Some Salt Bags
from Kerman Province

Like nomads elsewhere in the Zagros and throughout Persia, the tribes of Kerman Province have woven containers of a peculiar shape known as namakdan in Persian to transport and store salt and a variety of loose materials like seeds, nuts, and so on. The salt bags of the tribes of Kerman represent, in my mind, the most attractive group of such weavings made in Persia or anywhere.

In an article entitled "Salt Bags from Iran" (Hali, Vol. II, No. 3, Autumn 1979, pp. 198-205), I published three examples from the Kerman region and discussed the salt bags of that area as a whole. More recently, Parviz Tanavoli has expanded our knowledge of Persian salt bags in general, and ones from Kerman in particular, in his excellent publication, Bread and Salt (Tehran, 1991). The numerous examples illustrated in this book are identified with a degree of precision seldom seen in the literature on Persian tribal textiles. Tanavoli's initial research on the weavers and tribal weavings of Kerman Province is published in two Hali articles, "The Afshars, Part I: A Tribal History" (Issue 37, Jan/Feb 1988, pp. 23-29) and "The Afshars, Part II: Tribal Weavings from Kerman" (Issue 57, June 1991, pp. 96-105). Another useful article that sheds light on this group of salt bags is P.R.J. Ford's "Flatweaves of Kerman Province" (Oriental Rug Review, Vol. XII, No. 2, December/January 1992, pp. 18-24). Basing his remarks on information provided by a family of rug merchants in the Tehran bazaar with agents in the provinces, Ford states: "I cannot attempt to distinguish, as Parviz [Tanavoli] does, between the genuine products of the Turkish-speaking Afshars and those of the Persian-speaking tribeswomen and settled villages who clearly outnumber them" (p. 18).

The considerable size and diversity of the tribal population long inhabiting Kerman province has doubtless been responsible for the diversity of the salt bag types produced there. According to Tanavoli, "The fact that the Afshars of Kerman have the greatest share in the production of rug weaving in that region is not in doubt, but it would be a mistake to attribute all the Kerman tribal weavings to the Afshars as there are a number of other tribes besides the Afshars who weave in Kerman... The study and recognition of Afshar weavings is extremely difficult and complicated. For many years the tribes have mostly settled down and intermingled. This makes the distinguishing of their individual weavings very hard" (Bread and Salt, p. 17-18).

At this late date, how can one come to know exactly who was responsible for what in the realm of tribal textiles in Persia? The surest way is, of course, firsthand observation of what is still being used and woven by groups whose traditional lifestyle or identity remains intact, i.e. through systematic fieldwork. Unfortunately, the obstacles facing foreign researchers in Persia are such that few have been, or will be, capable of an undertaking of this sort. The logical candidates for this task are the Persians themselves, especially female researchers, who would have an entree to the weavers denied males in an Islamic society. It is the material thus identified that becomes our true anchor pieces, enabling us through comparisons based on physical and aesthetic characteristics to make confident attributions of similar objects found out of their original cultural context.

Lacking this underpinning, we are usually forced to rely on information originating in the marketplace in Persia, ideally from as close to the makers and users as possible. This might come from the local merchants who buy directly from tribesmen passing through the bazaar, or from agents or pickers who seek out the tribes. Such information is transmitted through the chain of local, provincial, and national (i.e. Tehran) markets. The knowledge found in provincial markets tends to be much more localized than in a place like Tehran, the ultimate destination of material from all over the country. It also tends to be more specific.

The accuracy of this type of information, I suspect, is dependent in large part upon the level of intellectual curiosity possessed by different dealers. Those who have it to a high degree take a real interest in questions of provenance and use and are likely to query their sources and transmit what they have learned honestly and accurately. Those who do not possess this quality are probably the principal source of the stories and misinformation that make the marketplace a far from perfect purveyor of knowledge. Whatever its strengths and weaknesses, it is the wisdom of all those who have participated in the rug business in Persia that provides the greatest part of what we know about the provenance of Persian rugs and textiles. This includes the work of A. Cecil Edwards, a British rug dealer whose long experience in Persia is embodied in the well respected volume, The Persian Carpet (London, 1953).

As mentioned above, Ford's source of information on the weavings of Kerman Province is the Samadi merchant family of the Tehran bazaar. Tanavoli's, while not cited in Bread and Salt, is apparently the carpet dealers in the bazaars of Tehran, Sirjan, and Shahr-e Babak, whose help in identifying various pieces is acknowledged in Part II of his previously noted Afshar article (p. 140). The basis for the attributions in my 1979 article is what I learned in the marketplace in Tehran in the 1970s. To help expand the picture of weavings from Kerman Province provided by the above publications and to apply, where possible, any insights furnished by Tanavoli's and Ford's work, I have selected some of the salt bags from the Kerman area that once formed part of the salt bag collection my wife Suzan and I built in Tehran during the early 1970s.

My favorites among the salt bags of Kerman Province are Nos. 1 and 2, finely woven in weft-faced plain weave patterned by weft substitution. They obviously come from the same group of weavers, as evidenced by their identical elaborate black and white edge finish and unusual closure panel, including the tassels. In both, the body and neck are treated as an integral unit of design. There are no direct parallels to either salt bag in Tanavoli's or Ford's work.

