National Institute of Informatics - Digital Silk Road Project
Digital Archive of Toyo Bunko Rare Books

Sec. v]   MANUSCRIPTS IN TIBETAN, SOGDIAN, TURKISH   921

were frustrated by the outbreak of the war, and finally destroyed by his lamented death, which took place in 1916.

It is impossible to foresee when and where a competent scholar may be found capable of continuing the labours which were so brilliantly begun by M. Gauthiot on the Sogdian texts of Chien-fo-tung. In the meantime it must suffice here to call attention to a point which presents a distinct antiquarian interest. The paper and outer appearance of the Sogdian rolls resemble so closely that of our Chinese texts of the Tang period from the cave that local production in Tunhuang or the adjoining region suggests itself at least as a possibility. This assumption would well agree with what M. Pelliot has been able to prove from historical notices among Chien-fo-tung manuscripts as to the existence of Sogdian colonies in the Lop tract, and probably eastwards also.'1 It is, therefore, of importance to note that, according to an observation kindly communicated to me by Professor F. W. K. Muller in the autumn of 191o, the Sogdian Buddhist text in the big roll Ch. ci. 00I shows distinct evidence of having been either translated from a Chinese vérsion or at least prepared with the help of Chinese materials.12

We still have briefly to survey those manuscript remains which, though showing different scripts, are all in Turkish language. In regard to these it is particularly gratifying to know that -those texts among them which are the most interesting by their contents and writing, and in all probability also the oldest, have been fully published and adequately elucidated through the care of exceptionally qualified experts. The place of honour may justly be accorded to the remains in Turkish ` Runic ' script, on account of their intrinsic philological value and the fact that it was Professor V. Thomsen, the famous decipherer of that script, first discovered in the Orkhon and Yenissei inscriptions, who did me the honour of undertaking their publication. The remains are few in number ; but among them is what in Professor Thomsen's words ` must decidedly be characterized as the most remarkable, comprehensive, and also best preserved of all the MSS. found hitherto written in the Turkish Runic script'.13

The little book, Ch. 0033, written on fifty-eight leaves of excellent paper of the Tang period and in a fine calligraphic hand, is complete from beginning to end, including the colophon.14 Its state of preservation is perfect, not even the glue which fastens the sheets at the back having loosened. The text, as Professor Thomsen's translation and comments show, comprises sixty-five concise stories and was composed primarily for the purpose of a divination book (called ïrq-bitig, ' fortune-book ', in the text itself). The linguistic interest of the text is great, both on account of the rich supply of words contained in it ' and because there is strong internal evidence to prove that, unlike most, if not all, early Turkish text fragments in this script so far found, it is not a translation from another language. In Professor Thomsen's opinion most outer and inner criteria speak in favour of its being of Manichaean origin, and the great care bestowed on the neat writing, both of text and of rubrics, certainly creates this impression. The cyclical date given in the colophon cannot be exactly determined ; but Professor Thomsen is inclined to assign the manuscript approximately to the beginning of the ninth century. In the colophon the scribe names two students, evidently Manichaeans, ` staying at the residence (or the college ?) of Taigüntan', a locality which still awaits identification.

Manuscripts in Turkish language.

Divination book in Turkish Runic' script.

" Cf. Pelliot, Le ` Cha Tcheou Tou Tou fou fou king', etc., Journal Asiat., janvier—février, 1916 pp. 115 sqq. Cf. also above, pp. 306, 327, 654.

12 With this may be compared M. Gauthiot's remarks on the very close relation existing between the Sogdian text of the Dirghanakha-sielra published by him from a Ch'ien-fotung manuscript in the Pelliot collection and the Chinese translation of the same by I-tsing dating from the beginning of

1374

the eighth century A. D. ; see Gauthiot, Le Sùlra du religieux Ongles-longs, Mémoires de la Société de Linguistique, i912, xvii (reprint), p. 2.

13 Cf. Thomsen, Dr. 2W. A. Stein's manuscripts in Turkish Runic' script from Miran and Tun-huang, JR.A.S., 1912, pp. 190 sqq.

14 For the reproduction of four double pages, see Pl. CLX ; for two more, J.R.A.S., 1912, Pl. H.

6B