THE QARAI RUGS OF TURBAT-I-HAIDARI part3
Baluch rug designs are derived from a disparate variety of sources - trees, leaves or cockscombs, tiles, Baluchistan flatweaves, mosques, Turkic sources including Seljuk, and Turkoman weavings. Rarely are classical carpets a source of inspiration, though two other well known exceptions exist.
One of these, found primarily in the vicinity of Farah in Afghanistan, is based on l7th and l8th century Caucasian carpets. This influence is directly attributable to the transfer of several Lesghi clans to this district by Nadir Shah in the second quarter of the 18th century.34 The second case, designs derived from 17th century vase carpets, runs fairly parallel with our hypothesis in that the innovators are a group of tribes relatively indigenous to Khurasan. This design is found primarily on Arab rugs from the Birjand-Qain-Firdaus triangle of south and west Khurasan. The Arab tribes of this district predate the arrival of most Baluch tribes by a millenium, having arrived with the Arab conquest. Like Turbat-i-Haidari in the 18th century, the Qain district thrived during the Safavid dynasty, peaking in the 17th century, thus providing an economic and cultural environment that would have allowed access to court carpets of this period. Under these circumstances only the Amir and possibly a few prominent khans would have the wherewithal to possess such magnificent carpets, but access to these could easily be gained by their subjects thorugh visits to the divan khaneh (hall of audience) which is open to the public. The proximity of Kerman in itself is another possible explanation for the migration of this design to the Arab tribes of the Qain district.
A similar scenario could have easily existed in the Turbat region in the 18th century when this district achieved economic and political importance under the leadership of Ishak Khan. Ishak's court has been historically established as being urbane enough to desire such carpets and wealthy enough to have them. Eighteenth century Kurdistan garden carpets were the current state of the art just prior to Ishak's ascent and could be considered logical luxuries for a ruler of his stature.
The source of the Herat carpet influence is clouded by doubts as to their origin. If indeed these carpets were made in Herat as early scholars claim (this is disputed by modern authorities), or in east Khurasan as Cecil Edwards35 speculates, then geographic proximity would have played an important role in explaining the source of this design in Qarai rugs. If India was the origin of Herat carpets, one cannot preclude the possibility of one or more reaching Turbat-i- Haidari during Ishak Khan's reign since he promoted a lively transit trade with the subcontinent through Hindu traders, some of whom maintained permanent residences in this town under Ishak's protection. If Herat carpets are of Isfahan origin, then the reasons36 for their widespread imitation in Qarai iconography are too remote to be considered. The structure of this particular group of "Baluch" rugs reveals a notable departure from mainstream Baluch weaving in that they are usually constructed with an offset or depressed warp. This is a deviation from true Baluch structure, a method of weaving based on flatweaves with all the warps set on one level. Another significant structural element found in these rugs is the selvedge, which is almost invariably composed of four cords wrapped in dark brown goat hair. The flatweaves of Baluchistan and the Baluch pile rugs of Afghanistan are generally made with two cords and one cord selvedges. Occasionally, pile rugs from Afghanistan do have four cords but this characteristic is not prevalent in any one group and very well may be a Khurasan influence.
The following is the typical construction of a Qarai rug:
Warp: Z2S ivory wool, depressed or partially depressed.
Weft: Natural camel hair, brown or olive dyed camel hair or wool, two shoots.
Knot: Asymmetrical, open to the left; horizontal, 7; vertical, 8; 56 knots per square inch. The range of the knot count is surprisingly narrow. The horizontal count ranges from 7 to 10 with 8 vertical knots occurring in 50% of the rugs examined.
Sides: Invariably four cords of 2Z(Z2S) ivory wool secured by the wefts and overwrapped in a continuous figure 8 manner with dark brown goat hair.
Ends: Two narrow bands of weft float on either side of a wide band of multicolored weft-faced plain weave chevrons or diagonals resolving at each color change into stepped diagonal weft substitution, all on a weft-faced plain tapestry ground. Another type that is frequently found is a simple kilim of narrow colored bands of weft-faced plain tapestry.
