 |
 |
In The Plateau Style
by Tom Cole
Originally appeared in HALI 131, © 2003
|
.
 |
|
 |
|
Mirror images, the central devices in a Tibetan rug and Mongolian felt |
| Few old Tibetan rugs that verifiably pre-date the late 19th century have survived the intensive use to which they were subjected, and the history of Tibetan rug weaving has been obscured by a dearth of documentation and observation. However, despite its apparent isolation, art from the Tibetan Plateau has, over many centuries, synthesised ideas and images derived from contacts stretching from Central Asia and East Turkestan, through Nepal and India, to Mongolia and China. | In the pages of HALI, and elsewhere, there has been some debate concerning the sources of designs encountered in rugs from the Tibetan Plateau. In previous articles (HALI 49, 1990; Oriental Rug Review XII/4, 1992, ORR XIII/3, 1993), I have speculated on the nomadic and/or tribal origin of certain Tibetan pile rugs. However, all the rugs featured in those discussions were examples of familiar types for which motif comparisons, and by inference shared iconography, can be found in the wider Central Asian carpet design repertoire. |
 |
|
Figure 1. An unusual and quite old Tibetan rug, one of a pair |
| Some years ago, however, a mirror-image pair of unusual old rugs (1) were found in Tibet which do not just depict a related theme or bear a cursory resemblance to tribal rugs from Central Asia. Woven in the characteristic looped-pile technique associated with the Tibetan Plateau, instead they have a design that is apparently of ancient 'Mongol' origin. The same design, consisting of a roundel depicting a 'lion-dog' motif in the central field of a compartmentalised 2-1-2 format, also appears on an apparently old 'Mongolian' felt cover (3). Taken together, the rugs and the felt appear to give credence to the idea of a historical relationship between the 'Tibetan' people and the hordes of Genghis Khan's Mongolia and, by association, with all the steppe tribes who were conquered by, or sought protection under the aegis of, the Great Khan. In A Thousand Years of the Tartars (1894), a history of the steppe nomads based on original Chinese sources, beginning with the first references to the nomads “whose country was on the back of |
a horse”, up to the conquests of Genghis Khan, Professor E.H. Parker wrote: “The result of a great battle was that many of the women passed over to new masters; the captive youths became warriors, usually under their own chiefs, but subject to the supreme control of the conqueror... Slave and master lived very much in the same way, the only difference being that one did the menial work whilst the other enjoyed himself; meanwhile the women, habituated to the idea of passing from one man to the other, even in their own tribes, only had to undergo the rough excitement of an extra embrace from a man who was not of their own particular choice… Under such circumstances, though the main distinctions of Hsiung-nu and Tungus are always preserved, it is not to be wondered at that languages got intermingled, tribes hopelessly mixed up, and customs interchanged.” |
 |
|
Figure 2. A khan of the Hsiung-nu seated on an audience carpet with a similarly compartmented field. |
| Modern maps define Mongolia as an area of steppe grasslands sandwiched between the two great powers of Asia, China and Russia (Siberia). But the combined territories of Inner and Outer Mongolia (the former is part of the Peoples Republic of China) bear little relation to the vast area that fell under Mongol control or influence even before the period of great expansion in the 12th century ad and the subsequent 'Pax Mongolica'. Various tribal groups found it politically and economically viable to align themselves with other hordes, and we may assume that oral histories and cultural traditions were shared by many of the steppe tribes. If one embraces the idea that the style and manner in which a tribal woman is taught to weave never changes, even if she |
marries into or is taken captive by another group, perhaps our rugs were woven by a Tibetan woman married to a Mongol man? Or did some Mongol weavers use this knot as well? The story of the Mongol rug weaving tradition is obscured in historical ambiguity. The clearest early documentation of 'Mongol' rugs is their depiction in the well-known Song dynasty scroll paintings (late 12th or 13th century) that tell the story of the return to China of Lady Wen-chi, a 2nd century Han noblewoman who was abducted while travelling in Gansu by nomadic Hsiung-nu raiders, spending twelve years as the wife of a Mongol chief before being ransomed by a Chinese embassy. The Hsiung-nu of this period are generally associated with the Turkic hordes of a later date. |
 |
|
Figure 3. A felt of apparent Mongolian origin |
| The scroll paintings depict rugs with the ubiquitous 2-1-2 medallion composition and the field divided into three defined areas, also seen on our rugs and felt. This arrangement is also associated with certain Anatolian rugs of the 15th and 16th centuries, which in all likelihood derive their design inspiration from an earlier prototype, carried westwards by the people who came to inhabit much of Anatolia, the Turko-Mongol hordes from the Central Asian steppe. One in the series of thirteen paintings clearly shows the compartmentalised field format of what is apparently an |
audience rug (2). The medallion in the central compartment is framed by a minor border of back-to-back 'C' motifs, while other less easily identifiable motifs decorate the end panels. The motifs within the compartments may be compared to those seen on textiles of undetermined attribution but most likely of steppe provenance. As I suggested in HALI 67 (1993), many of the textiles presently attributed to 'Chinese' sources should perhaps be considered from another perspective. Examination of specific design elements reveals some interesting comparisons. |
 |
|
Figure 4. An early "Chinese" cotton textile, Eastern Han Period, 25-220 CE |
| While the checkerboard design has probably been with us since man first attempted weaving, the form of its rendition on the rugs is peculiar. The squares of colour are not contiguous, unlike most Tibetan checkerboards with which we are familiar; the blue squares float freely but in an orderly fashion on the brown ground. In a fragmented textile (4) attributed to China in the Eastern Han period, a human figure in one of the compartments is clearly not Chinese and appears to have Indo-Scythian (Central Asian) racial characteristics. To the right of the figure is a similarly disjointed checkerboard arrangement, possibly part of a wide border. The adjacent reptilian body is also quite an intriguing figure to encounter on an apparently early Central Asian textile, as it is reminiscent of the omnipresent dragons of later Chinese and Tibetan rugs and textiles. This textile also | shows the ubiquitous sawtooth minor border, seen too on the 'Mongolian' felt, where it replaces the 'pearl' outer minor border of the pile rugs. This so-called pearl border, also seen here in another early Central Asian textile fragment (5), is commonly encountered as a minor border within a main outer border, usually the 'Greek key' or a related 'swastika' variant. In these rugs, however, it is drawn to almost the same scale and serves to minimise the impact of the meandering 'T' form. Curiously, this motif also constitutes the outermost border on the audience rug of the Hsiung-nu chieftain in the Wen-chi scroll painting. While the history of the pearl border has been commonly (perhaps mistakenly?) associated with Sasanian art, it may well be derivative of a steppe aesthetic. |
 |
|
Figure 5. Silk textile fragment, Han Period, 206 BCE - 220CE |
| Another Han period embroidery (6) shows patterns bearing a close likeness to the devices employed in the corners of the Tibetan rugs under discussion. In the rugs, it almost appears to be a swirling cross (swastika variant?) overlaid upon an 'X'. A comparison of (6) to a drawing reproduced in James Opie's Tribal Rugs, (p.38, fig.4.7) depicting a 3rd millenium bc Middle Eastern animal-headed swastika is interesting; there may be some relationship to the later Ordos 'animal-style' art of the steppes. Swastikas capped with animal heads also appear in 19th century rugs attributed to the 'Baluch' groups of Khorasan and western Afghanistan. | The central medallion in the rugs and the felt is a curiosity which defies attempts directly to correlate the primary element with motifs in other examples of steppe art. The flames surrounding the roundel may represent the influence of Sasanian art, but the lion within presents a puzzle in this discussion of design origins. One may point to a Chinese source for this lion (or lion-dog), but it is drawn in a decidedly different manner to that normally encountered on Ningxia 'lion-dog' rugs or in other media. Apparently standing on its front paws, with a coin or ball grasped in its teeth, the pose is quite distinct from that which we see in Chinese textile art. |
