Posted by Steve Price on 11-25-2007 01:15 PM:

Ersari(?) Mina Khani Juval

Hi People

I know we've already got active threads on tribal uses of the mina khani design, but I'd like to start a new one with a different focus. In this thread, I'm not concerned with who invented the mina khani pattern, how it got to where it is, or whether the motifs are poppies or aerial views of encampments. Instead, I'd like to take a look at the question of whether they were all produced by the same Turkmen group (I think they were). Here's one:

It is 73 inches wide, 35 inches deep, and was most likely an inch or so deeper when it was intact. It's knotted asymmetrically, open to the left, at 90 knots per square inch (7.5 horizontal x 12 vertical).

There are a number of published Turkmen weavings with mina khani fields: juvals, torbas and carpets. Every one that I've found with this field design has the same border, and those which include technical descriptions are knotted asymmetrically, open to the left. The juvals are quite large (as this one is), at least 60 inches wide. The elem design seems to be the only obvious difference between any of them in terms of design or palette.

The similarities among all of them (field and border, palette, large size, unusual As-L knot type for Turkmen) suggests that they were produced by the same group. Does anyone know of a Turkmen mina khani piece that doesn't fall within this group?

Steve Price

Posted by James Blanchard on 11-25-2007 03:02 PM:

Hi Steve,

Could you be more specific in what you mean by this "type"? Do you mean large juvals, or any Turkmen juvals, rugs or carpets with the mina khani pattern? If it is the latter, I would be less inclined to agree with your proposition of a single weaving group, unless you are prepared to stipulate that it is "Middle Amu Darya Turkmen" or "Lebab Turkmen", which could include Ersaris and/or Uzbeks, among others. I have a couple of examples that I will try to post that don't seem to fit within a common grouping, for one reason or another.

With respect to structure, beyond the AS-L knotting, can you indicate any other common structural characteristics that this group of weavings might share, such as alternate warp depression, warp and weft colours and materials, etc.?

By the way, I'm curious to know to which weaving group you would attribute the Turkmen mina khani group of weavings.


Posted by Steve Price on 11-25-2007 03:40 PM:

Hi James

By "type", I mean any Turkmen pile weaving with a mina khani field. All the ones I can find in print have pretty much the same palette, the juvals and torbas are large even by Ersari standards, and they seem to all have As-L knots. I suspect that they share other technical characteristics, but don't really know which ones.

What may be the best of the type is Plate 89 in Mackie and Thompson's TURKMEN, a juval. Thompson attributes it to Ersari, which seems reasonable but is pretty broad. My inclination is that it belongs to an Ersari (or, more properly nowadays, Middle Amu Darya) subgroup, and that all the mina khani Turkmen things I've seen belong to this same subgroup. Which subgroup is it? I have no idea.


Steve Price

Posted by Richard Larkin on 11-25-2007 04:12 PM:

HI steve,

I'm glad you included palette, because I think all of them have this range of colors, characterized particularly by the very strong (and good) red and yellow. I can't recall any offhand outside this range of characteristics.

Rich Larkin

Posted by James Blanchard on 11-25-2007 05:19 PM:

Hi Steve and all,

I'm sure everyone is getting a bit tired of seeing this carpet, but I wonder if it doesn't call into question Steve's suggestion that all Mina Khani Turkmen rugs are from the same group. Although knotted AS-L, this one is less tightly knotted than many, it has grey warps, dark brown wefts, and no warp depression. It also lacks the lattice and the white flower elements, but I would say that the drawing of the other floral elements is pretty typical, as is the palette. A noted writer and dealer indicated that he thought this was Uzbek (from the 19th century, for what it's worth), from the Middle Amu Darya region, citing the AS-L knotting and dark brown wefts.

As a contrast, remember the rug that John Howe presented on Turkotek ( John's rug is also knotted AS-L, but compared to mine the knotting is much more fine, the warps are ivory (not grey), and there is considerable alternate warp depression. So other than the AS-L knotting, there is not much structural homology between the two.

I have also personally seen and handled another small, square rug with the same design, but the palette was quite brown. Based on the palette and handle, I would not have placed it in the same group as either of these.


Posted by Richard Larkin on 11-25-2007 05:51 PM:

Hi James:

The large Mina Khani variant you posted with the accompanying corner detail: The large image shows a warmer coloring than the detail. Which is more accurate?

I mention it because, in my head, there are two somewhat distinct Ersari types (or MAD): a warm red/yellow type and a cooler red/pale yellow type. Your detail corner hints at the latter. Incidentally, I also associate the latter group with the eight pointed stars with the white dots at the tips.

Rich Larkin

Posted by James Blanchard on 11-25-2007 07:38 PM:

Hi Rich,

The second picture is a better representation of the colour of my carpet, which is why I included that picture.


Posted by Richard Larkin on 11-25-2007 07:54 PM:

Hi all,

I'm guessing James' rug is in a different line than the ones Steve is referring to. Being the variant example that it is, it might be the exception that proves the rule as far as Steve's proposition is concerned.

Rich Larkin

Posted by Steve Price on 11-25-2007 08:20 PM:

Hi James

Are the Turkmen mina khani weavings of the type represented by the one I illustrated and the one in TURKMEN the output of some specific tribal subgroup that use a characteristic palette, wove large bags, used the mina khani and that border together, and As-L knots? My survey of published examples may have just been too limited to sample enough of them to define the group crisply. On the other hand, I could be wrong. I hate it when that happens.


Steve Price

Posted by Richard Larkin on 11-25-2007 09:33 PM:


Is there a typo in the last post?

Rich Larkin

Posted by Steve Price on 11-25-2007 09:51 PM:

Hi Rich

"Represent" wound up in the sentence twice, a result of sloppy editing. I just fixed it. If there's another, it must be in my blind spot.

Steve Price

Posted by James Blanchard on 11-25-2007 09:58 PM:

Hi Steve and all,

There is another subset of Central Asian mina khani type rugs. Here is one referenced on Barry O'Connell's site, that belonged to Dennis Dodds. I have seen some others in this genre, that looked later, were much larger and looked more urban/workshop in origin. Although I think this is clearly a different group, it suggests to me that there was more than one weaving group that used the mina khani design, or a variant thereof.


Posted by Richard Larkin on 11-25-2007 10:11 PM:

Right, Steve.


That ex-Dodds is a killer, no?

Rich Larkin

Posted by James Blanchard on 11-25-2007 10:32 PM:

Hi Rich,

Yes, I think the ex-Dodds MAD rug is excellent. Vis-a-vis the subject of this thread I think it is useful to draw attention to the analogies between the border on the Dodds and the border on the large juval illustrated by Steve. I think they are related, and it is interesting to speculate whether the border design is a formalization of the approach on the Dodds type, or whether they are different interpretations of an earlier prototype. I think the latter is more likely, for some reason.


P.S. here is a close-up of the border of the ex-Dodds.

Posted by Richard Larkin on 11-26-2007 07:46 AM:

Hi James,

Thanks for the close shot. That is some rug.

Good point on the border similarities. I would think the Dodds rug is older.

Rich Larkin

Posted by Gene Williams on 11-26-2007 08:19 AM:


Hi Steve

This is the sole one of the type I have, and am not even sure if it belongs since its minus the "asterix" motif. It has very fine wool and is a very pleasant little thing; I think the board said it was a relatively late Ersari sometime last year.


Posted by James Blanchard on 11-26-2007 08:46 AM:

Hi Rich,

I would agree that the Dodds is older, but I think the borders are from the same design lineage. I am less certain that the later pieces were derived from ones like Dodds, or whether both came from another source. I would also point out that a similar type of border (with large floral designs) is not uncommon on large "Beshir" carpets, usually with a Herati design in the field.

One more thing... does it strike anyone else that the devices separating the main gul devices in the border of the Dodds rug look very much like the internal devices in Ersari temirjen guls? on the type of pieces that Steve and Gene have shown, the border looks to have a degenerated version of that.


Posted by Steve Price on 11-26-2007 08:59 AM:

Hi All

Here are a couple of images to make it easier to see the border and field details of the juval with which I opened this thread.

