Posted by louis_dubreuil on 09-04-2006 12:48 PM:

bothe rug

Bonjour à tous

After a quite silent august, again on Turkotek.
I have got this simple little rug for the simplicity and the balance of its design that gives to it a quite "heraldic" look. On a clear undyed natural wool ground the fields shows thirty bothes with a very simple design. The borders are also very simple and narrow.

Dimensions : 90x72 cm
Knotting : symetrical 29 Lx 26 W for 10x10 cm (754 Kn/100 cm2)
warps : two plyed S spun wool, pepper and salt, medium to coarse (up to 1 mm). Very depressed warps
wefts : not so coarse (1/2 mm), natural wool (brown, light brown, beige), one shoot
Colours : natural beige, natural brown, blue, light yellow, light red (synthetic, faded), fadded green (appears like a dirty yellow), purple red (in botom and upper borders).

The rug is low pile in the field and with some parts of long pile.
On the back the level of warps that are visible are very prominent.
In the field back the wool of the pile is worn in a special manner : the loop of the knot that is wrapped on the outer layer of the warp is abrased and cut. In those worn parts the warps appear nude, without any wool of the pile visible. The ply of wich the base of the knot is cut normally would be gone, but the knotting is so tight by the two levels of coarse warps that the plys are still all here. I suppose that, for being so worn on the back, the rug was used on a very abrasive area (stone surface with presence of sand, by ex).

I think this rug is from Caucasus, but I would like some precisions about the origine, and some commentaries about the weaving cracteristics.

Amitiés à tous


Posted by Tim Adam on 09-04-2006 04:58 PM:

Hi Louis,

Rugs with this design are known from the middle Amu Darya region. Below is another example.


Posted by Richard Larkin on 09-04-2006 11:33 PM:



Does that mean Turkoman?


It is unusual for single wefted rugs to have depressed warps, no?

Are you sure there is a single shot of weft between each row of knots? I find the rug very intriguing.

Rich Larkin

Posted by James Blanchard on 09-05-2006 12:19 AM:

Hi Tim and Louis,

It is an interesting rug. I have little to offer in terms of attribution, but your posts raised a couple of questions for me.

1. Is that fading pink-red fuschine? If so, that would probably help with the dating.

2. If it is fuschine, would that be compatible with a Turkoman origin? I don't think I have seen fuschine in Turkoman rugs, but my experience is limited.


Posted by louis_dubreuil on 09-05-2006 02:41 AM:



In fact there are two shoots of wefts, one straight, the second corrugated as it is classical in depressed warp weavings. The staight shoot is difficult to see because of the warp depression and the thin diameter of wefts compared to the coarse warps.
I do not think this rug could be turkoman, due to the use of symetric knotting on depressed warps.
About the colours I think the red is synthetic but I am not sure is was made of fushine. There are small signs of red bleeding. Only the tips of the pile are faded. The color of the not faded part of the pile is the same in the front and in the back of the rug. The part of the design that were that seems green are deeply faded and have turned to a dirty yellow. Yellow details are visible only on the back.

I think that the attribution can be made only by structure and by the use of undyed wool for the field (maybe also by the border), the bothe design being not a discriminating argument.



Posted by Filiberto Boncompagni on 09-05-2006 04:08 AM:

An interesting rug, but it doesn’t “speak” Caucasian to me.
It looks much more like the rug posted by Tim, the only trouble is the symm. knotting. I don’t think they used it in the Amu Darya region. Did they?

And… (I know you’re all going to hate me, but I do it for the sake of Science):
It’s not fuschine, nor fushine but fuchsine.

OK, OK, this is the last time I do it, or sooner or later somebody is going to tell me “go to fucksine yourself”.



Posted by James Blanchard on 09-05-2006 04:52 AM:

Hi Filiberto,

Thanks for the correction, in the name of science. I haven't heard FUCHSINE (FOOK-seen) pronounced in conversation, and the phonetics would have embedded the correct spelling. I probably subconsciously associated the word with "fuscin" from way back in my medical training.

At least if you do "fuchsine" yourself the effects will wear off with time...


Posted by Filiberto Boncompagni on 09-05-2006 04:55 AM:

Thank you, I rather prefer not to try.

Posted by Richard Larkin on 09-05-2006 09:06 AM:

I agree with Filiberto, it doesn't seem Turkoman (though I can buy Turkoman for Tim's example); and Caucasian is at best a default category. Louis, can you describe the handle or feel of the rug? And the quality of the wool. Do the selvedges seem to be original? If so, what are they?

Rich Larkin

Posted by Gene Williams on 09-05-2006 10:28 PM:


Hi Louis,

As far as design goes...does this look familiar. I'll try to do some structural analysis tomorrow.


Posted by Jack Williams on 09-05-2006 11:01 PM:

outback afshar?

Tom Cole published an article entitled "Outback Afshars" available on his web site

In it he discusses the NE persian afshar population, the sudden appearence of a set of apparently archaic afshar rugs that do not fit the norm. He notes use of cotten wefts, cotton in general, speculates on some of them being Afshari gabbahs, talks about the irelationship between some apparent baluch, arab baluch and afshar.

Based on the design and thought provoked by Tom's article, I wonder it this is an afshar gabbah? The boteh full field is one that seems afshari, but the design is so simple that a gabbah seems a resonable guess. And the afshar used the symetrical knot, at least part of the time.

Also of note, in his (Tom's) interview with Jerry Anderson, Jerry diverges from the conventional explanation of the afshari presence in east persia...or perhaps expands on it to include groupings more northerly near meshed. Jerry's explanation might work especially for a ne Persian Afshar element that, like black holes, seems to be proven to exist not by being seen, but primarily because of their influence on other groups.

That interview, read for it's brief touch on the Afshari-arab baluch element is also interesting to think about, especially since seemingly outrageous little tid bits from that interview keep turning up accurate...such as the Khazahs, converting to judism, being defeated by the Virangians, being the anscestors of the Mushwanis. Keep in mind he quoted that fact to Tom long before the internet existed, and Jerry did not exactly live his life in a library. woulld be nice if people who throw up a rug with a bold statement of fact about it would attribute the rug, and explain where it, and their, information comes from so that it can be checked out, or further research done.

Posted by Tim Adam on 09-06-2006 09:02 AM:

Hi Jack,

If you have a problem with my posts, why don't you simply ignore them? I'll do the same with yours.


