The Tribal Gol in Turkmen Carpets

by V.G. Moshkova

Adapted from Turkoman Studies 1


Reproduced here with permission by
Humanities Press/Oguz Press,Michael Franses & Robert Pinner





FOREWORD

by Tom Cole

I have discovered through working on re-creating this article and the reaction of the few who previewed it for typographical errors and aesthetics that many people with a real interest in Turkmen weavings have not thoroughly read Moshkova's text, myself included. Skimming the text and picking up on some of her theories is different from actually reading it, word for word. I realize now that the colour illustrations accompanying this reproduction assist the casual reader a great deal.

Studying this text of Moshkova's observations and studies is not without interest, even at this late date. Some of her conclusions are obviously wrong, while others are of lasting value. Her live gol/dead gol theory is somewhat interesting but probably flawed. Her detailed examination of the ertmen gol and its degeneration is probably correct, though demeaning its appearance on torbas and trappings is, again, probably incorrect. Identifying the patterns seen on smaller bags and weavings as a means of the weavers' personal expression outside the codified strictures of tribal design and gols is obvious and undoubtedly correct. Moshkova's assertion that the kepse gol is seen only on Yomud bags of recent origin is apparently incorrect as well. While old examples of chuvals or torbas with this ornament are rare, they are not unknown. For all the errors and dated material presented here, her contribution to Turkmen rug studies is still with value, and, thus, presented here as it has never been seen before, fully illustrated in colour

The translation of Moshkova's original text in Russian is somewhat clumsy, but, for the most part, I have left it as it was originally published, with only minor changes, ie. 'Turkmen' instead of 'Turkoman', and 'Ashgabat' instead of 'Ashkabad'. More familiar terminology has been added in parentheses, ie. ukudj (okbosh) and osmulduk (asmalyk).

I hope this presentation will help those who are just now discovering an interest in Turkmen ethnography, culture and weavings, making the Turkmen aesthetic that much more accessible.



Nowadays it would be rare to meet a carpet specialist who is unfamiliar with the appearance of Turkmen carpets. The world-wide fame of these carpets is well deserved, not only for their exceptional artistic qualities but also for the charm of their simple rhythmic designs and composition and their unusually rich range of colour shades. The local inhabitants of Central Asia were, however, far from imagining the role which carpets and carpet products play and have played in the land of whose population produced them.
 Carpet making in Turkmenia is true folk art. With few exceptions the entire older Turkmen population knew the art of knotting carpets. It is only when you have seen a Turkmen woman at her loom, watched her quick hands flying like birds over the weaving of her carpet, witnessed the perseverance and energy she brings to her work and the dynamic strength of the whole process, only then can you understand how superb a worker the Turkmen carpet maker is - the uncrowned queen of Turkmen folk arts.



Plate 1. "The wedding caravan which brought a rich bride from her parental home to her new life was always fitted out with numerous brightly woven decorations. The camel's neck were decked with woven collars; the leading camel was also adorned with a five-cornered rugs, the osmolduk (asmalyk), decorated with richly luxuriant tassels. " - Moshkova


With enormous care and love, and using a simple but unusually artistic pattern, the Turkmen woman made and decorated with her own hands even the most primitive items in everyday use. A knotted curtain, the ensi, was hung at the outer entrance of the yurt, to protect against the burning rays of the sun, against the cold and the dust afghan which intruded everywhere; a small rugs, the germech, lay across the threshold.

Decorative woven strips, yolami and boi, strengthened the wooden frames of the hut, and encircled on the outside, its roofing, enlivening the cheerless grey tone. The earthen floor of the hut was softly covered with patterned koshma, double-sided woven carpets, and carpets called khali; the family gathered round the hearth and chance guests sat on the woven odzhak bask and dip khali.
 Household utensils and all domestic goods and shattels were kept in woven torba and chuval suspended from the latticed walls of the hut.

Small loads taken on journeys were placed into double-bags, khordjin, which were thrown over the saddle; the horses of the Akhal Tekke and Yomut tribesmen were covered with woven saddlecloths, taynakcha, and embroidered horse and saddlecloths, cherlik. Even the donkey received its share of attention from the carpet makers: woven donkey saddle-girths are not unusual in Turkmenia. Especially useful during the journeys and at the stopping places were the diz-torba, knotted bags in which salt was carried; there were bags for mirrors, a rare possession, aina kap; for spoons, chemche-torba; spindles, iksilik; and so on.




Plate 2. Assorted weavings of the Yomud that do not adhere to the strict design format dictated by tribal tradition, but rather were left to the individual whim of the weaver. Included in this were knee covers (a) for the camel in the wedding procession, utilitarian weavings such as a spindle bag or knife sheath (b), the 'okbosh' (c). a gun cover or scabbord (d) and chanteh (personal bag (e)..



The wedding caravan which brought a rich bride from her parental home to her new life was always fitted out with numerous brightly woven decorations (Plate 1). The camel's neck were decked with woven collars; the leading camel was also adorned with a five-cornered rugs, the osmolduk (asmalyk), decorated with richly luxuriant tassels. Another small rug, the khalyk, was placed on the camel's chest and its long tassels almost reached to the animal's feet. These woven decorations harmonized beautifully with the bright embroidered cloths which were thrown over the neck and head of the camel.

Special splendour was added to the wedding procession by the bundles uk (roofing poles) for the wedding yurt. Interwoven with woolen braid, these uk swayed to and fro in woven festive containers, ukudj, reminiscent of huge warriors' quivers of arrows (okbosh).
 Along with all these decorative items, the young wife brought to her new family home her dowry: chuval, torba and, of course, carpets. Their quantity was strictly prescribed by custom. All these woven things were made with particular love and care; their high quality was regarded as evidence of the maturity of the young girl, and brought honour to the mother who had taught her daughter a high standard of artistry and craftsmanship. On the arrival of the young girl into her new home all the woven goods were hung in the yurt, special prominence being given to the khalyk which had adorned the chest of the wedding camel and now hung over the entrance. As time passed this khalyk became a special decoration, a kind of lambrequin on the door, the kapunuk.

