TRÜBNER’S COLLECTION OF SIMPLIFIED GRAMMARS OF THE PRINCIPAL ASIATIC AND EUROPEAN…

EDITED BY REINHOLD ROST, LL.D., Ph.D.

I. HINDUSTANI, PERSIAN, AND ARABIC. By the late E. H. Palmer, M.A. Price 5s.

II. HUNGARIAN. By I. Singer. Price 4s. 6d.

III. BASQUE. By W. Van Eys. Price 3s. 6d.

IV. MALAGASY. By G. W. Parker. Price 5s.

V. MODERN GREEK. By E. M. Geldart, M.A. Price 2s. 6d.

VI. ROUMANIAN. By R. Torceanu.

VII. TIBETAN. By H. A. Jäschke.

Grammars of the following are in preparation:—

Albanese, Anglo-Saxon, Assyrian, Bohemian, Bulgarian, Burmese, Chinese, Cymric and Gaelic, Danish, Finnish, Hebrew, Malay, Pali, Polish, Russian, Sanskrit, Serbian, Siamese, Singhalese, Swedish, Turkish.

London: TRÜBNER & CO., Ludgate Hill.

[V]

Preface.

The present new edition of Mr. Jäschke’s Tibetan Grammar scarcely needs a word of apology. As the first edition which was lithographed at Kyelaṅ in 1865 in a limited number of copies has long been out of print, Dr. Rost urged the author to revise his grammar for the purpose of bringing it out in an improved form. The latter, prevented by ill-health from undertaking the task, placed the matter in my hands, and had the goodness to make over to me his own manuscript notes and additions to the original work. Without his personal cooperation, however, I was unable to make any but a very sparing use of these, adding only a few remarks from Gyalrabs and Milaraspa, with some further remarks on the local vernacular of Western Tibet. Indeed, special attention has been paid throughout to this dialect; it is the one with which the author during his long residence at Kyelaṅ had become most familiar, and with which the English in India are most likely to be brought into direct contact.

Besides the above mentioned additions, I have taken a number of examples from the Dzaṅlun, to make clearer some of the rules, and, with the same view, I have altered, here and there, the wording of the lithographed edition. [VI]The order of the paragraphs has been retained throughout, and only one (23.) has been added for completeness’ sake.

The system of transliteration is nearly the same as in the Dictionary, only for ny, ñ is used, and instead of e̱, ä (respectively ā̤) has been thought to be a clearer representation of the sound intended. For the niceties of pronunciation the reader is referred to the Dictionary, as in this Grammar only the general rules have been given.

Finally I must express my warmest thanks to Dr. Rost, to whose exertions not only the printing of this Grammar is solely due, but who also rendered me much help in the correcting of the work.

Mayence, May 1883.

H. Wenzel.

Abbreviations.

[VII]

Errata.

The corrections listed below have been applied to the text.

[1]

Phonology.

1. The Alphabet. The Tibetan Alphabet was adapted from the Lañc̀ʽa (ལཱཉ་ཚ) form of the Indian letters by Tʽon-mi-sam-bho-ta (ཐོན་མི་སམ་བྷོ་ཏ) minister of king Sroṅ-tsan-gam-po (སྲོང་བཙན་སྒམ་པོ་) about the year 632 (s. Köpp. II, 56). The Indian letters out of which the single Tibetan characters were formed are given in the following table in their Nāgari shape.

[2]

It is seen from this table that several signs have been added to express sounds that are unknown in Sanscrit. The sibilants ཙ་ ཚ་ ཛ་ evidently were differentiated from the palatals. But as in transcribing Sanscrit words the Tibetans substitute their sibilants for the palatals of the original (as ཙི་ན་ for चीन), we must suppose that the sibilisation of those consonants, common at present among the Hindus on the Southern slopes of the Himālaya (who say tsār for चार, four etc.), was in general use with those Indians from whom the Tib. Alphabet was taken (cf. also the Afghan څ‎ and ڂ‎ likewise sprung from چ‎ and ج‎). ཝ་ is differentiated from བ་, which itself often is pronounced v, as shewn in the sequel; in transcribing Sanscrit, ब and व both are given, generally, by བ only. ཞ་ seems to be formed out of ཤ་ to which it is related in sound. ཟ་ evidently is only the inverted ཇ་. ཨ་ corresponds with Sanscrit अ. འ་ is newly invented; for its functions see the following §§.—The letters which are

