Bird Rugs of South Persia by James Opie Illustration 1. Detail of cover illustration. Birds confront, separated by a "pole" device. This is an aspect of a classic design form.

Approaching this from another side, what role do the designs in a weaving have in this respect? Could a rug which was copied from a city carpet pattern be tribal? In approaching the fascinating "bird" or "chicken" rugs of southern Iran, this examination of our terminology is useful as these rugs evoke precisely these questions. Are these nomadic products or from villages? Is this a borrowed design or a traditional, local one? Indeed, are these unique bird designs "tribal" and, if so, under what definition?

The comments above suggest two different perspectives from which this question of what we call "tribal" can be viewed. One is based on the way of life of the weaver. Where did she live and weave? Among nomads? In a village or town? Or, as we find in recent years, even in a city setting?

Illustration 2. Luri rug, 8'4"x5'3", fourth quarter 19th century. Bird rugs are almost always products of the Khamseh. The rare Luri example is simpler in design than Khamseh examples. Courtesy of an Oregon collection.

The other perspective focuses not on the way of life or habitat of the weaver but on the designs in the weaving itself. Many rugs which are sold as "tribal" are somewhat modified copies of urban designs. At times the geometric format which a design acquired after it was simplified or "tribalized" over the course of decades tends to conceal its earlier urban source.

These tribal and village rugs which owe very much to an earlier urban design scheme present one of the ironies of the tribal rug market at this time. Some of the most sought after and expensive of tribal rugs are pieces which contain designs which were lifted from urban weavings. One would expect that the most purely tribal examples would be avidly pursued, examples which owed nothing to urban design traditions. This is one of the factors which makes these bird rugs so fascinating and, potentially, important as a collector's group. Here we see designs which have no antecedents in Persian city weaving. Nor do we find this design family appearing in the rugs and kilims of other regions. It is something that appears to have indigenous roots in southern Iran, roots which may reach very far into the past. In my view, these rugs are truly tribal , whether they are the work of village weavers or nomads. To find a family of tribal designs that are not traceable to an outside source is very rare. On this ground alone, the bird rugs of southern Iran are worthy of active interest on the part of museums, collectors, and students of tribal culture.

A careful look at these bird rugs provides a useful reference point for clarifying degrees of tribalness . To hazard a general approach to defining this term, I suggest that the most tribal examples are those that were made by tribally affiliated weavers which contain traditional, indigenous designs which cannot be traced to any urban or non-urban outside influence. The pinnacle of this class would be those weavings which were made by nomads for their own use, but village products are not excluded from this first class.

Illustration 3. Uncertain village, 3'6"x3'3", late 19th or early 20th century. From Tribal Rugs of Southern Persian, page 209.

Somewhat less pure in this sense are those weavings which come from nomads or tribe-related villagers but which contain designs which were borrowed from another tribe. There are a great many examples of this type, including the work of entire tribes. Still less purely tribal are the rugs which are, in essence, copies of urban weaving designs. These are usually village or town products, often made in commercial workshops owned by members of the tribal aristocracy. Many examples are finely knotted and some display an element of genius in the use of color and the improvised arrangement of motifs. They can be marvelous examples of village work. But in terms of a search for what is genuinely tribal, this last category is certainly not on the same footing as the other two groups.

This approach to the subject can be useful in our efforts to sort the vast and complex subject of tribal rugs into some useful order. In such an effort, the bird rugs of southern Iran are an excellent subject. Few types of rugs give us so much visual pleasure and, along with this, so much to study, enhancing our intellectual enjoyment.

Before looking into them in greater depth, it is useful to examine the tribe from which most of these pieces originate, the Khamseh (Hahm-say, no "k" sound) Confederacy, formerly one of the most important tribal groups in southern Persia. Some readers will know that Khamseh means five in Arabic and that the Khamseh Confederacy was a political grouping of five south Persian tribes, founded in the 1 860s to counteract the power of the neighboring Qashqa'i (Gosh-guy-ee, three syllables, emphasis on the last) tribes. No tribal cluster so exemplified the complex tribal map of 19th century Persia as did this confederacy, composed of Arabs, Turkic factions, long-standing Luri elements, Persian villagers, and mixtures of these linguistically and culturally divergent groups. The confederacy was disbanded in the late l950s. (Fredrik Barth's book, Nomads of South Persia, includes few comments about weaving but gives the most complete picture available about one of the strongest remaining Khamseh tribes, the Basseri.)

