Home Page Discussion

Salon du Tapis d'Orient

The Salon du Tapis d'Orient is a moderated discussion group in the manner of the 19th century salon devoted to oriental rugs and textiles and all aspects of their appreciation. Please include your full name and e-mail address in your posting.

A Turkish "Village" Rug Fragment

by R. John Howe

A few months ago, while traveling, I encountered a large fragment of a Turkish "village" rug in a dealer's gallery.

It was coarse. It was torn. It had holes. There were areas with bare warps (only). There were other areas with warps flying. The edges had been eaten at badly and there were places with low pile or bare structure (warps and wefts). I could see the drawing was not impeccable, among other things the weaver had run out of warp and had miniaturized the main border on one end and there were some awkward, one-armed spandrel-like devices in the field. The fragment was irregular in shape, bowing out at the sides and with an "S" curve in the shape of its top edge. If you layed it out so that ends and sides were roughly horizontal and vertical, the field pattern was at an angle. If you rearranged it so that the main field design was square, the piece ran uphill left to right at both ends.

I bought it.

Of course, it had its points. Chief among them was a large, bold, square-ish device with arches at both ends, drawn in a bright blue with poly-chrome ram's horn instrumenting on a light red ground that took up most of the field and that spoke to me. In addition to this attractive red and blue, the rug had a softer yellow, an olive green, what looked like a good purple and some white, used sparingly but well. It had four diamond devices in the four corners of the field that,

together with the large central device and its internal instrumentation, followed the very ancient 2-1-2 schematic. The square-ish central device had its own central medallion, an olive green octagon, edged in white with four "ram's horn" devices, one off each side and the other two off the top and bottom. This green octagon had an eight pointed-star at its center that was faintly echoed at the center of each of the four diamond devices in the corners of the field. The rug had one major and two minor borders edged, effectively with four lines of a red and white barber-pole stripe.

The main border was an old friend: a poly-chrome quartered and four-eyed medallion, ubiquitous in Caucasian and Kurdish weaving. The minor border was clear "Z" rendition of a design seen more frequently as an "S." And the borders were mostly there all the way around.
The square-ish central device has arches drawn at both its ends, each filled with instrumentation that would have triggered "mother goddess" allusions not long ago (but we won't do that here).

I brought the rug home and asked a conservator to take her time and to sew it carefully on to a backing. She built my vocabulary by telling that what I was asking for was "couching," a form of unobtrusive embroidery. When her careful work was completed, I hung this piece on one wall of my office at work. It measures 61" X 91" and dominates that wall.

(Note: Gentle readers not particularly interested in such rugs should probably stop here. I am going to examine this rug somewhat more extensively than it's true place in the world of our interest likely deserves. You are forewarned.)

Technical Description:

Turkish rug
61" X 91"
Symmetrical knots
Vertical: 5 per inch
Horizontal: 5 per inch
Knots per square inch: 25
Back side appearance:
No warp depression
Yarn spin: Z
Warp: 2 plies, tan or undyed natural, occasional dark brown, wool
Weft: red, single unplied, two and often three strands, 2 and 3 picks.
Selvages: Gone
End finishes: Gone
Colors: Light red, bright blue, olive green, gold, purple, white.(outlining color?)

I been living with this piece for about a month, and I've been looking at a few books trying to figure out where it might have been woven. The Turkish portion my rug library is not extensive and it has been, largely, a discouraging experience. In my initial literature search, I encountered no rug that was similar enough to mine for me to get a hint about a plausible attribution. At one point I looked at Sevan Caucasian designs, and Jon Thompson, does say at in his Oriental Carpets (Caption on page 115) that "a related form of the (ed. Sevan) field design is also found in Turkish carpets." Yes, but where?

One experienced collector visited my office and commented on the red wefts, said the purple looked good, described the bright blue I liked as mottled and questioned the olive green. He was polite but quiet. (Well, it's not a great rug, just a modest one that I like.) We talked briefly about the strange purple one-armed spandrel devices and some other seeming instances of conventionalizing. He made no definite geographic attribution. At one point I joked that perhaps it had "no noticeable pedigree." He smiled ambiguously. I decided that if I didn't find out what it was soon that I might begin to describe it with a pleasanter sounding phrase like, "no other instance known." (Am I a rug collector or not?)

