Posted by James Blanchard on 07-17-2007 09:24 PM:

Two weavers, one rug

Hi all,

In a separate thread on Jaf Kurd bags Steve Price showed one that was obviously made by two weavers; one that was not able to render much of a design and the other that exhibited some experience to finish the job. Here is a picture of that bag.

I have a south Persian rug with which I have recently become re-acquainted and in so doing, I have discovered that it too was likely a product of two different weavers. The transition between weavers is not as obvious as in the above Jaf Kurd bag, and in fact it was a transition in the type of knotting that first alerted me to the change-over. In the picture below, the rug changes from symmetrical to asymmetrical knotting apprxomiately at the two superimposed red lines, which seems a clear indication that the weaving was taken over by someone new. It is also obvious that the second weaver was not as experienced or skilled as the first, especially in rendering the "Shekarlu" border. Beyond that, the wool and colour palette seems identical, so I think the new weaver used the same materials as the first, suggesting that the weaving was kept "in the family". Now for the speculation. Could this be evidence of a "blended" family, with a mother-in-law getting her daughter-in-law started with a traditional design that was new to the younger weaver? She did a heroic job of trying to render the border and other designs, but using her own (asymmetric) knotting technique.

I find this interesting, and like the rug better now than I did before.

I would appreciate any thoughts or opinions about this phenomenon. Are there any other examples like this out there?


Posted by Camille Khairallah on 07-18-2007 02:57 PM:

Hi James and all,

Thanks for starting that thread that, I am sure and not speculating, will carry a lot of funny surprises.
This is one of my ex "teacher/student" rugs:

The mother (teacher) traced a line and told her daughter: Let's see what you can do... I heard her.. No, the rug "merchant told me.
One week after, the poor girl was punished without Swiss chocoate for a whole month... But... I cannot confirm.

Some other pieces will follow.



Posted by Gene Williams on 07-18-2007 06:19 PM:



That Baluch (or whatever it is related to E.Iran with the semi-Doktor-I-Qazi designs in the field) is amazing.

I have nothing like it. But, Jack has a prayer Turkoman prayer carpet with is likewise interesting with two totally different kinds of weaves.. (I watch him like a hawk).


Posted by James Blanchard on 07-18-2007 07:19 PM:

Hi Camille,

That Baluch rug is a great example. To me it looks like even the first weaver wasn't very conversant with what is a common Baluch meander border. The second version of the main border looks like an attempt at a Sistan border.

To me there are two potentially interesting components. One is the transition from one weaver to another that results in a poorer (or better) drawing. The second relates to different weaving techniques. For those who put a lot of stock into the association between structure and tribal affiliations, those examples where there is a transition of knotting techniques seem to offer some evidence of admixture between different tribal weaving traditions. In your example, is the weaving technique the same by the two weavers?


Posted by Chris Countryman on 07-18-2007 09:23 PM:

An alternative explanation

Greetings all,
Camille's Baluch may be the work of one weaver with a serious problem. The changes in the rug remind me of the drawings found in psychology books done by patients having psychotic episodes. That border is really scary!

Probably not, but if it was one weaver she was having something worse than a bad hair day.


Chris Countryman

Posted by James Blanchard on 07-18-2007 10:48 PM:

Hi Chris and all,

I kind of wondered about Camille's rug. One could imagine that an inexperienced weaver was making a mess of the original border selection (meander) and made an unfortunate decision to switch to the seemingly more straightforward parallelogram design. The execution of the central designs and the minor borders seem to be a bit of a struggle throughout.


Posted by Jack Williams on 07-19-2007 12:04 AM:

Thou shalt knot warp in the middle of the rug.


That is a very interesting discovery. Obviously, your first weaver was much better than whoever took over, and the two had learned a different method when young. This whole phenomena raises an interesting point, that probably doesn’t have a lot to do with your rug...but...

...At least as far as Turkmen weavings between about 1800 and 1890 or so, the knot might knot be a hot way of identifying provenance. The endemic slavery practiced by the Turkmen of all stripes, makes the use of a certain knot unreliable. A rug woven by a persian slave using As-left, depressed warps, while weaving a pattern dictated and overseen by big Tekke moma, is weaving a Tekke rug...not a persian rug, even though she is not a Tekke.

What doesn’t make sense in Camille’s posted rug is that the field design, which presumably is a pretty complex in its own right, remains pretty consistent. But the border, which should have been an easier pattern was totally messed up.

In Camille’s rug, I bet that she knew the central motif knot count by rote…but couldn’t extemporize the new border. In your rug, both the border and field suffered when the weaver changed.

Here is an another example of a weaving oddity, followed by a botential Baluch belief belying blockbuster.

