Warps on two levels

Hello John and all,

Part of John Wertime’s Rug and Textile Appreciation Morning program at the TM explored the question of whether structures in certain rare sumak weavings can assist in identifying provenance. In particular, Wertime focused on several bags with “warps on two levels” (his term) or “fully depressed warps” (another term sometimes used by others).

We’re all familiar with fully depressed (i. e., 90 degrees) warps in many pile Persian city rugs such as Saruks and Kashans, but the phenomenon in flat weaves has not been much discussed.

As with pile rugs, having warps on two levels in sumak bags produces a thick, heavy, leather-like, firm handle that would be very durable. The thicker the warps themselves, the “thicker” the handle.

This Baghdadi Shahsavan (southernmost of the Shahsavan groups) mafrash end panel was plate #1 in Wertime’s Sumak book and has the characteristic “thick” handle. This panel has warps of brown and ivory yarns twisted together, some cotton in the warps and the wrapping wefts, some countering of the weft wrapping and twining at the top and bottom.

Although in Sumak he attributed this mafrash side panel to the Kazak area of the Caucasus, Wertime now wonders whether it should be grouped with the Baghdadi end panel on the basis of the thick handle and the presence of twining – something not normally encountered with sumak.

Several other sumak weavings with thick handles and depressed warps were brought in to the program. This is one of mine that Wertime also believes may be Baghdadi:

This well-known khorjin half also has depressed warps and a thick handle.

The use of warps on two levels must have been developed in order to make very durable utilitarian weavings.

The audience must have brought in seven or more reverse sumak pieces with Zili designs such as this one of mine:


Because of different colors, materials and designs, we could easily believe that putting warps on two levels is a relatively widespread practice in NWP. However, Wertime is currently exploring whether there is a primary connection to the Baghdadi Shahsavan.

All of the pieces shown above were shown at the program and there were many, many more excellent examples of sumak weaving. These RTAMs at The Textile Museum provide an outstanding learning experience.

Wendel

Wendel -

Say a little about how the depression of alternative warps in these sumak pieces is accomplished.

In pile weaving, it seems to be accomplished by making at least one set of structural wefts taut and/or by "crowding" the number of warps in a given width.

Is it the same in sumak? I assume that any sumak with deeply depressed alternate warps has a structural set of warps and wefts and that the patterning wefts (the ones that wrap) are separate from them. Correct?

Thanks,

R. John Howe

Hi John,

You posted:

quote:

---

Wendel -

Say a little about how the depression of alternative warps in these sumak pieces is accomplished.

---

quote:

---

In pile weaving, it seems to be accomplished by making at least one set of structural wefts taut and/or by "crowding" the number of warps in a given width.

---

quote:

---

Is it the same in sumak? I assume that any sumak with deeply depressed alternate warps has a structural set of warps and wefts and that the patterning wefts (the ones that wrap) are separate from them. Correct?

---

Hi Wendel -

What you have written above is useful.

First you confirm my understanding that sumak with fully depressed warps cannot be "weft-less sumak."

You wrote in this context:

"...sumak wrapping has structural or ground wefts, just as does pile weaving. John Wertime would contend that conceptually there is not a great deal of difference between pile and sumak wrapping except that the wrapping wefts are cut in the case of pile."

Me: So I think your answer to my question is "yes."

You also talk about depressed warps and how that is accomplished. Here you say in part:

"...It seems obvious that a weaver must set up the loom and the warps specifically for the degree, if any, of depression of the warps. When the warps are on two levels (90 degrees of depression), there are twice as many warps per linear inch horizontally as when there is no depression. The warps would have to be as close to one another as possible. The weaving cannot pinch or “crowd” the warps; the weaving comes off the loom the same width as the warps were set up initially..."

Me: It is true that the number of warps included in the orginal set up of a given weaving on a loom contributes to whether warps will be depressed or not. The word "crowded" actually comes from Marla and refers to a situation in which the combined width of warps planned to be included is higher than the available horizontal distance in the planned width of the weaving. Such crowding makes it necessary that all the warps to be included cannot be on the same level.

So there will be some depression occasioned by this "crowding," but there is another variable of concern, namely the "order" of depression.

Since what is desired, usually, is that only alternate warps be depressed, the weft also plays a role in this depression by defining the order and frequency of warp depression. And, of course, if this "depression-defining" weft is also taut, it can contribute to the depression itself as well.

(Note: I would not argue that there are not sometimes instances of full depression of alternative warps that are mostly a function of taunt wefts, only that if there is "crowding" the crowding contributes to the depression.)

