The colophon of the manuscript, which is added above (in its place) at the end of the Hymn, perhaps deserves a little more attention than I have given it. It seems to bear the marks of very negligent copying and abbreviating from an older manuscript; with inadvertent omissions, and some supplementing by the copyist. Translated literally, keeping the order of words, it runs thus: "And is finished by my hands this book blessed and blessing Heziran who by his name [is, or, am] Deacon fool Baryachâ Jalûiâ who [is] from Bar Kazi who [is] from Qeritha [i. e., village] Nahra." Whether the "blessed and blessing" refer, as usual in subscriptions, to some place of writing [omitted by the copyist]; or, as most natural to the construction as the words stand, to the book; or, as is frequent, to the month; is a question that suggests itself at the start. I incline to the first supposition of the three. The omission of the preposition before "Hezîran," as well as the misapplied plural points over the word, suggest that a full date was present in the archetype, and that in the date occurred the phrase "in the month Heziran." The latter part of the colophon, after the diamond punctuation-mark, probably refers to the latter copyist.

THE SOCIAL AND MILITARY POSITION

OF THE

RULING CASTE IN ANCIENT INDIA,

AS REPRESENTED BY THE SANSKRIT EPIC.

BY EDWARD W. HOPKINS,

PROFESSOR IN BRYN MAWR COLLEGE, BRYN MAWR, PA.

PREFACE.

THIS essay, in its original form, was read before the Oriental Society in May, 1886. Further contributions to the subject, made as reported in the subsequent Proceedings of the Society, have now been incorporated into the work, and the point of view of the whole somewhat extended.

My first intention was to record the data furnished by the Mahabharata in regard to the Warrior-caste. I have since been led to add matter illustrative of my topic from works more or less parallel to the Epic, and this paper now offers de entón an inquiry into the conditions of civilization in the Middle Ages of India from the point of view of the rulingpower. Into wider questions of pan-Aryan interest I have through lack of space refrained from entering: for example, into that of land-ownership and village communities, where a new and thorough investigation of India's position is needed. I believe no especially Epic study of Hindu civilization has yet been attempted. My authorities are, therefore, chiefly the native texts.*

*A study of the Vedic period is presented by Zimmer's Altindisches Leben. Weber's Collectanea (Indische Studien, vol. x.), Lassen's Indische Alterthumskunde, and Müller's India touch on some of the points here discussed. Of Wilson's Art of War and RajendralalaMitra's Indo-Aryans I shall speak more particularly below. On Epic antiquities Muir has some scattered remarks and a few special studies in his Sanskrit Texts. To these general acknowledgment is due. The term Epic I limit, for convenience, to the Mahabhārata, although,

The following abbreviations require explanation. Unprefixed numerical references imply that the quotation is from the Mahabharata, Bombay edition.*

=

=

R. Rāmāyaṇa, ed. Gorresio; M. Manu's law-book (manavadharmaçãstra); G.=Gautama's law-book (dharmaçāstra), ed. Stenzler; Vas. Vasistha's law-book (id.), ed. Führer; Ap. = Apastamba's law-book (dharmasutra), ed. Bühler; B. = translation of Baudhayana by Bühler. V. P. and Ag. P. denote respectively the Vishnu and Agni Puranas. The names of other Purāņas and the authors of the House-laws (grhyasūtra) are when quoted given in full.

I. INTRODUCTION. ORIGIN OF THE EPIC.

In order to a better understanding of the material from which are drawn the chief quotations preferred in this essay, a word will be necessary in regard to the present and past condition of the Hindu Epic. The poem is of obscure origin. History fails us, and who can trust Hindu tradition? More than this: the work when analyzed appears to be inwardly inconsistent. In the same heroes we discover different characters. Opposite tendencies seem at work. The highest god is at the same time a tricky mortal. The chief knights are depicted now as good and now as sinful men. The original theme is, as it were, diverted from its course.†

conversely, this Epic is regarded by native authorities as an ‘art-poem,' as was long ago pointed out by Müller (tyayā ca kávyam ity uktam tasmāt kāvyam bhaviṣyati, Mbh. i. 1.72; Müller, Ancient Sanskrit Literature, p. 41). The part of this Epic embraced by the twelfth and thirteenth books I call pseudo-Epic. In regard to the origin of the Mahabharata I have briefly discussed Holtzmann's general argument (Epos) in my Introduction, but ignored his speculations on Epic Buddhism: in respect of which I can say only that they fail to convince me of his demonstrandum. On the important subject of the critique of our received text of the poem I have not touched in this essay. [See now a paper thereon by the writer reported in the Proceedings for October, 1888.]

*A Pathfinder or concordance of references for the Bombay and Calcutta editions equated by verse-decades has been prepared by the writer, and will soon be published.

