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

is itself traced back by etymologists to Skr. karpūra, « camphor », BERNEKER saying that kāfūr is
Persian, most others that it is Arabic. As a matter of fact, it is common to both languages, but, as
a rule, the -p- ought to have been retained in Persian, whereas in Arabic, which has no p, the original
-p- could be rendered only as -f- or -b-. So it seems that kāfūr is an arabicized form, which was
adopted by the Persians at an early date; but it is somewhat surprising to find the arabicized form
in Latin as early as c. 540.
As to Arabic kāfūr, LOKOTSCH explains its derivation from karpūra by saying that, already in
Prākrit, there was a form kappūra (it is the Pali form); but kappūra ought to have given in Arabic
*kafūr rather than kāfūr. On the other hand, we must not forget that Skr. karpūra probably
represents a pre-Aryan word, as the case must be also with Skr. karpāsa, «cotton» (see «Cotton»),
and that in Javanese the word for « cotton » is kapas (Malay kāpas), just as the Javanese word for
« camphor » is kapur (Malay kāpur). In the case of karpāsa, I have pronounced in favour of a deri-
vation of kapas from karpāsa (Pali kappāsa), because I think that the cultivation of cotton is
extremely ancient in India, but the case of camphor is different. In the past, it was primarily an
Indonesian product, and it may well be that it is not kapur which is derived from karpūra, but kar-
pūra which has been formed on a pre-Aryan word very close to an original Indonesian form of the
kapur type. In any case, since there was in the first centuries of our era a maritime intercourse
between Arabia and Indonesia, it would seem to be a natural solution to suppose that Arabic kāfūr
does not represent Pali and Prākrit kappūra (< Skr. karpūra), but was borrowed directly from the
Indonesian kafur. GERINI's objections (Researches, 810) are of no value, since they are based on
the assumption that the other Malay word kāpur, « lime », « plaster », most probably an Indonesian
word [cf. CABATON, Dict. Čam-Français, 57], is to be traced back to Skr. karpūra, which never had
that meaning. I leave, however, the question in abeyance, because some camphor was also gathered
in southern India; moreover, most of the modern Indian dialects now have the form kapūr, in Mah-
ratī kāpūr (cf. J. BLOCH, La formation de la langue marathe, 309), which also may account for Arabic
kāfūr. Tibetan ga-bur (> Mong. gabur), which puzzled LAUFER (Sino-Iranica, 591), is merely one
of the many sonorized transcriptions in that language, of the same type as Skr. keśara, « pistil »,
> Tib. ge-sar, or Skr. kuṅkuma, « saffron », > Tib. gur-gum and gur-kum. KOVALEVSKIĬ's
Mongol gatbura or gadbura (Dictionnaire, 2431) almost certainly rests on the usual misreading
d instead of r in a Tibetan original which probably gave the correct Sanskrit form karpūra; LAUFER's
explanation, based on a mistaken restoration of the ancient pronunciation of a Chinese transcription,
is a failure.
The Indian word for « camphor » occurs in Chinese only in Buddhist texts, as 羯 布 羅 chieh-
pu-lo (*kjāt-puo-lâ; not with sonant initials as in LAUFER, Sino-Iranica, 591) and 劫 布 羅 chieh-
pu-lo (*kjop-puo-lâ). The first form is the one employed by Hsüan-tsang (Mémoires, II, 123; Vie,
193) and, as usuel with him, correctly renders Skr. karpūra. the second, clearly based on a Prākrit
kappūra, occurs in a translation made by Bhagavaddharma c. A. D. 650-660 (No. 1059 of Hōbōgirin)
fascicule annexe; it is not listed in Nanjiō). A third form 羯 婆 羅 chieh-p'o-lo (*kjāt-b'uâ-lâ)
is given in GILES's Dictionary (No. 9412), STUART, Materia Medica, 157, and TARANZANO, Voca-
bulaire, I, 239; and is said by GILES to represent Malay kāpur. The source of this information is the
Pên-ts'ao kang-mu (34, 38 b), where the form chieh-p'o-lo is ascribed to the Pên-ts'ao yen-i, and LI