Pramāṇa · Search · About · API

Ancient Indian Historical Tradition c9: Chapter IV. The Age of the Original Purana

Ancient Indian Historical Tradition c9

Language
English

Chapter IV. The Age of the Original Purana

c9.2

The age of the Puranas may now be considered according to the evidence obtained from themselves and from other sources. In doing this, the discussion must proceed from later known facts to earlier evidence.

c9.3

The Vāyu Purana existed before A. D. 620, because it is referred to by Bāṇa in his Harṣa-carita,² and the writing in a MS. of the Skanda in the Royal Library of Nepal shows that that Purana also existed about that time.³ Verses praising gifts of land are quoted in various land-grants, that are dated; and some of those are found only in the Padma, Bhaviṣya and Brahma Puranas, and thus indicate that those Puranas were in existence before A. D. 500 and even long before that time.⁴ Some of those verses, which occur in grants of the years 475-6 and 482-3, are declared in some grants to have been enunciated by Vyāsa in the Mahābhārata.⁵ They do not however (as far as I know) occur there, but are found in the Padma and Bhaviṣya, and nowhere else. Such an error, citing the Mahābhārata instead of the Puranas, in a land-grant, which was not a learned treatise, is venial, especially as Vyāsa was believed to be the author of all those works. The mistake shows the pre-eminent position held by the epic then. The important point however is that the grantors assert that a book was the original authority for those verses; not popular lore nor unknown compositions. It follows therefore that either the Padma or the Bhaviṣya or both existed before A. D. 475 and even much earlier, and a similar conclusion, though not so clear, may be drawn as regards the Brahma.

c9.4

² Chap. III, paragraph 4. Cowell and Thomas, Translation, p. 72. See V. A. Smith, *Early History of India*, 3rd ed., pp. 21-22.

c9.5

³ *Trans. VIth Oriental Congress*, vol. iii, p. 205. The MS. was sent to Oxford for inspection.

c9.6

⁴ *JRAS*, 1912, pp. 248-55. Gar may perhaps be added because verses 3*b* and 7 occur in it (ii, 31, 14 and 4).

c9.7

⁵ *JRAS*, 1912, 253-4.

c9.8

Further the Matsya, Vāyu and Brahmāṇḍa say in their accounts of the dynasties of the Kali age that they borrowed their accounts from the Bhaviṣya;¹ and the internal evidence therein shows that the Bhaviṣya existed in the middle of the third century A. D., the Matsya borrowed before the end of that century, and the Vāyu and Brahmāṇḍa borrowed in the next century.² The present Bhaviṣya, as presented in the Śrī-Veṅkaṭeśvara edition, does not contain that account, but another altogether corrupt and false, and the reason is that the Bhaviṣya has been freely tampered with in order to bring its prophecies up to date and the ancient matter utterly vitiated: but those three Puranas show what it contained in the third century, as regards the dynasties of the Kali age.

c9.9

¹ My Dynasties of the Kali Age, pp. vii-viii. JRAS, 1915, pp. 141-2, 517-18. These three Puranas existed before.

c9.10

² Id., pp. xii-xiv, xxiv, xxv. JRAS, 1912, pp. 142-3.

c9.11

Next may be considered the mention of the Bhaviṣyat Purana in the Āpastambīya Dharmasūtra (chap. III), and the inferences that may be drawn therefrom.

c9.12

The ‘Purana’, as already pointed out, first came into existence as a collection of ancient legendary lore, and this, its original nature, is an essential fact. Āpastamba obviously refers in his citations of Puranas (chap. III) to definite books. Now the Bhaviṣyat Purana plainly professed by its title to treat of ‘the future’, and its title is a contradiction in terms. The first inference therefore is that such a name could not have been possible until the term Purana had become so thoroughly specialized as to have lost its proper meaning, and had become merely the designation of a particular class of books. It would have required the existence of a number of books called Puranas to produce that change, and manifestly they must have had their own special names to distinguish one from another, and so convert their common title Purana into a mere class designation.¹

c9.13

¹ Compare the Journals of the Royal Asiatic and other Societies, where the title ‘Journal’ has completely lost its original meaning.

