Posted by R. John Howe on 06-25-2006 10:05 AM:

David Zahirpour: Oriental Rugs from S.W. Iran

David Zahirpour: Oriental Rugs from S.W. Iran
A Rug and Textile Appreciation Morning
The Textile Museum
June 24, 2006

Reported by R. John Howe

On this date David Zahirpour presented a “rug morning” program at the TM on southwest Iranian rugs.

David preparing for his session.

David was introduced by the TM’s new Curator of Education, Linda Powell.

Ms. Powell told me that she has come from a position in the Hirshhorn Museum. I told her we would introduce here a bit to the rug and textile community.

David provided a handout which indicated that he framed this group of “rugs” (he included a number of bags and one textile) a little differently than do many who treat this area.

In it he began by writing “There are six (6) groups of rug weavers in S.W. Iran. Five of them are mainly tribal and one is workshop.

David began his session by ticking off the groups in his handout.

• Weavings from the town of Shiraz (his “workshop" category, although he acknowledged at points that "workshop" modes have crept into some S.W. "tribal" weaving.)
• Gashgai tribe (spellings vary throughout)
• Kaskoli (there is a large group and a small one and he counts them separately since they seem to have no interaction)
• Khamseh tribe
• Gabeh (David’s handout seems to treat this as a tribal group but perhaps that is a typo and he intended only “tribal,” something to which most would agree)

I noticed immediately, that David’s listing did not include some weaving groups often indicated by others as among southwest Persian weavers. In particular, he does not include Afsharis, Lurs and Bahktiaris.

Here, for comparison, is Opie’s listing of weaving groups in his first volume:

• Quashqa’i confederacy
• Khamseh confederacy
• Lurs
• Bakhtiyaris
• Afshars

Why David’s particular listing? David was born in Iran and was there for at least part of his youth. He talked, in this session, of visiting (apparently a hard thing to do given their reclusive-ness and general hostility to outsiders) a “small Kashkuli” village. So it is likely that his geographic sensibilities about his native country are more honed than are those of many of us.

So I took a look at Opie’s map in which he indicates where these various groups were. Here is a scan of most of it.

Now this is a partial scan of a map that covers two pages and I cannot get it all "in." Note that the region on the far left is a more northern group of Lurs and that the unlabelled area to the right of the more southern Lur territory is where the Qashqa'i reside.

As you can see, David has defined “S.W. Iran” tightly.

He said to me that he excluded the Afsharis because they are too far east, really in the Kerman area.

Similarly, he excludes the Bakhtiaris because they are located well north of his most northern recognized group, the Quashqa’is. (Bakhtiaris are mostly marketed through Isfahan to which most of them are quite close.)

While David might defensibly exclude the northern group of Lurs, (who are “west” but hardly “south,” being on an approximate line with Arak and at their furthest extent with Hammadan), it is harder to see why he also excludes the southern Lurs who are parallel on an east-west line with his included Qashqa’is. He seemed to say after that he felt the Lurs were an ill-defined group and, in truth, David’s treatment of attribution of all the rugs he discussed in this rug morning was conservative. The main distinctions he seemed comfortable making were among Shiraz weaving, Qashqa’i weaving, that of the Khamseh Confederacy and of the two “Kashkoli” groups.

Most of the literature on southwest Persian weaving treats “Kashkoli” weaving as a species of Qashqa’i weaving. Why raise it to the level of “Qashqa’i” or “Khamseh” as a separate group? I am not sure. David, though, is also a dealer and subject to the pressures that market usages in some areas. He made a point of saying that “Kashkoli” weaving is the highest quality weaving in S.W. Iran. He said that he has seen very few old “Kashkoli” pieces in his career and only one that had silk in it. It may be that David sees “Kashkoli” weaving as distinctive from Qashqa’i weaving because of its recognized quality. And in his defense, the “Kashkoli” usage is not like “Serapi” (that is ONLY an indicator of high quality) because there are two actual “Kashkoli” tribes who weave.

Moreover, the weavings of the “Kashkoli” seem to be to an extent distinguishable from other “Qashqa’i weavings on technical grounds. “Kashkoli” weaving is said to have “deeply depressed alternate warps, red wefts and a fine weave.” The “Qashqa’i have Turkish origins and might be suspected most usually to use symmetric knotting. And some of the coarser types of Qashqa’i rugs do. The “Kashkoli” pieces seem invariably to have asymmetric knots, but it is not entirely clear that other finer “Qashqa’i pieces do not. My reading makes me think that there is some market influence in some of these distinctions.

Nevertheless, it is useful occasionally to encounter something like David’s groupings that makes one reconsider the accuracy of something we might think we “know.”

David began his discussion of pieces by acknowledging that he had not brought an example of a “Shiraz” pile rug, but in response to my question about the nature of this category, did provide some indicators for identifying such weavings.

He said that most “Shiraz” weaving exhibits three to five wefts, a noticeably loose weave, most often the ground color is red, and that medallion designs are frequent. He acknowledged that there are some Qashqa’i weavings that are marketed through Shiraz because of the proximity of some of those in this tribe to this city but that the “Shiraz-type” weaving was pretty distinguishable from them. In my admittedly limited experience, few “Qashqa'i pieces could accurately be described as “loosely woven.”

David had brought one Shiraz piece, a new flatweave, shown below.

It has good colors and marked graphics.

The most markedly Qashqa’i piece in the room had been brought by someone in the audience.

The owner of this piece said that its strongest dimension is the very high quality of its wool. It was complete although not sewn up the sides. Here is a look at it opened up.

David had a number of bag faces on the board and treated them generally as likely Qashqa’i.

(I asked him after about one of them with botehs in its field, saying that most of us would be tempted to call it Afshar. He looked at its back (it has red wefts) and agreed that Afshar is probably right.)

Here is one small, but complete, flatwoven Qashqa’i bag among this collection.

Its design is simple, even humble but not without some charm.

And its back is not undecorated.

Off to one side there was a bag face (also described as likely Qashqa’i) that attracted my eye.

Although rather crudely drawn and fragmentary, I was attracted to it for some reason, perhaps the treatment of its borders.

Another Qashqa’i example was a small kilim (draped but never opened up fully).

A little closer look.

David drew attention to its typical ends and borders.

David had brought one textile, what most of us would call a jajim, although he did not use this term or refer to its function.

I cannot remember whether he explicitly said that it is made in narrower panels and then sewn together, as many such pieces are, but he said he had owned it for many years and rhapsodized over the fact that it is made entirely of very fine wool.

The closer image below reveals its range of colors.

David had one long, narrow tent band.

A closer look.

Closer yet.

I think he attributed this piece to the Qashqa’I as well. He said that once a dealer had come into the store with a great many bands and horse trappings and the like and that this is the piece he bought that day.

David also had a contemporary example of Kashkoli weaving, the quality of which he had praised as he listed his categories of S.W. Iranian rugs.

He said that Kashkoli rugs show some curvilinear elements, as in the small medallion in this one.

But are composed of mostly rectilinear drawing. (This is largely true, but one thinks of at least one important exception: the Kashkoli produced entirely successful versions of the “mille fleur” design that are markedly curvilinear and that goes back to Mogul versions of the 17th and 18th centuries.)

David said that herati fields are frequent in Kashkoli rugs and this piece had one.

Here is a border section a little closer.

And a corner.

He rated the wool and the weaving of this piece as superior. David says that this one of only two Kashkoli rugs he has been able to buy recently.