First, the border. Degenerate? I beg your pardon, did someone say "degenerate"? Harrumph!

Incidentally, you get a pretty good look at the elem on this one. I don't recall seeing that motif before (which doesn't mean that I haven't seen it before, just that I don't remember it if I did).

The field motifs.


Steve Price

Posted by James Blanchard on 11-26-2007 11:16 AM:

Hi Steve,

Sorry about the "degenerate" comment... I remember reading about "design degeneration" in Turkmen weavings somewhere. Wait a minute... now I remember... the author was someone named S. Price.

Steve, actually I like this format and like the example you have shown, and the elem does seem unusual to me as well.

Do you agree that the two borders are related? Do you think there are signs of design degeneration in the border of yours and others like it, or do you think they are they apples and oranges?


Posted by Richard Larkin on 11-26-2007 11:38 AM:

Hi all,

You may recall in the opium poppy thread that Bob Kent posted a Kurdish Mina Khani he had scanned out of Mumford's book. In the text accompanying the plate, Mumford says that this essentially Kurdish design was "copied in Khorrassan and Turkestan...." Writing that in about 1900, I wonder what his source was, and what kind of timeline he was thinking about. He also claimed that the copyists did not achieve the fluidity in the drawing of the design the Kurds did.

I know we aren't supposed to be speculating about origins and derivations (degenerations?), but I wonder whether the appearance of this pattern in Ersari weavings represents a longish development, or whether it sprang up among them, almost in the nature of a weaving fad. When Steve put out his proposition that maybe all of the Turkoman Mina Khani rugs were from a single Ersari type source, one could feel the doom coming on of the odd exception. But certainly, most of them seem closely related and in line with Steve's notion. Furthermore, they seem to have the feel of the Beshir types of weavings that were apparently very prominent commercially in the local area in the 19th and early 20th centuries, judging from picture books of various khans and the like around Bokhara in that period.

Rich Larkin

Posted by Richard Larkin on 11-26-2007 11:52 AM:


Commenting on your question, I don't know what to say about the relationship between the two borders. What does strike me is that both the border and the field of the Dodds rug seem to be moving away from the norm; particularly the field, if one is to take the Kurdish examples as the sources for the Turkestan "copies." There appears to be a closer match between the medallion elements of the design in Steve's juval, for example, and the 19th century Kurdish models, than between the latter and the Dodds piece. Yet, that one seems almost certainly old among the examples we are seeing.

Rich Larkin

Posted by Steve Price on 11-26-2007 12:17 PM:

Hi James

I'm sure that the border designs are related at some level. Whether the more common one or the version on Dodds' rug is further along in the process of degeneration or whether both are more or less equally degenerate is probably unknowable. Anyway, when people talk about degeneration in tribal weavings, it's usually taken to mean that some traditional elements are modified in ways that suggest that the weavers lost the meanings that those elements had in the past. That's how we avoid defining every change as degeneration. Neither the mina khani design nor any of the borders we've seen in this thread are likely to have terribly deep roots in the Turkmen traditions.

It seems unlikely that the Turkmen who wove these things invented the mina khani, and the pieces that suggest an Ersari subgroup could have all been done in the same urban workshop. There's nothing about them (designs or sizes) that suggests that pastoral nomads were the weavers. Juvals that are six feet wide and three feet deep make more sense as furnishings in a stationary home than as portable storage bags.


Steve Price

Posted by Gene Williams on 11-26-2007 12:59 PM:

Are you Sorry Arsari are Ersari?

Hi all,

The impression is the Ersari (more often spelled Arsari in non-rug literature) are one monolithic tribe and they move around. They aren't and they don't. I have access to that excellent compilation of works on the tribes of Afghanistan Adamac’s “Historical and Political Gazetter of Afghanistan,” 6 vols, Graz, Austria, 1972 and reprinted to 1990 (which compiles a great deal of comments from many travellers/observers of Afghanistan over 150 years). Just to put the tribe and its location into perspective, I’ll quote a bit from it and give a listing of the clans, all from Vol. IV. You can decide whether its worthwhile.

Before we get into it though: Note the existence of two Kizil Ayak sections of the Arsaris below. Note all the existence of a Chobash Clan. I think I will be the first to state absolutely that many of James’ best carpets are Arsari, Kara-Bakaul Clan, Bark division (for obvious reasons).


“The Arsaris are now wholly concentrated on the Oxus and in districts adjacent to that river. They are a remarkably orderly people, exceedingly industrious and admirable farmers. They make carpets and felts, but the former are far inferior to those of the Sariks and Takkas (sic). They invariably live in owehs (yurts), which are even better than those of the Uzbaks (sic), and delight in going out into the chol with their flocks for a month or two in the spring. They seem to possess few horses and few camels, and to be entirely devoted to agriculture, sheep and cattle farming or production of silk. The dress of the Arsaris is much the same as that of the Uzbaks, only quieter in colour…It appears that there are no Arsari Turkomans anywhere but on the Oxus and in its vicinity.

“Tribal divisions: The main divisions of the Arsaris are as follows: Ulu Tapa, Gunesh, Kara and Bakul. The last appears by many to be a subdivision of the kara and is often spoke of as Kara-Baraul. Members of all the clans and sections live intermixed; …but the holdings of each subdivision are sufficiently distinct and segregated to be irrigated from a separate channel.

“Location: Primarily live along the southern bank of the Oxus along nearly the entire length of Northern Afganistan….It is said..that on the right bank the Uzbak element is giving way to the Turkoman, to which it is inferior, it would seem, in vitality, industry and tenacity of purpose. There appears to be little doubt at least in the province of Afghan Turkistan, that the Uzbaks are falling back..while the Arsari population is advancing and spreading and both banks of the Oxus, from Kilif to Chaharjui, will no doubt soon be an exclusive Arsari possession…

If you’d like, here are the major Clans and divisions of the Clans:

…..-- Breli
…..-- Chaharshangi ***fodder for the Charshango/Chaharshango vs Kizal Ayak adherents *** (but hold the cigars (opium?) and champagne (kurdled mare's milk)-see below)
…..-- Chai
…..-- Chatmak
…..-- Nasimili
…..-- Tabashurli
…..-- Tazargan
…..-- Togachi
…..-- Wachal


Chobash *** (Richard take note)***

Chodar (different from the Chaudor)

…..-- Agi
…..-- Ark Bator
…..-- Chukkur
…..-- Ikbash
…..-- Kawak
…..-- Kizilayak - ****note the existence of Kizalayak division of Gunesh clan of Arsari***
…..-- Koinli
…..-- Kubasakhal
…..-- Lumma
…..-- safe Bai
…..-- Suleiman
…..-- Umar

…..-- aranji
…..-- Bakawal
…..-- Bark *** James favorite Clan/Division ***
…..-- Chakash
…..-- Chopar
…..-- Dali
…..-- Harum
…..-- Inchka
…..-- Islam
…..-- Kannaj
…..-- Karabainli
…..-- Kazan
…..-- Koresh
…..-- Lumba
…..-- Mengajak
…..-- Pandji
…..-- Paraj
…..-- Perrik
…..-- Kara
…..-- Tagan
…..-- Titan
…..-- Ugam
…..-- Yezark

Khoresh ***(I don't believe David is a Turk name)






Ulu Tapa (Uluqtepe)
…..-- Abdal
…..-- Achi Beg
…..-- Akadri
…..-- Chatrak
…..-- Chaubash
…..-- Danaji
…..-- Ibdal
…..-- Itbash
…..-- Karaja
…..-- Khatab
…..-- Kizil Ayak *****Note second possibility for Kizil Ayak*****
…..-- Maqkri
…..-- Misar
…..-- Sukhi

Yaz Kuli

Yaz Murad.

There may be more Clans...Anyway, have fun…wait until the European carpet merchants find out about this.


Posted by Steve Price on 11-26-2007 01:20 PM:

Hi Gene

By my count (probably off by one or two), there are 72 known possible candidates if I want to believe that some Ersari (or Arsari) subgroup is the source for the mina khani things like the one with which I opened this thread. Narrowing it down should be easy - there isn't a single Mushwani variant on the list.