Third, you raise the issue of "What qualifies as a knot?" This is an arena in which my macrame experience, and my related contact with nautical knotting, applies. It is not importantly germane to our consideration of sumak structures, but may still be useful to review.

First, the word "knot" is ambiguous in the various literatures in which it is used. Some usages are not demanding at all. Referring to the useful link you provide a "simple hitch" is not very firm at all.

http://www.realknots.com/knots/hitches.htm

But at the other end of the continuum of knot structures that some argue the term "knot" should be restricted to, we have knots that are "firm on the basis of their own construction."

A "square knot": two overhand knot tied one on top of the other in the manner shown in the drawing in this link is one of the simpler knots that is firm on the basis of its own construction.

http://www.inquiry.net/outdoor/skills/b-p/knots.htm

In pile weaving, the asymmetric knot does not meet this latter requirement at all since it is composed of a "simple hitch" around one warp and an "inlay" on an adjoining one. The firmness of asymmetric knot is entirely dependent on the pinching action of the wefts above and below it.

The symmetric knot is closer to a knot that is firm on the basis of its own construction but still requires constant tension on its "pile" ends (which are cut off as the weaving proceeds) for its firmness. For this reason, it too is dependent to an extent on the pinching actions of the wefts above and below it for its firmness.

I got a small credit from Marla Mallett in her first edition for raising and arguing this issue with her before publication. He husband asked to be introduced to the person who had made so much trouble about the word "knot."

I'm a bit off topic with this latter dissertation on "What is a knot?", but I think your comments clarify most of the issues in my previous question.

Regards,

R. John Howe

Hi John,

I also don't want to deviate too much on the topic of knots, but the subject is interesting. You said:

quote:

---

The symmetric knot is closer to a knot that is firm on the basis of its own construction but still requires constant tension on its "pile" ends (which are cut off as the weaving proceeds) for its firmness.

---

Wendel -

As I said, the word "knot" is ambiguous, in part because it has been used historically to refer to different structures.

And there is an associated debate about what range of structures should be referred to with the word "knot."

So it's not (sorry) a question of being a little bit "pregnant" since what counts as a "knot" is not an "either-or" thing, but rather a continuum about which folks disagree about the applicable portion.

To show how stern some "knotters" can be, some naval knotters do not consider the "square knot" example I offered to be a "firm" knot since under sufficient tension it can change structure and slip. Here's link that makes this argument.

http://www.42brghtn.mistral.co.uk/knots/42ktreef.html

You need to be careful getting interested in "knotting." It's a field like rug collection that can consume you. Here are a few links in warning:

Here are some serious "knotters."

http://www.igkt.net/

But if you're not (again, sorry) careful you can find youself in a field of mathematics.

http://www.ccs3.lanl.gov/mega-math/workbk/knot/knot.html

http://www.popmath.org.uk/exhib/knotexhib.html

http://mathforum.org/library/topics/knot_theory/

If you venture into any of the mathematical knot theory sites you'll find that neither the symmetric nor the asymmetric knots can qualify as mathematical knots because the latter are restricted to structures that have no loose or dangling ends.

Aren't you sorry you went on about this?

Regards,

R. John Howe

Hi John,

Perhaps I will regret the foray into knots, but I was the one who first something that I really don’t know very well: knot theory. I’ve already spent entirely too much time browsing through some of your fascinating links.

Some of the sites are relevant to rug studies. One site has several versions of the Celtic knot, which is essentially the same creature as the Holbein interlace. There is a relationship between knot theory, principles of symmetry and that considerable portion of Islamic art that is based upon mathematics.

I have no training in mathematics myself. I’m generally much more concerned with aesthetics than ethnographic significance. Viewing textiles in a mathematical context may seem cold and distant, but I think doing so is far more productive to rug and textile appreciation that pondering the “meaning” in them.

Now, isn't anyone intrigued by the fact that sumak wrapping is sometimes done with warps on two levels? Or is that just too arcane?

Wendel

Wendel -

There was one place where Wertime was talking about sumak with warps on two levels where he said that this "extra weft wrapping" structure was on in which actual knots were formed.

The pieces in question, I think, were those from the Hashtrud-Miyaneh area.

In my virtual treatment of this part of his presentation I used the piece below.



As I indicated in my reporting this piece was not in the room but there were maybe five of this type brought in.

Here is what I reported about them in my virtual telling of Wertime's lecture.