Then one night I began to look more systematically through the books I had and toward the end of Christopher Alexander's book came upon this rug.

The first thing I noticed was the bright blue. Then as I examined the other colors it seemed to me that it also had the lighter red, the soft yellow (although it had a lot more of it), a more saturated darker purple, but nevertheless a purple, and it had the same olive green that my collector friend had seemed to question. So the color palette seemed very close.

It also seemed to me that the border systems were almost identical, my weaver had reversed the minor "S" borders in the Alexander piece to make them "Z" borders but except for that reversal, the drawing of the borders was exactly alike.

The piece now seemed a possible, and I began to examine it in detail. That's when I saw that my central square-ish device with the arches at both ends was also present. One simply had to drop the four extensions of the spotted purple bands at the sides of this central device, as well as the "V-shaped" endings of both arches.

(Note: I have rotated this image to the left so as to get it all on the screen.)

Alexander makes much of the drawing of this central device in his rug. (Alexander dates his rugs very aggressively and has been taken to task on this point. He estimates that this piece was woven in the 16th century.) He believes it to be a version of a massive 13th century Seljuk border design transplanted to field use. And there is a Seljuk prayer rug in the Alexander collection that has some striking similarities to Alexander's "Red Rug with Spotted Purple Band" and even, I will humbly say here, to mine.

Here is the Seljuk prayer rug:

I began to examine the photos of this rug closely too. Sure enough: there is the same bright blue used in some of the same places that it is used in my rug: that is, in the bands that are internally instrumented as my central device is. This instrumenting doesn't include the precisely opposed double ram's horn device that is used in mine. Instead double ram's horns are alternated, facing in opposite directions.(The precisely opposed version of the double ram's horn device is also present in a more complex form at the main border of the Seljuk piece.) And there are traces in the instrumenting of this same band in the section that forms the top of the arch. The red is more saturated but it is the field color and in the close-ups I could see that there is a row of "Y" shaped devices at both ends of the Seljuk rug, some of which are in gold, green (again more saturated than in my piece) and purple. Again the overall color palette is similar.

Note: I have rotated this image 90 degrees left so that so that you can see it comprehensively.

And if the arms are dropped from the upper corners of the central square-ish device and those that serve as an incomplete re-entrant device at the bottom are flipped vertically and joined with an arch we have the two arches in the "16th century" Alexander piece and in mine. Weavers are practiced in making such moves since a great many rug designs are rotations of a quarter portion in two directions or of a half in one direction.

The main "legged" and "mirhabed" device in the Seljuk rug is one about which Alexander resorts in his analysis to nearly spooky language. He talks about its "being-nature." He denies saying that it "resembles a creature" but says rather that the "abstract structure" of this device "is so conceived that to a depth extremely rare in any carpet, or in any work of art, one feels the presence of a being behind that form." I think that, for me, there may be something to this. I described an Ersari ikat chuval design in a recent salon as sometimes seeming nearly to "breathe." I think that one reason that the square-ish central device in my rug, with its arches at both ends, speaks to me, is that even with the loss of its "legs" and some other manipulation, it retains taxonomic echoes and exerts some of the aesthetic power of the "13th century" form of the central device in the Seljuk prayer rug.

Other interesting aspects of the Seljuk rug, include that fact that there are spandrel-like devices inside the central square-ish figure in the same approximate places that they occur in mine. There is also a central medallion but it does not have the octagon shape of the central medallion in my rug and the most central device is not an eight pointed star but rather a six-sided figure. This Seljuk prayer rug has some real differences but its similarities to the "16th century" Alexander rug and to mine are tantalizing.

Because he is interested in such things, I decided to see if Alexander might offer an explanation of the strange shape of the one-armed spandrel-like shapes in my rug.

And Alexander's discussion does seem to offer a basis for an interpretation. First, he uses spandrel devices in Turkish village rugs as one of his examples of drawing that produces both effective positive and negative space. He begins by examining one of his own drawings of a common Turkish spandrel usage.

The sketch on the left is Alexander's rendition of the positive space in such spandrel designs. The image on the right darkens the negative space in such spandrels to make it more evident. You can see that they both include trefoil devices pointed in opposite directions. The spandrel design itself is in its overall outside outline, a kind of double-winged device.