c9.14

The next inference is that the foregoing change implied long usage—that is, the Puranas began long before Āpastamba’s time. His sūtra is estimated by Bühler as not later than the third century B.C., and possibly 150–200 years earlier.² His citation of the Bhaviṣyat as an authority shows that it was no new work then, but had acquired an acknowledged position of dignity, which it could not have attained to in less than half a century. Hence the Bhaviṣyat cannot be placed later than the early part of the third century B.C., and even possibly earlier still by the above 150–200 years. At that time the title Purana had completely lost its original meaning, and the question arises, what length of time would have been required to bring about that result. There can be no definite pronouncement on this, but the time cannot have been less than two centuries, considering the conditions of literature in those times, and was probably much longer. Hence Puranas must have existed at least as early as the beginning of the fifth century B.C.; and this lower limit would be shifted 150–200 years earlier if a prior date be given to Āpastamba. It is quite probable therefore that the Matsya existed long prior to him, as indeed his citation of it indicates (p. 44).

c9.15

² SBE ii, p. xliii.

c9.16

The third inference from the name Bhaviṣyat is that before that Purana could have been composed about ‘the future’ there must have been some general consensus of opinion when ‘the future’ began. The Bhaviṣyat and Bhaviṣya are referred to here as distinct, the former being the Purana cited by Āpastamba, and the latter that which existed in the third century A.D. as mentioned above, and which, modified by the continual tampering to which it has notoriously been subjected, we have now. When the Bhaviṣyat was composed, whenever that was, obviously everything after that time was ‘future’, so that the third century B.C. at the latest and all after time fell into ‘the future’; and it would have included the two preceding centuries if Āpastamba should be antedated. It is however pretty obvious that ‘the future’ must have commenced before the Bhaviṣyat was composed, otherwise there would have been little to speak of beyond vague prophecies and teleology, subjects hardly attractive enough of themselves alone to win general interest in a new ‘Purana’.¹ Hence ‘the future’ had probably been reckoned as having begun some time before, so that the author could have commenced with interesting tales of what had taken place before launching out into talk about the real and unknown future. This inference is entirely supported by the statement ante that the Bhaviṣyat began with the story of Kṛṣṇa’s son Sāmba soon after the Bhārata battle, and by the following further declarations.

c9.17

¹ That was done as in the Kalki Upapurana, but it was a later elaboration and very small in scope.

c9.18

There is some definite information concerning, first, what was ranked as ‘past’, and secondly, what was considered to be ‘future’. First, all the epic and Puranic traditions that deal with kings and princes, and less markedly with rishis, stop short soon after the great Bhārata battle. Till that event they are fairly copious, and after that they take in the reign of Yudhiṣṭhira and his brothers, and, in the early and final chapters of the Mahābhārata, speak of their successors Parīkṣit II and Janamejaya III;² but there are no traditions about Janamejaya’s successors, nor about any of the kings of the great dynasties of Ayodhyā and Magadha after that battle beyond a few curt allusions in the list of the kings of the Kali age.³ ‘The past’ therefore was to that extent regarded as ending with the decease of the Pāṇḍavas, or later with Janamejaya III.

c9.19

² See Table of Genealogies in chap. XII.

c9.20

³ Dynasties of the Kali Age, pp. 1 f. Erroneous account in Hv 191 f.

c9.21

As regards ‘the future’ there are these data. The Vāyu professes to have been narrated in the reign of Asīmakṛṣṇa or Adhisīmakṛṣṇa, the great grandson of that Janamejaya and the sixth in generation from Arjuna, in the Paurava line.⁴ The Matsya (50, 66) takes the same standpoint. Both definitely declare his successors to be future.⁵ Both treat Divākara, king of Ayodhyā, and Senājit, king of Magadha, as reigning contemporaneously with him, and say they were respectively the fifth and seventh in succession from Bṛhadbala and Sahadeva who were killed in the great battle. Both declare their successors to be future, and so also does the Vāyu’s counterpart, the Brahmāṇḍa.¹ These three Puranas thus start the ‘future’ kings in those three great dynasties with the sixth or seventh successors of those who took part in that battle; that is, they make ‘the future’ begin some five or six clear reigns after that battle, or about a century after it, if we put aside the extravagant lengths given to the reigns of Senājit’s predecessors in Magadha.² This point is more definitely discussed in chapter XV.

c9.22

⁴ Vā 1, 12; 99, 255-8. Bd has lost the latter passage in a large lacuna.