How could the Turkmen not make the most beautiful carpets in the world, when their first steps in childhood were taken in a child's rug, the salatchak, and the departed were mourned and taken on their journey in a funeral carpet, the ayatlik? (Footnote 1)



Plate 3 a., b., c. "... it has been possible to establish a whole category of 'religious' designs, recreated quite clearly by carpet makers throughout Turkmenia, to which a religious-magical meaning is ascribed."

Plate 3 d. Detail of a fantastic kapunuk with a variety of tertiary devices on the white ground consisting of amulets or fertility symbols. Elena Tsareva has speculated that the triangular forms with prongs extending downward symbolise clouds shedding rain nourishing the earth. Russian Ethnographic Museum, St. Petersburg



Turkmen carpet are, so closely bound to the life of the people, is not only a most valuable memorial to their creative skill. It is also a rich source for the study of their ethnography and history.
A compartive study of the individual ornaments enables us to trace the changes which took place over the years and to discuss their historical development. Alongside some obviously new carpet patterns (sometimes borrowed from other cultures) are very ancient designs which have survived of their specific significance. If one recalls that in the past, as now, the decorative art of the Turkmenia was in the hands of women, the preservers of the old traditions, and was always most closely linked with the narrow domestic circle, then it is not surprising that one finds in the carpets features of great antiquity.
 The presence of all these living elements in the Turkoman carpet creates a genuine desire in the researcher to discover what the carpets can tell us of the complex history of the people.

In the researcher's attempts to discover the meaning of individual carpet designs and composition, the most useful information - apart from that gained by the comparative study of the carpets - is that which is obtained from the carpet makers themselves.

Experience has shown that material of this kind, carefully checked many times and thoroughly evaluated, can help us to discover the original content of some of the ornamental complexity in Turkmen carpets.



Plate 4. " In a central position in the ensi is a design called 'trace of the puppy' kuchukizi, an image of a dog as a sacred animal in the lives of the Turkmen. Also in the ensi is the pattern mihrab which clearly has a protective meaning." - Moshkova


By this technique it has been possible to establish a whole category of 'religious' designs, recreated quite clearly by carpet makers throughout Turkmenia, to which a religious-magical meaning is ascribed. An example is the design which bears the name gul-aidi (Plate 3 A). This always appeared on funeral carpets, double-sided rugs and children's rugs. The design dagdan, illustrating an amulet famous in Turkmenia, has the same meaning (Plate 3 B). The extent to which this representation remains alive can be judged from the fact that it is found on almost every carpet made today. It is inserted at the very beginning of the carpet on the first ornamental border. Many patterns of this kind appear on the ensi (door hanging) which undoubtedly belong to the group of archaic carpets which conceal many unexplained features. In a central position in the ensi is a design called 'trace of the puppy' kuchukizi, an image of a dog as a sacred animal in the lives of the Turkmen (Plate 3 C). Also in the ensi is the pattern mihrab which clearly has a protective meaning .(Plate 4) We can see how the silent patterns of the carpet come to life if examined from the view of the people who made them.

Even now the Turkmen, especially the women, understand clearly the tribal origin of carpet ornamentation. Each tribe has its own particular stock of carpet patterns. These concepts of the tribal origins of carpet patterns and composition have been taken up more or less correctly by European researchers who are interested in Turkmen carpets, and have found their way in to the typological classification of carpets in the literature which distinguishes between Tekke, Pende (Salor), Yomut carpets and others.

The carpet ornamentation of the Turkmen represents a single stylistic group, despite local and tribal differences.



Plate 5a. An example of Ersari weaving (detail from the border of a fragmented main carpet) depicting a very Persianate form of the ubiquitous boteh design. It is only the curvilinear lines of the boteh which suggest a Persian influence. Actually, this image is very realistic given the bird-like appearance of the boteh, complete with a distinctly beaked head looking back over the shoulder, a classic image seen in ancient textiles and weaving from ancient Inner Asia. Ersari Carpet (fragment), 18th century, Private Collection, USA

Plate 5b. Below, another example of Ersari weaving exhibiting a pattern thought to be derived from silk ikat weaving. Given the association with Bokhara, it is not surprising to see the Ersari women weaving patterns that have nothing to do with the traditional design pool of the Turkmen. 19th century, Private Collection



Moreover, even a rapid glance at the carpet designs of modern Turkmen tribes, including those which have their own very pronounced tribal features, shows the many common elements. This is not always a simple case of one tribe borrowing a design from another; on the contrary, the designs show us the traces of tribal groups which are now separate or, sometimes, of the merging of groups which were once separate tribes. Similar motifs of this kind can be found in the designs of the Tekke, Yomut, Ensari and Saryk tribes, all of which, as we know from history, are linked by a common origin.

Among all the shared features which bear witness to the common culture of the Turkman carpet, tribal differences in modern carpets stand out sharply. Tribal ornamentation is most 		distinctly represented in those Turkmen tribes which have most strongly preserved their own integrity, tradition and territory. Among these must be included be included such groups as the Merv and Akhal Tekke, and the Yomut.

The greatest loss of tribal patterns and the intrusion of foreign elements took place among those tribes whose tribal organization disintegrated. For example, the Ersari, who lived for 300 years near Bukhara under the immediate rule of the Buhkara khanate, lost their tribal traditions to a significant degree. In Ersari carpet ornaments more than in those of any other tribe, we can feel the diffusion of the many alien elements to which they were exposed. (Plate 5 a, b) The same can be said about the Chodor.




