarn (in yarn) or dyeing the cloth (in piece); while multicolored fabric could only be produced by weaving with dyed yarns. Thus, the dyeing process had to occur in the wool or in the yarn as I wanted to leave the locks in the natural color. Plus, I did not have a pot or stove big enough to dye the finished cloth.
I created a subtle design effect by using dyed yarn for the warp and weft while leaving the locks in their natural state. I used walnut hulls to obtain a light brown yarn. I gathered five pounds of walnuts in the hull for dyeing, valiantly fighting off vicious squirrels. No one warned me about the grubs that live in the walnut hulls. Ick! Ewwww! Time to scrub the kitchen!
I boiled the hulls for two hours to create the dye bath. Since walnuts are a natural source of tannin, it was not necessary to add mordant the yarn prior to dyeing. Then I added in wet, unmordanted wool, dyeing first the warp and then the weft. That way, any color change between the two dye baths would add variety and subtle patterning to my weaving. I simmered each batch for two hours. Then I rinsed the dyed yarn in warm water until all the excess dye was removed. Finally, I hung the yarn to dry. If there was any hue change between the dye baths, I could not discern it.
My next step was to size the Z-Twist warp yarn to prevent fuzzies, limit breaking, and to add stability to the yarn during the weaving process. I chose a gelatin sizing because I had previous success with it when weaving very sticky mohair. Flaxseed would have been the choice in the Middle Ages, but since I was using my precious hand-spun wool, I decided to go with a formula that had proven itself to me. And again, I needed to clean the kitchen since the sizing ended up all over everything.
I used straight 2/2 twill for my base fabric and inserted locks of wool into the weave. Firmly beaten in and later fulled, the locks are held in place without knots. I used a draft found in NESAT as a starting point and built a 2/2 twill with locks inserted in every fifth shed.
The draft I created : Pile Weave Draft
The red on every 5th row indicates the inlay of wool locks
I chose to make the cloth 21" wide so instead of wasting cloth, I could use selvages  for the edges and middle seam of the Irish Brat, making for a sturdier garment. Length is a function of the purpose of the cloth and the amount of yarn available to the weaver for the project. I chose a length that would (when doubled) give me a 7-foot brat. My thread count is 6 epi, chosen to provide ample room to clear a shed  and allow for the diameter variance of hand-spun yarn.
First, the Z-twist yarn was measured out on a warping board to produce the length and width of the desired cloth. Next, the loom was warped. This was the foundation upon which the weaving happened. I warp back to front, so I start by securing the warp to the back beam. The warp yarn was passed through heddles, attached to one of four harnesses to form a pattern according to the draft (see Figure 1) and through the reed. Then, the yarn was tied to the front beam to provide tension. The S-twist yarn was wrapped around the spool and put in the shuttle for the weft.
Then the cloth was woven. Weaving is accomplished by interlacing two sets of yarn to form fabric. By manipulating the four sheds up and down in combinations, one can change which set of yarn is on top, thus creating a pattern. The shuttle was passed into the opening between the two sets of warp yarn to place the weft yarn. The weft is beat into place with the reed. Every fifth shed, I would insert locks by wrapping them around the warp threads that were raised. As I finished with each spool, I bound the yarn off at the selvage by overlapping 1-2" and binding the old yarn into the floating selvage for a few beats. The process was repeated with the change of the spools until all available warp yarn is woven or one has woven the cloth to its desired length.
After weaving, the cloth was cut off the loom, the waste yarn was removed, and loose ends were woven into the cloth. I sewed the two pieces of cloth together at the selvages because I wanted them to felt in the wash, if possible.
Prewash size and look
The cloth was then washed, fulled, lightly felted, and air-dried. The fulling and felting occurred in the bathtub. I used Ivory soap instead of fuller's earth (which is period) due to a desire not to clog my pipes. I also needed the hot, soapy water to remove the gelatin sizing I used on the warp threads. After washing the cloth, I proceeded to felt the locks.
I rolled the cloth first from fringe to fringe and walked across it for several minutes, then reversed the roll. I repeated the process until I notice the locks had started to tangle together. The fulling occurred at the same time as the felting. The yarn expanded and filled in the mesh, helping to secure the locks in place.
The front and back of the Irish Brat
After drying, I noticed that some of the locks had cowardly fled their carefully assigned positions to either escape or to felt in with other locks. I also noticed that the felting had occurred differently in places. Some areas felted into a "confused mass" as expected. Others resisted my efforts entirely. And some matted down into the curls I hoped to develop. As my sample had felted evenly, I was surprised. My assumption was that because I used fleece from the entire sheep, instead of picking prime locks, the fibers had slightly different properties and acted accordingly.
The ends were twisted to prevent raveling. Twisting the ends to prevent the edges from unraveling until the cloth is used finishes the cloth. Hemming or knotting the ends would also have provided the same function. As this cloth was made into an Irish brat, the fringe would add decoration and serve as an additional watershed.