c9.23

⁵ Vā 99, 270. Mat 50, 77.

c9.24

¹ Vā 99, 282–3, 300. Mat 271, 5–6, 23. Bd iii, 74, 113–14 as regards Senājit; its reference to Divākara is lost in the lacuna.

c9.25

² And the 60 years assigned to Parīkṣit II, MBh i, 49, 1949.

c9.26

The Viṣṇu and Bhāgavata in their accounts speak of Parīkṣit as reigning and his successors as future in the Paurava line,³ but treat the first kings of Ayodhyā and Magadha after that battle, who were his contemporaries, as future.⁴ These two Puranas thus make ‘the future’ begin some thirty years after the battle as regards Parīkṣit, but immediately after the battle as regards the two other dynasties. The Garuḍa speaks of Janamejaya as reigning in the Paurava line and his successors as future, but apparently treats ‘the future’ in the two other dynasties as beginning after that battle.⁵

c9.27

³ Viṣ iv, 20, 12–13; 21, 1. Bhāg ix, 22, 34–6.

c9.28

⁴ Viṣ iv, 22, 1; 23, 1–3. Bhāg ix, 12, 9; 22, 46.

c9.29

⁵ Gar i, 140, 40; 141, 1, 5, 9.

c9.30

As regards ‘the future’, then, these statements offer two limits of commencement, an upper, the end of the Bhārata battle, and a lower, about a century later.⁶ Everything prior to the former was ‘past’, everything posterior to the latter was reckoned as definitely ‘future’, and the interval between them was intermediate, regarded sometimes as ‘past’ and sometimes as ‘future’. The Matsya, Vāyu and Brahmāṇḍa contain the undoubtedly oldest account of the kings of the Kali age and assert the lower limit. The Bhaviṣyat and Bhaviṣya, by including the story of Kṛṣṇa’s son Sāmba, took the earlier limit. The Viṣṇu, Garuḍa and Bhāgavata, which were later than the Bhaviṣyat, practically adopted its view. Āpastamba’s citation of the Bhaviṣyat is perfectly compatible with either reckoning, for it merely indicates that ‘the future’ had already begun before the third century B.C.

c9.31

⁶ Further remarks about the Kali age, chap. XV.

c9.32

The reckoning then was this. ‘The past’ ended and ‘the future’

c9.33

began at the close of the Bhārata battle, or at the latest about a century afterwards.¹ The difference is not material for the present purpose. The transition manifestly implies a definite stage in the position of tradition, and indicates that previous tradition must have been collected then and formed into a definite compilation, which closed ‘the past’. That would have been the precise consequence of the formation of the original Purana. Tradition says that the original Purana was composed about that time (p. 21).² The two therefore agree and the former corroborates the latter. True Puranas multiplied, the Bhaviṣyat was devised, and the Puranic brahmans had ample time to begin incorporating brahmanic matter, before Āpastamba’s date.

c9.34

¹ Estimated at 950 and 840 B.C. respectively, chap. XV.

c9.35

² Vyāsa may have begun it, and Romaharṣaṇa and his disciples would have completed it.

c9.36

These conclusions are corroborated by certain statements in the Kauṭilīya Arthaśāstra, which may next be considered. It belongs to the fourth century B.C., about a century earlier than the latest date for Āpastamba.

c9.37

Kauṭilya says, ‘The three Vedas, the Sāman, Ṛc and Yajus, are the threefold (scripture). The Atharva-veda and the Itihāsa-veda are also Vedas’.^3 He calls the Itihāsa a Veda and puts it on the same footing as the Atharva-veda. Clearly therefore the Itihāsa was something as definite and well known as that Veda. He also defines the Itihāsa thus—‘Itihāsa means the Purāṇa, Itivṛtta (history), Ākhyāyikā (tale), Udāharaṇa (illustrative story), the Dharmaśāstra and the Arthaśāstra’.^4 All these terms are obviously generic,⁵ and Purana here means Puranas. As the Itihāsa was a Veda and definite, its component parts cannot have been indefinite, hence the Puranas were not an indefinite collection of ancient tales but must have been compositions certain and well established in character then. This is corroborated by another passage which says that a minister skilled in the Arthaśāstra should admonish a king, who is led astray, by means of the Itivṛtta and Purāṇa.⁶ Here also the terms are generic, and the serious purpose for which the Puranas were to be used shows that they were not mere ancient tales but were definite and instructive compositions.