peculiar to Sanscrit are expressed, in transcribing, in the following manner. a) The linguals, simply by inverting the signs of the dentals: thus, ཊ་ ट, ཋ་ ठ, ཌ་ ड, ཎ་ ण. b) The sonant aspirates, by putting ཧ་ under the sonants: thus, གྷ་ घ, ཛྷ་ झ, ཊྷ་ ढ, དྷ་ ध, བྷ་ भ.1 [3]

2. Remarks. 1. Regarding the pronunciation of the single letters, as given above, it is to be born

in mind, that surds ཀ་ ཏ་ པ་ are uttered without the least admixture of an aspiration, viz. as k, t, p are pronounced in the words skate, stale, spear; the aspirates ཁ་ ཐ་ ཕ་ forcibly, rather harder than the same in Kate, tale, peer; the sonants ག་ ད་ བ་ like g, d, b in gate, dale, beer. 2. The same difference of hardness is to be observed in ཅ་ ཆ་ ཇ་ or c̀, c̀ʽ, j̀ (c̀ʽ occurs in church; c̀, the same without aspiration; j̀ in judge) and in ཙ་ ཚ་ ཛ་ or ts, tʽs, ds. 3. ཞ་ is the soft modification of s̀ or the s in leisure (French j in jamais, but more palatal). 4. ང་ is the English ng in sing, but occurs in Tibetan often at the commencement of a syllable. 5. ཉ་ ñ is the Hindi न्य, or the initial sound in the word new, which would be spelled ཉུ་ ñu. 6. In the dialects of Eastern or Chinese-Tibet, however, the soft consonants ག་ ད་ བ་ ཇ་ ཛ་, when occurring as initials, are pronounced with an aspiration, similar to the Hindi

घ, ध, भ, झ, or indeed so that they often scarcely differ from the common English k, t, p, ch; also ཞ་ and ཟ་ are more difficult to distinguish from ཤ་ and ས་ than in the Western provinces (Exceptions s. §§ 7. 8).

3. Vowels. 1. Since every consonant sign implies, like its Sanscrit prototype, a following a, unless some other vowel sign is attached to it, no particular sign is wanted to denote this vowel, except in some cases specified in the [4]following §§. The special vowel signs are ེ, ི, ོ, ུ, pronounced respectively as e, i, o, u are in German, Italian and most other European languages, viz. ེ like ay in say, or e in ten; ི like i in machine, tin; ོ like o in so, on; ུ like u in rule, pull. It ought to be specially remarked that all vowels, including e and o (unlike the Sanscrit vowels from which they have taken their signs) are short, since no long vowels at all occur in the Tibetan language, except under particular circumstances, mentioned below (s. § 9. 5, 6 ). 2. When vowels are initial, ཨ is used as their base, as is ا‎ in Urdu, e.g. ཨ་མ་ ama, ‘mother’. 3. འ is originally different from ཨ་, as the latter denotes the opening of the previously closed throat for pronouncing

a vowel with that slight explosive sound which the Arabs mean by أ‎ (همزة‎), as the a in the words: the lily, an endogen, which would be in Tibetan characters ལི་ལི་ཨན་; འ་ on the contrary is the mere vowel without that audible opening of the throat (as Arabic ا‎ without ء‎), as in Lilian, ལི་ལི་འན་. In Eastern Tibet this difference is strictly observed; and if the vowel is o or u the intentional exertion for avoiding the sound of ཨ་ makes it resemble wo and wu: འོ་མ་ ‘the milk’, almost like wo-ma, འུག་པ་ ‘the owl’ = wug-pa. In western Tibet this has been obliterated, and འ་ is there spoken just like ཨ་.

