Collected Articles Of LA Schwarzschild On Indo Aryan 1953 1979

Author(s): Royce Wiles
Publisher: Australian National University

Page 18

THE POSSESSIVE ADJECTIVES 09 LATE PRAKRIT 131 mahakera(ka) > mahaeraa) > erata) > meras > meri. thakera(ka) > thacrata) > tuberala) te > terac > teni.
The early contraction of the vowels we in hiatus may be put down to the Inck of accentuation of the word, while the loss of al is due to the analogy of the pronoun.
In the plural there was no doubt a similar influence of the Prakrit form of the genitive plural amhuinam, tumlinam, which was replaced by the Arabhrama period by amhala, wali, and anha, who 18 would normally go to a pointed out by Alsdorf: "Es ist auch unglaubhaft dass aus der Kontraktion von ac oder i jemals etwas anderes entstehen sollte als e." Therefore an analogical influence is also responsible for the plural forms:
ambakera(ka) > macrofa) amháng > amhiva.
Inhalera(ka) > merala) tumhānaux > kuhära. There was thus not a variation of postposition within the same dialect, but the analogy of the personal pronouns was responsible for the difference of development.
132 THE POSSESSIVE ADJECTIVES OF LATE PRAKRIT and that kura does not occur until very late. On account of this late occurrence and the fact that kurz is restricted to Bengali, Hoernle thought it was a double genitive formed very late from ku + na (Gaudian Grammar, p. 236, footnote). Though this explanation is no doubt incorrect, it is quite probable that lara is only a late deformation of Zara itself in Bengali. It is certainly almost impossible that such a late form, restricted as it is to Bengali and not even common to the other Magadhan languages, should have given rise to the whole of the possessive formation of Western Apabhram and the languages derived from it.
The reason for the adoption of this hypothetical kvira by Pischel and others seems to have been a phonetic one: waha and tha
+ kvira would easily give makar and tuhara, whereas the form vara, given first of all by Beames, presents certain phonetic difficulties in view of the question of the yarudi aka ay seems to have had a double development according to whether the y resulting from the l was pronounced or not in late Prakrit, so that the group sometimes gives and sometimes. There seems to have been a certain amount of hesitation and dialectal and chronological variation in the development of this group, which is further obscured by interborrowing and the tendency to substitute the suffix -ers for what would have been the correct ending of the word. It may be possible to explain a certain number of the inconsistencies by such a substitution of suffix rather than phonologically. Thus it appears that in Apabhrama in the case of some words, and in others in the early stages of the modern vernaculars the ending era was favoured. But it is not at all clear whether in those words akin has been replaced by akara which has developed to era via ayara, or whether there has been substitution of the suffix era <kera. Examples are for instance janera, janeri, which is found in Digambara, but not in Western Apabhrama (it is used even adjectivally once in the Bhavisattakaha), zrnera mentioned in Hei, 168, while the Kumarapalacarita hus kanniara, Maharastri Prakrit already has nalieri, cf. Hemacandra Delīnamamala ii, 10. naliara nalikera; gamereu cours in the rather late Sadbhāşacandrika of Laksmīdhara for gamina (gramavira). A particularly important example is the modern Hindi, Panjabi etc. andhera as opposed to Gujarati andhárm, which has been discussed by Helmer Smith 1
The postpositional affixes that are considered by Beames, Pischel, etc., to have brought about the formation of the W. Apabhramba waharan, Wira and kara are both inattested in Prakrit and Apabhrama. Pischel thinks that the absence of Kara in the works of the grammarians who mention only kerala) is due to the fact that a word vira existed already in Sanskrit and that there was therefore no need felt to explain the Prakrit kāru <kirys. But there is also no sign of Kāra from Marya in the texts, karā only occurs na suflix to form agent nouns, i.e. representing the Sanskrit Live and not viry. This makes it likely that the adjectival kira of possessive meaning that Pischel postulates did not exist at all It is scarcely justified phonetically the regular development of Xvirya in Prakrit is kajja; as a learned form, Axirya might possibly giev kara, without epenthesis, just as tīrya > türa and gambiry
> gawbhira, a development which is unusual with the vowela, norya > acera, etc., not acara. Pischel insists that "Nobody can doubt that dra of Bengali originally was mura and that kira and era are only modifications of the same word, vix. Mirya." But Chatterji la shown that the postposition responsible for the creation of the Bengali genitive and plural forms W era originally
R ichel, op. cit.
1 Heimer Smith, HSL, 31. p. 116.
- 16