Posted by Sean R. Roberts on 06-18-2006 11:48 PM:

Bukhara Beshir

This is a Beshir rug that I picked up Saturday in Kazakhstan. The dealer, who I know well, said it came from Bukhara. This also seemed to make sense to me since I have seen many many Beshir carpets in Bukhara. Aside from just showing it off (because I like it), I am posting it for the following reasons:

1) to see if anybody can venture to guess its age

2) to see if anybody knows why there were so many Beshir carpets in Bukhara at the turn of the century. I imagine that most Turkmen carpets in Bukhara at this time came from the Mary area. Is this home to the Beshir Turkmen??

Full length:

Center (excuse the toes):


Corner with view of back:

Posted by Filiberto Boncompagni on 06-19-2006 02:51 AM:

Hi Sean,
The matter of Beshir identification was discussed a while ago in mini salon 13, in the thread Erik Risman: A Little More on "Ersari/Beshir"

Posted by Sean R. Roberts on 06-19-2006 04:06 AM:


Thank you very much for alerting me to this mini-salon. It is, however, a little unsettling to one who collects Central Asian rugs. It seems to me that the mini-salon makes a case for re-categorizing virtually all Central Asian rugs that are found in the Oases of Uzbekistan--they are all to be known as "Middle Amu Darya" rugs. Perhaps, the certral key to the mystery of these rugs is that they simply defy categorization (something suggested by piling them all into an amorphous category). On the other hand, the strange part of the discussion in this mini-salon is that it focuses on what most people call Beshir rugs. The Beshir rugs, to me seem to be the only rugs from this "Middle Amu Darya" region that have definite attributes. The rug I have posted, for example, seems to be a definite example of a certain type of rug that I have always associated with a "Beshir Turkmen" design. Whether that has any correlation to an ethnographic category of Turkmen or not is an interesting question. Another interesting question is whether the entire "tribal attribution" puzzel for Central Asia may in fact be a flawed paradigm for understanding the rugs one finds here. As an anthropologist myself, I would also suggest that such an "unraveling" of accepted categories of Central Asian rugs may be reflective of the flexible ways in which identity has always been expressed in the region.

Now, if it is not easily attributed to certain weavers; is it at least easily datable?

Posted by Steve Price on 06-19-2006 05:14 AM:

Hi Sean

The "Beshir" appellation has traditionally referred to rugs and carpets marketed through Beshir, believed to be woven by urban Turkmen but with a design vocabulary not derived from tribal Turkmen.

Bukhara has long been a market city for Turkmen weavings, and many dealers still use the term "Bokhara" as nearly interchangeable with "Turkmen". Some old books that include Turkmen carpets label them as "Bokhara". In view of all this, it doesn't surprise to me to learn that Bukhara is now a market for "Beshir" carpets.

The size of your piece virtually precludes any origin outside of a workshop, so I'd assume that the colors were selected by a wholesaler or merchant and that the carpet was made to be sold. The Beshirs that I've seen attributed to the 19th century have lively palettes, with lots of accent colors, including a medium blue. Yours appears to be restricted to four colors: red, black, white and gray. The palette, and the gray warps, suggest to me that it was woven in Afghanistan around 1950-1975.


Steve Price

Posted by Steve Price on 06-19-2006 05:59 AM:

Hi Again, Sean

You mentioned that you are an anthropologist. A day or two ago, Gene Williams posted a link to an article in Journal of Anthropological Archeology (21: 443-463, 2002), dealing with design diffusion through Turkmen groups. Here is a link to the article. It's certainly an interesting paper, and relevant to topics that come up often here.

Being an academic, I am pretty familiar with the hazards of taking the findings of one paper without being conversant with the field and whatever controversies may be going on or have gone on in the past. Not being an anthropologist, I don't know the goings-on among anthropology pros. Do you know of any reason why I (or our readers) should be reluctant to take this paper as authoritative?


Steve Price

Posted by Sean R. Roberts on 06-19-2006 07:32 AM:


Thank you for sending me to this article. Being a social anthropologist and not an archaeologist, I am not sure that I will be the best critic to evaluate the article. In general, I am skeptical of attempts to quantify cultural evolution in the way that we do biological evolution. That aside, the theoretical approach to the issue does not neccessarily dispell the paper's data analysis, which looks quite interesting. I will read it with the interest of both an anthropologist and a rug collector!

PS--Why do you think this rug would have been manuafactured in northern Afghanistan and not in Mary/Merv? Could it not have been manufactured in the eastern Turkmenistan in the early 20th century? It is indeed possible that they were selling northern afghan rugs in Bukhara in the late 1990s, but the access to Afghanistan from Uzbekistan during the Taliban years was somewhat limited. Besides, there was definitely a significant indigenous market for Beshir-design rugs in Bukhara before the Bolshevik revolution. This is seen even in the Museums in Bukhara (such as the Emir's summer palace), which are decorated with enormous Beshir-designs of the five medallion variety.

