An Anglo-Saxon Cyrtel
Very little direct evidence of women’s costume in Anglo-Saxon England survives from the period. Most of what we know of the garment shape and style is derived from scant illumination evidence and from the three female figures in the Bayeux Tapestry with some conjecture from Continental sources. Fabric patterns and styles have survived in some archaeological sites and grave goods. This garment is my best, educated guess as to a basic cyrtel or sleeved over-garment typical to an Anglo-Saxon upper class woman (thegn) living around the era of the Conquest of England (1066). A cyrtel was the basic dress worn by women during this period. It would have been worn over an undergarment whose sleeves may have peeked from beneath the looser sleeve of the cyrtel. The gown was ankle length, without a train. The sleeves could be long and full, like Countess Judith of Flanders, or close to the wrist like those of Aelfgyva from the Bayuex Tapestry. The images suggest that the garment was fitted at the waist (indicate by the placket at Judith’s waist and the folds of fabric on Aelfgyva – the latter may be artistic license). In recreating this garment, the drape is of utmost importance – the evidence suggests that the fabric was heavy enough fall in loose folds against the person and not cling or fly away from the person.
In recreating this garment, I spun the wool to a sett found in archaeological finds, dyed the wool with dyes that would have been available to a women of the period - choosing a color combination that was possible, though not necessarily found in the archaeological record and wove the fabric using a broken lozenge twill that, judging by the quantity of found textiles from the period, was favored by the Anglo-Saxons. I cut and sewed the dress modeled from a found gament that provides a similar drape, and hand-sewed it using hand-spun thread and stitches that have been documented to the this era.
My hope is that in using the techniques and materials found in the period I will have created a cyrtel that is exemplary of the period. And, through this documentation, I have put into perspective the amount of time and effort that was needed to create even a simple garment.
All yarn in this gown was hand spun using a spinning wheel with the exception of the sewing thread for which I used a low whorl spindle. Before the 14th century, yarn would have been entirely spun using a spindle as spinning wheels in any form were not yet introduced to England (Walton-Rogers, 1745). I chose to use a wheel in the interest of having this project finished within a nine-month time. I found in the past that I spin approximately 100 yards of singles yarn per hour from combed fiber on a low whorl spindle. In the same period of time, I can spin over 350 yards per hour on a spinning wheel. Even with the speed of the spinning wheel, the spinning of the yarn for his project took over 100 hours.
I used commercially prepared combed wool top to save time as well. Analyzed yarns from existing textiles indicate a majority were spun from parallel fibers likely prepared using wool combs or directly from the wool staple. (Walton, 315) Worsted yarns are preferred for long wearing fabrics with good abrasion resistance and which show woven patterns well. From previous experience I have learned the preparation of the fiber from the raw wool, including grading, scouring, and combing would have added at least 10 hours per pound of wool to the project - an additional 60 hours.
All yarn spun for the fabric itself was spun singles (no ply). The thread used for the seams of the cyrtel was spun Z and plied S to reduce abrasion during stitching.
The warp yarn was spun at approximately 36 wraps per inch with an angle of twist of approximately 25°. I spun approximately 9,000 yards of yarn Z (fig. 1) for the warp and dyed using madder, a dye found in many of the textiles at Coppergate (Walton, 1989, 400). (see Appendix A for process) I used alum as a mordant (Walton, 401) with cream of tartar as a leveling agent and soaked the yarn in the madder overnight. The color was not as rich and red (fig. 4) as I had hoped, but several factors could have been at work there: I was dyeing three pounds in a crowded pot resulting in a lighter color, I only soaked the madder for three days as opposed to a longer period to extract the dye so less dye was present in the bath, and I may have had the temperature of the dye bath too high to extract the best reds. The result was a nice coral color, which could also be considered an exhaust bath color in the period. In addition, I did not stir the dye bath often enough which gave the yarn uneven color across the skeins. I tried to minimize the differences while warping by setting up four skeins (or balls) at a time and mixing them as I warped. It did not work as well as I had hoped and gave some striping to the finished fabric.
