Posted by R. John Howe on 11-26-2002 07:05 AM:


Daniel, Guido -

You say toward the beginning of your initial salon essay that a Hali, issue 70, Levi article on related Kurdish rugs produced some controversy.

The controversy you refer to, I presume, is the "proto-Kurdish" debate resulting from Levi's use of this term to indicate that many Kurdish designs seemed to him to be versions of earlier Persian ones. This led many Kurdish advocates to protest, saying that many Kurdish designs are rooted firmly within the Kurdish tradition and are not borrowings from others. They saw the use of the term "proto-Kurdish" as a denigration of the creativity exhibited in Kurdish weaving.

Now the Levi rug you show, is one that he explicitly called "proto-Kurdish." I take it that you do not intend in this salon to raise this debate again, but rather have some distinctive aims in mind.

Are there some questions and issues arising from this group of rugs you show and analyze that we should consider?


R. John Howe

Posted by Guido Imbimbo on 11-26-2002 11:49 AM:

Proto-kurdish rugs and Tuning-fork group

Dear John,

You raised the difficult a recurrent question regarding the "meaning" and "utility" of the term "proto-kurdish" cloned by Alberto Levi almost 10 years ago.

As you mentioned, many have criticized this term but I note that the word is still with us after so long time and it is used continuously in the rug community and market place. I even observed some tentative to extend the suffix "proto" to other rugs groups. As it is the case of other neologisms, the simple fact that a term has been newly introduced grants vast popularity to it and to its "inventor".

What did exactly Levi mean introducing the term "proto-kurdish" rug?

To my understanding his definition is "semantically" not completely clear. While infact it defines this group as "a family of relatively early Kurdish rugs" (p. 87 of the article) than has been greatly influenced by the so-called 'vase' carpet group than he fails to clarify if the term means a newly originated group of carpets.
Actually, in the first part of the article he seems to contradict the term "proto-Kurdish" because in recalling at least other two much earlier carpets (the Burrell floral lattice garden carpet and a Lahore version carpet) he attributes them to Kurdish manufacture.

Michael Wendorf has dedicated a Salon (and one of the next ICOC lecture) to analyze the possibility that the Kurdish weaving tradition has very precise and old roots, may be even ancient origins.

I believe therefore that it is difficult to accept that the group of Kurdish rugs introduced by Levi in his article represents - as the term "proto" seems to indicate - the ultimate and earliest version of a newly generated group of Kurdish weavings.

To me the word "Proto-kurdish" rugs means "early" Kurdish rugs. The major contribute of Levi in that article is in identifying an unknown body of relatively early Kurdish rugs that have a strong connection with the previous Safavid and vase carpet tradition. Under this point of view, the term "Proto-kurdish" has served very well to in attracting the attention of the public (and of myself).

John, you have also asked if the are any questions and issues arising from the specific group of rugs we have presented in the Salon.

I would like to point out just three aspects in relation to the Tuning-forks group and, more in general, to the "Proto" (i.e. early) Kurdish body of rugs.

1) In our analysis we have found confirmation that several sub-groups of the "Proto-kurdish" rugs, even characterized by a great variety of designs, are all closed related. For example, the "bud & flower" border found in the Group A of the "Tuning-fork" rugs is precisely used in specimens of other sub-groups of early Kurdish rugs (see Table 2).

2) The number of known rugs in each sub-group of the body of "Proto-kurdish" rugs is relatively low (see Table 4 in the Salon) if compared with other rug productions in other areas (for example Caucasian). My question is how do we explain this low number of rugs?

3) We found also that the early version of the sub-group of rugs analyzed in the Salon has exerted a direct influence in the subsequent production (Group B), with the adoption of designs and motif, even if often simplified and sometimes degenerated. This process is common also the tradition of carpet productions in other regions and it is a sign of a strong identity. This continuity gives strength to the idea that we are in front of a "distinctive" group of rugs.



Posted by Steve Price on 11-26-2002 12:28 PM:

Hi People,

The term "Proto-Kurdish" is odd, so poorly defined as to make it of little value (in my opinion). On the other hand, it is not the only such term to be introduced into the rug lexicon. "Pseudo-Chodor", apparently coined by Azadi and used in Wie Blumen in der Wuste, baffles me even more. At least the "Proto-" prefix suggests antiquity. The "Pseudo-" prefix doesn't suggest anything that makes sense to me within the context of tribal attribution.


Steve Price

Posted by R. John Howe on 11-26-2002 09:18 PM:

Guido, Steve -

Actually, I tried explicitly NOT to raise the debate about the term "proto-Kurdish" and assumed that Daniel and Guido were not interested in that.

Apparently, this term is now such a lightning rod that one cannot say it without triggering some aspect of this debate.

Guido, with regard to your points 1 and 3, although the rug literature is literally full of design analysis, I always wonder when it is offered how much it tells us since designs can travel with such ease. I have come to believe that analysis of designs is simply enjoyable to some in the rug world regardless of what it demonstrates. But I don't think you can lean on it much for establishing groups. Design similarities, it seems to me, may often be interesting but irrelevant. Mumford, writing in 1900 said about Kurdish weavings: "The general effect and finish must be relied upon to distinguish them, as their patterns are too widely used in other rugs, both Persian and Turkish, to be at all characteristic."

I can only offer a conjecture about your second point: why so few published rugs in this group. Could it be that Kurdish rugs did not draw much interest until fairly recently?

For example, Cecil Edwards does not have the term "Kurdish" as a direct reference in his index. He does discuss Kurdish rugs under three categories: Senneh, Bijar and tribal rugs. At one point he says that "Kurdish weavers---normally produce coarse and clumsy tribal rugs." (This is not to say that the early rug literature did not sometimes have positive things to say about Kurdish weaving. It did, and Eagleton reviews some of it at the beginning of his book and seems to say that, at least in taxanomic terms, the old "chestnuts" had a kind of "handle" on Kurdish weaving.) Still, there was a taste of the prejorative in such references as "Kurd Bijar" or "Senneh Kurd" that can still be encountered in the market.

So to a lesser extent, perhaps, but somewhat like Balouch weaving, we may have relatively few published examples of particular Kurdish sub-groups because there was not for some time much interest in Kurdish weaving beyond Bijar and Senneh.


R. John Howe