^{page 341 note 1} Cf. Kane, P. V., History of Dharmaśāstra (Ancient and Medieval Religious and Civil Law), Poona (Bhandarkar Oriental Research Institute, 1930–62), vol. v, part 2, p. 839Google Scholar for a collation of the Purānic passages touching on pañcalaksana.

^{page 341 note 2} Kirfel, Willibald, Das Purāna Pañcalaksana, (Kurt Schroeder, Bonn, 1927).Google Scholar

^{page 342 note 1} Shastri, M. Haraprasad, ‘The Maha-Puranas’, Journal of the Bihar and Orissa Research Society xiv, 3 (Sept. 1928), 326.Google Scholar

^{page 342 note 2} Hazra, R. C., ‘The Purānas’, in Bhattacharyya, Haridass (ed.), The Cultural Heritage of India, 2nd rev. ed., Calcutta (Ramakrishna Mission Institute of Culture, 1953–1962), 11, 246–7.Google Scholar

^{page 342 note 3} Out of the welter of such theories, the following may be cited as being illustrative, if not exhaustive: Pargiter, F. E. (ed.), The Purāna Text of the Dynasties of the Kali Age (Oxford University Press, London, 1913), introduction p. xviiGoogle Scholar note; Hazra, R. C. (op. cit. pp. 246–53Google Scholar); Kane, P. V. (op. cit. pp. 853–5Google Scholar); Gyani, S. D. (‘The Date of the Purānas’, Purāna 1, 2 (Feb. 1960), 213–19Google Scholar; 11, 1–2 (July 1960), 68–75Google Scholar); Dimmitt, Cornelia and Buitenen, J. A. B. van (ed. and tr., Classical Hindu Mythology: A Reader in the Sanskrit Purānas (Temple University Press, Philadelphia, 1978), pp. 4–13).Google Scholar

^{page 342 note 4} Dimmitt, and Buitenen, van, op. cit. p. 5Google Scholar. Inquiry into these matters, of course, continues. Two recent examples thereof may be cited as particularly promising: Stephan Hillyer Levitt, ‘A Note on the Compound Pañcalaksana in Amarasinha's, Nāmalingānuśāsana’, Purāna, xviii, 1 (Jan. 1976), 5–38Google Scholar, in which the author argues that the standard scholarly interpretation of pañcalaksana is incorrect; and Bonazzoli, , Giorgio, , ‘The Dynamic Canon of the Purāna-s’, Purāna xxi, 2 (July 1979), 116–66Google Scholar, which provides important data on the stages in the evolution of the Purāpas’ understanding of their own boundaries. Both studies are based on close textual analysis, but neither raises up for consideration the sort of explicitly religious issue that we turn to below.

^{page 342 note 5} Rocher, Ludo, ‘The Meaning of purāná in the Rgveda’, Wiener Zeitschrift für die Kunde Südasiens xxi (1977), 5–24.Google Scholar

^{page 343 note 1} Ibid. p. 6.

^{page 343 note 2} Ibid. pp. 12, 16.

^{page 343 note 3} Ibid. p. 20.

^{page 343 note 4} Cheever MacKenzie Brown has nicely captured the relationship between Purānas and Vedas, between smrti and śruti, by referring to the Veda as revealed truth, and to the ‘Purāna as revealing truth’. Cf. his God as Mother: An Historical and Theological Study of the Brahmavaivarta Purāna (Claude Stark and Co., Hartford, VT 1974), chap. 1, esp. pp. 18–19.Google Scholar

^{page 343 note 5} Rocher, , op. cit. p. 24.Google Scholar

^{page 343 note 6} Gupta, A. S. ‘Purānas and their Referencing’, Purāna vii 2 (July 1965), 323–6.Google Scholar

^{page 343 note 7} The data that are summarized here are drawn primarily from Ibid. supplemented by Kane's more exhaustive collection of relevant passages ( op. cit. pp. 815 ff.Google Scholar).