David offered one Khamseh example.

He asked if we knew the source of this design with its connected, anchored medallions. He said that it is based on the Persian love of water and on the pools that often mark Iranian gardens.

And, of course, it is hard to consider a Khamseh rug without noticing “chickens” and sometimes other seeming animal and even human forms that populate the filler devices with which they overflow.

At the end David pass around a border fragment of an older Qashqa’i rug.

Our thanks to David for making us reconsider some things we thought we knew about S.W. Iranian rugs.

And a special thanks to Filiberto for some extraordinarily swift work getting these images up on the web, so that we are able to provide you with a version of this TM rug morning less than 24 hours after it began.

Comments invited.


R. John Howe

Posted by Filiberto Boncompagni on 06-25-2006 12:03 PM:

Actually John, the work was entirely on you!


Posted by Richard Larkin on 06-25-2006 01:46 PM:


Thanks for bringing us this and other interesting reports. You are doing a great service to the afficionados out in the hinterlands.

My initial interest in oriental rugs came about in the mid sixties when I was living in Riyadh. The city hadn't had much contact with western visitors and the rug market there was isolated from the market elsewhere. Most of it took place in the old Dira Suq (long since gone, I'm told, sad to tell), and the dealers there, who catered mostly to a strictly local clientele, didn't quite understand why we were interested in old rugs. It's hard to figure because the new merchandise in the suq was mostly atrocious, but that was the attitude.

A staple run of rugs in the suq was Southwest Persian stuff, old and new. We westerners barely knew what we were doing at the time, learning as we went. We had the little Preben Leibetrau book in hand, turning pages and trying to find rugs we would come upon.

I bring all this up to mention that the Saudi sealers in the suq had their own nomenclature for the older rugs. I never heard the term "Khamseh" (for a is the Arabic word for "five," of course). The coarser rugs, of the sort that would tend to have domestic fowl well represented in the design, were universally called "Arabi." This no doubt reflected the often read comment that the Khamseh tribes include people whose ethnic origins are Arab. Ordinary quality Southwest Persian goods got called "Shiraz," although the dealers were well aware that the Araby and other rugs also came from Shiraz in the general sense. Better qualities merited the name "Qashqai." Special quality or age would bring up the name of a sub group of the Qashqai. I recall the terms "Bulli" and "Turki" being used. I remember noting that the "Turki" would in fact be symmetrically knotted. I don't know whether the dealers took that into acccount in the appelation. They didn't seem to be interested in structural questions.

The Saudis themselves were very conscious of tribal affiliation, and I took their attributions to have been made in that context. They assigned these labels very confidently, and in answer to direct questions about who made the rugs. At the same time, I realized, especially later, when I saw catalogs like The Qashqai of Iran (Whitworth Art Gallery, 1976), that the nomenclature being used by the Saudi dealers was substandially insufficient.

Extrapolating from that experience, I have always considered the efforts of dealers and collectors to attribute origins of weavings to tribes, villages, national groups, etc., to be excessively dogmatic. The necessary background information, which must be a huge body of data, is not in hand in sufficient quantity to allow for comprehensive explanations. Furthermore, in my opinion, there is far too much reliance on the principal that weaving style and technique can be matched up uniformly with ethnic tribal identity. Reading Thomas Cole's interview with Jerry Anderson about "Baluchi" (we thought) rugs exemplifies the point. Whatever might be thought about the things said in that interview, it is clear that other attempts to encompass the field of "Baluchi" weaving have been severely lacking in factual underpinnings.

A longwinded way to the point that most nomenclature sets for the attribution of rugs to historically and factually valid weaver contexts are far too simplistic. Although they may be generally accurate in many cases, at best, they are systems of convenience. They can be judged in regard to their usefulness in that way. I imagine Mr. Zahirpour's working vocabulary would be found to be such a system.

I hope that isn't too argumentative. My best point was the first one: Thanks, John, for bringing these things to our attention.

Rich Larkin

Posted by R. John Howe on 06-25-2006 07:00 PM:

Mr. Larkin -

We are glad you like these reports.

They do take a little effort by more than one of us, and it's good to see that they do not always disappear in cyber-space like a stone thrown into a very deep, very dry well (i.e., you don't even get to hear it "splash").

I think, as you say, that it is true that often our efforts to provide names and rubrics that let us understand better the rugs and textiles we collect are prone to error.

And often (and this may be unavoidable, since those we most want to talk to are dead now) the categories imposed are to an extent "ours" not necessarily those of the weaving cultures of interest.

But we have the pieces and it probably is not entirely wrong-headed to analyze them and to try to note and record similarities and differences encountered. But we need to be careful about the sorts of statements that can be made on the basis of this kind of analysis.

And it does seem to be true that there is a kind of "more, results in, less" phenomenon in play. You mention the various treatments of Balouch weaving and it seems to me that many students and collectors of this type have become more and more modest about the statements they tend to make about them.

Similarly, if you compare descriptions of various S.W Persian rugs in Opie's second broader book "Tribal Rugs" with those of his more focused previous book "Tribal Rugs of Southern Iran," I think you will find that he has become visibly more cautious about attribution statements at the "sub-tribe" level, something that the first book seems rather full of.

So I agree (I do not read your post as contentious at all) that we need to be both cautious and tentative about what we think we know. And we need to remain alert to the possibility that an unfamiliar grouping such as the one that David used in his session might have a more legitimate basis than those of us who have "read all the books" might suspect.

I think I have said before that in this context the song by the King in "The King and I" contains great truth.

It's lyrics are in part something like this:

There are times that I'm not sure that I'm convinced of what I absolutely know.

Often times I find confusion in conclusion I concluded long ago.

In my head are many facts that as a student I studied to procure.

In my head are many facts of which I wish I was more certain I was sure.


It could be worse. We could be astronomers and have to deal with the fact that not just our major explanatory theories, but even our paradigms (what a nice thing to find a place to use that word in a non-jargon-ish application) have repeatedly be up-turned and overwhelmed in recent years.

So, serious, interested, modest and tentative seem to be the hallmarks best suited to our pronouncements about the rugs and textiles we collect and in which we are interested.

Thanks for the "splash."


R. John Howe

Posted by Richard Larkin on 06-25-2006 07:10 PM:

By the way, John, I agree with you on that fragmentary bag you showed. I think it's the most interesting piece in the group.

Rich Larkin

Posted by Sue Zimmerman on 06-25-2006 08:27 PM:

Hi John,
I don't think it's worse for astrophysicists. I think the astronomers know what hit them and have adjusted pretty well. Rugdom just still suspects something is not as it seems. How can paradigm shifts be bad if they lead us closer to the truth? Where's the downside? Sue

Posted by Jack Williams on 06-25-2006 11:24 PM:

Much appreciated.

John that was a very nice presentation. We do not get much in that way down here. And I found Richard's comments to be quite interesting background. Stories about the earlier days have a value above just the mention of the carpets themselves. Thanks.

I wonder sometimes about the lingua franca of dealers though. It seems to me they tend to develop references within the network that don't necessarily have a lot to do with the facts on the ground. My brother has refered to this that the Pakistani dealers will confidently call a rug by a local trade name that seems to have little relation to other than a location. Gene Williams has previously noted that Jerry Anderson was pretty skeptical of...for instance...charchango used as a "tribe" rather than a location.

Personally, having an idea of where, from whom, allows me a connection to history and a vanishing lifestyle and that increases my enjoyment of a carpet immensely. Without that, I could almost just as well be looking at abstract art at the local college art class exhibition (nothing wrong with that at all though).