Posted by Steve Price on 06-19-2006 08:01 AM:

Hi Sean

Gray wefts on Turkmen-related rugs are usually associated with an Afghan origin. I don't actually know how reliable this indicator is; it most likely came through dealers. On the other hand, dealer information on pieces made in fairly large numbers by weavers under contract to their suppliers seem to me more likely to be correct than most marketplace lore about antiques.

But I hope nobody is in a position in which the accuracy of my guess has significant consequences.


Steve Price

Posted by James Blanchard on 06-19-2006 09:41 AM:

Hi Sean and all,

A couple of months ago I posted a "Beshir" rug that has some similar design elements as yours, but a somewhat different palette. I have read that rugs with these patterns have been woven over long time period. Some thought mine to be quite recent, others (including me) think it is 19th century. I am re-posting a picture of mine, without being knowledgable enough to offer a definite opinion about the age of yours or mine. I would note, as Steve has done, that some of the older examples I have seen tend to have a wider colour palette with varying field colours as seen on my example.



Posted by Steve Price on 06-19-2006 09:55 AM:

Hi James

That's what I'm accustomed to seeing with 19th century attributions. My memory of the thread discussing it has evaporated (probably a bad link in my brain's equivalent of the Windows registry). What made some people attribute it to the 20th century?


Steve Price

Posted by James Blanchard on 06-19-2006 02:19 PM:

Hi Steve,

As I recall, Dave Hunt was suspicious of the colours and the selvedge. I think that the colours are actually very good, but perhaps aren't always well-represented through digital images.

By the way, the wefts on mine are brown wool and the warps grey wool. I have seen several examples of old Beshir rugs that are said to be 19th century that have grey warps. So I am not sure whether that would disqualify Sean's as an older example per se.


Posted by Chuck Wagner on 06-19-2006 06:54 PM:

Hi all,

I have to agree with Steve. The sawtooth minor border, selvage, and red & blue kilims are other indicators that this is probably mid-20th century Afghan production utilizing a somewhat coarser version of a typical Beshir design.

Chuck Wagner

Chuck Wagner

Posted by louis_dubreuil on 06-20-2006 04:04 AM:

sovietic ?

Bonjour James

Your carpet seems to me to be a very good Soviet workshop product. First the palette of colours is not very "xix° beshir" (lack of yellow, replaced here by a dull gold colour, and light blue). Second the design is too crowded and excessively well drawn to be really beshir. Third the wool quality is well in the upper soviet standard with a very soft handling despite a quite long pile and a lustrous aspect. The knot count is certainly higher than that is usual in XIX° beshir.

Meilleures salutations


Posted by louis_dubreuil on 06-20-2006 07:46 AM:

another remark

Hi James

Another remark that makes me to think to a sovietic workshop product, is the dimensions of your rug. Generally Beshir rugs of this type producted on the XIX° c. have greater dimensions (more than 3 meters long) has they were made for local "bourgeoisie" and used more often for covering stands and dais in diverse outdoors urban ceremonies than in rooms in private houses (see various historic pictures depicting such public manifestations). Your rug has dimensions that fit perfectly with western market and indoor use in private houses.


Posted by Steve Price on 06-20-2006 08:03 AM:

Bonjour, Louis

I can't find any mention of the dimensions of James' Beshir rug, and a length in the neighborhood of 2.5 to 3 meters looks reasonable to me from the photo of someone standing behind it and holding it up. It also appears to be more than twice as long as it is wide, not at all typical of the sizes most often seen in our part of the world (in meters, 2 x 3 and 2.5 x 4 are the most common approximate sizes.

Do you see dimensions in one of the posts here? Maybe James sent them to you privately?


Steve Price

Posted by James Blanchard on 06-20-2006 08:07 AM:

Bonjour Louis,

Thanks for the feedback. You are offering some important criteria for distinguishing between "Sovietic" and true "Beshir" production, which I haven't heard or read about before.

I have a few questions based on some of my own reading and a bit of internet surfing to look at other examples of what have been called "Beshir" rugs.

1. Design crowded and well-drawn - I have seen several published examples of this particular design (the "Beshir Herati" design), and they all look quite crowded to me. Have you seen more "open" examples that would mark them as true Beshir? As to the quality of the drawing, it was my understanding that there was long-standing "city" production of such rugs and carpets in controlled circumstances prior to the Soviet era. Many of them were very large and elaborate, so I expect that well-drawn rugs were being produced before the Soviet era.

