^{page 1017 note 1} LSI, Vol. 3, Pt. 1, p. 47.

^{page 1018 note 1} See “The Vocalism of Sino-Tibetan”, JAOS. 61 (1941), 19.

^{page 1018 note 2} Where two or more forms of a verb are given the first is the perfect. Old Bodish present forms are indicated by a hyphen, as bgod-.

^{page 1018 note 3} Imperative cited because the final vowel of the root becomes a semi-vowel before the infinitive ending -a. The -s; of the imperative is not a part of the root.

^{page 1019 note 1} “Vocalism” (see p. 1018, n. 1), 17: 11.

^{page 1019 note 2} The final example in the table seems to contradict the statement that b- prefix is not found in Sbalti verbs unless the Old Bodish present as well as the past of the verb had b-, but is really an “exception that proves the rule”. Sbalti p‘lsol-ba “supplicate, plead, implore, appeal” was compared with O.B. btśol (pf.), ātś‘ol (pres.) “to entrust (with), commend (to)”, as no other Old Bodish verb was found with which it compared more closely. As the semantic difference was great, the entry was made tentatively and with considerable misgiving. When the phonetic tables for Sbalti were completed, this comparison stood out so obviously as not belonging in the table that the semantic grounds for scepticism were confirmed phonetically. The Sbalti and Old Bodish forms are therefore two different verbs. The comparison was retained in the table, nevertheless, to illustrate the value of Sbalti as a phonetic check on comparisons between modern Bodish dialect words and Old Bodish forms.

Additional evidence that b- prefix is not preserved in Sbalti verbs unless occurring in the present tense of the corresponding Old Bodish verb could be cited from Sbalti verbs corresponding to O.B. r- (pres.), br- (pf.) verbs. F. W. Thomas, op. cit., p. 196, suggested, because of the rarity of 6- prefix in “Nam”, that the b- preterite of that language was borrowed from Old Bodish.

^{page 1020 note 1} The evidence is not clear for Chinese.

^{page 1020 note 2} “Tibetische Studien”, Bull, hist.-phil. Acad. de St. Petersbourg 8 (1865), 298 ff.

^{page 1020 note 3} Op. cit. (p. 703, n. 1), p. 19.

^{page 1020 note 4} See my “Prefixed n-, ng-, in Tibetan”, Sino-Tibetica, No. 1 (Berkeley, 1936).

^{page 1021 note 1} Archæological work in Tibet has hardly been begun, yet F. W. Thomas has recently made known a number of old languages from Tibet or border regions. More intensive work will no doubt reveal others.

^{page 1021 note 2} According to the laws of prefix sandhi in Old Bodish, which prevented a prefixed consonant of the same varga as the initial, we can understand why g- could not occur before guttural initials; but why not before labials? I can only suggest that at the time these prefixes were formed the labials were labiovelar in Old Bodish. If so, it was probably only local, as I have found no evidence of labiovelars elsewhere in Sino-Tibetan.

^{page 1022 note 1} Op. cit. (p. 1020, n. 2), pp. 268–9.

^{page 1022 note 2} From the writer's manuscript material. However, most of these may be found in the Linguistic Survey of India or the grammars of these languages. Kukish also has an imperative singular variously recorded as -ō, -o;, -ó, -à’. Tsangla (Bodish) occasionally adds -o to the usual imperative ending -i, resulting in -yo.

^{page 1022 note 3} Schiefner, op. cit., considered the “in den Prasensformen vorkommenden E- und O-Laut als Schwachung und Tr¨bung des A-Lauts” of the perfect, and cited O.B. -an> -en in some of the modern dialects, -al >-el, -ad and -as > -e; written Burmese -ak, -an > spoken Burmese -et, -en; and added citations of the interchange of -e- and -o- with -a- in Old Bodish. While realizing that most of the above changes occurred in modern times, he argued that similar changes occurred in the prehistoric period. This view of the recurrence of the same phonemic phenomena in the same languages, which Co nrady accepted, might be called a cyclic theory of phonetic change Hans Nordewin von Koerber further developed this type of theory of phonetic change in his Morphology of the Tibetan Language (Los Angeles, San Francisco, ca. 1935). Yet even by this theory the difficulty with Schiefner's evidence is that the modern changes were due to certain finals, while none of the existing finals, prefixes, initials, or suffixes could have caused all the morphophonic changes to be found in the present compared with the perfect.

^{page 1023 note 1} For the equations for Burmese vowels, see “The Vocalism of Sino-Tibetan”, JA0S 60 (1940), 302–337; 61 (1941), 18–31.

^{page 1024 note 1} See p. 1018 for West Bodish examples. One might infer that the stop prefixes of the perfect (and sometimes future) b- and of the future (g-, d-) had reduced primitive -e- or -o- to the neutral vowel -a-. This appears at first to have been the case in West Himalayish, where corresponding to Old Bodish mt‘on “see”, we have Bunan t‘ann, Thebor taṅ “look”, tan “see”, Kanauri, Mantśati group taṅ. We might assume NNW Him. *bt‘aṅ, Kanauri and Mantśati group *btaṅ as the perfect from which these forms arose, aspiration being preserved after *b- in NNW Himalayish but not in the Kanauri and Mantśati groups. But corresponding to O.B. t'os “hear”, we find Kanauri and Mantśati t'as-, showing that the supposed prefix had nothing to do with the vowel change.

^{page 1024 note 2} -s in the -e- present of verbs of Type Sa was noted by Stuart N(orris) Wolfenden, Outlines of Tibeto-Burman Linguistic Morphology … (London, 1929), pp. 67–8. Of the two -o- present verbs of Type 8a one has -a, the other does not have -s with any form—one indication that these two -o- present verbs were formerly not of this type and have been only partially assimilated to it.

^{page 1026 note 1} Types are the same as in Section A.

^{page 1027 note 1} Balti Grammar (London, 1934), 53, 55.

^{page 1027 note 2} Linguistic Studies from the Himalayas (London, 1920). For Sino-Tibetan -t: -d, see JAOS 60 (1940), top p. 311. I use surds in reconstruction of finals, as *-et.

^{page 1027 note 3} Through a typographical error given as future, the future being actually omitted.

^{page 1028 note 1} Mark Aurel Stein, Ancient Khotan (Oxford, 1907), App. B, “Tibetan Manuscripts and Sgraffiti Discovered by Dr. M. A. Stein at Endere”, ed. by L. D. Barnett and A. H. Francke: I. “Fragments of the Śalistamba-Sūtra”, fol. 3 (Mdo xvi, ff. 192 a-b), line 2, ādzind-to; Laufer, Berthold, “Bird Divination among the Tibetans”, TP 15 (1914)Google Scholar, 61, ādzind-la, -na (from Si-tui sum-rtags, ed. by Gtsug-lag ts‘os-kyi snan-ba of Si-tu, Khams, in 1743, and reprinted by Bengal Secretariat Press in 1895 (the original work, in which d-drag was carefully noted, is not available to me); F. W. Thomas, “Tibetan Documents concerning Chinese Turkestan: IV. The Khotan Region”, JRAS (1930), p. 52, ādzind- (in name of regiment).

^{page 1028 note 2} Thomas, op. cit., p. 257, adrend.

^{page 1028 note 3} Laufer, loc. cit., āp‘end byed.

^{page 1028 note 4} Stein, op. cit., fol. 2 (Mdo xvi, ff. 191 b–192 a), line 1, lend-pa; Laufer, loc. cit., lend.