An abstract will indicate this. Of two possible heirs to the throne of Hastinapura, Pandu, the younger, having succeeded to the sovereignty on account of the blindness of his elder brother, Dhritarashtra, finally grew weary of ruling, and, retiring into the woods, where he died, left his kingdom to the blind Dhritarashtra. The latter, regarding Yudhishthira, Pandu's eldest son, as rightful heir, caused him at first to be proclaimed crown prince; but subsequently, persuaded by Duryodhana and others of his sons, reconsidered the matter, yielded to sin, permitted Yudhishthira with his four brothers (called the Pandus, as opposed to the Kurus, Duryodhana and his brothers) to be enticed out of the city, and then settled the whole kingdom on his own son. But the Pandus, at first expelled and in mortal danger, after proceeding to Panchala and forming an alliance with that king by a polyandrous marriage with his daughter, returned to Hastinapura, backed by Pan

From outside sources we know only that the poem is mentioned in the Sutra of Açvalayana, and seems to be intended in a description of a Hindu epic given by Dio Chrysostomos, in a fragment that may have come from Megasthenes. In the event of the description being original with the first, 100 A. D. may be set as the date of this information; with the second, 400 B. C.* What other accounts we have are not less doubtful in date. Thus, the poem is known to the Mahabhaṣya; but the earliest date of this work is 140 B. C.,+ while Panini's evidence is negative, mentioning characters but not the poem by name. Of the war, only the Epic gives an account, and the date of the conflict is matter of inference. Thus, Schroeder reckons that it antedates the Yajur-Veda, because the Kuru-Panchala alliance therein recorded must have been the result of the war; but this is absolutely uncertain. Analysis led Lassen to suppose that the original poem was an account of a war between Kurus and Panchalas, not between Kurus and Pandus. There is no very weighty reason for the view thus expressed. The poem itself asserts that its theme is the Kuru-Pandu war. Objections offered to believing this are based on the fact that the Kurus are an old family, known in more ancient literature, while the Pandus are not. The working-over of the poem is also thought to be attested by the fact that its introductory part states it to have had different beginnings and different lengths8800 couplets, 24,000, and 100,000; but as, aside from other proofs of recent time, it is evident that the last length could not have been noted till the work had been completed, this whole

chala's influence, made terms with their relatives, and took half the realm. In a corner of this they founded and occupied a new town, Indraprastha; and here, after years of conquests, they held a celebration that awakened the envy of Duryodhana, who soon challenged Yudhishthira to a deceitful game of dice. In its course the latter played away his newly acquired greatness, and then gambled again with the understanding that the loser should this time become a hermit. He lost, went into the woods with his brothers, and remained there in accordance with his promise for twelve years. At the close of one further year he found an ally, invoked anew the aid of Panchāla, elected Krishna (Vishnu) as his aid, marched against Hastinapura with a large force, and routed the yet larger army of Duryodhana by means of desperate and unscrupulous fighting on the part of the Pandu knights and the unfairly used influence of Krishna (whose help the Kurus had scorned). He found no one to oppose him within the town, and had himself crowned king of both Hastinapura and Indraprastha; and finally, after a long reign, laid down the crown in order to climb up to heaven in company with his four brothers and the family wife: the successful accomplishment of this journey terminating the story. * The different views on this subject have lately been set forth by L. von Schroeder, in his Indiens Literatur und Cultur, p. 464. Weber thinks the mention in Ãçvalāyana an interpolation. Compare Lassen, Ind. Alt., i. 589-592.

Weber, Lit., p. 201, 241.

statement can only be regarded as one of comparatively late origin, belonging to the final development of the Epic-a time when the writers knew little in regard to the working-over of their inherited verses. At present the text is overburdened with extraneous matter, tales, laws, moral codes, theologies, metaphysics, quite stifling the original body of living poetry.

From another point of view, efforts have been made to prove not only a change, but a complete inversion (in our present story) of the original theme. This criticism bases itself on the want of unity in the characters. Starting with the two-fold nature of Krishna-Vishnu as man and god,* and with the glossed-over sins of the Pandus, the critic argues that the first poem was written for the glory of the Kurus, and subsequently tampered with to magnify the Pandus; and that in this latter form we have our present Epic, dating from before the fourth century B. C.; since the worship of Vishnu was in Megasthenes time triumphant over that of Brahma, and it is with the cult of the former god that the Pandus are bound up. The first poem would thus be completely changed, or, as Schroeder in describing the theory says, 'set upon its head.' Schroeder's exposition of the theory, being the latest outcome of this criticism (we are indebted to Adolph Holtzmann for its tone), will serve as at once the clearest and most recent explanation of how the Epic may thus have been inverted. The original poet (he says in substance) lived at a time when Brahma was the highest god (700 to 500 or 400 B. C.); and this singer was a child of the Kuru-land. He heard reports of the celebrated Kuru race that once reigned in his land, but had been destroyed by the dishonorable fighting of a strange race of invaders. This tragical overthrow he depicted in such a way as to make his native heroes models of knightly virtue, while he painted the victors (Pandus, Panchalas, Matsyas), with Krishna, hero of the Yadavas, at their head, as ignoble and shamefully victorious. This is the old Bharata song mentioned in Açvalayana. After a time Krishna became a god, and his priests, supported by the Pandus, sought to make Krishna (Vishnu) worthy to be set against Buddha. Their exertions were successful. Vishnu in the fourth century became the great god, and his grateful priests rewarded their helpers, the Pandus, by taking the Bharata poem in hand and making a complete change in the story, so as to relieve them of the reproaches of the old poet. Finally they worked it into such shape that it praised the Pandus and blamed their opponents. About this time they inserted all the episodes that glorify Vishnu as the

* On Krishna as shepherd, see Lassen, Ind. Alt. i. 770.

+ Lit. u. Cult., p. 479.

1