4. Syllables. The Tibetan language is monosyllabic, that is to say all its words consist of one syllable only, which indeed may be variously composed, though the [5]component parts cannot, in every case, be recognised in their individuality. The mark for the end of such a syllable is a dot, called ཚེག་ tʽseg, put at the right side of the upper part of the closing letter, such as ཀ་ the syllable ka. This tʽseg must invariably be put at the end of each written syllable, except before a s̀ad (§ 10), in which case only ང་ ṅa retains its tʽseg. If therefore such a dot is found after two or more consonants, this will indicate that all of them, some way or other, form one syllable with only one vowel in it: ཀ་ར་ ka-ra, ཀར་ kar (cf. §§ 5. 8 ).

5. Final consonants. 1. Only the following ten: ག་ ང་ ཏ་ ན་ བ་ མ་ འ་ ར་ ལ་ ས་ (and the four with affixed ས, v. 5) occur at the end of a syllable. 2. It must be observed, that ག་ ད་ བ་ as finals are never pronounced like the English g, d, b in leg, bad, cab, but are transformed differently in the different provinces. In Ladak they sound

like k, t, p e.g. སོག་ = sock, གོད་ = got, ཐོབ་ = top. 3. In all Central Tibet, moreover, final ད་ and ན་, sometimes even ལ་, modify the sound of a preceding vowel: a to ä (similar to the English a in hare, man), o into o̤ (French eu in jeu), u into ṳ (French u in mur). In most of the other provinces ག་ and ད་ are uttered so indistinctly as to be scarcely audible, so that སོག་, གོད་ become sŏʼ, gŏʼ. In Tsang even final ལ་ is scarcely perceptible, and final ག་, particularly after o, is almost dissolved into a vowel sound = a: སོལ་བ་ so-wa, [6]དཀོན་མཆོག་ kon-choa.2 4. Final ས་ is sounded as s only in Northern Ladak; elsewhere it changes into i or disappears entirely, prolonging, or even modifying at the same time the preceding vowel. Thus

the following words: ནས་ ‘barley’, ཤེས་ ‘know’, རིས་ ‘figure’, ཆོས་ ‘religion’, ལུས་ ‘body’, are pronounced in Northern Ladak: năs, s̀ĕs, ris, c̀ʽos, lŭs; in Lahoul: nai, shei, rī, c̀ʽō, lū; in Lhasa, and consequently by everyone who wishes to speak elegantly: nā̤, s̀ē, rī, c̀ʽō̤, lṳ̄. 5. In some words final ས་ occurs as a second closing letter (affix), after ག་ ང་ བ་ མ་, as in ནགས་ ‘forest’, གངས་ ‘glacier-ice’, ཐབས་ ‘means’, རམས་ ‘indigo’; these are pronounced in N. Ladak: nacks, gaṅs, tʽaps, rams, elsewhere nack (in Ü: nā), gaṅ (ET ghang), tʽap, ram. 6. ན་ before པ་ and མ་ is especially in ET very often pronounced m, e.g. ཉན་པ་ ñäm-pa, ཉོན་པ་ ñöm-pa, སྙེན་པ་ ñem-pa.

6. Diphthongs. 1. They occur in Tibetan writing only where one of the vowels i, o, u have to be added to a word ending with an other vowel (s. §§ 15. 1; 33. 1; 45. 2 ). These additional vowels are then always written འི་, འོ་, འུ་, never ཨི་ etc. (cf. § 3. 3); and the combinations ai, oi, ui (as in བཀའི་, མགོའི་, བུའི་) are pronounced very much like ā̤, ō̤, ṳ̄, so that the syllables ནའི་, ཤེའི་, རིའི་, ཆོའི་, [7]ལུའི་ can only in some vulgar dialects be distinguished from those mentioned in § 5. 4. 2. The others ao, eo, io, oo, uo, au, eu, iu (བཀའོ་, སྐྱེའོ་, བགྱིའོ་, འགྲོའོ་, འདུའོ་, གའུ་, བྱེའུ་, ཁྱིའུ་) are pronounced in rapid succession, but each vowel is distinctly audible. In prosody they are generally regarded as one syllable, but if the verse should require it they may be counted as two.