A Turkmen woman spinning wool, Central Asia. Photograph by SM Dudin, circa 1902


The Chodor, well known as amongst the most ancient of Turkmen, were mentioned in the lists of the old Oghuz tribes of Makmud' Kashgarski and Rashi-al-Din. A study of the ornaments used by the Chodor is therefore of special interest. When the Chodor settled in the Khiva khanate surrounded by many other tribes, they lost to a significant degree the individual features of their tribal organization, and the tribal connection to those Chodor who had settled in different territories. Mixed marriages, even with non-Turkmen (Kazakh) took place among the Chodor more than among any other tribe of Turkmenia. For the Turkmen, this was exceptional, even in the recent past, since up to the present day, they have maintained strong marriage traditions within the bounds of their own large tribes. During the period of subjugation to the Khiva khanate, particularly after the last invasion by Russia, the Chodor became strongly involved in the trade of the area. It was the Chodor tribe who introduced carpets to the markets of the Khiva and sometimes the Bukhara khanates. This was reflected in their own carpet ornamentation.



Plate 6. An example of the 'dyrnak gol' of the Yomud groups as seen on main carpets


The original Chodor ornament, called ertmen, became deformed in a short time, within some 20-30 years (Plate 26, 29). Moreover the Chodor also adopted a series of patterns from neighbouring tribes, such as the Arabatchi, and used them in a new way: this new pattern has the appearance of a medallion filled with the letter "H" (see, Plate 8 caption for clarification & corrections). Widely known in Central Asia, this ornament can be seen in the ornament can be seem in the carpets of literally all chaykharakh. Over these years the Chodor's own original pattern, the ertmen, ceased to be used, and was preserved only in old, small pieces.

 Thus Chodor ornaments underwent great changes and took on many new elements. In determining which ornaments were borrowed from one tribe or another, the names, which were oftern preserved, help to indicate the place or tribe of origin. Thus there are designs passed to the Yomut from the Tekke tribe, called by the former Tekke "naksha" (i.e. Tekke pattern). The name mar-gol denotes a Merv design used by the Kizil Ayak tribe, who borrowed this pattern from the Merv Saryk or Salor. The Ersari have a pattern called kharatiy derived from Herat. There are a great number of similar examples of this kind: in the stock of designs of nearly every tribe there are ornaments whose names indicate their origin.



Plate 7a. (above) The Yomud 'kepse' gol, detail of a very old main carpet, 18th century, Private Collection

Plate 7b. (below) An old Yomud group chuval with the rare depiction of the kepse gol design, seemingly contradicting Moshkova's theory that this gol appears in only late bags and trappings While certainly not as old as the rug featured above, it is old, probaly circa 1860 or so, a solid 3rd quarter of the 19th century piece. Though not of the early classical period, I still think it may be presumptuous to call its appearance in bags such as the one pictured below as 'late' and removed from the tradition..



The prayer rugs, bags (torba) and the many other small items in everyday use are the pieces that show best those elements that are foreign and those that were created by that tribe. This is demonstrated by the large variety of ornaments seen on pieces made by the different tribes. Evidently it was making these small pieces that the Turkmen woman felt free to use her powers of invention and was least bound by tribal custom (Plate 2 ).
 It is clear that the stock of designs employed in Turkmen carpet are is very complex and can be analyzed only by painstaking study. It needs careful comparison of the individual ornaments used by the different groups of Turkmen tribes on the basis of the ethnographical and historical materials we posses and on the way of life, the relationships and the places of settlement of each tribe.



Plate 8. The Arabatchi "tauk nauska" gol (above). Though Moshkova claims the 'H' form of this motif without the suggestion of animal heads to be old, we know now that it is a later version of this older form. A Chodor version of this gol is seen below with a very colourful palette suggesting real age, circa 1850 or before. Interestingly, the minor gol is one associated with the Yomud, the 'erre' gol, more familiar when seen on Yomud chuvals, but seldom used on main carpets of either the Yomud or Chodor. To associate this gol type with any particular Turkmen tribe is dangerous and possibly erroneous. It has been used by more than one tribe dating to antiquity, including the Arabatchi, Chodor and Yomud as well as by non-Turkmen Central Asian weavers including those of the Uzbek, Karakalpak and Kighiz tribes. The 'tauk nauska' device is also incorporated into the original primary gol distinguishing Salor main carpets (Plate 9 c). Possibly, Moshkova had not seen a Salor main carpet with this gol at the time she offered her commentary.



It has already been shown that among the carpet ornaments of each tribe, with the newer elements, sometimes recently borrowed, there is a stock of very stable forms. Such changes as these ornaments have undergone over the years, have taken place slowly, without altering the general composition or the main design elements.

The more stable ornamental forms appear mainly in the patterns placed in the central area of the main floor carpets and are considered the basic tribal patterns, a fact which is borne out in many cases by their names: amongst those we know are the Tekke gol, the Saryk gol, the kepse gol (Plate 7) and the dyrnak gol (Plate 6). The last two are believed to be the basic tribal patterns of the Yomut. Among the Ersari tribe that gulli-gol and the temirdzhin (temirchen) gol are well known in the form of several variants (Plates 10 & 11). The Arabatshi tribe regard the tauk-nuska as their tribal pattern (Plate 8).
 Examination has shown that all the Turkmen tribes who engage in carpet making and who have preserved their tribal traditions to some degree have their own tribal gol, which appear most frequently on large carpets. The tribes who lost their independence, by settling in separate groups among other tribes, while remaining carpet makers, lost their tribal gol. Examples of this are the Ikdir (Igdyr) and Imreli tribes who use Yomut patterns, since they live in the immediate neighborhood of the Yomut and in the past were dependent on them. In the same way the Ali-Eli people, who live among the Tekke tribe, use Tekke designs, as do the Nukhurli who in the past were also dependent on the Tekke tribe, living in the Nukjur settlement near Bakharden.