The weft was spun at 28 wraps per inch with a twist angle of approximately 25°. The 7,500 yards of the weft warn was spun S (fig. 1) and was dyed using Queen Anne’s Lace (daucus carota) also know as Wild Carrot. (fig. 4). I did not have access to weld (at least not for free!) nor did I have time to grow a plot large enough to dye the three pounds of wool. I used 5 plastic grocery bags and picked the plants from a vacant lot cutting the flowers with about six to eight inches of stem. Interestingly, there is seed evidence for weld in use during this period, but no textiles have survived with the dye still in the wool like that of the alizarin (madder) or indigotin (woad). Since so many plants will give yellow dye and weld dyestuffs were found at Coppergate (Walton, 1997, 1766), I feel confident in using the color in the weft. It is possible that the yellows from weld are more fugitive than that of madder and woad and as such did not survive as well. I used alum and cream of tartar as the mordant and soaked the yarn for 10 minutes. The yarn was so bright I was afraid what would happen if I left it in the pot overnight! (fig. 5) I did not set the twist in any of the yarn specifically - I felt that the dye process would take care of setting the twist for me. I was right. I was concerned I would not dye enough fiber or spin enough yarn from the dyed fiber and would be forced to try to match color at a later time. As such I chose to dye yarn as opposed to the fiber itself.
The fabric was woven in broken lozenge twill (fig. 6). This is a pattern very common to the Anglo-Saxons and found in grave finds all over England. (Bender Jørgensen, 1992, 154-158 – Table of Anglo Saxon Textiles by Elizabeth Crowfoot published within this volume) The weaving was performed on a four-harness rising shed, jack loom (fig. 7 & 8), but would have been likely created on a warp-weighted loom during the period. (Henry, 1998, 157-166) The eleventh century was a transition period from weaving as a domestic craft to a profitable profession. During this time there is a decrease of loom weights found in archaeological sites which indicates a move to the larger, horizontal loom. The weaving width on the horizontal loom is much narrower than that of a warp-weighted or vertical loom which impacted how wide I could weave and how I would lay out the cyrtel on the fabric. I set the warp at 24 ends (threads) per inch which is the coarser end of the found Anglo-Saxon textiles - low range is 22 ends per inch reaching to over 80. (Bender-Jørgensen, 1992, 35 - Fig. 40 ). I sett the warp 30” wide in the reed for a total of 720 ends plus 4 ends at each side for a floating selvedge. Twill weaves tend leave gaps on the sides resulting in wear and abrasion on the side warps. I used a floating selvedge to counteract this effect. I threaded the heddles from front to back, padding the warp on the back beam with paper to prevent warps straying into the hills and valleys created by the other warp threads while winding on. I finished by tying the warp to the apron rod, adjusting the tension across the fabric and I was ready to weave!
I planned to weave the weft in at 18 picks per inch, but the fabric had other ideas. It wanted to be woven at 12 picks per inch. I suspect I had the warp stretched too tight to pack in well or the weft was too tightly spun. I was pleased by the low occurence of warp breakage on this project. I had been intimidated by the thought of using a singles warp, but I found it to be just as responsive as plied yarn. The warps that did tend to break (I had 12 warp breaks) were overspun in the areas the breaks occurred. I think the dyeing helped to set the twist and make the yarn more stable. In addition, I used Suave Styling Spritz as a sizing when the warp starting sticking together and getting fuzzy. I sprayed the warps behind the heddles before I would advance the warp.
I washed the yardage merely to finish the fabric - to let the yarns relax and to shrink the fabric. I did not want to full the fabric as fulling and teasing the nap were later finishes and were seldom used in Anglo-Saxon textiles. (Walton-Rogers, 1771) Fulling, by virtue of swelling the fibers and felting themselves together would have obscured the pattern I worked so hard to weave. I finished the fabric by stomping on it in my bathtub just as one would stomp grapes (in soapy water) to provide gentle agitation in the water. I rinsed the fabric in the bathtub and put the fabric through the spin cycle of the washer (no agitation!) to spin out as much water as I could before drying outside on the deck. (fig. 9).
