of stone, on rocks, on coins, deciphered only of late years, through the extraordinary ingenuity and perseverance of Mr. James Prinsep, have verified the names of races and titles of princes-the Gupta and Andhra Rajas, mentioned in the Puráñas-and have placed beyond dispute the identity of Chandragupta and Sandrocoptus; thus giving us a fixed point from which to compute the date of other persons and events. Thus, the Vishnu Purána specifies the interval between Chandragupta and the Great War to be eleven hundred years; and the occurence of the latter little more than fourteen centuries B. C., as shown in my observations on the passage,' remarkably concurs with inferences of the like date from different premises. The historical notices that then follow are considerably confused; but they probably afford an accurate picture of the political distractions of India at the time when they were written: and much of the perplexity arises from the corrupt state of the manuscripts, the obscure brevity of the record, and our total want of the means of collateral illustration.

The fifth book of the Vishnu Purána is exclusively occupied with the life of Krishna. This is one of the distinguishing characteristics of the Purána, and is one argument against its antiquity. It is possible, though not yet proved, that Krishna, as an Avatára of Vishnu, is mentioned in an indisputably genuine text of the Vedas. He is conspicuously prominent in the Mahábhárata, but very contradictorily described there. The part that he usually performs is that of a mere mortal;

1 See Book IV., Chapter XXIV.

?

although the passages are numerous that attach divinity to his person. There are, however, no descriptions, in the Mahábhárata, of his juvenile frolics, of his sports in Vrindavana, his pastimes with the cow-boys, or even his destruction of the Asuras sent to kill him. These stories have, all, a modern complexion; they do not harmonize with the tone of the ancient legends, which is, generally, grave, and, sometimes, majestic. They are the creations of a puerile taste and grovelling imagination. These chapters of the Vishnu Purána offer some difficulties as to their originality. They are the same as those on the same subject in the Brahma Puráňa: they are not very dissimilar to those of the Bhagavata. The latter has some incidents which the Vishnu has not, and may, therefore, be thought to have improved upon the prior narrative of the latter. On the other hand, abridgment is equally a proof of posteriority as amplification. The simpler style of the Vishnu Purána is, however, in favour of its priority; and the miscellaneous composition of the Brahma Purána renders it likely to have borrowed these chapters from the Vishnu. The life of Krishna in the Hari Vamsa and the Brahma Vaivarta are, indisputably, of later date.

The last book contains an account of the dissolution of the world, in both its major and minor cataclysms; and, in the particulars of the end of all things by fire and water, as well as in the principle of their perpetual renovation, presents a faithful exhibition of opinions that were general in the ancient world. The meta

1

1 Dr. Thomas Burnet has collected the opinions of the ancient world on this subject, tracing them, as he says, "to the earliest

physical annihilation of the universe, by the release of the spirit from bodily existence, offers, as already remarked, other analogies to doctrines and practices taught by Pythagoras and Plato, and by the Platonic Christians of later days.

The Vishnu Purána has kept very clear of particulars from which an approximation to its date may be conjectured. No place is described of which the sacredness has any known limit, nor any work cited of probable recent composition. The Vedas, the Puráñas, other works forming the body of Sanskrit literature, are named; and so is the Mahábhárata, to which, therefore, it is subsequent. Both Bauddhas and Jainas are adverted to. It was, therefore, written before the former had disappeared. But they existed, in some parts of India, as late as the twelfth century, at least; and it is probable that the Purána was compiled before that period. The Gupta kings reigned in the seventh century.* The historical record of the Purána which mentions them was, therefore, later: and there seems little doubt that the same alludes to the first incursions of the Mohammedans, which took place in the eighth century; which brings it still lower. In describing the latter dynasties, some, if not all, of which were, no doubt, contemporary, they are described as reigning,

people, and the first appearances of wisdom after the Flood." Sacred Theory of the Earth, Book III., Chapter III. The Hindu account explains what is imperfect or contradictory in ancient tradition, as handed down from other and less carefully perpetuated sources.

* More recent researches have rendered this conclusion doubtful.

altogether, one thousand seven hundred and ninety-six years. Why this duration should have been chosen does not appear; unless, in conjunction with the number

of

years which are said to have elapsed between the Great War and the last of the Andhra dynasty, which preceded these different races, and which amounted to two thousand three hundred and fifty, the compiler was influenced by the actual date at which he wrote. The aggregate of the two periods would be the Kali year 4146, equivalent to A. D. 1045. There are some variety and indistinctness in the enumeration of the periods which compose this total: but the date which results from it is not unlikely to be an approximation to that of the Vishnu Puráňa.

It is the boast of inductive philosophy, that it draws its conclusions from the careful observation and accumulation of facts; and it is, equally, the business of all philosophical research to determine its facts before it ventures upon speculation. This procedure has not been observed in the investigation of the mythology and traditions of the Hindus. Impatience to generalize has availed itself greedily of whatever promised to afford materials for generalization; and the most erroneous views have been confidently advocated, because the guides to which their authors trusted were ignorant or insufficient. The information gleaned by Sir William Jones was gathered in an early season of Sanskrit study, before the field was cultivated. The same may be said of the writings of Paolino da S. Bartolomeo,' with the further disadvantage of his having

been imperfectly acquainted with the Sanskrit language and literature, and his veiling his deficiencies under loftiness of pretension and a prodigal display of misapplied erudition. The documents to which Wilford1 trusted proved to be, in great part, fabrications, and, where genuine, were mixed up with so much loose and unauthenticated matter, and so overwhelmed with extravagance of speculation, that his citations need to be carefully and skilfully sifted, before they can be serviceably employed. The descriptions of Ward' are too deeply tinctured by his prejudices to be implicitly confided in; and they are also derived, in a great measure, from the oral or written communications of Bengali pandits, who are not, in general, very deeply read in the authorities of their mythology. The accounts of Polier3 were, in like manner, collected from questionable sources; and his Mythologie des Indous presents an heterogeneous mixture of popular and Pauránik tales, of ancient traditions, and legends apparently invented for the occasion, which renders the publication worse than useless, except in the hands of those who can distinguish the pure metal from the alloy. Such are the authorities to which Maurice, Faber, and Creuzer have exclusively trusted, in their description of the Hindu mythology; and it is no marvel that there should have been an utter confounding of good and bad in their selection of materials, and an inextricable

1 Asiatic Researches.

2 View of the History, Literature, and Religion of the Hindoos, with a Description of their Manners and Customs.

3

Mythologie des Indous, edited by la Chanoinesse de Polier.