PLATE ONE
Arabatchi Engsi
Archaic period

Plate One is the most significant engsi yet to be published. The complex pattern and highly articulated individual designs have not only influenced the development of other Arabatchi engsi but have also influenced, to a greater or lesser degree, other types of trappings as well. These features have facilitated the recognition of the connection between engsi and the cult of the shaman.

Supposedly engsi were hung in the doorway of the yurt as protection from extreme weather. Another possibility purports the existence of the mirhab-like niche seen in most examples, like fig.1, signified the engsi was a prayer carpet. This might have been the case for some examples made for or by Turkmen converts to Islam and likewise both of these suggestions may be more valid when considering examples from the pre-conquest or colonial weaving periods. But archaic period engsi were different. This description will explain their iconography and demonstrate, by comparison with another Arabatchi engsi of the same type(fig 1), the apparent differences that separated the earliest cult engsi from those which may very well have been used as normal domestic door-rugs or as prayer carpets.

Plate One, Plate Two which was made by a Yomut sub-group and one other, a still unpublished Saryk engsi that was formerly in a well-known collection in London, England and whose whereabouts are presently unknown, are the only existent nomadic archaic period engsi known to this writer. Although Plate Three has been assigned an archaic period label, it is most probably the product of a settled and not a nomadic weaving group and is in any case a later example.

Is it possible these are unique survivors from three different weaving groups active during the period of Turkmen history when engsi functioned as part of the shaman’s world of cult magic? Unfortunately this question can not be conclusively answered at this time. But because this example contains such a potent set of archetypal icons, it has been possible to isolate several of these symbols to help to answer this intriguing question.

The enigmatic Y designs in each of the four panels have been referred to by all previous writers as candelabra or animal head. These descriptions refer to the two distinct elements appearing at the end of the upper arms of the Y. One or the other of these appears in almost every pre-conquest engsi. This attribution is rather surprising as both these were called eagle or bird by the Turkmen themselves. Plates One, Two and Three show the typical western Turkmen version, while Plate Four illustrates the eastern style. The following, published here for the first time, provides a better explanation for these ornaments and demonstrates their connection to the cult of the shaman.

The widespread ancient pan-Asian belief in a cosmic tree that connected the earth to heaven was also central to the ideology of the shaman. In some Siberian groups the creator God was represented as an eagle and his children as bird spirits living on the cosmic tree. In addition, shaman from southern Siberia areas that are in even closer proximity to the Turkmen are directly connected to this myth, as it is part of their initiation and ascension rituals. The shaman is further attached to this myth in that his drum, a very significant part of his power and ritual paraphernalia, is supposedly made from wood taken from the cosmic tree. Also, in many indigenous legends he is born of a great bird in the branches of this tree.

In many Siberian villages a tall pillar placed the center of the community is thought to have magical powers in its connection to heaven and the land of the gods. In some instances these pillars have birds at their tops and fig.2 illustrates a group of such pillars, as they appeared in Dolgan villages in Siberia. Notice the tallest is capped with a double-headed bird surrounded by other single-headed ones. Is this double-headed type bird the source for this engsi’s Y icon, and does the myth of the cosmic tree and these bird pillars provide the source for the design in the field of the engsi? And is the middle border, which completely surrounds the field and is divided into boxes each containing a tree, another reference to the cosmic tree and its connection to the cult of the shaman? These are fascinating questions that serve to highlight the connection the icons in this engsi maintained with cult mythology.

The bird-pillar has a long history beginning as early as the late Paleolithic period where it appears in cave-wall paintings. One particular painting(fig.3) shows this icon, a large bison and a human figure. This scene has been interpreted by archaeologists and pre-historians in various ways. One very plausible version presents the idea the figure is a shaman in a trance and his outstretched arms connecting the bison and bird depict communication with the spirits to assure a successful hunt and kill of large animal prey.

Another important icon of this period, the animal in a recumbent pose was also found at a number of archaeological sites. Fig.4 depicts a bovine with its head turned back from its body at a 180 degree angle, the typical posture portrayed on these works. Sensitively carved from mammoth bone or ivory and dated circa 20,000-10,000BC, these artworks are undoubtedly some of the most beautiful known from this time. Unlike the bird icon, no later religious or cult traditions explain this symbol nor have any other sources identified its meaning. However enigmatic this icon remains, it can still help to unravel the mystery of the origins of this engsi.

