Archive for December, 2007

The Anatolian Yastiks of Dennis Dodds and Zinaida Vaganova

Posted in Uncategorized on December 13, 2007 by rjohn

On December 8, 2007, Dennis Dodds, conducted a “rug morning” program at The Textile Museum on the subject of “Regional Differences in Anatolian Yastiks.”



Dennis is a long-time student and collector of oriental rugs and textiles and is currently, the head of The International Conference on Oriental Carpets.  He is one of those directly responsible for the splendid conference this organization held last spring in Istanbul.

Dennis collects seriously in a number of areas, but yastiks have been one of his favorites from the beginning.  He began to collect them in Anatolia in the early 70s, and has worked to highlight them ever since.  He organized exhibitions of yastiks at the ICOC conferences in Philadelphia and Istanbul.  And the only real book on yastiks was published by Brian Morehouse, in conjunction with the Philadelphia exhibition in 1996.  A large part of one volume of the catalog published on ICOC XI, in Istanbul, provides images and text on that yastik exhibition.

So Dennis in particularly well-placed to treat the yastik format.

He said, early, that all of his yastiks have never been presented together before and that he, himself, had not seen them all so arrayed.  I took as many photos as I could, without intruding on Dennis’ presentation, but can’t claim that some that he presented have not been omitted.

In this session, Dennis followed the broad categorization that Morehouse uses in his book, attributing yastiks to “western,” “central,” and “eastern” Turkey.  Sometimes more specific attributions are possible, but attribution is often something about which different plausible and defensible suggestions can be made.

As he introduced Dennis, Dan Walker, the TM Director, asked him to be sure to say what a “yastik” is (sometimes we rug people begin talking as if everyone lives in our world and speaks our language).

So Dennis began by saying that Anatolian “yastiks” are small textiles, that served as cushion covers.  Many Turkish interiors are furnished with divans and these are padded with cushions for sitting on and against comfortably.  “Yastiks” are usually seen to be the front-side of the cushions used to ease one’s back while sitting on a divan.

Dennis seemed to suggest that “yastiks” were basically a format woven only in Anatolia.  He pointed to the fact that most Central Asian bags were woven with the warps parallel with their shorter sides.  But cushions were used by many ethnic groups and one format, the “balisht” woven by the Balouch, seems very similar in size.  Balishts also have there warps parallel with their long sides as yastiks do and open at their narrower ends.

Here is the image of a yastik.


Yastiks vary in size, but are usually about 2 feet wide, and not much more than 3 feet long.  The one above is 2′ X 2’8″.  Yastiks can be either pile or flatwoven, but Dennis seems partial to pile versions.

Dennis next said, in his introductory remarks, that yastik designs seem generally to be divisible into two broad groups.  Those with designs that feature “medallion” devices and those that seem to draw on “textile” patterns.

The first piece he showed was a small rug (not a yastik) that Dennis suggested was the sort of thing from which the “medallion” yastik designs were derived.  Here is that rug.


This rug features a central medallion often referred to as of a “Holbein” type.  This description results from the fact that a 16th century German painter, Hans Holbein (the younger), used Turkish rugs with such medallions in his paintings so frequently that such medallions are now called by his name.

Dennis started with yastiks he estimates were woven in western Turkey.  The first five, he said, were likely woven in the Menderes valley in southwest Turkey.

Dennis estimated that the piece below was woven in the first half of the 19th century.


It is very well drawn and has a good range of pleasing colors.  It is unusual because it still has its back, which is also very colorful.


Another yastik with a seemingly Holbein format ancestry is the one immediately below.


This piece appears as the third image in the Morehouse catalog where he describes it has having features often seen in weavings from the Kutahya and Menderes River valleys, but also sometimes seen in Bergama area yastiks.

The next two rugs have similar medallion designs.  The first one below, Dennis estimated to have been woven in the Menderes Valley, near Dazgir, in the 18th century.


Again there is sophisicated drawing and a good range of good colors.

The second yastik, of these two, was estimated to the 19th century.  This is a design more pared down by abstraction and presented in more muted colors.


