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

836 PICTORIAL REMAINS FROM THE THOUSAND BUDDHAS [Chap. XXIII

information on artistic points received from Mr. F. H. Andrews and on inscriptions from Dr. L. Giles and Mr. A. D. Waley. and the valuable iconographic indications which expert Japanese scholars like Professor Taki and Mr. Yabuki were kind enough to furnish on their visits to the collection.

For reasons of practical convenience it has been necessary to preserve in this Descriptive List the order of the ' site-mark ' numbers originally given to the objects as they emerged from the bundles searched at the cave-temple or subsequently from those which had to be carried off, in bulk as it were, to be examined later at the British Museum.13 It is obvious that with this numerical order of entries dependent on chance the Descriptive List itself could not serve for a systematic review of the pictorial relics. For this purpose it is necessary to arrange them first into definite groups, and by comparison of the individual pieces within each to determine the essential characteristics of their iconography and artistic execution. This task was the primary one among those to which M. Petrucci had proposed to devote the second main portion of his study.14 Since death has stayed his hand, I feel now compelled to attempt it in the interest of the present work, but with a scope necessarily reduced to my restricted competence.

The serious limitations imposed upon me are obvious. While familiar to some extent with the iconography of Graeco-Buddhist art and such remains of Buddhist art in the Tarim Basin as I had the good fortune to bring to light myself, I have never found leisure so far for a systematic study of the religious art of the Far East or Tibet, nor even for that of Central-Asian Buddhist art as displayed by the frescoes, etc., recovered in the northern oases of Chinese Turkestan. To these limitations must be added my want of Sinologue qualifications and the fact that the present chapter has to be written far away from my collection and solely with such help as Miss Lorimer's descriptions, plates, inventory photographs, and memory furnish. May the circumstances already referred to and the claims made upon my time by other tasks help to excuse any shortcomings in this chapter.

For the classification of the paintings and drawings I can fortunately avail myself of the guidance afforded by M. Petrucci's memorandum and by the briefer, but equally helpful, notes with which M. Foucher favoured me on his visit to the collection in 19 to." It is certain that neither the varying material on which the pictures are painted, silk, linen, or paper, nor such differences of style and date as examination may reveal in the present state of our knowledge, would form an adequate basis of classification. A grouping by subjects thus becomes necessary, and iconographic considerations make the following scheme appear to me the most convenient.

The first place may be suitably allotted to the silk banners which represent legendary scenes from the life of Gautama Buddha. It is significant for these scenes that their treatment is frankly and purely Chinese. We shall next proceed to the specifically ` iconographic paintings, which divide themselves into two broad categories, according as they are consecrated to individual divinities or to whole groups or assemblages of divine figures. Allowing in the former the first place, as seems fitting, to the rare representations of Buddhas, we shall then review the far more abundant pictures showing different Bodhisattvas, whether alone or with attendants and worshippers. Among them it will be advisable to treat first the very numerous banners. The single divinities shown in them can rarely be definitely identified. But they may conveniently be grouped by the type of figure represented, according as it adheres more or less closely to models

Systematic grouping of pictorial remains.

Limitations of survey.

Classification of paintings by subjects.

'3 The contents of bundles searched at the Thousand Buddhas are distinguished by ` site-marks' showing small Roman numerals (e.g. xxi, xxxviii, lv, etc.) as bundle numbers at the beginning. Those objects which were recovered amongst manuscript bundles, etc., at the British Museum are

marked with consecutive numbers, preceded by oo, after the general 'site-mark' Ch. (e.g. Ch. 0017, Ch. 00452, etc.).

74 Cf. below, Appendix E, I.

75 See above, p. 833.