2. Structure - I was not aware that Beshir production in the pre-Soviet era was so crude. Mine is around 80-90 kpsi. Would that disqualify it from being pre-Soviet?

3. Palette - Perhaps it is difficult to make colour comparisons of computerized images of rugs, but I have seen a few examples that seem to have similar palettes. Here are some from links on Barry O'Connell's Oriental Rug Notes site (

Stephen Louw's "Cloudband Beshir": This looks to have a "brown-green" that is very similar to mine. As I have mentioned previously, mine has sections where green wool has been plied with brown wool to create this effect. I wonder if Stephen Louw is looking in and could confirm whether these green patches are created in the same way. (By the way, the drawing of the edging of the field looks very much like on mine).

"Thompson Ersari Beshir" - The "yellow" on this looks very similar to mine.

"Dodds Beshir" - This has a darker yellow that looks similar, and also has a light blue that is similar to mine.

Please know that I am not trying to be defensive of my own rug, but rather trying to learn a bit more about rugs in general, and in this case, about Beshir rugs in particular. So I welcome frank feedback and critiques of my rug(s) and observations.

Merci bien.


Posted by James Blanchard on 06-20-2006 08:21 AM:


Bonjour Louis and Steve,

I don't have the rug at hand, but I think that it is almost 7 ft. by 3.5 or 4 ft.

I have also read that many such rugs were in larger dimensions, but that doesn't seem to be the rule. I would point to the "Cloudband" version in Thompson's "Carpets from the Tents, Cottages and Workshops of Asia" (p. 76) which is 1.87 m x 1.2 m; Stephen Louw's Cloudband Beshir (referenced above ), which is 2.55 x 1.2 m; Thompson's "Ersari Beshir" (referenced above), which is 1.6 x 1.02 m; Another Thompson Beshir (linked to O'Connell's site, and shown below), which is 1.58 x 1.17 m.

So, I am not yet sure how distinguishing the smaller size is in terms of manufacture of this group of rugs.


Posted by Sue Zimmerman on 06-20-2006 08:56 AM:

I have tried to get to Gene's posted article via Google because from my computer the article in the embedded file just shows gray lines where text should be. I can't get to it. Is this because my computer is old and unupgraded or something?
Anyone know? I really would like to read the article. Sue

Posted by Filiberto Boncompagni on 06-20-2006 09:23 AM:


The article is a PDF file. To read it you have to download Adobe Reader:



Posted by Sue Zimmerman on 06-20-2006 10:43 AM:

Thanks, Filiberto. I have downloaded Adobe and can access the PDF file but only the map, the gull drawings, and the article's title show up. I've had this problem of grayed out text at a few other sites. Most sites show up fine-even those with hundreds of pages. I just don't get it. Sue

Posted by Filiberto Boncompagni on 06-20-2006 11:00 AM:

Hi Sue,
I found this on the web. Hope it can help:
I have got Adobe Acrobat Reader installed and I still cannot view a PDF file

On occasion difficulties can be experienced viewing a PDF file, these problems can range from a blank white screen in the browser window to a frozen computer. To overcome this problem you can download the file to your computer and view it. To do this you must position your mouse pointer over the link which takes you to the PDF file and right click the mouse button. Select 'save target as', then specify where you want to save the file on your computer and the file will download. You should then be able to open the file with Acrobat Reader without any difficulty.

Posted by Sue Zimmerman on 06-20-2006 12:33 PM:

Thanks, Filiberto. That didn't work either but I pushed the Acrobat button between the hand and the printer buttons and suddenly text appeared! So I'm going to start thinking of computer stuff as similar to other mechanical household stuff--jiggle and mess around with it enough and it somehow gets fixed, like the dishwasher latch I've been messing with this morning. Thanks again, Sue

Posted by Sean R. Roberts on 06-21-2006 01:06 AM:

I agree with the others who suggested that James' rug is likely newer due to its size--perhaps an early . Maybe not all Beshir rugs, but definitely the "Besh Ai" design older rugs I have seen in the Bukhara area are much longer than 7 foot. The rug I posted is about 12' x 6', and it is shorter than most "Besh Ai" design rugs I have seen. In fact, when I bought this one, there was an older "Besh Ai" design rug the guy was selling in the same price ballpark. It was about 3 foot longer than mine--- about 15 feet long! I didn't get it mainly because it had alot of creases in it which made it very uneven and unable to lay flat at all. From Louis' comment about the use of xix century Beshir rugs, I am now thinking the creases were due to it being draped over things. I wish I had taken a picture of that rug since it was an interesting example, but it was mostly red, yellow, and black (maybe with grey, can't remember). But I am pretty certain that it did not have blue in it (and definitely not the amount of blue that is seen in James' rug).