My above rug has a 5 inch stretch in the middle woven with As-2, (asymmetric, open right, not depressed, possibly single wefted), while before and after that section the knot used is As-3 (asymmetric, open left, depressed about 30 degrees, double wefted). This almost should be impossible...but apparently somehow by using thinner wool the size of the rug is unchanged ...and the design remains beautiful... except the wool looks shinnier in As-2 area, and the pattern is offset one-half knot on front..while maintaing a ruler straight line in the back. In this case there was no loss of weaving skill,which led me to speculate this is a refugee camp rug.

But check this out, the NERS presentation of Mark Hopkins’ marvelous Baluch collection (every time I look at his focused collection I see something new that wows me). The structure of the rugs listed below might henceforth require us to look at knots on both sides of the rug, at least in Baluch.

Nos. 2, 12, 14, 21 – Pile: "Asymmetrical, open to left on left to right on right half…"

Regards, Jack Williams

Posted by Marty Grove on 07-19-2007 10:03 AM:

Apposite weaving

G'day all,

These curious pieces always intregue me when I see them, and you could say we see something similar reasonably often when you look at lots of rugs, but when one mentions that some may weave with the knot lay left, whereas the other weaver may do the knot with a lay right, then does that cause an abrupt 'feel' difference in the pile?

Ive a rug in which the lay of the pile just will not corrospond to the direction I expect it to take and wonder whether difference in knot lay may be a reason...?


Posted by Marty Grove on 07-19-2007 10:16 AM:

Sorry, above perhaps I should have said laid open left, open right.


Posted by Camille Khairallah on 07-19-2007 10:47 AM:

Hi all,

I do not like the word "speculation" because it sounds negative; I prefer to say that most suggested "probabilities" could be true.

Unfortunately, I sold this Baluch to a Spanish collector who lives in Seville and I don't have his address to ask about technical details.
As far as I remember, the knot was asymmetric and I do not think it changed with the change of weaver. But thanks for opening my eyes on that point. Since the time I started issuing certificates with technical analysis, I seldom examined more than two knot-places in one rug.

Jack, for us to whom the Baluch meander is a common design, we might find the field pattern more elaborate, but for a beginner who knows nothing of both patterns, the repetitive pattern is much easier to execute than the border for two reasons:

1- The single motif has a simple evolution that starts in one knot and ends a few rows later.
2- The weaver who weaves any repetitive pattern has nothing but to repeat a sequence of knots along a given row. Of course, the sequence depends of the complexity of the design but it is not the case here.

As for the border:

1- It starts at different points on a given row, 1 knot for the bottom of the hook and maybe 4 others for the line.
2- It has a continuous evolution and requires a lot of attention.

So, even though the second weaver (beginner) chose a discontinuous and far simpler border model, she couldn’t make it.

James, in my rug, there were two distinct weavers because even the field motifs underwent a mess, not as dramatic as the border but you can see it through the detail pictures.



Posted by Camille Khairallah on 07-20-2007 06:20 PM:

Hi guys,

Here is another 2-or-more-weaver-kilim that was woven in either Saveh or Hasht-Rud (Azerbaijan).

From the first glance you can see the dramatic mess. In fact it occurs on the lower left side, a little bit in the center, and along the two thirds of the right side.

One can imagine that the messy areas were woven by an aged lady whose eye-sight decreased considerably, but the ends of the messy areas at both sides are marked by a human figure that appears to be for me a little girl as if to say “here I signed out”.
This is my interpretation.
Still, the right side outer border "madakhil" is rather regular...

Could it be read in a different way?



Posted by Richard Larkin on 07-21-2007 09:52 AM:

Very interesting stuff, my friends, all of it. That Baluch put up by Camille, wow! Do the Baluch observe the Islamic prohibition against the consumption of alcohol?

Rich Larkin

Posted by James Blanchard on 07-21-2007 11:15 PM:

Hi Camille,

I see now that the drawing of the field in the lower part is also better than the top, which I think is consistent with the "two weaver" hypothesis. The first weaver also seems to have struggled with her drawing of all elements.

I can't really explain the kilim. I expect that someone who has more intimate knowledge of the flat-weaving process would have some ideas.



Posted by Camille Khairallah on 07-27-2007 08:00 AM:

Hello everyone,

I thought this thread would be a more active one as I had alone three pieces in mind, but I'm a bit sort of disappointed .

This is the third and last of my two-weavers' rugs:

I don't have it anymore but as far as I remember, it is a North-West Persian rug either from Goravan or -more likely- Meshkin.
The design is clearly reminescent of Mazlaghan (Hamadan group), but all the rest (colours, technique, etc..) points to North Persian Azerbaijan, and the approximate date of manufacture should have been 1930.

The major "trembling" lines of that rug are the achievement of what I guess again is a beginner (note also the Picasso's goat-inspired-chiken in the detail picture), but what is relatively more accurately done is the triangle-based medallion center that I guess was the work of someone else.



Late PS: To my knowledge, never a rug of more than 60 cm (2 feet) wide is woven by a single weaver, but James meant of course two weavers using different techniques or of diffferent expeience.

