Talk by Jon Thompson to
International Hajji Baba Society
September 11, 2005

How the Great Carpets Were Made


Dr. Thompson’s talk was an instance of his belief that there are valuable things yet to learn about the carpets that exist if we study them closely.

He began his presentation with a basic review of how rugs are woven.

He then indicated that a variety of methods used were used to guide weavers as they produced particular designs. In each case he presented examples and noted aspects of the designs produced in particular ways.

First, he said there is “improvisation.” In such rugs the weavers work without any pattern to guide them. He showed rectilinear designs that are the frequent result of such improvisation.

“Memorization” is the first of several ways in which desired designs were conveyed to weavers. He showed mothers working with daughters at a loom as an instance of the use of memorization. Rugs woven from memory tend generally to have small-scale rectilinear designs and a fairly narrow color palette. Thompson said that the memorization of rug patterns is analogous to the learning in language and music.

A second method used to convey rug patterns to weavers is copying from other rugs and textiles and other media (for example, from bookplates, or pottery or the designs on and in buildings). This method might be called the “picture-guided” method. Note that this method can allow the weaver some latitude since there is often no precise knot plan provided (although the use of another knotted rug would permit knot for knot copying if the piece meant to be copied was at hand). Thompson has shown elsewhere that the Ardebil Carpet is full of irregularities. He believes that this suggests that the weavers were following shapes not counting knots.

A third method of guiding the weaver as he/she produces patterns in rugs is the “exact knot plan” method. This method is intended to eliminate weaver variation (although that is not always achieved). Under this approach the design is recorded on graph paper with each square on the sheet of graph paper representing one knot. The color of each knot is the information recorded. The equivalent of modern graph paper, Thompson said, existed as early as the 12th century and drawing straight lines was in some places an established craft.

A fourth method of conveying rug patterns to weavers is the “digital” method. Here, the information on the knot plan is converted to a linear text that can be read or called out. Such digital systems tend to convert that knot plan design to written notations that can be read directly. Codes are not usual.

Thompson believes, as was illustrated once above, that a close examination of a rug can often disclose which method was used to guide the weaver in weaving a particular pattern. For example, he cited two high quality carpets known to have been woven to a very precise pattern. Thompson found while examining these two carpets that there was a small irregularity that occurred in precisely the same place on these two weavings. He takes this as evidence that the weavers were following a digital system, likely called out, rather than a knot plan.

Thompson made some additional comments about tendencies he sees resulting from the various ways in which rug patterns were/are conveyed. He said that in general patterns move from more restrictive to less restrictive media. This is similar to an argument Marla Mallett has made but Marla’s argument moves at the level of more or less restrictive weaving techniques (for example, more restrictive slit weave tapestry to less restrictive sumac or pile weaving). Thompson made this argument between different media like metal work versus weaving.

Thompson sees “law-like, irreversible” movement as rugs are copied and recopied. The usual move is toward simplification of older designs, although new features also sometimes appear. Thompson said that those copying a given pattern don’t understand it and as a result do not copy it accurately. He provided some examples of particular rug patterns as they were copied from a 15th century origin to a recently purchased 21st century example.



Some pictures
click on the thumbnail for the full picture.

The speaker, Dr. Jon Thompson, talking to Richard Isaacson


Some of the 37 IHBS members who attended the presentation














return to home page