Posted by Steve Price on 09-06-2006 09:10 AM:

Hi Tim

What am I missing here? I don't see any reference by Jack to your post. The only mention made of it seems to be Richard Larkin's asking if the middle Amu Darya attribution means Turkmen.

It certainly looks to me to be well within the range of what people attribute to the Ersari group, by the way. If it's yours, congratulations. If it isn't yours, it's a nice looking rug anyway.


Steve Price

Posted by Tim Adam on 09-06-2006 11:39 AM:

Hi Steve,

I think the last paragraph in Jack's post was spurred by my first post in this thread.



Posted by louis_dubreuil on 09-06-2006 02:10 PM:

more technical precisions

First : I have made an error in knot type. the rug's knots are As3, i.e. assymetric open to the left with the warped part on the upper warp (view from the front). The difference of level of the warps makes difficult the view of the path of the thread under the lower warp. For identifying the knot I have processed as follows : from the back of the rug I have selected a single red knot . This knot, in the front face has the wraped part visible on the right warp. To see the path of the other end of the thread I have pulled (on the back side) with a pair of pincers the right nod. If the end of the thread comes from the right side of the warp the knot is assymetric (this was the case), if the end of the thread comes from the left side of the warp, the knot is symetric (two ends into the warp interval). So this data, I presume, changes the possibilities.

Note fot the Society For The Prevention Of Cruelty To Rugs : after this destructive test I have replaced the loose thread with a hook)

Second : selvedges. Two cords.

Third : the handle. Coarse, due to the coarse warps and the coarse diameter of the two thread pile, but with a medium pliability. The rug can be folded without effort and without damage to the foundation. This is not the handle neither the structure (several wefts) of a Gabbeh. The wool is quite mat with no gloss and lacks of springiness ( a quite slack wool).

I hope those new precisions can help the identification.

Amicales salutations à tous


Posted by Gene Williams on 09-06-2006 02:15 PM:

A baluch in the woodwork

Bonjour Louis et - e Buongiorno Filiberto,

Ok I admit it...I snuck in a Baluch Bag one noticed.. I swear I wasn't trying to provoke anyone or claim that Louis' very attractive minimalist rug is Baluch or related to Baluch... (Now I can't speak for Chuck).


Posted by Richard Larkin on 09-06-2006 05:05 PM:


Hey gang:

Louis' rug is a real puzzler. It is one of those sui generis types, I guess. Persian knotting suggests some South Persian "tribal" production, or possibly Sultanabad area village production. These are desparate speculations.

Regarding Tim's rug and Jack's comment. My own question to Tim about his Amu Darya post was to inquire whether he thought Louis' rug was Turkoman. It seems clear Tim's rug is Turkoman (and nice). Notwithstanding the resemblance in the boteh, however, I don't think Louis' rug seems Turkoman. As far as Jack's last paragraph is concerned, I wondered what it meant when I read it, but I didn't take it to refer to Tim's post.

Gene, I knew that was a Baluch you put up there, but I didn't want to have to be the one to say so.

Rich Larkin

Posted by Tim Adam on 09-06-2006 07:43 PM:

Hi Rich,

I still owe you a response. I deliberately did not propose an attribution because the structure and the look from the back are clearly (?) not MAD/Turkmen. I'd agree with Filiberto that it does not have a Caucasian 'look' to it. I have seen this type of back before, but don't remember where. My guess is it is Persian of sorts.

The synthetic dyes seem to be very poor, which makes me think that it might be from the earlier synthetic period when dyers were still experimenting with the new synthetics. The design is clearly related to the small rug (engsi?) I posted, but I think the design has simply been copied.



Posted by Richard Larkin on 09-06-2006 08:42 PM:


You could be right about the early aniline dye syndrome. I have seen a few rugs I considered very old that had just the bad purple that fades to gray at the tips. I think of that purple as the forerunner. The crummy pinky red I think of as slightly later, but still 19th century. I don't know what to make of the green, which Louis says has faded to "dirty yellow."

I think it is a stretch to suggest the design is copied from a Turkoman source on the basis of your posted piece. The resemblance is there, though not exact (extra curl at the top of Louis' and the little projection at the bottom), and there are many versions of the boteh out there, everywhere.

Rich Larkin

Posted by Tim Adam on 09-06-2006 08:52 PM:

Hi Rich,

Last year there there was a bag, attributed to Beshir, auctioned at Rippon Boswell, which had a format/design/coloring very similar to Louis' piece. This is the one I had it mind, but unfortunately I do not have a picture of it. Maybe someone else saved the image from the RB website, and could post it for us.



Posted by Richard Larkin on 09-06-2006 09:36 PM:


It just doesn't seem to have the Turkoman feel, wouldn't you say? Even the "out of the box" Beshirs and Ersaris.

Notice I haven't gone near Baluch.

Rich Larkin

Posted by Chuck Wagner on 09-06-2006 09:50 PM:

Hi All,

No Gene, I don't think it's Baluchi. At least, not without a closer look. And Louis, is it possible to get a closer look at the back ? That one isn't bad but it's hard to see the knots.

As for the rest of you, if any of you has a copy of Neff & Maggs, I'd start poking through it; that weave looks familiar. I'm about 1800 miles away from my own.

I have no particular problem thinking that this might be from along the Amu Darya somewhere, because there are rugs from that area with fully depressed warps. I own one. That said, the first thing I thought when I saw the back was that it might be an old village piece from up in the Tabriz area.

Chuck Wagner

Chuck Wagner

Posted by Jack Williams on 09-06-2006 10:04 PM:



Help me understand declaratively tabbing the dyes in Louis' carpet as obvious synthetic, and bad synthetic at that. What charateristics make them seem so? The colors on the back seem to reflect the frontal colors pretty accurately, especially given the tanish cast the structure of the carpet gives to everything when viewed from the back. Also, none seem especially electric and the parts with the least wear, the borders, look pretty good.

Having bought quite a few carpets from pictures alone thus having considerable experience judging true colors from internet ones, and given the extreme wear in the pile of this one, I really wonder how one could be so certain of the synthetic nature of the dyes. Even though louis stated certain dyes as bad, I have trouble seeing it all from the pictures. But I am open to the possibilities...just that given my academic mindset, I would appreciate few indicators you use to so positively judge such a thing.