Layout, Seaming and Finishing
In order to create a garment similar to that of the women illustrated in the Bayeux tapestry and Countess Judith, I looked at similar cut garments in close chonological proximity. I found similar drape in the coronation robes of the Holy Roman Emperors of the 12th century (Payne, 1965, 164) which is later, but is very simple in cut. The Thorsberg costume - dated somewhere in the first through third centuries which is much earlier - is a very simple T with sleeves sewn into place (Payne, 1965, 137). The Egyptian child’s tunic (fig. 13) is dated to the 9th-10th century AD. and has a cut identical to my layout. It is loom-based narrow and has gores to add fullness. (Harris, 61). The cut would have been similar to the shape of a “T”. I have deviated in the cut of the garment from a typical “t” tunic recreationists favor (fig.10) of a wide fabric with a minimum of seams to a loom based cut: a piece for the body, gores for fullness and separately pieced sleeves.
I believe this more accurately reflects the use of fabric in the period. Since the fabric would have been hand-spun and hand-woven, it would have been far more economical and less wasteful to have extra seams in a cyrtel than to spin and weave the extra fabric to add fullness.
The sleeves are of moderate cut - not too tight to hide the garment underneath, not too full to restrict the ability to spin or perform other tasks. Illuminated pieces show some sort of ornamentation at the wrists and hem (Owen-Crocker, 1986, 139) although I did not include any ornamentation at this time. The length of the dress is ankle length (Owen-Crocker, 1986, 139).
The fitting of this gown owes itself to two plackets at the side, sewn into place at each wearing. It is seen in the illumination of Judith Countess of Flanders (fig. 12 & 13). It would be useful to have such expansion available when pregnancy was frequent, but a close fit was fashion. There is scant evidence of this in other sources so this may be an anomaly or a representation of a half-hidden girdle (Doyle, 1999, 27) (Owen-Crocker, 1986, 137).
I sewed all of the seams using a running stitch (Walton, 406). I then felled – that is, I rolled the edge of the fabric over itself and sewed into place – all of the seams to prevent wear and fraying (fig. 14). I used an overcast stitch to stitch the fells into place (Walton, 406) (Crowfoot,156). I used a rolled hem on the sleeves, neck and hem. (Walton, 405)
The time I invested in this project gave me a unique perspective on the amount of work our predecessors invested in a single piece of clothing. The whole of the project took over 200 hours from the spinning to the dyeing to the weaving and the sewing - and I took a significant short-cut using prepared wool and a spinning wheel. It is difficult to put into words how much respect I have for the women who had to make all of their garments from the sheep to the spindle to the loom by their own hands. The spinning is the most time-consuming part of any textile. While I spent only 100 hours spinning the yarn on my spinning wheel, I could easily have spent 300 hours or more using a spindle to spin, let alone preparing the fiber. The weaving took very little time when compared to the spinning and the sewing even less time. With the time spent in mind, I think I better understand how and why the cyrtel was cut so as not to waste any of the precious time invested in the creation of the fabric.
What would I change were I to do this over again?
I would do a better job at laying out the garment and fitting before I started spinning. As a result of inaccurate planning I had an extra three yards of fabric left over from this project. A mantle for me!
I would dye the warp with the Queen Anne’s Lace and the weft with madder. The madder is more expensive and harder to obtain. I would have had a better color result dyeing less fiber. I would also have soaked the madder for a longer period of time.
I would spin finer and set the warp at a minimum of 36 ends per inch. Fabrics in period were very fine indeed and I would like to reflect that.
I hope this project illustrates the time-consuming nature of textiles. From the spinning to the weaving to the sewing I invested over 200 hours. I had some benefits of technology (a spinning wheel) and used prepared wool, but in the end it is a product of my hands, my heart and my mind and that is what really connects me to the spinners and weavers from a long time ago.
Timeline: all time in hours unless noted.
Spinning yarn on a wheel: 100 hours Note: The spinning alone on a spindle would have taken 300+ hours (Calculation based on timing of 100 yards per hour on a spindle)
Time spent mordanting & dyeing yarn:
Queen Anne’s Lace 12
Time spent on weaving the fabric:
Warping the loom: 28
Weaving the fabric: 30
Finishing the fabric: 4
Time spent sewing and assembling:
Getting up courage to make the first cut 1 day
Sewing seams 6
Finishing raw edges 8
Total hours invested in project:* 201
* Not including this documentation.