Colors: DARK BLUE, red, deep rich aubergine, maroon brown, and ivory are the most frequently used colors. In older rugs a light powder blue may be found as well as a pale buff or light lemon yellow that is light sensitive. This yellow combined with blue produces a lovely primary green that turns to teal on the exposed pile side.
Materials: Wool, goat hair, and camel hair. Silk is occasionally found in the pile but rarely in examples made between 1864 and 1900.
Illustration 1. Qarai rug, 2'4"x5',5" Courtesy of Scott and Ann Murphy, Photo by Dennis Anderson
After establishing a structural criteria for this group of "classical" designed rugs, a variety of rugs with lattice designs emerged as well as a few rugs with Turkoman designs (Illustration 3) and archtypal Turkic motifs. The following is a list of design observations:
Guard stripes: In almost all cases the guard borders consist of short diagonal bars in two alternating colors.
Minor borders: In most cases a distinctive reciprocal running latch hook in ivory and dark brown.
Main borders: Several types occur frequently; the three most popular types are: 1. a sophisticated border featuring a maze of foliage based on Kurdistan garden carpets (Illustration I); 2. a Turkoman line with multiple latch hook appendages (Illustration 3); 3. multicolored diagonal bands resembling some Afshar borders. This may be an elaborated form of the guard stripes or adopted from Afshars who inhabited Khurasan prior to the 20th century.
Field: Surprisingly, lattice patterns are the predominate field design found in rugs with this structural composition. The lattice forms are fairly evenly divided between a diagonal four-sided diamond grid and a hexagonal lattice pattern. Generally speaking, the earlier examples have Turkic archetypal devices and Turkoman motifs filling the spaces formed by the lattice, while later rugs seem to prefer the primary flowers favored by the Mina Khani patterned rugs from this tribe.
Illustration 1. Qarai rug, Photo by Dennis Anderson
The Mina Khani design is the second most favored field decoration in this structural group. The treatment of this design is done in classic Turkoman format. A large red flower, derived from Turkoman chemche guls, serves as the primary aspect (or "gul"), a white five-petaled flower assumes the secondary position, while tertiary devices vary from rug to rug though they are seldom missing (Illustration 4).
The number of petals in the secondary white flowers vary. Generally, the oldest examples tend to have five petals while six and bifurcated four petal flowers are favored in most later rugs.
Other odd designs occur in this structural group but only rarely. Of these Turkoman patterns predominate and usually are archaic in drawing and iconography. With the exception of the chevron end finishings, Baluch designs are rare.
An idiosyncracy of this group of weavings is a propensity for integrating square shapes into the design. This tendency manifests itself in a number of ways, such as in the centers of the flowers found in the Mina Khani or in the reciprocal space between the hexagons in the lattice pattern. Interestingly, the square was the tamga (brand or tribal symbol) of the Qarai tribe in the 13th century.37
The evidence that this group of weavings are not the product of Baluch tribes is fairly conclusive. The four-cord selvedge, depressed warp set, and weft substitution chevron kilim ends deviate from the structure of Baluch weaving found in Baluchistan38 and Afghanistan. Weavers frequently change their designs but seldom do they alter the way they were taught to weave.
From a design perspective this group of weavings deviates from traditional Baluch work in one important aspect. The reliance on classical iconography found in the Mina Khani group is a foreign influence that is not only atypical of Baluch aesthetics but is difficult to explain in terms of origin. Baluch tribes are generally too impoverished to possess such luxurious furnishings, seldom are Baluch tribesmen invited to court or any other polite place that these carpets exist, and they would have no use for such a large carpet should they happen to steal one.
Proving conclusively that these rugs are from the Qarai tribe is less tenable, but indications do exist. Historically, the prevalent tribe of the Turbat-i-Haidari district were the Qarai up until the turn of the century. Lady Sheil,39 in the mid 19th century, mentions a 5:2 ratio of Qarai to Baluch. Qajar politics from the 18th century onward were devoted to breaking the political and economic strength of the Qarais, and dwindling population statistics demonstrate the success of this campaign. By the turn of this century the Qarai were almost indistinct from their Baluch neighbors. A study of the weavings of this region parallels the history. Later examples of Mina Khani and lattice design rugs generally have less or no warp depression, two cord selvedges are more frequent, and weft float and soumak work in the kilim ends become more prevalent. Early examples present a cohesive structural and design group that breaks down as the century progresses much the same as the tribe.
