^{page 757 note 1} Bulletin de l'Ecole française d'Extrême-Orient, vol. xiv, No. 4, 1914.Google Scholar

^{page 757 note 2} Attempts in this direction have not been many in the past. We have from the pen of C. de Harlez a study on “Le chinois parlé au vi^e siècle. A.C., d'après l'I-li” (T'oung Pao, vol. ix, pp. 215–25, 1898)Google Scholar, and from Jametel, M., “L'argot pékinois et le Kin-ping-mei” (reprint from Mém. Soc. sinico-japonaise, vol. vii, Paris, 1888, pp. 18)Google Scholar; but de Harlez and Jametel, on the ground of their literary sources, treat only of the style and phraseology of colloquial speech, not, as M. Maspero does, of its structure. His work, therefore, is entirely original.

^{page 758 note 1} Maspero, , loc. cit., p. 34.Google Scholar

^{page 758 note 2} Maspero, , loc. cit., p. 34.Google Scholar

^{page 758 note 3} The interesting coincidence with Lo-lo a-sa (Nyi dialect) and a-so (A-hi dialect), “who?” old Burmese a-su (see Houghton, B., JRAS., 1896, p. 33)Google Scholar, Lo-lo a-mi, “what?” (Burmese a-be, “what?”), Mo-so a-ne (“who?”), and a-tse (“what?”), may be pointed out right here, as well as the apparent relationship of Chinese šui, Tibetan and Newari su, and Lo-lo so, sa.

^{page 758 note 4} Maspero, , loc. cit., pp. 25–6.Google Scholar

^{page 758 note 5} Maspero, , loc. cit., p. 34.Google Scholar

^{page 758 note 6} Giles, , Chinese-English DictionaryGoogle Scholar, No. 12045 (likewise in Palladius).

^{page 759 note 1} Pelliot, , Bull, de l'Ecole française, vol. ix, p. 573, n. 1, 1909.Google Scholar

^{page 759 note 2} Liétard, A., Bull, de l'Ecole française, vol. ix, p. 553, 1909Google Scholar. The first element, me, in the Lo-lo word, is identical with Mo-so mö (m′ö), Si-hia mei, Tibetan mig, Burmese myak, Gešitsa dialect of Tibetan muk, Chinese muk . Lepcha a-mik and Southern Chin a-mi are analogous in form to ancient Chinese a-tu. A further relationship of the word tu(du) might possibly be given in the series ta of the T'ai languages.

^{page 759 note 3} Watters, (Essays on the Chinese Language, p. 366)Google Scholar states that the expression a-ko is in very common use among the Chinese as a respectful mode of address. He is quite correct in assuming that, if the same word was chosen for the rendering of Manchu age or agu (not, as written by him, agü), this was partly due to the meaning of the Chinese term. A-ko, accordingly, in this case, is not the Chinese transcription of a Manchu word, but the assimilation of a pre-existing Chinese term to the latter.

^{page 759 note 4} Arendt, C., Handbuch der nordchinesischen Umgangsprache, pt. i, p. 282.Google Scholar

^{page 760 note 1} For example, A-yü (Rocher, E., T'oung Pao, vol. x, p. 347, 1899)Google Scholar, A-hêng (Chavannes, , Mémoires historiques de Se-ma Ts'ien, vol. i, p. 178Google Scholar; vol. v, p. 196); A-jung mentioned in Tain shu (see Pétillon, , Allusions littéraires, p. 274)Google Scholar; A-po mentioned in Sui shu, Ch. 51 (Chavannes, , Dix inscriptions chinoises de l'Asie centrale, p. 28, n. 3)Google Scholar. Giles, in his Dictionary, cites A-hiang as the name of a fairy who assists the God of Thunder by pushing his oar. In Mayers (Chinese Reader's Manual, p. 1)Google Scholar we read of A-kiao A-man and A-to u . Tao-yün elegantly spoke of her uncle as A-ta-chung-lang (Lockhart, , Manual of Chinese Quotations, p. 130)Google Scholar. Compare also a-weng (“grand father” or “father-in-law”) in Pétillon, (loc. cit., pp. 126, 259)Google Scholar, a-kia (“mother-in-law,” ibid., p. 259), and a-p'o (“vieillard,” ibid., p. 418).

^{page 760 note 2} Einleitung in die Kawi Sprache, p. cccliv.Google Scholar

^{page 760 note 3} Inserted in his work De l'influence de l'écriture sur le langage (Darmstadt, 1835).Google Scholar

^{page 760 note 4} Loc. cit., pp. 144, 244, 256.