Plate 9a. (Above) The classic 'chuval' gol of the Salor used on these large bags with majenta silk. Possibly early 19th century. Silk used in such profusion is seldom seen in 18th century Turkmen weaving.

Plate 9b.(Below) Another Salor gol seen on chuvals and one appropriated by the Saryk and used extensively on their own carpets and bags in the late 19th/early 20th centuries. Details, Dudin Collection, Russian Ethnographic Museum, St. Petersburg.




Plate 9c. (Below) The Salor gol seen on the main carpets of this tribe. Note the tauk nauska elements in the center of the gol, and the floral motifs incorporated, similar to the Ersari 'gulli-gol' seen in Plate 10. Ex-Robert & Lesley Pinner Collection, 18th century Obviously this gol was never reproduced on chuvals by any tribe upon their defeat by the Tekke (circa 1842). Moshkova's reference to the appropriation of the Salor gol by other tribes includes the two gols seen above (Plates 9 a, & 9 b). Late Saryk weavings, both carpets and chuvals, from Pendeh and Afghanistan reveal the use of these gols, but the gol of the Salor main carpets has not been regularly used in this form by any other tribe.



The indications are that when a tribal lost its independence it was deprived of the right or the ability to produce its own design and that the use of the gol of the dominant tribe was then a symbol of subjugation to that tribe.

Particularly significant in this respect are the so-called Evlyad tribes (Shikh, Makhtum, etc.), who lives as 'holy' tribes among various Turkmen groups : they are often excellent carpet weavers and they invariably use the tribal designs of the group which shows them hospitality.

The stability of the gol is explained by the fact that they very clearly reflect the emblems of their tribe. Probably their use was at some time strictly regulated, and later became a long-lasting tradition. The examination of this gol is the main theme of this article.
 The gol must not be confused with the similar sounding term 'gul' or 'gul'cha'. To the Turkmen carpet weaver, they are very different. 'Gul' means flower, 'gul'cha' little flower, a certain shape of pattern which in outline resembles a flower. The term 'gol' can be applied only to the tribal ornaments placed in the centre field of the carpet, less often on other pieces.

There is no satisfactory translation of the word, 'gol'. The interpretation of the Salor gol as the Salor 'rose for example must be rejected. The carpet weavers themselves understand the term gol simply as design or pattern. (Footnote 2)

For want of a more satisfactory interpretation we consider the carpet gol in the sense of a tribal ornament or pattern, perhaps an emblem, and we must try to reveal its pictorial form.



Plate 10. The "gulli-gol" of the Ersari, seen exclusively on the main carpets of this Turkmen group.


When we regard the gol as the emblem of the Turkmen tribes and of the peoples who over the centuries grew, lived and died, we would expect to find a reflection of the living process in the carpet designs themselves. Our observations have led us to the conclusion that within the large variety of Turkmen medallion designs we can indeed find 'live' gol and 'dead' gol. The live gol are the gol of those Turkmen tribes, which in the recent past had not lost their importance and among these can be numbered, the Ersari, Yomut, Saryk and Tekke tribes, all of which have preserved their tribal gol. These gol persist in the shape of the basic field ornament of the main floor carpet; the principle carpet displayed by the Turkmen. Probably the fact that the main carpet had this important display function explains why it is here that the gol of the living active tribes is preserved. In the past this gol appears to have been used on no other piece. This tradition of the tribal gol has been preserved in the Tekke tribe: until the recent past the Tekke gol has not been found on any article in everyday use except the main carpet and, according to all the Tekke carpet weavers, the great tribal gol of the Tekke tribe was strictly reserved for this carpet (Plate 22). The Ersari tribe has the same tradition: the Ersari tribal gol are never found on small piece but only on main carpets (Plate 10).



Plate 11. The "temirchin" gol of the Saryk, an interesting gol that Tsareva identifies an element of with fish. Curiously Moshkova identfies it as an Ersari ornament. 18th century, Private Collection, USA


However there are also cases where gol from large carpets are transferred on to the small items. The Yomud kepse gol (Plate 7) and dyrnak gol (Plate 6), for examples, can be found on bags of recent manufacture, although they are not shown on old bags. Nor, as we have indicated, has the Chodor tribal ornament, the ertmen appeared on carpets for along time: it has been replaced by the lighter tauk nuska gol borrowed by the Chodor tribe from the Arabatshi tribe (see caption, Plate 8). The ertmen was transferred entirely to small items, chuval and torba, but although it has been widely used on such items, the carpet weavers still consider it as the Chodor tribal 'gol'. In the same way the famous Salor gol was transferred to chuval. The Salor gol lost its significance as a tribal ornament when the tribe, having suffered defeat from the the Tekke and Saryk at the end of the 19th century, ceased making carpets altogether. At this time, the famous Salor gol was taken over by the Saryk and reproduced by them in modernized form on chuval (Plate 9 b, 12), still however, preserving the old name Salor gol. The reproduction of the Salor gol on carpets of the Saryk and Merv Tekke occurred in the colonial period, and must be seen as Russian influence, since Russian custromers created the demand for carpets of this design.