^{page 343 note 8} The text is given by Kane, , op. cit. p. 816Google Scholar, note 1325. This translation is that of Whitney, and Lanman, : Atharva-Veda Samhitā, Harvard University (Cambridge, 1905), p. 646.Google Scholar

^{page 344 note 1} tān upadiśati purānam vedah soyam iti kimcitpurānam ācaksitā.

^{page 344 note 2} The distinction between itihāsa, ‘history’ (usually analysed as iti ha āsa, ‘thus it truly was’), and purāna is not always clear. For an extended discussion of these terms, cf. Gupta, A. S., ‘Purāna, Itihāsa, and Ākhyāna’, Purāna vi, 2 (July 1964), 451–61Google Scholar. For a discussion of these and other terms in the larger context of Indian historiography, cf. Majumdar, R. C., ‘Ideas of History in Sanskrit Literature’, in Phillips, C. H. (ed.), Historians of India, Pakistan, and Ceylon (Oxford University Press, London, 1961), pp. 13–28.Google Scholar

^{page 344 note 3} Gyani, , op. cit. p. 219Google Scholar. The alleged original unity of purāna is, of course, compatible with either the divinity or the humanity of its origin. That it inclines toward the former, however, will be seen in the next paragraph.

^{page 344 note 4} Cf. the references given by Gupta, , ‘Purānas and their Referencing’, p. 324Google Scholar and Kane, , op. cit. p. 829.Google Scholar

^{page 344 note 5} Winternitz, M. (A History of Indian Literature, Mrs Ketkar, S. (tr.), Russell, and Russell, (New York, 1971), 1, 527 nGoogle Scholar) quotes this passage from the Visnu Purāna without specific reference. Some Purānas see Vyāsa as an incarnation of Brahmā or of Śiva: cf. Kane, , op. cit. p. 857Google Scholar and note 1390.

^{page 344 note 6} Hazra, R. C., Studies in the Purānic Records on Hindu Rites and Customs (The University of Dacca, Dacca, [1940]), p. 2.Google Scholar

^{page 345 note 1} The three passages (given by Kane, , op. cit. p. 817Google Scholar, note 1328) are 1. 6. 19. 13; 2. 9. 23. 3–6; 2. 9. 24. 6. The text of the last of these reads punahsarge bījārthā bhavantīti bhavisyatpurāne.

^{page 345 note 2} ākhyānaiś cāpyupākhyanair gāthābhih kalpajoktibhih/purānasamhitām cakre purānārthaviśāradah//. The text is given by Hazra, , Studies, p. 5 nGoogle Scholar; Kane, , op. cit. p. 858Google Scholar, note 1392; and Gupta, , ‘Purānas and their Referencing’, p. 325 nGoogle Scholar. The latter two authors identify the text as coming from Brahmānda Purāna 11. 34. 21Google Scholar; Vāyu Purāna 1. 60. 21Google Scholar (varia lectio: kulakarmabhih); and Visnu Purāna iii. 6. 15Google Scholar (v.1.: kalpaśuddhibhih).

^{page 345 note 3} This tradition is widely attested. Cf. Gupta, , Purānas and their Referencing’, p. 325Google Scholar; Hazra, , ‘The Purānas’, p. 244Google Scholar; Agrawala, V. S., ‘Original Purāna Samhitā’, Purāna viii, 2 (July 1966), pp. 235–9Google Scholar; Dikshitar, V. R. Ramchandra, ‘The Purānas: A Study’, Indian Historical Quarterly viii, 4 (Dec. 1932), 753Google Scholar 5; Pargiter, F. E., Ancient Indian Historical Tradition (Oxford University Press, London, 1922), pp. 22–3.Google Scholar