The curled-leaf design used in the innermost vertical borders has for long been considered an important Turkmen symbol. Its meaning has remained a puzzle, but now a solution can be offered. It is not a curled-leaf but rather a recumbent animal derived from the icons described and pictured above. At the bottom of each border the weaver has articulated the first of these designs(figs.5 and 5a) in a completely different fashion than the rest(fig.6). The striking resemblance these have with the Paleolithic carvings seems highly unlikely to be accidental. The important choice of their placement at the beginning of the border, the conceptual point from which the others emanate, and their deliberate representations are very convincing.

Another Arabatchi engsi(fig 1) has been illustrated here for comparison even though they seem quite similar. Yet there are a number of subtle differences demonstrating why this engsi should be considered a later, classic period weaving and Plate One as the archetype from which it has been modeled.

The most obvious is inclusion of a single mirhab-like niche above the field. This style, as opposed to Plate One’s different format, has been influenced by the standard engsi layout associated with another Turkmen weaving group known as Tekke. The classic period Tekke engsi (fig.7) illustrated with Plate Three better demonstrates this and also the connection Tekke ensgi maintain with the simple square in a box designs (fig.8) under the mihrab and in the two panel areas on either side of fig.1. These are a rather simplified interpretation of a göl known to the Turkmen as aina(fig.9) that are actually a stand alone main göl design frequently used on many pre-conquest and colonial period Tekke trappings, called chuval and mafrash(fig10).

Like a botanist would graft the stem of one plant onto the root of another, so has the weaver of fig.1 grafted the top of a Tekke engsi onto an Arabatchi base. The Tekke had conquered and absorbed the western Arabatchi by the early 19^th century and fig.1 most probably dates from around this time period. Design alteration and recombination like these are frequently encountered in Classic period weaving and were indicative of the beginnings of the breakdown in the individual weaving cultures of the Turkmen.

The small-four-niche format is not the only proprietary feature of Plate One, as it also contains the complex and significant Turkmen icon, known as kejebe(fig.11). This highly enigmatic design, which has been entirely omitted from fig.1, is found on a cluster of long trappings, known as torba, attributed to the Turkmen group called Salor(fig.12). The inclusion of this mysterious icon here, as part of an archaic period engsi, signifies its original context was engsi related. The appearance on the later torbas, none of which are earlier than classic period, must then be considered an adaptation and re-use.

Very prevalent throughout the classic period, reproduction of archaic period iconography typically involved subtle but significant losses to the three-dimensionality and presence the originals were able to capture. Invariably lower levels of articulation, misunderstood color sense and inability to create the correct proportions, as all were the case here, highlight these shortcomings(fig.13).

The kejebe can not be examined without considering the niche that always surrounds it. Thought to have been used as the storage place for carved female figurines during the Paleolithic Period, the niche has a long history as an important symbol. These stone statuettes are considered to be the oldest existent sculptural artworks and have been dated circa 30,000 B.C and a number of them have been recovered from cave sites in the northern Mediterranean region. The figure contained under the niche in Plate One is not as easily understood and while the anthropomorphic drawing might suggest a figurine, such an interpretation is impossible to substantiate.

Above and between each of the four niche is another unusual design that also defies explanation(fig.15). But when these highly articulated symbols are doubled as a mirror image(fig.16), an uncanny resemblance with the form of a main göl known as dyrnak (fig. 17), which is used by the Yomut Turkmen on their main carpets, is created. This göl is atypical, as it sometimes appears in the same weaving in both a simple and a more complex version, and because it was used on a number of very old main carpets, especially those with multiple major göls, it is believed to be one of the most ancient and archaic.

Interestingly enough, the kejebe remains together with fig.15 in every instance of its use as a permanent design set, regardless of the type of weaving, period or group identification. The only difference being these later examples all show stylized and highly codified interpretations of the archetype Plate One originates.

