On Early Apbhramsa

Author(s): H C Bhayani
Publisher: H C Bhayani
Catalog link: https://jainqq.org/explore/269488/1

Page #1
On Early Apabhramsa Harivallabh C. Bhayani 1. By the seventh century A.D., Apabhramsa was established as a language of literature besides Sanskrit, Prakrit, and Paisaci (Bhamaha, Dandin). The grammatical tradition as represented by Hemacandra's treatment of Apabhramsa takes it to be one language like Sanskrit and Prakrit (Prakrit in the sense of Maharastri). The occasional observations of grammarians such as Namisadhu and Bhoja about the varieties of Apabhramsa relate to its use, with a touch of some regional dialects and occurring in some minor compositions, as a piece of novelty. This is illustrated in Namisadhu's commentary on Rudrata's Kavyalamkara (II 11-12) and in Bhoja's Srngaraprakasa (pp.102-103). From the linguistic point of view, however, the language of the earlier period of Apabhramsa (6th to 12th centuries A.D.) was standardized but not yet uniform, open as it was to the continuous impact of regional spoken dialects, besides the influence of Sanskrit and Prakrit. Grammarians have also taken note of some varying features of Apabhramsa. As the Apabhramsa works of the first two centuries are all lost and since the manuscripts of the earliest preserved works are much later than the works themselves (and, consequently, modernized considerably in orthography and grammar), we have to rely on a few scattered remnants as clues to get a glimpse of the form of the language as used in the earliest stratum of Apabhramsa literature. 2. In his treatment of Apabhramsa, Hemacandra has provided for the following: (1) change of intervocalick, kh, t, th, p, and ph respectively to g, gh, d, dh, v, and bh which were otherwise as a rule elided (non-aspirates) or changed to h (aspirates), refer to Siddhahemacandra 8-4-396; (2) optional preservation of r in clusters with r as the latter member (8-4-398); (3) change of Skt. bra- to bruv- or bro- (8-4-391). Besides, in a general way, Hemacandra has said many a time that the changes are the same (or generally the same) as in Sauraseni, which evidently relates to the features noted under 8-4-396 (refer to 8-4-260, 267, 274, 275, 422, 446). These rules have been appropriately illustrated in the commentary. BERLINER INDOLOGISCHE STUDIEN (BIS) 7.1993, pp. 1-7.

Page #2
H.C. Bhayani 3. We now find near-total absence of the first and the third of the above-mentioned features in the vast majority of the illustrations cited by Hemacandra, and in the whole of Apabhramsa literature known to us. The second feature is also found but rarely. It is nevertheless somewhat more common than the other two which seem to be quite exceptionall. In this connection we may note a few remarkable facts. In order to illustrate 8-4-446, Hemacandra has cited the following verse: sisi seharu khanu vinimmavidu khanu kamthi palambu kidu, radie vihidu khanu mumda-malie jam panaena tam namahu, kusuma-dama-koyamdu kamaho Il Herein we find forms with -d- derived from original -t-. The metre is 'Matra' (SC.IV 8 ff. and CH. V 17 ff.) which has 15+12+15+12+15 Matras in its consecutive five Caranas. 4. The verse cited under 8-4-422 to illustrate substitution in Apabhramba of drehi for Skt. drsti is as follows: ekkam ekkau jai-vi joedi hari sutthu savvayarena, to-vi drehi jahim kahim vi rahi ko sakkai samvarevi, daddha-nayana nehem palutta 11 The verse is also cited by Svayambhu (SC.IV 10.2) under the name of a poet Govinda. Svayambhu differs in more than one reading from Hemachandra, and the original features do not seem to be preserved, as we have joei for joedi and ditthi for the characteristic drehi. Here also the metre is Matra. We have a present third person singular form with the -edi ending and another form (drehi) that preserves --- in a cluster. 5. Noting briefly this characteristic of Apabhramsa, Namisadhu has cited in his commentary on Rudrata's Kavyalamkara (II 11-12) the following verse-line to illustrate the preservation of -r- in a cluster: gotru gamjidu malidu carittu This seems to be the first Carana of a Matra stanza, and it also exhibits the characteristic of the change of -- to d.. 6. The Apabhramba illustration cited in SP. on p.121 (jaivioga dahi xx etc.), although considerably corrupt, is most probably in the Matra metre and contains present third R. PISCHEL has noted cases of preservation of r in Hemacandra's illustrations (under $8 28 and 268), as well as cases of -- > -d-etc. (under the relevant section). 2 This verse also occurs in SP. (p. 309). See 'Prakrit Verses in Works on Sanskrit Poetics', Pt. I, Appendix, p. 13, no. 24. -- Refer for the two Matras < and > as discussed in (3) and (4) to L. ALSDORF Apabhramsa-Studien, Leipzig 1937, pp. 106-07 and pp. 55-56. ALSDORF's discussion does not affect the points mentioned by DR. BHAYANI. THE EDITORS.

