________________
HERMAN TIEKEN
THE DISTRIBUTION OF THE VARIANT ENDINGS -U, -AÜ AND -Ā IN THE APABHRAMŚA VERSES IN HEMACANDRA'S
PRĀKRIT GRAMMAR
In his edition of Puspadanta's Apabhramsa text Harivarśapurāna (1936, pp. 163–164), Alsdorf draws attention to the occurrence of extended and long endings alongside short ones in the inflection of the masc. and the ntr.: e.g. -aü or -aü beside -u (nom.-acc.), -ae beside -e (loc.), and -ā beside -a (nom.-acc. plur.). According to Alsdorf, the occurrence of the three long and extended endings mentioned just now was conditioned by the rhythm of the word: they all involve short endings (the loc. -e is a short vowel) and appear mainly after heavy syllables.2
With reference to the ending -ā of the nom.-acc. plur. Alsdorf (p. 154) mentions a category of words which are found only with long or extended endings (possessive pronouns: merā, amhārā, tumhārā, infinitives: mārevā, postpositions: -kerā, and the word -gārā) and words which are found often with these endings. The latter category consists of past participles and adjectives. In this connection Alsdorf again draws on the rhythmic explanation since all occurrences of the ending -a concern positions after a heavy syllable.?
In the case of the fem. ending -ī (beside -i) the situation is similar: of the 161 instances of the nom./acc. in -ī, 109 are past participles, 32 adjectives and 9 present participles. The remaining 11 instances are "ordinary" nouns. In 157 out of these 161 instances the long -7 is found after a heavy syllable (p. 165).
In his Apabhrarnsa-Studien (1937, pp. 7-14) Alsdorf discusses these variant endings once more in connection with their occurrence in the Apabhramsa verses in Hemacandra's Prākrit grammar (IV 329ff.). On the basis of this new material yet another long ending could be added to the set mentioned above, namely -a for the nom.-acc. sing. of the masc. and the ntr. Alsdorf was mainly concerned with their "derivation”. He argues that the endings represent different stages in the development of Middle Indo-Aryan to New Indo-Aryan. Thus, he considers -u to be the regular form in Apabhramsa, showing shorthening of final -o, and
Indo-Iranian Journal 41: 1–31, 1998. © 1998 Kluwer Academic Publishers. Printed in the Netherlands.