Introduction

Source

James D’Alwis, An Introduction to Kachchāyana’s Grammar of the Pāli Language; with an Introduction, Appendix, Notes, &c., Colombo: Williams and Norgate, 1863. Public domain by age. Reproduced from a scan of the University of California copy, available here as a PDF (15 MB), as a reading edition: the text was recognised from the page images and then proofread against them page by page. The print’s spelling and its accents are kept as printed (long vowels carry a grave accent, Pàli, Kachchàyana, and in places a circumflex; the niggahīta is written n), and so are the author’s capitals and small capitals; quotation marks are normalised to straight quotes, and the italics of Pāli words in the print are not reproduced, except in the Corrections, where an italic letter marks a lingual or a long vowel and is kept. Footnotes are numbered through the book; the print’s symbols (* † ‡ § ‖ ¶) are replaced by these numbers. Sinhalese characters are shown as [Sinhalese]. The printed page numbers are kept as hidden comments of the form <!-- p. xvii (PDF p. 21) -->. A word that could not be read is followed by [?].

This page is one part of D’Alwis’s book, reproduced as printed. The front matter and the contents are on the main page.

THERE is hardly a country on the face of the Globe which presents greater facilities for acquiring a knowledge of the Pali, than Ceylon; and, perhaps, no nation possesses greater advantages for its study than the Sinhalese. Pali, like the Sanskrit and the Sinhalese, forms a necessary part of the course of education pursued by the natives.¹ Our Alphabet is common to these several languages,² and the affinity which the Pali bears to the Sinhalese, both verbally and grammatically, renders its study far more easy to the people of this country than even to the Burmese.

Although the Sinhalese, as a language, has been latterly neglected; the Pali, from its being the dialect in which the Buddhist scriptures are recorded, has always been the principal study of the largest portion of the Ceylonese, who are followers of Buddha. From the period when it became the sacred language of the land, kings and princes have encouraged its study; nobles and statesmen have vied with each other to excel in its composition; and in it laymen and priests have produced some of our most elegant works. The names of Batuvantudàve, Hikkaduve, Lankàgoda, Dodanpahàla, Valàna, Bentota, Kahave, and Sumangala, amongst a host of others, are familiar to Pali scholars, as those of the learned who are even now able to produce compositions³ by no means inferior to those of a Buddhagosa or a Paràkkrama, though, like the modern Sanskrit, certainly more artificial than some of the more ancient writings.

The number of books, too, in the Pali language, is greater than in the Sinhalese; and, though those on Religion far exceed those upon other subjects, it is, nevertheless, a fact, that the Pali literature of the Sinhalese is not deficient in works upon other branches of Oriental Science. It presents indeed a proud array of extensive volumes on Prosody, Rhetoric, Medicine, and History. On Grammar alone there are no less than forty Pali works;⁴ whilst in the Sinhalese there is but one, the solitary Sidat-Sangarà. From the constant study of Pali in the Buddhist monasteries of this island, the books in that language are found to be comparatively free from errors: and it is a well known fact, that the Buddhist priests, unlike the Brahmans, are willing to give Pali scholars, whether Buddhist or Christian, free access to their libraries.

Advantages like these, combined with others, enabled the Hon’ble George Turnour, late Colonial Secretary of Ceylon, to attract the attention of Orientalists to the high claims of the Pali language as existing in Ceylon. In the prosecution of his labours with such a praiseworthy object, he drew attention, in his elaborate Introduction to the Mahàvansa, to some of the Pali works formerly extant in Ceylon, and, amongst them, to Kachchàyana’s Grammar, which he then regarded as extinct. This, in the very outset of my Pali studies, after many years’ devotion to Sinhalese literature, I ascertained to be a mistake;⁵ having added it to my library, in a purchase of Pali books which I had then (1855) recently made from the collection of the late lamented F. D’ Levera, Esq., District Judge of Colombo. Shortly afterwards I communicated the fact to some of my friends in Europe; and the repeated communications which I have received from them, especially from Dr. Rost of Canterbury, urging upon me the necessity for the publication of a Pali Grammar, and expressing a curiosity to examine Kachchàyana, have induced me to publish a Chapter from it, as an Introduction to a fuller translation.

In laying this before the public, I propose to give a brief account of some of the Pali Grammars known in this country, including a notice of the age and author of the work here presented; and also an Essay on the relations of the Pali to the Sanskrit.

The terms Pali and Màgadhì are at the present day indifferently employed in Ceylon, Ava, Siam, and even China, to express the sacred language of the Buddhists; and, being confined to those countries, the term Pali is not met with in any of the Indian writings.

Màgadhì is the correct and original name for the Pali. It was not so called in consequence, as some suppose, of the mission of Asoka, the king of Magadha, to introduce Buddhism into Ceylon.⁶ It had received that name before the age of that monarch,⁷ and was so called after the ancient name of Behar. It was the appellation for the ancient vernacular language of Magadha. It was the designation for the dialect of the Magadhas.—Magadhanan bhàsà Màgadhì.⁸

Pàli is comparatively a modern name for the Màgadhì. It has not originated from ‘the region called Pallistan the (supposed) land of the Pali,—our Palestine.’ It does not come from Palitur in Tyre—the so-called ‘Pali tower or Fort.’ It has no historical connection with ‘the Palatine hills of Rome.’⁹ It was not called after the Pehlve, the dialect of the Sessanian dynasty, nor is it derived from “Palli a village, as we should now-a-days distinguish gunavàri ‘village,’ ‘boorish,’ from Urdu, the language of the Court.”¹⁰ Nor does it indeed mean “root” or “original.”¹¹

Like àli the word pàli originally signified a ‘line,’ ‘row,’ ‘range,’¹² and was gradually extended to mean ‘suttan,’ from its being like a line;¹³ and to signify edicts,¹⁴ or the strings of rules in Buddha’s discourses or doctrines, which are taken from the Suttans.¹⁵ From thence it became an appellation for the text of the Buddhist Scriptures, as in the following passages:—Thereyàchariyà sabbe Pàlin viya Tamaggahun:—‘All the three preceptors held this compilation in the same estimation as the text (of the Pitakattàya).’¹⁶ Thera vàdehi pàlehi padehi vyanjanehicha. ‘In the Thera discourses as in the text (of the Pitakattàya); and in an expression as in a letter.’¹⁷ From thence again Pàli has become the name of the Màgadhì language in which Buddha delivered his doctrines.

The Pali has also received the designation of Tanti, ‘the string of a lute,’¹⁸ its Sanskrit cognate being tantri. From that signification it seems to have been originally applied by the Brahmans to tantru, ‘a religious treatise teaching peculiar and mystical formula and rites for the worship of their deities, or the attainment of super-human power,’ or, ‘that which is comprized of five subjects, the creation and destruction of the world, the worship of the gods, the attainment of all objects, magical rites for the acquirement of six super-human faculties, and four modes of union with the spirit by meditation.’¹⁹ The Magadhas, before their secession from the Brahman religion, probably used the Màgadhì term, tanti in this sense; but when they embraced the Buddhist faith, they used it to signify the doctrines of Gotama, as in the following passages:—(i) Sammà Sambuddho pi te pitakan Buddha vachanan Tantin àropento Màgadhì bàsàyeva aropesi—‘Buddha who rendered his tepitaka words into Tanti (or tantra or doctrines) did so by means of the Màgadhì language’—Vibhanga Atuvà. (ii) Tivagga sangahan chatuttinsa suttanta patimanditan chatu satthi bhànavàra parimànan tantin sangàyetvà ayan dìgha nikàyo nàmà’ti—‘Having rehearsed the Tanti (the doctrines) which contain 64 banavàra embracing 34 Suttans composed of 3 classes, (this was) named Dìghanikàya’—Bodhivansa.

From its application to the Buddhist doctrines, Tanti has become a name for the sacred language itself of the Buddhists—viz., the Màgadhì or Pàli. Thus in Buddhagosa’s Atthakathà:—‘Why was the first convocation held? In order that the nidànan of the Vinaya pitaka, the merits of which are conveyed in the Tanti (Pàli) language, might be illustrated.’²⁰ Thus also, in the Bàlavatàra, in a part of the passage which answers to §58 in the Rev. B. Clough’s version, where it is left untranslated:—

Eva maññà pi viññeyyà
Sanhità tanti yà hità;
Sanhità chita vannànan
Sannidha’byava dhànato.

That is to say, ‘In this wise know the rest of the combinations which are susceptible in the Tanti (language.)—Sanhita is the combination of letters without a hiatus.’

For the elucidation of the grammar of this language there are three schools; or, in other words, all Pali Grammars extant in Ceylon may be divided into three classes, viz., (1) Saddanìti; (2) Moggallàyana; and (3) Kachchàyana.

1. There are but few treatises which come under the first.

2. Under the second head there are several, all which have been written upon the principles laid down by Moggallàyana, the writer of Abhidhànapadìpikà. Owing to the omission of the Introduction and Conclusion of that work in the edition published by the Rev. B. Clough, oriental scholars have expressed various conjectures as to its date. As a help, however, to those who may be engaged in antiquarian researches, and with a view to fix the date of Moggallàyana, the omissions are here supplied.

ABHIDHANAPADIPIKA.

1. Tathàgato yo karunà karo karo
’Payàtamossajja sukhap padan padan
Akà paratthan kalisam bhave bhave
Namàmi tan kevala duk-karan karan

2. Apùjayun yam muni kunjarà jarà
‘Rujàdimuttà yahimuttare tare
Thità tivattambu nidhin narà’narà
Tarinsu tan dhamama’ maghà pahan’ pahan

3. Gatan munindo’ rasasûnutan nutan
Supuññakhettan bhuvane’sutan sutan
Ganampi pànì kata sanvaran varan
Sadà guno’ ghena nirantaran taran

4. Nàma lingesu kossallam
’Attha nichchhaya kàranam
Yato mahabbalan Buddha
Vachane pàta vatthinam.

5. Nàmalingàn’ yato Buddha
Bhàsitassà’ rahà n’aham
Dassayanto pakàsissam
Abhidhàna’ padìpikan.

6. Bhìyo rùpan tarà sàha
‘Chariyena cha katthachi
Kvachà’ hachcha vidhànena
Neyyan thìpun napunsakan.

7. Abhinna lingìnan yeva
Dvando cha linga vàchakà
Gàthà pàdanta majjhatthà
Pubban yantya’pare paran.

8. Pumitthiyan padan dvìsu
Sabba linge cha tìsviti
Abhidhànan tarà rambhe
Neyyan tvanta mathàdi cha.

9. Bhìyo payoga màgamma
Sogate àgame kvachi
Nighandu yuttin chànìya
Nàma lingan kathìyati.

’I adore Tathàgata, who is a mine of compassion, and who, having renounced the beatific nibban within his reach, conferred happiness on others, performing all the difficult-to-be-accomplished acts in metempsychosis, the fountain of sin.

’I (adore) the sin-scaring Dhamma, to which holy sages, devoid of decrepitude and disease, have paid reverence; and by conformance to which the high and the mean, both (amongst) men and other beings,²¹ have crossed the tri-annular²² ocean (of metempsychosis.)

’And ever (do I adore) the supreme priesthood, (like unto) a merit-(producing) field, who have become the legitimate sons²³ of Buddha; and who receive reverence—are illustrious in the (three) worlds—preserve the sanvara²⁴ like life itself—and ever practise an abundance of virtues.

’Since an intimate acquaintance with nouns, and (their) genders, is essential to the (ascertainment of) the correct significations (of words), and is a powerful help to those desirous of mastering the word of Buddha;

’I shall publish the Abhidhànapadìpiká,²⁵ illustrating nouns and (their) genders, according to their application in the language of (the discourses of) Buddha.

’The masculine, feminine, and neuter are to be distinguished, chiefly, from their different forms; sometimes from the association of words (context); and sometimes by specific rule.

’[In this work] dvanda compounds will consist (of nouns) of the same genders. When words which denote the genders occur at the end or the middle of a line in a verse, (such words) refer to the (names at the) beginning (of that line); (but where they are placed at) the commencement, (they refer to) the remaining words (of the same line.)

’Know that the term dvìsu denotes both masculine and feminine; that tísu signifies all the genders; and that words ending in tu or (preceded by) atha &c., are given to express the commencement of a series of names.

‘Nouns and (their) genders are (here) illustrated, according to their application, chiefly in the Buddhist works, and sometimes after the usage adopted in Lexicons.’

The above is the Introduction to the Abhidhànapadìpikà; and I cannot conceive why it was omitted in the translation of that work by Mr. Tolfrey, and was left out by his publisher, the Revd. B. Clough. At the conclusion of the same book are also nine stanzas, which are likewise left out in the publication above mentioned; and which, since they enable us to fix the date of the work, are here subjoined:—

1 Sagga kando cha bhù kando
Tathà sàmañña kandakan
Kandattayanvità esà
Abhidhàna padìpiká.

2 Tidive mahiyan bhujagá vasathe
Sakalattha samavhaya dípani’yam
Iha yo kusalo matimá sanaro
Patu hoti mahámunino vachane.

3 Parakkama bhujo nàma
Bhùpàlo guna bhùsano
Lankàya’ mási tejassì
Jayì kesari vikkamo.

4 Vibhinnan chiran bhikkhu sanghan nikàya
Tayasmin cha kàresi sammà sammagge
Sadehanva nichchà ’daro dìgha kàlan
Mahagghehi rakkhesi yo pachchayehi.

5 Yena Lankà vihàrehi
Gàmà’ràma purìhicha
Kittiyáviya sambodhi
Katá khettehi vápihì.

6 Yassà’ sàdhàranan patvà
’Nuggaham sabba kàmadam
Ahampi gandha kàrattam
Patto vibudha gocharam.

7 Kàrite tena pàssàda
Gopuràdi vibhùsite
Sagga kandeva tattoyà
Sayasmin patibimbite.

8 Mahà Jetavanà khyamhi
Vihàre sàdhu sammate
Sarogàma samùhamhi
Vasatà santa vuttinà.

9 Saddhammatthiti kámena
Moggallànena dhìmatà
Therena rachitá yesà
Abhidhànapadìpikà.

’The Abhidhànapadípikà consists of three sections—on Heavenly, Earthly, and General subjects.

’It interprets the names of all objects in Heaven, Earth, and the Nága regions. A sensible person who excels in this, will master the words of the great sage.

’There was in Lankà a Monarch named Parakkamabáhu—celebrated, successful, endowed with virtues, and valorous as a lion.

’He in the right manner (in the legitimate mode) reconciled²⁶ the bhikkhus and Sanghas of the three Nikàyas;²⁷ and, with unceasing love, long extended his protection to (them) as to his own body, with valuable objects of maintenance.²⁸

’He established to profusion in Lankà, in the same manner that it was filled with his renown,²⁹ monasteries,³⁰ villages,³¹ parks,³² cities,³³ fields³⁴ and tanks.³⁵

’Being the special object of his wish-conferring patronage, I too, have acquired the privilege of authorship peculiar to the learned.

’Desirous of perpetuating the Saddhamma, the Abhidhánapadìpiká was composed by the erudite Moggallána thera,

’Of mild deportment, dwelling amongst the Sarogàma³⁶ fraternity (who were) received by the virtuous with approbation; and (residing) in the Vohára called the Mahà Jetavana;—

‘[A monastic Establishment] adorned with the temples, ornamented porches, &c., which were built by him (the aforesaid king) as it were a portion of Heaven reflected in his Tank.’

Here we have sufficient data to fix the date of the Abhidhánapadípikà. It was composed by a thera named Moggallàna, who had been patronized by king Parakkama. His acts, which are here related, can only be identified with those of “the heroic and invincible royal warrior, gloriously endowed with might, majesty, and wisdom; and radient with benignant virtues,”³⁷ “the most martial, enterprising and glorious of the Sinhalese Sovereigns,”³⁸ who, according to history, was Parakkamabàhu of Polonnoruva. He ascended the throne in 1153 A. D.; and when we notice that, that sovereign, who reigned for thirty-three years, turned his attention to the internal improvements which are here mentioned, in the latter part of his reign, and after he had brought his local and foreign wars to a termination; we may assign to the Abhidhánapadípikà a date at the latter end of the second half of the twelfth century. This, therefore, is posterior to the Amarakosa, which is much after the fashion of the Abhidhànapadípikà. To show their correspondence we need only present the three following introductory stanzas from the first named work.

’The masculine, feminine, and neuter (genders) are to be known chiefly by their different forms; sometimes by the association of words; and sometimes by specific rule.

’Here, with a view to distinct elucidation (nouns of) different unspecified genders are not rendered into dvanda compounds. Neither are they, without order, jumbled together; nor indeed expressed by eka sesha.³⁹

‘The term tríshu (denotes) the three genders; and dvayoh the male and female. (Where a certain) gender is expressly negatived, the remaining ones (are meant); and, where words ending in tu (occur or) atha, &c., they do not refer to the preceding (words).’