Star and octagon motifs identical to those of Nos. 1 and 2 are seen in several rows on a rug in extra-weft wrapping (sumak) from the Kerman area published by Jenny Housego (Tribal Rugs, pl. 117). The Housego rug may well be closely related to one placed by Ford (Fig. 7) in the Sirjan (Sa'idabad) area where, he says, such products were made by the Osturi and Buchaqchi tribes as well as in the town of Sirjan itself. Tanavoli cites the Buchaqchi Ashraflu clan as belonging to the Afshar tribe. Other writers seem to consider the Buchaqchis as separate from the Afshars. I have not seen the Osturis mentioned elsewhere. According to Ford, "There is a village of Ostur not far from Sirjan and it could be that many Osturi pieces come from that area" (p. 20). Tanavoli says the largest Afshar community in Kerman Province is found in the Sirjan area, and he ascribes many of the examples in his various publications to them.

The possible relationship of Nos. 1 and 2 to the Housego sumak rug, which in turn can probably be located geographically by its relationship to Ford's Figure 7, might suggest that they are products of the Sirjan area. If so, who wove them? Buchaqchis? Osturis? Afshars? Some other group?

Salt bags of the same structure as Nos. 1 and 2 were evidently woven in several different parts of Kerman Province by different groups. Tanavoli points out a Baluchi product of the region (Bread and Salt, pl. 97) and two from the Afshars of Baft or Bardsir (pls. 101 and 102). A salt bag published by Ford (Fig. 13) together with a spoon/spindle bag (Fig. 12) is attributed to the Jamal Barez (better termed Jabal-e Barez), a mountain range in the southeastern part of Kerman Province. These two bags exhibit strong parallels to salt bag No. 3, of a somewhat heavier and courser construction than Nos. 1 and 2. Ford's figures 12 and 13 and salt bag No. 3 seem to be of a group with the sofreh he illustrations as Figure 11. However, Tanavoli attributes an identical sofreh to the Afshars of Sirjan, which lies some distance away in the western part of Kerman Province. This difference of opinion might be explained away if one tribe could be linked to both areas. However, I have not been able to do this. Is Ford right on the bags and wrong on the sofreh? Right on both? Wrong on both? The difficulty of acquiring correct information in Persia and the dangers of drawing conclusions from it are patent.

More straightforward is the argument for No. 4, woven in extra-weft wrapping with a row of weft-faced plain weave patterned by weft substitution at top and bottom. a piece similar in structure, composition, and use of undyed cotton in the stars and in the border is seen in Tanavoli's Pl. 99, attributed to the Afshars of West Kerman in the caption and to the "Afshars from around Sirjan" in the text (p. 51). The shiraki cover of Ford's Figure 4 contains a border of stars which are strongly reminiscent of those in these two salt bags. Ford attributes this cover to the village of Pariz, which is not too far north of Sirjan. The border of No. 4, a repeating S motif in a hexagon, is identical to that of Tanavoli's Pl. 113, which he calls Afshar from Bardsir, a town in the general area of Sirjan and Pariz.

No. 5, with its plain weave ground patterned by small rosettes and its surrounding sumak border may also be related to the shiraki cover of Ford's Figure 4.

Two bags with similar composition, coloring, and use of undyed cotton, Nos. 6 and 7, are strongly evocative of the sumak rug of Ford's Figure 7, discussed above, which he attributes to the Sirjan area and calls Osturi in the caption.

No. 8 features a rendition of the botteh that appears commonly in Afshar bags and rugs, like the one Tanavoli published in Hali 37, p. 28. The use of the chevron motif in reciprocal weft weave seen in the closure panel and in stripes above and below it and at the bottom of the bag is a very common feature in the bags of the Kerman area. Also very common in pile and flat-woven textiles of the region are the diagonal stripes of the border.

Large stars in a grid of squares appear in different flatweaves of the Kerman area, like salt bag No. 9 and a horse cover published by Housego in Tribal Rugs (Pl. 118). The back-to-back bird or animal motif seen in Tanavoli's Bread and Salt Pl. 106 from the Afshars of Sirjan appears in No. 9 as well.

The basic layout of No. 10 and Tanavoli's Shahr-e Babak Afshar salt bag (Pl. 112 in Bread and Salt) is identical, as is the principal motif in the body of both. All the elements of design seen here, including the saw-toothed outer guard, commonly occur in other salt bags and weavings of the region. The colors of No. 10 are noteworthy for their bright, clear quality which comes through despite the substantial wear it has seen. Is No. 10 older than most? Is it from the Afshars in the western part of Kerman Province?

Of the salt bags from Kerman Province with which I am familiar, none more closely resembles the work of the Qashqa'is than the densely filled, very finely woven bag with a large dominating lozenge medallion, No. 11. The features that put it in the Kerman region are its extra-weft wrapping instead of pile structure typical of Qashqa'i bags, its coloration, and other details like the black and white overcasting of the sides, multi-colored wrapping of the extra warps on either side of the neck, and its plain red back with a few narrow stripes. I have not seen any other bag of this type.

Boldly patterning the body of No. 12 is a large lozenge made up of a repeating diamond motif filled with a cruciform figure. A single diamond fills the neck, which is surrounded by the same border as the body. Once again, undyed cotton is used generously and effectively. The saw-toothed guard seen in salt bags No. 6 and No. 10 and Pls. 99 and 112 of Bread and Salt show up here as well. Is this another example from the western region of Kerman Province?

Two pile salt bags, Nos. 13 and 14, neither with a design that is particularly common in the pile weavings of the area, are included. The distinctive Kerman coloration that ranges from peach to orange comes through quite clearly in both examples. The back of the neck of No. 13 is virtually identical to the front except for a change in the ground color from blue to red.