Lastly, I am not quite ready to concede the Afshar possibility. The "gabbahs" Tom Cole talked about in his article referenced above are worth looking at. They are speculative and labled as such...but the structure is worth a glance. The article itself is well worth the read for the information alone. But tom does describe the structure of the subject rugs...worth considering?



Posted by Tim Adam on 09-06-2006 11:58 PM:

Hi Rich,

I am not saying that Louis' rug is Turkmen. Just that the design is very similar to a Beshir design. Maybe the weaver copied from a Beshir bag. Maybe the Beshir copied the design from some common Persian source. Hard to tell.



Posted by Richard Larkin on 09-07-2006 08:42 AM:

That wasn't Gene


That Tabriz area suggestion is a good one, although I'd expect symmetrical knotting from that venue for this kind of rug. The non-saturated color and the dry wool point that way.


I took Louis' characterization of the dyes as accurate. The light red looks bad to me on the face picture, but I acknowledge the color isn't screamimg from the back view. If I am seeing the faded green Louis mentions (left side vertical row, second from bottom...?), it might bother me as well. By the way, I wouldn't eliminate Afshar here, either. That'e one of my favorite default categories.

Incidentally, I understand why Louis picked up this rug. It gets to you.

Rich Larkin

Posted by Steve Price on 09-07-2006 09:06 AM:

Hi Jack

I'd be grateful if you'd try to control your aggression a little better, and to read more carefully before popping off. To be more specific: your most recent post challenges Tim about what you call ... declaratively tabbing the dyes in Louis' carpet as obvious synthetic, and bad synthetic at that ... I really wonder how one could be so certain of the synthetic nature of the dyes. Even though louis stated certain dyes as bad, I have trouble seeing it all from the pictures.

First, if you have doubts about whether the dyes are synthetics, they should be directed at Louis, who identified them as such in the thread's opening post. Second, the suggestion that your impression of the dyes in the digital images trumps Louis' impression from having the piece in his hands smacks of a self-assurance that I find hard to accept.


Steve Price

Posted by Richard Larkin on 09-07-2006 09:43 AM:


Bonjour Louis,

If you are disposed to post any more pictures of your interesting rug, it would be very helpful to see a closer detail of the border, both side and end.

Thanks for this entertaining thread.

Rich Larkin

Posted by louis_dubreuil on 09-07-2006 10:19 AM:

back pictures

Here are three pictures showing details of differents parts of the back.

This first one shows the border region that is not worn. The visible nods are the left (for front vision) and correspond to the non wraped part of the knot. The second (right) nod is not visible from the back.

This second picture show a worn part of the back. The parts where the warp is visible correspond to the eroded/abrased nods.

On this third picture, following a warp line in the red bothe , we can see different levels and stades of nod erosion.

The warp depression gives to the back a corrugated look, this effect being amplified by the coarsness of the warps.

Salutations à tous


Posted by louis_dubreuil on 09-07-2006 12:48 PM:


Here are the pictures of borders

Posted by Richard Larkin on 09-07-2006 01:35 PM:

Merci, Louis,

In the two images of the border sections, it appears there might be some red staining on the exposed warp ends, as though from some bleeding red dye. Am I seeing this correctly? If that evidence is there, could it indicate that the rug once had a flatwoven end?

Rich Larkin

Posted by Horst Nitz on 09-07-2006 04:46 PM:

Hi all,

please, let me chip in with a few observations:

This second set of images suggest the "abbration" on the back was caused by moths.

Dyes look fine to me, that is, no synthetics.

It does not look Turkish.

It has a Turkoman appeal.

It looks Kerman in some aspects but also might be from Khorasan or from some Afshari group that had settled in Northern Iran - why?

The Afshari seem to love white or ivory ground rugs and are fond of botehs. The majority of their produce is in the symmetric knot, but according to Eiland and Eiland some rugs are done in the asymmetric knot at least down in the south. The Turkoman appeal of Afshari rugs from Kerman is something Jenny Housego has already commented on in Tribal Rugs . I would like to add that the same applies to Afshari rugs from Easten Turkey.

According to Housego, but also to Azadi, some Afshari groups settle in Khorasan. Neither of the two offers images. It also might be possible that some Afshari groups in other parts of the country assimilated to neighbouring groups and begun using the asymmetric knot, the symmetric knot being the more traditional one to them.

For an Afshari rug from Kerman I am missing the peculiar electric blue.



Posted by Tim Adam on 09-07-2006 08:47 PM:

Hi everyone,

I remembered where I had seen this particular type of back before. Louis' bag may be from East Turkestan. The coarse weave and asym. open left knot would be consistent with this attribution. What do you all think?

A (in my mind) clearly related piece is the Beshir bag I mentioned above. Although I do not have an image of the bag sold at RB, I found a similar bag in Moshkova.


Posted by Richard Larkin on 09-07-2006 08:55 PM:


Bravo!! (For coming up with that image.)

Rich Larkin

Posted by Jack Williams on 09-07-2006 11:17 PM:

I'll second Horst

Good morning all:

I’ll second Horst's comments about Afshar (again) of Louis' rug and add a bit. Here is a rug taken directly from Tom Cole’s site, his article on “Outback Afshars” first published by Hali, 117, 2001; Note the back view of the rug structure compared to Louis'. Comments below.

Horst’s comments that Louis’ rug with the all-boteh motive still has a ring of Afshari because of the design: there are many examples of Afshar use of the allover boteh design, and they also use vase motives, even poppie plants as all-over, large scale fill for the field. Here are some general examples.

The colors: the colors in Louis' rug seem to be colors used in many afshar rugs…of particular interest to me is what may be a peachy-pink in the border and some botehs..which has been described as a marker for Afshar. Even the light mottled green-yellow may be ok (I’m not sure how to change green to yellow in a rug, unless the blue dye has independently faded out. To make green, I’ve read that usually the wool is dyed blue first, then over-dyed with yellow, and from what I understand, blue usually is fast and natural, even the synthetic indigo is supposedly chemically identical to the natural dye. The other way to create green is to use iron in the mordant of wool intended to be dyed yellow. The iron will turn the yellow or gold to green, which may be what happened here). I’m happy to concede that confident identification of the color and dyes in this carpet is difficult given the wear, front and back. But if I were looking strictly at the pictures, I might decide the colors were good and they could well be colors common to Afshar weavings.