Steve Price

Posted by Gene Williams on 11-26-2007 01:32 PM:

The second Grand Unification

Hi Steve,

Remember what JA claimed..that Afshar really should be spelled Afsar and that they are closely related to ... you know who...and that brings in the Kurdish interaction. Grand unification? or Grand consipiracy?

Gene (I am in my bomb shelter tonight..peace brothers)

Posted by Richard Larkin on 11-26-2007 01:55 PM:

Hi Gene:

Do you know when the quoted section was written? In what time period were these things being said of the Arsari?

BTW, it is these endless lists of tribes and subtribes, and names that seem like other names, ad infinitum, that make me despair of making any real sense of the paltry collection of terms customarily employed in the attribution of rugs and trappings.

Rich Larkin

Posted by Steve Price on 11-26-2007 02:01 PM:

Hi Rich

I agree - it greatly exceeds my attention span. On the brighter side, arriving at the realization of how hopeless accurate attribution is has freed my mind - now I can just enjoy the the aesthetics.


Steve Price

Posted by Richard Larkin on 11-26-2007 02:10 PM:

Hi Steve,

As to the aesthetics, I note that Gene's source opined that the carpets of the Arsari are "far inferior" to those of the "Sariks and the Takkas." Maybe they were. I've seen some really nice Arsari rugs, though; and some big ho-hummers from the Sariks and the Takkas. I guess it depends on the rug.

Rich Larkin

Posted by Gene Williams on 11-26-2007 02:12 PM:

When written


My impression is that the book and others like it draw a great deal on writings from the late 19C and especially from about 1910 to 1938. But some of those themselves were compilations of earlier writings. Adamec of course published his Gazeteer first in 1972. The one I have in the office (its late here) is 1985. I'll look tomorrow. But its my impression from spellings and such that its dated to about 1925, possibly to the "Military Reporton Afghanistan..1925.

Whatever, the list of clans should bring us back to earth on variants of a design.


Posted by James Blanchard on 11-26-2007 02:22 PM:

Hi Steve,

I agree that the mina khani design is probably a relatively late addition to the design repertoire of the Turkmen weaving group(s) that adopted the design. Hans Konig suggests that these designs that he attributes to Persian influence came rather late to the Ersari/MAD weaving groups. Dodds' rug doesn't look very late though, so perhaps there were some of these designs kicking around the region earlier.

I do think the border is perhaps the most interesting and potentially informative aspect linking these weavings. The motif that looks very much like a component Ersari temerjin gul element appears to be much better articulated in the Dodds piece, and "degenerated" in the others, if one accepts that it is a design that has some authentic tribal roots in Turkmen weavings (i.e. in the Ersari temerjin tribal guls).


Posted by Gene Williams on 11-26-2007 02:43 PM:

Uzbek or Ersari


I never get tired of seeing that Ersari/Uzbek carpet you posted. It just plain tickles the anti-establishment fancy and I can see from it why JA would start thinking about the random pattern of stars. Note the above mixture of Uzbek and Ersari in N. Afghanistan along the Amu Darreyh/Oxus...which might explain the confusion between the two.

I'd also bet that Steve was right about the design..that commercial "Beshir is hot" currents came down from Bukhara and voila, variants began to be woven especially by the sedantary A/Ershari and maybe the Uzbeks. Re-read Edward about the late 1890's rebirth of the Persian carpet industry, the rise and fall of the "Shawl" designs, etc. We've already established that Turkomen were marketing their textiles in Bukhara and Samarkand and probably other market centers since at least the early 1800's. And its known that huge caravans left the Central Asian Oases bound for Russia and other places, 5,000 camel loads at a time..back into the 1500's. I assume that carpets were part of the loads.


Posted by Steve Price on 11-26-2007 05:30 PM:

Hi James

I suspect that most of what we think of as the Turkmen design vocabulary was in place 300 years ago, probably much earlier. If that's the case, the mina khani could have entered their vocabulary any time since then (say, 1750 or 1800, for instance). That would be pretty late for a new entry into their vocabulary, but not many people would call a Turkmen rug dating from before 1850 a "late Turkmen weaving".

I'm a little surprised at the repeated assertions that the Dodds rug is the oldest piece in this thread. It might be, but I don't see much evidence except that it's beautiful.


Steve Price

Posted by James Blanchard on 11-26-2007 06:54 PM:

Hi Gene and all,

First, Gene, stay safe!

An article in the NERS Newsletter (October 2005) has some interesting insights from Erik Risman and Peter Poullada related to MAD weavings. They suggest a divide between West and East bank MAD weavers, with the former ("West Bank") producing more traditional Ersari tribal designs and the latter ("East Bank") having more external influences and producing the mina khani and other Persian-influenced designs.

Steve, I agree that we don't have much to go on to confirm that the Dodds rug is the oldest. However, I have seen some carpets that have a similar design that appear to be late 19th or early 20th century, and they seem to be substantially more recent than the Dodds if we consider the drawing, etc. Not much to go on, but I still think that the Dodds carpet is likely mid-19th century or earlier and I somehow doubt that the others in this thread are as old or older. But that is just conjecture.

In terms of a somewhat tangential analogy for my carpet, below is a rug sold as part of Sotheby's "Thompson Sale". I found it on Barry O'Connell's site ( Note the similarity in the main border with segmented 8-pointed stars within multitoned squares.


Posted by Jack Williams on 11-26-2007 08:39 PM:

Here is the truth according to M. Inak Ani

Steve, et. al.:

Below is one of the three Turkmen rugs I own with variants of this design. This one is 2’6” x 3’8”, with variable color brown (dark to natural looking) weft, double twist brown-mottled warp not depressed, Asy open left knot, about 11 x 8+, h x v, about 90 knots per inch. The dyes look pretty good and there are indications this was hung on a wall. I think it turn of 20th C. or so.

I’ve been interested in these Turkmen mina khani designs for some time. And of course the Turkmen m-k design has been used in all sizes of rugs from utilitarian Khordjins to palace-size carpets.

But...maybe we ought to define the category. In my opinion, in order to be called a true “mina khani” rather than some varient flower-lattice, I propose that the Turkmen mina khani design needs to have three types of base or primary flowers and usually a fourth “floating” flower (I call it that because with its white color contrasting with the darker base-flowers, it seems to "float" adding depth that often is what makes these weavings so interesting, at least to me).

Here are prototypes of the four usual types of flowers.

Even on the same rug, these flowers frequently vary somewhat in color combination and design. But the Turkmen m-k will almost always include a ‘british flag’ flower (#1), a ‘diamond’ flower (#2), a ‘double hump’ flower (#3), and usually a depth- offset blossom… [usually] 5-petaled white ‘floating flowers’ (#4), though these are not required.

Steve, I agree with one of your points, that these mina khani designs were likely woven by a single “group.” However, I disagree with several of your other contentions.

1. There was no use of a consistent border design at all and the design of the border is not an indication of age;

2. There is not a requirement for the use of a consistent knot;

3. There is no consistent structure or size.

In my opinion the “group” responsible for weaving this pattern was the “Lebab Turkmen” on the M.A.D....Ersari Yurt weavers settled on the right bank of the Amu Darya and influenced by the commercial desires of Bukhara and the Uzbeks.

Lebab Turkmen is a term I first encountered reading Vambury. Later a good case for their existence was made by Peter Poullada, see: . Because this “Yurt” or group had varied ethnic backgrounds including refugee groups from Salor, Saryk, embedded with Ersari, and goodness knows what slaves, I doubt that one can establish this group based on structure... but defining the group by the field design makes sense.

I suspect the m-k design was probably a fairly late comer to the Turkmen repertoire...say late 18th C. at earliest. [note: Craycraft has written that the Turk mina khani design is an offshoot of the standard Turkmen gul-and-secondary-gul design and that it is a derivative old Turkmen design. I disagree.] To my knowledge, use of the design was pretty much confined to the M.A.D. and the polyglot bunch of refugees that were called Ersaris that settled in that area.