The main border in the piece above is one to remember.  Dennis found versions of it in other yastiks and had additional comment.  The abrash of the lighter blue field is a usage that is attractive to me.

The next yastik below was also estimated to have been woven in the Menderes Valley region around Dazgir.  This rayed medallion also appears in the 18th century Dazgir piece shown above.


Notice the treatment of the endings, top and bottom.  Red, flatwoven kilim ends are embellished with small, pile, polychrome decorations.  The strong yellow “jumps” at you a bit.

The next piece, below, features a “Memling” gul (a medallion device named for another western painter who often used rugs with similar devices on them in his paintings).


Dennis estimated that it was woven in the 18th century. I find the strong graphics of this small piece particularly attractive.

The yastik below was placed with the Holbein medallion group.


Another instance of good use of a somewhat wider color palette, and of sophisticated composition and drawing. Dennis said it is 18th century. Notice the unusual, delicate treatment of the lappets.

The yastik immediately below was described as from the Kutahya area.


This piece appears as the second one in the Morehouse “Yastiks” volume. The “saw-tooth” borders in contrasting light and dark colors are characteristic of pieces from this area. Note that the coloration is a little different from that of many “Bergama” weavings.

Dennis said that the piece below was woven in the Kozak area, a mountainous region north of Bergama.


The device on the white-ground main border of this piece often appears on Anatolian “heybes” (saddle-bags).

Two yastiks, exhibiting different medallions, completed Dennis’ treatment of pieces from western Turkey.

The first of these has a darkish ground in outer areas of the field but this is enlivened by three white-ground Memling guls down its center.


Morehouse suggests that this piece has border, palette and end finish features suggesting that it was woven in northwest Anatolia.

The last western Anatolian yastik Dennis had brought, was this attractive, squarish piece with a nice, white-ground border.


A relatively wide palette and simple spacious use of good color and strong graphics mark this piece.  The use of a distinctive blue in the medallion, the borders and lappets serves to integrate this piece, shaping perception so that it is experienced initially as a “whole.”

Dennis now moved to the group of his yastiks that he attributed to central Anatolia.


The first yastik in this group is estimated to have been woven in the Obruk area.


Two large octagan medallions on a red field are highlighted by projecting white latchhooks. The Obruk attribution is based in part on the similarity between the main border design and some used on Obruk kilims.

A second central Anatolian example below is attributed to Aksaray.


The central medallion of the yastik above is very similar to that on Number 119 in the “Yastik” catalog.  Morehouse places his piece in his eastern Anatolian group.  Dodds diverges from Morehouse about his own piece.  Because of its construction, palette and the many distinctive “cross” devices, Dennis relates it to Plate 33 in the “Atlantic Collections” catalog, which he sees as from Aksaray in central Anatolia (Mehmet Cetinkaya concurs on the basis of fragment of a nearly identical carpet that he owns).

Another nice little piece with a Holbein-derived medallion is this one.


I like its spacious simplicity. Trying to clarify how best to describe it I wrote to Dennis after this session. Here is his response about this piece and its estimated age. I quote:

“This yastik displays a medallion that is seen in a carpet fragment from the Beyshehir trove, now in the Mevlana Museum in Konya. “In The History of the Early Turkish Carpet, Kurt Erdmann illustrates that carpet in fig. 70 with the caption: ” ‘Holbein’ type I carpet fragment from Beyshehir. Mevlana Muzesi, Konya.”

“In Carpet Fragments, Carl Johann Lamm discusses “Anatolian Holbein Carpets of the 15th Century,” pp. 50-62. He illustrates two fragments with medallions of this design in Color Fragments 26 and 27, on pp. 85 and 86 which he dates to the late 15th century.

“By the 16th century, this medallion had moved from an overall repeat pattern to a solitary and central position in the well-known group of rugs referred to by Erdmann, op.cit., fig. 37, as ‘Ushak prayer rug of the end of the 16th century with opposed prayer niches…”

“In Oriental Rugs from Atlantic Collections, our yastik is illustrated on p.21, plate 21, where the date is shown as “c. 1800,” with the caption referring to its probable design origins in the “small-pattern ‘Holbein’ medallion” type.