^{page 760 note 5} Mélanges asiatiques, vol. i, pp. 301–3.Google Scholar

^{page 762 note 1} See Rockhill, , Diary of a Journey through Mongolia and Tibet, p. 88Google Scholar. de Lacouperie, T. (Les langues de la Chine avant les Chinois, p. 71)Google Scholar noted that in certain dialects of the Miao tribes of Southern China a prefix a- is joined to terms of bodily parts (a-pu, hand; a-t'au, foot; a-biu, ear).

^{page 763 note 1} The name A -nu appears as that of an author of two grammatical works in the Tanjur (vol. cxxiv, Nos. 2, 3)Google Scholar. Schiefner supposed the identity of this A-nu with Anubhūti, which is purely a conjecture unsupported by evidence. True it is that there is also a Sanskrit proper name Anu; but it does not follow therefrom that a Tibetan name A-nu is borrowed from India. On the contrary, it appears as a genuine Tibetan word, being derived from nu-bo (“younger brother”) in the same manner as, for instance, a-jo from jo-bo (“elder brother”). There seems also good reason to believe that the Tibetan tradition ascribing the authorship of these two treatises to T'on-mi A-nu (T'on-mi assuming his father's name) is well founded (see Huth, , ZDMG., vol. xlix, p. 284, 1895).Google Scholar

^{page 763 note 2} Rockhill, W. W., Land of the Lamas, p. 189.Google Scholar

^{page 764 note 1} Jäschke has queried this word with an interrogation-mark. It was derived by him from the dictionary of I. J. Schmidt, who on his part culled it from the Tibetan-Mongol dictionary rTog-par sla-ba, where it is indeed thus given and rendered by Mongol sain. The word presumably belongs to a dialect of Amdo.

^{page 764 note 2} Das, Chandra (Tibetan-English Dictionary, p. 1342)Google Scholar states that a-gyis is “an interrogative pleonastic term signifying: have you done it or done so?” This is possible; in this case, however, a has nothing to do with the prefix here in question, but is the interrogative particle a largely employed in Eastern Tibet (cf. Desgodins, A., Essai de grammaire tibétaine, p. 26)Google Scholar. This phrase, accordingly, has no right to be in the lexicon any more than several others listed by Das under separate headings, as though they were independent expressions; as, for instance, a-t'ul, explained as “a colloquial expression of doubt as to whether an enemy would be vanquished”; a-ñan, “hesitation to listen to one's advice”; or a-drag, “doubt as to whether a thing is good or bad.” In these examples we simply encounter a verbal form or adjective prefixed by the interrogative particle a; and they should have been entered under the latter as catchword, which, however, is not given under a.

^{page 766 note 1} Cf. Col. Mainwaring, G. B., Grammar of the Róng (Lepcha) Language, pp. 111–12, Calcutta, 1876Google Scholar, and Mainwaring-Grünwedel, , Dictionary of the Lepcha Language, p. 439Google Scholar. Grünwedel compares with the first category Burmese č'a (“to be hungry”)—a-č'a (“hunger”); with the second, Burmese im (“house”)—a-im (“sheath”). See also Vossion, L., Grammaire franco-birmune d'après A. Judson, p. 22.Google Scholar

^{page 767 note 1} Tibetan lče, accordingly, is composed of two elements, le + če, the former being preserved by Gurung, etc., and Si-hia la (glossary of Ivanov), the latter corresponding to Chinese she .

^{page 768 note 1} Cf. Grierson, , Linguistic Survey of India, vol. iii, pt. i, p. 183Google Scholar. I do not believe that the a, as here marked, is long (cf. also the editor's remark on pronunciation on p. 182).

^{page 768 note 2} Joshi, , Grammar and Dictionary of Kanāwari, pp. 29, 32, Calcutta, 1909.Google Scholar

^{page 768 note 3} “Note on the Languages spoken between the Assam Valley and Tibet” (JRAS., 1902, p. 134).Google Scholar

^{page 768 note 4} Vol. iii, pt. i, p. 589.

^{page 769 note 1} Dictionnaire français-lolo, p. (21).Google Scholar

^{page 769 note 2} A-bi is known as the inventor of Lo-lo script (Devéria, , “Les Lolos et les Miao-tze,” p. 7Google Scholar, extrait du Journal Asiatique, 1891). The Lo-lo adopt as personal name also the terms of the zoo-zodiac under which they have been born, this term being linked with the prefix a-; for example, A-nu (“born in the year, month, or day of the monkey”), A-ie (“Mr. Rat-Year”), A-jo (“Mr. Sheep-Year”); see Vial, P., Les Lolos, p. 37 (Shanghai, 1888Google Scholar, publication of Siccawei).