Fig.12 presents what is considered to be the seminal version but notice that the kejebe figure has been removed from a richer environment, placed in an empty niche with narrow long proportions and changed by additions of new ornamentation. Hooks, hand-like extensions and a base of four triangles have somehow become attached to the figure itself. And now the figure floats within the niche rather then being firmly anchored at the bottom. The niche has also undergone alteration with four hooks and a double hook design called kotshak having been added in addition to being surrounded by an outline of red, purple and black circles with open centers.

Fig.15 was not forgotten, it has been retained and now appears as a nice, neat and far less mysterious medallion(fig.15a). Another kejebe torba(fig.14), attributed to the Saryk, has materials, design, wool quality and surface patina that are appreciably earlier than fig.12 and this version of these large kejebe torbas should be considered as the prototype for the group of Salor examples. Notice that the niche is broader, the kejebe figure more fixed and not floating within the niche(fig.14a), no other hooks and only a kotshak has been added, the round circles are only at the sides and do not continue at the top of the niche and the medallion between the niche retains more of the proportion and character of those from Plate One. It is not surprising then to find a recumbent animal icon(fig.14b) in the large medallion of this Saryk kejebe torba whereas fig.12 and all the other Salor kejebe torba never contain reference to this icon.

This process of design change was not a unique occurrence, it was invariably repeated again and again as archaic period weavings were reproduced during the classic and later periods. However ability of certain icons, symbols and designs to retain their original character and others to remain together as fixed sets throughout long periods of time was testament to their importance as cult icons as well as to the highly proscribed and strict fundamental nature of the Turkmen weaving culture. And even though most post-classic period weavers were prone to even greater propensity for omission and addition, certain groups still recognized these sacred sets and maintained their association.

In conclusion, because fig.1 is actually a hybrid, the omission of the recumbent animal borders as well as the large horizontal center panel that separates the four-quadrant field is quite typical. In their place a simplified rendition of the horizontal border, which is above and below this panel on Plate One(fig.22), has been substituted. The inclusion of additional borders, borders that have been created by simplification, or entirely new ones not associated with the archaic period original or its weaving group, were also flaws of the post-achaic weaving periods.

The weaver of fig.1 has included alien designs, as well, and two borders typically found in the weavings of the Ersari, another Turkmen weaving group, suddenly appear. The first is located above the horizontal box and tree border over the niche and may be another design derived from fig.15. The second, a simple juxtaposition of colored triangles and dots, completely surrounding the field and niche is often found on many weavings of the Saryk, another Turkmen group.

CAPTIONS

Fig.1 Arabatchi engsi classic period from Between the Black Desert and the Red Weidersperg Collection Fine Arts Museum, San Francisco
Fig. 2 Bird Pillars from Dolgan Villages, Siberia after Holmberg pg 334 fig. 13
Fig. 3. Paleolithic bird pillar design incorporated in shaman scene after Maringer pg. 168
Fig.4. Recumbent animal carving of reindeer ivory after Maringer fig. 38
Fig. 5a. Recumbent animal border from Plate One
Fig. 5b. Recumbent animal border from Plate One
Fig. 6 Major border from Plate One showing normal curled-leaf drawing
Fig. 7 Tekke engsi classic period from Turkmen Studies plate XIV
Fig.8 Detail fig.1
Fig.9 Typical aina major göl
Fig.10 Mafrash with aina göl design classic period from Turkmen: tribal carpets and traditions plate 41
Fig.11 Kejebe icon from top panel in Plate One
Fig.12 Salor kejebe torba classic period from Between the Black Desert and the Red Weidersperg Collection
Fig. 13 Detail fig. 12
Fig.14 Saryk kejebe torba classic period from Tent Band Tent Bag Plate Five
Fig. 14a Detail of kejebe niche fig 14
Fig.14b Detail showing recumbent animal from major medallion fig 14
Fig.15 Bird icon from kejebe panel Plate One
Fig.15a Detail of fig.12
Fig.16 Bird icon doubled to create darynak göl
Fig.17 Typical darynak göl and link to the carpet it appears on Fig.22 Detail from Plate One

No photographs or text may be reproduced without the written consent of the copyright holder, the Weaving Art Museum, Inc.