Page:Vedic Grammar.djvu/114

I. ALLGEMEINES UND SPRACHE. 4. VEDIC GRAMMAR. pūrva-vát, pratna-vát ‘as of old', which are accusative neuters of the adjective suffix -vant (unaccented); and some adverbs from the comparative in tara, ut-tarám (AV.) ‘higher', but it-tara- as adjective; similarly ava-tarám, paras- tarám and parā-tarám ‘farther away', vi-tarám ‘more widely'² sam-tarám (AV.) 'still farther'. Similarly the instrumental dívà 'by day', but divá 'through heaven'; and the dative aparaya 'for the future', but áparaya 'to the later'; and the ablatives apakat 'from afar' (ápāka- ‘far'), amắt ‘from near' (áma- AV. 'this'), sanát 'from of old' (sána- 'old')³. 104 6. Accent in Sandhi. 108. 1. When two vowels combine so as to form a long vowel or diphthong, the latter has the Udatta if either or both of the original vowels had it; e. g. nudasvátha for nudasva átha; távét for táva it; kvét for kvà íts; ágat for á agat; pitéva for pitá iva, śató for satá u; nántaras (VI. 63²) for ná ántaras. The contraction of i+ i is, however, accented 7,6 the (dependent) Svarita having here (in í ) ousted the preceding Udatta; e. g. diviva (RV. AV.) for divi va. This is the praślista, 'contracted', Svarita of the Prātiśākhyas 8. 2. When i and with Udatta are changend to y and 7, a following unaccented vowel receives the Svarita; e. g. vy ànat for vi araṭ. The uncontracted form with Udatta must, however, almost invariably be read in the RV. This is the kṣaipra Svarita of the Prātiśākhyas. Here the enclitic Svarita assumes the appearance of an independent accent. 3. When á is elided, it throws back its Udatta on unaccented e or o; e. g. sūnávé 'gne for sūnáve ágne; vó vasaḥ for vo ávasaḥ. But when un- accented a is elided, it changes a preceding Udātta to Svarita; e. g. sò 'dhamáḥ for só adhamáh. This is the abhinihita Svarita of the Prātiśākhyas. Here also the enclitic Svarita (in ó à) has ousted the preceding Udātta (as in diviva) ¹º. 7. The Sentence Accent. HASKELL, Vocative-Accent in the Veda, JAOS. 11, 57. BÖHTLINGK, ein erster Versuch über den Accent im Sanskrit (Mémoires de l'Académie imp. de St. Péters- bourg 1843) p. 38 ff. - WHITNEY, JAOS. 5, 193 ff., 387 ff. AUREL MAYR, Beiträge aus dem Rigveda zur Accentuirung des verbum finitum (Sitzungsberichte der phil.-hist. Classe der kaiserlichen Akademie der Wissenschaften, Wien 1871, p. 219 ff.). WEBER, IS. 13, 70ff. --- BÖHTLINGK, Sanskrit-Chrestomathie² p. 356. — WHITNEY, Sanskrit Grammar 591 ff. DELBRÜCK, Altindische Syntax (Halle 1888) 21-29. OLDENBERG, Die Verbalenklisis im Rgveda, ZDMG. 60, 707-740. - ¹ Cp. WHITNEY 1107 a. 2 See op. cit. 1119. 3 Cp. op. cit. 1114 a. 4 Cp. BENFEY, Vollständige Grammatik 64. 5 But when a Svarita is followed by an unaccented syllable, it of course remains; e. g. kvèyatha for kvà iyatha (VIII. 17). 6 Except in the Taittiriya texts which follow the general rule (diviva). 7 This also takes place in sudgātá (TS.) for sú-udgātá a good Udgatr' (TS. VII. 1. 81: B.). 8 Cp. HAUG 75. — 109. The vocative.-a. The vocative, which whether it consist of a single word or a compound expression, is invariably accented on the first syllable, retains its accent only at the beginning of a sentence or Pada ¹¹; | 9 So called because 'uttered with a quick' (ksipra-) pronunciation, the semivowel replacing the vowel. 10 Cp. WACKERNAGEL I, 251, ba; BRUG- MANN, KG. 45, 2. 1¹ This applies to the second as well as the first Pâda of a hemistich (as some of the above examples show), thus indicating the independent character of these Pâdas, which is obscured by the way in which the re- dactors of the Samhitas apply the rules of Sandhi and mark the dependent Svarita.