^{page 345 note 4} In this same vein, we may note that a single Purāna can affirm both that Vyāsa is the author of all eighteen Purānas and that the different Purānas have separate authors ( Bhavisya Purāna 1. 1. 58Google Scholar and 111. ii. 28. 10–15, cited by Gupta, , ‘Purāpas and their Referencing’, pp. 333–4Google Scholar). This is not, of course, to say that the same individual composed both passages, for, as Hazra has shown (Studies, passim), most of the Purānas have been subjected to multiple recastings. What is significant is that the final redactor did not see fit to smooth over the apparent logical incompatibility of these two traditions by editing out one of them: each is affirmed as a different facet of, or perspective on, the ultimate truth. For further reflections on the ‘Purānic spirit’, cf. Brown, , op. cit. pp. 17–19.Google Scholar

^{page 346 note 1} This edition - The Mahābhārata, for the first time critically edited by Sukthankar, Vishnu S. (and others), Bhandarkar Oriental Research Institute (Poona, 1933–59Google Scholar), 19 vols. – is the end-product of a process of debate, and trial-and-error, that reaches back into the nineteenth century. For a fascinating account of that process, cf. Sukthankar's ‘Prolegomenon’ to vol. 1 of the critical edition.

^{page 346 note 2} For the sake of convenience, and because we are here concerned with the Purānas as scripture, I will rely primarily, though not exclusively, on various articles that have appeared in Purāna journal. Most of the participants in this debate have written in these pages, and it is possible to extrapolate from what is said there to see how the work of other scholars (e.g., Lévi-Strauss, Dumézil) would fit into the debate.

^{page 346 note 3} Cf. Sukthankar, , op. cit., passim.Google Scholar

^{page 347 note 1} Lévi, Sylvain, ‘Review of The Mahā-Bhārata, for the first time critically edited by Vishnu S. Sukthankar’, Journal Asiatique ccxxv, 2 (Oct.-Dec. 1934), 282Google Scholar, quoted in English translation in Biardeau, Madeleine, ‘Some More Considerations about Textual Criticism’, Purāna x 2 (July 1968), 116.Google Scholar

^{page 347 note 2} Biardeau, , ‘Textual Criticism’, pp. 116–418Google Scholar. Biardeau overlooks the Purānic tradition of their divine origin, a matter to which we shall return below.

^{page 347 note 3} Ibid. p. 121.

^{page 347 note 4} Ibid. p. 122. Biardeau, points out (‘The Story of Arjuna Kārtavīrya without Reconstruction)’, Purāna XII, 2 (July 1970, p. 302 nGoogle Scholar) what she thinks is a contradiction in Sukthankar's own attitude to this matter of authority: in his lectures on the Mbh (given in 1942, but not published until 1957, as On the Meaning of the Mahābhārata, The Asiatic Society of BombayGoogle Scholar), he relied on the Vulgate rather than on the critical edition of the text. However, since the critical edition was not completed until 1959, it is hard to see how he could have used it in 1942.

^{page 347 note 5} Biardeau, , ‘Textual Criticism’, p. 123.Google Scholar

^{page 347 note 6} Biardeau, , ‘Arjuna Kārtavīrya’, p. 293Google Scholar. This essay, which Biardeau intended as a preliminary demonstration of her alternative to a text-historical interpretation of the Purānas, has been supplemented elsewhere by more sustained, and very impressive, ventures: e.g. ‘Études de Mythologie Hindoue’, Bulletin de l'École Française d'Extrême Orient LIV (1968), 19–45Google Scholar; LV (1969), 59–105Google Scholar; LVIII (1971), 17–89Google Scholar; also comptes-rendus of ‘Conférences de Mlle Madeleine Biardeau’, Annuaire de l'École Pratiques des Hautes Études, Section des Sciences Religieuses LXXVII (1969–70), 168–73Google Scholar; LXXVIII (1970–71), 151–61; LXXIX (1971–2), 139–46.

