An Ancient Carpet in Urumchi

by Chris Walter

From Oriental Rug Review, Vol. 8/2, December/January, 1988

The Xinjiang Uygur Autonomous Region Museum in Urumchi, capital of Xinjiang Region of China, formerly known as East Turkestan, holds an extremely rich collection of apparently very old carpets and other textiles. It appears that the importance of this collection, and some of the pieces in particular, is little known in the West and, I might venture to say, even in China or by the Museum itself.

The subject carpet

Writing in this publication a year ago ( ORR , Oct. 1986) I commented briefly on one intact carpet in particular which was dated by the museum to the Han Dynasty period (206 B.C .- 220 A.D.). This apparently met with incredulity from many quarters. It is well known by most people who concern themselves with the history of carpets and related textiles that the oldest knotted pile carpet ever found up to this point is the "Pazyryk Carpet", from a frozen tomb in the Altai Mountains of Siberia, dating from the 4th century B.C. Following this, the carpet which is recognized by many as being the second greatest in age, and about which much has been written in the last year ( ORR , 1987), is the so called "Fostat Carpet", purportedly found in a rubbish heap near Cairo and having been dated, with some contention, from the 9th century A.D. The Fostat Rug, unlike the Pazyryk, is far from complete. Does there exist another knotted pile rug which is intact, virtually complete, and is far older than the Fostat fragment, though still remaining second in age to the Pazyryk Carpet? Such would seem to be the case.

When commenting on this rug a year ago after my first trip to Xinjiang I mentioned that the only description of this rug and other artifacts in this museum as well was in Chinese and Uygur (Arabic-Persian script). Only careful translation and cross-checking enabled me to ascertain that the date given was Han Dynasty Period (206 B.C .- 220 A.D.). The labeling remains the same today; however, this time I arranged to meet with one of the staff of the Museum, an Uygur with whom I would speak in Turkish, who assured me that the date being described for this piece and some other non-pile textiles was indeed as I had assumed. Additional information I received gave added veracity to the ascribed date.

Firstly, the piece was found by excavation at a site in the Hotan area, to the east of Hotan itself, in the Lop District near Sanpra village, thus the site name given is Lop Sanpra (not to be confused with Lop Nur, in the central Taklamakan). The excavation at this site by Chinese archeologists has been recent, commencing in 1984, so the rug has been uncovered since then, exactly when I don't know. Its relatively recent discovery and the lack of communication between this area and the outside world may also account for the unknown status of this rug.

Unlike the Fostat Carpet, but like the Pazyryk, it was apparently found in situ with other contemporary objects. It is well known from the discovery of ancient manuscripts and other types of textiles that burial in the dry sands of the Taklamakan was unsurpassed among archeological sites in facilitating the preservation of perishable materials. There were reportedly (Silk Road Tour, Urumchi, II/86, a Chinese publication) 52 tombs uncovered at this site, this carpet apparently having been found in one of them. We may presume that the dating of this rug was originally based on its association in site with other contemporary objects, pottery, textiles, and perhaps even manuscripts. The most startling piece of information, however, is that since last year a carbon 14 test has been run on the carpet (Silk Road Tour, II /86, p. 24) with a date being given of between 2290-1715 years old. The 500-600 year span is due to the normal range of accuracy or error for Carbon 14 testing. This date corresponds very closely with the date previously given on the basis of archeological strata and/or association with contemporary objects.

According to the above source this site remains from a period of occupation by the Western Saka people, also identified as "Sicitai" by Herodotus, usually written as Scythian in English. These are, I believe, the same Indo-European speaking people, known from contemporary Chinese manuscripts as the "Big Moon People", whocontrolled the Southern Silk Road oases at the time. More important still, these "Big Moon People", probably one of the Saka tribes, are very likely closely related to the Scythians to whom most scholars have attributed the frozen tombs in the Altai Mountains, where the Pazyryk carpet was found.