c9.38

³ Book i, chap. 3 (p. 7).

c9.39

⁴ Book i, chap. 5 (p. 10).

c9.40

⁵ See JRAS, 1914, p. 1022.

c9.41

⁶ Book v, chap. 6 (p. 255, lines 1, 2)

c9.42

Kauṭilya enjoins that a prince should spend the afternoon in listening to the Itihāsa;¹ and in order to hear the Puranas the prince would need some one to recite them. Hence among the officials whom a king should retain with salaries are mentioned ‘the Paurāṇika, the sūta and the māgadha’.² The Paurāṇika here is manifestly one specially conversant with the Puranas; and he is distinguished from the sūta and māgadha. He is also referred to in another passage in a similar setting, apart from them,³ and separate. Kauṭilya uses the word Paurāṇika also to distinguish the Puranic sūta and māgadha from the two castes of mixed origin who were so named, but the former had ceased to exist then (p. 18). Hence the Puranas were old in his time. The three first mentioned, the Paurāṇika, the sūta and the māgadha, were quite different. Of these the latter two mean the mixed castes of sūtas and māgadhas who had succeeded to some at least of the functions of the ancient sūtas and māgadhas;⁴ and the Paurāṇika was the person, whether brahman or other, who made the Puranas his speciality. His office proves that the Puranas were well known and established compositions in the fourth century B. C.; and the fact that the original sūta and māgadha were only known then from them shows also that the Puranas went back a considerable time before that century.

c9.43

¹ Book i, chap. 5 (p. 10, line 15).

c9.44

² Book v, chap. 3 (p. 245). Cf. MBh xii, 85, 3203.

c9.45

³ Book xiii, chap. 1 (p. 393).

c9.46

⁴ Were they employed for ākhyāyikās and udāharaṇas?

c9.47

It thus appears from Kauṭilya that Puranas, definite works, existed at least as early as the fourth century B. C., possessed an authoritative position, and were not novel works then, but went back a long time previously as the Puranic sūta had completely disappeared.

c9.48

The Purana was regarded with high respect even by the brahmans who upheld the Vedas specially. Thus the Atharva-veda says⁵—The ṛcs, and the sāmans, the metres, the Purāṇa, together with the Yajus, all gods in the heavens, founded upon heaven, were born of the ucchiṣṭa. The Chāndogya Upaniṣad says⁶—The (hymns of the) Atharvāngiras are the bees, the Itihāsa-Purāṇa is the flower : and this simile, as expressed, implies that those hymns drew their sustenance from the Itihāsa and Purana, which must therefore have been ancient like those hymns. Both these passages imply that the Purana was something definite, like the other compositions mentioned, and was not a novel thing then.

c9.49

⁵ xi, 7, 24. SBE xlii, 229. Cf. MBh viii, 34, 1498.

c9.50

⁶ iii, 4, 1. SBE i, 39.

c9.51

Further the Śatapatha Brāhmaṇa calls the Itihāsa-Purāṇa and certain other compositions ‘honey-offerings to the gods’, and commends their daily study.¹ It also appoints the Itihāsa and Purana for recitation by the priests, calling each a Veda.² These passages show that the Itihāsa and Purana were definite compositions. Similarly the Śāṅkhāyana Śrauta Sūtra and Āśvalāyana Sūtra³ say—On the eighth day he tells the story which begins with Matsya Sāmmada... He then says, ‘The Itihāsa-veda is the Veda, this is the Veda,’ and recites an Itihāsa. On the ninth day he tells the story which begins with Tārkṣya Vaipaśyata... He then says, ‘The Purāṇa-veda is the Veda, this is the Veda,’ and recites the Purāṇa.⁵

c9.52

¹ xi, 5, 6, 8. SBE xliv, 98. Weber, Hist. of Indian Lit., 93.

c9.53

² xiii, 4, 3, 12–13. SBE xliv, 369.

c9.54

³ Śāṅkh xvi, 1. Āśval x, 7, inverting the procedure of the two days; SBE xliv, 369, note 3. Max Müller, Sanskrit Lit., 37, 40.