7. Compound consonants. 1. They are expressed in writing by putting one below the other, in which case several change their original figure.

Subscribed consonants. 2. The letter y subjoined to another is represented by the figure ྱ, and occurs in connection with the three gutturals and labials, and with m, thus ཀྱ་ ཁྱ་ གྱ་ པྱ་ ཕྱ་ བྱ་ མྱ་. The former three have preserved, in most cases, their original pronunciation kya, kʽya, gya (the latter in ET: ghya s. § 2. 6). In the Mongol pronunciation of Tibetan words, however, they have been corrupted

into c̀, c̀ʽ, j̀ respectively, a well known instance of which is the common pronunciation Kanj̀ur i.o. kangyur, or eleg. ka-gyur (བཀའ་འགྱུར་). པྱ་, ཕྱ་, བྱ་ are almost everywhere spoken without any difference from ཅ, ཆ, ཇ (except in the Western dialect before e and i, where the y is dropped and པ, ཕ, བ alone are pronounced). མྱ is spoken ny = ཉ. 3. r occurs at the foot of the gutturals, dentals, labials, of ན, མ, ས, and ཧ, in the shape of ྲ. In some parts of the country, as in Purig, these combinations [8]are pronounced literally, like kra, khra etc., but by far the most general custom is to sound them like the Indian cerebrals,

viz. ཀྲ, ཏྲ, པྲ indiscriminately = ट ṭ; ཁྲ, ཐྲ, ཕྲ = ठ ṭh; གྲ, དྲ, བྲ = ड ḍ (in CT: ḍh); only in the case of བྲ the literal pronunciation br is not uncommon. In ནྲ and མྲ both letters are distinctly heard; ཧྲ sounds like shr in shrub, and so does སྲ generally. In Ü this r is dropped nearly in all cases: thus, ཕྲ pʽa, སྲ sa etc. 4. Six letters are often found with an ལ beneath: ཀླ་ གླ་ བླ་ ཟླ་ རླ་ སླ་; in these the ལ alone is pronounced, except in ཟླ་, which sounds da. 5. The figure ྭ, sometimes found at the foot of a letter is used in Sanscrit words to express the subscribed व, as in སྭཱ་ཧཱ་ (cf. § 9. 6 ) for स्वाहा; and is now pronounced by Tibetans = ō: sōhā; in words originally Tibetan it now exists merely as an orthographical mark, to distinguish

homonyms in writing, as ཚ་ tʽsa, ‘hot’ and ཚྭ་ tʽsa, ‘salt’; but, as it is spoken, in some words at least, in Balti (e.g. རྩྭ་ rtswa ‘grass’), it must be supposed that, in the primitive form of the language, it was generally heard.—Note. Of such compounds, indeed, as ཕྱྭ་ ‘lot’ it is difficult to understand, how they can have been pronounced literally, if the v was not, perhaps, pronounced before the y.

Superadded consonants. 6. r over another consonant is written ⸆, and 11 consonants have this sign: རྐ་ རྒ་ རྔ་ རྟ་ རྡ་ རྣ་ རྦ་ རྨ་ རྩ་ རྫ་, above ཉ་ it preserves [9]its full shape, as better adapted to the form of that letter: thus, རྙ་. In speaking it is seldom heard except provincially, and in some instances in compound words after a vowel thus, ཨུ་རྒྱན་ Urgyán, Urgyén, ancient name of the country of Lahore; རྡོ་རྗེ་ dórje ‘vaj̀ra’. Ladakees often pronounce it = s: རྟ་ sta ‘horse’ elsewhere ta. 7. Similar is the usage in those with a superadded ལ (namely: the surds and sonants of the first four classes, the guttural nasal, and ཧ), which latter is often softly heard in WT, but entirely dropped elsewhere, except in the case of ལྷ, which is spoken = ལ in WT, but with a distinct aspiration = hla or lha in ET. 8. ས is superadded to the gutturals, dentals and labials with exception of the aspiratae,

then ཉ་ and ཙ་. It is, in many cases, distinctly pronounced in Ladak, but dropped elsewhere3. 9. ག་ ད་ བ་ ཇ་ ཛ་ with any superadded letter lose the aspiration mentioned in § 2. 6 and sound = g, d, b, j̀, ds. 10. རྗ་ རྩ་ རྫ་ often lose even the inherent t-sound in pronunciation and are spoken like j̀, s, z.