Having a chance to sit in on a presentation via your efforts is really appreciated. Was there some degree of skeptism about the catagorys of rugs presented and the slight lack of clarity about location? or am I reading that into the essay? Thanks again.

Jack Williams

Posted by R. John Howe on 06-26-2006 07:29 AM:

Hi Jack -

You ask "Was there some degree of skepticism about the catagorys of rugs presented and the slight lack of clarity about location? or am I reading that into the essay?"

Well, you can see from the fact that I asked questions of David both during and after the presentation, that at least I thought his categories were unusual. There were, perhaps only a couple of experienced collectors in the audience for this presentation and TM "rug morning" audiences tend to be accepting. So there was no general skepticism visible.

I have sometimes heard sharp questions from the floor in some rug mornings. One I remember occurred during a presentation on contemporary rugs woven with natural dyes and handspun wools. The speaker seemed to suggest that this was a continuation of the historic weaving tradition in this area despite lots of outside influences.

One experienced person standing in the door objected loudly to this characterization saying that this was mere commercial propaganda for this particular contemporary line.

So while sometimes serious questions are raised from a TM rug morning audience, usually most folks in them are preoccupied with hearing about, seeing, and getting their hands on the material presented. So the more general audience posture is an accepting one.

Sue -

I was mostly joking about paradigms shifts, but in truth they can be "painful" for those working in a particular field of reseach. For example, when "oxygen" was "discovered," the reigning theory of burning was the "Phlogiston Theory" (i.e., that there was a substance in things that caused them to burn). That whole generation of scholars had to die off before a new conception of burning was accepted. They apparently had too much invested in the previous conception and couldn't make the shift.

Human beings are still full of human foibles as they work in life's various sectors. Not everything is obvious idealism.

But I was mostly just celebrating a less than jargonish chance to use the word "paradigm."


When I was first in grad school, pretentious language was multiplying in the social sciences under the urge to "scientize" these fields.

I had one very bright professor who poked fun at this sort of thing with a doggeral poem in which he advised such things as "Let your every "thought" be a "conceptualization."

The ending line was likely fully understandable only to folks who could remember a particular popular song from the days of the Depression, but was a thing of teasing, humorous genius.

It was: "Brother, can you paradigm?"


R. John Howe

Posted by Sue Zimmerman on 06-26-2006 08:45 AM:

Hi John,
Actually I don't take paradigm shifts too seriously. They're born they mess up then they die just like everything else. Words don't know. Sue

Posted by Richard Larkin on 06-26-2006 03:11 PM:


I can't argue with a thing you said in the post of June 25. These tentative attribution models, etc., are very convenient for discussion purposes; and certainly, if we qualify every reference to them, they lose their usefulness. There is a tendency among us (me included) to treat them as hard standards that require adherence. The impassioned discussions surrounding Turkoman attributions come to mind when the point under discussion is, for example, the use of some minor decorative motif. The discussion is fine, but one has to wonder how certain a commentator can be in discussing such issues when the truth is, we barely understand the contexts in which the weavings originated. My hero in these things has generally been Murray Eiland.

The truth has to be that there were many thousands of significant facts having to do with political conditions, family or tribal relationships, market factors, individual circumstances, fashions, and untold other circumstances about which we will never learn, that profoundly influenced the nature and production of the weavings that fascinate us today. Not that we should stop searching and thinking. Every now and then, something like the Wright-Wertime studies comes along and provides welcome grist for the mill. As long as we proceed with the kind of caution (and humility?) that you suggest, we should be all right.

The disease (dogmatism) has infected those even at the highest levels, I believe. Do I recall correctly that Erdmann refused to credit the Pazyryk Carpet as a specimen of piled weaving?

Rich Larkin

Posted by Steve Price on 06-26-2006 03:51 PM:

Originally posted by Sue Zimmerman
Hi John,
I don't think it's worse for astrophysicists. I think the astronomers know what hit them and have adjusted pretty well. Rugdom just still suspects something is not as it seems. How can paradigm shifts be bad if they lead us closer to the truth? Where's the downside? Sue

Hi Sue

The short answer to your question is that not every paradigm shift brings us closer to the truth. The downside of accepting every new paradigm without being critical is that it brings us into chaos. We see an awful lot of resistance to critical thought here - people becoming irritated if their fascinating new idea isn't accepted straight away.


Steve Price

Posted by R. John Howe on 06-26-2006 04:08 PM:

Hi Richard -

I had not heard of Erdmann's doubts about whether the Pazyryk Carpet was pile weaving.

An initial suspicion in that regard might be justifiable since things seem to progress from flatwoven structures to knotted pile.

The Pazyryk piece was so much older than anything else we had, when it was first found, that a rug scholar, especially, might be forgiven for predicting that it would almost certainly be flatwoven.

On the other hand, if Erdmann persisted in that view after the piece had been analyzed and not just its knotted character, but even the character of the knots, had been determined, then, that would seem more like an instance of dogmatism or as some say of a "privileged hypothesis," i.e., one about which one's confidence is not affected following its refutation with empirical data.

As far as I can tell, the fact that one may know a good deal about an area is no gaurantee that one might not also harbor the suspicion that "only I know" or "only I can see." I expect that may be why some experienced people not only disagree with one another, but sorrow reciprocally about each other's analyses and abilities.


R. John Howe

Posted by Richard Larkin on 06-26-2006 08:58 PM:


I believe I had heard that Erdmann, having asserted that the origins of pile weaving were much more recent than the burial site discovered by Rudenko, insisted that the Pazyryk carpet was a felt. The story continued that he refused to inspect it. If the story is apocryphal, I apologize to all concerned.

Turning back to Mr. Zahirpour's presentation at the Textile Museum, and that funky and fragmentary bagface that caught your eye, were you sure that it was Southwest Persian tribal? I don't necessarily doubt it, but it is little outside the norm of the typical pieces. I wouldn't be shocked to learn that it was peripheral to that region.

Two colors get my attention, the (what looks on my screen like) auberginish color inside the red hooks of the central medallion and the greenish-brownish color of the inner part of the medallion. Was the latter color the same as appears in the reciprocal devices of the border? What did you make of those colors, if anything?

Sorry to play twenty questions with you. Not only do you do all the heavy lifting on the reporting end, but you have to deal with these infernal questions.

Regards (and thanks again),


Rich Larkin

Posted by R. John Howe on 06-26-2006 10:26 PM:

Mr. Larkin -

One never need to apologize for questions and comments here on Turkotek. That is what we are about.

Monitors are funny things and I'm not sure I'm seeing what you are asking about. More, I'm not the person to answer since I really don't follow S.W. Persian weaving as some do here. (I have been hoping that Pat Weiler would speak up but he may be traveling at the moment.)

If I had a question myself about this fragmentary piece you ask about it would not be so much about whether it should be included in a S.W. Iranian grouping (do you have another suspicion?), but rather about whether it might more likely be Luri rather than Qashqa'i.

Here below are two images of a classic Qashqa'i khorjin with a much more disciplined version of this design.