Here is a Bergama rug that has spandrels of this sort in the four corners of its field. Notice that in this rendition in an actual rug, the overall image of the spandrel device has lost some of the symmetry it had in Alexander's sketch. It is heavier in the half that is closer to the center of the rug. I have looked at a few of them and this is a frequent tendency in the drawing of spandrels in Turkish village rugs.

Let's look at one of these spandrels more closely.

The imbalance that I just mentioned is clear in this blow-up but notice also that the trefoil device with its branches pointing toward the center of the rug that were connected to the outside corner of the spandrel in the sketch version is now disconnected. Notice also that this trefoil itself is also unbalanced with one "plume" longer than the other. This usage is also common is such spandrels.

I am now ready to offer two set of reasons for the seeming strange drawing of the spandrel-like devices in my rug. One of them, I think, is fairly straightforward, the other might be seen as more fanciful.

First, I think the overall device the weaver of my rug intended to draw was a version of the kind of spandrel in the Bergama rug. And she followed a usage visible in other Turkish rugs of overbalancing the body of this device towards the center of the rug. This is, I think, the likely source of the awkward seeming rectangular shape of the "body" of the one-armed spandrel in my rug. The weaver was simply following the guidance of an item in the Turkish weaving vocabulary concerning the drawing of such a spandrel. Problem: one cannot make the body of the device an inward moving rectangle and still have room to draw the other "wing." I think my weaver simply dropped the other wing off when she saw that she didn't have room to include both it and the rectangular shape of the body. (It is interesting that she appears to have had room to include the inside "wings" if she had made the "body" of the spandrel a smaller square, but she apparently decided that the "rectangular" part of her weaving vocabulary in the drawing this device was more important than was a second spandrel wing. Steve Price wrote a few years ago in the ORR about the likely importance of noticing the decisions weavers make about what to give up and what to preserve in such situations.)

Now let's be more fanciful for a moment. Why is the red trefoil placed as it is in the rectangular body of this one-winged spandrel devices? I have already alluded to the first possibility when I noted that in the Bergama rendition the trefoil with the branch pointing inward is disconnected there as it was not in Alexander's sketched version. So part of its placement has to do with this lost connection. It is still positioned approximately as it would have been had the stem continued to the outside corner of the spandrel.

Now for the fanciful part. Alexander first shows that there are some interesting "bird" and "animal" forms in the Konya weaving vocabulary (Oh yes, that's where Alexander says both the "13th" and "16th"century pieces with which I've been comparing with mine are from. They were likely woven in the Konya area.).

Alexander provides this sketch of a Konya "bird."

He also provides a similar drawing of a Konya "animal."

Notice the "tail feathers" on both of these creatures. Note that they are they have the kind of imbalance that we saw in the trefoil device in the spandrels in the Bergama rug and also in the shape of the similar red device in the one-winged spandrel-like shape in my rug. But why, if the trefoil devices on my piece are "tailfeathers," are they placed as they are on this awkward rectangle? Well, in a part of his analysis that may stretch credulity for some, Alexander suggests that the spandrel device itself can be seen as a "bird-like" form. A bird with wings outstretched viewed from overhead (it looks more like a B-1 bomber to me but never mind). The "beak "of such a "bird' would be the outside corner of the spandrel and the two "wings" are real out-stretched wings of a bird in flight. If you are with me and Alexander this far, it is not a too much further stretch to suggest that the placement of the red, unbalanced, former trefoil device on the purple rectangle is not too outlandishly far from where such a bird's tail feathers might in fact be. It is closer yet to the forward seeming position of the "tail feathers" on Alexander's Konya "animals."

So I'm feeling a little better about this humble rug fragment. It is still gauche, a bit crude, more than a little conventionalized in it drawing. But I think it has telling similarities to some noteworthy rugs woven in the Konya area. And that is my current guess about where it was woven too, sometime during the 19th century. To use language I have objected to on this site, it may be a poor country "cousin" to the two fine rugs in Alexander's collection, but I think it is a "cousin" none the less. My humble and abused but now rescued and well-conserved piece is not one "without discernible pedigree."

Compare with: Alexander, Christopher, A Foreshadowing of 21st Century Art, New York, Oxford: Oxford University Press, 1993, pages 279 and 127.

The piece above has been fun to write and I plead guilty to having dramatized things a bit. Despite this, I would truly be interested in comment and critique on this piece and on what I have written.