A bit more tenuous, but worthy of notice, is the archetypal Turkic treatment of design in these rugs. The rendition of the major flower in the Mina Khani design is based on an archaic formula of laying a vertical cross over a diagonal cross. This is the format for the compartmentalized design seen in the Pazyryk carpet and perpetuated by the chemche gul in Turkoman rugs. The 2-1-2 arrangment of these flowers, the propensity to work squares into the designs, and the archetypal Turkic devices used to fill the lattice designed examples are characteristics that suggest a design pool originating in the central Asian steppes, not Baluchistan.
Illustration 1. Qarai rug, 3'2"x6'3", This rug appeared in an ad for Ashik, Albuquerque, NM
by Michael Craycraft
1 Early observers (12th-14th centuries) tend to spell this tribal name Kerait while those of the 19th century favor Karai. Kerait is probably the more correct, while Karai is the latest usage. (Ed. We have changed the author's spelling to Qarai, our reference source being Cambridge History of Iran.)
2 Rhys, E., The Travels of Marco Polo, New York, 1908, p. 142.
3 Czaplicka, M.A., The Turks of Central Asia, Oxford, 1918,p. 57.
4 Conolly, A Journey to the North of India, Vol. I, London, 1834, p.290.
5 An in depth account of Togril and Genghis may be found in The Secret History of the Mongol Dynasty by Yuan-Chao-Pi-Shi. A good summary of this is contained in Barthold's Turkestan Down to the Mongol invasion.
6 Rhys, op. cit., p. 116-7.
7 Czaplicka, op. cit., p. 38.
8 Malcolm, J., The History of Persia, London, 1811, p. 147, Vol.2.
9 Wolff, J. D., Narrative of a Mission to Bokhara, New York, 1845, p. 134-5.
10 Yate, C.E., Khurasan and Sistan, Edinburgh, 1900, p. 54.
11 Malcolm, op. cit., p. 147.
13 Yate, ibid.
14 Malcolm, op. cit., p. 147.
15 Sheil, Lady, Life and Manners in Persia, London, 1856, p. 401.
16 Yate, op. cit., p. 53.
17 Skrine, F.R., The Heart of Asia, London, 1899, p. 153, footnote.
18 Prestor is an abbreviated form of "presbyter" which means "priest" in old Latin.
19 Malcolm, op. cit., p. 147-8, Vol. 2.
20 Ibid., p. 194.
21 Frazer, J.B., Journey into Khorasan, London, 1825, p. 25, appendix B.
22 Omit 23 Brill, E.S., First Encyclopedia of lslam, Vol.8, Holland 1927 p. 1219.
24 Lal, M. Travel in Punjab, Afghanistan, and Turkistan to Balkh, Bokhara, and Herat, London, 1846, p. 120.
25 Smith, E., Eastern Persia, London, 1876, p. 352.
26 Bellew, H. W., From the Indus to the Tigris, London 1874 p. 349.
27 Doyle, J.A., The Cambridge History of Islam, Vol. 5, Cambridge, 1968, p. 310.
28 Conolly, op. cit., p. 290.
29 Lal, op. cit., p. 119.
30 Bellew, op. cit., p. 349 and Smith, op. cit., p. 353.
31 Curzon, G. N., Persia and the Persian Question, Vol. 2, New York, 1892, p. 509.
32 Craycraft, M., Belouch Prayer Rugs, Point Reyes Station, 1982, p.75.
33 Bogolubov, Carpets of Central Asia, England, 1973, Plate 34.
34 Elphinstone, M., The Kingdom of Caubel, Vol.1, Oxford, 1972, p.420.
35 Edwards, A.C., The Persian Carpet, London, 1953, p. 164.
36 Malcolm, op. cit., p. 149.
37 Czaplicka, op. cit., p. 43.
38 Konieczny, M. G., Textiles of Baluchistan, London, 1979.
39 Sheil, op. cit., p. 401. 5,000 Qarai families to 2,000 Baluch.