While writing Tribal Rugs of Southern Persia (Portland, 1981 ), I began to realize that some of the designs to be found in Khamseh weavings were probably indigenous to south Persia and had come into the Confederacy through tribal elements who had lived in the region from earlier periods, before Turkic and Arab groups forced their way onto the scene. The possibility of finding tribal motifs which could not be traced to any outside source intrigued me and rugs from this group began to interest me more and more. At that time, even more than now, the marketplace clearly favored the formal, urban style rugs from the region, particularly from the Qashqa'i tribe. In 1 980 some rugs in the mille fleur design were selling for as much as $50,000 a piece. It was not until my book was published and it was again possible for me to stand back from the subject that my bias favoring the most traditional designs, many of which are found in Khamseh work and which were not highly regarded in the market, became deeply rooted. Some number of collectors and other students of the subject share this leaning, but, as yet, no broad movement favoring what can be recognized as truly tribal weavings has developed. If this should occur, the name "Khamseh" will no longer take a back seat to the more popular "Qashqa'i" label and the Bird Rugs in particular may become a highly collectible group.

As the illustrations accompanying this article suggest, these rugs have more going for them than a pure pedigree. The best of them are not only tribal rugs; they are culturally significant objects of art. The best examples, noted for their delicately balanced colors and rich design conceptions, are worthy of much greater interest. These are weavings which retain a subtle quality which repeated viewings never exhaust.

Tribal Rugs of Southern Persia refers to these designs in the same terms which one would hear in Iran, as "chicken" rugs. (See cover rug.) Dealers there speak of the morgh (chicken) design. Having sold pieces with the "chicken" label for many years, I don't wish to pull the rug out from under any collectors, but lately a personal preference for the term "bird rugs" has emerged, based on observations concerning the possible origins of the designs. These birds do, indeed, resemble chickens (Illustration 1, cover detail), which are found scratching about in every village and tribal camp in Iran. But the original form of these fowl was probably something of a less domestic nature. In brief, I believe that we can trace the design to a rather common theme in ancient West Asian art, including Persian art, which often depicted pairs of animals or birds facing each other. These pairs of facing creatures were commonly separated by a tree, bush, or a human form. If this possible origin of the design is correct, it would suggest that some versions of this design were sustained for very long periods within the weaving traditions of long-standing south Persian tribespeople who later becamepart of the Khamseh Confederacy. For the moment it is best to approach this idea as a thesis. A more complex presentation is necessary and is possible.

Left Illustration 8. Khamseh rug. 6'8"x4'3", second half 19th century. Accurate dating of older is difficult. The narrow borders point to an older, purely tribal form. Light greens and blues suggest a date prior to 1880. Courtesy of a Montana collection. Right Illustration 9. Khamseh rug. 6'x4'7", c. 1900. This two medallion farmat. Embellished with chevron half-medallions, is simpler and more direct than most three medallion versions. Courtesy of a Massachusetts collection.

The early racial history of Iran is complex but we know that Arab and Turkic elements, both of which were powerful in the Khamseh Confederacy, were late-comers to the scene. Arabs came with the spread of Islam from the Arabian peninsula. Turkic people came in waves from further east. If the Khamseh Confederacy were composed solely of immigranl tribes, it would be harder to see how specific, old Persian designs survived among them. But the Confederacy contains old Persian elements, also. Some of these were absorbed into the invading tribes. Other splinter groups of original stock were not fully absorbed and continued to function as largely separate tribes, speaking their old Persian dialects and no doubt carrying forward other aspects of Iranian tribal culture. One ethnic component of southwestern Iran, the Lurs, stands out in this respect. It is known that there were ancestral Luri elements, or groups closely related to them, such as the Leks, in the regions before the Arabs entered Iran in the 7th century and before the Turkic groups began arriving several hundred years later. Although rare, there are some Luri examples of rugs with the bird motif (see Illustration 2), and we find other highly traditional indigenous designs in other Luri weaving.

One design which has clear antecedents in ancient Persian art and which appears in many bird rugs is the two-headedanimal. A close relationship between a two-headed animal form in a Khamseh rug and a 6th to 9th century Luristan bronze was illustrated in previous writings on this and other designs which can be linked to ancient Persian art. (See Oriental Rug Review, September, 1986, front page and page 2). The present article fits into the same general picture and expands the discussion to include these remarkable bird rugs.

Southern Iran, while not as isolated as Luristan, was not on major trade routes and thus was sufficiently isolated for local design traditions to survive through the past 1,000 years, a turbulent time for many Asian tribes. And so we find this remarkable phenomenon of quite old rug designs, sustained among nomadic and village weavers, even into the 20th century.

This explains to my satisfaction how the bird design and other very traditional forms came to be woven in a confederacy dominated by Turkic and Arab, that is, by non-Persian elements. From the viewpoint of their woven art, the old, conquered residents won out. The general flow of design ideas through the centuries favored the stronger, local design traditions. By following similar lines of examination and logic, we will be able to trace the origin of many of the more traditional Qashqa'i designs, as well. As a rule, Qashqa'i designs are either Luri in origin or were adapted from urban weavings.