Plate 12. The chuval gol as it is seen in the weavings of the Saryk. This gol is apparnetly derivative of the Salor chuval gol (Plate 9 a) according to Moshkova, appropriated subsequent to the defeat of the Salor in the 1840s. Though Saryk chuvals are rare, I believe one may encounter early chuvals using this or a form of this gol type. On the other hand, the appearance and use of the gol seen in Plate 9 b corresponds closely with this time frame and may be the supporting evidence for Moshkovaa's theory regarding ornaments of rugs being "taken over by the victors, to be used as models for reproduction of patterns.".


The observations lead us to conclude that in the ornamentation of chuvals we can find today representations of those gol which gradually lost their tribal importance. Here we approach the group of gol which can be called 'dead gol'.

From the example, of the Salor we see that, following the defeat of the tribal group and the loss of opportunity to engage in the art of carpet weaving, the stock of carpet ornaments didn't disappear, but was taken up by the conquerors and used by them in their own carpet products . It is easy to establish the route which this took. Shortly after the carpets and carpet 		products of the defeated tribe had been destroyed, the ornaments were taken over by the victors, to be used as models for reproduction of patterns. It is true that, in the case of the Salor, the special relationship of this tribe could have had special significance: as we know, they were considered to be the most noble and aristocratic tribe, a status inherited from their ecenomic position in Turkmenia over the centuries. To this day, the Salor are regarded as the fathers of Turkmen carpet weaving and it is not surprising therefore that their stock of carpet designs was so widely adopted and used by their conquerors, the Saryk, and the Tekke of Merv. (ie. Plate 9 b)



Plate 13. Though Moshkova describes this 'chuval gol' form to be a distinctly Yomud attribute, she is mistaken. The gol she refers to is seen below in Plate 14 is, among others, used by the Kizil Ayak as well as the Yomud. I believe it may have degenerated from the gols we see above, a detail of a very old and fine Tekke chuval.


In any case, we have here an example of the transference of tribal ornament from a carpet to a chuval. A similar example can be seen among the Chodor who, under the influence of the growing trade in carpets, stopped reproducing the ertmen on their main carpets. However they preserved the ertmen when it had lost its significance as a tribal ornament, on torba and chuval. A similar phenomenon took place among the Saryk, who at the beginning fo the 20th century employed their tribal patterns widely on chuval.

In consequence, dead gol, those which had lost their signifi- cance as tribal ornaments, were preserved on small items of carpet-ware and over the years began to be used as simple ornamental motifs. Another good illustration of this can be seen in the comparison of the tribal gol of the Salor group of Turkmen. 		Although the largest tribal groups of modern Turkmenia, the Tekke, Saryk Ersari, and Yomud, are now widely dispersed from the shores of the Caspian Sea to the banks fo the Amu-Darya, it is known from historical sources that in the 16th century they formed one tribal group, known as the 'outer Salor' with the original Salor tribe as its head. This group inhabited a comparatively small area on the eastern shore of the Caspian.

After the disintegration of the Salor union, the tribes which had belonged to it dispersed throughout Turkmenia. For the small individual groups of the former times, the Tekke, Yomut and Saryk, this period of resettlement was highly important, each grew into a large tribal union and during this time the 17th to 19th centuries, all three began to play an increasingly important political role in the history of Turkmenia.



Plate 14. A gol, as seen on a Kizil Ayak chuval (right), that corresponds to the gol cited in Moshkova's theory of an ornament that is no longer recognized with any meaning, and used for strictly decorative purposes. For some reason, Moshkova, in her text, assoicated this gol (diagram, left) soley with the Yomud.


In their carpet ornamentation each tribe took its own path, forming a separate group, each with its own identity, although maintaining a common Turkmen style in colour, geometrical character of the design, and some ornamental details. Moreover, it the carpets of each of these tribes were preserved the basic designs, shapes and ornaments in which we are still able to see a common source; a form of tribal gol which these tribal groups evidently shared at the time when they lived together by the Caspian Sea as one tribal group.

We have discussed above the tribal gol of the Tekke and Ersari tribes both of which have been used to present day on the main		 carpets of these tribes. The Yomut gol of the same character was preserved on chuval and was very widely used in this form; a Saryk ornament of a similar kind was reproduced in the recent past on large carpets and to the present day can be found on chuval (Plate 9 b). Even a passing glance at these ornaments is sufficient to see in them the elements which testify to their common origins. This undoubted similarity serves to illustrate how the historical connection (in this case, the common origin of this group of Turkmen tribes) found its expression in the tribal gol



Plate 15. An 'eagle group' kapnuk from the National Museum, Ashgabat, Turkmenistan. 3'8" x 1'7". Curiously, Moshkova identifies it as a 'khalyk', probably due to the fact it has only two "arms" rather than three, a more commonn feature of Yomud group kapunuks that are later than this example. See Turkoman Studies 1 for an overview of Yomud kapunuk.



However there is more to be learned from the examples we have shown. It is apparent that both the Saryk and especially the Yomud gol can be included in the group of deal gol, since they are no longer recognized by the carpet makers as tribal ornaments and are used simply for decoration. The old Yomut gol, used today exclusively in the field of chuvals, now bears the name chuval gul and not gol, and is used here along with other purely ornamental patterns (Plate 14, see caption).

The conclusion lies near that old gol, now out of use, which belonged to tribes that have disappeared from the historical scene, can be found preserved in the small items among the carpet products of Turkmenia. It is not improbable that among the present rich stock of designs of Turkmen medallion		 patterns, gol have been preserved in altered form which once belonged to tribes as the Ikdir (Igdyr) which having lost their importance as independent tribes, began to use the gols of foreign tribes. The tribal patterns of the old Turkmen tribes which played a leading political role in Turkmenia, Iran and Asia Minor in the 11th to 14th centuries, may not have disappeared without a trace. In the Turkmen territory, the tribal gol of these old Turkmen and Oghuz tribes have come down to us as the gol of the Salor (Plate 9 c) and the Yomut. The Ikdir (Igdyr), another ancient group, included among the old Oghuz tribes, have lost their tribal gol. In all probability one must seek in the carpets of Asia Minor the gol which have disappeared since, from the 11th century onward, the Oghuz tribes moved into Asia Minor and into Iran



Plate 16. Though Moshkova claims the classic ashik form seen in some asmalyks is a later feature, I believe she is mistaken. This seven sided asmalyk (above) is quite old, certainly contemporaneous with other weavings exhibiting the more "archaic' pattern seen in the lattice designed example directly below, though probably not quite as old as this paritcular one.


