Last weekend was the Kingdom of Northshield arts and sciences competition. Here is my entry!
The tablet woven band of a Merovingian Queen
The monastery of Chelles, 25 kilometers east of Paris, was founded around 658-659 CE by Queen Bathilde, widow of Clovis II, then regent of Neustria for her son Clotaire III. The Benedictine nuns were to remain there without much interruption for more than eleven centuries. When the abbey fell victim to the disestablishment of the Catholic Church in 1792 during the French Revolution and was dismantled, the remains were stripped of their precious metals and were moved to the local parish church, where they stood untouched until 1983 when they were excavated and during this event the almost complete contents of two burials, that of the first abbess of Chelles, Bertille (Canonized Sainte-Bertille, pre-congregation), who died around 704-705, and that of Queen Bathilde (Canonized Sainte-Reine-Bathilde in 860 by Pope Nicholas I), who died in Chelles around 680.
Each tomb contained a bundle of bones and a bundle of dust and organic debris. In a third tomb a fifth package contained various stiffened and brittle pieces of fabric, with a label written in 1544: “linges demourez de béate Bauteur et de béate Bertille” . This last package contained the mortuary clothes of the two saints, gathered. It is not known why this was done in 1544.
At first a young English slave, then the wife of King Clovis II, then the mother of the future Clotaire III, Childeric II and Thierry III, Bathilde had an interesting life. At the death of Clovis II in 657, Bathilde assumed the regency of their son Clotaire III. She founded the monastery of Chelles around 658-659. Six or seven years later, when her son came of age, he forced her to retire to Chelles. She died there around 680 and was buried in the main churchyard of Sainte-Croix monastery. Her body was exhumed in 833, by order of Louis the Pious. Her remains have always been preserved with care and honour ever since. The skeleton is quite complete despite numerous but tiny samples taken over the centuries.
The bones of the queen make it possible to restore her silhouette. Slender, she measured about 1m 58cm. Originally blonde, she wore her hair long and divided into two large braids, held by a carefully decorated silk cord.
She was buried wrapped in a large semicircular yellow and pink silk cloak. Her belt was a silk tablet woven band decorated with a woven piece depicting animals. Another tablet woven band, but of uncertain use, depicting geometric patterns was placed on her. This documentation is about that second tablet woven band.
Materials
Tablets
Original: Tablets can be made from various materials. There is evidence of tablets made from wood (5th -4th century BCE, 9th century), bone (4th century, 7th century, 12th century), bronze (5th – 10th century) , just to name a few examples. Given the era the tablets used were likely made of wood or bone. 50 tablets were used for the original piece, 3 for each border and 44 for the design.
Mine: The tablets I used were made from sturdy cardboard I recycled from the back of desktop calendars. It makes for sturdy tablets that are eco-friendly and budget friendly. I used 50 tablets for my piece, 3 for each border, 44 for the design.
Shuttle
Original: If used, the shuttle was likely made of wood or bone.
Mine: My shuttle is made of wood. It is light but sturdy and can hold about 50 yards of 20/2 silk.
Tension Method
Original: There are multiple options the original weaver could have used for a loom. The two most comment method of warp tension in the 7th century were stretched between the weaver and a fixed point and stretched between 2 fix points. Given the length of the fragment the second method is more realistic.
Mine: Although I usually weave with my warp stretched between myself and a fixed point (back-strap), I wove this project on an inkle loom. My reason is simple; I have 2 young kittens that absolutely love to play with dangling warps.
Silk
Original: The band was woven in silk of three different colours, red, blue, and yellow. Considering the era, the silk was likely dyed using madder for the red, woad for the blue, and buckthorn berries for the yellow, all 3 used with alum mordant to create the desired colour (madder with iron mordant will produce browns, greys and purples. Buckthorn berries with iron mordant will produce green).
Mine: I used 20/2 silk I bought from Dixie Weaver at Gulf Wars last month and some red silk I had left bought from Recreated Textiles. I deviate from the original by using 2 shades of red as I ran out of the lighter red while warping. I was short 1.5 tablets so I removed the borders I already warped, switched them with the darker red, finished the 1.5 design tablets I had left and warped the last 3 border tablets with the darker red. Both shades of red are achievable with madder.
Cost
Original: At the time the band was made the Byzantine Empire had the monopoly on manufacturing and selling silk due to some silkworm being smuggled to Europe during Justinian I’s reign. Silk was only sold to authorized buyers. The Franks and Byzantines were not on great terms therefore it is unsure if any Frank would have been on the list of authorized buyers, especially a Catholic establishment. The Franks may have bought their silks from China. Either way, it is likely that silk was an expensive luxury.
Mine: In silk only this project costs about $60. Warping, weaving, and patterning took about 115 hours. If we go by federal minimum wage ($16.65/hr) this project is worth $1,914.75 in time.