^{page 348 note 1} Bedekar, V. M., ‘Principles of Mahābhārata Textual Criticism: The Need for a Restatement’, Purāna xi (July 1969), 219Google Scholar. This is a very tricky matter, by no means as straightforward as Bedekar implies. To pursue it here, however, would take us too far afield. For further discussion, cf. pp. 109–24 of my ‘The Crystallization of the Worship of the Goddess: The Sources and Context of the Devī-Māhātmya’, unpublished Ph.D. dissertation (Harvard, 1977).Google Scholar

^{page 348 note 2} Bedekar, , op. cit. p. 221.Google Scholar

^{page 348 note 3} Bedekar's quotation of Sukthankar occurs Ibid. p. 225. Because Bedekar's quotation has a number of errors in transcription, the version given here follows Sukthankar's original ( op. cit. p. ciiGoogle Scholar) and is slightly fuller than Bedekar's.

^{page 348 note 4} All-India Kashiraj Trust (Varanasi, 1968 and 1972Google Scholar, respectively).

^{page 348 note 5} Gupta, A. S., ‘A Problem of Purānic Text Reconstruction’, Purāna xii, 2 (July 1970), 305.Google Scholar

^{page 349 note 1} For a fuller account of Gupta's proposal, cf Ibid., passim, esp. pp. 310–21.

^{page 349 note 2} On Biardeau, , cf. the references given in notes on p. 347Google Scholar above. O'Flaherty's tour de force remains her Asceticism and Eroticism in the Mythology of Śiva (Oxford University Press, London, 1973Google Scholar). If one identifies the emphasis on myth as the crucial feature of what I have called ‘structuralism’, then it becomes relevant to cite here the work of Stig Wikander, (‘Pāndava-sagen och Mahābhāratas mystika forutsattningar’, Religion och Bibel IV (1947), 27–39Google Scholar) and Dumézil, Georges (Mythe et épopée, vol. 1Google Scholar: L'Idéologie des trois fonctiones dans les épopées des peuples indo-européens (Gallimard, Paris, 1968)).Google Scholar

^{page 349 note 3} Review of Hinduism: New Essays in the History of Religion, edited by Smith, Bardwell L., Brill, E. J. (Leiden, 1976Google Scholar), Journal of the American Oriental Society, XCVIII, 3 (July-Sept. 1978), 326Google Scholar, commenting on Hiltebeitel's essay, ‘The Burning of the Forest Myth’. Elsewhere, Hiltebeitel claims that ‘the Indo-European perspective of Dumézil and the Purānic, one might say “Hindu”, perspective of Biardeau are both valid, and…to borrow from a Sāmkhya similitude, they may at some points be as necessary to each other, in making a way through the Mahābhārata forest, as the blind man and the lame’ ( The Ritual of Battle: Krishna in the Mahābhārata, Cornell University Press (Ithaca and London, 1976), p. 140Google Scholar). If it looks as if Dumézil's position is being variously categorized here, it is because van Buitenen ‘has apparently made a convert of Dumézil’ on the matter of the utility of the critical edition: cf. Hiltebeitel, , ‘Burning’, p. 209Google Scholar and n.

^{page 350 note 1} Cf. Biardeau, , ‘Textual Criticism’, pp. 116–17, 122Google Scholar; Bedekar, , op. cit. pp. 218–20, 223–4.Google Scholar

^{page 350 note 2} Cf. the essay of Deshpande, V. V. (‘Nature and Significance of Itihāsa and Purāpa in Vedic Purusārtha Vidyās’, Purāna xvi, 1 (Jan. 1974), 47–66Google Scholar; XVI, 2 (July 1974), 245–60Google Scholar; XVIII, 2 (July 1976), 197–211Google Scholar) in which he seeks to develop an adequate scholarly approach to the Purānas by first exploring the various assumptions that have guided Western historical scholarly writing since the Renaissance. Cf. also Pathak, V. S., ‘Ancient Historical Biographies and Reconstruction of History’, in Problems of Historical Writing in India, Proceedings of the Seminar held at the India International Centre, New Delhi, 21–25 Jan. 1963, pp. 11–21.Google Scholar