Moggallàyana’s Grammar, to which we may assign the same date that was given to the Abhidhànapadípiká, is written in imitation of Kachchāna’s style; and contains rules, supplementary notes or Vutti; and examples. The same writer has left behind a large commentary on this work. It has been further illustrated by Totagamuva⁴⁰ in his Panchikà-pradìpa, a Sinhalese work, held in high esteem among the learned. Piyadassi, a pupil of Moggaláyana has, moreover, written an abridgment of his master’s work, called the Pada-Sádana, from which the following is extracted:—

Satthànam karunà vatà gatavatà
Pàram param dhìmatà
There nà ’tuma pàdapanjara gato
Yo sadda satthà disù.
Moggallàyana vissute niha suvach
‘Chàpo viníto yathà
So’ kàsip Piyadassi nàma yati ’dam
Byattan Sukhappattiyà.

‘With a view to facilitate (study) this specific (work) has been composed by the ascetic named Piyadassi, who, like a paroquet which was taken into the cage of his (feet) tuition, was trained in the science of grammar &c., by the wise, generous, and erudite⁴¹ Moggalláyana thera of (world-wide) renown.’

There is also a Commentary on the above by Ananda, a pupil of Medankara; from which we select the following introductory remarks:—

1. Yassà tulan samadhi gamya parappasàdan
Sampàditam parahitan vipulan mayedam
So sangha rakkhita itìrita nàma dheyyo
Bhànùva bhàtu suchiràya mahàdisàmi.

2, Saddhà dhanassa patipatti paràyanassa
Sallekhiyena likhità khila kibbisassa
Odumbarà bhi hita pabbata vàsi kassa
Medankaravhaya mahà yati pungavassa.

3, Sìladi settha patipatti parànugena
Sisso-rasena saparattha ratena tena
Ananda nàma pathi tena tapo dhanena
Sankhepato nigadito Padasàdhanattho.

1. ’May the Supreme Lord named Sangharakkhita, by the acquirement of whose unparalleled patriotism⁴² this great benevolent (deed) has been achieved by me, long exist like the sun!

2 and 3. ‘The Padasàdana has been concisely commented upon by the ascetic known by the name of Ananda, (who is) bent upon doing good to himself and others; (who has) adhered to the principal (religious) usages of Sìla &c.; (and who is) a pupil⁴³ of the preeminent chief-priest named Medankara, that dwelt on the Mountain called Odumbara,—(was) rich in faith, (was) attached to (religious) duties, and (had) scratched off all sin by Sallekhiya.’⁴⁴

Payogasiddhi by Vanaratana, in the reign of one of the Buvanekabàhu’s, is also a Pali Grammar upon the basis of Moggaláyana, from the Commentary to which Turnour has quoted in his introduction to the Mahavansa.

3. The next, and, by far the most numerous, class of Pali Grammars are founded on the principles enunciated in the Sandhikappa, usually called after the name of its author Kachchàyana. This, as I have already stated, is extant in Ceylon; and, from a list of Burman Pali works in my possession, I find that it is also found in that empire. The several other editions or revisions of Kachcháyana’s Grammar, which, as remarked by Mr. Turnour, “profess, according as its date is more modern, to be more condensed, and methodized than the preceding one”, are the Rúpasiddhi and Bàlavatára.

Buddhappiyo commences the Rùpasiddhi in these words:

“Kachchàyananchàchariyan namitwà; nissàya Kachchàyanawannanàdin, bàlappabodhatthamujun karissan wyattan sukandan padarúpasiddhin.”

“Reverentially bowing down to the Achárayo-Kachchàno, and guided by the rules laid down by the said Kachchàyano, I compose the Rùpasiddhi, in a perspicuous form; judiciously subdivided into sections, for the use of degenerated intellects (of the present age, which could not grasp the original.)”—Mahàvansa, p. xxvi.

The following is in the conclusion of the same work:—

“Wikkhyàtànandatherawhaya waragurùnan Tambapanniddhajánan sisso Dípankaràkkhyo Damilawasumatí dípaladdhappakàso Bàlàdichchàdi wásaddwitayamadhiwasan, sàsanan jotayí yo, soyam Buddhappi-yawho yatí; imamujukan Rùpasiddhin akàsi.

“A certain disciple of Anando, a preceptor who was (a rallying point) unto eminent preceptors like unto a standard, in Tambapanni, named Dípankaro, renowned in the Damila kingdom (of Chola) and the resident-superior of two fraternities, there, the Bàládichchá (and the Chudàmanikyo), caused the religon (of Buddho) to shine forth. He was the priest who obtained the appellation of Buddhappiyo (the delight o Buddho,) and compiled this perfect Rùpasiddhi.”—Mahàvansa, p. xxvi.

Before I notice the principal and the oldest work with which we are concerned, viz., Kachchàyana’s Grammar, I shall mention the names of some of the principal Comments thereon to which I shall have occasian to refer in the course of these notes; viz.

Nyàsa or Mukhamatta Dìpana
Kachchàyana bheda
Kachchàyana bhede Tìkà
Kachchàyana bhede Vannanà.
Kachchàyana Vannanà
Kachchàyana Sàra
Kachchàyana Sàra Tìkà
Sandhikappa Atuvà
Sandhikappa Viggahà.⁴⁵

All these have been written expressly for the purpose of elucidating the text of Kachcháyana, which, I need scarcely say, is held in the same high estimation by Buddhists that Pánini is by the Brahmans.

Kachchàyana’s Grammar is divided into eight books. The first treats on ‘Combination’, the second on ‘Declension’, the third on ‘Syntax’, the fourth on ‘Compounds’, the fifth on (Tadhita) ‘nominal Derivatives,’ the sixth on ‘Verbs’, the seventh on (Kitaka) ‘verbal derivatives,’ and the eighth on Unnádi Affixes.’

These are found subdivided into Chapters or Sections. But, all the aphorisms do not exceed six hundred and eighty seven.⁴⁶ The following extract embraces the writer’s introductory remarks, together with the first Section of his Grammar:—

Setthan tiloka mahitan abhivandi yaggan
Buddhan cha dhamma’ mamalan gana’ mutta mancha
Satthussa tassa vachanattha varan subuddhun
Vakkhàmi sutta hita’ mettha su Sandhikappan.
Seyyan Jinerita nayena buddhà labhanti
Tanchà’pi tassa vachanattha subhodhanena
Attan cha akkhara padesu amoha bhàvà
Seyyatthi ko pada’mato vividhan suneyya.

’Having reverentially bowed down to the supreme chief Buddha adored by the three worlds, and also to the pure dhamma, and the illustrious priesthood; I now celebrate⁴⁷ the (pure) Sandhikappa in accordance with the Sutta, to the end that the deep import of that teacher’s words may be easily comprehended.

‘The wise attain to supreme (bliss) by conforming (themselves) to the teachings of Buddha. That (is the result) of a correct acquaintance with the import of his word. The sense, too, (is learnt) by a [non-ignorance] knowledge of characters and words. Wherefore, let him who aims at that highest felicity hear the various verbal forms.’

Lib. I. Section 1.

1. Attho akkhara saññàto.

The sense is known by letters.

2. Akkharà pàdayo eka chattàlisan.

The letters, a &c., are forty one.⁴⁸

3. Tattho dantà sarà attha.

Of these the eight ending with o are vowels.

4. Lahumattà tayo rassà.

The three light-measured (are) short.

5. Aññe dìghà.

The others, (are) long.

6. Sesà byanjanà.

The rest are consonants.

7. Vaggà pancha panchàso mantà.

Each (set of) five to the end of m (constitutes) a class.

8. An iti niggahìtan.

The An⁴⁹ is a dependent.

9. Para samaññà payoge.⁵⁰

Other’s names in composition.

10. Pubba’ madho’-thitam’ assaran sarena viyojaye.

Let the first be separated from its (inherent) vowel, by (rendering) the preceding⁵¹ a consonant.

Such is the sententions brevity with which the Rules in Kachcháyana’s Grammar are expressed. The author adopts three modes of explaining them. First, Vuttiyá or Várttikas, comments to supply the deficiencies in the Suttas, and to render them clear; secondly examples; and thirdly explanatory notes on some of the principal grammatical terms in the shape of questions and answers. To these again are occasionally added a note to mark the exceptions to the Rule. In the examples mention is made of several of the places and towns which were rendered sacred by the abode of Gòtama, such as Sàvatti, Patáli, Baranasí &c.⁵² There is also much correspondence between the Páninya Sutras and those given in Kacchàyana. E. G:—

1. Apàdàne panchamì Panini III. 4, 52.
Apadane panchami—Kachchayana.

So likewise:—

2. Bhùvàdayo dhàtavah. I. 3, 1.
Bhuvadayo dhatavo.

3. Kàlàdhvano ratyanta sanyoge. II. 3, 5.
Kaladdhana machchanta sanyoge.

4. Kartarì krit. III. 4, 6.
Kattari kit.

5. Asmadyuttamah. I. 4, 107.
Amhe uttamo.

Again, the text of Páninì is altered to meet the exigencies of the Pali Grammar, thus;

6. Tìnas trìni trìni pathama madhyamottamáh
Dve dve pathama majjhimuttama purisa. [I. 4, 101.

Tradition represents that, the whole work (including Vutti) as we now have it, was written by one and the same person. But this is contradicted by another Tradition. On this, I shall offer a few observations hereafter.

From its language, the Pali Grammar appears to be a very ancient work. It was probably written at a time when literature was usually carried on in the concise Algebraic form of aphorisms. This is put down by Pr. Max Muller at 600—200 B. C., embracing the period at which the founder of Buddhism flourished, and which upon the best proof on which a date may be fixed in Asiatic History, was not only posterior to the Mantas (which are identified with the Védas⁵³ the Vedangas and Veyyàkarana, but subsequent to the Mahà Bhàrata,⁵⁴ Rámàyana; Asvalàyana⁵⁴ and Paràsara; and upon other data,—477 B. C.

To the indefatigable labours of the learned translator of the Mahávansa, whom I have so frequently mentioned in the course of these notes, were Europeans first indebted for the information as to the probable date of Kachchàyana’s Grammar. In order to render my observations clear, Mr. Turnour’s authorities on the subject are here extrated.

“In the commentary on the Rúphasiddhi we find the following distinct and important particulars regarding Kachchàyana, purporting to be conveyed in his own words:—

“Kachcháyano sigrifies the son of Kachcho. The said Kachcho was the first individual (who assumed that name as a patronymic) in that family. All who are descended from that stock are, by birth Kachchàyanà.

“(If I am asked) who is thìs Kachchàyano? Whence his name Kachcháyano? (I answer), It is he who was selected for the important office (of compiling the first Pali Grammar, by Buddho himself; who said on that occasion): ‘Bhikkhus from amongst my sanctified disciples, who are capable of elucidating in detail, that which is expressed in the abstract, the most eminent is this Mahàkachchàyano.’

“Bhagawà (Buddho) seated in the midst of the four classes of devotees, of which his congregation was composed (viz. priests and priestesses, male and female asceties:)—opening his sacred mouth, like unto a flower expanding under the genial influence of Surio’s rays, and pouring forth a stream of eloquence like unto that of Brahmo—said: ’My disciples! the profoundly wise Sàriputto is competent to spread abroad the tidings of the wisdom (contained in my religion) by his having proclaimed of me that,—‘To define the bounds of his omniscience by a standard of measure, let the grains of sand in the Ganges be counted; let the water in the great ocean be measured; let the particles of matter in the great earth be numbered’; as well as by his various other discourses.

“It has also been admitted that, excepting the saviour of the world, there are no others in existence whose wisdom is equal to one sixteenth part of the profundity of Sàriputto. By the Achàrayos also the wisdom of Sàriputto has been celebrated. Moreover, while the other great disciples also, who had overcome the dominion of sin and attained the four gifts of sanctification were yet living; he (Buddho) allotted, from amongst those who were capable of illustrating the word of Tathàgato, this important task to me,—in the same manner that a Chakkawatti ràja confers on an eldest son, who is capable of sustaining the weight of empire, the office of Parinàyako. I must therefore render unto Tathágato a service equivalent to the honor conferred. Bhagawà has assigned to me a most worthy commission. Let me place implicit faith in whatever Bhagawà has vouchsafed to propound.

“This being achieved, men of various nations and tongues, rejecting the dialets which had become confused by its disorderly mixture with the Sanscrit and other languages, will, with facility acquire, by conformity to the rules of grammar propounded by Tathàgato, the knowledge of the word of Buddho ‘Thus the Thero Mahá Kachchàyano, who is here (in this work) called simply Kachchàyano, setting forth his qualification; pursuant to the declaration of Buddho, that “sense is represented by letters”, composed the grammatical work called Niruttipitako.’⁵⁵ Mahàvansa p. xxvii.

Before I notice some of the objections urged against the above tradition, it may perhaps be convenient to refer to the various other Pali writers who have given it the sanction of their high authority.

Kachchàyana commences his work, as we have already seen, with Atthó akkhara saññàto; and it has already been shewn from the passage quoted by Turnour, that that Suttan was declared by Buddha himself. This is more clearly stated as follows in the

SUTTA NIDDESA.

Attho akkhara saññáto-ti ádi máha; idan suttan kena vuttan? Bhagavatà vuttan. Kadà vuttanti—Yama Uppala nàmakà dve Brahmaná Khaya-vaya kammatthánan gahetvà gachchhantà Nadì-tíre Khaya-vayanti Kammattháne karìyamàne eko udake machchan ganhitun charantan bakan disvà, udaka bakoti vicharati. Eko ghate patan disvá ghata pato ti vicharati, Tadá Bhagavà obhàsan munchitvà attho akkhara saññàto-ti vàkyan thapesi. Tesan cha Kammatthánan patitthahi. Tasmà Bhagavatá vuttanti vuchchati. Tan ñatvá mahà Kachcháno Bhagavantan yàchitvà Himavantan gantvá Mano-silà tale dakkhina disá bhàgan sísan katvá purátthima disàbhimukho hutvà attho akkhara saññáto-ti àdikan Kachchàyana pakaranan rachi.

’It is said that ‘sense is represented by letters’ &c. By whom was this suttan declared? It was laid down by Bhagavà. (To explain) when it was declared:—Two Brahman (Priests) Yama and Uppala, having learnt (from Gotama) the khaya-vaya branches of Kammatthánan⁵⁶ went away; and, whilst engaged in abstract meditation repeating ‘Khaya-Vaya’ on the banks of the Nadí, one of them saw a crane proceeding to catch a fish in the water, and began muttering⁵⁷ Udako bako ‘water-crane.’ The other, seeing a ghate-patan ‘a cloth in a pot’ began muttering ghata-pato. At this time Bhagavà by means of a light, which he issued, declared the sentence, Attho akkhara saññàto—‘The sense is represented by letters.’ Their Kammatthànan was also effectual. Wherefore it is said that this Suttan was declared by Bhagavà. When Mahá Kachchàna learnt this, he proceeded with Bhagavà’s permission to Himavanta. Reclining in the Mano-sila region with his head towards the south, and facing the east he composed the Kachchàyana-pakarana consisting of (the Suttans) attho akkhara saññàto &c.’

In the atthakathá to the Anguttara Nikáya Mahà Kachchàyana is spoken of;⁵⁸ and the Tìka to the same work contains further particulars which are quoted in the following extract from

THE KACHCHAYANA VANNANA.

Achariyà pana lakkhana vutti Udáharana sankhátan iman Kachchàyana gandha pakaranan Kachchàyanattherena eva katanti vadanti. Tena áhá eka nipàta Anguttara tíkàyan ‘Mahà Kachcháyanatthero pubba patthanà vasena Kachcháyna pakaranan, Mahà Nirutti pakaranan, Netti pakaranan, chàti pakaranattayan sangha majjhe pakàsesi.’

’Teachers say that this Kachchàyana gandha pakarana (Text or composition) which numbers lakkhana (Rules), vutti (supplementary notes), and Udàharana (Examples), was composed by Kachcháyana thera himself. Wherefore the Tíka to the Anguttara of the Ekanipáta says; ‘the thera Mahà Kachchàyana, according to his previous aspirations, published in the midst of the priesthood the three compositions, viz. Kachcháyana Pakarana, Mahà Nirutti Pakarana, and Netti Pakarana.’

Of the three books here mentioned the Netti’ Pakarana is also extant in this Island; and it has been suggested by my Pandit that the style of this work, of which I give a specimen,⁵⁹ would seem to differ from that of the Grammar. There can be no question of this. The language of a work such as the Pali grammar, in which (to adopt the words of Professor Max Muller in respect of Pànini) the author “does not write and compose, but squeezes and distils his thoughts, and puts them before us in a form which hardly deserves the name of style,”⁶⁰ cannot bear any comparison to the style of a work on religion, where the object was to convince and fascinate the reader, and not to cramp the writer’s ideas in small sentences with a view to render a facility to those who committed them to memory. Even in other languages works on science and religion written by the same person present the same diversity of style which the subject may demand. The one may be terse, sententious, and dry; and the other full, flowing, and elegant as is the case here. The difference of style, therefore, in these two works, does by no means furnish any ground for overthrowing the authorship ascribed to this Páli Grammar. And, I am again reminded by my learned Pandit that the metre of some of the gáthàs in Kacchàyana are different from those in the text-books of Buddhism such as Dhammapada; and that that difference would favor the belief that this grammar was written long after the Buddhist era.