Page #3
On Early Apabhramsa person singular forms in -adi (see 'Prakrit Verses in Works on Sanskrit Poetics', I, Appendix, p.6, no.9). 3 7. The illustration of the Radda metre given under CH. V 23 is as follows: lumdhidu camdana-valli-pallamki sammilidu lavamga-vani, khalidu vatthu-ramaniya-kayalihim I ucchalidu phanni-layahim, ghulidu tarala-kallola-lavalihim || cumbidu mahavi-vallarihim, pulaida-kami-sariru I bhamara-saricchau samcaradi, raddau malaya-samiru II Herein we have numerous forms showing the change -t->d-. It should be noted that in the numerous illustrations given by Hemacandra to illustrate varieties of the Matra metre (under CH. V 17-22) the present third person singular forms have the usual Maharastri ending -ai. The above-cited illustration is the only exception. 8. In Virahanka's Vrttajatisamuccaya (IV 31) the following Duvahaa (dvipathaka, doha) occurs in the definition of the Radda metre: eahu mattahu amtimahu, jamvihi duvahau bhrodi (? bhodi) | to tahu namem radda phudu, chamdai kai-janu bhrodi (? brodi) || The form bhodi in the first line is given by Hemacandra (8-4-273, 274) as a characteristically Sauraseni form. bhrodi (most probably corrupt for brodi) in the second line is an -preserving form. 9. Hemacandra has given bruv (Skt. bra-) as characteristic of Apabhramsa (8-4-391), and he has cited bruvaha, broppi, and broppinu as illustrations. 10. In a passage in Uddyotana's Kuvalayamala that gives the reactions and the advice of the village elders to the hero who has prepared to burn himself on a pyre to atone for his sins the language is full of rustic colloquialisms (Part I, p.63, lines 18-26). All the four relevant statements are given in the Matra metre (lines 18, 20, 22-23, and 25-26). The notable forms are: (18:) praraddhaum, prai (for praim), protu, bhrati, samprati; (20:) viraidu, samprati, brollitaum, praraddhu; (22-23:) pravu, vrata; (25-26:) bhramiti, pravesi, mitra-drojjhu. The abundance of r-preserving forms and one form with the change -t- > -d-besides those that preserve the original intervocalic -t--are noteworthy3. The Kuva 3 A.N.UPADHYE has noted these traits. See Kuvalayamala II, Introduction, pp. 80 and 82. Forms like pravu (< Skt. papam) with intrusiver have been noted as a peculiarity of Apabhramsa by Namisadhu (see '5' and '13') and Hemacandra (8-4-399, illustration vrasu < vyasah). This trait was inherited by Gujarati as can be seen from numerous instances like karod, srapa, saran, kalindri, karolio, truthvi, trabu, mathravati, etc. See Anusilano, pp. 149-53, and Vyutpattivicar, p. 156. - In Puspadanta's Mahapurana