c9.55

⁵ The Commentator on the Śāṅkhāyana notes, The Purana uttered by Vāyu should be narrated here. Both Vā and Bḍ were uttered by Vāyu, and were one originally.

c9.56

The story above mentioned that begins Matsyaḥ Sāmmadaḥ⁶ appears to be that told in Viṣṇu iv, 2, 19 f. and Padma vi, 232, 33 f. about the rishi Saubhari. While practising long austerities he saw a fish named Sammada or Sāmamada joyous with an immense and happy family,⁷ and aroused thereby, he married the fifty daughters of Māndhātṛ, king of Ayodhyā, and maintained them in great magnificence and happiness.⁸ That is probably what the above brahmanical passage cited. What is there called an itihāsa is found in two Puranas now. I have not found the story about Tārkṣya Vaipaśyata, but it was a story about birds.⁹

c9.57

⁴ The Āśval. Sūtra reads Vaipaścita (better ?).

c9.58

⁶ Rigv viii, 67 is attributed in the alternative to Matsya Sāmmada; Anukramaṇī and Vedārth.

c9.59

⁷ Śatapatha Brāhm (xiii, 4, 3, 12) says, King Matsya Sāmmada and his people were water-dwellers. This cannot refer to the Matsya country and people, for their country was anything but watery; so matsya must mean ‘fish’. Vedārth on Rigv viii, 67 says he was a fish.

c9.60

⁸ This story is further considered in chap. V.

c9.61

⁹ So Śatapatha Brāhm, loc. cit.

c9.62

As pointed out above (p. 54), a collection of tradition must have been made within a century or so after the Bhārata battle, thus closing the ‘past’ and its traditions, whence all subsequent occurrences belonged to the ‘future’. This conclusion is confirmed by a general survey of tradition in the Puranas.

c9.63

There is much traditional history including fairly copious genealogies down to the time of that battle and the death of Kṛṣṇa and the Pāṇḍavas, and then all the genealogies stop short except those of the three great kingdoms of Hastināpura, Ayodhyā and Magadha, although other old dynasties continued to exist, such as those of Pañcāla, Kāśi, Mithilā, &c.¹ There is a little historical tradition of the century or so that followed the battle, yet only concerning the first five Paurava kings in the first of those three kingdoms, and nothing about the two other kingdoms. After that century or so there is no historical tradition, and the genealogies of those three kingdoms are given in prophetic form, but were manifestly compiled long afterwards out of Prakrit chronicles.² Yet there were traditions about those ‘future’ kings, as, for instance, about the kings in Buddha’s time and about king Udayana of Vatsa,³ and none such are noticed in the Puranas, as far as I am aware. These facts, much traditional history down to the death of the Pāṇḍavas, a very little for a century or so following, and then none whatever in the Puranas, prove that there must have been a closing stage in tradition during that century or so—that is, that the original Purana must have been compiled about that time.

c9.64

¹ My Dynasties of the Kali Age, p.

c9.65

The absolute dearth of traditional history after that stage is quite intelligible, both because the compilation of the Purana had set a seal on tradition, and because the Purana soon passed into the hands of brahmans, who preserved what they had received, but with the brahmanic lack of the historical sense added nothing about later kings—just as the Bhāgavata Purana, which was composed about the ninth century A.D., added nothing to its account of the kings of the Kali age beyond where the Vāyu stopped some four centuries earlier. With the same lack however they have introduced in their own additions to the Puranas notices of brahmans who were later, such as Āsuri, Pañcaśikha, &c.⁴ This marked change at that stage betokens the compilation of the original Purana and the change in the custody of tradition from the sūtas to the brahmans. It is clear evidence that ancient traditions were collected while they were still well known, and while there was still a class of men whose business it was to preserve them carefully.

c9.66

²³ See chap. XV. ² Id. p. 10 and Appendix I. ³ Meghadūta i, 31. ⁴ Discussed in chap. XXVII.

c9.67

The Brāhmaṇas are in accord with this conclusion. The late Śatapatha has far more tradition than the earliest Brāhmaṇas. They were composed while the Purana or Puranas were in their infancy,¹ but it was composed after the Puranas had become established, and by quoting their tradition shows that they existed and that the recluse brahmans who studied the Vedas had at length become aware to some extent of the contents and importance of the Puranas.

c9.68

¹ See chap. XXVII.

Linked texts (23)