Plae 17. A very old examle of the Yomud asmalyk with a more archaic field pattern. Private Collection, USA


The examples we have cited of tribal gol transferred to the smaller carpet products cause us to look closely at the designs of these items, since they often preserve versions of tribal gol in their more archaic form. In this respect, important material can be found in the ornamentation of the piled wedding decorations of the Turkmen, which have now been preserved only by the Yomut in their so-called osmolduk (asmalyk). This is the well known five cornered rug which hung at the sides of the wedding camel which bore the bride. Modern osmolduk, made this century during the colonoal period, were frequently decorated with the ashik design which filled the central field of the osmolduk. In the two or three osmolduk of fthe early 19th, perhaps even of the late 18th, century which are preserved in the museums of Moscow and St. Petersburg, the ashik design has a more archaic form. It has the appearance of a coloured rosette with white ziz-zags, rhomboid in shape and in the centre of each is the representation of a bird, sometimes with an animal, treated quite realistically.(Plate 17, Footnote 3) The similarity in composition of similar articles and the preservation of several elements of the design during its transformation into a simplified geometric pattern, leaves no doubt that the ashik design is a later version of an old ornament in the form of a bird (see caption, Plate 16).



Plate 18. 'Khalyk', Tekke tribe, 19th century, Private Collection, Germany. An extraordinary example of the khalyk, complete with tassels as well as decorated with an unusual design.



Plate 19. Detail image of a very old khalyk in the National Museum, Ashgabat. In my experience, the ornamentation seen here for weavings of this function is unique.


Another knotted wedding decoration was the khalyk - a rug which hung from the chest of the animal in the shape of rectangle (in old pieces it was five sided) with two sections hanging down the sides and finishing in a twisted fringe (Plate 18). Khalyk have gone out of use today, as have all the knotted wedding decorations. However in some tribes, for example among the Saryk, the so called 'kapunuk' was preserved into the recent past as a special decoration for the yurt door (Plate 15). The ancestor of the kapunuk was probably the khalyk.
 Very few khalyk have survived to the present day The khalyk known to the author are literally only those pieces preserved in the museums of Central Asia. The design of one of these khalyk is all the more interesting in that it consists of an octagonal medallion with the form of a bird (Plate 15). The colour scheme of this piece, the treatment of the border and tweaving technique leave no doubt that this khalyk is of Turkmen origin (curiously, Moshkova refers to it as a khalyk when it is undoubtedly a kapunuk, Footnote 4).



Plate 20 a, b. Details of two asmalyks from the Dudin Collection in the Russian Ethnographic Museum, St. Petersburg. Note the two different types of drawing, the upper photos with a running bird flanked by dogs, while in the example below, the bird is apparently standing. In both, these birds are seen bearing something on their backs. Birds, in traditional Central Asian/Inner Asian mythology bear souls from one world to the next. Interestingly, all these birds depicted bear something on their backs as are those birds seen in Anatolian rugs..



The fact that the bird design is found in Turkmen wedding decorations is of interest. Even during the 19th century, Yomut weddings were marked by the inclusion of a tribal ornament into the knotted wedding decorations of the girl who married into another tribe. Yet, in not one of the wedding carpet pieces of the Yomut that has survived do we find the reproduction of a known Yomut tribal ornament, neither old nor new. The reason for this is hidden in the nature of the marriage tradition among the Turkmen in general and of the Yomut in particular. Those Turkmen groups, which Euriopean researchers call tribes (Yomut, Tekke, etc.), the Turkmen themselves and the historical sources commonly call peoples 'khalk'. Until recently, marriages among the Turkmen were always confined with this khalk, (ie. Tekke women married Tekke men and Ersari married only within their own group). The Turkmen regard the the 'khalk' as a 'people' and those sub-divisions of the 'khalk' which European reserarchers now call 'families' are the tribal groups. Turkmen marriages then take place between different tribal groups within the larger unit. There has been very little research into this matter and marriage relationships in Turkmenia are still unclear. Nevertheless the work on this subject by Prof. Preobrazhensky on the northern Yomut leaves no doubt that although marriages between individuals of different groups within the 'khalk' used to be the custom, the rule has been broken with increasing frequency in recent times. Nowadays, only one rule applies, that marriages take place within the 'khalk'. For a long time, however, marriages have taken place with close family groups.



Plate 21. Bird forms seen on an asmalyk attributed to the Arabatchi, Russian Ethnogrpahic Musuem, St. Petersburg.


In the light of the information given about marriages among members of the Yomut tribe, we can begin to understand the absence of the known tribal gols, old or new, from their wedding decorations. It would clearly be wrong to try to find Yomut tribal gol in khalyk which belong to the 'khalk'. On the contrary , in wedding decorations, only the gol of the smaller family subdivisions were used. Later, because of the intermingling of marriage groups within the Yomut 'people', the smaller groups lost their significance; the reproduction of the gol became simply a tradition.

Thus these ornaments, medallion with the forms of birds and animals which were used in ancient Tukrmen wedding 		decorations, must be seen in the light of ethnographical study as gol of one or other of the Yomut family subdivisions. They represent an ornament preserved in the most archaic form, in which the realistic illustration of birds and animals was often used in the composition.