^{page 1029 note 1} From the above account of -d drag, where it is only suffixed to transitive present forms, it should not be inferred that this was the sole function of -d drag in Archaic Bodish. Enough examples have been collected by the writer to indicate that sometimes it was either part of the root or could be added to any stem. The interpretation of the texts is often too obscure at the present time to determine the function of -d drag in such instances. But one may suspect -d drag is the residue of various suffixes which have been reduced to -d. Schiefner and P. Cordier had begun the study of “bisuffixes” in Old Bodish and recently Walter Simon has considered the subject more extensively in “Certain Tibetan Suffixes and Their Combinations”, HJAS 5 (1941), 372–391, and “Tibetan daṅ, ciṅ, kyin, yin, and ham”, BSOS 10 (1942),.954–975, showing that some postpositions were the result of joining a postposition to a postposition. The Sbalti present tense suffix -ed is an instance of this type, as -e is the genitive ending of the verb, which is something in the nature of a noun, expressing, as Read defined it, “not so much an action as merely a coming to pass.”This genitive termination replaces the final root vowel a of an infinitive, as gwa “to go”, gwe “of going”; bya “to do”, be “of doing”(Read 38). To this genitive ending additional inflections have been added. One of these is -d to form the present tense ending -e-d. Another is the addition of -n to form the present participle, as rb-e-n “writing”(inf. rby-a, pf. rbi-s). Now -n is an inflection which many modern Bodish dialects, and even the later classical literature, have added to adjectival roots as an adjectival termination, and the use of the present participle is similar to that of an adjective. The adjective, excepting numerals and adjectives of quantity, precedes the noun (Read 20). Recalling the statement above about the nominal nature of verbs, the adjectival (i.e. adverbial) nature of the present participle will be clear from the following examples in which the present participle precedes the verb just as an adjective precedes the noun:— k'o hrts-e-n oṅ-e-d he dancing comes. “He comes dancing.” mo hlu teṅ-e-n soṅs she song giving went. “She went singing.” de ri la ri-dak ofi-e-n yod that mountain on ibex coming are. “Ibex are (continually) coming on that mountain.” •ṅa si rb-e-n yod I by writing am. “I am writing (continually).” This construction is duplicated in nearly every respect in Burig. The Bodish suffix *-et is not a remnant of the copula yod as Berthold Laufer suggested (TP, ser. 2, 15 [1914], 63). He was “inclined to think that byed (phonetically byod or iod) has arisen from a contraction of bya + yod, lit. ‘he is doing’”. As O.B. -od > Gtsang, Dbus -ó', it was not unreasonable to assume—in want of precise phonetic data—that in some Bodish dialect -od or yod might have become umlauted to -od, then unrounded to -ed, -et. But this last step did not occur in West Bodish where we have the greatest extension of this present tense suffix. Nor is Laufer's statement that O.B. byed is phonetically byōd or bōd true for any dialect on which we have published material and two others on which I have examined manuscript material. Nor does my examination of the manuscript material left by Laufer reveal any private collection of modern Bodish linguistic data or that Laufer knew much about the phonetics of present-day Bodish, and I am “inclined to think” that he was performing a little wishful guessing about the pronunciation of byed. Bacot, op. cit., pp. 53–4, didactically stated that d-drag was a suffix of the past tense, citing bstand as the “passé réel”. He does not mention the archaic present forms, stond, stond-pa, also given by Laufer, loc. cit., which seem to run counter to Bacot's theory. Since d-drag occurs not only with present and past forms, both transitive and intransitive forms of the same root, but also with nouns and conjunctions, it seems probable that d-drag is sometimes part of the root, but sometimes a suffix —as in the present tense of the verbs noted above. Conrady, op. cit., p. 45, and Laufer, TP, 15 (1941), 62 fif., 63 n., considered -d as a causative termination in Old Bodish. The use of *-et in the transitive present of Old Bodish probably should not be compared with the causative -at of Garo, but the latter with the O.B. -d causative in such verbs as skyed, bskyed “generate” (in comparison with skye, skyes “be born”) where the root vowel of the verb is never affected and where the suffix has nothing to do with tense. Peter A. Boodberg discussed -t in the transitive formation of Chinese in an article known to me only by the title: “The Morphology of Final -n and -t”, Notes on Chinese Morphology and Syntax 3 (Berkeley, 1934).

^{page 1030 note 1} The cerebral ṣ may be a typographical error.