________________
FEBRUARY, 1932)
ALPHABETS AND PRONUNCIATION
[ $$ 128-130
As o is in Ks. always the epenthetic result of a followed by t-mätrā, it follows that many words are written sometimes , and sometimes , and in some cases the form is the only one used. Thus, one of the terminations of the genitive is ok, but it is invariably written uk", and will be so written in the following pages. The reader must also be prepared to meet other instances of exchange in the case of other vowels in the foregoing list of pairs.
128. It is probable that in Dardio languages the distinction between dental and cerebral mutes is not so sharp as in India proper. In Kašmiri the two sets of mutes are distinguished in writing, but in poetry th is allowed to rhyme with th, a state of affairs which would be impossible in the Ganges Valley. According to Lorimer, in şina, which is not a written language, the sound of d is rare, and only two possible instances have been known in which it is found in words cognate to Sanskrit. In the same language, he maintains that the only independent t-sound in existence is more like the English alveolar t, i.e., neither cerebral nor dental. This is disputed by G. Bailey, who says that both t and are heard in the language. Also, none of the Dardio languages, including Kašmiri, have (except in sporadio instances) any sonant aspirate consonants, gh, jh, dh, dh, or bh. In words in which in other languages such sonant aspirates occur, the corresponding unaspirated letters are substituted, except in the case of jh, which is represented by 2. Thus, H. ghorā, Ks. gur", a horse ; Pr. bujihai, Kx. bözi, he will hear; H. buddha, Ks. bud", old; H. buddhi, Kg. böd, wisdom; H. bhai, Kš. bāya, a brother.
E... Yindarzuth rhymes with oth in Rāmavatāra-carida, 699.
? I have noted one or two words in Lorimer's liste which can be compared with others in Indian languages. Lorimer himself compares bado, great, with H. bard; and also the 8. root bud., dive, may be referred to the Skr. vrud-, sink, a word, moreover, which, according to the PWB., has been noted as need only in the Ks. Rajatarangini. Most of the words containing 4 are also found in the contiguous non-Aryan Burušaski.
3 See the discussion in JRAS., 1924, 33 fl., 190 ff. • So also in the so-called Tocharisch of Central Asia. Soo Grierson in JA., X, xix, 340.
129. On the other hand, no KĀtmiri word can end in an unaspirated surd. When such a surd comes at the end of a word it must be aspirated. Thus : Base. Nom. Sing.
but Dat. Sing. trak., a certain measure. treekh ($ 105)
tralas. kāls-, glass.
kātsh
kātsas, loaf-, a ram.
kaeth ($ 105)
katas. rat-, blood.
ræth ($ 105)
ratas. tāp-, sunshine.
tāph
tāpas. This aspiration is frequently omitted by careless writers in the Sāradă or Nagari character, and is never indicated when writing or printing in the Persian character.
130. One other fact that frequently obscures the derivation of Dardic words may here be mentioned. This is that the vowels e, e, i, i, o, u, and i cannot commenoe any word or follow another. wel, although in writing they are often shown as initials. In such cases & semi-vowel is inv. iably prefixed in speaking, in order to aid the pronunciation. To e, e, i, and i is prefixed y, and to o, u, and is prefixed w. Thus, yelat- (Ar. illat), a defect; yer,
63