Bender Jørgensen, Lise. 1992. North European Textiles until AD 1000. Aarhus University Press. Aarhus C, Denmark. Bender Jørgensen, Lise & Rinaldo, Christina (eds.). 1998.
Textiles in European Archaeology. Report from the 6th NESAT Symposium, 7-11th May 1996 in Borås, Sweden.
Banck, Johanna. “Ein merowingerzeitlicher Baumsarg aus Lauchheim/Ostalbkreis - Zur Bergung und Dokumentation der Textilefunde.” pp. 115-124.
Farke, Heidemarie. “Der Männerkittel aus Bernuthsfeld. Beobachtungen während einer Restaurierung.” pp. 99-106.
Gabra-Sanders, Thea. “A Review of Viking-Age Textiles and Fibers from Scotland: an Interim Report.” pp. 177-185.
Henry, Phillippa. “Textiles as Indices of Late Saxon Social Dynamics”. pp. 157-166.
Mannering, Ulla. “The textiles from Nørre Sandergård Vest”. pp. 107-114.
Rœder Knudsen, Lise. “An Iron Age Cloak with Tablet Woven Borders: a New Interpretation of the Method of Production.” pp. 79-84.
Wild, J.P., Cooke, W.D., Cork, C.R., and Fang Lu, L. “Vindolanda: Some Results of the Leverhulme Trust Programme.” pp. 85-95.
Boucher, François. 20,000 Years of Fashion; The History of Costume and Personal Adornment. Harry N. Abrams. NY. 1983.
Broughton West, Barbara. “Weaving Medieval Fabric: Wool Diamond Twills from the First to the Twelfth Centuries in Europe.” Tournaments Illuminated. Issue 106. Spring 1993. pp. 30-34.
Buchanan, Rita. 1995. A Dyer’s Garden: From Plant to Pot Growing Dyes for Natural Fibers. Interweave Press. Loveland, CO.
Buchanan, Rita. 1987. A Weaver’s Garden.Interweave Press. Loveland, CO.
Burnham, Dorothy. 1973. Cut My Cote. Textile Department. Royal Ontario Museum. Ontario. Canada.
Collingwood, Peter. 1982. The Techniques of Tablet Weaving. Watson-Guphill Publications. New York.
Crockett, Candace. 1991. Card Weaving. Interweave Press. Loveland, CO. Crowfoot, Elizabeth;
Pritchard, Frances; Staniland, Kay. 1992. Medieval Finds From Excavations In London. Textiles and Clothing c, 1150 - c. 1450. Museum of London. London, U.K.
Davidson, Mary Frances. The Dye Pot. Self-published, Gatlinburg, TN. 1974.
Doyle, Sarah. “Clothing of Norman Women in the Late 11th and Early 12th Centuries”. Chronicle. Volume 9, Issue 1 (No. 53). Spring 1999. pp. 24-27.
Harris, Jennifer, editor. 1993. Textiles: 5,000 Years. Trustees of the British Museum.
Hill, Mary. “Medieval Dyeing.” Spin-Off.Fall, 1989. Pp. 66-69.
Mayer, Anita Luvera. 1984. Clothing From the Hands that Weave. Interweave Press. Loveland, Colorado.
Owen-Crocker, Gale R. 1986. Dress in Anglo-Saxon England. Manchester University Press. Manchester. U.K.
Priest-Dorman, Carolyn. An SCA Handweaver’s Bibliography. 10 May, 2000. <http://www.cs.vassar.edu/~capriest/textilebiblio.html>.
Payne, Blanche. 1965. History of Costume. Harper & Row, NY.
Regia Anglorum: Members Handbook. 1992
Walton, Penelope. 1989. The Archaeology of York: The Small Finds 17/5. Textiles, Cordage and Raw Fibre from 16-22 Coppergate. Council for British Archaeology.
Walton Rogers, Penelope. 1997. The Archaeological of York: The Small Finds 17/11: Textile Production at 16-22 Coppergate. Council for British Archaeology.
Wilson, Sir David. The Bayeux Tapestry. Alfred A. Knopf. NY. 1985.