Here is what Opie says about this latter piece and about Qashqa'i vs Luri weaving more generally. "...the 'latch hook' designs around the main medallion have the appearance of animal heads with eyes that supports my opinion that this Qashqa'i design was derived from Luri sources." Opie goes on to say that this design is found in old Luri bags and on old Luri rugs but rarely on old Qashqa'i rugs. He thinks the Luris are the more fecund designers and weavers, who were absorbed into some Qashqa'i tribal groups. He says that Luri weaving is more "fruitful" than that of the Qashqa'is but that the latter are "unsurpassed" in fineness of weave, choice of colors, refinement of design and with filling space with small designs.

One indicator often used to identify Luri weaving (at least initially) is that it often contains irregularities (as does this fragmentary example of David's). Some time back Pat Weiler hosted a fine salon discussion of Luri weaving that you can find here:

Within in it Daniel DeDeschuyteneer put up the small Luri bag in the link below and made the comments contained in it.

Notice that this piece has many similarities to David's including the trefoil border (which appears on Qashqa'i and other S.W. Persian flatweaves, including both of those that David showed in his rug morning).

I don't have information about knot or materials on David's fragmentary piece, but I think a closer examination might show that it is more Luri than Qashqa'i.


R. John Howe

Posted by Richard Larkin on 06-27-2006 05:41 AM:

Hi John.

Excellent links, thank you. I'm aware of Jim Opie's views in regard to the weavings of the Luri, as I knew him years ago. The bagface about which I inquired in the David Zahirpour group looks like it came from the same place as the excellent one discussed in the Danial DeDeschuyteneer link.

Rich Larkin

Posted by R. John Howe on 06-27-2006 08:54 AM:

Richard -

Not to go on about it unduly but you mention "Pazyryk felts." Of course, you likely know there were some.

They were quite wonderful and included some rows of horses with riders similar to those that appear on the border of the rug.

My own personal favorite Pazyryk felts are the stuffed geese.

And while we are talking about Luri weaving, we have perhaps licensed my putting up the piece I brought to David's rug morning as a possible S.W. Iranian item.

Here's the other side:

This is a piece I bought entirely on the basis of its colors at the Seattle ACOR a few years ago.

It is done in a coarse sumak and you can see its drawing contains irregularities and made me think it might possibly be Luri.

It has another aspect worth noting: it is a bag made in one continuous strip. This raised the possibility that it might be a "constructed" piece (i.e., a bag made by cutting down and sewing up a strip from something originally larger). The consensus to date (and David agreed) is that is it not such a "constructed" piece since it has indications of "finish" at both ends and at its sides under the heavy goat hair selvege.

It is crude and a bit rude, but to me exhibits a colorful exuberance that strongly suggests a Luri weaver.


R. John Howe

Posted by James Blanchard on 06-27-2006 11:33 AM:

Hi John,

I would also like to offer my thanks for your continued efforts as a "roving reporter" of interesting rug-related events and exhibitions.

I am interested in the "Khamseh" bird rug that you showed from David Zahirpour's session. To me, it looks uncharacteristically "wonky" for this type of rug. It seems to have been woven by someone who was not well-versed or practiced in this traditional design. It looks like the weaver wasn't even very comfortable drawing the usual Khamseh "murgh" (chicken/peacock), and so replaced them with simple quadripeds here and there. Although I have little experience in this area it makes me wonder if this is actually a Khamseh, so I will ask Patrick's question... "could it be Luri?"

For comparion, here is an example of a Khamseh bird rug that I own that has the same basic design, but is executed in a more traditional or controlled or strict manner. In fact, this one of mine is about as strict in its design symmetry (with the exception of some problems in the resolution of the main border) as any I have seen in this general category, suggesting perhaps that it was manufactured in a more controlled setting. I would mention that mine is symmetrically knotted, which is in contrast to another I have that is assymetric. Does anyone know whether this is a significant distinction (e.g. Arab vs. non-Arab Khamseh?).


Posted by R. John Howe on 06-27-2006 04:18 PM:

Hi James -

I can't answer your question of whether the drawing in David's "Khamseh" is irregular enough to make us suspect it might be Luri. Usually, one looks for a number of coalescing indicators.

I can say that Opie says explicitly that many "Arab-Khamseh" rugs have asymmetric knots but then bestows that attribution on one that he says does not (pages 74-75).

We need to buy Pat Weiler a laptop and a deal on a wireless connection at the hotels he uses when on the road.


R. John Howe

Posted by James Blanchard on 06-27-2006 07:33 PM:

Thanks John.

I also seem to recall either Opie or Eiland writing that Arab Khamsehs tended to produce rugs with crisper drawing than others, but I don't have my books with me to verify that.


Posted by Richard Larkin on 06-27-2006 09:02 PM:


That bag you acquired in Seattle is a little gem. Did David Zahirpour have anything of interest to say about it? My working and ill-informed method for cataloging pieces that seem South Persian but not standard Qashqai or Arab (Khamseh) is to ascribe them to the Luri or Bakhtiari. Luri for your little item sounds as good as anything. On the other hand, the colors are light and bright, qualities I don't generally associate with Luri weaving.

Mr. Blanchard's rug is quintessentially Arabi in my estimation. The related item in the Zahirpour group somewhat less so, but I wouldn't take it for Luri work. I believe there is a very broad range of style within the greater Shiraz weaving area (i. e., Qashqai/Khamseh) within which Mr. Zahirpour's rug fits quite comfortably. On the matter of distinguishing design related Luri work from the work of other tribal groups in the South Persian area, I think the relative paucity (in Luri weavings) of "busy" filler motifs is diagnostic.

Incidentally, I would have said that a lot of what I would call Arabi work was less precise in drawing that other South Persian "tribal" work, if anything.

Rich Larkin

Posted by R. John Howe on 07-05-2006 09:56 PM:

"Shiraz" in the Rug Literature

Dear folks –

Before this thread gets archived I thought it might be useful to explore a bit how the term “Shiraz” has been used over time in the rug literature. I had had the impression that “Shiraz” was a usage based on the fact that the city with this name is the location through which most southwest Persian rugs are marketed.

David, you will note above, seems to treat it as a species of rug. In fact, he pushed back at my question of whether the “Shiraz” usage was similar to that of “Bukhara” which seemed for a long time a catch-all way of describing most Turkmen rugs. He defended his position by giving technical indicators for recognizing Shiraz pieces.

So I thought I would look at what the literature has said over time.

I started with Hawley (a 1937 edition of his book originally published in 1913) because he is one of the first of the older rug book authors who treats the technical aspects of rugs systematically. He includes a chart at the end of each major segment where he summarizes the technical features of particular types of rugs in that grouping in convenient to compare matrix.

Hawley treats “Shiraz” as a type of rug. He has a section on it in his text, in which he refers to Martin as also considering it a specific type and says that it is a weaving tradition rooted in the 15th century. He describes “Shiraz” rugs as being made of very high quality glossy wool, the dyes include striking reds and blues and the effects of the colors in them is often like “stained glass.”

Here is Hawley’s technical description of Shiraz rugs from his Persian rug matrix: (I am not always using his precise language.)

Knot: Usually asymmetric but there is a lesser use of symmetric knots.
KPI, 7-12 horizontal and 8-12 vertical.
Warps: Wool (occasionally goat hair), usually warps are about level but in some instance are depressed.
Wefts: Wool, two picks between rows of knots
Sides: Selvedged rather than overcast.
Lower end: Webbed rather than selvedged; warp loops
Upper end: Webbed
Nap: Medium
Weave: Medium rather than fine or coarse
Texture: Loose

Hawley refers to “Afshari” and to “Luristan” rugs in his index, but not to Bakhtiaris, Khamsehs or Qashqa’is and mentions the rugs of Luristan only once, in a discussion of what Marla Mallett would call “weft ease.”