^{page 770 note 1} A comparative series of this word in the various Lo-lo dialects is given by Liétard, A. (Bull, de l'Ecole française, vol. ix, p. 552, 1909).Google Scholar

^{page 770 note 2} Liétard, A., Bull, de l'Ecole française, vol. ix, pp. 290, 294, 555, 1909Google Scholar. In Kachin also the prefix a- enters into the formation of colour adjectives; for instance, a-čyaṅ (“black”). Likewise in Thādo, which belongs to the group of Northern Chin languages, and is spoken in southern Manipur: a-vom (“black”), a-yeṅ (“green, yellow”; cf. a-yeṅ, “turmeric,” in the same language), a-boṅ (“white”), a-wa (“bright, light”; from wat, “to shine”), a-yiṅ (“dark, dense”); see Hodson, T. C., Thādo Grammar, pp. 61, 64 (Shillong, 1905)Google Scholar. The same feature occurs in Lepcha: a-nok or a-tyaṅ (“black”), a-bok or a-doṅ (“black and white”), a-dum (“white”), a-foṅ (“green”), a-hyir (“red”).

^{page 770 note 3} A. Liétard, ibid., p. 289.

^{page 770 note 4} Au Yun-nan. Les Lo-lo P'o, p. 217 (Münster, 1913).Google Scholar The premature death of Father Liétard, who died on July 5, 1912, in Chao-t'ung, Yun-nan, before the publication of his important work, is an irreparable loss to science.

^{page 770 note 5} The word šö apparently is a Chinese loan-word, derived from ši ; and it is particularly interesting that Maspero, M. (loc. cit., p. 13)Google Scholar has discovered the Chinese counterpart of the above Lo-lo term in the form a-ši (“maître, moine”). The affix p'o means “male”.

^{page 771 note 1} More examples will be found in Liétard, A.'s “Essai de dictionnaire lo-lo français” (T'oung Pao, 1911, pp. 17–21).Google Scholar Lolo a-nö (“milk”) is comparable with Tibetan nu-ma, Chinese nou ; Siamese nom; a-ṅo (“fish”), with Tibetan ña, Lepcha ṅo, Mo-so ṅi, Chinese ṅü, ṅ

^{page 771 note 2} Bull, , française, de l'Ecole, vol. ix, pp. 549–56, 1909.Google Scholar

^{page 771 note 3} Burmese, kräkGoogle Scholar is doubtless related to Tibetan (written language) sgog-pa, (“garlic”), Tibetan, Yün-nangau-paGoogle Scholar, Tibetan, Sung-pančon-grogGoogle Scholar (Potanin, , Tanguto-Tibetan Border-Land of China, vol. ii, p. 395, in Russian).Google Scholar Cf. further Lepcha, sun-guGoogle Scholar (“garlic”), A-hi Lo-lo šo, Lo-lo-p'o šu, Nyi Lo-lo še-ma.

^{page 772 note 1} Bacot, J., Lea Mo-so, p. 59.Google Scholar Mo-so ne apparently is identical with Chinese na and Mo-so tse with Tibetan či.