We can try to confirm the hyposthesis that the realistc drawing preserve features of ancient tribal gol by comparing them with more recent gol which may contain elements of the earlier forms.



Plate 22. a) Kush - bird, b).Dunguz Burun - pig's snout, c).Chakkan - one who bites (scorpion?), d). Kovacha - cotton ball


e) Detail image of the primary gol from a Tekke main carpet, mid 19th century, Private Collection, USA


In our analysis, we consider the gol belonging to the Salor union to be the most important and their content is at least in part revealed from what we have learned about the Tekke gol.

The individual elements of the Tekke gol have been named by the carpet weavers as (a) kush - bird, (b) Dunguz burun - pig snout, (c) chakkan - one who bites or stings, and (d) kovacha - cotton ball (Plate 22).

From this list it is clear that the basic elements of the Tekke gol take the form of a central plant motif with birds or animals at the sides. In the more recent version of the Tekke gol, these birds 		look more like elements of a plant than of an animal. In the Ersari gol, these patterns clearly have the shape of a trefoil, and are called 'gul' or flower, thus losing any connections with the original form (Plate 10). Here the Saryk gol (Plate 12), which has survived in the fragment of an old carpet is of interest. The gol has only one bird illustration and it is particularly valuable because it serves as an intermediate stage between the Tekke gol and the old Yomut gol, which has no 'birds', but has taken on the character of sharp geometirc angles.

Although, we cannot yet discuss the history of the development of these motifs, it is useful to emphasize the presence of birds and animals as elements of all the gols in Turkmen carpets.



Plate 23. Similar to the Tekke gol, this example is seen on a very old chuval attributable to the Arabatchi tribe. Note the similarities as well as very minor differences in the design. This 'lobed' gol shape is thought to be older, more archaic. The relationship of the Arabatchi to the other Turkmen tribes is nebulous. Some have speculated they are a branch or sub-group of the Ersari. Moshkova seems to relate them to the Chodor but even the Chodor have proven to be an enigma, exhibiting a design pool somewhat different than that seen in weavings of other Turkmen tribes. This specific lobed gol form appears to very particular to the Arabatchi and it is difficult to determine which is "older", the Tekke or Arabatchi gol. What may be possible to say is that the 'kurbaghe' secondary gol may originally be an Arabatchi ornament from which, possibly, the Tekke variant might be derived. Private Collection


Plate 24 Arabatchi Chuval,.76m x 1.37, (4'6" x 2'6"), late 18th/early 19th century, Metropolitan Museum of Art, NY, 22.100.40a, Gift of James F. Ballard


Interesting material also comes from the old tribal gol of the Chodor, the ertmen, an ornament which has not been sufficiently studied. It is not difficult to see that the figures within the gol are a combination of birds and plants.(Plate 26) The five leaved flower clearly has the head of a bird at the top, and a tail and feet at the bottom (Plate 27). Another important part of the ertmen are the three rhomboid zig-zag figures joined by a common stem. On two of these the outline of a bird with spread wings is clearly marked. The general composition represents a plant with two birds placed at either side facing each other. It is interesting that the Chodor have not been able to establish the names of these ertmen elements. This is probably explained by the fact that this is an old gol, which went out of use early and is now rarely found. However, it contains the form of a bird, which even recently was not recongnized although its name was preserved elsewhere.

In Chodor troba from the Amu-Darya, made not by those of the northern Tashauz, but by a group which settled between Daynau and Sayat at the end of the 18th century, we find a design which in pattern, general them and from the technical evidence is arguably the same ertmen, in an altered form. Here no trace of the bird remains, but instead there are only the rosettes of a flower. (Plate 26 c, Plate 29 a, c) Nevertheless, the whole composition of the Amu Darya Chodor is called kushly (birds).



Plate 25. The 'ertmen gol' of the Chodor as seen on a Chodor main carpet, 19th century, Private Collection, Germany




Plate 26. Diagrams from Moshkova, two of which retain the true form of the ertmen gol with birds ("kushli") on either side of the central tree.




Plate 27. A detail image of a Chodor trapping depicting the ertmen gol with birds flanking the central tree deivce. The birds are depicted with both tails and feet below the "five leaved flower".


Thus, when we compare the ornament of the Chodor of the north with those of the Amu Darya area, we conclude that the bird drawing found in the classical ertmen, testifies to the presence, at one time, of a realistic bird representation which gradually became simplified, took on new elements and changes its form to a floral rosette. In this sequence, the design of a Chodor chuval in the Museum of Arts in Ashgabat, can undoubtedly be taken as an intermediary link. This still contains elements of the bird, but these are no longer recognized by the carpet makers. Thus, the analysis of the ertmen shows that birds are an integral part of the composition in the family gol of the Chodor.
 Now let us turn to the tribal ornament of the Arabtshi tribe, which is rlated to the Chodor tribe. The tauk noska is reproduced from a carpet in a private collection in Ashgabat, and, judging from the technical details of the border and the general composition is about 100 yeaers old. It one takes away the upper bird-like heads, this ornament resembles the letter 'H' and in this form it has been used widely both in the work of the Aratcshi themselves and of the Chodor (Platte 8) who borrowed this pattern and took it to the Amu Darya region. In Simakov's reproductions, an interesting carpet carries a gol of similar type (Simakov does not relate it to the Arabatshi), but of a more archaic form with two birds standing side by side without a horizontal cross piece.)



Plate 28. A later variant of the 'ertmen gol'. Still the birds are evident but clearly somewhat deformed from the earlier version seen above.




Plate 29. Line drawings from Moshkova showing the degeneration of the ertmen gol, with only the central diagram portraying some relevance to the original concept of incorporating bird motifs into the ornament.




