Since I raised a possibly tangential point at the beginning, perhaps I should be the one first to take Michael Wendorf's advice to restore the discussion to a footing more closely aligned with Daniel's original intent.

As a possible way of doing this, I have looked a bit for some examples that might be seen to fit into the set of weavings that Daniel is alluding to in this salon. One of the difficulties is that frequently the examples encountered do not include technical descriptions.

But the catalog "Mideast Meets Midwest" done in 1993 in conjunction with ACOR II in Chicago, has a Kurdish section. Although most of the pieces in this section are seen to be from the Bijar area (something that seems, if I understand properly, to put them outside the group of rugs Daniel is pointing to here), the section includes descriptions by Jim Klingner, whom Daniel referred to, and it provides technical descriptions.

Here is the lead piece in that section.

Klinger says that this rug is from the western Persian mountains and is related to the one that is Plate 10 in Eagleton. He says this asymmetric design could be sourced in either the "afshan" or "mina khani" designs.

I would take it that this rug is likely within the group to which Daniel is alluding in this salon.

Here is the technical information:

Age: Late 19th century
Warp length: 100 in. Weft length: 50 in.
Colors: 7
Pile fiber: Wool
Knot: symmetric Vertical: 7 Horizontal: 6
Density: 42 per sq. in.
Warp fiber: tan wool Spin and ply: 2ZS
Offset: 0 degrees
Weft fiber: red wool Spin and ply: 2ZS
No. of shoots: 2 to 3
Edge finish: 3 cabled warps
End finish: weft twining at top end

Here also is the second piece in that section:

Now it is likely that this second rug is outside the group that Daniel is trying to delineate since Klingner says that it is "possibly from Quchan," which would put it in Khourassan but he also says that it is similar to a rug that Jenny Housego (Plate 140) places in Veramin, which while not in the western mountains, does seem to be within the geographic area to which Daniel is referring.

Here is the technical description of this second piece:

Age: early 20th century
Warp length: 91 in. Weft length: 51 in.
Colors: 13
Pile fiber: wool
Knot: symmetric Vertical: 7 Horizontal: 7
Density: 49 per sq. in.
Warp fiber: tan and white wool
Spin and ply: Z2S Offset: 0 degrees
Weft fiber: tan and white wool
Spin and ply: Z spun singles No. of shoots: 2
Edge finish: 3 cabled warps
End finish: weft twining at bottom end, new warps were added and the length of the rug was extended.

Klingner also offers the following summary of indicators of the weavings of the village and mountain Kurds:

· Most are small or short runners. Short runners, about three feet by nine
feet, are particularly common in Anatolian Kurdistan.
· Virtually all rural Kurdish rugs are symmetrically knotted.
· Pile weavings have short kilim aprons and these are often decorated with
Twining, extra weft brocading or some other decorative technicque.
· The selvages are usually reinforced with pile yarns in different colors.
· Warp ends are frequently cross braided.
· The longer rugs are often crooked. The is especially common in Anatolian Kurdish rugs.

This is in 1993 and some things have been learned about distinctive structures since then.

It might be useful to put up other likely examples.

Regards,

R. John Howe

If we are to discuss weavings from the Kurdish weaving heartland, it probably excludes both Quchan and Veramin. The Kurdish weaving heartland in this context would be more narrowly confined to historic Kurdistan.

The first rug you display belongs to Moe Jamali and is almost certainly from the area around Sauj Bulaugh or modern day Mahabad. It belongs to the group of rugs identified by Alberto Levi in Hali 70 from this area that he labels "proto-kurdish" but which generally contain persianate designs. As previously stated, I do not believe that the term proto-kurdish is appropriate, defensible or proper for this group of rugs - a group I nonetheless find very beautiful and well-colored and that has become well respected in the rug world for the color and wool quality.