Posted by James Blanchard on 06-21-2006 09:44 AM:

Hi Sean and all,

This discussion is similar to a few that have ensued when I have displayed rugs on Turkotek. Some discussants have claimed that the displayed rug is a new reproduction, whereas others have expressed the opinion that it is 19th century.

For the present example, diverse criteria have been given for attributing the rug to later production:

1) The fuller pile and suspicion of a particular red (thereby attributing it as a modern production).

2) Beshir rugs pre-dating Soviet production were not made in the size of mine.

3) Older rugs with that particular "Beshir Herati" design are invariably larger.

4) The palette of mine is found in later Soviet production, particularly the lighter blue and the yellow.

5) The crowded design and good drawing is not found in pre-Soviet production.

6) The fineness of the weave (80-90 kpsi) is not found in the pre-Soviet era.

My somewhat limited reading and experience have never uncovered these criteria, and in fact, I have found several examples that seem to contradict these analyses.

The difficulty for a novice such as me is that these divergent opinions often remain unresolved, so it is difficult to progress in understanding. I could understand if the question related to distinguishing age within a few decades, but I am surprised that distinguishing between 19th century, Soviet-era, and recent Turkmen production seems so controversial.

I hope that I don't sound overly defensive, but I should say that it is sometimes puzzling to get such diverse feedback on what seem to be fairly straightforward examples.



Posted by Steve Price on 06-21-2006 10:27 AM:

Hi James

I think a lot of the inconsistency in attribution criteria comes from the fact that so much that is really not known is written and said with so much apparent certainty and authority. It then gets heard or read, and repeated.

Here are some basic facts to keep in mind:
1. I'm not aware of any existing Turkmen rug that can be documented to have been collected by 1875. Perhaps there are some that I've not heard of, but probably not many.
2. There are only a handful of Turkmen rugs that have been dated by C-14 analysis. I'm skeptical about the applicability of the method to rugs, but let's assume it to be reliable.
3. There are some very good color plates of Turkmen rugs, made from the actual examples in the late 19th century by the Russian general, Bogolyubov. Another Russian, Simakov, published a book that included drawings of some Turkmen rugs a few years before Bogolyubov did, in 1880 or so.

To cut to the chase: except for the very small number of Turkmen rugs that have been dated by C-14, there probably isn't a single documentable specimen (or even a drawing of one) made before 1875 from which to construct a database of characteristics. In my opinion, this casts a very long shadow on pre-1875 date attributions of Turkmen weavings. Although I am familiar with the criteria on which such attributions are usually based, I am always ready for those to change like the latest food superstitions.

Later weavings can be attributed with more reliability, mainly because the dealers who commissioned, bought and sold pieces during the commercial period often (but not always) knew their sources and knew when the rugs were woven. Contemporary production (ignoring rugs made with the intent to deceive the buyers into thinking that they are antique) is usually the easiest to identify.

Much of the Soviet era workshop production has easily recognizable characteristics, too. I'm not familiar with Soviet era Beshir carpets, and have nothing to contribute to a list of criteria for recognizing them. Others seem to be very confident about some of those criteria. I ask, and I mean no disrespect in this, if those people would kindly provide us with the sources of their information, so it can be evaluated.


Steve Price

Posted by Unregistered on 06-21-2006 11:25 AM:

flux in turkmen conventional wisdom

Good morning James.

To me, there seems to be a lot of uncertainty in Turkmen rug provenance and age, especially from pictures. And it sometimes is disconcerting how people can be so sure with just a glance at some pictures. I suspose such judgements are born of experience and a quick impression of design and color. However, Turkmen research has long concentrated on structure and a combination of other factors as equally important.

It seems to me that current scholarship has placed a lot of conventional assumptions about Turkmen carpets in some doubt. Therefore, conclusions using conventional wisdom can be erroneous even from very experienced people. I think one reason for this is because many attribution principles were postulated before the full power of the internet was existent. The researchers were numerically limited in the number of examples they could access to support theories.

But now, it is a new situation. For example, in the course of researching a Turkmen carpet chuval, I've found some great sites, looked at maybe 4,000 pictures of Turkmen carpet articles of all ages. Here are two sites that will provide you with hundreds if not thousands of Turkmen examples of all ages. JBOCs book alone has a huge section on Bashire carpets complete with drawings structure analysis, and color illustrations.

You may be able to come some supportable conclusions as to structure, color, design and pallette, on your own. Then a post that references these conclusion adds quite a bit. If you eventually conclude that your rug is newer than you initially thought, or was made by little green men on mars, so what, if you like it.

Regards, Jack

Posted by Steve Price on 06-21-2006 11:39 AM:

Hi Jack

Would you kindly send me your full name, so I can insert it into your post? We have a number of Jacks who participate, and people like to know who they're talking and listening to.

JBOC's book? Is this something new, or did you mean to refer to his website?


Steve Price

Posted by Jack Williams on 06-21-2006 12:24 PM:

name etc.

Steve, I (Jack Williams) made that previous post. Sorry, but when I deleted my cookies here at work, I also knocked out your site recognition of my computer...and then forgot to log-in.

RE: the book I referenced, sorry again a mistake. I cannot edit the previous post so if you would I would appreciate if you would correct the name of the author of the book I referenced. No, this is not JBOCs web site, it is a scanned, on line copy of the following:

"Oriental Rugs, Volume 5, TURKOMAN" by Uwe Jourdan, Introduction to English version by Ian Bennett, English translation printed in England by the Antique Collectors Club LTD, Woodbridge, Suffolk, 1989 (I think). This book is a vertbatium English translation from German. In this 350+ page book, the author uses and is "largely reliant for its illustrations on pieces sold by the German auction house of Dr. Fritz Nagel in Stuttgart... augmented by illustrations of a number of rare weavings....[etc.]."

It may be that this particular book has not received a wide distribution in the US. But I was amazed at the detail, drawings of individual elements, and the huge number of beautifully photographed and described examples. Steve it is well worth anyone's time... hope I haven't bored anyone if this is a well known reference generally already available to all. Regards,

Jack Williams

Posted by Steve Price on 06-21-2006 12:28 PM:

Hi Jack

Thanks. I thought it was probably your post, but couldn't be sure.

The Jourdan book is, as you note, very well done. It is part of an excellent series, which includes Bennett's book on Caucasians, Parsons' on Afghan rugs, and authors whose names I can't bring to consciousness right this minute on Turkish and Persian rugs.


Steve Price

Posted by James Blanchard on 06-21-2006 05:41 PM:

Hi Steve and Jack,

First, thanks to both of you for providing more references on Turkoman weavings. It is appreciated. I think I will try to find a copy of Jourdan's book.

I recognize the constraints on accurately dating Turkmen weavings from the 19th century, but I am still intrigued by the seeming lack of consistency between different observers. Since I am not familiar with the level of experience or knowledge of those who post on Turkotek it is difficult to sift and sort between divergent opinions. Perhaps this is unresolvable on a forum such as this, but as Steve mentioned additional sources of information or examples might help in some circumstances.



Posted by Sean R. Roberts on 06-21-2006 11:57 PM:

As we are talking about resources on Turkmen rugs, let me add a plug for Moshkova (I think that is her name)--this book is a translation from Russian of several publications written by a Soviet specialist on Central Asian rugs. I like this book because it is the first one I have seen that seems to base most of its information on rugs and sources to be found in former Soviet Central Asia. I suspect that many other sources published in the 20th century base their findings more on rugs to be found in Afghanistan and Iran. This is yet another issue that complicates our knowledge of Turkmen carpets. In terms of the limited nature of knowledge on this issue, we should also remember that studying rugs is a hobby for most people. Even people who deal in rugs have no real reason to be a full fledged expert in attribution and dating... Some academics may be able to do some in-depth research on rugs and textiles, but these are few and far between. But, I think it is the mystery that makes rug collecting so addicting!

Posted by Steve Price on 06-22-2006 06:00 AM:

Hi Sean

One of the things that makes Moshkova's book so interesting and educational is the editing and annotation by the late George O'Bannon. O'Bannon did a great job of bringing the information in it up to date and pointing out errors and their sources. It is a good illustration of the temporary nature of so much scholarly work, something many people don't understand. Thanks for bringing it up.

Your statements about ruggies are completely on the mark, of course, but I would point out that knowing the attribution criteria and their bases is important in making buying decisions.


Steve Price

Posted by Sean R. Roberts on 06-22-2006 06:56 AM:


I agree, but actual attribution may be less important in buying decisions than is perceived attribution. Afterall, perception is reality in markets focused on collectors.... Or, maybe in any market for that matter. I am curious, however, about the statement you made concerning C-14 dating. Is the reason that few rugs have been dated this way that the process is expensive? Or is it that there is not enough interest in accurate dating to push people to have their carpets dated this way? This may be a stupid question since I have no idea how C-14 dating is even done.

Posted by Steve Price on 06-22-2006 08:11 AM:

Hi Sean

C-14 dating is fairly expensive (I think it's a couple of hundred dollars per sample), and the nature of the method is such that it is useless for anything less than, perhaps, 300 years old. Not many rugs are even suspected of being in the range in which C-14 dating would be helpful.

Vis-a-vis perception = reality. It does, to a certain extent. But the collector doesn't want to pay the price of an antique for something that isn't, so he really needs to know enough to protect himself from gross misperceptions. I think it's also useful to be aware of the fact that the crriteria for attributions change, and this can change the value of a piece later on.


Steve Price

Posted by Jack Williams on 06-22-2006 09:34 AM:

RE: book

James, no need to search for that Uwe Jourdan book. The link I gave...

...leads directly to a scanned, on-line copy containing all pages. If you want, you could exhaust your color printer printing a copy. Also, an abriged version of some of Moskova's writings, with comments, can be accessed at the following site:

Regards, Jack Williams

Posted by Sue Zimmerman on 06-22-2006 10:07 AM:

I read the article in the Journal of Anthropological Archeology.
If rugs are called "woven assemblages" by scientists who's focus is on "cultures" what do they call the makers of such things, "woven assemblage facilitators"?
As an artist, I'm kind of an alien to rugdom and science and middlemandom, and seem to speak in a different native language, in a way, I guess, than most of you do. I am wondering if such languages are even translatable into each other.
The terminology in this article, from my '"native" vantage point, looks anything but neutral and because of that it may explain, in part, why field work for "social scientists" ends up as it usually does.
I'm dead serious here. Maybe Sean and other social scientists can lend some insight on this matter. If no one can it seems rather futile, at best, for me to comment further. Sue

Posted by Steve Price on 06-22-2006 10:25 AM:

Hi Sue

The article wasn't aimed at people like you and me, who are not professionals in the field. Those folks use a lot of what looks like jargon, but is actually a language that they have evolved that conveys their meaning with as much precision as they can muster and in as small a number of words as possible.

I suspect (but don't know for sure) that their term, "woven assemblages" is intended to be broader than "rugs" but more restrictive than "textiles". They probably call those who make woven assemblages "weavers", maybe "assemblers". "Facilitators" seems pretty unlikely, since it means something different.

I'm sure the language of the anthropologist professional is translatable, and can think of no way a newly minted professional anthropologist could have learned it if that isn't true. It probably requires more effort than throwing up one's hands in frustration and expecting them to change it to a language we already understand.


Steve Price

Posted by Jack Williams on 06-22-2006 11:13 AM:

integrated approach

Sue, I understand your point.

Biological sciences have used this approach to retrospectively affiliate species, trace the independent devoplement and the evolution of new species. Some of the most spectacular sucesses in this scientific approach are the retrospective tracing of languages and the affiliation of languages into groups.

This article was quite a light bulb to me, especially as a lot of discussion of provenace in rugs seems to end up arguing points of design, structure, etc., especially those designs that seem to have been phylogensis vs ethnogensis arguments as an indicator of provenance.

In my opinion, anadoctal evidence can tend to support the phylogensis thesis...Eiland, for instance, notes the lack of change in rug patterns used by the Kurdish population near Meshed, 200 years after they were settled there, despite being in very close contact with other weaving cultures including Turkmen, Baluch, Afshari, etc.

For me, being aware of the phylogenesis model helps integrate other information. For instance, the ability of a member of a certain tribe to invariable identify what tribe another belonged to simply by the way he wears his hat makes sense when read from a phylogenesis standpoint. This example is taken from "The Turcomen Tribes," Edward O'Donovan, 1882, found on A quote from that article:

"...In Merv itself the distinction between the clans is kept up with the utmost formality. Personally I never could discover the difference between them, but the Turkomans had no difficulty in telling to what clan a man belonged at first sight, On asking once how to distinguish the wearers, a native pointed out that a peculiar way of knotting the sash and wearing the hat always indicated a member of the Sultan Aziz clan, a peculiar tie of the sword belt one of the Burkoz, and other minute points of the dress the members of the other clans. My eye could never be sufficiently trained to tell a man's clan at first sight by the cock of his hat, or the tie of his sash: but my Turcoman friends never erred in the matter, which is a somewhat important one in their society."

What the article does do is place "design" as a phylogenesis marker equivilent to structure. My guess is this is where it can get controversial, and it might prove that design is only some factor, less than 1:1, of structure. It also seems to me that the gist of Turkmen research the last 20 years was toward weighting the structure of a carpet more heavily than the design component...could be wrong though.