I freely admit the force of the learned Pandit’s observation, that the difference of metre (if, such were the fact) between confessedly Gotama’s gáthàs, and all other Pàli writings (including Kachchàna’s grammar) would establish a line of demarcation between two periods of literature. And I also confess that I have been disappointed in my search after a verse in the Vasantatilaka metre, such as Setthan tiloka mahitan &c.,⁶¹ in any of the original writings in the Buddhist works. But, leaving this question for the determination of future researches, and of more competent scholars than myself, I may be permitted to remark that the absence of a particular metre in Buddha’s lectures does not necessarily prove that it was unknown in his time. Different metres suit different compositions. There are some which are peculiarly adapted to the genius of the oldest form of the Pàli, before it received the elaborations of poets. Such’are the Anustubh, the Tristubh, the Anustubh-tristubh, the Jagatì, the Tristubh-Jagatì, and the Vaitálíya metres which we frequently meet with in old writings. The Vasantatilaka is generally unsuited to the Pali, as it is to the Sinhalese, and other so-called Pràkrita dialects. Although compositions in that metre are occasionally found in works of both those languages; yet they have all the evidence of being forced, and abound with compounds, of which the really old Buddhist writings are comparatively free. This is doubtless the reason why this particular metre (Vasantatilaka) is not to be found in the lectures of Gotama, which were expressed, as occasions presented themselves, with a view to impart religious instruction, without study, and without any attempts at ornament. The case with the grammarian was, however, different. Except in the Suttans which were designed for instruction, he seems to have studied ornament. Perhaps too, he was fond of display. He, as we learn from himself, did not despise Sanskrit grammatical terms. He had no reason, therefore, to reject really Sanskrit metres in which he could with elegance depict the virtues of his Teacher. Hence the adoption in this grammar of metres which are not met with in original Buddhist writings.

There is not, therefore, I apprehend, sufficient evidence to set aside the popular tradition as to the author of this Grammar, which I perceive is supported by various considerations and inferences which may be drawn from several historical facts. That the Màgadhí which the Buddhists denominate the Pàli, was an actually existent form of speech in Behar at the time Buddhism arose, may be easily believed.⁶² Yet, before this period when the Magadhas had but one common religion, and, perhaps, cultivated the language in which that religion was taught with greater zest, it is not improbable that the treatises to elucidate the vernacular Mágadhì were, as the Sinhalese Grammar are at the present day, limited. And this is obviously the reason why Kachcháyana, as he himself declares, had borrowed technical terms from Sanskrit authors.

PARA SAMANNA PAYOGE. (Vutti)—yá cha pana sakkata gandhésu samaññà ghosà’ ti và aghosà’ ti và tá payoge sati etthà’ pi yujjante. ‘In composition other’s appelations. [Vutti.] Such (Grammatical) terms as are called ghòsa, (sonants) or aghósa (surds) in Sanskrit (gandhas⁶³) compositions are here adopted as exigency may require.’

By ‘Sanskrit’ sources, perhaps, the writer meant the Pràkrit Grammars by Sanskrit writers, or such Rules of Pànini as are indicated in the following extract from the Kavikanthapàsa by Kedàra-bhatta:

Pànini bhagavàn Pràkrita
Lakshana mapi vakti Sanskrità danyat:
Dìrghàksharancha kutrachi
Dekàn màtrà mupaitìtì.

That is:—‘Pànini, the Rishi, speaks also of the lakshana (or Grammatical Rules) of Pràkrita, besides the Sanskrit; and (says) that in some languages a long letter⁶⁴ becomes one syllabic instant.’ But, whether this inference be well founded or not, it is quite clear that this was one of the earliest Pali Grammars, which borrowed its technical terms, not from Pali but, Sanskrit writers. Its object too, was to fix the Rules of that language; since at the time Kachchàyana wrote his Grammar the Mágadhì, like other “dialects,” had a tendency to become ‘confused by its disorderly mixture with the Sanskrit and other languages.’ This is a fact. As a vernacular dialect, the Màgadhì scarcely maintained its character in Asia till the age of Asòka. Two hundred years had hardly elapsed before its development was ‘intermediate between the Pali and Sanskrit.’ There must therefore, have been a gradual declension in Pali literature during those two centuries. Indeed this could not have been otherwise when we ascertain that the very pupils of Gotama had resorted to other languages for the elucidation of Buddhism. Hence the necessity for a compilation like Kachchàyana’s—a work on Grammar ‘for the easy comprehension of the word of Buddha’—Sukhèna Buddha Vachanan ugganhissanti-ti; and as Kachchàyana himself says in the opening of his work, written, not according to the vernacular dialect of the Magadhas, but ‘in accordance with the (language) of the Suttans’—vakkhàmi sutta hita’ metta su Sandhikappan.

These are not all the circumstances from which it may be inferred that Kachchàyana’s Grammar was written at the first dawn of Buddhism. As was the case with all ancient nations, a sacred literature gave rise in Magadha to philological sciences. Religious doctrines when disseminated through a written medium, naturally led the Buddhists to Grammatical inquiries. The necessity for rescuing the dhamma from corruption, and for preserving their correct interpretations, as already seen, was a powerful inducement to Gotama’s disciples to fix the rules of their Preceptors’ words, the Jina vachana, as the Màgadhì is called in most ancient works.

It may, however, be asserted that neither was writing known six centuries before Christ, nor, consequently, were Buddhist doctrines recorded at that date. I have discussed this question elsewhere, and I may here state, as the result of those investigations, that at the time when Buddhism first started into existence, writing was known in Magadha as much as painting.⁶⁵ It was practised in the time of Gotama.⁶⁶ Buddhist doctrines were conveyed to different countries by its means.⁶⁷ Laws and usages were recorded.⁶⁸ Little children were taught to write.⁶⁹ Even women were found able to read and write.⁷⁰ The character used was the Nágari.⁷¹ Vermilion was the ‘ink’, and metal plates, cloth, hydes, and leaves constituted the ‘paper’ of the time.⁷² That Buddhist annals therefore, were reduced to writing from the very commencement, is not only reasonable, but is indeed capable of easy and satisfactory proof.

To return to the subject. The literary qualifications of the théra Kachchàyana, seem to have been indeed such as to warrant the belief that he devoted his time to the elucidation of the language of Buddhism. He was, as is abundantly proved in the Pali works, a distinguished member of the Buddhist Church. He is also mentioned in the Tibetan Buddhistical Annals, as one of the disciples of Gotama; and it is expressly stated of him, that ‘he recited the Sútra on emancipation in the vulgar dialect.’ By ‘the vulgar dialect’ Mons. De Korési doubtless meant the language to which Colebrooke had previously given that appellation—the Màgadhí. Gotama himself states that of all his pupils Mahà Kachchàyana was the most competent to elucidate his doctrines. In the very language of the sage, which is here quoted from the Ekanipáta of the Angutta Nikáya: Etanaggan Bhikkhavè mama sávakànan bhikkhúnan sankhittena bhàsitassa vitthàrena atthan vibhajantànan, yadidan Mahà Kachchàno—‘Priests, he who is Mahà Kachchàyana is the chief of all the bhikkhus, my pupils, who can minutely elucidate the sense of what is concisely expressed.’ That this supremacy refers to the literary, and not the theological, attainments of Kachchàyana appears from the following comment which we extract from the atthakathà to the Anguttara Nikàya.

Aññe kira Tathàgatassa sankhepa vachanan Attha vasena và pùritun sakkonti vyañjana vasena và; ayan pana thero ubhayenàpi sakkoti: tasmà aggo-ti vutto.

‘Some are able to amplify the concise words of Tathàgata either by means of letters, or by [shewing] their sense. But this thera can do so in both ways. He is therefore called chief.’

In the Nyàsa or the Mukha matta-dìpaní, which is supposed to be the earliest commentary on Kachchàyana’s Pali Grammar, and, as may be proved, older than the Rùpasiddhi, the author of this Grammar is not only identified with the Kachchàyana thera, whose ‘intellectual supremacy was extolled by Buddha’; but his memory is thus respected by an ‘Obeisance.’

Kachchàyanan cha muni vannita buddhi-‘massa
Kachchàyanassa mukha matta’ mahan karrissan
Pàramparà gata vinichchaya nichchhayan cha.

‘Also (bowing down to) Kachchàyana, whose intellectual attainments had been complimented by Buddha, I shall comment upon the positive conclusions (Rules) which have been handed down by tradition as the very oral (teachings)⁷³ of this Kachchàyana.’

With reference to the name Kachchàyana in the above extract, the following passage occurs in the Nirutti sàra Manjusa: wherein also the writer acknowledges the consummate scholarship of the Grammarian.

Kasi Kammàdinà vyàpàrena kachchati dippatíti Kachcho, thera pità tassa apachchan putto Kachchàyano Neruttukànan pabhava bhùto pabhinna pati sambhido etadagga tháne thapito khìnásavatthero, tan pana natvàna.

’By reason of the occupation of ploughing &c. [comes] Kachchati ‘he shines.’ Thence Kachcho, (the name of) the thera’s father. His son is Kachchàyana—a thera, who was an arahanta, who was placed in the highest position, who had attained the patisambhidà,⁷⁴ and who was the first cause (source) of all Neruttikà, [Grammarians or] philologers.’

Although it is stated⁷⁵ that Kachcháyana was residing at Avanti, the pachchanta or ‘the foreign regions’⁷⁶, it is however expressly stated that this Grammar was written in the Himavanta; and from the mention of the principal towns celebrated by the presence and abode of Gotama, and especially that which had risen from a small village to the importance of a populous city in the time of the sage, I mean Patàliputta,⁷⁷ it may be inferred that the writer took for his examples such of the names as were then of recent celebrity.

As we have already seen, the Màgadhí scarcely maintained its original purity in the Magadha desa until the second ecumenial convocation; and the code of the Vajjian heretics, which was made at this time, and which may be clearly identified as the Nepal collection,⁷⁸ was in point of language “intermediate” between the Pali of Kachchàyana and the Dhammapada; and the Pràkrit of the Pillar-dialect.

These are important facts, which prove that the Grammar in question was composed in the golden age of the Pali literature—before it became interlarded with the Sanskrit, and before the language was so much neglected, (as at the time of the second convocation), that, except those who maintained the orthodox version of the Scriptures, literary men were unable to pay correct attention to ‘the nature of nouns, their genders, and other accidents of Grammar, as well as the various requirements of style.’⁷⁹

Nàma lingan parikkhàran àkappakaranàni cha
Pakatibhàvan vijahitvà, tañcha aññan akansute.

Thus, when the uniform and popular tradition stated in the above extracts, which there is no reasonable ground to set aside, is coupled with the many inferences to which I have directed attention; it is, I apprehend, very clear that Kachchàyana, the author of Sandhi-kappa, was one of the eighty eminent disciples of Gotama. As such, he must have flourished in the latter-half of the sixth century before Christ.

Against this popular belief I have been referred by several friends to another name of Kachchàyana, the author of the Abhidharma Jnàna prasthàna, mentioned in the following account of Hiouen-thsagn, the Chinese traveller of 629—645, A. D.

Après avoir fait environ cinq cent li, au sud-est de la capitale (de Chìnapati), il arriva au couvent appelé Ta-mo-sou-fa-na-seng-kia-lan (Tâmasvana-sanghâ râma), ou le convent de la Forêt Sombre. On y comptait environ trois cent religieux qui suivaient les principes de l’école des Sarvàstivàdas. Ils avaient un extérieur grave et imposant, et se distinguaient par la pureté de leur vertu et l’élévation de leur caractère. Ils approfondissaient surtout l’étude du petit Véhicule. Les mille Buddhas du kalpa des sages (Bhadrakalpa) doivent, dans ce lieu, rassambler la multitude des Devas et leur expliquer la sublime loi. Dans la trois centième année après le Nirvàna de Sàkya Tathàgata, il y eut un maître des Sàstras, nommé Kâtyâyana, qui composa, dans ce couvent, le Fa-tchi-lun (Abhidharma-jnàna-prasthàna.)—Mémoires sur les Contrées occidentales par Hiouen-thsang, liv. iv. p. 200.

‘Having travelled about five hundred li southwest of the capital (of Chínapati) he arrived at the monastery called Ta-mo-sou-fa-na-seng-kia-lan—(Tamasvana Sanghàràma) or the monastery of the dark-forest. About three hundred Religieux are reckoned in this place, who follow the principles of the Sarvâstivàdas school. They maintain a grave and imposing exterior, and are remarkable for purity of virtue, and elevation of character. They are engaged in the profound study of the little vehicle. The thousand Buddhas of the Kalpa of the wise men (Bhaddrakalpa) were bound to assemble, in this place, the whole multitude of the Devas, and expound to them the sublime law. In the three hundredth year after the nirvána of Sàkya Tathàgata, there was a master of the Sâstras, named Kátyâyana, who composed in this monastery, the Fa-tchi-lun (Abhi-dharma-jnâna-prasthâna.)’

Here there is nothing to establish the identity of persons. The age too, given by the Chinese pilgrim, does not throw any light on the subject. If Kàtyàyana, the author of Abhi-dharma-jnâna Prasthâna lived 300 A. B., he flourished after Asòka, and, according to the same authority quoted by Cowell, in a ‘Monastère fondè par Asoka’; and at a time when he was sure to have figured very conspicuously in the Buddhistical annals of Ceylon.

The absence, however, of any notice regarding him, proves, as I apprehend, what the Chinese traveller himself indicates, that Kàtyàyana of Tamasvana Sanghàràma was one who followed the principles of the Sabbatti vàda⁸⁰-school, and, therefore, one of the seventeen sects mentioned in the Dípávansa—‘who distorted the sense and phraseology (of the scriptures); omitted a portion of the original (compilation) and of the gàthás; substituted others (in lieu of them); disregarded the nature of nouns, their genders, and other accidents, as well as the various requisites of style; and corrupted the same by different substitutions.’

Now, it is quite clear, from the evidence contained in the above passage, as well as from that furnished by the style of the Nepal scriptures, and also from the statements in the Chinese accounts, that the language used by these sectarians was, as stated by Professor Burnouf, ‘one intermediate between the Pali and the Sanskrit’; that it was called the Fan, or the Brahman language, as that word is unmistakeably used in the following passage—Le dieu Fan (Brahman) et le roi du ciel (Indra) établirent des règles et se conformèrent au temps;’ and that it was a language with the dual number, and therefore the Sanskrit, as moreover the name Abhidharma jnàna prasthàna, the work itself attributed to Kàtyayàna clearly indicates.

It is also stated that the Buddhists had an object in ascribing this Grammar to Kachchàyana. Professor Max Muller traces the animus falcendi, to a ‘tendency of later Buddhist writers to refer the authorship of their works to names famous in ancient Brahmanic history.’—p. 303. There is, I believe, no foundation for this assertion. ‘One swallow does not make summer.’ Much less does a single instance prove a practice. One solitary exception is here made the rule, especially in regard to a question of custom—‘the tendency of a people to do certain acts.’ If, therefore, no other names but “Kachchàyana” can be pointed out in support of this allegation, the allegation itself, I apprehend, is disproved.

But, to take a brief excursus into the subject, nothing can be clearer from the history of Buddhism, as we find it in the annals of Ceylon, than that the Buddhists, in imitation of their teacher, have always attempted to draw a distinction between themselves and other sects, especially the Brahmans. This will be found to be the case, even where there is no substantial difference between them. Although Buddhism, I am persuaded, arose out of Brahmanism, and although the very doctrines of the former are built upon those of the latter; yet there is scarcely a single subject upon which the doctrines of both are identical. Gotama, it would also seem, never lost an opportunity to draw some distinction between his own doctrines, and those of the Brahmans. Take, for instance, the doctrines of ‘the Creation’; of [àtman or attha] ‘the soul’; of [Nirvána or nibban] ‘eternal bliss’ &c. &c.; and none can fail to perceive the attempt that is made by Buddhists to vary the Brahman doctrines. In this endeavour they have got into confusion, as in the case of Nibban, which even Nágasena pronounces to be ‘a mystery.’

I may also allude to the institution of Castes. Gotama varied the Brahmanical doctrine on the subject. He differed from them as to its origin. He abolished that distinction among the priesthood. Although he never preached against it in a social point of view, and never affirmed it to be sinful, and pernicious to society; yet to set aside the pretensions of the ‘twice born’ he declared the universal equality of mankind in a religious point of view.