The particular pattern in the rug you display has no direct connection to the Mina Khani pattern. It is directly derived, in simplified and enlarged scale, from the Avshan pattern. The Avshan pattern has a long history in many areas of Persia and the Caucasus and Murry Eiland, Sr. has written a fairly exhaustive history of the pattern in a recent Hali. In addition, it is worth noting that Charley Ellis thought the Avshan had its origins in India and connected it to an early fragment in the Textile Museum's collection that dates to the 16th century. See his book on Early Caucasian Carpets for the reference. It bears repeating that this pattern was aimplied, adapted and adopted by weavers in many places. For example, Jim Burns has fantastic examples of such simplified but nonetheless vigorous rugs with their roots in the Avshan pattern from no less than three distinct weaving groups. See his book, Traditions in Weaving for a Caucasian example. He also owns an early Kurdish example woven on an ivory ground. Still other examples of this type are known from around Bidjar which may or may not be Kurdish weavings. Check out a rug that belonged to Dennis Dodds in Atlantic Collections to see it woven on a light brown ground and recall Michael Rothberg's rug on display at the last ACOR, this is the same rug depicted by Levi in Hali 70.

What do any of these rugs have to do with a Kurdish tradition? Well, it shows what Kurdish weavers were capable of in terms of wool and color and confirms what we already knew about their willingness and ability to adapt and adopt designs. But this does not really add much else.

The second rug strikes me as being more likely from somewhere in Azerbaijan than Quchan or Veramin and the analysis is much the same execpt the color is not quite as good as the avshan varient you posted. Here we see so-called memling guls. Although this rug is probably woven by Kurdish hands, I again fail to see what it adds to our understanding of Kurdish weaving traditions.

Reading the descriptions of these rugs and a general set of characteristics for these rugs written in 1993 by a well known proponent of Kurdish rugs like Klingner is also a bit disappointing and humbling. It really says very little of real use.

As a start, Daniel has made a point of indentifying oblique interlacing bands, often called plaited or braided in the literature, as an identifier of weavings that appear to come from the Kurdish weaving heartland and that seem to represent a vein that is a tradition of long standing. Likewise, oblique interlacing, two-pick oblique interlacing and oblique wrapping. I would add brocaded aprons, skirts or borders to this list as well as weft-float patterned aprons or skirts.

I would also add selvedges with two sets of wrapping yarns in contrasting colors or, as Marla Mallett has suggested, split symmetrical wrapping.

Finally, I would look for rugs that in addition to some of the details above, contain some goat hair in the foundation, multiple colors or shades of undyed wools and patterns that can be understand as having a relationship or origin in more restrictive weaves such as brocades. These rugs seem to be part of a long continuum that, seen in a larger picture, hint at the scope and depth of a long tradition as long as that of weaving itself.

Thank you for your interest. Michael

I might add that bands of oblique interlacing and two-pick interlacing have not been used exclusively by Kurds, although bands of oblique wrapping MAY be limited to Kurdish work. It is too soon to know for sure, as we are just starting to sort out these structures that normally have all been erroneously lumped together in rug books under the term "braided."

Likewise, brocaded details are not limited to Kurdish work. I don't understand what you mean when after mentioning brocading in the aprons and skirts you add, "...as well as weft-float patterned prons or skirts." Are you talking about weft-substitution, complementary weft, or what?

Perhaps you can tell us whether or not the two-colored "wrapped and bound selvages" like my example in Figures 15.61 and 15.62 in "Woven Structures" don't also appear in Kurdish work? I should point out for others that the selvage I've called "split symmetrical wrapping" appears in my Second Edition as Figure 15.67 and it is on the website UPDATES-3 page, but it is one of the things that didn't make it into the first dition of the book. f course a wide variety of selvages have been used in Kurdish work--both interlaced and overcast--in addition to these specialized constructions.

As for the use of darker colored wools, I have a Turkmen horse cover from Eastern Anatolia in front of me right now that combines a warp-faced ground with inlaid brocading and knotted pile. Large areas of the pile are undyed medium gray. Other pieces from Eastern Anatolia that are not Kurdish also frequently include natural browns and grays--in both the pile and ground weave structures. So though such usage may be common in Kurdish pieces, we probably need to be careful to not imply that it is an exclusively Kurdish practice.

Many thanks for all of your contributions.

Marla

I want to encourage you to be patient with those of us who are interested in "Kurdish" weaving but who have not studied it or collected it as you have. You have suggested a couple of times so far that particular contributions do not advance things much. I think it not a realistic objective for the salon that the rest of us provide advances on the knowledge of Kurdish weaving that you have accumulated.

I think things will go better if you are prepared to be a little bored and to help edify the rest of us. And in truth you have tried a bit.