Finally, the back of Louis' rug, with the dominating brownish wefts that seems to “wander,” varying in thickness seem familiar to some Afshar carpets I have. It may be hard to see from the picture taken from T.Coles site, but the weft structure and appearance has at least a superficial similarity to my eye to Louis’. I would not be too concerned about the open left knotting. The Afshar of East Persia are widely reported to have intermarried and been influenced by the Persian villagers (see Edwards, et. al.) to the point that tribal identity has just about disappeared…though the design influence has remained reportedly reflected not only in the Turkish-nomadic remnant population but in the weavings of the surrounding villages.

Because of that, I would speculate, I repeat, speculate, a Khorasan provenance for Louis’ rug, which combined with the Turkish tribal traditions of the Afshari would explain a resemblance some see to Turkmen rugs (which incidentally is not immediately apparent to me). That there are Afshari in Khorasan seems to be generally agreed upon, though their specific location has not been pinpointed. Tom Cole refers to them as follows:

"Afshar tribal rugs present inherent problems to scholars and collectors. There is some doubt as to which of the myriad types of south Persian weavings should be classfied as Afshar, and we still have only a partial understanding of who the Afshar actually were. Parviz Tanavoli (HALI 37, 1988; HALI 57, 1991), Murray Eiland Jr. (Oriental Rugs from Pacfic Collections, 1990) and James Opie (Tribal Rugs, 1992) have contributed much to our grasp of the Afshar attribution, but the attempt to assign precise attributions to identifiable subtribes is often of limited value. However, the appearance of a hitherto unfamiliar group of weavings, apparently representing the early aesthetic of an unidentified group of Afshar weavers, means that the problem must be addressed.

"The Afshar were Central Asian Turkic nomads, part of the ancient Oghuz Turkmen horde. They eventually populated areas of eastern Anatolia and, since the 16th century, have been present in Azarbayjan (northwest Iran). In relatively recent history, Afshar tribes have come to inhabit areas of northeast Iran (Khorasan) and northwest Afghanistan, and are perhaps best known in conjunction with studies of tribal rugs and peoples from the Kerman region of southern Iran.

"After the seizure of power by the Afshar chieftain Nader Quli Khan -- who was crowned Nader Shah Afshar of Persia in 1737 -- diverse tribal groups swore allegiance and subsequently identfied themselves as Afshar. In the wake of Nader's triumphant sack of Delhi in the early 1740s, Afshar clans remained scattered from Kabul to Khorasan, living among the Afghan tribes, the Pathans, Tajiks, Uzbeks, and the tribes of the Chahar Aimaq Confederation. Khorasan itself remained under Afshar control until 1796. This patchwork of groups may help account for the confusing interpretation of later tribal census surveys by, among others, A. Cecil Edwards (The Persian Carpet, 1953). It is also unclear to what extent the Turkic peoples of the Kerman region are Afshar, or belong to other elements of the Qizilbash Turkmen Confederation, a significant power in the past of which the Afshar were at times a dominant member. We cannot therefore be certain of specfic attributions for rugs woven within this cultural milieu.”

Jerry Anderson et al also refer to Afsharis in Khorasan, explaining their original relationship to “Ersaris,” and the crossover rugs, Afshar-baluch-kurd-chahar aimaq are usually identified from Khorasan somewhere around Meshed.

What is to my eye especially unique is the minimalist nature of the design of Louis’ rug. With the extremely simple, but very narrow borders, it could be one of T.Coles “outback Afshar” and therefore could have significant age.

The only other idea I have is just a stab…Uzbek Jadagir (sp).

As far as Louis' rug being "East Turkistan?" I'm sorry to disagree. E.Turkistan rugs are generally limited to what used to be called "Samarkand" rugs from the oases of Sinjiang province, Khotan, Yarkand, and Kashgar. From a reveiw I did for my own edificaton of the rugs from these cities, none seemed to have the appearance of Louis' rug, and I haven't seen botahs in E. turkistan rugs. But then again, who knows?

Regards, Jack Williams

Posted by Robert Alimi on 09-07-2006 11:22 PM:

Here's a related example.

This image was published in ORR, Vol 15, No. 2 (Dec/Jan, 1995) in Mark Hopkins' report on the Second International Congress on Turkish and Central Asian Carpets. The related quote is "What attracted the most attention of all, however, was a striking 10-foot long "mystery piece," a pile rug whose pale camel field, multicolored botehs and depressed warps had the experts calling it everything from Bijar to Ersari."


Posted by richard tomlinson on 09-08-2006 08:03 AM:

hi louis

do you have some better pics of the front of the rug, especially the red botehs? perhaps that would help with the dyes debate.

in the main image, the colours look pretty washed out, especially the red. i would still put my money on this rug having synthetic dyes.

i like minimalist design rugs, especially ones with a single border. however, this rug doesn't do it for me because what i see are poor dyes (natural or synthetic - they are still poor)


richard tomlinson

Posted by Steve Price on 09-08-2006 08:07 AM:

Hi Richard

Thanks for emphasizing that natural and good are not synonyms for describing dyes, just as synthetic and bad are not synonyms. People lapse into using them that way, and reminders like yours come up much too infrequently.


Steve Price

Posted by louis_dubreuil on 09-08-2006 12:40 PM:


Bonsoir à tous

Here are pictures of the diferent bothes on the rug.

The colours on my monitor are quite near of the one on the rug, but maybe, as usual with digital pictures, a little bit more saturated.

This picture of the whole rug is more faithfull than the one at the begining of the thread.

We can see some areas where the red has lightly run (upper left corner, the other slightly coloured parts that appear in the field -as in botom near the border- are artifacts from the camera).

The red, as it is visible in the following bothe closeup, is pinky. The colour is quite the same on the back. Tips of the wool are faded. I don't know if this dye is natural or chimical.

The darker color on the right side is due to a shadow. The double cross motif within the bothe is of what I name grey/green and is darker and greener than the natural light beige of the ground. The bothe we can see on the right is of this colour.

The blue bothes seems to be made of indigo with faded tips (but it could be synthetic indigo)

The following bothe has the grey/green colour, and is more grey than green. I don't know if it is a faded colour or an original one. The back has the same colour.

There are bothes with diferent shades of brown or grey brown, all seem to be natural undied wool, except the purple brown that could be a natural brown wool dyed with red (I don't if this is common).