“For publication, this is a reasonable date for the piece.”

And in a second indication about attribution of this piece, Dennis wrote:

“I think this yastik should be placed in Central Anatolia and the Konya/Karapinar area, owing principally to the lack of outlining in the major design elements, i.e., the ‘kilim style’ that Dr. Mae Beattie identified as a style predominantly used in that area.”

(End of Dodds’ quotes)

The next yastik Dennis presented was the one below.


This handsome piece has the same design as another that Morehouse selected for the cover of his “Yastiks” catalog. That piece also appears as Number 55 in that volume. Dennis’ piece above is Number 56.

There has been some interesting comment about this design. As it happens, I own what is likely a later version of it and when I published it on the internet, Michael Bischof, long involved in the natural dye movement in Turkey wrote me to say that this design is claimed as their own by a village near Karaman in the Karapinar area. This village is predominantly Yomut Turkmen who moved to this area over 200 years ago from Khorusan in Iran. Morehouse mentions aspects of this connection in his catalog comments on these two pieces and once acknowledged to me in conversation that there might be something to Bischof’s claims.

This design is also an occasion for a discussion of whether designs tend generally to become more conventionalized as they come forward in time.

Dennis argued, on the basis of some pieces estimated to be older but with quite simple, even abstracted designs, that simplicity of design may be an indicator of an earlier period, when other factors are also considered. But it can be unreliable. Especially when discussing yastiks, their small size limits the amount of design elements that a weaver can use, so even later yastiks might display simple designs.

Morehouse argues explicitly that this particular design becomes more simplified in younger pieces. The number of leaves in the lappets moves from four to two; the cruciform devices inside the “insect” devices in the borders lose their definition; and the use of filler designs in the field increases. (In a subsequent post I will examine five examples, including the two mentioned here to see what they suggest.)

Dennis estimates this Karapinar yastik to the 18th century.

The next three central Anatolian yastiks Dennis had brought featured two medallions with cross motifs.  The piece below is the first of these.


The abrashed light green-ground medallions are effectively placed on a wider red field the latter emphasized with latch hooks in a distinctive blue.  Dennis also sees this as an 18th century piece.

A second instance in this small design grouping is the one below.


Dennis estimates it as having been woven in the 19th century.

And the third is this one.


This example is seen as late 19th century.

To the right was the yastik below with a three, stacked device design.  This piece may be attributable to the Cappadocia area.


Its border was described as having “kufesque” elements.

At the top far right of his central Anatolian examples, Dennis provided the white-ground piece below.


Among its noteworthy features are its long hexagonal medallion, an, almost certainly, cochineal red, and dark wefts.

Dennis described the pieces in his lower row of central Anatolian yastiks as instances of “all-over” designs.

The first piece in this grouping is the one below.


Here is the second.


Dennis described these two “all-over” repeat designs as of the “textile” variety. He suggested that they have roots in Seljuk usages. He attributed them to the Gelveri area or, perhaps, to Aksaray.

The vertical stripes in the next piece clearly indicate that this is a “textile-derived” design, the second of the two design groups Dennis stipulated earlier.


The colors are bright and clear and seem quite different from many yastiks we have seen. This piece is attributed to the Kirsehir area.

To the right of the piece above was another with wide borders and striping in its small field. Dennis said this combination is a central Anatolian indicator.  This piece also has more subued colors.


This yastik appears as Number 82 in the Morehouse catalog where the description notes that it has attribution features that point in several possible directions. The use of lappets is characteristic of Mucur, it has colors that suggest Kirsehir, but its softer palette points possibly to the Sivas area.

The piece below is attributed to the area south of Konya.


It  has a soft wool, with a long pile and wide selveges. Its designs are taken entirely from border patterns.

The next central Anatolian piece above was another white-ground design.


It features small repeating medallions embelished with pinwheels.

The last central Anatolian yastik Dennis presented was the one below.


The design a series of repeating squares each with some vague internal instrumentation approximating an “X.”

Dennis now moved to eastern Anatolian pieces in his yastik collection.

(Note:  In the interest of complete accuracy, the photo immediately below is, in fact, from a little further on.  The lower row here are not from Dennis’ collection, but rather are yastiks that members of the audience brought.  I just needed a nice transition picture and this is it.   Just so folks don’t write me about this “mistake” later. 🙂  )


The first of Dennis’ eastern Anatolian pieces is the one below.


Dennis described the piece above as a “rug” design: a medallion surrounded by “leaf” forms that seem near “armatures.

The second is this one.  It is estimated to be from the same approximate region as the one above: Mudjur or Sivas.


Dennis indicated that the palette of these two pieces is much like that of some in the central Anatolian group.

The next piece in the eastern Anatolian group was this one.


This piece appears as Number 111 in Morehouse’s catalog. He says its composition with “boxed medallions flanking a central hexagon” is unusual.  In his discussion of it, Dennis also called attention to the “square boxes with hooked edges.”  Morehouse is complimentary about this piece and thinks there is a suggestion of a date below the top medallion.  No precise attribution is attempted.

The next piece features an elaborate “flower-form” medallion on a plain field.


The red ground seems likely from cochineal and the floral medallion seems almost European.  Dennis  said there was an example of this group in a recent Grogan auction that had an inscribed date of 1877.  He also said that this piece may have been woven in the Gordes region where the Medjidiyeh style was very popular in the third quarter of the 19th century.  Someone in the audience suggested that it may have been woven by an Armenian weaver.

Morehouse offers seemingly similar pieces in his Numbers 97 and 98. He says they project a “Mejidiye aesthetic” and relates them to Numbers 95 and 96, attributed to Kirsehir. For this reason he places his examples in central rather than eastern Anatolia.

Dennis’ next eastern Anatolian piece was this one.


This yastik appears as Number 116 in Morehouse’s catalog. Dennis described it as from Malatya with a likely cochineal red. He said that the steeper angles in the diamond designs were created using offset knotting, something that suggests that its weavers may have been Kurdish. He also pointed out its herringbone selvedges.

The piece below appears in Morehouse’s central Anatolian grouping, Number 102 on the bases of palette and the paired flower and rosette in the main border.


But Morehouse admits that on the bases of its multiple stacked and hooked medallions and its multiple selvedges, it could defensibly be placed further east in the Sivas region.  This is, apparently what Dennis has done, for he includes it in his eastern Anatolian grouping.

The last piece on the far right of Dennis’ top row of eastern Anatolian yastiks was this one.


Dennis attributed this piece to Kirsehir or east of there. It has a distinctive pale yellow ground main border like the piece directly below it on the board.

This lower piece appears as Number 105 in the Morehouse catalog.


Morehouse says that its palette and several of its border patterns are characteristic of the Sivas area.

The next piece in Dennis’ eastern Anatolian grouping was the one below. He attributed it to Mucur. It appears in Morehouse’s catalog as Number 107.


Morehouse comments on the “bird’s head motif” in the main border and the “eight-petaled rosettes in the field grid.”

Dennis’ eastern Anatolian pieces included this yastik with rich and seemingly unusual colors.


This piece was published as Number 138 in Morehouse’s catalog, where he says that the combination of red-orange shades with blue or blue-green “is characteristic of weavings produced in Malatya.”  He adds that the “S” border is also seen “throughout” this area and “to the north.”

Bob Emry, who took a good set of notes in support of my efforts here, describes the piece below as “funky,” which it assuredly is.


Dennis suggested that it is Kurdish.  He also offered an interesting possible interpretation of the three squares within the compartments.  Dennis feels that these may be abstracted residues of the Ottoman “cintamani” design that has three “balls” arranged in a similar triangular fashion.

Here is a black and white image of the cintamani design on the fabric from a garment of a known Ottoman sultan, of the 15th century.


My feeling in the TM sesssion was that, while Dennis’ suggestion is imaginative, to be fully convincing it would need to account for the seeming absence of the “wavy stripe forms” that occur beneath each set of balls in the Ottoman version. Perhaps the base of the compartment could be seen to play that role.  It is also true that weavers sometimes took only parts of designs into their pieces.

In fact, Dennis argues that there is evidence that the cintamani design was so adapted.

In a subsequent conversation he suggested why the lack of wavy lines may not disqualify his cintamani interpretation of the piece above.  He talked further about his suggestion that “the small squares, arranged triptically, are stylized versions of the round cintamani balls, made square because of the coarse knotting.”


He said:  “[ed. The image above] of a “semiantique Paotao Chinese saddlecover” is from a recent Bukowskis auction in Sweden and shows how the 3 balls can exist independently from the stripes. This arrangement, sans stripes, is also observable in 16-17century Ottoman textiles and ceramics.”

This horsecover design seems to me a pretty convincing instance of Dennis’ claim.

The last of Dennis’ eastern Anatolian yastiks was this one.


It is Number 131 in the Morehouse catalog. Although all of the devices in this piece are from the standard Anatolian repertoire, Morehouse calls attention to how unusual their comoposition is. The design retains device usages that would be necessary in kilim weaving but which are not in a pile piece. There is an indication that one such design usage is found in Malatya kilims.

Well, that’s the end of Dennis’ TM Rug and Textile Appreciation Morning presentation on his Anatolian yastiks.  Dennis took questions and then folks moved to the front of the room and after-session conversations started.

In the image below, Wendel Swan talks to Dennis about a particular piece.


My thanks to Dennis for permitting me to share this with you on this virtual basis and for saving me from some reporting errors.

My primary purpose in this internet post is to make it possible for more people than the number who can fit into Mr. Myers’ former living room to enjoy these valuable, interesting TM programs.

I am not particularly attempting to foster conversation, but if you have a comment, or question, and/or some images that seem congruent with this presentation that you want to share, send it/them to me at together with your permission to quote you in part or wholly.  I may use the materials submitted in subsequent posts I make regarding this yastik session.

You are also encouraged to forward this URL to others whom you think might enjoy this TM “rug morning” program.


R. John Howe (further comment below)


Further comment on Dennis’ “cintimani” suggestion.

You will recall that Dennis suggested that the “three squares” design of the Kurdish yastik below


was possibly an instance of the “cinitmani” design albeit without the stripes/lips that are often seen to be part of it.


In support of his conjecture Dennis provided an example of a saddle cover that has the “three triangulated balls” of the cintimani design without the stripes/lips.


I have received two comments bearing on this suggestion.

The first is from Bertram Frauenknecht, saying:

“…the Cintamani is considered a fertility symbol, to be found under Ottoman rule. The version without the lips is much older. It is called ‘cat paw’ – design nowadays.

“I found the oldest version on the dress of a female Etruscan dancer, 600 BC, in an Etruscan tomb north of Rome, Italy.


“It is said that the Etruscans came to Italy from Anatolia.”

[Ed: Bertram continues: And the image below] “shows a Karapinar fragment from the 15th century, which…[ed. has] the ‘cat paw’ on the white field.”


My thanks to Bertram for these two interesting further examples.

And Brian Morehouse writes, saying first that [ed. It’s] “always good to hear that Yastiks are being discussed and viewed.”

But he is skeptical of whether the fact that the cintimani design sometimes occurs without the stripes-lips, can be taken to indicate that that is what the weaver of this particular yastik intended.

Here are his words: “…With regard to the issue supporting Dennis’ cintimani concept: All things are possible, but the chance of a humble villager even seeing a similar motif is highly unlikely given the rustic nature of the piece.

[Ed.]”…as rug collectors we are allowed the privilege of having an imagination, but as scholars we must be more skeptical, [ed. while still] not devoid of possibilities…Whatever the design may represent all that concerns me is….is it compelling as a woven work of art? I think it is so unclear whether this design is representative of the cintimani design that it should simply stand on its own merits and if it transfers that design in a meaningful way all the better. I will leave that up to the viewer.”

My thanks to both Bertram and Brian for these interesting additions.

R. John Howe

Click the link below to return to the home page and other posts listed on the right.