In your last post you said in part:

"…Although this rug (ed. the second rug I offered in the preceding post) is probably woven by Kurdish hands, I again fail to see what it adds to our understanding of Kurdish weaving traditions.

(ed. one possibility might be that Daniel included a Veramin piece as his initial example and you have apparently excluded Veramin from the area in which it is most hopeful to look for traditional Kurdish weaving. By process of elimination we are defining the primary area of geographic interest more precisely. This could be seen as a small advance.)

"Reading the descriptions of these rugs and a general set of characteristics for these rugs written in 1993 by a well known proponent of Kurdish rugs like Klingner is also a bit disappointing and humbling. It really says very little of real use.

"As a start, Daniel has made a point of identifying

"oblique interlacing bands, often called plaited or braided in the literature, as an identifier of weavings that appear to come from the Kurdish weaving heartland and that seem to represent a vein that is a tradition of long standing...Likewise, oblique interlacing, two-pick oblique interlacing and oblique wrapping.

[ed. here is a direct scan of the ends and sides of a rug fragment of mine (please ignore the dog hair, this piece is on the floor at the moment) that for awhile I thought might be Kurdish precisely because it seems to exhibit a braided end finish.]

(ed. continuing: Some advanced collectors, with it in their hands, have suggested that it is "Caucasian" If they are correct such end finishes may not be very conclusive indicators of Kurdishness.)

"I would add brocaded aprons, skirts or borders to this list as well as weft-float patterned aprons or skirts.

(ed. Klingner mentions aprons but not with this specificity)

"I would also add selvedges with two sets of wrapping yarns in contrasting colors or, as Marla Mallett has suggested, split symmetrical wrapping.

(ed., actually Klingner, as general as he is, said this with perfect clarity)

Finally, I would look for rugs that in addition to some of the details above,
contain some goat hair in the foundation,
multiple colors or shades of undyed wools and patterns that can be understand as having a relationship or origin in more restrictive weaves such as brocades.

These rugs seem to be part of a long continuum that, seen in a larger picture, hint at the scope and depth of a long tradition as long as that of weaving itself."

In addition to these indicators, Daniel mentioned in his initial essay that
weftless sumac is a structure seemingly used only by eastern Anatolian Kurds.

He also said that Marla indicates that only Anatolian weavers from the Malatya area use paired warp tapestry. So this is a likely traditional Kurdish indicator too.

So this is the sort of thing we should be looking for. It's going to be difficult since most of the available literature does not include (since most of these technical features have only recently been identified) the needed technical information.

So I'm a bit puzzled about how to go forward with sharing possible examples "in the Kurdish tradition."

Here, just to suggest the difficulty (not, mind you, with any claim to "advance" things seriously) are three bag face images from Julian Homer's 1988 sale catalog "Truly Tribal." Homer labels all three Kurdish. He says technical analysis was done on all of these pieces but he does not include it in the catalog. Ian Bennett provided some comment but it mostly focused on design and he did not comment on these pieces.

Can we tell whether these pieces are "in the Kurdish tradition" or not?

Regards,

R. John Howe

I quoted a longish passage of Michael's last post above but tried to highlight the indicators he listed in red so as to convert them into a kind of more easily seen listing.

Unfortunately, I forgot that when my Word document gets converted to HTML all that red highlighting is lost. This makes the longish quote of Michael's previous and available comments seem pointless. I may try to re-insert the red color if I can but I apologize for what surely seems like an unneeded, even a little mindlessly long quote of what Michael already said.

Regards,

R. John Howe

Marla Mallett has given me permission to quote her with regard to the direct scan I made above of a piece I thought perhaps was Kurdish because of the braided end finish but that others have indicated is Caucasian.

Marla: "The end finish you showed
is a two-pick interlaced band, and it is a type that is used not only by
Kurds but which appears very frequently in the southern Caucasus as well. It
is not a guarantee at all that the piece is Kurdish. What is on the other
end? Warp loops or a heading cord perhaps? Or another band?"

Me again: The end shown is that at which the rug began. The other end has what seems to be the same finish with the "nodes" of the braid being noticably larger.