Page #4
H.C. Bhayani layamala was completed in 779 A.D. 11. In the following Prakrit verse, occurring in the Vasudevahindi of Sanghadasagani (about 6th century), we find an admixture of Apabhramsa forms: pasim kappim cauramsiya reva-paya-punniyam sediyam ca genheppi sasi-ppabha-vanniyam maim suyam pi ekkalliyam sayani nivanniyam savva-rattim ghosei samana-savanniyam || (p.28) Here the absolutive forms kappim and genheppi are specially noteworthy. We are reminded of similar forms such as broppinu, broppi, jeppi, gampi, gameppi, jineppi occurring in the illustrations cited by Hemacandra (8-4-391, 440, 441, 442). The second illustration given by Hemacandra in 442 contains also the "Sauraseni' form kiladi. 12. From the above-noted instances it is quite clear that in the case of the Apabhramsa verses in the Matra and Radda metres there was a consistent tradition of using -r- preserving forms and forms with voiced stops. Matra, along with its extension Radda, seems to be among the earliest Apabhramsa metres. Svayambhu (ninth century) was familiar with a narrative poem of Govinda possibly composed wholly in the Radda metre. He has cited six Matras of Govinda, one each of Chailla and Suddhasila, one anonymously; one Radda of Jinadasa, and one anonymously. But due to the late date of the manuscripts these verses have not preserved their original phonological features. Bhamaha's remark that self-standing verses, i.e. Muktakas, were composed in such metres as Gatha (Prakrit), sloka (Sanskrit), and Matra (Apabhramsa) also supports our observations. 13. It appears that such poems in the Matra and Radda metres represent an early stratum of Apabhramsa poetry as compared to the bulk which consists of the later "Sandhibandha" and other types of Mahakavya in standardized Apabhramsa that developed under the impact of literary Maharastri. The illustrations under Siddhahemacandra 8-4-396 with voiced intervocalic stops, and several forms which preserve --- in clusters (PISCHEL 268) can be assigned to that early stratum. Phonologically, these latter traits had come to be known as distinguishing marks of Apabhramsa as can be seen from the following two illustrations of the Bhasa-slesa of Sanskrit and Apabhramsa given by Rudrata: (972 A.D.) we find three such instances: vruhu < budhah (16, 11.7), vrasu < vyasah (98, 8.6), vrahiu > vadhitah (99, 3.5). "But due to the late date of the manuscripts, these verses have not preserved all the original phonological features. Kavyalamkara 1.30: anibaddham punar gatha-sloka-matradi tat-punah | Here the word matra has so far invariably been misinterpreted.

Page #5
On Early Apabhramsa 5 dhiragacchadume hatamududdharavarisadassu | abhramadaprasaraharanu ravikirana tejassu || (Kavyalamkara, IV 15) kridanti prasaranti madhu, kamala-pranayi lihanti | bhramara mitra suvibhrama, matta bhuri rasanti || (ibid. IV 21) Namisadhu, in his brief summary of Apabhramsa grammar (which most probably has been reproduced from an earlier source) has given four phonological characteristics of Apabhramsa and has illustrated them appropriately: (1) Preservation of the postconsonantal in a source-cluster as in bhrayaru. (2) Interpolation of an unorganic -- as in vracalau. (3) Change of intervocalic-t->d- as in 'gotru gamjidu... (already cited under '5' above). (4) Preservation of r as in 'trna-samu ganijjai". The absolutives in -ppi also seem to be an archaic feature. In the passages discussed above we find together forms with voiced consonants, absolutives in -ppi, and r-preserving clusters. We can also note in this connection the form prassadi, for which Hemacandra has provided a separate rule (8-4-393). Now it is well-known that, in the transmission of Apabhramsa texts, original forms with r-clusters (as also those with r in the Sanskrit source-form) were changed to r-less forms under the influence of Prakrit. See ALSDORF's discussion in his Introduction to the Harivamsapurana (pp.137 ff.). ALSDORF has observed that the manuscript 'C' used by him for editing the text has seventy-five instances of such forms". Even in later works we occasionally find instances of r and r preserved. The Kavidarpana (ca. 12th century), for example, has 'trasai mrga-ganu' in the verse given to define and illustrate the Apabhramsa metre Pancananalalita (p.23, verse 14.1). 14. Quite obviously, we do not know for certain whether the special Apabhramsa features noted above characterized early Apabhramsa in general, or were regional, i.e. 'dialectical', traits. Scholars have suggested that the preservation of r was possibly a distinctive characteristic of the Vracada variety of Apabhramsa. But there is little actual evidence to support this speculation. Since long it has become clear that modern efforts to interprete and account for variations in literary Apabhramsa as sectarian (e.g. H. JACOBI, L. ALSDORF) or regional varieties (e.g. G.V. TAGARE) were based on wrong assumptions and on misunderstanding of the available evidence, which itself was partial." The language of the Vasudevahindi-Madhyamakhanda (ca. 700 A.D.) provides 6 This trait is also inherited by Gujarati which has numerous words, derived from Sanskrit, which have preserved from the cluster in the source-word. When I undertook to analyse the Apabhramsa of the Samdesarasaka in 1945, I was apparently guided by the views of H. JACOBI and L. ALSDORF regarding the dialectal variation in Apabhramsa (Samdesarasaka, Introduction, pp.4 and 25),