One reason I say it was a late comer is that the design did not seem to evolve or change very much during its weaving life and its use is not evident in other Turkmen groups, Tekke, Youmout, Saryk etc. Therefore, it seems reasonable that the design arrived in Ersari camps after they were safely in the M.A.D. I believe that the use of the design began late 18th C. at earliest, and pretty much ceased to be woven in the 1920s-30s, about the time of the forced collectivization of the Turkmen and the mass migration into Afghanistan.

Here is another chuval I own. This one is, 3’4” x 5’9,” no depressed warps, Asy open left knot, different border, elem, etc. It has a stenciled number on the back indicating it was imported into the Netherlands probably in the 1930s. I think it is considerably older. Unfortunately it is pretty worn and has some moth damage. It has similar warp and weft structure, knotting is less fine, maybe 10h x 7v, or 70 kpsi.

Here is a long rug, one of my favorites. It is 2’9” x 6’3”, green silk highlights, knot Asy open left, 11+v x 9+h or about 100 kpsi. It also has the three cord selvedge overcast in brown wool. The double wefts are brown, no offset warps, warps are double twist, ivory and/or a natural gray with flecks of brown.

Finally, here are some two more examples to illustrate that design variations in the border, elem, even in the base flowers, etc., but especially in the little floating flower. I don’t have the structural information on these items. But you can that first one of these has 4-petal white floating flower, and the second rug no floating flowers at all. I have also seen the floating flower represented as a bud, not yet unfolded – and still regret not buying that chuval.

Summary. I think Poullada’s writings define this particular design well as being sourced in the M.A.D. It likely first was introduced after the Ersari arrived in the late 18th C., and possibly originated from Kurd weavings, or other Persian sources. The design was limited to being woven in the M.A.D., and possibly upstream into Afghanistan, by Ersari groups.

After the flight of the M.A.D Turkmen to avoid forced collectivization, I believe weaving this design declined and eventually almost stopped completely. Therefore, most of the items with this design are likely to be pre-depression and many considerably of the few "design tells age" rugs in greater rugdom.

Re: Border: I have collected pictures of about 50 different Turkmen m-k designs. While the border on Steve's rug is echo'd quite a bit, there are quite a few distinct borders. I doubt border design has much to do with age or geographic provenence, though it might indicate some sub-group origin. Perhaps I'll make up a collage of borders.

Regards, Jack

Posted by Steve Price on 11-26-2007 08:58 PM:

Hi Jack

First, a clarification: I don't think I said that I thought the border was a criterion for age. If I said something that could be interpreted that way, it wasn't my intention.

My initial suggestion that the border on the first piece shown was associated with the Turkmen mina khani field was based on a small sample of published specimens, and is obviously not the case. I don't know that I've seen that border with any other field design, though (this should bring forth a volley of examples showing that border with every conceivable Turkmen field).

The one with which you opened your last post is small enough to make me wonder whether it belongs in the same group at all. I'm also reluctant to lump together a lot of different rugs and bags on the basis of field design alone. Size, knot type, border design, selvage, all vary. That doesn't suggest a small weaving group to me although some subset of these may be the products of such a group.

I do agree that the mina khani field is probably a latecomer (after 1700, most likely after 1800) to the Turkmen, since it appears to be restricted to Ersari-like weavings.


Steve Price

Posted by Patrick Weiler on 11-26-2007 09:32 PM:

Early Bird?

This piece was advertised in Hali 109, spring 2000, by Asad and Dawn Kahn.
It is described as:
"Arabatchi "group" torba face, with a unique interpretation of the Mina Khani design. Asymmetrical knots open to the left, camel hair warps and wefts. Finely knotted, well saturated dyes. Circa 1800 or before. 15" x 54".

It fits the description, with three flowers and the white "floating" flower as noted by Jack Williams. It has the "badam" border. There is no discernable elem.
Steve, I assume you have tested your warps and wefts to see if they are camel hair?
It falls in the early range of your expectations.

Patrick Weiler

Posted by Steve Price on 11-26-2007 10:00 PM:

Hi Pat

I don't think mine has any camel hair. The foundation is wool, except for a few cotton wefts.

I tend to be skeptical about date attributions, especially early ones, and I'd be surprised if the criteria used to place that one at the beginning of the 19th century (or earlier) are based on anything solid. I'm skeptical about the Arabachi attribution as well, just because every mina khani design Turkmen piece seems to be so firmly within the Ersari group. Is this the only Arabachi piece - indeed, the only non-Ersari Turkmen piece - with a mina khani field? The border on it looks mighty Ersari-ish to me, too.

Pretty bagface, though. That means it's very old, right?


Steve Price

Posted by Richard Larkin on 11-27-2007 08:03 AM:

Hi Patrick,

What a piece! Sure looks old to me. I agree with Steve, the Arabatchi attribution is hard to figure.

Rich Larkin

Posted by Bob Kent on 11-27-2007 08:21 AM:

precise tribal attributions and dates

I agree with a few of steve's comments on this thread: there are so many groups of people, and there is so little clear evidence of who-made-what weaving, that precise tribal attributions are probably as much fantasy as fact. In the baluch world, the "A Look at the Word Tribal" ORR article by Eiland (it's on the web) points out that Karai/Quarai attribution was made without a really clear group of weavings being identified (a lot of baluch rugs have depressed warps and/or 4-cord sides, or both). More importantly, the rugs were attributed to a group that is known to exist but for which there is no evidence that they wove _anything_. And that "attribution" is still used. He also points out that the mina khani design on some of the baluch rugs comes from older persian weavings and probably had no tribal significance.

similar situation for dates. ...

Hali 42 page 17: "It is not difficult to come to believe that we can date most turkoman rugs with a fair degree of accuracy, when we are really following, and in turn supporting, a dating system which is conventional rather than one based on evidence. The fact is that we have sufficient evidence to date fairly accurately most - but not all - turkoman carpets made in the last 100 years. We can make more or less intelligent guesses at the dates of a fair proportion of pieces made 50 years before that but, with rare exceptions, we have no evidence on which to judge the date of rugs made at the beginning of the 19th century or earlier."

This is a really old peice. How do you know? It has the XXXX design features that we like. How do you know that this signals great age? The old ones have it....

I think that with so many small centers of production, rugs that were made at the same time now recieve different datings. For all I know, two rugs made in the same village would now creatively recieve different tribal attibutions...

But because it is easier to ask $1500 for a "late 19th century" "arab quainat jamshidi rug" than to ask for $1500 for "one of these baluch rugs with knots open right and the vine meander border, 1920s for all anyone knows" this probably won't stop any time soon.

Posted by James Blanchard on 11-27-2007 08:56 AM:

Hi all,

I think that this is the third mina khani piece that has been attributed by someone or other to the Arabatchi. Remember John Howe's "Central Asian Attribution Puzzle"?

He was told by a dealer that he thought it was Arabatchi, primarily because it had a long, thin open-left knotting technique. The other reference I have seen was a juval with a classic mina khani design on a prominent dealer's site. I haven't posted a picture of that one because I am not sure if it is still for sale.

A couple of comments...

Risman mentioned that the MAD pieces with the badam (or "judor") border were usually "left bank" pieces, which denotes the more typical Ersari tribal pieces. There are now three pieces on this thread with the mina khani design, which I don't associate with the more tribal designs, that have the badam border (mine, John Howe's and this latest "Arabatchi" shown by Patrick).

With regards to age, I agree that the piece shown by Patrick looks "primitive", and the colours do look great. However, I am not sure whether "primitive" translates into old, or just that the weaver was a bit inexperienced with the design. Whatever the case, it does look quite old for some reason. This brings me to my last point. When we first discussed my mina khani carpet on Turkotek someone suggested that the badam border was a feature of later "Afghan" weaving. At the time I was skeptical, and I think that some of the other examples I have seen with the badam border (including Patrick's example) have firmed up my viewpoint on this. I think the badam border was used in the MAD area, and probably on some older pieces. Any thoughts on that?