Posted by Steve Price on 07-27-2007 08:09 AM:

Hi Camille

The central medallion looks to be drawn more skillfully than the rest of the rug, but I think it would be odd if one weaver just did the knotting for that element. Why? A pile rug is woven in consecutive horizontal rows, so the weaver who did the medallion would have to have tied a single knot at the lower apex of the medallion, the less skilled partner doing the rest of that row, leading up to and then following it. The medallion gets progressively wider as the weaving proceeds, then gets progressively smaller again. The process would involve an awful lot of handing the work back and forth.

I am puzzled by the difference in apparent skill in the central medallion compared with the rest of the rug, but that explanation seems unlikely to me.


Steve Price

Posted by Camille Khairallah on 07-27-2007 08:27 AM:

Hi Steve,

You are completely right; I had not paid attention to that idea...

But now that I referred to the size on my site, it is 108 x 190 cm and I guess that the 108 cm width could have allowed a mother in the center and her two daughters on her sides. But again, did she stay by the loom just to weave the medallion center?...
The question is still enigmatic.



Posted by Marty Grove on 07-27-2007 09:32 AM:

Wonky Work

G'day Camille,

I think a dearth of posts here is not lack of interest, but perhaps more because we instinctively pass by those rugs which display a too obvious poverty of skill. They are interesting yes, but not the sort of thing which we would expressively display to all and sundry.

In lots of rugs we see evidence of the opposite weaver; two females sitting beside each other, and occasionally we may see where a third gave a hand at the weaving while the one of the others was off milking or bringing the cheese to the simmer.

If the wonky-ness isnt too apparent, then we can tend to smile and like it more because it shows a bit of character, and often where one piece of work in the whole is completely out of whack, we might then nod as if to expect it as Allahs acknowledgement.

But in the rug where the whole of the design is just too poorly put together, then it is unlikely we will buy it with pride - so perhaps those here have few like those you show, or may be a little bit too shy, which is not a good argument really as there are threads here which can equal or surpass yours.

There is one particular rusty Turkman I remember, owned by a senior gent here, which displays this wonkyness admirably, but is not one necessarily, of multiple weavers.


Posted by Marty Grove on 07-27-2007 09:42 AM:

G'day James and all,

Of course saying as I have above, is not to detract from James's initial query for examples of two specific weavers are working, one to initiate the beginnings of the rug, and later an obvious change of interpretation or skill, utilising the same materials for the same design.


Posted by Steve Price on 07-27-2007 09:48 AM:

Hi Marty

I wonder whether the wonky Turkmen piece you're thinking about is this one. If so, I'm flattered to be called a gent. I get called lots of other things, too.


Steve Price

Posted by Marty Grove on 07-27-2007 09:53 AM:


G'day Steve and all,

Actually, no it wasnt that one Steve, although its a good example of James's thread; the one I have in mind is John's large rusty orange bag with the very wonky guls.

Which bag of course is not quite in the theme of James's thread. Just an indication of wonkyness really.


Posted by Marty Grove on 07-27-2007 10:34 AM:


Jamesand all,

It might be that the phenomenon of one of two weavers on the same rug being trained/accustomed in a cultural weaving style different from her partner, may not be so very uncommon. How many of us have carefully looked for it?

With all the tribal fractures occasioned over the past two hundred years, and their peoples fleeing strife, and who must eventually settle some place where there is a bit of security enough to be able to weave to sell for their support or advancement, there the family might come into contact with another family with an unmarried son and make a contract of marriage for their unmarried daughter.

The first family was from Kurdistan where they usually knot Symmetrically, and perhaps the boys family was from a part of Afghanistan where generally it could be said the weavers knot in the Asymmetric type.

The resultant rug woven by the mixed weaving in-laws would more than likely have the two types of knot in it, and one would assume after time the knotting would become the same as is generally used by the area in which the families find themselves.

While we accept that sometimes the Turkmen rugs have Symmetric knots nearest the selvedge and the rest of the rug Asymmetric, there are few occasions we find the two types intermixed, but that is not to say that it is rare, just fairly uncommon as we rarely notice it.

This is distinct from two weavers of different skill - rather a mix of two tribal cultures.


Posted by James Blanchard on 07-27-2007 10:59 AM:

Hi all,

As Marty has pointed out in his most recent post, my interest in starting this thread was not just to show that two weavers of much different skill sometimes worked on the same rug. In the rug I posted initially, the knotting technique is also different. To me, that could be direct evidence of how fluid design and structure could be in tribal weavings where there was close geographic and social proximity. Altough it is conjecture, I have a notion that my rug was started by a reasonably accomplished weaver who used symmetric knotting (Khamseh?), and completed by a less experienced weaver who used asymmetric knotting. If so, it is possible that even rugs woven within one S. Persian family would have different structures. I wonder if this structural fluidity is present in other tribal groups more than we might have thought.