One of the difficulties of this salon is that many of us have had little direct experience (that we know of) with non-Turkmen Central Asian weavings. So many of us may feel that there is little to say.

So one question we might ask is what seem to be the indicators that are used to identify such pieces.

I've begun by listing a few that seem to me often to be used by the experts.

Indicators of Non-Turkmen Central Asian Weaving:

1. One weft per pick (also called "single-wefted") (exceptions: some very finely woven pile Tekke mafrashes have only one weft between rows of knots; and some Turkmen tent bands of full pile or mixed technique also have this feature)

2. Knots tied only on raised warps in each shed. (Although, some Turkmen tentbands have this structure.)

3. Rugs sewn together from narrow strips rather than woven in a single piece.

4. Similar to Ersari weavings but different in some respect (e.g., wool "feels" different)

I invite others to add to this list or to question or revise the indicators I have listed here.

In this way we may assemble some views of how such weavings are distinguished from those of the Turkmen tribes. Notice that I am only asking about a dichotomy here, not about closer attributions.

Regards,

R. John Howe

5. It seems to me that a number of non-Turkmen Central Asian pieces have three or more picks of weft between rows of knots.

6. A number of them also have pile that is longer and sometimes much shaggier than do even the deepest Turkmen pieces in full pile.

Regards,

R. John Howe

In a post in another thread, Vincent Keers says in part:

"...Do you know, why they made the attributions as they are? Was it on colors?..."

Vincent, first notice that Jim Blackmon carefully said at the beginning of his presentation (and he put it specifically into the salon essay as well) that he doesn't want to claim to be very certain of the geographic attributions he has provided since the experts in the literature disagree.

This is a general situation with regard to Central Asian textiles. Kate Fitz Gibbon, the wonderfully knowledgeable student of things Central Asian, spoke before Jim Blackmon and in her presentation said that the designation "Lakai" is being used far too broadly and that many items so designated were made by weavers in other groups.

The same complaint has been made for some time now about the large western Turkmen tribal designations "Yomut" and "Ersari." These need to be broken down and while some modest successes have been achieved (e.g., the "fine-brown" Yomut grouping), there has not been much progress to date.

But to answer your question: I think these geographic attributions have been made on the usual bases, by looking at a number of indicators for several that point in similar directions. But geograpic attributions of this material is a very tricky business and as the "pickers" have traveled to more and more remote areas, non-Turkme Central Asian pieces these are reaching the market in greater numbers and often adding to the confusion.

There was a time when a non-Turkmen pile rug with one pick of wefts between rows of knots was thought likely to be Kyrghyz and this is often still the case. But analysis has shown that there are Uzbek and Khazak pieces (and perhaps some from other Central Asian groups as well) that have this same "single-wefting." And there are some Kyrghyz pieces with two picks of weft between rows of knots. So that indicator has become more tenuous.

So while it is agreed that more precise geographic attributions are needed, the bases on which they are to be made are still not clear. Hence, Jim Blackmon's caution.

If you had questions about a specific piece, we might be able to discover the basis on which the attribution provided here was made.

Regards,

R. John Howe

I hope it will work.

Is it knotted like:
2 warps/one knot + 2warps/no knot etc?
Or: 3 warps/one knot so the middle warp is free?
I can't figure this out, because I understand offset knotting is used in some of the rugs. This makes option one difficult.

Asymmetrical knots seem to be out of the question with these kind of constructions if the pilehight is normal. Because then the "last" open pile from the knot would ly down to easy. Am I right?

Best regards,
Vincent

This is a very important two-part question you have asked. I am sure you asked this to clarify the situation for many people who are somewhat startled by the unusual appearance of the backs of these weavings. So,

Part one:

In a "normal" rug, the weaver ties a knot around two adjacent warps, then another knot around the next pair of warps, and the same process is repeated across the width of the rug. THEN she lifts every other warp thread. This opens a "shed", into which she places a weft shot. So far, this is a "single wefted" rug. If she then lowers this set of warps and lifts the other warps, this opens a different shed. When she places a shot of weft across this open area, she now has a "double wefted" rug. The back of the rug shows no warp threads because, by lifting alternate warps to insert the weft, the "upper" set of warps is covered from being seen at the back of the rug by this weft thread. When the second weft is inserted, this thread covers the warp threads that were left visible by the previous weft thread. She then ties knots around each adjacent pair of knots and starts the procedure over until the rug is done.
This usually requires a couple of double-tall-non-fat lattes and some chit-chat.

Part two:

The way most pile tent bands ,and some of these rugs, are woven is a bit different. Instead of tying a knot around two adjacent warp threads, the weaver first lifts or raises every other warp thread as though she were going to insert a weft. But instead of inserting a weft, she ties a pile knot around every adjacent warp thread of only the warp threads that have been lifted or raised up. If a weft is inserted at this point, only warps without knots are visible from the back of the rug, because the row of warps with the knots is above the weft thread. Then she lowers this set of warps and raises the other previously unknotted set of warps and ties a knot around every adjacent pair of warps in this set of warp threads. A weft inserted now "hides" the knots and only the unknotted warp threads can be seen from below. The back of the rug now will only show a small amount of pile colors that peeks from between the wefts and gives us their characteristic appearance.