1 A very clear exposition of the ramification of Indian alphabets by Dr. Haas is to be found in the Publications of the Palaeographical Society Oriental Series IV, pl. XLIV. ↑

2 This is the form in which the word, chosen by the missionaries to express the Christian “God” (cf. dict.), has found its way into several popular works. ↑

3 This will be indicated in the following examples by including the s in parentheses, as (s)kom. ↑

Examples.

[11]

8. Prefixed letters. 1. The five letters ག་ ད་ བ་ མ་ འ་ frequently occur before the real, radical initials of other words, but are seldom pronounced, except in similar cases as § 7. 6. ག་ occurs before ཅ་ ཉ་ ཏ་ ད་ ན་ ཙ་ ཞ་ ཟ་ ཡ་ ཤ་ ས་; ད before the gutturals and labials with exception of the aspiratae; བ་ before ཀ་ ག་, the palatals, dentals and palatal sibilants with the same exception as under ད, then ཞ་ ཟ་ ར་ ཤ་ ས་; མ་ before the gutturals, palatals, dentals and palatal sibilants, except the surds; འ before the aspiratae and sonants of the five classes. In CT, to pronounce them in any case, is considered vulgar. 2. The ambiguity which would arise in case of the prefix standing before one of the

10 final consonants, as single radical, the vowel being the unwritten a,—e.g. in the syllable དག་, which, if ད is radical, has to be pronounced dag, if prefixed gā,—is avoided by adding an འ་ in the latter case: thus, དགའ་. Other examples are: གད་ gad (gʽäʼ) and གདའ་ dā; བས་ bas (bā̤, bʽā̤) and བསའ་ sā; མད་ mad (mäʼ) and མདའ་ dā; འགའ་ gā. This འ་ is added, though the radical be not one of the mentioned letters; as, བཀའ་ kā. 3. ད་ as a prefix and བ་ as first radical annul each other, so that only the following sound is heard, as

will be seen in the [12]following examples (དབང་ etc.). 4. Another irregularity is the nasal pronunciation of the prefixed འ་ in compounds after a vowel, which is often heard e.g. དགེ་འདུན་ pronounced gen-dún, gen-dṳ́n, but eleg.: ge-dṳ́n, ‘clergy’; བཀའ་འབུམ་ kam-bum, eleg. ka-búm, ‘the 100 000 precepts’ (title of a book).—Note. With regard to the aspiration of the soft consonants in ET the prefixed letters have the same influence as the superadded ones § 7. 9.

Examples.

9. Word; Accent; Quantity. 1. The peculiarity of the Tibetan mode of writing in distinctly marking the word-syllables, but not the words (cf. § 4) composed of two or more of these, sometimes renders it doubtful what is to be regarded as one word. 2. There exist a great number of [13]small monosyllables, which serve to denote different shades of notions, grammatical relations etc., and are postponed to the word in question; but never alter its original shape, though their own initials are not seldom influenced by its final consonant (cf. § 15). 3. Such monosyllables may conveniently be regarded as terminations, forming one word together with the preceding nominal or verbal root. 4. The accent is, in such cases, most naturally given to the root, or, in compounds,

generally to the latter part of the composition, as: མིག་ mig, ‘eye’, མིག་གི་ míg-gi, ‘of the eye’; ལག་ lag, ‘hand’, ལག་ཤུབས་ lag-s̀ub(s), ‘hand-covering, glove’.—5. Equally natural is, in WT, the quantity of the vowels: accentuated vowels, when closing the syllable, are comparatively long (though never so long as in the English words bee, stay, or Hindi راجا‎ etc.), otherwise short, as མི་ mī ‘man’, མི་ལ་ mī-lă ‘to the man’, but མར་ măr, ‘butter’.—In CT, however, even accentuated and closing vowels are uttered very shortly: