Journal Home Page Attribution Guides Collector Exhibits Links

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.

Dating Tribal Rugs.

by Yon Bard

Throughout the following discussion, questions posed in bold-face are suggested topics for discussion

From the beginning of my interest in rugs I have been skeptical of dates assigned to tribal rugs. Contributing to this skepticism were observations such as the following:

  1. Auction catalogers purport to differentiate between ‘late 19C,’ ‘ca. 1890,’ and ‘fourth quarter 19C.’ How do they do it?
  2. Like dress hemlines, date attributions move up and down as fashion dictates. A few years ago, dating appeared to be getting more conservative, but this trend has now been reversed with, and even 16C figures are being bruited about.
  3. Authorities often disagree on the dating of the same piece. For example, the by-now famous Cyr rug – a Yomud main carpet sold for almost $100,000–is seen by many as the equal of a very early rug in the Textile Museum, but is declared by at least one reputable expert to be a late 19C copy. See also story of Ersari ensi below.
  4. Many people assume that the most beautiful expression of any design is the oldest, with the questionable implication that designs arise spontaneously in full perfection and do nothing but degenerate subsequently.

To allay my doubts (hopefully), I would like to initiate a discussion on how to date tribal rugs. Although I’d like this discussion to encompass other tribal rugs as well, what I have to say is largely about Türkmen, because (1) I know more about them, and (2) they appear to span a wider range of dates: while almost all collectible Shahsevan, S.W. Persian, Yürük, etc., rugs are attributed to the late 19th and early 20th century, a sizable number of extant Türkmen is attributed to the early 19th century and before. Why this difference?

There are several methods for dating rugs but they all leave something to be desired:

  1. Scientific methods. Laboratory analysis can establish the presence of chemical dyes, which dates a rug to no earlier than, say, 1875. (Some people claim to be able to do this by eye. Do you think you, or anyone else, can?) On the early side, carbon-14 dating can reasonably establish a rug’s date if it was made, say, before 1750. Recent results have shown that some Türkmen in all probability are at least that old (see HALI 104 page 82 report on Basel conference.) Do you accept these findings? Unfortunately, even if these methods worked perfectly within their domains, we would still be left with the problem of dating the bulk of the pieces that we collect–those made between 1750 and 1875.
  2. Provenance. If we could trace the rug to its origin in space and time, we would know how old it is. Unfortunately, the oldest Türkmen acquisition records are those of the Russian collectors (Gobolubov, Dudin) who followed their army into Central Asia in the late 19C. We know the pieces that were acquired then, but how old they were at the time is only known from market-place hearsay, not a very reliable source. Even in our literate society, legends about family heirlooms can quickly depart from reality.
  3. Woven-in dates. Unfortunately, these are rare in tribal weavings, and even if present they may be either undecipherable or not creditable. I know of only two dated Türkmen, both rather peculiar: One, a ‘Salor’ in the Gregorian collection, has an Armenian inscription with the date 1888; the other, pictured in the HALI report, has a Hebrew inscription with the date 1660. The carbon dating of this rug is rather inconclusive, and the ‘experts’ estimate it to be 19C. So much for woven-in dates! Does anyone know of other dated specimens?
  4. Pictorial records. We can date many historical rugs because they are depicted in paintings of known date, or described by travellers. Unfortunately, no such records exist of our items of interest. The few foreigners who visited the Türkmen before the Russian conquest (e.g., Vambery) were too concerned with preserving their lives to pay much attention to the rugs.
  5. Ethnographic studies. Sometimes it seems that known historical events can be correlated with structural and thematic aspects of the rugs. E.g., when a tribe resides near cotton-growing areas, its weavings may be more likely to have cotton in the wefts. Or, when two tribes reside in proximity, there may be cross influences in the designs found on their rugs. Rugs with such features can be speculatively dated to periods when these events occurred. A tour-de-force application of these principles can be found in Kurt Munkacsi’s article Dividing the Chodor (HALI 77 page 96). Unfortunately, Kurt has indicated that he himself now has some doubts about the accuracy of his conclusions, and what we know of the tribes’ movements is probably fragmentary at best. Furthermore, once features are incorporated in a tribe’s weaving traditions they persist long beyond the original influence.
  6. Structural and esthetic considerations. The most common method of dating these rugs by auction house specialists, dealers, and rug-book authors, is to correlate age from various features such as wool and dye quality, spaciousness of design, or shape of guls. Close students of the subject have developed a myriad criteria by which age can supposedly be judged. To quote an example, here is what George O’Bannon has to say about an Ersari ensi (Vanishing Jewels p. 107): ‘Several factors support an early date for this rug: the small size for an Ersari ensi, the simplicity of the "candelabra" patterns, the outer minor border of white rosettes, and the "tree roots" at the base of the center vertical panels.’ Now the question is: where do such criteria come from? If we have no objective ways of dating ‘model’ pieces, how can we tell whether these criteria apply? Indeed, doubts about these accepted shibboleths are beginning to be voiced. The HALI report of the Basel conference, for example, quotes Hans Sienknecht as saying ‘The earliest pieces do not show great perfection. Similar design does not necessarily mean a similar date… In particular, carpets with the least crowded fields are not necessarily the oldest.’

Given the fallibility of all these methods, a few questions come to mind:

What are we to think of all the dates in the auction catalogs?

Do we buy a piece because a ‘trustworthy’ dealer touts it as ‘probably 18C?’

Do we still feel confident in dismissing a piece as ‘late’ or accepting it as ‘early’ based on our own judgment?

Here is a case in point: Sotheby’s NY had an Ersari ensi (4/13/95 lot 95) satisfying most of O’Bannon’s above cited criteria. Sotheby’s duly listed it as ‘early 19C.’ O’Bannon himself thought he saw a chemical dye in it, and altogether repudiated his criteria. Another expert saw no chemical dyes, but said the piece was ca. 1875. When asked how he knew, he replied ‘you can tell if you have seen enough pieces,’ but how did he learn to date pieces from observing many rugs of uncertain age? Was he merely perpetuating traditional lore without solid foundations?

Well, enough questions. Let’s go and date some rugs! And if you have interesting example whose age you wish to discuss, please post them here.

P.S. According to the HALI report, the Ersari ensi fragment (Tent Band–Tent Bag p. 120) on which much scorn was heaped in Salon no. 5 (PIC 1) has now been carbon-dated probably to the middle of the 17th century, but no later than 1806.

[ Post Message ] [ FAQ ]

Post A Message!




Optional Link URL:
Link Title:
Optional Image URL:

Scripts and WWWBoard created by Matt Wright and can be found at Matt's Script Archive