Posted by Richard Larkin on 11-27-2007 09:31 AM:

Hi James,

I don't know whether I'm one of the suspects in the promulgation of the "badam border = late" doctrine. I do think it tends to predominate more in later Afghanistan pieces than it did in earlier pieces. As far as the alleged "Arabatchi" Patrick posted is concerned, there seems to be a telling difference in the drawing of the border relative to most examples. The angle of photography looks oblique, so I can't tell very well how different the drawing really is. I think the drawing is less hackneyed in your example, too.

I have to take the time to reread your link to John's post of two years ago. It looks interesting.

Rich Larkin

Posted by Sue Zimmerman on 11-27-2007 02:14 PM:

I am quite confident that the Turkmen subgroup who first wove mini kahani designs came from, or at least spent weaving time, in Ushak.
Designwise, to see this, you can follow the five petal flowers to locate and access the source of these later translations, most efficiently, in the famous medallion Ushak carpets. That's where the first "flowery" version of the mini kahani design, at least that I have seen, were woven. It's pretty easy to see. Sue

Posted by Robert Emry on 11-27-2007 10:49 PM:

The same but yet so different

Hello overyone:

Here's a mina kani piece that appears to fit one of Steve's criteria---it's big--17 inches high and 73 inches long.

But what is it? I doubt that it was ever the face of a bag, but more likely a trapping, perhaps a camel trapping/asmalyk. So if it isn't a bag face, then maybe it isn't so big---maybe normal size for what it is. The knot is asymmetrical open right. Nice colors, and since it's mine, I believe it is really old (except for the top and bottom bindings, which were done in the machine age). Any thoughts?

Bob Emry

Posted by Steve Price on 11-28-2007 05:39 AM:

Hi Bob

It looks like there's a strong family resemblance to the Arabachi(?) piece from HALI that Pat Weiler posted:

The palettes, dimensions and border are very similar, although the knot types aren't.


Steve Price

Posted by robert emry on 11-28-2007 05:27 PM:

Additions and corrections!

Hello again:

When posted last evening, I neglected to mention another oddity of the long trapping (images in posting above)--it has a very limited amount of white cotton knotting---5 knots in the center of each of the larger rosettes (the diamond-shaped ones with the apricot color), and a few of the little white spots in the guard borders.

Then today I discovered that this trapping is even odder. Before last evening I had thought this piece has asymmetric knots open left. But before I posted last evening, I checked again to be sure and found they were open right. But today I discovered I was right both times. It is mostly open right, but about the middle third on the left half is open left. The open left knots are within the bracketed area on the new images here. This suggest to me that at least the middle third of this piece was woven by two women sitting side-by-side, one knotting asymmetric open left, and the other open right. What might this say about tribal affiliation?

Posted by James Blanchard on 11-28-2007 08:36 PM:

Hi Bob,

I like that piece, whatever it is! The light blue border gives it some additional visual punch and is unusual.

Could it be that the section with the AS-L knotting is a good and old repair (patch)? I note that there is some discontinuity in the drawing of the border in that section.


Posted by Richard Larkin on 11-28-2007 09:52 PM:

Hi Folks,

Terrific posts and images from y'all. I think Steve's original thesis is largely intact, in that most of these seem to come under the general M. A. D./Ersari umbrella. The last two images are striking, particularly on account of the sky blue borders. Yet, I get the impression that the one Patrick showed is a much different fabric than Bob Emry's. Maybe it only looks that way. We've waged the camel hair wars before, and I won't inquire how M. & M. Kahn knew there was camel in their piece, but I wouldn't imagine anyone would take the foundation materials of Bob's to be camel hair.

Rich Larkin

Posted by James Blanchard on 11-28-2007 10:06 PM:

Hi Rich,

I am not so sure that Steve's thesis is intact, unless he stipulates that he meant that these pieces come "under the general M.A.D. / Ersari umbrella". Based on my readings, there were a number of weaving groups in that region, including Ersaris, Uzbeks and Arabatchis (according to some). There have been suggestions through this thread that these mina khani pieces could have come from any or all of these groups. Am I missing something? Other than the AS-L knotting, it does seem that there was a range of foundation materials used, and variability in the presence of alternate warp depression. So I am not convinced that we even have a consistent structural group, especially by the usual rigidity of Turkmen structural classifications.


Posted by Richard Larkin on 11-28-2007 10:29 PM:

Hi James,

I get your point. Of course, it all depends on what is meant by "the same Turkmen group." There is a general tacit acceptance of the notion that it means a tribe, as though all the major names were intact and got the same special cut of meat at the big feasts as they got in the 11th century. And yet, it appears the tribal identity of many of these people, particularly Ersari related people, had become attenuated by the 19th century, especially in this region. It is by no means clear that relatively slight differences in "structure" break down on lines of tribal affiliation or ancestry for these weavings. I haven't even read the M. A. D. material, but I get the idea the region was a conglomeration of ethnic groups, with many of them apparently weaving in a commercial milieu. Perhaps I was construing Steve's point (erroneouosly) that the rugs all seemed to come from this context.

Rich Larkin

Posted by James Blanchard on 11-28-2007 10:36 PM:

Hi Rich,

If the concept is the "group" is defined by being not Tekke, Yomud, Salor or Saryk, then I would agree. I suppose that they could all fall within the Ersari category, considering the diversity of their weavings by the late 19th century. I was only pointing out that there were structural differences in some mina khani examples that would be considered significant in the general Turkmen context, and that others had attributed some of these to Uzbek and Arabatchi weavers in the M.A.D. region.


Posted by Sue Zimmerman on 11-28-2007 10:54 PM:

Hi Bob,
What were originally the slightly depressed warps mistakenly were pulled forward which pulled the knots tied after that wefting mistake into the opposite direction, too. The left end of the top white marking line in your last photo points out that boo boo. If you start looking around too much you will have the chance to be right quite a few more times.
It's usually a good thing when two people are working together on very unwieldy projects, such as this type of weaving, for each person to strictly keep track of what the other one is doing. How often that actually happens probably has something to do with this format not being undertaken all that often ever. It wouldn't surprise me if all of them were really old. Sue

Posted by Robert Emry on 11-28-2007 11:05 PM:


I'm certain that the section with asymmetric left knots is not a patch and not reknotting. There are one-knot discontinuities in the left borders that coincide approximately with the changes in knotting, but not exactly, and the shifts in the main and minor borders are not all on the same row of knots. And there is also a one-knot shift in the right border, where knotting is all Asym right.

With the machine serging of the top and bottom edges (I intend to remove this), it is hard to see much of the warps, but where warps can be glimpsed between knots and wefts, the warps appear to be a mix of mostly ivory wool or hair with a minor amount of darker fibers. Wefts are brown, wool or hair. Nothing that could be camel hair.

Posted by Steve Price on 11-29-2007 05:38 AM:

Hi Guys

Just to clarify: when I opened this thread, my notion was that the mina khani Turkmen weavings were done by some fairly distinct subgroup of what we generally lump into Ersari or Middle Amu Darya. I've modified my view to the suspicion that they were workshop products made in that region, many coming from the same workshop. It may not progress much beyond suspicion. There's a lot more variation than what I found in my initial cursory survey of the ones I turned up in print.


Steve Price

Posted by Jack Williams on 11-29-2007 08:07 AM:

mad about MAD minas

James, Steve, et. al, Re-reading these two references about the M.A.D. Ersari, has helped me get comfortable with diverse structural characteristics co-existing with a single weaving cultural design ethic. These references together seem to explain a lot, at least to me.

Elena Tsareva, on the Ersari – especially the M.A.D.

and Peter Poullada

Bob That is a beautiful and enigmatic weaving. I love the way the base flowers are interrupted by the border. I have not seen that endless feature in other Ersari mina khani weavings. Re: the use of both open-left and open-right knotting This feature is known, and in Baluch rugs it may be more common than has been reported. Here is an example of an Ersari prayer rug in which the change from open-left to open- right, then back to open-left, is obvious [the "bottom" of the rug is actual the "top" for weaving purposes...the change in knotting offsets straight lines].