Next I looked at Edwards “The Persian Carpet.” Widely respected and admired sufficiently by the Iranians themselves that they had it translated into Farsi. Edwards retired in 1947 after nearly 50 years in the rug business in Iran and published his book in 1953. He takes what seems at first a quite different view of “Shiraz” and of S.W. Persian weaving more generally than does Hawley.

Edwards: “…Shiraz, the capital, (ed. of Fars) is not a weaving center. There is no (ed. weaving) in the towns. It is, however, important to the merchant because it is the market for the thousands of rugs and small carpets which were woven by the nomadic tribes and settled villagers in Fars.

Every day dust-covered lorries or little caravans of donkeys bring them in from the plains or the mountains, 20-150 miles away…The name Shiraz which is commonly applied to these rugs, is therefore a misnomer. They should be called after the province where they were woven --- Fars.

Edwards divides the rugs of Fars into four groups: Qashqa’i, Khamseh, Mamassani and Hulagu (the latter two apparently his terms for those who speak “Luri’) and “Persian village.” He treats the “Qashquli” as one of eight Qashqua’i sub-tribes that he lists. He lists five (what else?) Khamseh subtribes and two groups indicated above that speak Luri. His category of “Persian village” rugs in Fars is defined as those made in the villages that surround “Shiraz” within a radius of 150 miles.

So perhaps Edwards’ point about the “term” Shiraz” is not as serious as it initially seems and is only that he prefers a “Fars” reference, since it seems likely that his “Persian village” grouping in his Fars classification is very similar to what others refer to with the term “Shiraz.”

Edwards ends saying that the Khamsehs and the “Persian villages” weave 85% of the rugs woven in Fars and that the Shiraz merchants do not have a good opinion of “Luri” weaving. He ends by saying that he hesitates to give advice to indigenous weavers but that those in Fars would produce a better product if they could be convinced to adopt cotton warps.

Next I looked at Jacobsen’s “Complete Guide,” published in 1962. Although Jacobsen is often too busy bragging or trying to impress the reader to actually write the considerable he no doubt knew about rugs, I went to him as likely reflecting pretty accurately the notions likely conversant about the “Shiraz” designation among dealers in the U.S. before the more scholarly work began to appear (Opie’s first book was published in 1981).

Jacobsen uses the “Shiraz” designation for a group of rugs, but defines their source geographically. Like Edwards he includes an area 150 miles distant from Shiraz, but his is a square rather than a circle. He seems explicitly to include both Qashqa’i and Khamseh within it. Jacobsen talks about “old Shiraz” which he says can, in rare instances, have “golden canary” grounds and be “as pliable as a silk handkerchief.” He also praises the quality of their wool and colors and says that 90% have wool warps. The newer Shiraz, he says, can be as coarse as 40 KPSI, but also as fine as 300 KPSI. He also volunteers that what are widely read as “chickens” in S.W. Persian rugs are more usually “nightingales,” and are supposed to be symbols of “contentment” and “happiness.”

This latter passage reminded me of a speaker who came to the TM with a wonderful Khamseh rug and said that the “bird-like” devices on it could not possibly be chickens. On a rug like this they must at least be eagles. Jacobsen provides several black and white images of rugs he labels “Shiraz,” but except for the “yellow-ground” rarity, they do not seem remarkable. Stacked medallions on two and a “mille fleur” niche design on another.

Opie, in his first book “Tribal Rugs of Southern Persia, 1981 works at the level of tribes and sub-tribes. He does not provide an index in this book, but it appears that he does not treat the “Shiraz” usage at all. His uses of this word seem invariably to describe dealers. This might well be expected since, like the “Bakhara” marketing center usage for Turkmen pieces, “Shiraz” is likely what he was moving away from. I also checked Opie’s later “Tribal Rugs” volume. It has an index with six “Shiraz” references, but all again are to the town or dealers in it, not to a group of rugs. So it appears that Opie avoids that particular Shiraz usage.

Last I checked what the two Eilands say in their most recent “comprehensive guide.” They begin by indicating that the “Shiraz” usage indicates only that a piece is estimated to have been woven somewhere in the Fars province. Their subgroups for Fars are:

Village production

As you can see, their categories are not parallel, since they do not attempt a further labeling of “village rugs,” and indicate clearly that “gabbehs” are a type of rug not a weaving group. (It occurs to me that, excepting for the omission of the Loris, and the reversion to the tradition of calling the “village rug” group “Shiraz,” the Eilands’ listing is nearly identical with what David offered. It may be that their treatment influenced the categories he adopted for his presentation.)

The Eilands also seem to suggest that Edwards' wish for cotton warps on S.W. Persian rugs has been realized to a considerable extent. They say that a cotton structure is increasingly being used and that sometimes cotton wefts are used with wool warps. Edwards' would wish for the reverse.

Anyway, that is what I can suggest quickly about what the rug literature has said about the “Shiraz” usage, in selected volumes over the last 50 years or so.

Comments and corrections are invited.


R. John Howe

Posted by Gene Williams on 07-06-2006 03:26 AM:

enjoyed it

very good post John and very informative. Thanks.


Posted by Richard Larkin on 07-06-2006 08:10 AM:


That was almost nostalgic. You certainly characterized Jacobsen accurately. And Hawley! Wow, the old guard.

Fars area rugs were and remain among my favorite types. I don't see cotton wefts (or warps) improving the aesthetics, with all due respect to Edwards. However, I am reminded that there is (or was up into the 1980's) a type of village rug in the market from that area that bore the trade name "Abadeh." It had cotton foundations, as I recall, and uninspiring designs vaguely suggestive of the area's tribal production. I realize there is a town of that name not too far north from Shiraz, and the rugs probably come from there. However, I recall having heard that the Abadeh rugs were being produced by tribal weavers who had gotten themselves organized, perhaps settled, for the job.

In light of having checked those written sources and having thought the matter over, do you have a sense of what types of rugs David Zahirpour has in mind when he insists there is a particular "Shiraz rug?"

Rich Larkin

Posted by R. John Howe on 07-06-2006 08:53 AM:

Richard -

About Edwards desire to move to cotton warps: it's not about aesthetics at all. Edwards was the long-time employee of an English rug making firm. The sub-title of his book is "A Survey of the Carpet Weaving Industry of Persia." His concern about the desirability of a cotton foundation is entirely commercial. A rug with a cotton foundation is not prone to the problems of becoming oddly shaped when taken from the loom and will lie predictably flat on the floor.

Once when lecturing publicly about Turkmen rugs, which Edwards cites as the other major group that uses a wool structure pretty much throughout, I joked that, if you live in a tent, the "flatness of the floor" does not come up much in conversation. But if you have hardwood and have invested a little money in an oriental rug, you likely will find it irritating if it adopts odd shapes or exhibits areas that "bubble up" off the floor. This was Edwards concern: to meet the requirements of the western customer.

About Abadeh. It is as you say rather close to the Shiraz area and the Eilands treat it in that section of their book. Some others treat Abadeh rugs in their Kerman section. I have a Persian dealer friend who says that the weaving tradition in Abadeh is rather recent and the Eilands report that in the 1950s a different "fabric" with a cotton foundation and asymmetric knots appeared there.

About what David thinks a Shiraz is, if you look back at my early paragraphs you will see that he described the group he sees pretty concretely complete with technical indicators (more than two wefts and a noticably loose weave are highlights). As I said there, I suspect this is one place where he has yielded to a market usage rather than the more recent "village rugs" one.