Regards,

R. John Howe

The third bagface you show is from the Qazvin area where Mafi and Rashwand Kurds are settled since since the 17th century. According to Eagleton the Rashwand Kurds are may be a part of the nomadic Rashwan of Anatolia who migrated as far as Aleppo and even Damascus in the 18th and 19th centuries. An exactly similar example as yours is illustrated in Housego’s book – plate 40 -

Thanks for your participation.

Daniel

I fully agree with Marla.
You says :
The end shown is that at which the rug began. The other end has what seems to be the same finish with the "nodes" of the braid being noticably larger.

This is not unusual and the weaver certainly interlaced larger groups of warps
in this second band.

Best,

Daniel

Here are three more images. Two of them have good technical data. The third does not but I provide them as possible fodder for your considerations here.

The first is given that troublesome label "proto-Kurdish" and is estimated to have been woven about 1800.

I put it up not to raise the "proto-Kurdish" terminology debate again but rather because the rug has very good color, including multiple reds and blues, a green and an aubergine and the design has a kind of crude simplicity one could associate with the work of weavers in remote locations.

The description say "browns oxidized to foundations," approximately 10 ft. 6 in. by 5 ft. 7 in.

The technical description is:
Warp: wool, Z2S, natural ivory and brown
Weft: wool, Z2S, red, two shoots
Pile: wool, symmetric knot
Density: 7 horizontal, 9 vertical
Sides: remnants of cords, no original overcasting
Ends: incomplete, warp fringe
Colors: madder red, deep madder, deep rose, deep blue, medium blue, blue-green, green, yellow, aubergine, ivory, walnut.

The second piece is labeled a "Kurdish yastik, late 19th century, original flatwoven ends, partially oxidized browns.

Here is the technical description:

Warp: cotton, Z2S, ivory
Weft: cotton, Z2S, ivory, three shoots
Pile: wool, symmetric knot
Density: 6-7 horizontal, 6-7 vertical
Sides: flatweave with cord of two warps wrapped in white cotton
Ends: 1/2 " flatweave, warp fringe
Colors: ivory, madder red, rust red, rose, pale olive green, beige, walnut

I notice that all of the yastiks from Eastern Analtolia in the Morehouse book, seem to have wool warps and wefts. Does the presence of cotton in this piece make us question whether it is in the Kurdish tradition we are trying to explicate here? It has apparent multiple reds and a beige and a walnut that might be undyed colors.

The third piece is another from Julian Homer's catalog "Truly Tribal," and is included, despite the fact that the drawing in the field has a seeming workshop regularity, because the particular rendition of the kufic border here seems to me an instance of something the Kurds are often lauded for: their ability to take a design and to produce a rendition of it that is not just their own but sometimes aesthetically superior to the original.

Homer, you will remember me saying, gives no technical data in this catalog.

I find this rendition of the Kufic border more appealing than I do the more precise, hard-edged older versions that one sees, for example, those often found on "Lotto" rugs. Rightly or wrongly, I associate this kind of drawing with traditional Kurdish weaving.

I am not sure that these pieces provide springboards for suggesting further things about traditional Kurdish weaving but I offer them to make that possibility concrete.

Regards,

R. John Howe

About the last three rugs you posted. The firts one is well known and was advertised in SNY 12/12/97.
You are correct when you say that this rug is one of the so-called "proto-kurdish" group. Interesting in this rug are the flame like palmette showing Zoroastrianism influence.I discussed this rug in my Salon number 7 and related it to the Meyer Muller rug.

http://www.turkotek.com/salon_00007/salon.html

Not any of these rugs are "authentic" Kurdish weavings. They don't show any of the "heartland" characteristics we have already noticed.

Thanks,

Daniel

First Marla: Maybe I was being a little redundant. The treatment in the skirts or apron can varyb but often I see what I call overlay - underlay brocading. My recollection is that there is a Jaf face on your website that belongs to Allen Arthur that has this. Of course, the larger point is that while there are some techniques that only Kurds seem to continue to use there are other that are used and have been used by many peoples. In identifying a rug as being Kurdish, one must look at a variety of characteristics in most cases including color and handle as well as structure.

Second, John. I am sorry you interpret my comments as impatience. Getting ones hands around these rugs is a humbling process that I personally continue to struggle with, I certainly do not mean to do any thing to dissuade you from your own inquiry and journey. Post all the images you want.

Take care, Michael