And in the NERS exhibit of the Baluch collection of Mark Hopkins linked here , there are several rugs that are amazingly open left on the left side of the rug, open right on the right side. See: . The rugs so identified (#s 2, 12, 14, and 21) are really worth a look. Given the small size of those weavings, one could easily conclude a single weaver changed the knotting method purposefully.

The picture (above) is of two “torbas.” The second one [it's cut down...but is about 59" long as is] is the one that looks to be similar to the two previously shown with the badam border…but the well known dealer offering it simply calls it an Ersari-Beshir mina khani torba, not Arabatchi or Martian, or whatever. These “torbas” are the only smaller weavings I can recall having both the mina khani design and the badam border. There are a couple of larger rug-carpet examples with the badam border, but no chuvals that I’ve seen.

The second torba is included to transition to a discussion of borders in these mina khanis.. I’m pretty sure Steve is right… the border on his chuval is the most common one found on mina khani chuvals (at least from my collection of pictures, gathered for the last couple of years). I’d guess about 60 percent or so of the Ersari mina khani chuvals have a variant of that border.

But…there are differences in a lot of the weaves that are really charming. Here are four chuvals with more or less that “ “A” border… The border on the second chuval is unusually eye catching as some of the border elements seem to “bleed,” intruding over the borders boundries, creating a cool 3-D effect.

Steve raised a question whether this ”A” border was ever used in conjunction with other field designs. I’ve looked, but haven’t found it used with anything but a mina khani. I think I may have a picture of a Bijar Kurdish mina khani with a very similar border. I’ll post it later, or in another line because I’m not sure it of its provenance or whether it is germane to our discussion.

There are other borders used in chuvals. The second most common that I’ve encountered is a variant of a common and simple Turkmen, Ersari design, the kochanak pattern, or "B" border. Here are three examples of its use in chuvals. And to show that there are other borders used in chuvals, note the beautiful mina khani with the “C” border, with a totally different design element.

We all know the mina khani design is not limited to chuvals or torbas. It is found on rugs and carpets, ranging from small to huge ones of palace size. The borders of mina khani rugs and carpets seem to have a good deal more variation than those of chuvals. Even the borders that more or less echo the most common chuval “A” border are artistically quite individualistic. Here are some examples of “A” border modified and used in rugs and carpets.

And here are two examples of the use of the kochanak pattern ("B" border) Ersari type border complete with badam minor borders, used with many other field designs. For some reason these two rugs look a little more modern to my eye...perhaps it is simply the use of mulitiple borders.

Here are some borders of some ornate rugs and carpets that are similar to mina khanis in the field design yet are missing some key elements. I propose that these type of rugs [of which there are a remarkable number published…perhaps because of their ornate sophisticated beauty] be excluded from the mina khani category. Their borders show evidence of being ornate derivatives of the “A” border…but in these carpets, borders and fields are embellished and modified to the point that their kinship to our mina khani design seems pretty distant and limited to abstract conceptional kinship, at least to moi. These borders do tend to link the carpets to the Ersari, because of some of the design elements.

Finally, there are some rugs and carpets with a seeming true mina khani design in the field, yet with borders that are not particularly associated with other m-k designs. Rather they seem to reflect somewhat more common non-mina khani Ersari type borders. Here are two examples, my long rug, and James’ room size carpet (James, Jourdan's book has several main Ersari carpets with badam borders that are attributed deep in the 19th C.)

I hope this helps us think about the possibility of applying some organization to the types of these mina khanis. I am tempted to speculate that the common “A” border and it’s derivatives are right-bank M.A.D. woven for mostly urban Uzbek taste and consumption. I wonder if the more simple “B” border and possibly the “C” etc. borders of the chuvals could be left bank, products of more nomadic and wilder Ersari? Where the use of the badam minor border would fall, especially on those “torbas,” is a question.

I’m pretty convinced that the Ersari mina khani design pretty much ceased to be woven in the late 1920s, about the same time as the demise of the “Beshir” designs. The reason is probably the world-wide rug depression during WWI and the Russian civil war that followed…and the mass migration of Ersari Turkmen groups to Afghanistan to avoid forced collectivization. Too bad… because these are beautiful designs.

Regards, Jack Williams

Posted by Steve Price on 11-29-2007 08:30 AM:

Hi Jack

Here, for everyone's amusement, is the elem on my piece and what you refer to as the B Border:

Looks like they have a little common DNA.


Steve Price

Posted by Sue Zimmerman on 11-29-2007 11:51 AM:

Hi Bob,
The one-knot discontinuities which coincide with the knot changes you are seeing speak to the boo boos I pointed out in my last post. This is a type of error that could not go unnoticed by the weavers. That they continued merrily along for so long points to the care in weaving indigenous to poorly supervised commercial interests.
This is not meant to be a comment on aesthetic but personally I prefer the Oushak medallion carpet's whole story version to the "Ersari", etc., mini kahanis edited one. That's just my opinion. I am not singling your weaving out as being commercial. I think all mini kahanis are. I can't see what difference which side of a river they were made on matters or how they have anything to do with anything of tribal importance. Maybe I'm just getting jaded. I'm beginning to think there is no such thing as "Tribal" weavings. Sue

Posted by Jack Williams on 11-29-2007 02:48 PM:

more minas


re: carefull..asking such questions (is anything out there tribal?) is to put one foot into the world of Baluch, a world like no other... A trip down the yellow brick Baluch boulevard is one from which few return unchanged, a veritable twilight zone of rugdom.

Steve et. al.

On the border of your chuval, I think perhaps that outside border you reference is actually the "top" of the chuval. The other end is where another border mirroring the one on the top, and the elem would probably have been. Still, I thoroughly understand the visual connection between several of you put it, the "DNA" seems to match.

I've made up a few more pictures...might as well dump as much data as possible. Strange how few images there are of Turkmen mina khani rugs on the net. I have collected quite a few through the years..but find one of my files with some great stuff is missing. I hope it wasn't on the Katrina-computer!

Above are a couple of more chuvals... a little on the pedestrian side or maybe just bad pictures. Both have variations of the"A" border. The top one though has a significant deviation from the standard Turkmen m-k theme in the field. It may not even be Turkmen!

Here are close-ups of the chuval borders, what I called "A", "B", and "C" - (wow! what imaginative imagery those names evoke). In the first group of "A" borders, Steve's chuval is in the upper left. Note the blue line along the bottom...which I think is the top of the bag. That is usually a marker for "top" in most Turkmen chuvals.

And here are closeups of a couple of chuvals with the "B" border. The minor side border of the first one is interesting in it's own right. Also included below is a closeup of that unusual chuval with the odd Turkmen border, ("C").

If I find my missing file, I'll post those as well. I hope everyone does so that a single repository will have a record of as many of these beautiful things as possible.

Regards, Jack

Posted by Steve Price on 11-29-2007 03:38 PM:

Hi Jack

If you're right about my piece being upside down, there would have originally been at least another three inches in the depth (assuming that what I think is an elem is actually a border and that it was at the top and at the bottom), plus the depth of the elem - typically four inches or more. In its present state, the piece is 6'1" x 2'11", so it would have to have been an enormous 6'1" x 2'6", maybe even bigger. Like an engsi for the portal to a 2 car garage (I'm told that the autos of the Middle Amu Darya around 1900 were those flattened sporty things).


Steve Price

Posted by Jack Williams on 11-29-2007 11:25 PM:

chuval size is normal..."ahhh but what is 'normal' grasshopper?"


I am reluctant to gainsay the moderator…but your chuval really is probably missing the Elem. The part of your chuval with the little “extra border,”... the pseucho kochanak device and blue line... is most likely the top of the bag. Most Saryk or Salor chuvals had that blue line. That little “extra kochanak-like border” occupies a design space above the main border that seems to be included on many chuvals in this mina khani group.

Here is what your chuval very possibly looked like before it lost it’s elem….

