III. Accent. General Character of Vedic Accent.
77
The Sutras, though not themselves accented, occasionally accent the Mantras ivhich they quote.
The Vedic accent, like that of ancient Greece, was of an essentially riusical nature. This is indicated by the fact that the accent exercises no influence on the rhythm of versification. The phoneticians of the Prati- E(akhyas, in describing it, speak only of its pitch, which is also indicated by the name of the chief tone, ud-atta 'raised', 'high' '. But that the Vedic accent was accompanied by some stress ^ is shown by certain phonetic changes which cannot otherwise be explained^.
In the Vedic accent three degrees of pitch may be distinguished : the high, properly represented by the udatta, the middle by the svarita ('sounded'), and the low by the an-udatta ('not raised'). That the Udatta originally- denoted the highest pitch in the oldest form of the Vedic language, is shown by the evidence of Comparative Philology, and especially of Greek which, as far as its peculiar laws of accent will admit, has the acute on the same syllable as in the Vedic language has the Udatta (e. g. A//6f : divas ; smd : sapid; oktm : astdu). This conclusion is also supported by the method of marking the Udatta in the Samaveda, the Maitrayani Sainhita, and the Kathaka. In the Rgveda, however, the Udatta has a middle pitch, lower than that of the Svarita, as is shown both by the way in which it is marked and by the account given of it in the Pratisakhyas. This must be an innovation, though an old one.
The Svarita is a falling accent of a dependent nature, marking the transition from an accented to a toneless syllable. It regularly follows an Udatta, to the rise of which its fall corresponds in pitch. It assumes an independent appearance when the preceding Udatta is lost in consequence of the vowel that bears the Udatta being changed to a semivowel in Sandhi. It is described in Panini i. 2 3't as a combination of Udatta and Anudatta, which means that it falls from the high pitch of the acute to the low pitch of unaccented syllables. According to the RV. Pratisakhya and the TS. Pratisakhya, however, the first part of the Svarita sounds higher than the Udatta. This means that, instead of falling immediately from the high pitch of the preceding Udatta, it first rises somewhat before falling to low pitch ^- It would thus have something of the nature of a circumflex in the RV.; only the rise in pitch above the highest level of the Udatta is but slight (corre- sponding to the initial rise of the Udatta from Anudattatara to Anudatta level), while the fall corresponds to the total rise of the Udatta^. The low tone of the syllables preceding an accented syllable (with Udatta or Svarita) is called an-udatta 'not raised' in the Pratisakhyas ^ When it follows a Svarita it is called the pracaya^ {svara), or 'accumulated pitch' (as several such unaccented syllables often occur in succession) which continues at the low level reached by the preceding Svarita till the syllable immediately
poem composed in the style of the Vedic hymns, is also accented, but with many mistakes; see Grube's edition in Indische Studien, vol. Xiv.
1 Cp. Haug, op. cit. 19.
2 The Vedic accent, like the Greek, was, after the beginning of our era, changed to a stress accent which, however, unlike the modern Greek stress accent, did not remain on the original- syllable, but is regulated by the quantity of the last two or three syllables, much as in Latin; cp. Haug 99, end.
3 See Wackernagel i, 218 and cp. OsT- HOFF, Morphologische Untersuchungen 4, 73.
4 In agreement with VPr. I. 126; APr. 1. 17; cp. Haug 73.
5 See Oldenberg, Prolegomena 483 f.
6 According to Panini's account the Svarita does not rise above Udatta pitch before falling; see Oldenberg, loc. cit.
7 See RPr. Ul. i; cp. Haug 91.
8 Haug 92 f.