Anyway, the scientific approach provides lots of interesting food for thought and it is scary to read the archived discussions and see in them the Phylo vs Ethno arguments. Regards,

Jack Williams

Posted by Sue Zimmerman on 06-22-2006 02:28 PM:

Steve, I understand what you are saying. It is not the use of jargon I'm talking about. That is translatable. It's the implications of word choices I find disturbing, as an artist, really. Artists and artisans are not mearly cultural conduits. Individuals create "material culture" cultures don't.
Cultures, in many cases, are what remains of previous civilizations. Art and craftmanship traditions, are, in many cases, what remains of original art. To me it is obvious that it is inspired individuals who move civilizations and cultures along. I see no comprehension of this revealed in the creepy choice of words of this scientific niche's jargon.

Jack, I'm having a problem with these quantitative models being applied to qualitative problems in general. For one thing, specifically, the results obtained through, and squeezed into, these models can be used to support or negate people's ideas while the truth can so easily slip through the cracks. It's a paradigm thing, I think, which is perhaps untranslatable.
For instance I don't understand how an ethnogenesis model can be applied to Turkmen weavings without basing it's possible use on the assumption that Turkmen tribes can be ethnically grouped according language, diet, social institutions, etc., at all. We know that tribes are groups of people. What is the rest but assumption? Assumptions, in this case, it seems to me, that are founded on the authors' sited rug expert's writings. These assumptions are not necessarily true, you know. Sue

Posted by Steve Price on 06-22-2006 04:40 PM:

Hi Sue

I'm confused by a number of things in your post. You write, Individuals create "material culture" cultures don't. I think, in the context of Turkmen pile woven textiles, the individual artisan is very subservient to the dictates of the culture in which she lived. Yomud women wove things that are usually easy to identify as Yomud; likewise for Tekke, Saryk, etc. Idiosyncrasies of individual weavers that allow us to identify two pieces as being from the same hand are so rare that we hardly ever even consider this.

You then go on to object to quantitative methods being applied to qualitative problems. I'm not sure the problems are qualitative (or that any problems can't be ultimately reduced to being quantitative). Anyway, what alternative would you suggest, bearing in mind that it has be be subject to tests that would falsify it if it's off base. And how could you possibly proceed in any direction without making assumptions?



Steve Price

Posted by Sue Zimmerman on 06-22-2006 05:06 PM:

Steve, I haven't a clue as to how what you are saying addresses what I said or how it refers the article's premises. Maybe someone can translate. Sue

Posted by Jack Williams on 06-22-2006 06:04 PM:

split personality

Sue, you have touched on a key element for evaluating this thesis. But in truth, they do incorporate all the facts you mentioned, language, pottery, traditions, religion, etc. in these type studies.

Collard and Tehrani are not concerned with rugs per se and their paper is not actually about rugs. They are archeological anthropologists who study groups of people and attempt to decode their history and interactions.

Some of the most common archeology tools for tracking cultures are...1) language, 2) pottery, 3) artifacts, 4) names, 5) housing, 6) burial traditions, 7) religion and religious rites, etc. T & C hit on using carpets to try to make some anthropological sense out of nomadic Turkmen history, which is not only mostly unrecorded, but probably fabulously wrong where it was recorded (for reasons, see; This approach also seemed to provide an opportunity to test opposing theories about how change takes place in cultures.

Art and qualitative measures are not important in this type of study unless quality is suspected of being a marker for an identifiable culture. Carpets are not important either, except they provide a ubiquitous point of reference both spatially and in time. Most of these archeology-anthropology people love pottery shards, but in the absence of pottery, rugs will do.

But these studies do not survive peer review if they are insular, ignoring other evidence. That is why the authors took great pains to cite other studies based on language, material cultures, “pottery assemblages,” etc. And when they sum up their results, they are careful to note the correlation of conclusions derived from the textile data with other evidence, such as similarity of clan names, and known geographic distribution of groups. As language is regarded as the most definitive marker, they are also careful to note the difficulty of tracking dialects within a language...which may mean that they were unable to get supporting data from language studies.

I particularly like their characterization of our hobby as being split between the phylogenesis people (most of us) who believe that rugs tell a story of a particular tribe because the structure, etc.; and the evil opposition. Those are the misguided ethnogenesis people who think that market forces have always dominated the carpet world, therefore there is no identifiable attribution with specific groups. The whole thing about provenance was created by carpet junkies to raise the price of their artifacts and everything was copied. In other words, just-because-it-looks-like-a –Tekke- doesn’t- mean- it- is- a- could- have- been- woven- by- a- Beluch- if- there- is- such- a- group, so quit trying to use carpets to define historical social units because they don’t.

Actually in the ruggie world, we have a phenomenon of split personality, phylo vs ethno, that often appears within an individual. A commentator takes a bold stand on a carpet being discussed because, “’s my carpet, I’m certain about provenance and attribution because phylogenesis markers clearly indicate my carpet’s place in a specific culture and time...!” The next day though, in a different line the same commentator posts something else, “...Oh, about YOUR carpet, sorry, it is fake, woven by Pakistanis somewhere using copied weaves and designs taken from others; i.e., your carpet is strictly a product of ethnogenesis and you are too inexperience to realize it, ‘take my word for it...’”

