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

> > > >
Color New!IIIF Color HighRes Gray HighRes PDF   Japanese English
0508 Notes on Marco Polo : vol.1
Notes on Marco Polo : vol.1 / Page 508 (Grayscale High Resolution Image)

New!Citation Information

doi: 10.20676/00000246
Citation Format: Chicago | APA | Harvard | IEEE

OCR Text


492   183. COTTON

probably the proper reading, the se being due to the contamination of the next name, in which se occurs. Now, Wên-su is the ancient name of U6-Turfan (north-west of Aqsu), and it may be that the cloth manufactured in the Mediterranean Orient was called « Wên-su cloth » by analogy with the one really coming from Wên-su. The T'ai-p'ing yü-lan readings are not to be neglected; for instance, for the name preceding that of « Wên-sê cloth » or « Wên-su cloth », our text of the Wei lio gives « f ft to-tai cloth » (or, in the Ch'ien-lung edition, A { to fa), but, in the T'aip'ing yii-lan, it is « gg ft lu-tai cloth », and lu-tai (*luk-d'âi) closely resembles the red brocade lolatai (or lulttai) of Kâsyari (cf. Brockelmann, 119, where SCHINDLER'S restoration into g ;'; lo-tai, *ldk-tdi, is valueless; lo-tai exists, but its meaning is quite different; cf. TP, 1929, 144). But, even if we read « Wên-su cloth » instead of « Wên-sê cloth », that would only bear evidence

   to the early renown of the textiles of Chinese Turkestan, without providing any indication as to   j

their nature, wool or cotton.

Another case is more intricate. One of the textiles most often mentioned in the Kharothi documents is koj'ava or kosava. LÜDERS (Textilien im alten Turkistan, 3-11) pronounced himself in favour of a Skr. original *kaucapa, itself of foreign origin, and rejected a possible, but

difficult derivation from Kuêi, the ancient name of Kuêà; *kaucapa would be the designation of a woollen blanket or carpet. In my review of LÜDERS'S paper (Oriental. Literaturzeitung, 1938, 184-185), I suggested that Ch. tt g;c ch'ii-shu (*kiuli2u or *g'iu-siau) might be a transcription of the same word (Hui-lin [A, viii, 93 b] says that ch'ii-shu was a foreign word of Central Asia

[ tjj Hu], and that the fabric was popularly known as   mao-chin, « woollen brocade »), and
remarked that the derivative form of *kaucapa, kaucapaka, occurred in ch. 44 of the Vinaya of

the Mnlasarvâstivâdin translated by I-ching, but in a form X p, {   ku-ch'ê po-chia which
would suppose *kochavaka. I added that I-ching, however, was not a very strict phonetician in his transcriptions. As a matter of fact, I have since found, in another Vinaya translation by I-ching (%K, v, 72 b), the same word transcribed n 4 z. kao-chê p'o (*kâu-t'siäp-b'uâ), i. e. kaucava. A text of the Kuang chih, quoted by the K'ang-hsi tza-tien, s. v..{ , (equivalent of

shu), seems to connect ch'ü-shu with po-tieh, i. e. in principle « cotton »; but the text is of doubtful origin, and need not detain us. More important is Hui-sin's gloss on I-ching's *kocha-

vaka (kaucapaka), which is said to be    À tt T   , «the name of a fine good po-tieh
cloth » (such is the text in A.,, ix, 156 a, and in Tokyo Tripit. of Taishô, 54, 713 3; the omission of tieh in SAKAKI's Mandvyutpatti, No. 5861, is an error). Now, a note by I-ching himself, following his transcription kao-chê p'o of kaucava, says that it was the name of a « carpet » (fit t'an; I do not think that the variant ;n t'an can be adopted here, except as a graphic equivalent of t'an, « carpet »). As a matter of fact, I-ching must have written on the authority of some

dictionary; kaucava was the designation both of a blanket worn as a garment and of a carpet; in the text translated by I-Ching, it could not be a carpet, since it was the first of the five garments allowed to the monks by the Buddha. But the word t'an used by I-ching implies that it was a woollen stuff. We should gain nothing by supposing that a confusion was made between *kau-

capa (kaucava, *kaucapaka, etc.) and -   rijjkao-t'an-po-chia of Ch. 18 of the same Vinaya
(3R, viii, 18 a), probably a transcription of *kautumbaka (cf. Divydvaddna, 55910: kautumba; Mandvyutpatti, No. 9163; kotambakam; Pali kotumbara), since the definition of *kautumbaka