 |
|
Figure 6. Wool embroidery fragment, Han Period, 206 BCE - 220CE |
| This particular rendition of the lion-dog also allows for a dualistic interpretation; one leg and haunch of the beast appear to resemble the head of an elephant. This manner of presenting two animal images within one is not uncommon in the Bactrian agate seals from Central Asian excavation sites. It should also be noted that the feet of the lion appear to be clad in boots resembling traditional Mongolian attire. The minor elements of the roundel also suggest an influence foreign to the traditional perception of Tibetan rug art. Two sets of three circular balls float in the roundel flanking the lion, reminiscent of the çintamani motif which Gerard Paquin (HALI 64, 1992) surmised has its roots in Mongol art. The manner in which 'pearls' surround the floral form at either end of the central compartment is again reminiscent of Sasanian (and Tang) textile art. The design comparisons discussed above leave us no wiser as to the context in which our rugs were woven, by whom, where, |
and for what purpose? Despite the use of a knotting technique associated solely with rugs from the Tibetan Plateau, the nature of their design and their unusual size (somewhat larger than standard 19th/20th century 'Tibetan' khaden), may indicate a non-Tibetan origin, as might their possible identification as 'audience' rugs, a format not commonly associated with Tibetan rustic weaving. But while their function, age and origin may be disputed, the unusual design format is not easily dismissed and distinguishes them from other known 'Tibetan' weavings. However, given that over the centuries Tibet was populated by diverse tribes representing different ethnic and language groups, as well as persistent references to the presence of 'Tartar' groups within the Tibetan realm, it should come as no surprise to encounter such atypical Tibetan rugs. Indeed their existence affirms the diversity of 'plateau style' weavings, one of the few art forms from the Tibetan Plateau which developed within the secular confines of the Tibetan home. |
 |
Detail of the Tibetan rug illustrating a design similar to htat seen in the early textile seen in Figure 6 |
| It is quite possible that other rugs made in the 'plateau style' by the same Tibeto-Mongol groups exist, but have yet to be recognised. This unusual design may provide some clues concerning 'Tibetan' production and the array of the ethnic groups we refer to as Tibetan. Just as importantly, the dissemination of the peculiar knotting technique (also known as the 'Sehna loop'), must be studied more closely. Which, if any, other Central Asian peoples employed this technique, when, and where were they located? Why has it survived in Tibet and nowhere else on the steppe? | It would be easier to dismiss the two rugs and the felt as representatives of an abberant design type, if it were not for C-14 tests carried out at Oxford in 1992. In these, the rug (1) and the felt (3) were found to have uncalibrated radiocarbon ages that indicated a possible 17th century date for both textiles. The result for the second rug is closely comparable. Interestingly, a felt of similar type, with a different design, yielded a calibrated age range of 1481-1655 when tested much more recently in Zurich (HALI 130, p.99). |
 |
|
Detail of the Tibetan rug showing the disjointed checkerboard compartment |
| The case for a relatively early date for both rug (1) and felt (3) must therefore consist of qualified speculation based mainly upon history as interpreted in various sources, but taking into consideration C-14 tests conducted during the formative stages of an evolving scientific process for the dating of textiles. It is certainly possible that (1) and its companion may be among the earliest surviving 'Tibetan' | rugs known, and the close correspondence between the C-14 results for the rugs and the felt, indicative of near contemporaneous manufacture, greatly diminishes the likelihood that the rugs are later copies of an earlier 'Mongolian' prototype, while both design and technique suggest a tangible relationship between the Turko-Mongol steppe tribes and the people we now know as Tibetans. |
|
Original photos and text appeared in HALI 131, © 2003 |
| <---previous article |