Based on cost of material and federal minimum wage, my band is worth $1,974.75.
Technique
There are several different tablet weaving techniques. The one used for this band is referred to as floatwork. In floatwork, 2 warp threads from a given number of tablets are kept floating on top and on the underside of the band with the 2 other threads twisting together and behaving as they would usually do in other techniques. There are 4 different sub-techniques of floatwork. The most popular and recognizable is the Snartemo, the name is sometimes used as synonym for floatwork. Here are examples of all 4 sub-techniques of floatwork:
Snartemo
The Snartemo band is dated to the late 6th century and were found on a farm just outside of Årnes, Norway. It differs itself from the other floatworks by being threaded with all Z tablets and having 4 different colour (green, yellow, red, blue) threads per tablets. There are several Snartemo bands, and the majority uses the floatwork technique.
Saint-Severin
The Saint-Severin band is a floatwork band from Cologne, Germany woven between the 7th and 9th century. It is similar to the Snartemo band when it comes to the use of 4 different colours (white, red, green, yellow) per tablets but the difference is in the threading. The Saint-Severin band is threaded with alternating SZ tablets. The geographical shapes and the checkered patterns are very similar to the subject of this presentation.
Øvre Berge
The Øvre Berge uses 3 different colours; yellow, red, and blue with the yellow thread repeated. Like with the Snartemo band, the Øvre Berge band is threaded in all Z tablets. This 6th century band was found in Lyngdal, Norway.
Chelles
This band is the one studied in this presentation. Like the Øvre Berge band it has 3 colours; red, yellow, and blue with red repeated. Contrary to the Øvre Berge band, the repeating colours are threaded at opposite corners of the tablets instead of next to each other. It follows the Saint-Severin threading with alternating SZ tablet slants.
Planning, warping, and weaving
How many tablets are needed?
I started by taking a zoomed in look at the image of the band using photoshop. I determined that forty-four tablets were used for the design and 3 or 4 for the borders, based on the width of the border I went with 3.
Usually, weavers will start out with a pattern but in this case, I used the picture of the fragment has my pattern. I know the amount of tablets will be an even number because all of the patterns are centered, with the angles meeting in a V right in the center of the width. There are some figures that are similar to other bands I’ve woven therefore I had a head start on how many tablets were needed to create certain designs. In this case I compared it to a Birka inspired band I wove for my fighting tunic. I’ve marked in blue how I counted the tablets on the Birka inspired band and marked with a line the similarity on Bathilde’s band. Half of the Birka band has 10 tablets which means that there’s 20 tablets total. The length of the dragon (?) on Bathilde’s band (blue mark) makes a little less than 1/4 of the band. This means the V shape (blue and turquoise) has 20 tablets. When I copy, paste, and flip horizontally the line from the Bathilde band it fits on each side with a bit of spare room (yellow and fuchsia). This means there’s 20 more tablets. If we look further up the band, we see that there’s about 2 pixels left on each side. 20+20+2+2=44.
With this knowledge and by observing the overall designs I was able to determine that 44 design tablets were needed to weave this. Also, the borders are slightly wider than the space left after counting 40 tablets, I concluded that the borders each had 3 tablets.
Colours
We know by looking at the fragment that there’s 3 colours, red, yellow, and blue. We also know by looking at the structure that there’s no relief on the surface of the band meaning that all tablets are fully threaded; we have 4 threads per tablet. Which colour is doubled?
By looking at the fragment we can see that yellow and blue are never touching, red is always sandwiched between yellow and blue. This is a good indicator that red is the repeated colour and it’s also an indicator that the red threads are not threaded next to each other. If they were threaded next to each other, yellow and blue would touch on the fragment. If we look at a sample from the fragment that consists of horizontal repeats, we can see the red borders yellow and blue.
Tablet slant
If we look at the shape marked by the blue and turquoise lines we notice a V, this indicates that the tablets are slanted S and Z at that exact point. We can safely assume that every V or inverted V point is where tablets are slanted S and Z. By looking at the length of the fragment we can see that all the design tablets are alternating SZSZSZSZ.
Thread spin
Another detail that can be observed is how smooth the surface appears to be. This is due to a small maybe insignificant detail, that can change how light reflex on the woven band. It is due to the threads being warped with their twists all in the same direction. I couldn’t see what direction the threads were on the original and without knowing which end is the start or the end of the band it is impossible to know. I decided to warp my threads all in S.
Calculation error and solution
I had a wee bit of a panic when I ran out of the red silk. I found another darker one, removed my 3 border tablets to change them for the darker red, took these threads to finish the pattern tablets and added 3 dark border tablets on the opposite side. I miscalculated by 1.5 tablets. The original only has one shade of red, mine has two.