^{page 350 note 3} Smith, Wilfred Cantwell, ‘Comparative Religion: Whither – and Why?’, in Eliade, Mircea and Kitagawa, Joseph M. (eds), The History of Religions: Essays in Methodology (The University of Chicago Press, Chicago and London, 1959), p. 52Google Scholar. Cf. also Smith, Wilfred Cantwell, ‘A Human View of Truth’, in Hick, John (ed.), Truth and Dialogue in World Religions: Conflicting Truth-Claims (The Westminster Press, Philadelphia, 1974), p. 35Google Scholar: ‘A counterpart for the social sciences and humanities of the verificationist principle in the natural sciences is the principle that no statement about human affairs is true that cannot be existentially appropriated by those about whom the statement is made.’

^{page 350 note 4} One might be tempted to minimize the difficulty here by asking whether Gupta's viewpoint is representative of the Hindu tradition as a whole. The problem with such a tack is: if we do exclude Gupta's view because it is non-representative, on what grounds do we do so?

^{page 351 note 1} Biardeau, , ‘Textual Criticism’, p. 120.Google Scholar

^{page 351 note 2} Keeping the desideratum in front of us may prove rewarding, even if our labors fall short of the ultimate goal: e.g., in sharing my views on this matter with Professor Harry Buck recently, he commented that, when examining manuscripts of the Rāmāyana in India, he had been delighted to see how the character of Rāvana became less and less fearsome, the closer one got to Srī Lanka. The work of Bonazzoli, , cited above at note 4 on p. 342Google Scholar is an example of how much more sophisticated Purānic scholarship can become in its reconstruction of textual history.

^{page 351 note 3} There is an implication to Biardeau's comments on the matter of authority which should be noted here. She felt, it will be recalled, that Sukthankar et al. placed authority in that text which was chronologically closest to Vyāsa, rather than in the opinion of local Brahmans, and Bedekar replied that Sukthankar had never claimed authoritativeness, in traditional terms, for his text. Biardeau's point may then be reformulated: what Sukthankar has done is to place authority for interpretation of the Mbh in the principles of textual criticism, of academic inquiry, rather than in the opinion of local Brahmans. A corroboration of this might be seen in Bedekar's capitalization of the words ‘Western Scientific’ when discussing the matter of authority. If this is the case, then there is an element of truth in Lévi's calling the critical edition of the Mbh ‘the recension of Poona’ (op. cit. ‘ p. 347Google Scholar): it is one recension, along side of many others, and they are differentiated from each other simply by the respective authorities to which (or to whom) they turn for the constitution of their individual texts. It should be noted, however, that even if the critical edition is thus seen as simply one of many recensions, it still retains one unique feature: it is also useful for a critical historical inquiry into the growth of the Mbh text tradition as a whole, and of Indian religion as reflected therein. Moreover, when commitment to text-historical inquiry is thus seen as being a matter of authority for the constitution of a text, it is even more clear that this commitment can be quite different from the commitment that is faith.

^{page 352 note 1} I am indebted to Prof. John B. Carman for this formulation of the matter.

^{page 352 note 2} This quotation comes from an article entitled ‘Delhi scholar rebuts new Ramayana theory’ that was sent my by a colleague in India in January 1976. The newspaper was not identified, and I have been unable to locate the specific source of the article.

^{page 352 note 3} At the risk of being self-indulgent, let me note that it is this same issue of subjectivity-objectivity, and the role of the person in the apprehension of quality, that commands Robert Pirsig's attention at the heart of Zen and the Art of Motorcycle Maintenance (Bantam Books, Toronto, New York, London, 1975Google Scholar): cf his analysis, particularly of the word ‘just’, in the statement ‘Quality is just what you like’, op. cit. pp. 223–34.Google Scholar