The double cross in this one is pale yellow (the yello is better on the back)

About the quality of the dying, I think that the coloured results depend for a large part of the quality of the whool and also of the quantity of its lanoline remains. I find the wool too flat, maybe because an excess of whashing before being dyed.

I have a question about the mothes and the quality of the wool. It seems to me that some qualities of wool, especially those that have a dried aspect and seem to be less dense, are prefered by mothes. Maybe because there is no lanoline left ?



Posted by Richard Larkin on 09-08-2006 04:14 PM:

Nice pix


Those are good images. I have a higher opinion of the rug in general now than I did after the original post.

The only color about which I might have reservations is the red. I can believe it is natural (the new images present it in a better aspect than the originals), and I can believe it is synthetic. It has a "bright" look about it, but not of the flagrant quality I initially feared. Incidentally, I seem to recall that some years ago, Paul Mushak published some results of his analyses that showed some reds to have been dyed twice, synthetic and natural.

The blue doesn't bother me much. The books are full of statements that synthetic indigo and natural indigo are essentially equivalent. As far as the white on those yarns is concerned, it looks less like tip fading to me and more like dye loss at various places in the fiber from insufficient dye penetration, a not uncommon phenomenon with indigo in my experience.

A question or two for Horst. What do you mean by, "It has a Turkoman appeal?" Also, would you agree that there appear to be several "types" of Afshar weaves extant, and that they may well reflect traditions that are not closely related, at least in recent decades? For example, the Southeast Persian Afshars, reasonably well understood by afficionados, and the less well understood weaves from more northerly groups in Iran? Then there is the interesting link to Tom Cole's article provided by Jack Williams. I must confess that when I read your Afshar profile as calling for symmetrical knotting in most cases, I was dubious; but when I pulled a little bag of the shelf, there they were, symmetrical knots. One must keep refreshed on these points.

Rich Larkin

Posted by Horst Nitz on 09-08-2006 05:32 PM:

Hi Folks,

Tim, East Turkestan is beyond my range of experience. However, Louis' rug also has a border that wants to be accomodated in an attribution. In other words, the border does not fit into an East Turkestan schema in my opinion.

Louis, my empathy with moths is not very high (bad experience), so I am unsure as to their most favourite diet. Those mean biests usually approach rugs from the back because this seems to be where they feel least disturbed while implanting their eggs. Eventually it is the larve that feeds on the wool. Sometimes bleached, white wool looks moth-eaten, more on the surface, but in fact this is due to brittleness. Soumaks hanging against a wall with their loose threads are most vulnerable in my experience. Now, they are often made of handspun wool with high lanolin count, whilst I would assume that natural white wool that was bleached white white, would contain less lanolin. In other words, may they fly on lanolin even more than my kids do on Hamburgers?

Richard, "Turkoman appeal" because those botehs are overdimensionally large and treated like göls almost; also the small motives in the border have a basic göl format, do you see what I mean?

I'll ask Steve to paste a few image here of another Afshari rug from Kerman province that links to East Turkish Afshari rugs from the "Yörük Triangle" and also has göl like main motives; their Holbein knot echoes in an even more stylized version in Louis' rug - the "double cross" Louis called it.

The image of the reverse side shows a structure like Louis' rug as well as the one Jack had send an image of.

The image of the border detail stems from a Hashtrud rug up in the north-west; maybe from an Afshari fraction of theirs ?

Cordially, Horst

Posted by Richard Larkin on 09-08-2006 06:53 PM:


Thanks for that image. It is most enlightening. Do we conclude from the East Turkish echoes that the Afshars in the Kerman area carried those Turkish elements with them and continue to reflect them today, or at least within recent time? No doubt, it has been mentioned already in one of these posts, but when did the Afshars become situated in Southeast Iran? I thought I read that one of the Shahs moved them forcibly quite a long time ago.

Regarding moth damage, I have always thought (based on my own experience) that the group of rugs most likely to show moth damage by a good margin was the Bijar group. I speculated that the extremely tight and heavy weave of those pieces enabled the moths to become well ensconced, and not easily dislodged, and thus able to do their damage. An unscientific observation, of course. In those pieces, the damage was most often on the pile side. However, I own a Baluchi rug that exhibits extensively the symptoms you point out on Louis' rug, with the backs of the knots cut. In my rug, the pile is intact, but staying in the rug somewhat miraculously by a kind of suspension. The warps are only very slightly depressed, and with any rough handling, I am sure the pile would fall out from the front. In the case of Louis' rug, as he mentioned, the pile is probably being held in place by the tightness of the weave in the two level warp syndrome.

I would like to make this comment about Louis' rug. Once one allows for the fact that it really does not show strong affirmative indications from any known specific venue, but rather provides evidence that is weakly suggestive of many venues, one must conclude that it could have been produced anywhere. To express it differently, every rug does not have to reflect regional, tribal, or other indicia that are generally relied on for attribution.

Kind regards.

Rich Larkin

Posted by Patrick Weiler on 09-09-2006 12:14 AM:

Offset knots


Another clue to the origin of this rug may be the offset knotting that is apparent from many of the close-up photos. It is fairly easy to spot in the worn white areas of the field and around some of the botehs.
Kurdish rugs are often woven with offset knots, but they are not the only ones. Varamin area bag faces I have studied almost universally have offset knots.
I am not aware of Afshar rugs with offset knots, but come daylight I will inspect the few Afshar pieces I have.
I have a Baluch balisht with offset knots, although it may be the only such Baluch in the entire universe with offset knots. (This is a test of the principle of rug studies that as soon as one states that a piece is unique, more of them show up like maggots on a carcass) It has symmetric knots and the warps are on one level, so it does not match the construction of Louis' piece.
Just another monkey wrench in the attribution of your rug.

Patrick Weiler

Posted by Jack Williams on 09-09-2006 11:46 AM:

Uzbek? D. Mehea et al?


I've been suggesting a Khorasan Afshar provenance for this rug, even posted reasons why. But the more I look at it the more confused and uncertain I become.

It is not so much the weave, though that inciteful observation by Patrick on the offset knotting is food for thought (actually, I don't know if Afshar carpets commonly include this feature, never even thought of it, and I am away from my own small group of Afshars so cannot check them).

But what keeps bothering me is the design. It is just too simple, especially the borders, to ring of a card carrying Afshar, or Kurd, or Baluch, or or or. Some Afshari rugs do have simple, narrow borders, but this goes to the extreme.