As far as size goes, don’t be concerned that the chuval will be "too big" if you add in the missing elem. Yours will look more balanced and it's size will be "normal" for the genre. Note, chuvals with the “A” border often seem to have two design layers in the elem. “B” border (kochanak pattern) chuvals do not seem to have that feature.

To illustrate the point about size, here is a 19th C. example from Jourdan, p. 298. Note the size…yours would be pretty close to the Jourdan chuval if it had the [probably] missing elem. On the Jourdan chuval, notice the "two design elements in the elem" and the fairly wide design space filled with different symbols, above the top main border. Even without the elem, yours is still pretty awesome if a little ragged on the bottom. The details of your border are really nice.

Also, here are two more examples of chuvals with the “A” border. Unfortunately I don’t have the sizes of these two chuvals…but I suspect they are both in the same class as Steve’s and the one from Jourdan. The first is from JBOC's site some time ago...I can't find it now.

The second one I found on this NERS Newsletter site a while ago. This particular newsletter contains an interesting and good discussion of the M.A.D. written by Yon Bard, but covering a presentation by Eric Risman. Several truly unusual Ersaris including this mina khani are shown in color in the newsletter. It is a link well worth a look. See: .

Jack Williams

Posted by Chuck Wagner on 11-30-2007 12:13 AM:

Hi all,

I've don't have much to contribute to this discussion other than to note that this particular motif, and the border design like that on Steve's (and other) pieces continued to be produced in and around the villages in the Imam Saheb area, north of Kunduz, right up to the time of the Soviet invasion. I would guess that there might still be production now, but I don't know.

The rug designs are called after another village in the area, Baba Sediq. This area is really a little too far upriver to be called Middle Amu Darya; there's not that much left once you get northeast of Kunduz but the uppermost reaches of the Amu Darya. Maybe we should start calling them not-lower-Amu-Darya pieces.

Parsons covers this area in his book on Afghan rugs. He also has a few pictures of earlier 20th century Afghan Turkoman pieces including this Mina Khani chuval:

It looks complete to me; I doubt there is any missing elem on this one. The quality of the drawing is quite good, I think.

A quick review of Jourdan's book on Turkoman rugs shows that the Ersari seem to have created quite a few long, narrow pieces with aspect ratios similar to Bob's.

There were almost no Yomud or Tekke counterparts; there were a few Saryk & Salor. It's a bit of a reach to draw conclusions from this, as there may be no relationship between the statistical spread of pieces in the book and the totality of Turkoman weaving.

Either way, ther Ersari wove a lot of big torbas. Here's one I own that is almost six feet wide (no, it's not a Mina Khani). It's not as narrow as Bob's but it's a big one:

Chuck Wagner

Chuck Wagner

Posted by Steve Price on 11-30-2007 05:38 AM:

Hi Gene

I'm not offended by having mistaken the top for the bottom on mine. I hadn't paid any attention to the blue lines, and am unaccustomed to seeing Turkmen borders with no "frames" around them except in elems. I like it anyway. As a point in passing, the well respected dealer from whom I bought it told me that the orientation was the one you pointed out. He was right, as he usually is.


Steve Price

Posted by Jack Williams on 11-30-2007 08:54 AM:

Steve, the close up of your border shows a very nice weave and pattern. I think you have a winner.

Here is one more small rug, 5' x 3'11". It is for sale so I will not comment about it, except that the border is unusual...a kochanak ""B" border" type with some archaic and interesting adds.


That is some chuval you posted. And the torba...what an absolute beauty. the difference between the worn out rags I have accumulated and the beautiful carpet items you have collected is striking. I agree about the length...the torba fragment I posted earlier has been cut down...but is 69" long!

The information that the pattern continued to be weaved until 1979 is interesting. Have you seen these newer types? Do you know who is weaving them? Do they stay true to the genre or have they evolved? I ran a search on the internet but only got geographic information.

Thanks, Jack Williams

Posted by Lavergne Philippe on 11-30-2007 09:02 PM:

Two more exemples

Juval: 170 x 98 cm

Ensi : 135 x 165 cm

What do you think about the larger one?
Is it an ensi, as I believe, and according with the dimensions?

Have you ever seen this large white and yellow border before? made, as you can see, with the elementary parts of the frequent design find at the bottom of many old juvals.

Bonsoir et bonne nuit !


Posted by Jack Williams on 12-01-2007 06:05 PM:

nothing comparable

Lavergne, that chuval is great..the wool looks lustrous and the design is crisp and well balanced. I really like the contrast in reds..the blood red of the base flowers and the lighter red of the field. It causes a double take at first...but is very striking. I might comment on that further later.

But the other little rug just leaves me staring with open mouth. I've never seen any thing quite like it...the size, color and pattern, and the intracate detail of the border is new to me. I love the field of course...and the border seems to be entirely composed of known elements. But whoever wove this was a real innovator and artist. Looking at the field with that thick border is like looking out of a window.

And the size and shape... this doesn't seem a normal Ersari but has more of an Afshar characteristic dimensions. What a rug.

In your little rug, I try to see a hint of the way its border was created reflected in the border of my long rug. The wider than normal size of the border, and the pattern created out of several known motifs combined together in a way that thickens the design. Well, maybe...but it might really be a stretch to compare the two or classify them together.

There is probably nothing about your border and composition that I've seen before or is comparable on an Ersari, or other Turkmen that I've seen. Congratulations.

Do you have any history on this rug? I would love to see some close ups, also of the elam area.

Jack Williams

Posted by Lavergne Philippe on 12-02-2007 01:40 PM:

Dear Jack,

I tried to do my best to show you close-ups of my 2 carpets, with possibility to see the back.

The pictures are made with a flash, because it is a rainy day in Paris. The field colours are not quite the same, the juwal is a little more brownish..

About the long carpet already shown (and with pictures named MK) the border made with small coloured squares is said by my afghan friend from the Kunduz area.

The story of the carpets?
The juval came by an usual way, Afghanistan, Peshawar, etc…

I bought the other one from a nice man, a fairly experienced rug dealer, with a family concerned by rugs for generations, in a big town of the south of France. He had it from an old person with a large family home, and things here for a long time.
In this south, Marseille and Provence, many people, I guess, were involved by Orient. Dealing, diplomacy, and so on.. (For exemple a family was staying in Cairo at the beginning of the century, and went back with carpets, etc).;
That is all I know… as usual.. We can dream… Things talk only with our knowledge, and there is no archives for that. From time to time, a few records…

About Ensis made by Ersari, you can see 6 exemples (N° 27 to 32) in Gols and Guls by David M. Reuben 1998.
They are not very larges 144 x 120 to 200 x 158 cm
One of them is a mina khani ( 185 x 134) with the lowest border of flowers ( as on the mine)
The lattice of the field is not made, there are only the 3 big motives. No little white flowers.

Bien amicalement à tous


Posted by Jack Williams on 12-06-2007 05:51 PM:

great balls of fire...

I would like to deal with some of the nuts and bolts that have surfaced in this line. Let me start with the subject of Arabatchi attribution of mina khani weavings.

Patrick W. posted pictures of an interesting torba attributed to Arabatchi Turkmen taken from a magazine, circa 2000. After looking on the net, I found what appears to be the same item, sold on Sotheby’s in April, 2004 for a significant sum. The Sotheby’s lot description of this item was “Ersari Torba, mid 19th C.”

The colors of the single picture of the Sotheby Sale may have an echo of the pallet of Arabatchie weavings. But unfortunately, there seem to be no details of why the attribution later fluctuated back to Ersari. It may simply be different dealers and their private estimations.

Looking further on the internet, on a well known site I found two other mina khani designs attributed to the Arabatchi by Mr. Craycraft. Both may be for sale so I’ll limit my comments. The first was attributed with a question mark.

Though it is one of the ornate decorative types that I prefer to exclude from our mina-khani definition, I posted some pictures because the border has a feature seen in some rug-carpet mina khanis (but not to my observation, chuvals)...the little line of “balls” used as a minor border. Here are three rug examples of the use of that same minor border (see three pictures below)

I think I’ve seen similar borders on Kurdish mina khanis, but not often on any other Turkmen weavings.