Plate 30. An unusually simple and spacious treatment of the ertmen gol design which adheres to none of the line drawings supplied by Moshkova.


The place of origin of these carpets is the Denau region of Chardzju territoriy, where the Arabatshi group has most completely preserved its identity and here a large number of caprpets with this pattern in the form of the letter 'H' have survived. The Arabashi regard this gol as their own and call it tauk nuska (Plate 8), figure of a hen. Here again we have the form of a bird, clear both from the outline of the ornament in its more archaic form, and from its name.
 The presence of bird, or bird and animal forms in four tribal gols of differeent Turkmen tribes provides us with sufficient evidence to relate these gols to the birds on the wedding osmolduk and to consider th latter a more archaic form of the gols. From all the material we have examined we can draw on conclusion: in the original composition of the tribal gols, birds sometimes together with animals and plants played an important part and their traces survived in more or less realistic form in the gols of Turkmenia today.



Plate 31. An interpretation of the 'ertmen gol' woven by an Ersari group in the Middle Amu Darya region of Central Asia.


The opportunity to draw on historical facts contribes greatly to the success fo this type of research. Data which supports our argument are found in the information about the Turkmen which were collectded by the 14th century author Rashid-al-Din. In his famous work, Djamiy-at-Taravrikh, which discussing the history of the Oghuz tribes, this author gives a detailed list of all the Oghuz tribes, with notes on the tamgha, which the tribes used to brand their stock, the names of the cuts of meat which the representatives of each tribe were given at the feasts and finally,the name of the ongon (totem) of each tribe. From this list it can be seen that the Oghuz tribes of that period were divided into six groups, which were associated by family, with four tribes in each, and to each of these groups was assigned the same piece of meat and the same totem. The following tribal totems are of special interest to us:

1. The Kayn, Bayat, Alkyrevli and Kaaeveli tribes had as their totem the Tsar's white falcon.
2. The Yazyr, Dyker, Dudurga and Yaparli tribes had the eagle
3. The Avshar, Karik, Bedkili and Karkin tribes had the eagle carrying a hare.
4. The Bayundur, Bechene, Dzhaguldur, and Chjepin tribes had the gerflacon.
5. The Salor, Eimur, Ala, and Yuntli tribes had the uch (no translation)
6. The Ikder (Igdyr), Byukdyuz, Yive, and Kynyk tribes had the chakir (no translation)



Plate 31. An early painting from Herat (circa 1420), depicting a woman seated on a rug with the layout of the field bearing a similarity to that which we see in subsequent Turkmen weavings. Additionally, the archaic 'Kufic' border pattern seen in old Anatolian rugs is evident.


It is clear from the list that Oguhuz tribal groups had totem birds of the eagle family. The information set out by Rashid-al-Din allows us to conclude that the different bird forms in the tribal gol of the Turkmen were used deliberately and were not chance phenomena.

Moreover, there is every reason to suppose that the primitive representations of birds, especially of the eagle family, on carpets reproduced in the paintings of 14th century artists, were the most ancient versions of the carpets of the Oghuz which carried tribal gol in the form of the totems of the tribes.

We have in mind the following:

1 The carpet published by Willhel von Bode from the painting by Lippo Memmi, Portrait of Mthe Madonna, 1330-40, Berline Museum, Dahlem invl. 1072, (Plate 32 a)

 2 The carpet reproduced by the same author from the painting by Nicolo di Buonacorso 'The Marriage of the Virgin', 1370s to 1380s (National Gallery, London, invl. 1109) showing an eagle like bird of unusually primitive form within an octagon.
(Plate 32 b) (Footnote 5)

To embark on an analysis fo these primitive versions of tribal gol is beyond the scope of this article. However it is striking that our conclusions, based on the historical and ethnographic material we have examined, is supported by the opinions of the researchers who attribute these 14th century carpets, with their primitive bird forms to Asia Minor which, by this period, had become the arena for the activities of the ancient Turkmen tribes.



Plate 32. a) Line drawing of the painting by Lippos Memmi, 1330-40, b) Line drawing of a painting by Nicolo di Buonacorso, 1370-80


FOOTNOTES

1. Ladyzhensky, visting western Turkmenia in the 1860s saw decomposing funeral carpets on Cheleken Turkmen graves. See: AS Berg, A History of Turkoman Research', Turkmenia, Vol. 1, 1929, p. 86

2. A term which became widely used after the TSSR filkkm production 'Salor Rose'. This inaccurate translation is due to the artist OF Mezgireva, who showed the knotting of carpet ornaments in one of hter paintings (Museum of Arts of the Trukin, Ashgabat)

3. The osmolduk were described in a work on Central Asian carpets by Fel'kerzam; they also attracted the attention of Gogel, a later researcher into Turkmen carpets.

4. This khalyk can be found in the National Museum, Ashgabat, Turkmenistan

5. Wilhelm Von Bode, Vorderasiastische Knupfteppiche aus alter Zeit, Leipzig, 1914, p. 109



My thanks to Michael Franses, Robert Pinner and Humanities Press/Oguz Press for permission to reproduce the original text and line drawings which originally appeared in Turkmen Studies 1 (1980).

My thanks also to Elena Tsareva for permission to photograph some of the weavings from the Dudin Collection in the Russian Ethnographic Museum, St. Petersburg during my visit in 2000 and my heartfelt gratitude to the women working in the bowels of the National Museum, Ashgabat, Turkmenistan for their gracious hospitality in allowing me to spend time with their collection in 2001.

I also wish to thank Seref Ozen (Cocoon, Istanbul) for going through his photo archive and offering additional images.

Dedicated to Robert Pinner