This technique obviously places each knot "offset" from the knots of the previous row of knots, because they are not lined up like little soldiers on the same columns of warp threads from the bottom of the rug all the way to the top. It is as though the knots zig-zag their way up the rug. This does leave one "empty" warp thread every other row all the way up the side of the rug, though, but those sneaky weavers have ways of compensating for that.

Partly yours,

Prtrick Weiler

On this

I was a little puzled. But part two makes it all clear. In this way the symmetric knots are "rubbing" shoulders, creating a sort of zigzag close fabric at the front side. (The left side of the next knot, fits in between the sides of the previous knot)
So, every knot needs 3 warps.
How many of the rugs in the salon, share this fabric? All of them? A couple? John says symmetric, single wefted, triple wefted, but now it seems that all rugs are triple warp knotted.

I'm ignorant, not dealing with these kinds of fabrics on a regular basis.
Getting more and more jealous.

Appreciate your sound explanation,

All yours,
Vincent

Good probes you are making. And Pat's responses seem very much on target to me.

Couple of additional thoughts.

First, I'm not sure that it's correct to say that each there are three warps involved in the symmetric knots that are used in this "knotted on alternate warps" structure. Two raised warps are wrapped with a symmetric knot as in conventional weaving and a third (unraised) warp does lie behind the center of this knot. But this third warp is not wrapped by the pile knots at all and forms part of the plain-weave back. This, I think, is why Marla says that this is not a "jufti" usage. In jufti knots, as you know, the extra warps are encircled by the pile knots.

Second, this "symmetrical knotting on alternate warps" is the primary structure of a number of these rugs in Blackmon's presentation but not all of them. I'm doing this from memory and may need correction but I think the pieces in this presentation that have this structure are 2, 3, 11, 12, 14, 15, 16, 17, 19, and I think, 26. I am not sure about 24. Blackmon said that its structure is "distinctive" and so it may be something else, but the back does seem faint.

About the numbers of wefts. Some of these pieces have only one weft between rows of knots. But some also have three or MORE wefts, sometimes many more. So I miscommunicated if I seemed to say "one-weft or "three." It's one weft and three or more.

About the ignorance you are feeling: the group you are in is large. None of us sees this kind of material much and although we could handle it after Blackmon's presentation, most of us have not much more experience with it than you say you do. This is why someone as experienced as Wendel cautioned me that he would not have much to say in this salon.

Such rugs are appearing more frequently in the market but they are still pretty unusual weavings to encounter. And at this level of quality, I would say that they are rare.

Regards,

R. John Howe

So if I understand it well.
The weaver raises 2 warps, leaves behind 1 warp, etc. So it's like
2.1.2.1.2.1 etc. and the next row is 1.2.1.2.1.2 etc.
This isn't what I learned from (partly mine) Patrick. Patricks way creates zigzag and offset knots.
With the 2.1 etc rhythm it's "impossible" to create offset knots.

I only hope people aren't bored with this.
But the Chinese seem to have invented offset knotting and I think the colors of these rugs point to China. But if no offset knots are available because of the structure, this is hard to follow.

Best regards,
Vincent

ps. Patrick. If you use Spell Check, your name can be found correctly spelled.

No, it not knots over 2 1 2 1 warps, etc.

Its (in an open shed, with alternate warps raised and the other have at a lower level), tien knot over two raised warps and continue doing so across the rug, insert weft, change sheds, (now a different alternate set of warps is raised). Again tie symmetric knots over each of the two raised warps in order. Insert weft. Close shed. Now the original set of alternate warps is raised again. Repeat.

It might help to go back and read Marla's description of how this weave is made. I quoted it in one of these threads.

The fact that there is a lower warp in the center of (but behind) each knot tied on the alternate raised warps does not mean that when the shed is shifted that there is only one warp to tie to. The weaver ties on the two contigous warps that are raised each time. That's how the weave alternates to form what I think may legitimately be called a species of "offset knotting" although not of the more usual kind.

Hope that helps,

Regards,

R. John Howe

Ohhh.....

So it's
2.0.2.0.2.0 etc.
0.2.0.2.0.2 etc.
2.0.2.0.2.0 etc.
This creates a lot of space inbetween the piles.
Doesn't look like offset knotting.
Looks like lazy work.

Thanks John,
best regards,
Vincent

Marla's site helped a lot.
It's
101010101 etc.
010101010 etc.
The 0 being the warp that goes at the back of the next knot in the next row.

Couldn't anyone tell me this from the start.
Or is my English to poor.
Think it's the alternate? that does it.
But maybe it's my brain, not functioning properly. One is in need for some mindspring once and a while?

Hope some other non English readers, had the same problem.
But no w're geting to the bottom of it.
So it's offset knotting all the way.
Chinese back in the picture.

Thanks,
best regards,
Vincent

English is the problem.
Reading back, I mentioned 3 warps, one in the middle free. Post 5 "Or..."
But because you said no..and I wanted to save the offset knotting in this story, I came up with 2.1. etc.

Next time I'll make a drawing.
Or maybe we should use a more nutral language like Esperanto? (Russian didn't work very well)

Never mind, I really appreciate your patience, because I sometimes think you're all pulling hairs out of your sculls, because of the text you're reading.

Best regards,
Vincent