Khattiyo settho jane tasmin ye-gotta patisàrino
Vijjà charana sampanno so settho deva mànus.

‘Amongst mankind, who are scrupulous in regard to their lineage, the Khattiya is supreme; but he who is endowed with Vijjà and Charana is supreme, both amongst devas and men.’

The Vijjâ and Charana are also thus defined by the Buddhists:—

Vipassanà-ñàna manomayiddhì
Iddhippabhedo picha dibha sotan
Parassa cheto pariyàya ñánan
Pubbènivà sànu gatancha ñànan.
Dibbancha chakkhà sava sankha yocha
Etàni ñànàn idhattha vijjà
Alankarun yà muni dhamma deham
Visesa sobhà guna majjhu peta.

‘Here the eight Vijjá are the (following heads of) knowledge;—abstract devout meditation; power to assume any corporeal figure whatever at one’s will; the various other attributes of iddhi;⁸¹ divine hearing; knowledge of that which is produced in another’s mind; knowledge of the state of previous existences; a divine perception; and the extinction of (distressful) desire. These, which are peculiar attributes or qualities, embellish the [dhamma] religious-body of Buddha.’

Sìlan varan indriya san varocha
Mattà sità jà-gariyà’nu-yogo
Saddhà hirottappa bahussutat-tan
Parakkamo cheva satì matìcha.
Chattàri jhànànicha tàni’ mànì
Tìpancha dhammà charanàni jaññà
Etehi vijjàhicha sampayogo
Sampanna vijjá charano munindo.

‘Know that these fifteen constitute the Charana; (viz.) observance of the supreme precepts, subjugation of the passions, eating ordinately,⁸² wakefulness,⁸³ faith, shame (for sin), fear (of sin), much hearing (study), prowess, retentive memory (sati), understanding (mati), and the four jànas.⁸⁴ By the association of these with the foregoing Vijjá the supreme Buddhà was endowed with Vijjá and Charana.’—Pradípikáva.

The anxiety of Buddhists not to identify themselves with Brahmans appears also from the meanings which the former attach to the very words borrowed from the latter, e. g.

Brahmachariyà, ‘the Brahman in his noviciate’ is interpreted to mean ‘the whole course of Buddhist religious duties.’⁸⁵ Take again the word Valabá. Between it and the Sanskrit word badavà, there is but the difference of the two dialects ; for the interchange, occasionally, of b and v, and the change of the Sanskrit d into l, in the Pàli, is well known. Taking this, therefore, to be the Sanskrit word badavà we find that the Buddhists, whilst adopting the Brahman word for “Aurva, sub-marine fire, called badava or bàdava, and personified as the son of the Saint Urva”, assign to it the like meaning of ‘a sub-marine fire’, but, with a view to differ from the Brahmans, explain it to be—not the deva of the Hindu Pantheon, ‘who, consisting of flames but with a mare’s head sprung from the thighs of Urva, and was received by the ocean’, but—as Milton describes it,
——‘a fiery deluge, fed
With ever-burning sulphur unconsum’d.’
“The unquenchable fire of hell, so rigorous that its contact with water only inflames it the more.”⁸⁶ And the badavàmukha ‘the mare’s mouth’, wherein the Hindu Urva entered, is also known to the Buddhists as valabámukha ; but they define it to be ‘a pool or hollow in the trough of the Sea, occasioned in stormy weather by the waves rolling towards the Meru or the Sakvala-gala.’

To return to the alleged ‘tendency of later Buddhists &c.’ I have closely searched, but in vain, for a single instance (Kachchâyana excepted) in which the Buddhists of any period might be charged with such a weakness.

Take, for instance, the Rùpasiddhi or Bàlavatàra, and other Pali Grammars which have already been noticed. There is no correspondence between the names of their reputed authors, and the “names famous in ancient Brahmanic history.” Take also such names as Mihindu, Buddhagosa, Ananda, Buddhapiya, Vanaratana, Moggallàyana, and Anòmadassi, and we look in vain for their celebrated prototypes ‘famous in Brahmanic history.’ True, some of the Buddhist names are Brahmanical, and for the simplest of all reasons, that Buddhism arose out of Brahmanism, and on Brahmanical soil ; and that some of ‘the most famous in Brahmanic history’, such as Amara, Purushottama, Hemachandra, &c., have embraced the new religion. Moreover, in the East, Brahman appellations were anciently, as they are at present, common names. Amongst the coolies in the Coffee and Cocoanut plantations of Ceylon do we meet with many a Parasi Rámă, Chandra, &c., &c. ; yet, with the exception of the Patronymic Kachcháyana, it is difficult to find a single name of a Buddhist writer, which can be ‘referred to similar names amongst the Brahmans famous in history.’

The reason too, for the anxiety evinced by Buddhists not to identify themselves with Brahmans, may be easily explained. It was to remove the reproaches of the Brahmans, such as the following, which Kumârilá casts upon the Sàkya fraternity.—“These Sákyas, Vaiseshikas, and other heretics, who have been frightened out of their wits by the faithful Mimansakas, prattle away with our own words as if trying to lay hold of a shadow.⁸⁷

Having thus noticed the arguments for and against the alleged date and authorship of this Pali Grammar, it is indeed a matter of curious inquiry, especially in view of the similarity between it and Pànini ;—’what relationship does the Pali Kachcháyana bear to its Sanskrit prototype? And here I shall first dwell upon the alleged identity between the author of the Pràkrit Prakàsa, and Kachcháyana. Professor Cowell, the erudite translator of the Pràkrit Prakàsa, thus notices the subject :—

“Kàtyàyana has always been the reputed author of the Vàrtikàs, or supplemental remarks on the ancient Grammar of Pànini; and both names are found in the accounts of the Chinese Buddhist Hiuan-thsang, who travelled in India in the first half of the seventh century of our era. Pànini is called Pho-ni-ni, and described as the founder of music, which appears to be the nearest Chinese expression for a Grammarian ; the passage relating to Kàtyàyana is as follows.⁸⁸ ‘Tchi na pou ti (erigé par les Chinois), limite de l’Inde du Nord. . . . . . . Au sud-est de la grande ville, à 500 li, monastère de Tha mo sou fa na (forêt obscure); lá a vécu le docteur Kia to yan na, 300 ans aprés le Nirvána⁸⁹ Monastère fondé par Asoka.’ The Buddhist traditions in Ceylon all agree in calling the author of the earliest Pali Grammar Kachcháyano ;⁹⁰ and although this is said to have perished, yet when we remember how very closely allied Pàli is to Pràkrit, and that Kachchàyano is simply the Pràkrit form of Kátyáyana, there can be little doubt that the Prákrit grammar of the one and the Pali grammar of the other, are only the Brahmanical and Buddhist versions of the same tradition.”—p. viii.

The learned Professor’s argument amounts to the following :—’Kàtyáyana alias Vararuchi was the writer of the Páninya-Várttikas. Kachchàyana, between whose name and that of Kàtyàyana there is only the difference of dialects, was the author of a Pali (Pràkrit) Grammar. Vararuchi was, moreover, the writer of the Pràkrit Prakàsa. Things which are equal to the same thing, are equal to each other. Therefore, Vararuchi was Kàtyàyana ;—Kàtyàyana, Kachchàyana ;—and Kachchàyana, Vararuchi. Therefore, the Pali Grammarian and the Pràkrit writer were identical !

This reasoning is certainly inadmissible. Identity of names does not prove identity of persons.⁹¹ It is possible to point out from History several Kàtyàyanas, and as many Vararuchi’s. They are, like Kàlidàsa and Kàsyapa in India, or, as Smith in England, common names. We have no better authority than the idle tale of a Vrihat Kathà, which abounds with the “marvellous,”⁹² to prove that Kàtyáyana was called Vararuchi ; and. even admitting for the sake of argument, and upon the authority of the Kathà Sàrit Ságara, and Hema-chandra, that such was the fact ; it is remarkable that, when people speak of the author of the Vàrttikàs, they generally name him Kàtyáyana—not Vararuchi; and that when they allude to the writer of the Pràkrit Prakàsa they call him Vararuchi, not Katyàyana—shewing that they were two different men. There is indeed no tenable evidence of the identity between Panini’s Commentator, and the author of the Pràkrit Prakàsa; none, indeed, to shew that the latter was the same individual that wrote the Pali Grammar.

The internal evidence, however, contained in the Pali and Pràkrit Grammars, satisfactorily proves that they were written by two different men, and at comparatively two remote times from each other.

Kachchàyana was a Buddhist, not only upon the authority of the Rúpasiddhi, but the testimony which confirms it, viz. the internal evidence of the fact in the Pali Grammar. Kachchàyana opens his work with a salutation to “Buddha of infinite knowledge,” whereas Vararuchi, I believe it will be admitted, was of the Brahman faith. This is not all. There is no correspondence whatever in either arrangement, sentiments, or words, between the two works. According to Professor Lassen (Inst : § 6.) “Each (of the six dialects, of which the Pràkrit Grammarians treat) “descends by one degree of purity below the preceding one, so that the last is more remote than any of the former, from the common source.” In this view of the case, the Màgadhì takes a third place in the list of “scenic dialects.” Yet it is a well established fact, even in the opinion of M.M. Burnouf and Lassen, (Essai sur le Pali, p.p. 138 ff.) that “when the Pali (Màgadhì) as a derivative from the Sanskrit, is compared with other dialects, which have the same origin, it is found to approach far more closely than any of those others to that common source. It stands, so to speak, on the first step of the ladder of departure from Sanskrit, and is the first of the series of dialects which break up that rich and fertile language.” This discrepancy, therefore fully proves that Vararuchi treats of Pràkrit dialects, especially the Mágadhí, of an age much later, as the language shews,⁹³ than the text-books of Buddhism. His grammatical rules of the principal Pràkrit, which Lessen denominates the Dialectus Præcipua’, are designed for a modified form of the Pali—after it found a retreat in Ceylon, and degenerated from the form in which we find it in Kachchàyana, and Dhammapada, and before it assumed the shape of the present Mahàràshtrí. This I shall endeavour to shew hereafter. Suffice it however to state here that the Pràkrit-Màgadhì of Vararuchi is different from the Pali, and from every dialect which is supposed to have risen from it.

It would thus appear, that the author of the Pràkrit Prakàsa, and Kachchàyana, were different persons ; and, upon the evidence of religion, it may be inferred, that the latter was also different from the Brahman sage of the Pàninya-Varttikas.

In view, however, of the correspondence between Pànini’s Sanskrit Grammar, and Kachchàyana’s Pali work—a correspondence which is not limited to one or two stray instances, but found in different chains of Sútras, and which may be detected not merely from the similarity of thoughts, but from the sameness of language—it may be inquired which of these works was prior in point of time ?

This question may be considered in two different points of view ; 1st, whether Kachchàyana availed himself of the same Grammarians to whom Pànini himself was indebted ? or, 2ndly, whether the Sanskrit author, whose grammatical terminology the Pali writer chiefly adopted, was Panini ?

So far as my researches have extended, and they are indeed very limited, the only circumstance which favors the first hypothesis is, that some of the technical terms in Kachchàyana, e. g., Panchamí and Sattamí, for the ‘Benedictive’ and ‘Potential’ moods, which are not found as a fifth and a seventh division of the verb, are different from the names given to the same by Pànini. The Balàvatàra explains (panchamì sattamí tyayan pubb’àchariya saññà) that ‘Panchamí and Sattamì are the appellations of former teachers’ ; and the Mahà Sadda Níti states, that these appellations are in accordance with Sanskrit Grammars, such as the Kàtantra,⁹⁴ a comparatively modern Grammar, as stated by Colebrooke. These statements however are of no value. The expressions “former teachers” and “the Grammars such as the Kàtantra” are too vague and indefinite. Upon their basis no conclusions can be drawn. By “former teachers” we may fairly infer those who lived before Kachchàyana, or before Pànini; and who can say that they did not likewise adopt the same appellations which “such Grammars as the Kàtantra” use in regard to the Benedictive and Potential moods ? It is indeed probable that Pànini, like Kachchàyana, adopted certain, and rejected certain other, technical terms, &c., of former Grammarians. My acquaintance with the Sanskrit literature is far too limited to draw any further inferences. But all circumstances considered (to some of which I shall hereafter refer), I cannot refrain from the conviction that Kachchàyana had Pànini before him when he composed the Sandhikappa. If such were the fact, should not the Sanskrit Grammarian be placed before the Buddhist era ?

This question, as indeed every matter relating to Asiatic History and Chronology, is one of considerable difficulty ; and I must most distinctly disclaim the slightest pretention to give any definite proof on the particular question, especially when I find such eminent Sanskrit scholars, as Wilson, Bohthling, Weber, and Max Muller, have failed to do so. All I desire however, in view of the evidence which the Pali Grammar reveals, and the historical incidents which the Pali Buddhist annals disclose, is to attract public attention to a few inferences and deductions which may be drawn from them, and which do not precisely accord with the views expressed by the learned scholars abovenamed.

Experience has proved, that whatever weight might be attached to facts stated in Brahmanical works, no reliance could be placed upon their chronological calculations. All that may be depended upon to a certain extent are their popular traditions, when supported by the testimony of other, especially the Buddhist, nations. The popular tradition then, as to the age of Pànini, which is current not only among the Brahmans of India, but among the Buddhists of Ceylon, is exactly what is stated by Colebrooke in the following passage.

“Pànini, the Father of Sanskrit Grammar, lived in so remote an age, that he ranks amongst those ancient sages, whose fabulous history occupies a conspicuous place in the Purànas or Indian Theogònies. The name is a patronymic indicating his descent from Panin, but according to the Paurànica legends, he was grandson of Devala, an inspired legislator.”⁹⁵

Against this popular belief, several writers have quoted the passage wherein Pànini mentions Yavanàni as a name of a lipi, or writing, “the alphabet of the Yavanas.” The inference sought to be deduced, viz., that the Yavanas, who were a “head-shaving race”⁹⁶ were Ionians, or Bactrian Greeks, who could only have been known in Asia after the invasion of Alexander the Great,⁹⁷ is indeed unfounded.

Few subjects in the history of the East, are capable of more satisfactory proof than that the Yavanas or Yonas had been known before Gotama Buddha.

The identification of Yavanas with Mohammedans, is also open, in the opinion of Professor Wilson, to the objection that the former are mentioned in works prior to the Mohammedan era.⁹⁸

In one of Asoka’s inscriptions, the Girnar, Antiochus is called the Yòna ràja, ‘the king of the Yonas.’ The Milindappanna speaks of Milinda as a Yona king. Whether he be identical with Meneander, and the Yonaka country with Euthydemia⁹⁹ remains to be proved. From the following extracts, however, we glean the facts ; that Milinda was born at Kalasi in Alasadda, 200 Yojanas from Sàgal; and that Sàgal was only twelve Yojanas from Cashmir.

Ràjà àha bhante Nàgasena yo idha kàla kato Brahma loke uppajjeyya yocha idha kála kato Kasmìre uppajjeyya kochira taran ko sìga taran’-ti. Samakan Mahà ràjà’ti. Opamman karohì’ti—kuhinpana Mahá ràja tava jàta nagaranti—Atthigàmo bhante Kalasi gàmo nàma yatthàhan jàto’ti—Kìva dùro Mahà rája ito Kalasi gàmo hotíti—Dumattàni bhante yojana satànìti—kíva dùran mahà ràjà ito Kasmíran hotíti-Dvedasa bhante yojanànìti—Ingha tvan mahà ràjà Kalasigàman chintehíti—Chintito bhanteti—Ingha tvan Mahà ràja Kasmíran chintehíti—Chintitan bhanteti—Kataman nukho mahà ràjà chirena chintitan kataman sígataranti—samakan bhanteti. Eva mevako mahà ràja yo idha kála kato Brahmaloke uppajjeyya yocha idha kàla kato Kasmìre uppajjeyya samakan yeva uppajjantìti.

“The king said, Lord Nàgasena (suppose) one who dies here (Sàgal) is born in the Brahma world ; and another who dies here is born in Kashmir : which of them is born sooner, and which of them later ? Priest—Monarch, at the same time ? King—Give an illustration. Priest—Monarch, which is the City of thy birth ? King—Lord, I was born in a place which is called Kalasi gàma. Priest—Monarch, how far is Kalasi gáma from hence ? King—Lord, about 200 yojanas. Priest—Monarch, how far is Kasmir from hence ? King—Lord, Twelve yojanas. Priest—Monarch, think quickly of Kalasi gàma. King—Lord, I have thought. Priest—Monarch, think quickly of Kasmir. King—Lord, I have. Priest—Which of them, Monarch, hast thou taken shorter time to think, and which of them longer ? King—Lord, equal time. Priest—So likewise, Monarch, he who dies here, and is born in the Brahma lòka ; and he who dies here, and is born in Kasmir, are both born at the same (period of) time.”