This was my first attempt at garb from the sheep out. I am very pleased with the overall results. My spinning improved dramatically over the course of the year. The project provided me with an excuse to play with natural dyeing. I learned that wool from the same sheep might behave inconsistently. I also gained a huge appreciation for the sheer effort it took our ancestors to keep their families clothed.
Amos, Alden, Big Book of Handspinning, Interweave Press, Inc., Loveland, CO, 2001
Barber, Elizabeth Wayland, Prehistoric Textiles, WW Norton & Company, New York, NY, 1991
Barber, Elizabeth Wayland, Women's Work: The First 20,000 Years, WW Norton & Company, New York, NY, 1995
Bersu, Gerhard and Wilson, David M., Three Viking Graves in the Isle of Man, Society for Medieval Archaeology, London, England, 1966
Biezerman, Suzanne and Searle, Karen, Finishes in the Ethnic Tradition, Dos Tejedoras, St. Paul, MN, 1978
Dunlevy, Mairead, Dress in Ireland, Holmes & Meier Publishing, Inc., New York, NY, 1989
Geijer, Agnes, History of Textile Art, A, Sotheby Parke Bernet Publications, Covent Garden, London, 1979
Hald, Margrethe, Comparative Study of Costume and Iron Age Textiles, A, National Museum of Denmark, Denmark, 1980
Jochens, Jenny, Women in Old Norse Society, Cornell University Press, Ithaca, NY, 1995, ISBN: 0-8014-8520-7
Jorgensen, L. and Munksgaard, E., eds, Archaeological Textiles in Northern Europe: Report from the 4th NESAT Sypmpsium 1-5 May 1990 in Copenhagen, Hungary, 1992
Jorgensen, L. and Rinaldo, C., Textiles in European Archaeology: Report from the 6th NESAT Symposium, 7-11th May 1996 in Boras, Goteborg University, Sweden, 1998
Jorgensen, Lise, North European textiles until AD 1000, Aarhus University Press, Aarhus, Denmark, 1991
Leggett, William F., Ancient and Medieval Dyes, Chemical Publishing Co, Inc, Brooklyn, NY, 1944
Thornton, Peter, Italian Renaissance Interior, The, Harry N. Abrams, Inc., Publishers, New York, NY, 1991, ISBN: 0-8109-3459-0
Turner, RC & Scaife, RG, Bog Bodies: New Discoveries and New Perspectives, British Museum Press, London, England, 1995, ISBN: 0-7141-2305-6
Walton Rogers, Penelope, Textile Production at 16-22 Coppergate, Council for British Archaeology, York, England, 1997
Walton, Penelope and Wild, John, Textiles in Northern Archaeology: NESAT III: Textile Symposium in York 6 -9 May 1987, Archetype Publications, London, 1990
Walton, Penelope, Textiles, Cordage and Raw Fibre from 16-22 Coppergate, Council for British Archaeology, York, England, 1989
Weigle, Palmy, Ancient Dyes for Modern Weavers, Watson-Guptill Publications, New York, NY, 1974
Wild, J. P., Textile Manufacture in the Northern Roman Provinces, Cambridge University Press, London, England, 1970
Wild, John, Textiles in Archaeology, Shire Publications LTD, Aylesbury, England, 1988
Wilson, Jean, Pile Weaves, The, Van Nostrand Reinhold Co., New York, NY, 1974
Zielinski, S. A., Encyclopaedia of Hand-Weaving, The Ryerson Press, Toronto, Canada, 1959
S-Twist is yarn spun clockwise.
Weft is the set of threads that go across the fabric at right angles to the warp.
Z-Twist is yarn spun counterclockwise.
Singles are one strand of yarn that contains energy in the twist.
Warp is a set of parallel yarns stretched on the loom.
Fulling is the process of shrinking and thickening the cloth after it's been woven.
Plied yarn consists of two or more strands of yarn, which are usually balanced and do not contain energy.
Warp-weighted loom is constructed of two upright beams connected at the top by a bar called the cloth bar, from which the warp threads are hung.
Horizontal loom is any loom where the warp is stretched horizontally.
Shed is the triangle formed by two parts of the warp and the reed when any of the treadles are depressed. Insert weft here.
Twill weave is based on a unit of three or more warps and three or more wefts, with the weft crossing two or more adjacent warps to create a pattern in the woven cloth. These can range from diagonal lines to complex diamonds.
Tabby weave is an even, balanced cloth where each weft passes under one warp. It is also known as plain weave.
 Locks are long pieces of unprocessed wool removed from the sheep by cutting.
 Carding is a process used to open and organize the fiber for easier spinning.
 Rolags are slender tubes of rolled-up, fluffy wool batts pulled off the cards.
 Mordant is a fixing agent used to help set the dye.
 Selvage is the part of the cloth that forms an edge itself (without hemming or fraying) on the left and right hand sides of the cloth.
Shed is the triangle formed by two parts of the warp and the reed when any of the treadles are depressed. Insert weft here.
Heddles are cords or wires that control one warp thread.
Draft is a graphic representation of the threading, tie-up, treadling, and draw-down.