Tim started me thinking when he suggest "East Turkman." While it is probably not such (Khotan, Khasgar et. al., ) perhaps the rug is indeed from outside of the geographic norm....say....maybe it is further north, at least from a tribal origin perspective. I am not particularly knowledgable about Uzbek weavings...but from what I've seen of their pile articles, it seems as if their designs tend to the simple, perhaps even minimalist. I've never seen a Uzbek with botehs, but lo-and-behold, I ran a search for "Uzbek boteh" and turned up a couple of fabrics (not pile rugs) that incorporate the motive.

Also, identifiable Uzbeck populations seems to have spread down the Amu Darya into northern Afganistan, even into Khorasan. Though we rarely identify rugs from this area as "Uzbek," it seems possible that relatively isolated pockets still weave patterns that are identifiably different from the usual suspects.

Danny, from the beautiful presentation of part of your collection on Jozan (all I can say is...WOW), I notice you have accumulated or dealt with gabbahs, Uzbeks, etc. Is it possible that Louis' rug is Uzbek? The longish pile in the un-worn areas might be an indicator.

I recognize we might never solve this mystery...but it is fun and instrutive to try. Besides, I like Louis' rug. True, the dyes in its current state look funky, but I suspect it is better in person, and once was a knock-out. Makes me want a new one...

Regards, to all,

Jack Williams

Posted by Horst Nitz on 09-09-2006 04:03 PM:

Hi all

There is no offset knotting that I can see.

The images I sent are ment to demonstrate that the rugs they are from, and Louis' rug share pedigree.

The filler figures in those botehs of Louis rug are much conventionalized Holbein knots; inside the blue göls of the sure Afshari rug from Kerman are Holbein knots, only less conventionalized. Borders are also very closely related when comparing Louis' rug with the Hashtrud (Afshar) from the north - octagonal shapes with little white dots in all for corners and one dot in the middle, sometimes also white. Just to mention it, the Hashtrud (Afshari) also has a boteh design. What more can you ask for when attempting to attribute an unknown rug?

Yes, Richard, the Afshari have apparently hung on to some of their designs for eight-hundred years. According to some writers they were part of the original Oghuz-tribes that first entered Asia minor in the 12th century. Some settled in East Turkey, some in West Iran from where they were moved first by Shah Tahmasp in the 16th century, again by Shap Abbas in the early 17th century and again by Shah Nadir in the 18th century. This explains why they are scattered all over Iran - one could speak of an Afshari Diaspora. As we know, people in exile, chosen or not, often hang on to the old ways with particular dedication

Regards, Horst

Posted by Jack Williams on 09-09-2006 08:20 PM:

A bunch of Afshar examples

Well, for what it is worth, here are three groups of Afshar rugs, as identified by the dealers. All are currently or recently for sale so I will withold attribution untiil they are no longer available from the various sites (Steve, is this ok?). Collectively, they may be instructive and may help develop a concensus around Afshar, N. Afganistan-Iran Khorasan.

Posted by Steve Price on 09-09-2006 08:43 PM:

Hi Jack

We do post images of pieces that are for sale when it is difficult to find alternative illustrative material. Please (everyone) avoid comments that might bear on their value or general desirability.


Steve Price

Posted by Richard Larkin on 09-10-2006 11:15 AM:


I am interested in distinctions among Afshari rugs, especially as to Afshari production from the greater Kerman region as contrasted with more northerly regions (Khorassan?). Horst offered some comments in this regard recently on this thread. Any additional comments from those in the know will be appreciated. I'm thinking Patrick Weiler, inter alia, must have some ideas.

The link Jack provided to the Tom Cole site was helpful.

Rich Larkin

Posted by louis_dubreuil on 09-11-2006 04:19 AM:

double cross motif

Bonjour Horst

You say in your post that the device inside the bothe can be derivated from Holbein motif. I do not think it is necessary to make this sophisticated genealogy for a so obviously clear and simple geometric motif. This motif is shared by several rug making group in all the Mideast as pointed by P. Stone's book page 208. Known as "dog foot print " by persian this motif is encountered in anatolian rugs, Bergama rugs, Kasim Ushag, kurdish rugs and it is noted in an Afshari border. P. Stone apparently forgets to cite this motif in some types of Tekke ensi, especially in some animal tree ensi types field borders (R. Pinner in Turkoman Studies I, page 141). I think this motif is autonomous and takes part of the old found of stabilized weaving motifs shared by the Oguz descendents.

For all : about the Afshar hypothesis, when I compare the general Afshar style, which is quite sophisticated and have a hudge "persian look", with the simplicity of my rug, I cannot easily make the connexion. Maybe there are two kinds of rugs made by the Afshari, as usual in some weaving areas : the "best" ones, with sophisticated design (high time consumming), for the market, the others, like gabbehs, with simple design (lower time consumming), just for domestic use. This fact can explain the rarity of the second category that is also not very well documented and studied.

Posted by Patrick Weiler on 09-12-2006 01:09 AM:

ersari example

Here is an image from Mideast Meets Midwest, the exhibition catalog from ACOR 2 in 1994.

It is called an Ersari, with Ersari structural features. It is 21" wide and 14" high, with 70 asymmetric-open-right knots per square inch, with some symmetric jufti knots along the lower edge.
The design is similar to Louis' rug, but these boteh's have a hook attached and these boteh's are shorter. Both borders are a diamond design.
There are no cross-motifs in the boteh's of this piece.
I do not think this piece is related to Louis' piece any more than I think his piece is Afshar.
The squarish format is common to Afshar rugs, though.
From Marla Mallett's web site there is this information about depressed warps:
"The vast majority of workshop carpets with severely depressed warps have asymmetrical knots. Persian Bijar carpets are one exception; they are symmetrically knotted."
She does not list other exceptions. Louis' rug does not look like a Bijar, either.
And she notes this:
"Symmetrical knots are typical in Turkish and Caucasian rugs, but they also appear in some Turkmen rugs, some North African weavings, and a good many Persian village rugs."
Again the Persian village rug possibility. The design is too plain for Afshar work that I am familiar with and the colors are too flat. This could have been caused by bleaching or washing, though.
As for the offset knots, look at the third photo in the close-up pictures. It shows a blue boteh with a red cross-hatch design in it. If you imagine "hanging" this boteh from a string, the string would go up in the "crook" of the boteh. Notice that the boteh is double-outlined with brown and red. The brown knot in the "crook" is offset from the red knot above it, and you can readily see that the red knot is offset, or lies between, the two blue knots above it.
Where does this lead us?
Perhaps this is a Persian village rug with offset knotting, depressed warps and symmetric knots!