Again :—Thero àhakuhin pana mahà ràja tava jàta bhúmìti’—‘Atthi bhante Alasando nàma dípo tatthàhan jàtoti’—‘kíva dúro mahà ràja ito Alasando hótìti’—‘dumattàni bhantè yojana satà niti.’

“The Priest asked, Monarch, where is the land of thy birth ? Oh! Lord, there is an island named Alasanda. I was born there. Monarch, how far is Alasanda from hence (Sàgala) ?—Lord, about two hundred yojanas”

In the following passage Isiodorus mentions Ságal and Alexandria in the same sentence—et Sigal urbs, ubi regia Sacarum, propeque Alexandria urbs et non procul Alexandriapolis urbs. From the Mahawansa, moreover, we learn that Alasadda was the Capital of the Yona country. The mention of dìpo in reference to Alasanda, in one of the above extracts, presents no valid objection against its identification with Alexandria ; for Pali writers, and Buddhists in general, like the ancient Greeks, had a very vague notion of the Geographical position of countries.

Perhaps the Milindappanna, as well as the Inscriptions do, not furnish conclusive proofs on the subject ; since they were clearly after the date of Asoka, who is expressly mentioned therein. Nor indeed are the Nàtakas of much value, for the same reason. But the same objection does not apply to Manu, or the Mahà Bhárata, in both which ancient works the Yavanas are mentioned.

Manu states¹⁰⁰ that the following tribes were originally Kshatriyas, but have gradually sunk to the state of Vrishalas (Sudras), from the extinction of sacred rites and from having no communication with Brahmans ; viz. Paundrakas, Odras, Dravidas, Kāmbojas, Yavanas, Sakas Pāradas, Pahlavas, Chìnas, Kirātas, Daradas, and Khasas.

“These tribes of Kshatriyas, viz., Sakas, Yavanas, Kāmbojas, Dravidas, Kalindas, Pulindas, Usìnaras, Kolisarpas, and Mahîshakas, have become Sudras from seeing no Brahmans.”¹⁰¹

The facts contained in the above extracts are supported in the Buddhistical annals ; and in quoting therefrom it becomes my privilege to adduce the authority to which Mr. Turnour referred, but which he failed to adduce, to prove that ‘Yavana (yóna) is mentioned anterior to Alexander’s invasions in the ancient Pali works’¹⁰² Whether the Buddhist Pitakattáya was written after the death of the Sage, or before (and that it was at the period of the Buddhist era is also capable of satisfactory proof), Gotama, whose age is firmly established, has spoken of the Yavanas ; and in special reference to the distinction of Aryas and dàsyas, which was recognized in the pachchanta (foreign) countries such as Yona and Kamboja.

In the Majjhima Nikàya, from which I shall again quote, Gotama is said to have asked :

Tankin maññasi Assalàyana ? suttante ’Yona Kambojesu aññesu cha pachchante mesu janapadesu vevannā ayyocheva dàso cha hòti—ayyo hutvà dàsohoti, dàsohutvà ayyohotì’ti.

‘Assalàyana, what thinkest thou of this ? Hast thou heard, that in Yona and Kamboja, and in other foreign countries, there are various ayyas¹⁰³ (superiors) and dàsas (inferiors) ; that superiors become inferiors, and inferiors superiors’ ?

It is said in the commentary that the above was said to illustrate (such a case as) this :

Bràhmano sabhariyo vanijjan payo jento Yonaka ratthan và Kamboja rattan và gantvà kàlankaroti,—tassa gehe vayappatto dàso hoti ; Bràhmanì dásenavà kammakàre navà saddhin vásankappeti ; etasmin dàrake jàte so puriso dàsova hoti ; tassa jàta dàrako para dàyajja sàmiko hoti-matito suddho pitito asuddho—so vanijjan payojento majjhima padesan gantvà bráhmana dàrikan gahetvà ; tassà puchchchismin puttan patilabhati, sopi màtitova suddho hoti pitito asuddho. Evan Bráhmana samayasmin yeva jàtisambhedo hoti—ti dassanattan etan vuttan.

‘A Brahaman, provided with merchandize, having gone with his wife either to the country of Kamboja, or the country of Yona, dies. There is a grown up dása or laborer in his house. The Brahmanì lives either with the dàsa or the laborer, and begets a child for him ;—that person is still a dàsa. The child that is born for him, who is pure as regards the mother, and impure as regards the father, becomes the lord of the inheritance. He (too), provided with merchandize, goes to the Majjhima region, and takes to him a Brahaman lass. She too gets a son, who is pure only on the mother’s side, but impure on that of the father. Thus according to the very observances of the Brahamans there is a distinction of tribes.’

Whilst the authority above quoted satisfactorily explains the reason why, as in the Hero and the Nymph, Kàlidàsa has applied the term Yavana to menial females ; it also establishes the fact that the Yavanas were ante Buddhistical.

It has also been stated by Professor Max Muller that since Pànini refers to the Unnâdi-sûtras, which mention dinàrah ‘the Roman denarius’, Jinah ‘synonimous with Arhat a Buddhist saint’, tirîtam ‘a golden diadem’, stûpah ‘the Buddhist topes’ ; the Sanskrit Grammarian was after Buddha.¹⁰⁴ The learned Professor himself has rendered it very probable that all these words were introduced into the Unnâdisútras ‘after the general spreading of Buddhism, and the erection of Topes in India.’ This was, however, upon the supposition that the Unnâdi sûtras, which are now extant, were identical with the Sûtras of the same name quoted by Pànini. But, the proof is indeed wanting to show that such was the fact ; and the non-existence of the many Grammatical works to which Pànini refers, and the anxiety evinced by the Brahmans to place that sage as Pàninyàdyah, may fairly lead to the inference that all those works, including the particular Unnàdi sùtras referred to by Pànini, had been long ago lost.¹⁰⁵

Since these sheets have gone to the press, and the two first sheets have been printed, I have received from England the invaluable work of Professor Goldstucker on the age of Pànini ; and I here avail myself of that consummate scholar’s remarks on the Unnàdi Sùtras, which directly bear upon the subject.

’It is true (says he) that this grammarian (Pànini) speaks twice of Unnàdis, but he never speaks of Unnàdi-Sùtras. The former term merely implies a list of Unnàdi affixes, and may imply, according to analogous expressions in Pànini, a list of words formed with these affixes ; but it can never imply a work which treats of these affixes and these formations, like the Unnàdi Sùtras which we are speaking of. Between a list of Unnàdis—affixes or words—and Unnàdi-Sùtras, there is all the difference which exists between a lexicographical and a grammatical work. All the conclusions, therefore, which are based on the identity of both, vanish at once—p. 159.

Again, says the same writer : ‘Had Pànini not written the five Sutras (1, 2, 53-57) in which he explains the method of his grammar, or had he explained all the technical terms used by him, the absence of a definition of such terms in the Unnàdi-Sùtras would not justify us in arriving at any conclusion as regards the mutual relation of the two works. But since we know that Pànini does not define all his terms ; and, on the other hand, that a treatise like the Unnàdi-Sutras uses those terms which are defined by him, and exactly in the same sense in which they occur in his work, the only possible conclusion is that this treatise was written later than the Grammar of Pànini.’—p. 170.

I have examined the Unnàdi-Stùras with the assistance of my Pandit ; but have not been able to find any correspondence between them and the Unnàdi in Kachchàyana’s grammar—a circumstance which favors my belief that the former work was also later than, the Pali grammar. This therefore accounts for the mention of ‘the Roman denarius’¹⁰⁶ and ‘the Buddhist sthûpa’ and ‘Jina, the founder of a Buddha sect’ &c. ; although the two last, I may remark, had an origin before the age of Gòtama, as may be shewn from the sermons of that sage himself.

M. Reinaud in his ‘Mémoire Géographique, Historique et Scientifique Sur l’ Inde &c. (Paris 1849) says.

‘Hiouen Thsang attributes to Pànini, as he does to many other notable personages of Buddhism, two existences ; the first he refers to an epoch in which the life of man was longer than at present, and the second about the year 500 after the death of Buddha ; that is, in the time of Vickramaditya, a century after the reign of Kaniska. In his first existence, Pànini professed Brahmanism ; but in in his second, he, together with his father, was converted to Buddhism.’—p. 88.

Founded upon this “Ghost-story”, it has also been supposed by Professor Weber that Pànini should be placed six centuries after Gotama Buddha, or at 140 A.D.¹⁰⁷ But the legendary tale which Hiouenthsang relates, and which is quoted below, far from countenancing this conjecture, merely places him, “at the epoch when the life of man was reduced to a hundred years.” This need not necessarily have been, as we again ascertain from the Buddhistical annals, after Gotama Buddha. It is stated in the Buddhavansa that.

‘At the particular period (of the manifestation of the great elect) the term of human existence was one hundred years; and that it therefore appeared to be the proper age in which his advent should take place.’

This subject may therefore be dismissed by simply subjoining the following translation¹⁰⁸ of the passages referred to in Hiouen thsang :—

’Having travelled about twenty li north-east of the town, Ou-to-kia han t’cha (Udakhànda ?) he arrived at the city Po-lo-tou-lo (Sâlâtura), the birth place of Rishi Po-ni-ni (Pánini) author of the treatise Ching-ming lun (Vyâkaranam.)

’During the times of a remote antiquity, the words of the language were extremely numerous ; but after the world had been destroyed, the universe was found void and waste. Some Gods of an extraordinary longevity descended on the earth to serve as guides to the nations. Such was the origin of letters and books. At the conclusion of this epoch their source enlarged itself, and became boundless. The god Fan (Brahman), and the king of heaven (Indra) established rules and conformed to the times. Some heretic Rishis compos,ed, each of them, some words. Men used them as models, carried on their work, and rivalled with each other in preserving tradition ; but students made vain efforts, and it was difficult for them to comprehend their meaning,

’At the epoch when the life of man was reduced to a hundred years, the Rishi Pánini appeared, who received instruction from his birth, and possessed an immense understanding. Grieved at the ignorance of the age, he longed to abolish all vague and false conceptions, to extricate language from superfluous terms, and to establish its laws. As he was travelling for the purpose of research and instruction, he met the God Tseu thsai (Isvara Deva), and set before him the plan of the work he was meditating.

’Very well said the god Tseu-thsai (Isvara Deva) you may reckon on my assistance.

’Having received his instructions the Rishi departed He then gave himself up to profound researches, and employed all the energy of his intellect. He collected a multitude of expressions, and composed a vocabulary which contained a thousand slokas ; each sloka consisted of thirty-two syllables. He sounded to their utmost limits, knowledge both ancient and modern; and having brought together, in this work, letters and terms, he enclosed it in a sealed envelope, and presented it to the king, who equally prized and admired it. He made a decree, which ordered all his subjects to study and teach it. He added that he, who should be able to recite it from one end to the other, would receive a reward of a thousand pieces of gold. Hence the reason, (thanks to the lessons of successive teachers) that this work is still held in great estimation. Hence it is that the Brahmans of this city possess substantial knowledge, and talents of a high order, and are always distinguished by the extent of their knowledge, and the rich stores of their memory.

’In the city of Po-lo-tou-lo (read So-lo-tou-lo, Sâlâtura) there is a Stûpa. It was in this place that a Lo-han (an Arhat) converted a disciple of Ponini. Five huudred years after is Jou-lai (the Tathàgata) had left the world, there was a great ’Olohan (Arhat) who, coming from the kingdom of Kia-chi-milo (Cashmire) travelled for the purpose of converting people. When he had arrived in this country, he saw a Fan-tchi (a Brahmachârin) occupied in whipping a little boy, whom he was teaching. “Why do you ill-treat that child ?” said the Arhat to the Fan-tchi.

‘I am making him study,’ replied he, ‘the treatise of the doctrine of sounds, (ching-ming Vyâkaranam) but he makes no progress.’

’The Arhat seemed amused, and suffered a smile to escape him. The old Fan-tchi said to him, “The Cha-men (sramanas) possess a tender and compassionate heart, and they pity the creatures that are enduring pain. A man full of humanity smiles upon occasion. I should wish to learn the cause.

’It is not difficult to make you acquainted with it, replied the Arhat, but I fear I shall produce in you a hesitancy of belief. You have, doubtless, heard of a certain Rishi named Ponini, who composed the treatise ching-ming-lun, and that he has left it behind for the instruction of the world. The Po-lo-men said to him—The children of this city who are all his disciples, revere his virtue, and the statue, erected to his memony, exists at this day.

’Well said the Arhat, this child, to whom you gave life, is actually that Rishi. In his former existence, he used his strong memory in studying profane writings : he did not speak, but of heretical treatises, and did not seek at all the truth. His genius and his science perished ; and he coursed though, without stopping, the circle of life, and of death. Thanks to a remnant of virtue, he has been permitted to become your dear son. But profane writings, and the eloquence of the age only impose a useless labour. Can they be compared to the sacred instructions of Jou-lai which, by a mysterious influence, affords understanding and happiness. ?

’In former times, there was, on the shores of the Southern Ocean, a whithered tree whose hollow trunk afforded an asylum to five hundred bats. One day, some merchants halted at the foot of this tree. As there prevailed at the time an icy cold breeze, these men, who were tormented with cold and hunger, collected together sticks and thorns, and lighted a fire at the foot of the tree. The flame increased by degrees, and soon set the withered tree on fire.

’At this moment there was one of the merchants who began, at mid night, to read with a loud voice, the collection of the O-pi-ta-mo (Abhidharma.) The bats, tormented as they were by the heat of the fire, listened however, with desire to the accents of the law, endured the pain without quitting their retreat, and there terminated their existence. In consequence of this virtuous conduct, they obtained the honor of being born again in the class of human beings. They left their families, gave themselves up to study, and, thanks to the sounds of the law which they had formerly heard, they acquired a rare understanding, obtained altogether the dignity of Arhat, and cultivated from age to age, the field of happiness.

’During this latter period the king Kia-ni-se-kia (Kanish’ka) and the Honorable Hie (Arya-Parsvika) assembled five hundred sages in the kingdom of Kia-chi-mi-lo (Cashmire) and composed the Pi-po-cha-lun (the Vibhásha sástra). All these sages were the five hundred bats who had formerly inhabited the cavity of the withered tree. Although I possess a limited intelligence, yet I am one of them. But, men differ from one another, either by the superiority, or mediocrity of their genius. Those essay their flight, while these creep in obscurity. And now, O man full of humanity, you must allow your dear son to leave his friends. In performing this act, that is, embracing the life of a religieuse, one acquires ineffable merit.

’Having finished his discourse, the Arhat gave proof of his divine power by his immediate disappearance.

‘The Bráhman felt himself penetrated by faith and reverence ; and, having loudly expressed his admiration, went and related the event in the neighbourhood, He also permitted his son to embrace the life of a Religieuse, and devote himself to study. As for himself, he was immediately converted; and showed the greatest esteem for the Three Gems. The men of his village followed his example, and, even at this day, the inhabitants are confirmed in their faith, day by day.’

Professor Bohtlingh, in his introduction to Pánini, advances the following arguments founded, as it would seem, on nearly the same authorities as those already quoted.

’As respects the age (he says) in which our Grammarian lived, I will produce some citations which will give some weight to the received opinion, that Pànini lived in the 4th Century, according to our chronology.

’Amara-Sinha, the most ancient lexicographer whose work is extant, lived, as universally received, in the middle of the first centry after Christ. In his work, we meet with a multitude of grammatical expressions and affixes, which occur also in Pànini. From this circumstance alone, one should not venture to decide absolutely on the high antiquity of Pànini ; for, as we have early enough remarked, Pànini’s grammatical terminology is easily discovered amongst his predecessors. By means of the following passages, however. I trust, I shall be authorized to draw a conclusion.

Amara Kosha (S 363, p I. and S. 378 12 and S 384 25 of Colebrook’s edition ) It is said, that the word Rātra at the end of a compound, is masculine except when a numeral precedes ; in this case it is a neuter. According to Pànini (II 4. 29), Ràtra at the end of a compound is always masculine ; Kàtyàyana also appears to maintain the rule unrestricted ; compare with II. 4. 29—S. 363. Z. 4. (Colebrooke’s) (S. 384. 26.) Pathah Sankvyavyayat parah (“patha, on a numerable or an undeclinable word following, is at the end of a compound neuter.)” Pànini (II, 4. 30.) allows only Apatha to be a neuter ; Kàtyàyana enlarges the rule, in the same manner as Amara Sinha ; compare with II. 4. 30.—S. 368. Z. 4. (Colebr. 385. 15.) Punayasudinàmyò twahah parah, “Aha, on punaya and sudina following, (is a neuter.)” With Pànini, Aha, at the end of every compound, is a neuter. Both opinions are given by Kātyāyana (compare with II. 4. 29.) The passage next following is most decidedly only half intelligible, without consulting our grammarian S. 374. Z. 3. (S. 393. 45. C.) Anághantàstetarak tàgharthe. The derivatives in An &c. in the signification of “coloured thereby” &c. (are all of three genders.) An is the first affix in that division of grammar in which the taddhitas and their significations are treated of ; compare IV. 1. 83. The first signification of these affixes, in the formation of adjectives, is Tenaraktan ; compare IV. 2. 1.