The rug itself is nearly square, about 2'4" on a side, and was apparently a saddle fur. It is so described by the museum and this is logical given its format. The Scythians were purportedly a culture which put great emphasis on the horse. The piece is a fairly coarsely woven, I would estimate in the range of 30-40 knots per square inch, although I could not handle the rug or examine its back. I believe it is symmetrically Turkish-knotted, although I could not be absolutely sure for the same reason. The Pazyryk Carpet, which may have been the product of a closely related culture, is considerably finer, in the range of 250 knots per square inch, The Pazyryk Carpet was possibly woven as part of tomb furnishing for an important person; although I have no information about the exact context of its discovery, this Lop Sanpra rug likely was intended for more everyday use as a saddle rug. Could it be that the requirements for ritual use of the Pazyryk Carpet account for its finer weave? Others have suggested that the Pazyryk Carpet was not woven by the same people who entombed it, but was rather acquired through trade. Such questions certainly require far more research. The time gap, the Lop Sanpra carpet being apparently somewhere between two and six centuries later, could point to a degeneration of weaving skill, or simply an entirely different culture with a less developed weaving tradition.

The colors of this Lop Sanpra rug, like those of the Pazyryk Carpet, are still quite strong. They consist of a blue, still quite vivid, probably indigo, a madder-like rust red of still moderate strength, black of moderate strength (not notceably corroded), and a yellowish green which seems to be considerably faded, so that it is hard to distinguish it from the apparently natural beige color wool which it is adjacent to in most places.

I was not allowed to photograph the rug but the museum did provide a recently printed photo taken for the previously cited local publication (Silk Road Tour, Urumchi, 11/86). The design consists basically of a lattice patterned field with vaguely humanoid forms filling the spaces between the lattice, and a right angle meandering vine border with leaf-like forms. There are tassles of wool bound to each corner, apparently attached at some later date as they do not seem to be part of the original structure.

Left
A flatwoven fragment
Right
A wool tapestry woven bag

The museum is rich in other non-pile textiles as well, some of them from the same site and attributed to the same period, plus others from different finds. One piece in particular which stands out is an extremely well executed flat woven image of a face. This was also discovered at the Lop Sanpra site, east of Hotan. In both its tapestry technique as well as the strikingly realistic drawing and execution of the human face it resembles some of the finer European tapestries from the 17th and 18th centuries. The material used here was wool and subsequent to its weaving it has been made into what seems to be some sort of bag. Another similar tapestry woven piece is equally stunning. It depicts a centaur surrounded by flowers on a black field. Both of these tapestry bags have been attributed to the same period as the carpet while both of them show a striking resemblance to Roman art and textiles of the same period. Around the first and second centuries A.D. this trading route connecting East and West was controlled and managed by the two major empires of the day, the Roman Empire and the Han Empire of China. East-West trading was mediated at this time by the Kushans, an Indo-European speaking people probably related to the Sakas whose empire was situated at first to the west of the Hotan region in Bactria (now Soviet Uzbekistan and Afghanistan) and later with its capital near what is now Peshawar in northwest Pakistan. The Western influence on the art of the Kushans, primarily Buddhist art, is well known. It seems likely now that this influence also extended further east along the Silk Road to the Hotan area to a people who were also Indo-European and probably Buddhist.

From a site called Zughanluk at Cherchen, also on the southern Silk Road, 700 kilometers east of Hotan, are a number of very interesting flatwoven pieces, including plain flatweave, slit tapestry weave (weft face), and weft float brocade. These are dated to an age of one thousand years, are in an excellent state of preservation, and in some cases retain very strong color. These later pieces are in a new exhibit at the museum just opened within the last year which also contains several mummies and a considerable amount of woven clothing, mostly from wool, which dates for the most part from between one thousand and two thousand years old.

It is my hope that this brief article may inspire further research into these recently discovered textiles of East Turkestan, so that they may take their deserved place in the literature concerning the history of world textiles.

The author, Chris Walter (facing)