Conclusively yours,

Patrick Weiler

Posted by louis_dubreuil on 09-12-2006 06:08 AM:


To Patrick

I suppose you have missed the post in which I made a correction about knots. After further investigations into the structure it appears that the knots are AS3. This info can open the field of possibilities.

About the offset knotting I haven't seen anything like that on the back of the rug. I think that the AS3 knots and the very depressed warp structure don't allow any offsettting of the knots, because the wraped part of the AS knot is always on the upper warp (seen from the front). If the knot is shifted on the the warp situated on the right or on the left, the wraped part of the knot would be made on the depressed warp. I don't know if it is technically possible and what can be the effect on the aspect of the back.

Posted by James Blanchard on 09-12-2006 06:42 AM:

Hi all,

Just a quick note to mention that there is a similar, simple boteh rug similar to the one referenced by Patrick shown in a small rug book by Ian Bennett. He attributes it to Uzbek weavers, without details about structure, etc. I'll have another look to see if the botehs have "hooks". Unfortunately I don't have a scanner so I can show a picture, not that it would help resolve things in any case.


Posted by Patrick Weiler on 09-13-2006 12:14 AM:



You are right, I missed your post showing the asymmetric knots.
Turkmen tent bands have extremely depressed warps, yet also utilize offset knotting extensively. The knot is tied before the wefts are beaten down, so the weaver could offset the knots easily enough.
Your rug is unusual, but vaguely familiar. Bakshaish carpets often have the same nearly square shape. I have seen Genje rugs with the same coloration. I have a Turkish prayer rug with the same colors, but with no warp depression. I have a Kazak with a multitude of those little tic-tac-toe cross devices, but also with a flat back and more saturated colors.
Zakatala rugs have some features of your rug, but I do not know if any of them have a depressed warp.
It will be by process of elimination that perhaps the origin of your piece may be determined.

Patrick Weiler

Posted by Horst Nitz on 09-15-2006 06:43 AM:

Hi all

In his 19th / 20th centuries epic family saga set against a Damascene background, author Rafik Shami lets one of his figures say: “Where do you think Christendom would be today if it hadn’t been for the powerful symbolism of the cross?”
This implies, cross is not just cross. While one is highly significant, others may simply be occurrences. You, Louis, make another cross significant, the “double cross” as you call it, by proposing it had been passed on by the Oghuze. I would be very interested to see the evidence with which you want to support this claim. So far it had been my impression that nobody knows what designs exactly the Oghuze carried on their quest westward.
Others have aimed lower, Louis, and have researched the earliest available evidence
on pile rugs after the Seljuk invasion. Those rugs were collected from the mosques of Anatolia and are now kept at the TIEM in Istanbul. They date to the 13th to 15th century and show several cross forms in transitory stages, Holbein related motives amongst them, kotchak forms also, but no double cross of the kind in your boteh rug or in the border of the last rug, Jack Williams had collected on the internet, or on the woollen socks I had bought at Kars bazaar some time ago: this motive is a late development and marks the present era stage in the conventionalisation of the Holbein motive. It hasn’t always been the “clear and simple geometric motif” of which you seem to be so fond.
More a side issue, how much explanatory value would you credit a ‘dog foot print’ label with in a Muslim culture? Small version of the ‘elephant foot’ for those large Turkoman göls? We might as well return to wineglass and crab borders.
Between 1973 and 1975, Ralph Yohe undertook several field trips to Eastern Turkey, where he researched rugs kept in local mosques in the Sivas, Erzerum, Kars, Diyabakir, Adana and Kayseri areas: Yohe R S (1979) Rugs of the Yörük Triangle. HALI Vol 2 No 2, 113 – 118. Rugs from this area share a number of features, “particularly wool quality and colours, and these represent the weavings of a number of separate tribes, including Kurds, Türkmen, Avshari and other groups of people, mostly now settled, which call themselves Yörük.” It should be added, that those tribes had been living in the region for centuries (the same applies to some Armenians), if not millennia (Kurds).
Among the rugs worth mentioning in this context are figures 7, 14 and 30. Fig. 7 has a camel-colored field with lattice-like tracings and hexagons, locally called Sarkislar göls (this design I take to be an ancestor of my rug of which I had send a couple of images). Borders of this design type may show stylised 8-pointed stars (as on a rug of this type in the TM), other geometric designs or have quite elaborate floral vine designs.
Fig. 14 features a light field rug filled with large octagons in which “double crosses” with inbound ends and a square centre are set. Fig. 30 has a large ‘Holbein knot’ design of a kind as it was woven in the area at least until the 1970’s.
Further research is needed still, but enough evidence is available to make it seem a fair statement to call this area a historic ‘cradle of designs’.
The Yörük would have carried designs west to Aegean Turkey throughout the preceding centuries, where eventually Harald Böhmer and the DOBAG project around him had juvenized them; since late middle ages some Kurdish and Afshari tribes had chosen to move into areas under Persian power, away from Ottoman overlords. From there, some fractions were eventually resettled in subsequent waves to other provinces of the empire, i.e. the Kurds mainly to Khorassan, the Afshari mainly to Kerman. They would have adopted some of their old designs and techniques to their new living conditions, while they also would have assimilated new influences. This is how the asymmetric knot would have come on the Afshari, and how new designs with a distinct Persian look would have stepped next to those with the traditional Anatolian appeal.
Images of the cross or of crosses on Afghan weaves seem to have an entirely different history to the ones being discussed in this context.
There does not seem to be a single publication covering rugs by the various Afshari groups comprehensively. I have let myself been guided by Jenny Housego’s Tribal Rugs, and by Heinz Hegenbart’s Rare Oriental Woven Bags.
This is what Eiland & Eiland (1998) in Oriental Carpets have to say: “Afshari products (of Kerman province) … tend to be slightly more square than Fars rugs. In design there is a great variation, … the boteh may be finely rendered or drawn in large, crude forms, often against an ivory background …”.
It seems that quite attractive rugs with natural dyes where produced in the south by Afshari village weavers well into 20th century, before around WW II deterioration set in as well.
There is not much more to say now except perhaps for one observation I have made.
There seems to be no shortcut solution leading to a reliable identification of rugs. Coping with uncomfortable feelings of uncertainty in the process is not achieved by making it all equal, or by exotic attributions - unless we want to call it a forerun to a discussion of the rugs from Narnia and Middle-Earth.