’It is indeed, by no means proved hereby that Pànini lived three centuries before Amara Sinha. But then this opinion will acquire probability, when it is stated that Amara Sinha is still more recent than Patangali. In this case we shall still have, between Pànini and Amara Sinha, four Grammarians ; Kàtyàyana, the author of the Paribáshā, the author of the Karika, and Patangali.

’Tradition makes Bhartrihari the brother of Vickramaditya, the author of the Karika. Were this point settled, Patangali would be at most a contemporary of Amara Sinha’s. This tradition is contradicted by another ; according to which Patangali is removed to a high antiquity, and constituted a mythological being in the shape of a Serpent. We shall not, however, take our refuge, by proving the worthlessness of one story by means of another, whilst we have at our command an historical testimony in the annals of Kashmere. The passage contains a grammatical difficulty, which may however, be removed by a small alteration. The verse is expressed in the Calcutta edition as follows (1. 176.)

Chandràcharyádibhirlabdhadesan tasmat tadàgaman.
Pravartìtan sahabhashyan svan cha vyákaranan krit.

’Troyer (in his recent edition of this chronicle, Ràjàtaringini) reads Chandra vyàkaranam for svan cha vyàkaranam, and translates “Tchandràtcharya and others after receiving the commands, explained his (the king Abhimanyu’s) Sāstra, and composed a large commentary and a grammar bearing the name of Chandra.” Seeing, for ought that I know, that nothing is said any where about Abhimanyu’s having composed a Sāstra, the word pravartitan can have no grammatical reference to tadàgaman ; for this word is of necessity a masculine. To join tadàgaman as an adjective to labdhàdesan, and to render it “to come thither (to Abhimanyupura) or to him according to the command received by him” would be too forced. We read labdhvádesan ; then can tadàgaman be easily united with Adesan. Professor Herr Lassen, whom I consulted on this passage, proposed to me to read tadàgame by which the difficulty would be removed. “The causal from. pravart has here, assuredly, no other signification than ‘to set up a thing, to introduce a matter.’” The full sense of the verses will accordingly be the following : “when the teacher Kandra and others had received the command from him, (the king Abhimanyu) thither, (or to him) to repair, they produced the Mahabàshya and composed an accurate grammar.” To corroborate this translation I put down here a quite similar passage from the same work (IV. 487.)

Desàntaràdāgamayya vyàchakshanàn kshamàpatih.
Pràvartayata vichcchinnan mahabashyan svamandale.

“After the king (Gayàpîda,) had brought in expositors from other lands, he introduced into his land the worn-out (no longer extant in a perfect condition ?) Mahabàshya again.” M. Troyer renders vichchinnan Mahabàshyan by “the large well divided grammar,” and remarks in a parenthesis, that this is Pànini’s Grammar. In the first verse that learned man has translated Mahabhàshya quite commonly “a large commentary,” as I conjecture, from this ground that it appeared to him improbable that the study of grammar was pursued already in the twelfth century according to our reckoning. (M. Troyer maintains strongly the chronology of the Cashmirian chronicle.) From this can we explain only his remarks on every passage : “The titles of the books vyàkarana, and Upadêsa, appear amongst the Buddhists, to be equivalent to those of the “purànas” and “tantras” (See the Memoir of Mr. Hodgson in the Transactions of the Rl. As. Soc. of Great Britain and Ireland, Vol II. parts I. and II.

’Kandra occurs in a memorial verse in union with the following ancient grammarians. Indra, Kasakritsna, Apisali, Sâkatàyana, Pànini, Amara, and Ginendra. Bhattogi mentions him and his followers, the Kandras, often.

‘The age of the king Abhimanyu, in whose reign Kandra lived, may be fixed in several ways, all of which lead to the same result. Under Abhimanyu, there appeared in Cashmere, the Bodhisatva Nagârjuna, whose birth the Tibetans place 400 years after Buddha’s death, therefore in the year 143 or 144 before Christ. His preaching as well as the reign of Abhimanyu may be accordingly fixed for the year 100. We maintain the same number, when we adhere to the chronicle of Cashmere. Asoka, the forty-eighth king of the second period is, without doubt, the grandson of Kandragupta. Asoka is removed from his grandfather forty-nine or sixty-two years ; the beginning of his reign will fall in the year 250 before Christ. Five kings, according to the annals of Cashmere, divide Asoka from Abhimanyu. Let us allow each of them as well as Asoka, to reign on an average twenty-five years ; we then have the wished for number of years for Abhimanyu. We arrive close to the same result when we follow the Chinese narrative. This narrative places Kanishka the last of the Turushka princes, and the direct predecessor of Abhimanyu 400 years after Buddha’s death, that is, in the year 143—144 after Christ. Now, since we have discovered, that Patangali’s Mahabàshya through Kandra in Cashmere, already in the year 100 before Christ, came into general use, we are fully authorized to put back the composition of the great commentary on Pànini’s Sûtras to the year 150. Between Patangali and Pànini, there are, as we have remarked above, three Grammarians known to us who furnish us with contributions to Pànini’s grammar. We need accordingly to place the interval between merely two or fifty years, to reach the year 350, in which, according to Kathâ-sarit-sâgara, our Grammarian is to be placed.’¹⁰⁹

Professor Max Muller in reviewing the above arguments says Professor Bohtlingk “endeavored to shew that the great commentary of Patanjali, which embraces both the Varttikàs of Kàtyàyana, and the sûtras of Pànini, was known in the middle of the second century B. C. It is said in the history of Kashmir, that Abhimanyu, the king of Kashmir, sent for Brahmans to teach the Mahàbhàshya in his kingdom. Abhimanyu, it is true, did not reign, as Professor Bohtlingk supposed, in the second century B, C., but, as has been proved from coins, by Professor Lassen, in the first century A. D. But even thus this argument is important. In the history of Indian literature, dates are mostly so precarious, that a confirmation even within a century or two, is not to be despised. The fact that Patanjali’s immense commentary on Pànini and Kàtyàyana had become so famous as to be imported by royal authority into Kashmir in the first half of the first century, A D, shews at least, that we cannot be very far wrong in placing the composition of the original grammar and of the supplementary rules of Kàtyàyana on the threshold of the third century B. C. At what time the Mahàbhàshya was first composed it is impossible to say.¹¹⁰ Patanjali, the author of the great commentary, is sometimes identified with Pingala ; and on this view, as Pingala is called the younger brother, or at least the descendant of Pànini, it might be supposed that the original composition of the Mahàbhàshya belonged to the third century. But the identity of Pingala and Patanjali is far from probable, and it would be rash to use it as a foundation for other calculations.’¹¹¹

All these arguments Professor Max Muller characterizes as “entirely hypothetical.” Indeed they are ; and the reader cannot fail to perceive that though it is quite correct to fix the date of Pànini at some time before Amarasinha, yet no valid ground has been shewn to determine that as having been after the Buddhistical era.

A fact, however, may be here cited from the Buddhistical annals, which apparently countenances the conjecture of Professor Max Muller. It is this ; that Dévala is mentioned as a contemporary of Gotama.

The Atthakathá to the Buddhavansa after alluding to the birth of Siddhatta before he became Gotama-Buddha says :— ’At that period a certain tápaso, named Kàladewalo, who was a confidant of the màharàja Suddhodano, and who had acquired the eight samàpatti, having taken his meal,—for the purpose of enjoying his noonday rest,—repaired to the Tàwatinsá realms, He there found the host of dewatà, in the Tàwatinsa realms, revelling in joy, and in the exuberance of their felicity, waving cloths over their heads, and asked, ‘Why is it that ye thus rejoice, in the fulness of the heart’s delight ? Tell me the cause thereof ?’ The dewatà thus replied, ‘Blessed ! unto the ràja a son is born, who seated at the foot of the bo tree, having become Buddho, will establish the supremacy of dhammo : and we shall be blessed with the sight of the many attributes of his Buddhohood, and with the hearing of his dhammo. It is from this cause that we rejoice.’

’Thereupon the said Dewala the tápaso, on hearing this announcement of theirs, descending from the supreme Dewalòka enchanting with its golden glitter ; and entering the palace of the monarch Suddhodana, seated himself on the pre-eminent throne erected therein. He then thus addressed the ràja who had accorded to him a gracious reception. ‘Ràja to thee a son is born : him I will see.’ The ràja caused the infant, richly clad, to be brought, in order that he (the infant) might do homage to the tàpaso, Dévalo. The feet of the great elect, at that instant performing an evolution, planted themselves on the jatà (top-knot of Devalo) which glittered, from its hoariness, like unto the fleecy white cloud impregnated with rain. There being no one greater to whom reverence is due than to a Buddho elect, who had attained the last stage of existence,—instantly rising from the throne on which he was seated, (Dewalo) bowed down with his clasped hands raised over his head, to the Buddho elect ; and the Ràja also, on witnessing this miraculous result, likewise bowed down to his own son.

’The tàpaso having perceived the perfection of the immortal attributes of the elect, was meditating whether he would or would not become the supreme Buddho ; and while thus meditating, he ascertained by his power of perception into futurity, he would certainly become so ; and smiling said, ‘This is the wonderful mortal.’ He again thus meditated : ‘am, I or am I not destined to behold his achievement of Buddhohood ?’ and said, ‘No, I am not destined : dying in the interval, though a thousand Buddhá be henceforth manifested, it will not be vouchsafed to me to participate in such a blessing : I shall be regenerated in realms inhabited by incorporeal spirits : never shall I behold the wonderful mortal : a mighty calamity is impending over me.’ Having thus divined, he wept.

’The bystanders remarking, ‘our ayyo (revered teacher) having this moment smiled, has now commenced to weep,’ inquired, ‘Is there any misfortune impending over the infant of our ruler ?’ The tàpaso replied, :unto him there is no impending calamity : beyond all doubt he is destined to become Buddho.’ ‘Why dost thou then weep ?’ ‘I am not destined to see so wonderful a mortal as this, on his attaining Buddhohood : most assuredly unto me this is an awful calamity. I weep in the bitterness of my own disappointment.’

If the Kàladèvala ascetic here mentioned ‘who had acquired the eight samàpatti,’ and Dèvala ‘the inspired legislator’ of the Hindu Pauranic legends were identical, we might indeed be warranted in placing Pânini, as ‘the grandson of Dèvala,’ in the third century B. C., or in the third century A. D. But, this is by no means satisfactorily proved. There is the same difference between their names as between Sàkatâyana and Kàtyàyana. The Pauranic legend is also contradicted by Bhottagi. Professor Bothlingk says : “Pànini is, according to Bhottagi, a descendant of Pâṇinâ, who is either a grandson. or more remote descendant of Panin.” Be this however, as it may. Without at all impugning the authenticity or genuineness of the Pali Atthakathà to the Buddhawansa, it may be stated that the identification of these two persons involves us in this difficulty, viz. that to other well-known Hindu works and writers, whom Gotama unmistakably mentions, we must in that case, assign a post-Buddhistic date.

At the time Gotama appeared, the Vedangas had been in existence. At the time the Vedàngas were composed, ‘the period of inspiration,’ according to Brahmans and even Buddhists, had long before ceased.¹¹² Their authors too, claimed no inspiration for themselves. They merely rendered the study of “the revealed literature,” easier. Devala, as an “inspired legislator’” must therefore belong to a period before the Vedànga literature, and anterior to the appearance of Gotama Buddha. Hence the non-identity between Devala and Kàladevala.

If, again, Pànini lived two or three centuries after Buddha, we are sure to have in his sùtras, some allusion to the sage or his remarkable doctrines, which, as M. Burnouf says, ‘found numerous recruits among those who were frightened by the difficulties of Brahmanical science’ There is however no such allusion ; and the word ‘Stûpah.’ if it were not a later introduction, means, ‘not a Buddhist tope,’ but simply ‘a heap of earth,’ as it is said to have been used in the Vedas.

To place Pànini after the Buddhist era (supposing that I have correctly fixed the age of Kachchàyana) is indeed to affirm that the proud Brahamans were indebted for their Grammatical principles to those who had ceceded from their Church, and who were availing themselves of the Brahaman literature ; and at a time too, when Buddhism with the language in which it was promulgated, was fast disappearing in Hindustan. This is indeed so very improbable, especially in view of the fact expressly stated by Kachchàyana, that he had adopted terms given by Sanskrit Grammarians ; that it may reasonably be concluded that Pànini was before Kachcháyana, and therefore before Gotama Buddha.

Professor Goldstucker says : ’Though Yáska be older than Pánini, and Pánini older than Kátyáyana, there still remains the mystery as to the era of Pánini. No work of the ancient literature, within my knowledge, gives us the means of penetrating it. But as the remotest date of Hindu antiquity which may be called a real date, is that of Buddha’s death, it must be of interest to know whether Pánini is likely to have lived before or after this event.

’Not only is the name of Sákyamuni, or Sákya, never adverted to in the Sùtras of Pánini, but there is another fact connected with this name which is still more remarkable.

’The great schism which divided ancient India into two hostile creeds, centres in the notion which each entertained of the nature of eternal bliss. The Brahmanic Hindus hope that their souls will ultimately become united with the universal spirit ; which, in the language of the Upanishads, is the neuter Brahman ; and, in that of the sects, the supreme deity, who takes the place of this philosophical and impersonal god. And however indefinite this god Brahman may be, it is nevertheless, to the mind of the Brahmanic Hindu, an entity.. The final salvation of a Buddhist is entire non-entity. This difference between the goal of both, created that deep and irreconcileable antagonism which allowed of none of the compromise which was possible between all the shades and degrees of the Brahmanic faith, from the most enlightened to the most degenerate. The various expressions for eternal bliss in the Brahmanic creed, like apavarga, moksha, mukti, nihsreyasa, all mean either “liberation from this earthly career” or the “absolute good ;” they therefore imply a condition of hope. The absolute end of a Buddhist is without hope ; it is nirvána, or extinction. This word means literally “blown out ;” but there is this difference, if I am not mistaken, between its use in the Brahmanic and in the Buddhistic literature,—that, in the former, it is employed, like other past participles, in any of the three genders, whereas in the latter it occurs only in the neuter gender, and there, too, only in the sense of an abstract noun, in that of extinction, i. e., absolute annihilation of the soul. I have no instance at my command in which nirvàna, when used in the classical literature, implies any other sense than the sense “blown out,” or a sense immediately connected with it. Thus Patanjali, when illustrating the use of this past participle, gives the instances : “the fire is blown out by the wind, the lamp is blown out by the wind ;” and Kaiyyata who, on the same occasion, observes that a phrase, “the wind has ceased to blow,” would not be expressed by “nirvàno vátah, but by nirvàto vátah,” corroborates the instances of Patanjali with one of his own : “blowing out (has been effected) by the wind.” But Pánini, who teaches the formation of this participle in rule VIII. 2, 50, which has indirectly called forth all these instances, says : “(the past participle of và with prefix nir is) nirvàna (if the word means) ‘free from wind,’ (or, ‘not blowing, as wind’).”

’This is the natural interpretation of Pánini’s rule. Kàtyàyana, it is true, gives a Várttika, which corrects the word aváte into avàtàbhidháne “(if it have) not the sense of wind (or of blowing) ;” yet it is very remarkable that Patanjali, in commenting on this Várttika, does not interpret its words in his usual manner, but merely adds to them the instances I have just named ; it is remarkable, too, that he introduces them with the observation : “(this Várttika is given in order to show) that (nirvána) is also or is emphatically used in the following instances.” Still he has no instance whatever for the sense stated by Pànini, and his word “also” or “emphatically” does not appear to be justified by the criticism of Kátyáyana, which simply corrects the word aváte into avàtábhidhàne without any additional remark.

‘In short, my opinion on this Vàrttika is analogous to that which I have expressed in previous instances. The sense of nirvàna, “free from wind (or not blowing),” had become obsolete in the time of Kàtyàyana, who merely knew that sense of it which found its ulterior and special application in the nirvána of the Buddhistic faith. But since there is no logical link between this latter word and the nirvàna, “wind-still,” of Pànini ; and since it is not probable that he would have passed over in silence that sense of the word which finally became its only sense, I hold that this sense did not yet exist in his time ; in other words, that his silence affords a strong probability of his having preceded the origin of the Buddhistic creed.’¹¹³

Dr. Weber after reviewing the remarks of Professor Goldstucker,¹¹⁴ concludes by exclaiming—“And this then is all wherewith Goldstucker is able to prop up his opinion of Pànini’s priority to Buddha—a daring undertaking indeed ! and at the same time an ignominy of all that speaks to the contrary, which excites surprize” ! The learned Doctor then proceeds to adduce proofs in support of his own opinion—that Pànini was later than Buddha ; and produces four items, such as the frequent mention of bhikshu, sramana, chìvara, munda, &c. &c. ; 2. That the Buddhists themselves consider Pànini as having lived after Buddha’s time. For (says he) Burnouf informs us from the Aryamanjusrí Mûlatantra, “It is thus that Sâkya predicts the future advent of Nagarjûna 400 years after him. He likewise announces that of Pànini, of Chandragupta, and of Aryasangha” ; 3. That Pànini’s vocabulary is proof of the proposition ; and 4. That no mention is made, among other names, of Pànini in the Rik or Rik. Sanhita.

