TAPESTRY FLOWERS
Shawl Weaving in the later part of the 20th Century

The various colored tojlis are arranged according to the pattern and for a simple coarse shawl, like those being woven in Kanihama, approximately 3-400 were necessary for a width of three feet. However that number was multiplied by three or four times to make the fine historic examples. Each tojlis was inserted one at a time (fig. K6 and K7) since no shuttle (a labor saving device employed in many other types of hand-weaving processes) or sheds (another labor saving device which separated adjacent warp thread to allow the shuttle to be quickly passed or thrown from edge to edge) were used. Not only did the tojlis have to be inserted one at a time but, each time the pattern called for another color to define the design, the two tojlis had to be interlocked or joined together.

Tapestry-twill weaving proceeded from side to side, with each tojlis going over and under successive pairs of warp threads. Figure K8 shows the wrapped tojlis and a section of weaving from another angle that might make this description easier to understand. By transversing over and under pairs of warps threads rather than single ones the resulting fabric gained a considerable level of additional strength, an important factor considering the fineness of the threads. Each over/under insertion(using pairs of warp threads) proceeded from one side of the fabric to the other making the required interlockings at every color join. When this was completed the weaver then packed the weft thread closer together (figs. K9, K10and K11). Figure K12 shows the weaver now ready to start inserting the tojlis to begin the process again.

The patterns of the shawls woven in Kanihama were quite simple, based on repeating the same pattern and once the weaver learned the required number of insertions, the work could proceed rather easily. There is little need for a diagram or someone to call out the required moves, however, this was not the case for the historic shawls with their very complicated patterns. It is believed the weavers of these shawls would listen to the master weaver call out each separate insertion, supposedly this was done because the weaving diagrams, called talem, were written in a code only the master could read. Figure K13 shows a talem and figure K14 the shawl cloth woven from this pattern, which by the way required four such pages written on both sides.

Designing of the shawls was done by a pattern maker (fig. K15). Another specialist then translated the pattern into the required weaving steps and wrote the code for1the talem. At Kanihama there were no talem in sight and it appeared the reason for this was the simplicity of the designs and the weavers familiarity with them did not require their use.