Have fun,


Posted by R. John Howe on 09-19-2006 04:36 AM:

Horst -

Pat Weiler and I have been talking on the side about possibly doing a salon on the "varieties of Afshar weaving."

It may be that a lot of the ground to be covered is being treated in this thread, but it appears that you have good sources on Afshari history.

If Pat and I go ahead, perhaps we could convince you to work with us on the historical stuff. Please write me an off-board email at and let me know if this is a possibility for, and of interest to, you.

Pat and I have no timetable and so can put such a salon together at leisure.


R. John Howe

Posted by louis_dubreuil on 09-21-2006 11:25 AM:

double cross motif

Bonjour Horst

About what I called double cross motif, I come back to my hypothesis : I think this motif takes place in the great family of what we can name "the basic tribal motifs", family having its roots in inemorial times. For this type of motifs the problem is always the same : are they pre-turkic anatolian or are they motifs carried westward by the turkic tribes as Oguz ? The fact is that motif and its relatives are found in a wide area including anatolia.

For this area the Turkish author Güran Erbek (Anatolian motifs, from Catal Uyuk to present ) classes this motif in the "eye" category. Eye drawing is used under several shapes in anatolian weavings (kilims and rugs) with the prophilactic aim to get over evil and other unfriendly spirits that could do harm to the weaver's family. This double cross motif is not identified by the author as a cross. It takes part of a drawing group made by a square divided to nine squares (sort of magic square). The following pictures are from the cited book and show different renditions of the "eye" motifs in anatolian weavings.

In those plates the motifs 46, 59, 63, 69, 70, 71, 72, 74, 75, 76, 79, 80, 90 are good ex of this "nine squares" type's variations of the "eye" motif.

I think that this hypothesis of a basic motif is more likely than the Holbein knot derivation. Maybe this is the Holbein knot that is a sohisticated (urban) rendition of this simple tribal eye motif !

This is an exemple of the endless problem of motif origins : tribal motifs can be seen by some scholars as derivation/simplification/standardisation of urban sophisticated designs (and this seems be true for some motifs, especially Turkmen rugs and some village productions anywhere in weaving areas, even in Berber productions). If generalized I find this way of thinking very suspect as it denies any original culture to the tribal groups, just enought able to copy and transform/distort/impoverish the rich urban weaving vocabulary.
But it seems also that the great majority of basic tribal motifs have a more ancient origin, as we can find them in very old artifacts as potery and cave engravings. Some scholars speak of "neolithic" origin for this type of basic motifs. Their shape is stabilized for a long time and they make part of the decorative and symbolic fund of those traditional societies. They are employed with very few variations by following generation of weavers, even if the original symbolic signification is lost or transformed by the times. This fact can explain why the same motif is seen here as "eye" and there as "dog foot print". Horst you pointed that it was curious that a symbol could be named "dog foot print" in an islamic society. I think this is quite normal as the culture of those societies are made of several layers of historical estates, the islamic layer beeing just one of them. Here in Europe, there are several traditions and symbols that are very far of the Chritian culture. Why could it be different in Anatolia and central asia ? Names of rug motifs can have also obcure and sensless origin as they can be fabricated by rug merchants just for the market (I think "elephant foot" is one of those merchants' inventions, as "Princess bukhara" in place of ensi or engsi).

amicales salutations


Posted by James Blanchard on 09-21-2006 02:34 PM:

Hi Louis,

That's a very interesting post on design motifs and possible relationships.

I have occasionally wondered about this possible progression/regression in the borders of Baluch-type rugs. I think the progression is relevant to the "9 square" group of motifs you discuss above.

The first two are taken from Pittenger's article entitled "Prayer Rugs of the Timuri and Their Neighbors", accessible at: The first border design has been referred to by Opie as representing a meandering "S" form. I have wondered whether the border design on the second two is an evolution of that design, with greater scale and a greater emphasis on the "cross" form. The third rug is a prayer rug of mine in which I think the weaver didn't understand the layout and instead of creating the typical large scale "S", she reversed the positionin of the triangles. Any thoughts?


Posted by Patrick Weiler on 09-21-2006 10:25 PM:

double crosser

Here is the Kazak I mentioned earlier. It has several "versions" of the double-cross/eye? motif. It has no warp depression, so it is not related to the construction of the boteh piece. It does have a similar squarish size, though:

It also has "bow-ties" and "candy canes". The condition is deplorable, but the colors and impact are more than enough to recommend it as a "keeper".
And there are numerous Baluch weavings with the double cross motif in the borders and I have a pile band with the double cross as the major motif. But I do not think Louis' piece is Baluch, either!

Patrick Weiler

Posted by louis_dubreuil on 09-22-2006 09:42 AM:

nine squares square


As this motif and its relatives are widespread from Caucasus to Turkmenistan, including Anatolia and Persia, it cannot be a good tribe marker. This is the same for numerous geometric motifs, generally used as field decoration between greater medlions : we can find them in all the the asian weaving area. Botehs and borders are also too widespread motifs to make precise tribe markers

I think that the solution of this rug identification is to be found in the graphic style (size, use of natural colours of wool, minimalism style, shape and space distribution of the botehs) combined with the structural caracteristics.



Posted by Horst Nitz on 09-22-2006 03:27 PM:

Hi Patrick,

I agree, it is a "keeper" as you were putting it.

I wonder whether you have noticed that the motives inside those hooked lozenges are drawn in the same Holbein style as the red main motives in the Karachoph type II rugs, the red colour seems to be rather similar as well, figure and ground colours are reversed.

Louis, I need to think it over a little whether I want to enter into a discussion of your last post eventually. My impression is you want to have it all your way, and whatever someone is saying in a different direction, it raises your defences.



Posted by louis_dubreuil on 09-22-2006 04:33 PM:

motif discussion

Bonsoir Horst

This forum is made just for anyone who has to expose some hypothesis or theory can show and develop his arguments.
I would be happy that you show us in a more detailed manner the Holbein motif genealogy and the way it could be at the origin of the double cross motif (drawings would be appreciated). Has it is very common in subjects like rug study, the discussion remains often open, because undisputable evidences are very difficult to find.