Taking the last ground first, it appears to me that (without entering into other questions which arise upon this point), the inference here sought to be deduced is of no greater weight than the like inference drawn by Professor Goldstucker from the fact, that no mention is made of Sàkya by Pànini. As to Pànini’s vocabulary, I fail to perceive anything which leads to a conclusion one way or the other ; and the words given by Dr. Weber as “actually Buddhistic terms” prove, in my humble opinion, nothing. For the Buddhists have scarcely any words which they have not taken from the Brahmans ;¹¹⁵ e. g. the titles veyyàkarana and Upadesa, to which reference is made in one of the above extracts, are for the Abhidhammapitaka, and “the hymns of joyous inspiration” of the Buddhists. It would also appear from the Buddhist works that the Lokàyata or the Jainas¹¹⁶ had an existence before Gotama. The Jainas had doubtless their bhikkhu mendicants, their Samana or Sàvaka hearers. They wore chivara robes ; and had, like the Yavanas, their heads bare. It would, therefore, to say the least, be rash to affirm that the other words given by Dr. Weber as “actually Buddhistic terms” had not been known to other pre-existing Sectarians ; more especially as we find in the text books of Bhuddhism that Brahmans had frequently addressed Buddha with the epithets “S(r)amana bhavat Go(w)tama.”

We are also told that the Buddhists themselves consider that Pànini is after Gotama Buddha. This is a mistake. There is no such belief entertained by the Buddhists in Ceylon. There is no mention of Pànini in any of the Ceylon Buddhist works. The authorities referred to are from the Nepaul works, and they are indeed no authorities at all. The prophecies which are related in them are the interpolations of seceders from the Buddhist Church. The predictions given in some of our own books regarding persons who lived after Gotama, such as Wijaya, Asôka, Nàgasena, &c., are the additions of zealous Buddhists, anxious to uphold the characters of whom they wrote, and to procure for their acts all the authority and weight with which such a prediction on the part of the sage was calculated to invest them. And, I may conclude by remarking, that the works themselves, in which these pretended prophecies¹¹⁷ are recorded, are comparatively modern works ; and therefore not the text-books of Buddhism.

To return to the subject. The proof adduced by Professor Goldstucker is not the only evidence on this matter. It is capable of more satisfactory proof. The best mode in which the dates of authors may be ascertained, in view of the scanty information which Asiatic biography affords us, is by the references which are made by writers whose dates have been ascertained. Now, few dates have been better ascertained than that of Gotama Buddha. If it be not 543 ; it is assuredly 477 B. C. If, therefore, the personages, who figure most conspicuously in Brahmanic history, are unmistakably mentioned by Gotama, there can be but little doubt of their existence having been anti-Buddhistic. By an investigation into the Buddhist Literature, we obtain a result which, to say the least, is satisfactory. According to Shadgurusishya’s Commentary on Kàtyàyana’s, Sarvànukrama : (and here I am indebted to Pr. Max Muller for the extract, see his Sanskrit Literature, p. 230, et seq.) Saunahotra, a descendant of Bhàradvàja of the race of Angiras, who entered the family of Bhrigu, took the name of Saunaka ;’ the Reverend Asvalàyana was Saunaka’s pupil ;’ and Kàtyàyana studied the works of both Saunaka and Asvalàyana.’ The same authority places Vyàsa about the same date, if not a little anterior to Saunaka. And Vyàsa, we learn from other Brahmanical sources, was the son of Paràsara. If therefore, we adjust these names according to their dates, we obtain, 1 Paràsara, 2 Vyàsa, 3 Saunaka, 4 Asvalàyana, and 5 Kàtyàyana. These facts may perhaps be relied upon : but I must object to their being applied to the “idle stories” of Kathà Sàrit Sàgara of Dr. Somadeva of Kashmir, or to any chronological calculations being built upon their basis.

I have already had occasion to refer to this, which is the same work in substance as the Vrihatkatha. It is confessedly not a book of any authority.—It is a compilation of fables, abounding with the “marvellous.” It was composed without reference to History or Chronology. It has not even followed the chronological system of his contemporary Kalhana Pandit. The author has strung together various stories without order or date ; and, I am persuaded, no one would be more amused than Somadeva himself, if now alive, at the historical importance attached by Europeans to his “ghost stories,’ lost in the unfathomable depths of his “Ocean-of-Rivers-of-Stories.” It may indeed be asserted, (says Professor Wilson) that the Càthà Sàrit Sàgara, or rather the Vrihat Cathà, is not a much better guide than the Bhòjaprabanda, and that a collection of idle tales is bad historical evidence : it must be remembered however, that those tales are not of Sòmadeva’s invention : he has only the merit of telling them in his own way, and of having collected them together from various quarters. Thus we have most of the legends relating to Vikrama, which constitute the Sinhàsana Dvàtrinsati and Vetala Pancha Vinsati, and we have also a very considerable portion of the Hitopadesa or Panchatantra comprised in this selection.’¹¹⁸

Be the authenticity of the matters in Sòmadva’s work, however, as it may. It cannot for one moment be maintained that it lends any authority to the identity (upon which several writers have based their inferences) between king Nanda in connection with whom Kàtyàyana is mentioned, and the predecessor of Chandragupta :¹¹⁹ My pandit has shrewdly intimated to me the probability, in view of the age of Asvalàyana, as it appears from Buddha’s discourses, and the shifts to which Somadeva resorts to connect Kàtyàyana’s story with other fables, that the mention of Channakka and Nanda, had led the writer to identify some previous Nanda with the predecessor of the Sudra king. This is not unlikely in the same manner that the Nepaul Buddhistical writings, to which M. Burnouf assigns a Cashmirian origin, had mistaken Dharmàsoka for Kàlàsoka : and I need hardly add that no inference can be drawn from the mention of Channakka. No arguments are indeed necessary to prove that the scheming, and treacherous Purōhita Brahman Channakka, who figures so conspicuously in the Hindu Nàtakas and in our own Buddhistical annals, was different from the venerable sage Saunaka, the preceptor of Asvalàyana “celebrated among the rishis as the glorious, having seen the second Mandala, and who heard the collection of the Maha Bharata.”

Nor has this, I believe, been attempted. But the authority upon which the identity between the writer of the Varttikas and the minister of King Nanda of Pàtaliputta is attempted to be established, would have us believe that the former was also the contemporary of Pànini ‘and actually defeated Pànini in a grammatical controversy.’ ! !

To return from this digression : the dates of Paràsara and Asvalàyana, however, may be ascertained from the Buddhistical annals. True it is that we cannot fix them exactly ; but if it can be shewn that they had an existence before Gòtama Buddha, it is sufficient for our purpose. Now, any one who has the slightest acquaintance with the history of Buddhism, and the disputes which the principal fraternities of Brahmans are said to have had with Gotama, cannot fail to identify the youth mentioned in the following extract, (the first) with one of the descendants of Paràsara, the Hindu sage ;—and likewise the mànavo in the second extract, with a descendant of the notable Asvalàyana of Hindu legends.

In the Majjhima Nikàya, from which I have already quoted, at p. xlv., the following passage occurs :—

1. Evan me sutan : ekan samayan Bhagavà Kajangalàyan vihariti Mukheluvane. Athakho Uttaro mànavo Pàràsariyante vàsi yena bhagavà tenupasankami……. …………… Deseti no Uttara Pàràsariyo brahmano Sàvakànan indriyànan bhàvananti ? Deseti bho Gotama Pàràsariyo brahmano sàvakànan indriyànan bhávananti.

‘Thus have I heard. When, at a time, Bhagavà dwelt at Mukheluvana in Kajangala, a youth (named) Uttara, a pupil of the Pàràsariya fraternity, went to the place where Bhagavà was……………… …… (Gotama inquired) Uttara, does the Brahman (your teacher) of the Pàràsariya fraternity teach Indriya bhàvanà to pupils ? Sir, Gotama (replied Uttara) the Brahman of the Pàràsariya fraternity does teach Indriya bhàvanà to pupils.’

In the Assaláyana Suttan, where a dialogue is given between Gotama and one of the Assalàyana family, a distinguished member of the Brahman fraternity, as to their alone being ‘the highest’ race, ‘the purest,’ ‘the projenitors of Mahà Brahmà,’ and who had ‘sprung from his mouth ;’ the following passage occurs :—

2. Evan me sutan : ekan samayan bhagavà Sàvattiyan viharati Jetavane ………… Tena khopana samayena nànà verajjakànan brahmanànan panchamattàni Brahmana satáni Sàvattiyan pativasanti kenachadevakaraniyena. Atha kho tesan Brahmanànan etadahosi : ’Ayan kho samano Gotamo chatuvannin suddhin paññàpeti ; konukho pahoti samanena Gotamena saddhin asmin vachane patimantetun’ti. Tena kho pana samayena Assalàyano nàma mànavo Sàvattiyan pativasati ;—dhàro vuttasiro sàlasavassuddesiko jàtiyà ; tinnan vedànan pàragú sanighandu ketubhànan sakkharappabhedànan itihàsa panchamànan ; pàdako veyyakarano lokàyata mahà purisa lakkhanesu anavayò.

’Thus have I heard : At a time Bhagavà dwelt at Jetavana in Sàvatti ; and at that time about five hundred Brahmans of different countries,¹²⁰ also resided there for some purpose. They thus thought : ‘this Samana Gotama proclaims the purity of (all) the four classes : who is able to dispute with Samana Gotama on this matter ?’ At this time there lived at Sàvatti a youth named Assalàyana. He was young, head-shaven,¹²¹ and about sixteen years of age (from his birth.) He had mastered the three Vedas, which, with (the supplements) Nighandu, Ketubhà, and the distinction of Akkhara (letters) &c., have Itihàsa for a fifth.¹²² He was a Pàdaka,¹²³ and a Veyyàkarano (grammarian). He was accomplished in Lokàyata,¹²⁴ and in the science of Purisa-lakkhanà.¹²⁵

More direct evidence than the above, can scarcely be adduced of the identity of persons in Asiatic History : and, if, as I apprehend, Assalàyana here named, was a descendant of Asvalàyana, ‘celebrated amongst the Rishis’, the claims of Pànini to an antiquity remoter than Gotama, are undoubted. It may thence be concluded, that the Buddhist Grammarian availed himself of the Pàninya Vyàkarana, between which and the Pàli aphorisms there is so much correspondence.

But, there is the same correspondence between the Sanskrit Varttikas, and the Pali Vutti. Was, therefore, Kàtyàyana anterior to the Buddhist Pali Grammarian ?

I believe it is not stated that Kàtyàyana was a pupil of Asvalàyana. All that is alleged in respect of the former, is, that ‘he, having mastered the thirteen books of Saunaka and of his pupil, composed several books himself.’ What time, therefore, elapsed between them does not appear ; nor is there any valid reason to admit between them ‘only an interval as large as that between teacher and pupil, or between father and son.’¹²⁶ If, however, such was the fact, Kàtyàyana may, for the reasons already adduced, be placed before Gotama ; for, all the testimony adduced in favor of Pànini’s antiquity applies equally to Kàtyàyana.

But, supposing for the sake of argument, that Katyàyana’s Varttikà were post-Buddhistical, and after the age of Kachchàyana ; and that it is very improbable that the former availed himself of the language of a Buddhistic Pali writer : I beg to submit that this hypothesis does not shake the testimony in regard to Mahà Kachchàyana’s identity with the author of Sandhikappa ; for all that may be fairly inferred in that case, is, that the Vutti in the Pali work, like the Varttikà to the Sanskrit Grammar, were written at a subsequent period and by a different person.

In noticing this question, it must be borne in mind, that although tradition in one voice ascribes the authorship of the Pàli Suttans in the Sandhikappa to Mahà Kachchàyana, yet that writers are divided in their belief as to the Vutti having been written by that distinguished hierarch of the Buddhist Church.¹²⁷ This very difference of opinion disproves the alleged “tendency of later Buddhist writers to refer the authorship of their works to names famous in ancient Brahminic history.”—(Max Muller, p. 303.)

Such are the facts and circumstances connected with the age and authorship of this Grammar, to which I desire to attract public attention. It is indeed possible that future researches may enable me to adduce more satisfactory proofs upon those points, or materially to qualify the inferences and conclusions here drawn. But, so far as my humble researches have hitherto extended, I incline to the opinion that this Pali Grammar was written by Mahà Kachchàyana in the latter-half of the sixth century before Christ.

As already stated this work is intended to illustrate the grammar of the language of Gotama Buddha’s discourses. This may lead to the inference that it was, in some degree, different from another dialect which had also received the appellation of Màgadhí. Be this however, as it might. The Páli is essentially the language of Buddhism. Nothing is known definitely of the state of its cultivation previous to the establishment of Buddhism by Gotama.

All that may be confidently advanced of times previous to the Buddhist era, is that in remote antiquity a tribe of people settled themselves under Bhàrat in the Aryadesha or Aryàvarta, the region commonly known as Central India, between two lines of mountains the Himalayà on the North-East and the Vindhya on the West.¹²⁸ According to the traditions current in India there were numerous kings from the war of the Mahà Bharat to a comparatively very late period ; but all the dynasties, though existing in different parts of the Aryadèsha, were founded by one and the same race of people, whom we may designate the Arians, consisting of four classes, the Kshestriyas, or the royal (military) tribe—the Brahmans, or the sacerdotal class—Vaisya, or the commercial—and the Sudra, or the servile.¹²⁹ Of the several dynasties, one was that of Magadhas. It numbers a connected chain of thirty-five kings from Sahadeva, who reigned at the termination of the war of Mahà Bharat, to the Buddhistical era in the reign of Ajàtasatta.

The religion of the Magadha people was doubtless Brahmanism from a very early period, though at the Buddhistical era, it branched off into different sects.¹³⁰ Yet the undoubted existence at this time of a cultivated dialect, peculiar to the Magadhas, called the Màgadhí, proves that, whilst sharing with the Brahmans their religion, the Magadhas had a language of their own, fundamentally the same as the Sanskrit, and exhibiting the nearest relation to the earliest form of the language of the Brahmans.

Although there are now several dialects, including the Pali, which receive the name of Pràkrita, it may nevertheless be gathered from a variety of circumstances to which I shall hereafter refer, when I come to speak more particularly of the relation which the Màgadhi bears to the Sanskrit,—that the language which had at first received the name of Pràkrita is either now entirely lost, or has been absorbed into other forms of speech, leaving behind but few, if any, traces of the parent stem.

It is also very probable from historical and philological considerations, that the Màgadhi and the Sanskrit are two of those forms. Against this view of the question may be urged the absence of a Màgadhí literature before the Buddhistical period, and the undoubted existence of the Sanskrît.

The existence of a Sanskrit literature, before the Buddhistical era, was owing to the pre-existence of Brahmanism ; and if Buddhism had existed in Magadha before 628 B. C., we should doubtless have some evidence of the existence of the Màgadhi. But, such was not the fact. Till Gotama proclaimed his religion, and sought to disseminate it, far and wide, throughout Asia, the kings of Magadha had no necessity for a display of their language. Till then they had no religious feuds ; no sectarian animosities. Till then there had been no contentions with any who held an antagonistic faith. Both the Magadhas and their Arian brethren were the adorers of the same gods. Their sacred books, through which alone the existence of a literature may be ascertained of nations of antiquity, were identical with those of the Brahmans. Although they had a language of their own, the Màgadhi ; yet they had little to record in it apart from religon, to the exposition of which the Brahmans had laid exclusive claims ; and if they had a literature of their own, which doubtless they had, its destruction through the agency of Brahmans, is sufficiently proved from the political changes which Magadha underwent in the fifth century, and, above all, from the fact that not many centuries after their promulgation, the sacred scriptures of Buddha were re-transferred from Ceylon to India.

When, however, the Magadha kings had embraced a new faith, a necessity arose which had not existed before ; and that was the recording of the doctrines of Buddhism in books. And, although they shared the fate of their scientific works, yet by the early transfer of the former into different countries at a time when the Buddhists had not been subjected to the oppressions of the Brahmans, we have some means of knowing the existence of the Màgadhi, and that it had already attained the refinement it now possesses, at the time of Gotama Buddha’s advent.¹³¹

The absence, again, of a literature is not proof positive of the non-existence of the language itself, just as the absence of Sanskrit Inscriptions of a data anterior to the Buddhist era, is no evidence of the previous absence of the Sanskrit language. The non-existence moreover, of historical or literary records, testifies nothing further than the imbecility, or the apathy of a people, who had no taste for literary pursuits, or a regard for history, or historical proceedings. This, indeed, was the case with the early Brahmans, as well as the early Magadha princes. At first both had a common interest, and possessed in common a literature which was inseparably connected with their religion also common to both ; and, therefore there was no inducement to raise monumental erections, which in comparatively later times were the result of a rupture of those ties which had formerly bound them together—the consequence of a new faith—the work of vain kings, wishing to proclaim th supremacy of a particular princedom, and to disseminate the tenets of a newly embraced creed.

