国立情報学研究所 - ディジタル・シルクロード・プロジェクト
『東洋文庫所蔵』貴重書デジタルアーカイブ

Sec. v] RELICS OF TEXTILE ART FROM THE TOMBS OF ASTANA   677

Standing quite by itself at Astàna and of particular interest is the design of the figured silk in the face-cover vii. I. o6 (Pl. LXXX). It consists of a lozenge lattice pattern, formed by bands of heart-shaped leaves, and of an eight-pointed star in the field of each lozenge, containing a flower with four heart-shaped petals. The general design as well as the decorative details show unmistakably close resemblance to those found on the one hand in a late Hellenistic fabric from Antinoe at Berlin, and on the other in a Byzantine silk at Liège which Professor von Falke dates approximately from the first half of the seventh century.23 I believe we may safely recognize in this interesting Turfân relic an import from the silk industry of the Near East. The pieces of a ` Sasanian ' fabric used in the curious patchwork ix. 2. 019 (Pl. LXXVII) are too small to permit of a determination of its special character.

We now turn to a small but distinctly instructive group of silks which illustrate the influence exercised by Western designs upon what unmistakably is the work of Chinese weavers. First among them may be noted two fabrics in which there is clear evidence of Chinese manufacture, not only in the style of treatment, but also in the peculiar ` warp-rib ' weave. We have previously noticed this distinctive technique as characteristic of all the oldest Chinese silk textiles, whether from the grave-pits of Lou-lan or the Limes of Tun-huang.24 That we should find some specimens of it surviving in the Astâna tombs cannot cause surprise, since we find it also represented in at least one relic preserved in the Tun-huang hoard for the deposit of which the terminus ad quern is about four centuries later.25 The design of the fabric, pieces of which have been sewn together in ix. 3. 02 (Pl. LXXX) to form what looks like a cushion cover, shows two rows of medallions of ` Sasanian ' type one above the other, with the usual pearl border. Within both there are seen pairs of winged horses, treated in an animated style recalling the horses which appear in silks preserved at Nara,26 but in varying poses. The floral motif used as base in both rows of medallions is also Chinese, and resembles that found in a corresponding position in a printed Chinese silk from Chien-fo-tung, imitating a ` Sasanian ' design.27 In the other ` warp-rib ' fabric, i. 5. b. (DI (Pl. LXXIX), a much-decayed face-cover, we see parts of ` Sasanian ' medallions which are filled with confronting swans on either side of a flowering tree and with subsidiary flower shapes. A flying crane appears in the spandrel. Chinese feeling pervades the design notwithstanding its stiff ` Sasanian ' setting.

In the better preserved fragment v. 2. 0I (Pl. LXXVIII, LXXXI), also a face-cover, the arrangement of the two oval medallions, one placed above the other, each with an outer pearl border and an inner of fleur-de-lis scrolls, is obviously borrowed from ` Sasanian ' models. But the pairs of confronting animals, two in each, as well as the foliate scrolls at the base, are treated with Chinese freedom. The cockatrices in the upper medallion closely resemble those found in silks from Nara,2s while what remains of the lower one shows a finely designed winged lion. Pairs of running deer and of sheep-like animals appear in the spandrels. It is curious to note that the Chinese weaver has preserved something of the ` stepped ' outlines of the Western models of this design. The weave is characterized by Mr. Andrews as a transition from ` warp-rib ' to twill. Among the remaining specimens of this ` Sino-Sasanian ' group the fragment of a face-cover, i. 3. a. oI (Pl. LXXIX), well preserved in its colours, is of particular interest, because the geometrically treated flower ornament filling the spandrel between the medallions is almost identical in design

Western lattice pattern with heart shapes.

Western designs in Chinese stuffs.

Other Sino• Sasanian ' patterns.

23 Cf. v. Falke, loc. cit., i. p. 33, Fig. 34 ; ii. pp. 7 sq., Fig. 228. We have, perhaps, a later derivative from the same source in the pattern of Ch. 00169, Serindia, iv. Pl. LV.

24 See above, i. pp. 233 sq.

25 Cf. Serindia, ii. p. 963, Ch. oo1 i8 ; Andrews, Chin. Figured Silks, p. 14, Fig. io.

26 See v. Falke, loc. cit., i. Figs. 11o, 11r.

27 See Serindia, iv. Pl. Cxvi. A, Ch. 00291.

28 See v. Falke, loc. cit., i. Figs. 118, 119.

II   4 s