Weaving
I started with the top left design (see figure 6). I did not follow a written pattern and preferred to use the image of the fragment as a guide. On the image, the design I started with is slightly narrower than the rest of the band and strangely enough, so is the first design on my band. I’ve noticed that the band seemed to naturally go narrower in the sections between the designs. I corrected this after noticing but the beginning of my band remained narrow, like the original fragment. I always start my bands using small sticks to find my width, this was useless in this case.
For the designs with significant damage, I used similar designs from the band as inspiration since most designs have a “twin” with a different coloured background and often mirror image.
I took the liberty to modify the designs that, in my opinion, looked too much like a swastika.
This band is woven using the floating thread technique which means that the design is created by keeping certain threads on the top of the warp. This is done by avoiding twisting them by turning the tablets back and forth until we need to change colour, then we turn them forward-forward or back-back. Having good warp tension is essential.
It is essential to know the placement of your threads on your tablets. The tablets are threaded red, blue, red, yellow. This means that if you turn forward 1/4 turn and see Yellow on the warp the next time you turn forward 1/4 turn you will get Red but if I go backwards, you will still get yellow. The rule here is: When turning forward, whichever thread that ends up the furthest away from you will be the one showing. When turning backwards the thread that ends up the closest to you will be the one showing.
Conclusion
This project was extremely fun to weave. I’ve only woven about a dozen bands using only a picture of a fragment as a pattern and it’s always a treat. I love to weave bands that are not repetitive, I get bored easily when I’m always weaving the same couple of designs. This one was a different design every time.
I think this would be considered an advanced pattern, but I would recommend new weavers to still give it a shot just to observe the behaviours of the threads. This technique and its sub-techniques have so much to teach newer members that books can’t teach.
What I would do differently
- Now that I know about the narrowing issue I would weave a sample of a few centimeters, leave an unwoven section on my warp, and start my band. I would also plan out my pattern in advance. I loved the challenge of weaving directly from the image of the fragment but it meant a lot of unweaving and trial and error until the design appeared just right. I did take patterning notes while weaving it, so it is possible for me to weave it again following my notes.
- I would also plan my warp better; I would make sure I have more yarn than needed. I was lucky to not have any threads break but if a bright red thread had snapped, I would have been in big trouble.
- This point isn’t a correction but rather a challenge, I would like to weave this piece again with silk I would dye myself using period appropriate dyes.
- I did contact the Musée Alfred Bonno before and during the weaving of this band to get pictures and information about the band. Both times I was told that the museum was closed, and the collection was in storage until 2026. I will wait until the museum opens again and see if I can get more information, details, and photos of the band before weaving it again.
Gallery
Side by side comparison
Sources
- BENGTSSON Anders, La vie de Sainte Bathilde: quatre versions en prose des XIIIe et XVe siècles, publiées avec introduction, notes et glossaire, Volume 54 of Etudes romanes de Lund, Lund University Press, 1996
- BERTHELIIER-AJOT Nadine, Chelles à l’époque mérovingienne. In: Revue du Nord, tome 68, n°269, Avril-juin 1986.
- BOYER Raymond, LAPORTE Jean-Pierre, Trésors de Chelles: sépultures et reliques de la reine Bathilde (vers 680) et de l’abbesse Bertille (vers 704) ; catalogue de l’exposition organisée au Musée Alfred Bonno David Coxall, responsable du Musée, Soc. Archéologique et Historique, 1991
- Catalogue de l’exposition organisée au Musée Alfred Bonno, Chelles, 1991
- COLLINGWOOD Peter, The Techniques of Tablet Weaving, Echo Point Books & Media 2015
- COXAL David, Chelles : site de l’ancienne Abbaye royale: fouilles préalables à l’agrandissement de l’Hôtel de Ville (troisième tranche), 1991-1992, Ville de Chelles, 1994
- HEN Yitzhak, Culture and religion in Merovingian Gaul, A.D. 481-751, E.J. Brill, 1995
- LAPORTE Jean-Pierre, Le trésor des saints de Chelles Volumes 8-9 of Bulletin de la Société Archéologique et Historique de Chelles: Société Archéologique et Historique Issues 8-9 of Bulletin de la Société archéologique et historique de Chelles: Nouvelle série, Société Archéologique et Historique Chelles, Société archéologique et historique de Chelles, 1988
- LAPORTE Jean-Pierre, Les reliques de Chelles, une sépulture royale mérovingienne. In: Bulletin de la Société Nationale des Antiquaires de France, 1987, 1989
- THING Olvik, Viking Age Tablet Weaving Class Handout, 2011
- TORCHET C, Histoire de l’abbaye royale de Notre-Dame de Chelles, Volume 1, Retaux-Bray, 1889
- TWIST, Tablet Weavers’ International Studies & Techniques, Volume 30-1, Spring 2023 edition
- WOLLNY Claudia, Tablets at Work, Das Brettchenweben Grundlagen Technikbuch, Claudia Wollny Edition, 2017
Vefaradrekinn 