Do these origins matter? To this I would say that it depends on the interests of the reader. To understand tribal rugs in their broadest terms, it is helpful to find the sources of some key designs. It also helps to learn about the mixing of various tribal groups of quite diverse backgrounds. Mostly, we like to look at beautiful rugs and skip the fine print. But a deeper appreciation of what we are looking at, a deeper relationship with these objects, can come with the acquisition of understanding about the designs which they contain and the historical patterns of change and continuity which they manifest.

It is not too late to piece together a very broad picture of the movements of rug designs from Persian source groups to others. Such an approach, which emphasizes design relationships more than memorizing tribal and sub-tribal names, can lead to a useful and flexible general understanding.

Before discussing specific illustrations in this article, I would like to correct a point that is raised in the Khamseh Confederacy chapter of my book and which has regrettably influenced several other authors. It concerns the mistaken conclusion that these bird rugs were solely the product of the Arab tribe within the confederacy.

During the 1970s when I was buying rugs in Shiraz, there were, in fact, relatively new rugs of this design which local dealers referred to as "Arab." I took this information about new "Arab-Khamseh" rugs as reflective of l9th century Arab weavings as well. I now see this as an error and, candidly, do not know which of the five tribes produced them. I suspect that Arab, Turkic and Luri elements may all have woven versions of the design.

Left Illustration 4. Detail of a Luristan bronze showing faces and confronting birds. Courtesy of Robert Balsom. Right Illustration 5. Luristan bronze with bird heads facing away.

Tribal Rugs of Southern Persia included three excellent bird rugs on pages 75, 77, and 209. Illustration 3, a detail from page 209, is of particular interest in that it contains hints of the origin of the bird design. Looking at Illustration 3, we see the form of two birds facing each other, separated by a pole arrangement which is decorated with various devices. Among these decorating elements are humanoid faces. Illustrations 4 and 5 show Luristan bronze objects from the 6th to 9th centuries B. C. which also combine bird and human elements. Illustration 6 shows an Assyrian cylinder seal design from roughly the same period which conveys a variation of this theme. Illustration 7 depicts an earlier Assyrian seal showing winged creatures facing a sacred tree.

Left Illustration 6. Assyrian versions of the heroic theme of a man or woman confronted by pairs of birds or animals predate Persian examples. Right Illustration 7. Ancient Assyrian cylindrical seals were decorated with designs from sacred mythological tales.

These examples from ancient West Asian art are shown to suggest the broad family of design ideas which are roughly parallel to what we see in the bird rugs. The correspondences are not identical and I do not point to a specific object as the source of the theme of facing birds. It was widespread. That it touched on nomadic art of ancient Iran is evident from the Luristan examples, which were found in the graves of ancient nomadic inhabitants whose precise ethnic identity is uncertain. Nomads and villagers in the region quit making bronze art sometime in the 7th century B.C., but the need for woven articles of a utilitarian nature continued there, nearly to the present. These objects did not have to contain specific designs in order to serve their function, but the tradition for decorating them with ancestral motifs was firmly established and must have held deep appeal for the weavers and the users of these articles. We see a great deal of variety among the bird rugs within several families of the design. What we can view as the classic archetype is a two or three medallion design, with birds within the medallions and also in the field around them. The Khamseh rug on the cover of this journal is an exemplary piece. Illustrations 8 and 9 are also excellent examples. Illustration 10 shows an unusual format, with birds solely within the medallions, the field being taken up with a "cane" pattern. Tribal Rugs of Southern Persia includes an example without medallions on Page 77.

Illustration 11. Khamseh bagface, 1'9"x2', last quarter 19th century. This Khamseh bagface contains a simplified version of the bird design. Note the stylized animals in the outer border. From author's collection. Illustration 10. Khamseh rug, 6'2"x4'1", second half 19th century. Sizes of bird rugs vary, but most are in standard rug formats, roughly 6'x4'. The "S" designs, common in rugs of this type, are readily seen in this piece. Courtesy of a North Carolina collection.

That some Khamseh nomads continued to use this design on strictly utilitarian articles is suggested by Illustration 11, the front of a l9th century saddle bag. Closely related examples from the Qashqa'i tribe are known, some numbers of which appear to have been woven for commercial purposes rather than indigenous use.

Birds appear frequently in Qashqa'i rugs but to date I have seen no rugs of the classic bird rug type which come from this tribe. We often see the Qashqa'i label on these rugs, however. Perhaps the passion for this mislabeling will eventually fade.

There is a great deal to learn about and from these bird rugs. But, first and last, they are something to enjoy. Sorting out what makes the great examples great and the fair only fair is something that initially everyone needs to approach for himself or herself. But in any season we all can enjoy the charm of a colorful carpet of birds, and very old birds, at that.