into three distinct classes, adapted to the capabilities of his hearers.* 1st, The Vinaya, or "religious discipline," addressed to the Sráwaka, or auditors," who were the lowest class of the Bauddha community; 2nd, The Sutra, " aphorisms," or Principles of Faith, addressed to the Pratyekas or "distinct intelligences," who formed the middle class of monks; and 3rd, the Abhidharma, or 66 supreme law," or Transcendental Principles of Faith, imparted only to the Bodhisatwas, or "true intelligences," who were the highest class of the Bauddha community.

21. These three classes of doctrine are collectively called the Tri-Pitaka, or "three repositories ;" and the Tri Yanika,† or "three-means-of-progression;" and separately they are generally known as the LOWEST, the MIDDLE, and the HIGHEST means of ADVANCEMENT. These terms are of common occurrence in the Buddhist writings, and especially in the works of the Chinese travellers; from whom we learn that the pastoral nations of the Northern Hills, accustomed to active habits, were content with the LOWEST-MEANS-OF-ADVANCEMENT, while the more intellectual and contemplative people of India generally strove for the attainment of the superior degrees of Madhyima-Yánika, and Mahá

*Fo-kwe-ki, c. 2, note. Csoma de Koros, -Analysis of the Tibetan Works, in Prinsep's Journal, vol. vii. p. 145.

+ Yána means a vehicle of any kind, by which progress or advance is obtained.

>

Yánika, or "MIDDLE AND HIGHEST MEANS-OFADVANCEMENT."

22. The Tri-Pitaka were compiled immediately after Sákya's death, in B. c. 543, by three of his chief disciples, with the assistance of five hundred learned monks. The ABHIDHARMA was the work of Kasyapa, the head of the Bauddha fraternity; the SUTRA of Anánda, Sákya's favourite disciple; and the VINÁYA of Upáli.* The language in which these works was written, has been the subject of much dispute; but the account given by the Tibetans† is so probable, and at the same time so natural, that it ought, as James Prinsep has observed, to set the matter at rest. Their account is that the SUTRAS in general, that is, the Vináya, as well as the Sutrá proper, were first written in the SINDHU language; but that the whole of the Sher-chin, that is, the Prajná Páramitá, or "transcendental wisdom," and the whole of the Gyud, that is, the Tantras, or religious mysticism," were composed in SANSKRIT. This appears to be the only conclusion that anyone can come to who examines the subject attentively. For the Vinaya and Sutra, which were addressed to the people at large, as well as to the Sránakas and Pratyekas, must necessarily have been published in the vernacular language of the country; while

Prinsep's Journal, vol. i. p. 2; and Transactions As. Soc. Bengal, vol. xx. p. 42.

+ Csoma's Index to the Kahgyur, in Prinsep's Journal, vol. vi. See Csoma, in Prinsep's Journal, vol. vi. p. 503.

p. 688.

the abstruse and metaphysical philosophy of the Abhidharma, which was addressed solely to the learned, that is, to the Bráhmans and Bodhisatwas, would, without doubt, have been enunciated in Sanskrit, for the simple reason that its refined elegance of ideas, and delicate shades of meaning, could not be adequately expressed in any of the vernacular languages. The Tantras are of much later date; but the same reasoning holds equally good for them; as the esoteric mysticism of their doctrines could only have been expressed in Sanskrit. In a few words the speculative principles of Buddhism were expounded and recorded in Sanskrit,* while the practical system of belief, deduced from those principles, was spread abroad and propagated by means of the vernacular Prakrit.

23. In the Rúpasiddhi, which is the oldest Páli grammar now extant, and which the author Buddhapriya compiled from the ancient work of Kachháyana, a quotation from the latter is given, apparently in the original words. According to this account, Kachháyana was one of the principal disciples of Sákya, by whom he was selected for the important office of compiling the first Páli grammar, the rules of which are said to have been propounded by Tathagata himself. This statement seems highly probable; for the teacher must have soon found the

* See also Hodgson's opinion on this point. Prinsep's Journal, vol. vi. p. 683.

+ Turnour's Introduction to the Mahawanso, p. 26.

difficulty of making himself clearly understood when each petty district had a provincial dialect of its own, unsettled both in its spelling and its pro

nunciation.

24. A difficulty of this kind could only be overcome by the publication of some established rules of speech, which should fix the wavering pronunciation and loose orthography of a common language. This was accomplished by the Páli Grammar of Kachháyana, compiled under Sákya's instructions; and the language, thus firmly established, was used throughout India by the Buddhist teachers, for the promulgation and extension of the practical doctrines of their faith.

25. In the Buddhist works of Ceylon, this language is expressly called Mágadhi, or the speech of Magadha; and as this district was the principal scene of Sákya's labours, as well as the native country of himself and of his principal disciples, the selection of Mágadhi for the publication of his doctrines was both natural and obvious. It is true, as Professor H. H. Wilson has remarked,* that there are several differences between the language of existing Buddhist inscriptions and the Mágadhi of Páli Grammars; but these differences are not such as to render them unintelligible to those whom PRIYADARSI addressed in his Pillar edicts in the middle of the third century before Christ. The Pro

• Journal of the Royal Asiatic Society, vol. xii. p. 238.

fessor admits that the Páli was most likely selected for his edicts by Priyadarsi, "that they might be intelligible to the people;" but he is of opinion that the language of the inscriptions was rather the common tongue of the inhabitants of Upper India than a form of speech peculiar to a class of religionists; and he argues that the use of the Páli language in the inscription is not a conclusive proof of their Buddhistical origin.

26. The conclusion which I have come to is exactly the reverse; for it is a well known fact, that the Brahmans have never used any never used any language but Sanskrit for their religious writings, and have stigmatised the Mágadhi as the speech of men of low tribes.* In their dramas also the heroes and the Brahmans always speak Sanskrit, while the use of Mágadhi is confined to the attendants of royalty. Professor Wilson has, however, identified the Mágadhi with Prákrit, the use of which, though more honourable, was still confined to the principal female characters; but the extensive employment, in the dramatic works of the Brahmans, of various dialects, all derived from one common stock, seems to me to prove that they were the vernacular language of the people. In this vernacular language, whatever it was, whether the high Prákrit of the Saurasenas, or the low Prákrit of the Mágadhas, we know certainly that the Vinaya and Sutra, or the practical doctrines

* Colebrooke, in Trans. As. Soc. Bengal, vol. vii. p. 199. Wilson's Hindu Theatre, vol. i. p. lxiii. iv.