R. John Howe

Posted by Gene Williams on 07-07-2006 09:54 AM:

Iranian census figures

Hi John,

Theres a lot of Iranian census data around; a mini census in 1991, major census inn 1996...another mini one in 2001 and one is due out this year. Wonder how David's thesis holds up against the official census figures. its all in english on the net.

Here is one summary on Fars Province from the 1991 census which was not as detailed as one would mention of a "Gabbeh" tribe:

The following tribes are living in Fars Province: Qashqaie, Boyer Ahmad Olia, Arab, and Basseri Tavayef.

The families and clans of the said tribes are as follows: The Qashqaie tribe consists of the following clans: Darreh Shouri (5,265 families), Kashkouli Bozorg (4,862 families), Shesh Boluki (4,350 families), Kashkouli Kuchak (650 families), Qaracheh (430 families), Safi Khani (235 families), Rahimi (370 families), Farsi Madan (1,505 families), and Amaleh (5,397 families).

The Boyer Ahmad Oliya is made up of the following clans: Negin Taji (640 families), Sadat (290 families), Ardeshiri (150 families), Qaed Givi (500 families), Jalil and Babakan (610 families), Amaleh Yab (170 families) and residents of villages (130 families).

The Arab tribe is composed of the following clans:

Mohseni (930 families), Farsi (350 families), Shiri (580 families), Abda Yousefi (540 families), Rahimi Chanani (100 families), Jaberi (120 families), Amaleh Sheibani (155 families), Naqd Ali (80 families), Labu Mohammadi (250 families), Pir Eslami (60 families), Derazi (120 families) and Bani Abdollahi (30 families).

The Basseri tribe includes the following clans:

Abdoli (110 families), Ali Qoli (20 families), Labu Musa (260 families), Ali Qanbari (26 families), Zahrabi (40 families), Jackmei (350 families), Jochin (150 families), Ali Shahqoli (50 families), Hannani (80 families) and Nafar (40 families).

The Mohseni tribe consists of the following clans:

Rostam, Bekosh, Javid, Fahlian, Doshman Ziari (there are no figures available on the number of their families but they should not be very high) and Kurd Sholi (600 families).

As a whole, the Qashqaei, Arab, Basseri and Mamasani have not settled down in fixed places as much as other tribes. They still shuttle between summer and winter quarters.

They earn their living from animal husbandry, farming and weaving and sale of the best kinds of rugs, Gilim, jajim and Gabbeh.


Posted by Doug Klingensmith on 07-07-2006 05:34 PM:

thanks John

Thanks to John and all. Please rest assured that your work is appreciated. I would wager there are many "read-only" viewers like me who usually refrain from posting because we dont have anything salient to offer- just mooching up a free education.
As a research engineer I am painfully familiar with the danger and necessisty of categorical language.
As a novice collector of South/Southwest Persian/Fars/Arab Khamseh/quasi-Qashgai weavings, I will continue to refer to them as chicken rugs...
regards to all- Doug

Posted by Don Ruyle on 07-07-2006 08:37 PM:

Hello All:

Many Thanks to all the work to bring these "Rug Morning" presentations. Your work on this site is very much appreciated, John, Filiberto, and Steve, and as always, the follow-up draws intersting comments. By coincidence, about three weeks ago, I purchased a jajim similar looking to the one presented. I bought it at auction, and have no idea where it is from so now I have a name to call it,"Jajim", and can try to learn the when and where. Ah! The joy of collecting! It looks on my soffa. Best regards, and thanks again.


Don Ruyle

Posted by Don Ruyle on 07-07-2006 08:39 PM:

It looks GOOD on my soffa.

Don Ruyle

Posted by R. John Howe on 07-14-2006 08:28 AM:

Don et al –

I used “jajim” in a kind of throw-away fashion, describing the one textile that David brought into this rug morning.

Your indication above makes me wonder whether we should not say more about what is being referred to here.

First, the exact descriptive terms vary and the spellings often will too. In his “Oriental Rug Lexicon,” Peter Stone prefers a spelling of “jijim” to “jajim, but also says that similar pieces are referred to in various rug weaving societies as “alasa,” “cicim,” and “gadzhari.” They are all warp-faced, flat woven pieces made from narrow strips of various widths and sewn together to make larger pieces.

David’s piece is Persian and so we’ll stay mostly with that grouping. Parviz Tanavoli devotes a chapter of his book on “Persian Flatweaves” to “Jajims” and distinguishes those made in NW Iran, from those made in Khorasan and those made in Fars. Tanavoli also includes similar textiles called “mowj” in his treatment of “jajims” but admits that they are structurally a bit distinctive, being somewhat closer to balanced plane weave.

The uses of jajims are extremely varied. Tanavoli says that are used as floor coverings, blankets, rufarshi (carpet covers), “korsi” (a cover for a heating unit), a pack cover for belongings during migration, a curtain and to cover bedding at the edge of a tent.

Here are a couple of images of jajims in use.

The jajim in use on the camel above is not from Fars, but one made by Kurds in Khorasan.

In the image below, jajims are being used to cover bedding in a tent.

But let’s look at a few jajims from Fars.

The jajim below is early 20th century Qashqa’i. I have rotated the image to the left so that I can let you see more of it. So the warps are horizontal in this image.

There are two strips sewn together to make this piece which is almost 4 feet wide and over 14 feet long. It is entirely of wool.

The next piece is also Qashqa’i but of the “mowj” variety of jajim I mentioned above (in this case “balanced twill” weave).

Again two pieces sewn together to make a size of about 5 ft by 7 ft. Tanavoli says that most Fars jajims are rectangular in shape (those from other areas are usually more square-ish) and often have fringes top and bottom, as this one does.

A third example of a jajim from Fars is the “shisha-derma” below.

This term refers to another variety of jajim that is usually done in “black and white.” The structure is “warp-faced alternating-float weave.” It is again in two pieces, all wool, and about 8 feet by 5 feet. Tanavoli says that this is one of only two jajims of the shisha-derma variety that he has seen that were intended as floor coverings. He attributes it to the Qashqa’i in Ameleh, Fars.

While we are talking about jajims it might be useful to show you some examples from other areas of Iran.

Some of the most spectacular jajims are attributed to the Shahsavan on the far NW border of Iran. Tanavoli calls this piece Moghan, Azarbaijan.

He estimates it as mid to late 19th century. All wool, four panels, to make a piece about 6 feet by 5 feet.

The piece that seems to resemble most closely the one David brought to this rug morning is an example that Tanavoli attributes to Zanjan, in southern Azarbaijan.

Tanavoli says this piece is comprised of nine widths sewn together to achieve a size of over 5 feet by 8 feet. This jajim is all wool, warp-faced plane weave and is estimated to have been made in the late 19th century.

Flatweaves similar to jajims were also made in Central Asia. I will show you just one.

This is a warp-faced piece of Uzbek flatweave folded for display on the wall so that only about one third of its width is visible. It is again all wool, done in strips sewn together. It has a warp-faced weave. Such Central Asian pieces are colorful, attractive and are encountered with fair frequency nowadays in the market. Its vertical length in this image is about 6 feet.

That is my little survey of the jajim format. As usual, comments and corrections are invited.


R. John Howe

Posted by Marty Grove on 07-14-2006 09:28 AM:


G'day Mr Howe,

Another wonderfully comprehensive and interesting display.