Page #6
H.C. Bhayani good evidence for the use of Sauraseni in narrative prose, besides its long-established employment in the prose of Sanskrit drama under specific conditions. In the language of the Madhyamakhanda we quite frequently find the intervocalic -t- unchanged, and perhaps equally frequently we find it changed to d-. Voicing of unvoiced aspirates and preservation of voiced intervocalic stops are also frequently met with. See pp.37 and 41 of the Introduction to the Madhyamakhanda. Originally, the number of cases might have been greater still in view of the fact that the language of our rather late manuscripts is modernized. On p.42 of the Introduction to the Madhyamakhanda the editors wrote: "... here we have in the Vasudevahindi-Madhyamakhanda a Svetambara work whose Prakrit has Sauraseni phonological elements comparable to what we find in the Prakrit of Digambara works". It seems probable that, during the middle centuries of the first millennium, literary Apabhramsa was marked by Sauraseni phonological features while later Apabhramsa was characterized by Maharastri features. There were a few other peculiar features in Apabhramsa which also changed in the course of time. The source-work of Namisadhu's extract on Apabhramsa (see '13' above) and the concluding rule in Hemacandra's treatment of Apabhramsa, i.e. 8-4-446 (saurasenivat: this obviously overlaps on 8-4-396) - along with a few other rules - account for peculiar traits of the language of a specific body of texts which represent early Apabhramsa literature. REFERENCES Anusilano (in Gujarati), H.C. BHAYANI. 1965. Chandonusasana of Hemacandra, ed. H.D. VELANKAR 1961 (abbreviated as CH.). Comparative Grammar of the Prakrit Languages, R. PISCHEL (Transl. SUBHADRA JHA) 1965. Refer also to the Apabhramsa supplement: R. PISCHEL Materialien zur Kenntnis des Apabhramsa. Berlin 1902. Harivamsapurana (from the Mahapurana of Puspadanta), ed. L. ALSDORF 1936. 8 This 'archaic Apabhramsa (i.e. archaic fron the perspective of the later standardized Apabhramsa) has a parallel in 'archaic' Prakrit, from the language of Bharata's Dhruvas onwards. - In the Mscchakatika, the language of the gambler Mathura is Dhakki according to the commentator. In this dialect, nominative singular forms of masculine a-stems have the ending -u, which is also characteristic of Apabhramsa. It is noteworthy that Dhakki also has some forms that demonstrate the change --> -d- Similarly, the fact that, in manuscript 'C' of Canda's Praktalaksana, Apabhramsa is treated along with Sauraseni is quite significant.

Page #7
On Early Apabhramsa Kavidarpana (anonymous), ed. H.D. VELANKAR 1962. Kavyalamkara of Bhamaha Kavyalamkara of Rudrata with Namisadhu's commentary, Kavyamala Series, 3rd edition 1928. -- Namisadhu's commentary on II 11-12 is reproduced in L. NITTI-DOLCI's "The Prakrit Grammarians' (Transl. PRABHAKARA JHA) 1972, pp.166-169. Kuvalayamala of Uddyotanasuri, ed. A.N. UPADHYE. I (1959), II (1970). Micchakatika of Sudraka Prakrit Grammar of Hemacandra, ed. P.L. VAIDYA 1958. Prakrit Verses in Works on Sanskrit Poetics, ed. V.M. KULKARNI, I (1988), II (1990). Prakstalaksana of Canda, ed. A.F.R. HOERNLE 1880. Samdesarasaka of Abdala Rahamana, . ed. MUNI JINAVIJAYA 1945. With an introduction by H.C. BHAYANI. Stngaraprakasa of Bhoja, ed. G.R. JOSYER, Part I 1955 (abbreviated as SP.). JOSYER has published the complete SP in four parts (1955, 1963, 1969, no year). Svayambhuchandas of Svayambhu, ed. H.D. VELANKAR 1962 (abbreviated as SC.). Vasudevahindi of Sanghadasagani, ed. MUNI CATURA VIJAYA and MUNI PUNYAVIJAYA 1989 (reprint). Vasudevahindi-Madhyamakhanda of Dharmasenagani, ed. H.C. BHAYANI, R.M. SHAH 1987. Vittajatisamuccaya of Virahanka, ed. H.D. VELANKAR 1962. Vyutpattivicar (in Gujarati) H.C. BHAYANI 1975.