"...Wertime indicated that this is another form of sumak in which there is 90 percent displacement of every other warp in the wrapping. Another, even more unusual feature, he said, was that this piece is made with an “extra-weft wrapping” structure in which “a true knot if formed in the wrapping process.” Wertime used the example of the knot one uses to tie one’s shoes, but I think what he intends is that a lot of “knots” in the world of textiles are not “true knots” in the sense that they are not “firm on the basis of their own construction.”

(Ed.: My italics added here)

I am not sure what this "true knot" structure looks like.

Wendel, maybe you could ask him whether there is a drawing of it anywhere. Or do you know yourself?

This is a place where our discussion of "knots" intersects precisely with Wertime's session.

I think the occurence of what we might call a "true knot," that is, one that is "firm on the basis of its own construction" is a relatively rare thing in the weaving world.

Regards,

R. John Howe

John,

The Chapman bags have truly glorious colors. Too bad we didn't see them that day.

I posted a discussion of knotted reverse sumak earlier in this thread, wherein I said that this kind of wrapping is like a clove hitch, based upon Marla Mallett's illustrations. I would not call a clove hitch a true knot in that it would collapse if the warps were removed. It is not firm in its own construction.

See illustration 5.25 in Marla's book for an image of the reverse side of one of these. That one is mine.

Wendel

Wendel -

"Clove hitch" is good. I missed it somehow in your previous post.

I have only Marla's first edition which does not have a 5.25 illustration. That must be one of the changes made in her second edition.

Here is another image of a "clove hitch" for others who might have missed it above.

http://www.instructables.com/id/Tieing-a-Clove-Hitch/

If you rotated this image 90 degrees so that the wooden rod becomes the warp, that is the structure that Wertime is apparently referring to.


Regards,

R. John Howe

Hi John and Wendel,

I agree with John that the question, "to be or knot to be" is properly answered in the context of a continuum. One can argue about where on the continuum the answer is found. It is precisely not analogous to the question of pregnancy. I also agree that as to pile weaving, the symmetrical "knot" is fundamentally "firmer" than the asymmetrical knot. I wonder whether this makes the symmetrically knotted fabric appreciably more stable or durable than the other. It may be a question for physicists.

I find the following comment attributed to Marla Mallett to be very interesting:

“This warp depression is made possible and orderly solely by the way weft yarns are inserted. Alternating wefts are allowed different amounts of ease.”

If I understand this correctly, she means that the weaver chooses and controls the phenomenon by application of technique. It seems remarkable that such uniform and stable fabrics can be produced in this manner.

__________________
Rich Larkin

Hi Rich

That's exactly how warp depression is done in pile weavings as well. One weft shoot is a little slack ("sinuous") so it more or less half-encircles each warp, the next weft is taut, putting alternate warps into different planes.

Clever people, these weavers.

Regards

Steve Price

Bi level

G'day John and Wendel,

Assuredly there are some of us who are interested in these soumac weaving structures made on two levels - though I must admit to knot (sri) not having the faintest clue just how it is done.

If the design is created by wrapping threads around warps/1/2 or more, and the structure is stiffened by having one warp below the level of that beside it, and around these warps threads are wrapped, not individually on each warp separately, but threads which are wrapped around one, then around the next etc etc, and after all this, one of the warps in sequence must be below the other to be able to call the structure 'depressed warps', well then I find it extemely difficulty to visualise...

About the description 'knot' in rug weaving - Mumford refers to them as 'stitches', and he is not the only old writer who does - any idea just when stitches became knots?

Mumford also refers to Soumack, and an early mistake in 'Soumack' supposedly being a khanate in Northern Iran/Caucasus rather than a weaving 'type', which incorrect identification/provenance he made, and which in books by others thereafter, the mistake was perpetuated until he corrected it.

Apparently, in the earliest days of commercialism in rugs, the soumac type of rug was given the appelation of 'Kashmir', this being a tick of super quality, as the Kashmir shawls were considered the finest woven product of the time - so sayeth Mumford.

Regards,
Marty.

Oops!

Hi Steve,

Once again, you've set me straight. In reading Marla's comment, I was considering that the wefts dictating the depressed/two level structure were individual supplementary wefts that also formed the surface, giving the piece color and pattern. To apply these in a manner to dictate as well the character of the fabric as a whole would be truly astonishing. I realize, however, that it is the continuous wefts that are alternately slack and taut, thereby setting up the two level situation. As that funny lady used to say on TV, "Nevermind!"

__________________
Rich Larkin