A question though; none of the pieces shown, with the exception of those aboard the distainfully expressive camel, and perhaps in the tent, all the others appear unaged. Are they in fact representative of flatweaves which are modern, and readily available in the west?

These beautiful textiles are so durable, and useful in our western lives, for purposes obviously beyond the original intention of the weaver, and Im sure many of us would enjoy easy access to them.

Although there are many similar forms enjoying current popularity as the turkish machinemade types have shown, perhaps the price differential may not be so great as to preclude the handwoven types gaining some inroads.

One other question, a silly one for a rug lover, as perhaps the answer is one I should already know, however above was mentioned in the data analysis of Shiraz rugs - 'selvedged rather than overcast' - well, this has always interested me; the difference I dont really understand.

As I have accepted, selvedges are the cords at the sides which generally are thicker than the warps, over which the wefts wrap at the end of the row of knots, which is then overcast during the completion of the rug.

That is how I have always thought of it. Could you please explain the difference between the two 'types' of side finish please - selvedge and overcast.


PS. Perhaps on reflection, I should have said some of the Jajims show little wear. I see now most are not new. Nonetheless, are these types still readily available as new products now?

Martin R. Grove

Posted by R. John Howe on 07-14-2006 10:33 AM:

Hi Marty -

Yes, jajims are encountered with fair frequency in the market today. One likely has to contact the sorts of rug dealers one sees in Hali or on, when the latter is resurrected. They are not as frequent as pile rugs, or some flatweaves of different sorts (e.g. sumak bags), but are hardly rare. Marla Mallett occasionally has them.

Now about your question of the distinction between a "selveged" side treatment versus one that has been "overcast," notice first that my reference was to Hawley's usage. That's some time back, when technical features were only beginning to be examined, so I don't know with precision the nature of the distinction he would make if we were able to ask him.

But Marla Mallett in her book "Woven Structures" (which given your interest in such things you might want to buy on her web site) lists the follow categories of pile rug selveges:

Plain interlaced
Attached interlaced
Attached overcast

Now since she includes "overcasting" as a kind of "selvege" her distinctions are somewhat different from those of Hawley. But her definition of "overcasting" might help. An "overcast" selvege is one that is not "interlaced." That is the overcasting cord simply circles one or more warps. My guess is that Hawley's "selveged" usage may refer to a side finish that features some "interlacing" of the warp cords.

Hope that helps.

R. John Howe

Posted by Marty Grove on 07-14-2006 11:10 AM:


G'day Mr Howe,

And thanks, I have read it on Ms Malletts site, and at the time scimmed my mind over the different finishes on my stuff and all seems normal, as it does now again.

The 'interlacing' can be invisaged as having been done prior to the weaving beginning, say 4/6 warps, Im guess anyway, after which at the end of the rows, the shuttle chucks across and is wound back, starting the 'overcasting'.

Thats how I see it, and hopefully it corresponds with that which you see...

There are few passages about rugs on the more readily accessable sites which I havent looked at, but unfortunately I have the sort of mind which is immediate - meaning that at the time of reading, its absorbed and goes into the files; its there but unfortunately finding it again as quickly is more difficult - it makes me slow, but I get brainwaves every now and then, which makes up for it.

The biggest shock I had was my discovery on two really nice Caucasian type Ardibil mats, that the sides were attached after the weaving had been complete - I had read of it without the actual visual and tactile examination - and it was quite disturbing especially as I had to overcast repairs several inches where the selvege had detatched itself from the actual rug.

Many of my used Baluchi sides have had to be redone to protect from wear, but the cords were part of the structure.

Fortunately my rugs are not such that I cant have a go at preventative small actions in the care of them, but it would be a nuisance if it were necessary to have to send them off to say, Vincent or other, for professional attendance.

Ill have a Shiraz when I get home and get down and enjoy further speculations.......


Martin R. Grove

Posted by R. John Howe on 07-14-2006 11:47 AM:

Hi Marty -

You say in part:

"...The 'interlacing' can be invisaged as having been done prior to the weaving beginning, say 4/6 warps, Im guess anyway, after which at the end of the rows, the shuttle chucks across and is wound back, starting the 'overcasting'..."


It is my understanding that an interlaced selvege is often done with the weft strands themselves (although not necessarily). If so the selvege would be constructed row by row as the field work was completed.

Now if a cord other than the ground wefts was used to do the selvege that could concievably be done later (although it might still be easier to do at the end of each row as that is completed). I think the only time when a selvege might be done prior to weaving is when it is a separate thing to be attached after it is fashioned.


R. John Howe

Posted by Marty Grove on 07-14-2006 12:12 PM:


G'day Mr Howe,

Yep, thats how I see it would be done also. Just thinking of the structure I see how a reinforced side could be constructed, in all its ways.

From all those weavers, being taught by Mum, or the Master and the leader of the workers, an established method which has over almost aeons evolved to protect the structure from dissembling.

They sure are tough stuff, weavings, and its no wonder they survive so well for such a long period of time, and it amazes me the condition some get to and remain inherently strong, even though the pile, if piled, is no longer there, the sides are falling apart and the ends well ground.

From one side of the earth to the other, weavers have provided mankinds needs for fabric, and regardless the advent of machinery, handloomed material is still a mainstay of our civilizations.

No wonder we love 'em


Martin R. Grove

Posted by R. John Howe on 07-16-2006 06:40 AM:

Dear folks -

I pass David Zahirpour's shop frequently on one of my exercise walk routes and yesterday stopped to chat.

I confirmed with him that the striped textile he brought to his rug morning and that I discussed as the lead-in piece to a general characterization of the "jajim" format was in fact made in several pieces and sewn together. So this piece is an example of a jajim.

Second, I talked to David a little more about the "Shiraz" characterization and happened in that to say again that Cecil Edwards says that there was no weaving in the city of Shiraz proper during the time of his career (approximately the first 50 years of the 20th century).

David says that his father knew a family living in Shiraz who did weave, who wove rugs that they signed and that he, himself, owns one.

I think this is one of those occasions when conflicting information does not necessarily entail contradiction. Edwards, is talking about weaving visible to someone surveying the Persian rug industry and I think he is saying there was no "visible" rug production in Shiraz proper during his time. But Iran is a rug weaving society in which a great many women have weaving skills. So it is not surprising if individual weavers (even families of them) might be weaving without that coming to the attention of someone surveying the industry.

NW Ohio is not an area known to contain lots of knitters. But my mother knitted there steadily for 30 years and produced a great many items, some of which I still have. Her work would have been invisible to an "Edwards."

So I suspect that David and Edwards can both be correct without our having to give up on Edwards' basic characterization of weaving in Shiraz city.

This is the sort of thing that could lead someone to question whether the rug literature has much to say that is useful. I hope my suggestion above shows why it may be that conflicting indications do not always entail real contradiction.


R. John Howe

Posted by Steve Price on 07-16-2006 08:36 AM:

Hi John

I'm afraid I find your post confusing. Cecil Edwards wrote that there was no weaving in Shiraz during the first half of the 20th century. David Zahirpour says that he knows that there was.

If David is telling the truth (and I have no reason to suspect that he isn't), Edwards was mistaken, and this contradiction in the literature (treating David's spoken words to be part of "the literature") exists because Edwards was wrong. In fact, every contradiction in the literature is probably a result of at least one person being mistaken.