Regards, Jack Williams

Posted by Steve Price on 06-22-2006 06:07 PM:

Hi Sue

I'll take another crack at it.

You wrote,

Individuals create "material culture" cultures don't.

I interpret this to mean that you are objecting to the notion that the migration of artistic expressions from one culture to another (like the cultures of the different Turkmen groups, the subject of the article in question) can't be, or shouldn't be analyzed in terms of the culture, but are best explained as the results of individual creativity. If it doesn't mean this, then I missed the point completely and am responding to the mistaken interpretation I put on your words. If it does mean what I think it does, then my opinion is that it is mistaken. The reason I think migration of artistic expressions (designs and motifs on rugs, in this case) is best analyzed in terms of culture than of individuals is that it seems so obvious to me that those expressions primarily reflect the cultures and that the creative expressions of the artisans (the weavers) are distant seconds. The reason I think that is that I can easily recognize the culture from which most Turkmen textiles came, and I can't see how that could be true if the primary source of those expressions was the individual weaver.

You then went on, with

I'm having a problem with these quantitative models being applied to qualitative problems in general. For one thing, specifically, the results obtained through, and squeezed into, these models can be used to support or negate people's ideas while the truth can so easily slip through the cracks ...

First, you seem to be objecting to applying quantitative methods to the problem the authors were approaching, because such methods can be wrong sometimes. Is this what you mean? I can't find another meaning in your words. So, assuming that this is what you mean, I was wondering what non-quantitative methods could be used to approach the question. We had some conversation on here recently in which people simply insisted that what they believed (about dimensionality and movement in Turkmen weavings being part of the weaver's intent) would be self-evident and immune to testing. This approach, which we might call evangelical (I believe it, my belief is so beautiful that if you adopt it, you will experience the beauty, too), is one that I find unacceptable. Maybe we disagree on that.

You continued,

For instance I don't understand how an ethnogenesis model can be applied to Turkmen weavings without basing it's possible use on the assumption that Turkmen tribes can be ethnically grouped according language, diet, social institutions, etc., at all. We know that tribes are groups of people. What is the rest but assumption? Assumptions, in this case, it seems to me, that are founded on the authors' sited rug expert's writings. These assumptions are not necessarily true, you know.

My understanding of what you wrote is that you were objecting to the assumptions that Turkmen groups can be classified according to language, diet, social institutions, etc. Right? My response was, essentially, that there is no way to proceed in the investigation of any problem without making asumptions of one sort or another, and the assumptions these guys used seem reasonable. Do you think otherwise? If so, what assumptions would you use instead? I agree that the authors' assumptions may be flawed, but unless you (or someone) has alternative assumptions that are less flawed, theirs seem as good a way to go as any other, and better than most.

If this is still not clear, let's simply agree to remain mutually baffled.


Steve Price

Posted by Stephen Louw on 06-30-2006 05:03 AM:

Dear James

Sorry, I have been away for some time and have just checked in. You asked whether the green's on my Ersari/beshir/MAD rug were also interspersed with speckles of brown. Indeed, they are.

Incidently, that is one of the criteria I use to date - or at least cluster groups of analagous rugs. If you look at my rug, all of the areas of colour, and especially the blues, have a dominant colour tempered with frequent specks of a similiar colour. Thus, in addition to three clear blues, there are distinct variations within each blue area. (Not abrash).


Posted by James Blanchard on 06-30-2006 10:07 AM:

Thanks Stephen,

Perhaps you can look at this close-up pictures of my Beshir (which I posted earlier, and have attached below) to see if it looks similar to yours. Unfortunately, the colours in these photos are washed out in comparison to the rug itself, and I don't have it close at hand to take more photos. A close inspection of my rug shows that this colour mixing is not achieved by interspersing knots of different coloured yarn, but actually by plying brown wool with green wool to create this effect. So many of these knots and pile tufts have both brown and green wool. I haven't looked as closely at the blues, but they are also variegated so I suppose that they might have achieved that effect the same way.

If you read back in this thread, you might see that there has been some debate about the age of my Beshir, with opininions ranging from "modern", to "Sovietic" to 19th century. What is your conclusion vis-a-vis age in the context of this "colour-mixing" technique?



P.S. I really like your "cloudband Beshir". I noticed that mine has the same treatment on the edging of the main field on the inner side of the inner guard border ("saw-tooth" with projecting flowers or some such thing).