The next item, also attributed to the Arabatchie by Mr. Craycraft, is a chuval that at first look seems similar to those we have been discussing at length complete with the "A" border, though some of the design details seem different, such as the barberpole vines. It also may be for sale so I won't comment about the details of the chuval.

I confess I do not know what to make of these Arabatchi attributions...or what characteristics make a weaving an Arabatchi rather than an Ersari. But absent some documentation, I personally don’t see a necessity at this time of creating a new or sub-tribal mina khani category.

To my mind, the standard Ersari Beshir attribution works well with history, geography, and ethnography. To extend the pattern to the territory supposedly occupied by the Arabatchie would open questions about design dissemination.

Jack Williams

[add-edit: Since posting this, I've gone on Michael Craycraft's site and cannot find a reference to either of these items. A resonable conclusion would be that the pictures posted on another well known site are old news. I think we can assume that they are no longer for sale and can therefore be discussed.]

Posted by Jack Williams on 12-07-2007 09:50 AM:

minor point major pain

This is another minor point that I thought should be mentioned.

There is a very good archived line that definitely documents the use of stenciled numbers on Turkmen weavings imported into the Netherlands in the 1930s. Through records, it has proved possible to identify when certain rugs so marked were brought in and how much they were sold for.

In this line, three mina khanis have stenciled numbers on them. They differ from the stenciled numbers in the archived line in that they have no letter preceding them.

However, the two that I own were bought from dealers in the Netherlands. So it seems plausable that the stenciled numbers could indicate a 1930s importation date into Europe. In the case of the Canadian Textile Museum chuval that is attributed to 1880-1890(presumably by experienced people), the importation date would have been several decades later than the attributed weaving estimate of age, showing at least some of the items stenciled were not new production.

What would be interesting is to discover who was doing the collecting, on the ground, in Turkmenistan, and marking these with stenciled paint? Certainly a considerable number of items were so marked. Food for thought?

Posted by Jack Williams on 12-07-2007 10:02 AM:

Open left! open right! stand-up sit-down, fight! fight! fight!

And a mina khani good morning all.

Killing a few more rats that scurried out of this line...["killin rats" is a rural southern expression for tending to minor chores or details]

This is another comment intended to enlarge on a topic that was touched on. It might provide some food for thought.. and maybe another topic for discussion in detail at another time. It is NOT an attempt to subvert this line into a Baluch discussion [though now that I think about it, the Baluch mina khanis seem to fall into two or possibly three catagories…. later.].

Earlier, Bob Emory showed a "torba", with a particularly interesting and beautiful mina khani design. Here it is again for reference sake.

The item had a portion of the weaving that exhibited asy open-left knotting while most of the torba had asy open-right knots. I suspect the explanation of that anomaly simply lies in the use of two different weavers…as in the Ersari prayer rug also shown.

But, it was noted that Mark Hopkins’ fabulous collection of Baluch [see: ] included four rugs that have “asymmetric open left knots on the left side, and asymmetric open right on the right side.” The rugs and bags with this feature are generally so small that one would suspect a single weaver deliberately changed the orientation of the knots. Here are those rugs.

Data from NERS Exhibit

Hopkins, NERS exhibit, #2
Size: 1’10” x 1’10” (56 cm x 56 cm).
Warp: Natural white and brown wool – Z2S.
Weft: Brown wool – Z2S. Two shoots.
Pile: Asymmetrical, open to left on left half, open to righ on right half* h: 6 v: 7 kpi: 42.
Colors: (12) dark blue, light blue, blue-green, dark aubergine, light aubergine, brown, brown-black, yellow, pale green, pale red, burnt orange, white.
Selvages: 2 weft-wrapped double warps.
Ends: Weft-faced plainweave with weft substitution.

Hopkins, NERS exhibit, #12
Size: 2’10” x 4’2” (86 cm x 127 cm).
Warp: Natural white wool – Z2S. Slight depression.
Weft: Light brown wool – Z2S. Two shoots.
Pile: Asymmetrical, open to left on left side; open to right on right side.* h: 7 v: 7 kpi: 49.
Colors: (7) dark blue, medium blue, rose red, chestnut red, dark brown, camel, white.
Selvages: Missing.
Ends: Missing.

Hopkins, NERS exhibit,]#14
Size: 2’5” x 3’7” (74 cm x 109 cm).
Warp: Natural white wool – Z2S.
Weft: Brown wool – Z2S. Two shoots.
Pile: Asymmetrical, open to left on left side, open to right on right side.* h: 8 v: 9 kpi: 72.
Colors: (11) dark blue, medium blue, pale blue, red, dull red, yellow, dark brown, medium brown, gray brown, camel, white.
Selvages: Not original.
Ends: Missing.

Hopkins, NERS exhibit, #21
Size: 2’2” x 3’0” (66 cm x 91 cm).
Warp: Natural white wool – Z2S.
Weft: Brown wool in various shades– Z2S. Two shoots.
Pile: Asymmetrical, open to slightly to left on left and slightly to right on right.* h: 7 v: 7 kpi: 49.
Colors: (8) dark blue, medium blue, light blue, red, dark brown, medium brown, camel, white.
Selvages: 4 warp cords weft-wrapped with brown goat hair overwrap or in some areas with replacement wool.
Ends: Narrow weft-faced plainweave; rest probably missing.

*italic emphasis added

The existence of this "odd" feature on this many Baluch group weavings might cause us to wonder... 1. are there are other Baluch with similar features?...2. is this more common than we are aware of simply because no one has looked closely?..and 3. why? Unfortunately, Mark’s exhibit doesn’t include detail shots that might aid speculation why such weaving would be done in this manner. Still…it could be an interesting question…possibly akin to whether the Baluch took corrosion into account in their designs.

Jack Williams

Posted by James Blanchard on 12-07-2007 10:10 AM:

Hi Jack,

In the previous thread that addressed stencilled numbers I presented an Ersari small rug that I purchased in Pakistan, from a fellow who acquired it in Afghanistan. So some of those rugs seem to have made their way back to Central Asia.

Another couple of points on this. First, the local fellow in Pakistan seemed to be well aware of the dating of these pieces, since he told me "this piece is at least 70-80 years old because that is when they were painting these numbers on the rugs". Vincent Keers was the Turkotekker who informed all of us about this.

Second, there is at least a bit of evidence that the relative value of different types of weaving has changed in the last several decades. Vincent Keers was able to dig up the details of two of the rugs... my Ersari and a Tekke torba belonging to Marvin Amstey. The buying and selling price of mine in the 1930s was 3-fold higher than the Tekke torba. Although I do like my little rug, I am quite confident that today the relative value of the Tekke would be justifiably several fold higher than my Ersari. Maybe they were paying by area back then...


Posted by Sue Zimmerman on 12-07-2007 11:24 AM:

Yes, Jack, there are more boo boos in commercial weavings. Sue

Posted by Gene Williams on 12-07-2007 02:29 PM:

Paying by Square Foot

Hi James,

My experience with retail dealers is that they always bought by square foot (or square something); wholesalers also sold by square foot. The price could vary based on "fineness" but rarely based on art or age. (for instance in Karachi in the mid-70's we would pay 5 ruppees (then $.50 sq ft) for lowest quality..up to 50 ruppees (then $5.00 sq fot) for highest quality.

That's why I still negotiate with a seller based on square footage...and may explain the difference in price in 1930 for your Tekke and Ersari. Edwards has a marvelous proverb in his book from some Persian town I recall which went something like this...Father: "Son get the yardstick." Son: "Which one Dad, the one for buying or the one for selling?"


Posted by Gene Williams on 12-09-2007 03:53 PM:

Windsor and the Baluch

Hi all,

This has nothing to do with this excellent mina-khani Ersari line. Jack posted some examples from Mark Hopkins' collection There is one in the collection which is haunting.

It recalls Windsor's famous bag. The problem of course is that Windsor's bag had a cotton base and "hammered" wefts. Still, the design is close, right? Back to the Ersari (who are not a Baluch tribe).