The Emergence of a Kurdish Rug Type
by William Eagleton 

Oriental Rug Review, Vol. 9, No. 5 

For most of us a rug type exists only after it has reached international markets and has been seen and classified by collectors and scholars. Until then rugs and other weavings can be present in considerable numbers in their natural habitats without attracting the attention of the outside world. This was the case with many Anatolian, Iranian, and Iraqi kilims, bags, and trappings 20 years ago; however, it seemed unlikely that by 1988 any major rug type was still awaiting discovery. Nevertheless, such a weaving has recently emerged from the Kurdish heartland of Eastern Turkey.

Illustration 1. Hartushi common type

Later research has revealed that the rugs were woven by all of the 11 sections of the large Hartushi tribe, of which the Goyan were originally a part. Hence the preferred label for these is now Hartushi; but, if you are looking for one in Istanbul or Konya, you will find that they are usually referred to as "Herkis." This is a classic example of how a rug type comes to be mislabeled by the trade. Turkish dealers have long known, and carried in their inventories, rugs from the Van region labeled "Herki" after the large nomadic Herki and associated tribes which, until recently, moved back and forth within the frontier area of Iraq, Iran, and Turkey. The Herki rugs, however, are quite different in dyes, design, size and structure. The Herkis (Illustration 2) are almost all in the roughly one meter by two meter format, while the Hartushis are 125 to 160 centimeters wide and two-and-a-half to three and-a-half meters long, with some even larger sizes available. By chance, two of the rugs I found in the Mardin furniture shop were in a narrower runner format, but this appears to be atypical.

As can be seen in the illustration, the Herki rugs have rather complicated designs, either original or borrowed, and the fields are usually adorned with a variety of small devices that are typical of the tribal Kurdish repertoire. Most Hartushis, however, have small borders and large dramatic medallion type centers repeated in twos, threes, fours, or fives, or fractions thereof. The fields are relatively uncluttered.

Back to the process of discovery: it was not until September, 1987, in a stopover in Istanbul that I saw two of these rugs in a repair shop on the edge of the Istanbul bazaar. I asked a friend to try to find others, but it was some months later that I discovered that, if I had looked in the right places, I could have seen hundreds in Istanbul.

The next step was a visit a few months later to Konya where Hakan Tazecan, knowing of my interest in Kurdish rugs, told me he had just bought 35 Herkis. Would I like to see them? The prospect of seeing Herkis after four years in Iraq where they were available in large numbers did not excite me; but, when he displayed-his "Herkis," I realized he was in possession of a hoard of what we now call Hartushis. Shortly thereafter, similar hoards were found in Istanbul.

The mystery, however, had not yet been cleared up. Why had these rugs not been seen in the market before, and why had they suddenly appeared in such large numbers? All agreed that pickers were bringing them in unwashed from the East by the truckload and peddling them in the Istanbul and West Anatolian rug markets. Initially, the rugs were not as popular with dealers as they might be with collectors since they are very heavy, require considerable space, and are hard to manage, both by the boys who lay them out and by the repair crews. Despite full, fleecy piles, many of them have holes or other defects when they are collected from villages, and these must be corrected before they are salable. Some of them are deemed non-commercial because of the prominent use of bright orange or pink. In Turkey this can be remedied at little expense by substituting more acceptable yarns.

We have not yet had an opportunity to submit the fibers of these rugs for dye analyses. Many of the dyes, particularly in the later examples, are probably chemical, including certainly the pinks and the bright oranges. It is remarkable, however, that they rarely run with washing, although in 'some rugs the reds of a certain period (perhaps 40 to 60 years ago) have faded to abrashed light red. Many colors, however, remain vivid. The blues, blue-greens, yellows, and browns, and most reds are impressive. The overall effect is reminiscent of some of the crude, but much appreciated, Caucasian rugs of the late 19th and early 20th centuries.

Most Hartushi rug designs are native to the Kurdish Eastern Anatolian-Caucasian area, though an occasional piece is influenced by other Turkish weaving. The execution, however, is from the mind, or possibly from another rug, but not from a detailed drawing. Inquiries during our investigation in the Van region in June, 1988, led to the conclusion that the Hartushi rug designs, like the Hartushi (Van) kilim designs, are woven interchangeably by all sections of the large Hartushi tribe. Hence, there is no way to assign designs to specific sections of the tribe. The major rug designs bear little resemblance to the repeat designs found in the Hartushi kilims. We have found this to be the rule with other Kurdish tribal weavings as well. The only exceptions are some of the simple borders where there may be similarities between the kilims and the rugs.

The most persistent and notable field design is a shield which is referred to as a "turtle" in Kurdish and Turkish. It is found on about 20% of the Hartushi rugs. This dramatic device is seen in weavings from the Iranian and (allegedly) Soviet Azerbaijans. It is hard to believe that it developed from urban influence or was originally conceived to represent a turtle. It is best known to the rug world from soumack bagfaces attributed to the Caucasus, one of which appeared on the cover of the 1969 Textile Museum publication, From the Bosphorus to Samarkand, Flat Woven Rugs. Others recently seen at auctions are not only similar to those found in Hartushi rugs but are, in fact, identical in at least one case (Illustration 3, pg. 24), including minor field devices. Perhaps we should reexamine the claimed Caucasian parentage of the soumak bagfaces, though it is not suggested here that they are necessarily produced by the Hartushis. For those who favor antique origins, geometric human figures with arms held out and up can be found on artifacts from the period 1300 to 800 B.C. attributed to Anatolian peoples, including Hittites. In addition to the "turtle" (Illustration 3) and a variety of Memling guls (Illustration 1), eight-pointed stars are common, both as major field designs and as adornments. In fact, the centers of most "turtles" contain large, eight-pointed stars.

Illustration 4, Hartushi backs, some showing triangular add-ons

This is a question we put to a number of people in the Van-Hakkari region during a trip to eastern Turkey in June, 1988. The most common answer was, "Only recently did the pickers get into the villages to bring the rugs out," but this is, of course, an incomplete answer. It has been suggested that political developments in the region might have had an effect, and indeed some sources claim that rugs have been brought out of villages in northern Iraq near the Turkish border wherethere have been movements of people from Iraqi Kurdish villages to resettlement areas. The best explanation, it seems, is that those who own the rugs no longer prize them as highly as they do cash or new machine-made rugs which are given in exchange, and once the process got underway, the profit motive takes the pickers deeper into the mountains in search of more and more rugs.

It is not at all certain how many Hartushi rugs will be available for the many collectors who are looking for tribal rugs untouched by urban or commercial influences. If for several years there is a glut, this does not mean that they are inexhaustible. Tens of thousands of Jaf bags flooded the U. S. market in the 1 920s and '30s, but they are now rare where they originated in Iran and Iraq. There are probably 300 to 500 Hartushi rugs now in the retail and wholesale outlets of Istanbul, and several hundred more in the other centers of Turkey. Perhaps a thousand or two still remain to be brought, out of the mountains. It can be a risky business going into these villages unless the picker has somehow gained the confidence of the tribesmen and women. However, greed will win out and, before long, most of the Hartushi rugs will have been extracted, and many of them will be in Europe and the United States. During the past few months prices have risen, perhaps indicating that supplies are drying up -- or that demand is increasing.

It is likely that enough of these rugs will soon be available to form a group that will find its place in the rug vocabulary. When this happens, let us begin by discarding the "Herki" label and calling the rugs after their tribal origin, "Hartushis," or if attribution to the area rather than the tribe is preferred, "Hakkaris."