^{page 772 note 2} Cf. Cordier, H., T'oung Pao, 1908, p. 683.Google Scholar

^{page 772 note 3} T'oung Pao, 1912, pp. 6)Google Scholar 1, 614 et seq. It is a feature of particular interest that in the “History of the Yüan” (Yüan shi, chap. 61, p. 4)Google Scholar the name A-liang is written Me-liang, and the name A-hu Me-wu (ibid., p. 569). M. Chavannes explains this word Me as a tribal name, the Chinese being in the habit of prefixing to the name of a chief that of his tribe, which was gradually looked upon by them as his family-name. Thus Me was exchanged for Mu, the name of the Mo-so chieftains of Li-kiang, in 1382, when the latter themselves adopted the Chinese custom of family-names. This point of view is confirmed by Sü Hung-tsu, (1586–1647)Google Scholar, who says that all the chiefs of Li-kiang bore the family-name Me from the time of the Han down to the Ming, when T'ai-tsu altered it into Mu. Of course, neither the Mo-so nor the Lo-lo or Tibetans ever had family-names (“Les Lolos n'ont pas de noms patronymiques, ils n'ont qu'un nom personnel,” says Vial, P., Les Lolos, p. 37)Google Scholar; and it is solely Sü Hung-tsu's personal viewing of the matter when he takes Me for a family-name. But is Me really a designation of ethnic value which as such ever had currency among the Mo-so themselves? I venture to doubt it, despite the alternating forms Me-ch'a and Mu-ch'a pointed out by M. Chavannes. What is certain to me is, that the element me-, as shown by the Yüan shi, is a prefix on exactly the same footing as a-; Me-liang is identical with and the equivalent of A-liang. The labial prefixes, ma-, me-, and m- are very frequent in numerous Indo-Chinese languages (ma-, for instance, in Lepcha, and in Chulikata and Digaru Mishmi [Linguistic Survey of India, vol. iii, pt. i, pp. 614, 616]Google Scholar, me- in Miri [ibid., p. 589], m- in Miju Mishmi [ibid., p. 619], Tibetan, Mo-so, and Lolo); and what is particularly notable, the prefix ma- is interchangeable with a- in Kachin (Sten Konow, ZDMG., vol. lvi, p. 493, 1902).Google Scholar In this language nearly all personal names are combined with the prefix ma- (Hertz, H. F., Handbook of the Kachin or Chingpaw Language, p. 37, 2nd ed., Rangoon, 1902).Google Scholar Cf. also A-hi Lo-lo me-ne (“cat”) with Cho-ko a-ñi (“cat”). For this reason it is equally probable that at the time of the Mongols there was a period of the Mo-so language when the prefix a- could alternate with the prefix me-.

^{page 773 note 1} Sainson, C., Histoire particulière du Nan-tchao, pp. 116, 125, 145, 247, 258Google Scholar; see also the Index on p. 277.

^{page 773 note 2} Les langues de la Chine avant les Chinois, p. 63.Google Scholar

^{page 773 note 3} Vial, P. (Les Lolos, p. 33)Google Scholar States that the proper mode of writing Chung-kia is ; that is, “heavy cuirasses.” This is somewhat more sensible; but the chances are that Chung-kia originally had no meaning in Chinese, but that it is the indigenous designation of the tribesmen in question, which the Chinese, tant bien que mal, attempted to reproduce in their writing.—It is very curious that a prefix a- in connexion with proper names occurs also in Khmer, which belongs to the Mon-Khmer family of languages, that is not morphologically related to Indo-Chinese. Moura, M. (Vocabulaire cambodgien-français, p. 33)Google Scholar states, “a, devant un nom propre d'homme indique la familiarité, s'il s'agit d'un enfant; il marque le mépris, s'il s'agit d'une personne âgée.” Himly, K. (“Bemerkungen über die Wortbildung des Mon”: Sitzungsber. bayer. Akad., 1889, p. 274)Google Scholar has drawn attention to this prefix in Mon, Stieng, and Khiner: Mon a-kruim (“boastful”) from kruim (“to boast”); a-čak (“link”) from čak (“to unite”); a-gah (“that one”) from gah (“this one”). In view of the profound historical influence of Burmese upon Mon, the formation of nouns from verbal roots by means of a- in Mon might be ascribed to an impetus received from that quarter; for the rest, however, the entire question requires a special investigation in the Mon-Khmer group.

^{page 774 note 1} Grundriss der Sprachwissenschaft, vol. ii, p. 352.Google Scholar

^{page 774 note 2} Linguistic Survey of India, vol. iiiGoogle Scholar, Tibeto-Burman Family, pt. i, p. 595 (Calcutta, 1909).Google Scholar

^{page 774 note 3} ZDMG., vol. lvi, p. 493, 1902.Google Scholar

^{page 775 note 1} Tibetan-English Dictionary, p. 603.Google Scholar

^{page 775 note 2} T'oung Pao, 1914, p. 56, note.Google Scholar

^{page 775 note 3} ZDMG., vol. lvi, pp. 513–4, 1902.Google Scholar

^{page 776 note 1} Examples of the latter kind are cited on p. 492 of his article. Cf. also Hodson, T. C., Thādo Grammar, pp. 9, 13.Google Scholar In the Linguistic Survey of India (vol. iii, pt. i, p. 575)Google Scholar it is observed that the prefixes a-, e-, or u- in Aka are probably identical with the possessive pronoun of the third person, while the prefix na- is explained as being perhaps that of the first person.

^{page 778 note 1} In regard to Chinese see Watters, , Essays on the Chinese Language, p. 136.Google Scholar

^{page 779 note 1} See the Dictionary of Mainwaring-Grünwedel, , p. 440.Google Scholar