In my opinion, this simply illustrates the hazards of using absolutes. Had Edwards written that there was rather little (or nearly no) weaving in Shiraz during the first half of the 20th century, he would probably have been correct and there would be no contradiction.


Steve Price

Posted by Gene Williams on 07-16-2006 08:44 AM:

could both be right?

I always interpreted the passage in Edwards about shiraz to mean there was no significant commercial weaving to speak of in shiraz.

However, David doesn't look to be much more than say 50. that would mean that whenn he was say 10 in Shiraz it would be around 1966, 18 years after Edwards wrote his book.

So could both be right? Big changes took place in Iran from 1948 to 66. The shah was embarked on modernization. Nomads were settled down. Its reasonable to assume that weaving families migrated into Shiraz in the time from the end of WWII to 1966.

Gene ''

Posted by R. John Howe on 07-16-2006 04:05 PM:

Hi Steve -

Gene has apparently said better than I did what I think is the case. I think that David is unquestionably right that he has signed piece woven by a family in the city of Shiraz and that they wove a group of them.

I am not sure, but I think that Edwards could also be right since such an effort might well not be visible at the level at which his "commercially" focused survey was scanning.

But I also have to say that my speculation about how there might not in fact be a contradiction is just that. If Edwards was scanning at the level at which this (small) production existed, then he's clearly mistaken.

It's not a big deal. I was just trying to suggest how both could be true. I did so in part because of the conversation suggesting that the rug literature is full of error (something we know). I thought it a possible example of a seeming, but not necessary, rug literature error of the sort that someone might jump at in a conclusionary way perhaps too quickly.


R. John Howe

Posted by Steve Price on 07-16-2006 06:02 PM:

Hi John

There are no submarines in the small Virginia village in which I live. That statement implies a high level of certainty, and you can take it to the bank. If I knew only that I hadn't seen any but that some of the village was still unknown to me, I would just say that I don't think there are any submarines here.

It's probably another manifestation of what Jean calls my linear thinking.


Steve Price

Posted by Jack Williams on 07-17-2006 01:56 AM:

RE: Edwards

I think Edwards is fairly clear when a slightly more exact quote is taken from his work on p. 284.

"Shiraz, the capital, is not a weaving centre. There are no carpet factories in the town, nor is weaving carried on in the houses, as in the other weaving centres of Persia...."

...Because of the mention of "factories," and "weaving center," as John and Gene pointed out his comments about Shiraz are most likely to concern commercial weaving, not for-home use. It would seem that his reference to "carried on in the houses" is in context to his entire approach and concern with commercial activity as he pretty consistanly discusses "number of looms," "numbers of rugs produced etc. throughout his book. Forced to choose, I would side with Edwards vs even a reputabale modern carpet dealer... whose commercial interest may cloud the issue.

I recently read a new book entitled "The Root of the Wild Madder" which as intellectual property was about on the level of the title. However, the author, Irish, traveled fairly extensively ala Edwards through Iran within the last 4-5 years. His rug knowledge was rudimentary, and his self-depreciation in print become both painful and as hard to stomach as a silk stocking upper east-side New York lisp (when I saw his picture in the book, he just looked like someone who ought to be...slapped for being such a professional wimp or something).

However, in the book were were some quite interesting passages. He had extensive pictures of Quasquai weaving..including pictures of tribes women, under contract, weaving of an exact copy of the Pazyryk carpet for some rich German client of a merchant in Theran. He also talks a great deal about the Persian poet Hafaiz, of Shiraz. And he mentions a particular phsyco trait found in 1 of 2000 or so people who see words as colors.

His descriptions of the bazaars of Theran, Ishfahan, Shiraz and other cities is interesting if you can stand a self-centered, New York Times approach. Perhaps I ought to write a book review or something.

Regards, Jack Williams

Posted by Gene Williams on 07-17-2006 03:06 AM:

submarines in McLean

My son had a submarine he used in the bathtub. And...and...according to a reputable source (JL-1967), "We all live in a yellow submarine." ''

Posted by Steve Price on 07-17-2006 05:34 AM:

Hi Gene

Ya got me.

Goodbye, cruel world.

Steve Price

Posted by Patrick Weiler on 07-18-2006 01:21 AM:


There is a local dealer here in Seattle who is from Shiraz. He has a small rug, about the size of a balisht or yastik, which he says was woven in honor of his birth. He is about 60 years old and the rug is easily that or older. I would take this to mean that his family, from the Shiraz area, wove this and probably additional weavings. They most likely settled in Shiraz sometime in the 1920-1930 era when the Qashqai and other nomads were forcefully settled by the Shah. They retained their tradition of weaving, even if not commercially. It is also probable that settled tribal weavers in the village of Abadeh utilized their traditional designsto manufacture a standard type of rug for the market.
"The market" in this case is Shiraz. One of the first rugs I bought was described by this same Iranian dealer as "Mecca Shiraz". This term supposedly describes a finer type of Shiraz area production brought by pilgrims from the Fars area on their way to Mecca and sold along the way to finance their trip. Filiberto often has described this same phenomenon regarding pilgrims from the Caucasus stopping in Jordan to sell their carpets.

P.S. John, I recently received a Sprint Air Card to connect wirelessly via my laptop to the internet while on the road. Unfortunately, I recently found that it does not work in Alaska, where I mostly travel for business! Perhaps I could sell it to finance the purchase of a rug?

Patrick Weiler

Posted by R. John Howe on 07-18-2006 05:18 AM:

Ah Pat -

I'm glad you're back while this thread is still up.

You are reinforcing Steve's view that there probably were "submarines" in Shiraz after all. At least two sightings so far.

Please look back at a post by James Blanchard and see what you can say about his question there about whether a given piece looks "Lori" to you.

Most of us come to Seattle from the south and east, and it looks pretty good from that perspective. It must seem like heaven itself coming home to it from the northwest.

Welcome back,

R. John Howe

Posted by Patrick Weiler on 07-19-2006 01:27 AM:

Khamseh or Luri?


The piece Mr. Zahirpour characterizes as Khamseh appears to me to more likely be Luri, as you suspect. The "drawing" is askew. The length is twice the width, rather than the more stubby Khamsheh type. It has four medallions rather than the three that the Khamseh usually produce. The variety and plethora of different animals is a more Luri trait than Khamseh, who tend to stick with one type of animal in their designs. Only one of the medallions actually has birds in it. Notice the half-medallions protruding from the sides of the field. The designs within each half-medallion are completely different from each other. A khamseh rug would have the same design within each half-medallion. The borders are not typical Khamseh designs.

The James Opie book, Tribal Rugs, notes about Luri weavings:
"The irregular shapes of the medallions and relatively loose weave are typical Luri features...Luri weavers allowed greater latitude for designs to work themselves out in the course of the weaving process."
It is certainly understandable that this rug would be confused with Khamseh work. Most "bird" rugs ARE Khamseh, but Opie also notes:
"Luri "bird rugs" constitute a rare and appealing sub-group. Most south Persian examples of this type are the work of weavers within the Khamseh Confederacy. Luri versions are sufficiently different from those of the Khamseh to suggest that a united tradition, once shared by several native Iranian tribes, began to fragment several centuries ago and to develop regional characteristics."
Regarding the variety of animals and designs in this rug, Opie says "The improvisational approach to a frequent feature of Luri work."

I would consider the Zahirpour rug to be even more desirable as a Luri than a Khamseh. If I owned it, I would have a contractor add several feet to my living room so I would have enough space to hang it!

Patrick Weiler