The most important fact, however which demonstrates the existence of a Màgadhì literature before the Buddhist era is that furnished by the language itself, viz.—its high state of cultivation at the period above mentioned. This proves that it had been in existence for a long time previously. For it could not have attained its perfection in a day. It must have been the work of time, the result of ages, the slow progress of innumerable changes. A considerable period too, must have intervened between the time when it first started into existence, and the period at which it was so much celebrated as to induce Pali scholars to designate it the Múla bhàsà, ‘the original language’

There are two theories current with regard to the comparative antiquity of the Sanskrit and Pali. Some regard the former as the original, and the latter as a derivative from it ; whilst others affirm the superiority of the Pali over the Sanskrit, and assign to it an origin before the language of the Vedas. But, nearly all Brahman and European writers on the subject, are agreed in considering the several Indian dialects, which are generally designated the Pràkrita, as inferior in structure to the Sanskrit, and, therefore, as being deduced from it,

These are questions which are involved in doubt and obscurity ; and upon which there is still great misapprehension. The uncertainty which prevails on the subject, appears to me to originate in the minds of men, who, without paying sufficient attention to a most important consideration—the history of Buddhism and of its language, the Pali, suffer their judgment to be swayed by two circumstances, viz., the superior structure of the Sanskrit, and the comparative antiquity of its records.

Those who assign to the Sanskrit a superiority over the Páli do so upon the grounds that the former is more “finished” and “elaborate” than the latter ; and that the Pali cannot be traced, through its literature or religion, to a period before Gotama—whereas the Vedas and the Mahà Bhàrata¹³² are confessedly of a much anterior date.

I readily accord to the Sanskrit, as to its “wonderful structure” and “refinement,” the superiority, which one of its most accomplished and able students, Mr. Colebrooke, assigns, when he declares it to be—‘a most polished tongue, which was gradually refined, until it became fixed in the classic writings of many elegant poets, most of whom are supposed to have flourished in the century preceding the Christian era.’ I also admit the existence of positive evidence to prove that the Vedas, and therefore the faith it proclaims, had been known before Gotama established his religion by means of the Màgadhì.

But these admissions, I presume, do not affect the theory of some of our ablest pandits, viz., that at a very remote period, some one Prákrita [‘mother’] dialect, which can no longer be identified, or may not now be in existence in its original development, was the principal tongue of the Arians ; and that the Páli (not the dialect spoken at present in Magadh or Behar, nor the Mágadhì of the Indian Grammarians), and the Sanskrit, are both branches of the same unknown original stock.

In the investigation of this subject, it may not be unprofitable to notice, to the extent of our limited information and means, (1) the subdivisions of the principal Indian languages ; (2) the number of dialects comprehended in the term Pràkrita ; (3) the acceptation of that term by nations and literary men ; (4) its correct signification ; (5) the conjectures as to its identification ; (6) its relationship to the Sanskrit ; and (7) its high state of cultivation and decline in Asia.

I. As to the classification of Indian languages, we have the following interesting passage in the Dandialankàra, or as it is otherwise called, Kàvyadarsha.

Tade tadvànmayan bhúyas¹³³
Sanskritan Pràkritan tathà ;
Apabhransascha misranche
Tyàhuràptàschatur vidham.
Sanskritan nàma daivìvà
Ganvàkhyàtà mahàrshibih ;
Tadbhavan Tatsaman Desì
Tyanèkah Pràkritakramah.
Mahàràshtràsrayàm bhàshàm
Prakrishtam Pràkritam viduh ;
Sàgarah sùkti ratnànàm
Setubandhàdi yanmayam.
Saurasenìcha Làtìcha
Gowdìchà’nyàpi tádrisì ;
Yànti Pràkrita mityèva
Vyavahàreshu sannidhim.
‘Abhiràdi girah kàvye
Shv’apabhransa itismritah ;
Sastreshu Sanskrità danya
D’apabhransa tayoditam—
Sanskritam sarga bandhàdi
Pràkritam skhanda’ kàdiyat ;
Ousharàdì-ny’apabhranso
Nàtakà ditu misrakam.
Kathàdi sarvabhàshàbih
Sanskritenacha baddhyate ;
Bhútà bhàshà mayim pràhur
Atbhùt àrtham Vrihatkathàm.

That is to say—‘Preceptors declare that the (above) compositions consist of four kinds (of language) ; Sanskrit, Pràkrit, Apabhransa,¹³⁴ and Misra. The speech of the gods, which is defined by great sages, is named Sanskrit. The Pràkrit is of various orders ; viz. Tadbhava (born of the Sanskrit,) Tatsama (which is equal or similar to the Sanskrit), and Desi (provincial or local). The language current in Maharàshtra is known as the principal Pràkrita—that which is an ocean to gems of beautiful language, and that in which Sètubandha,¹³⁵ &c., are composed. The languages of Saurasena, Làta, Gowda, and such like,¹³⁶ are usually¹³⁷ treated under the very name of Pràkrita. Dialects like the Abhir๳⁸ &c., are (alone) reckoned as Apabhransa in poetry ; but in the Shàstras¹³⁹ all languages besides the Sanskrit receive that name. Compositions which are divided into chapters, are in the Sanskrit language ; those which are (skandaka) composed in one entire body, are in the Pràkrita ; those like the Aushra are in the Apabhransa ; and the Dràmas are (misra) in a mixed dialect. Historical writings &c. are composed in the Sanskrit, as well as in (other) dialects ; but the Vrihatkathà of marvellous import, is said to be only in the speech of the demons.’

Obs.—(a) The division of languages here spoken of, is founded upon the authority of Brahman writers, and their usages. That division is into four ; viz., Sanskrit, Pràkrit, Apabhransa, and Misra—epithets which have certain meanings, and which are descriptive of the different languages for which they are names. The Sanskrit is ‘the language of the gods’—that is, the sacred language of the Brahmans, in which historical writings, and the Shástras are composed, having subdivisions.

The Prakrit is various, or, in other words, has ‘several orders’ or ‘dialects’ ; and is divided into three classes, viz., tadbhava, tatsama, and desi. The first comprehends “derivatives,” those which have sprung from the Sanskrit which have undergone various mutations in course of time and which may yet be traced to the parent stem. The second includes ‘sister-dialects,’ those which are tatsama ‘co,-equal’ or, ‘similar,’ or ’bear a resemblance to the Sanskrit ; and which stand in fraternal connection with the Sanskrit—not in the relation of descent from it—not begotten by it ;—but sprung from the same shoot with it. The third is a name for provincial dialects, or non-Sanskrit languages,—those which cannot be traced to a Sanskrit origin, or do not exhibit any evidence of fraternity to it. They are dèsi, peculiar to a country, or, as the Sinhalese call them, nipan ‘born in a country,’ and therefore local.

Of these various Pràkrit dialects, the Mahàràstri is in the estimation of my authority, the best cultivated dialect. Perhaps, it was, at the time Dandi wrote, which is supposed to have been in the twelfth century of the christian era—¹⁴⁰ ‘an ocean to gems of beautiful language’—a dialect, which, like the ‘vast profound’ abounded in gems of lofty expressions—(that is ; rich in expression, and copious as a language), as evidenced by works like the Sétubandha &c. Prákrit compositions, unlike the Sanskrit works, which are subdivided into chapters, are only in entire books.

Although by Pràkrita such languages as the above are alone indicated ; yet (says Dandi) ‘the dialects of Saurasena ; Láta, Gowda, and such like (by which the commentator says the Màgadhí and Panchàli were meant) are usually treated under the very name of Pràkrita.’

The name of the third division of languages is Alpabhransa or ‘the ungrammatical’—‘a jargon ;’ such as the Abhiri and Chandàli—dialects used by herdsmen, and by persons of the lower orders ;¹⁴¹—for compositions in which, Dandi refers the reader to Oushra.

Mìsra is the name of the last division of languages,—that dialect which contains a Mixture of all the other classes of language—not a mixture of merely words, but entire passages of different languages interlarded with the Sanskrit as in the dramas.

Another division of languages, as stated by Dandi, and according to the Shástras of the Brahmans, is into Sanskrit and Apabhransa, the latter including all dialects besides the Sanskrit. Historical compositions are written in all the languages comprised under these two heads ; but Vrihatkatha alone (which abound with the ‘marvellous’) are to be found in the speech of the demons, by whom I understand the barbarians, or the aboriginal inhabitants of those provinces which were afterwards peopled by the Arian race.

Obs.—(b.) The above inferences which are forced upon us by the phraseology adopted by Dandi and by the explanations of his Commentator, authorize five important conclusions, 1st, That although the Prakrita has become manifold, or has assumed diverse forms ; yet it may be regarded as tatsama, or, as having a fraternal relation to the Sanskrit. 2. That the principal Prakrit dialect, which (whether rightly or wrongly we shall not here pause to consider) is generally regarded as the Maharastri—is rich in expression, and copious as a language ; 3. That although originally there was but one Prakrita dialect, many tongues have nevertheless been by the usage of Brahmans improperly comprehended under the appellation of Prakrit ; 4. That all dialects except the Sanskrit are by them designated Apabhransa ’the ungrammatical ;¹⁴² 5. That the Pali (Mágadhì) and the language of Punjab (Panchàla) may be regarded as two dialects bearing some particular relation to each other.¹⁴³

II. The dialects which receive the designation of the Pràkrita are many ; and they are to be generally found in the dramatic writers. The Commentators, Grammarians, and Rhetoricians, who have written on the languages contained in the Hindu plays variously define them, with reference to the patois of the drama ; and there seems to be much misapprehension, and much more confusion in their classification.

1. Monsieur Adolphus Fredricus Stenzler in his preface to the Play entitled the Mritcha Katika, after quoting from a Commentator states :

Dialecti non Sanskritœ, quœ in dramates adhibentur, sunt aut Prakritœ aut apabhransicœ. Pràkritœ dialecti sunt Saurasenì, Avanti, Prachyà, Màgadhì. Apabhransicœ dialecti sunt Sakàrà, Candàlì, Sabarì, Dhakki.’¹⁴⁴

2. The same Commentator, to whom Monsieur Stenzlir was indebted for the above, also states :—

‘Màgadhyavantijà Pràchyà Saurasenyardha Màgadhi
Bàlìkà Dàkshinàtyàcha sapta bhàshàh prakìrtitàh.’

That is : ’It is declared that there are seven dialects (in Dramatic works) Màgadhì, Avanti, Prachyà, Saurasenì, Ardha-Màgadhì, Bahlìkà, and Dakshinàtyà.¹⁴⁵

3. There was a tradition current, in the thirteenth century, that there were only six Indian dialects which received the name of bhàsha ; and hence the appellation of Shad-bhàsha parameshvara, to one who is a proficient in six languagesl These are enumerated in the Ratna-kosha ; and they are—

‘The Sanskrit, Pràkrita, Màgadhì, Saurasenì, Paisàchì, and Apabransa.’

4. In the Selalihini-Sandesa the philological acquirements of Totagamuva, a Ceylonese, who lived about 1415 A.D.¹⁴⁶, are said to have been so vast that Brahmans visited him from India to pay the homage due to a Shadbhàshà-parameshvara.¹⁴⁷ The six bàsas with which he was acquainted are enumerated thus—‘The Sanskrit, Pàli (Màgadhì), Pràkrit, Saurasenì, Paisàchi, and Apabhransa.’

5. Hemachandra, a Grammarian of the Jaina sect, who flourished in the thirteenth century, and wrote a Pràkrit Grammar as an eighth adhyàya after the seven adhyàyas of his Sanskrit Grammar¹⁴⁸ enumerates also six bashàs ; and they are Pràkrit bhàsha (which Mr. Cowell calls ‘principal pràkrit’) the Saurasenì, Magadhì, Paisachì, Chulika Paisachì, and the Apabransa bhasha.

6. Mr. Colebrooke gives the following translation of a passage¹⁴⁹ in a work on Rhetoric compiled for the use of Mànïkya Chandra a king of Tirhat, but the name of the work is not given :—‘Sanscrita, Pracrita, Paisachî and Màgadhì are in short the four paths of poetry. The gods, &c., speak Sanscrita ; benevolent genii, Pràcrita ; wicked demons Paisachi ; and men of low tribes and the rest, Màgadhì. But sages deem Sanscrit the chief of these four languages. It is used three ways ; in prose, verse, and in a mixture of both.’

‘Language again, the virtuous have declared to be fourfold, Sanscrita (or the polished dialect) Pràcrita, (or the vulgar dialect) Apabhransa (or jargon), and Misra (or mixed). Sanscrita is the speech of the celestials, framed in grammatical institutes ; Pràcrita is similar to it, but manifold as a provincial dialect and otherwise, and those languages which are ungrammatical are spoken ín their respective districts.’

7. Vararuchi in his Prákrit Prakasa gives a principal Pràkrita, supposed to have been the Maharastri, and enumerates three others, the Saurasenì, Magadhì and Paisachì, for the elucidation of which he assigns but a very small portion of his work.

Obs.—From the above authorities, taken in connection with the foregoing historical notices on the Pàli language, several important facts may be deduced. First, that there was originally but one language which received the name of Prakrit, and that since it became ‘manifold as a provincial dialect,’ its name has been applied to other tongues : and this is attested by another fact, that the farther back we go in point of time in search of the Prakrit, the fewer are the dialects treated of by authors under that name ; and that Vararuchi, the Pràkrit Grammarian of the times of Vikramàditya, treats chiefly of one Pràkrit dialect, the ’dialectus principua” of Professor Lassen. From the names, too, given to the Shad-bhasha of comparatively an ancient date, it may be inferred that the many varieties given in modern works under the name of Pràkrit are merely “the subtle refinements of a later age,” in order to distinguish the fifty-six¹⁵⁰ different languages “spoken in their respective districts,” some of them being Misra or mixed ; and that philologically speaking languages were originally three-fold, viz, those which have been designated by the epithets Sanskrita, Pràkrita, and Apabhransa. These in the language of the Brahman Rhetorician above quoted upon Mr. Colebrooke’s authority, were 1st, “the speech of the celestials framed in Grammatical Institutes ;” 2nd, a “dialect similar to the last ;” and 3rd, a “language ungammatical.”

The first, it would seem, became early fixed, as the dialect of the Vedas—the last was the spoken language of the rude aborigines of the Dekhan—and the second, the speech of different Arian nations, which underwent different modifications ; such as the Pali of the Sinhalese ; the Bactrian-Pali¹⁵¹ of the bilingual coins of the Greek kings ; the idiom of the Zendavastâ ;¹⁵² the dialect of the Nepal Buddhists, and the Jains ; the speech of the mass of the people with whom the Greeks came in contact after Alexander’s invasions ; the language of Asoka’s Inscriptions ; and the Prâkrita of the plays.

Although the original Pràkrit has thus undergone different changes in different countries ; yet it is clear that the differences which at first distinguished several Pràkrit dialects from each other, were not such as to render any one of them altogether unintelligible to the great mass of the people. This was, obviously, the reason why the Màgadhì, “the speech of the Brahmans and the Aryas”¹⁵³ is represented as having been generally intelligible to Gotama’s varied congregrations; why the ancient Translators of several Sinhalese books¹⁵⁴ into the Pali say, that by such transposition they “would render a facility to both the inhabitants of this island, and of other lands;” and why Buddhaghosa, a consummate Sanskrit scholar of the 5th century, preferred the Pàli to the Sanskrit in translating the Sinhalese Atthakathà.

III. As to the acceptation of the term Pràkrita amongst nations and literary men;—

1. Mr. Colebrooke, according to a forced etymology, says ’the most common acceptation of this word is outcast, or men of the lowest class; as applied to a language it signifies vulgar.¹⁵⁵

2. Cowell, the translator of the Prákrita Prakása, says “Pràkrit is the common name given to the various dialects which sprang up in early times in India from the corruption of the Sanskrit, and as the word is used by grammarians, it signifies, ’derived, thereby to denote its connection with the original Sanskrit.”¹⁵⁶

3. The authority upon which the above opinions are founded, is taken from Hemachandra, who defines—Prakritih Sanskritam tatra-bhavam tata àgatam và, pràkritam—‘Pràkrit has its source in Sanskrit, and is that which springs or comes from it.’

Obs.—Though willing to admit the above to be the acceptation of this word amongst the Brahmans generally, to distinguish their so-called heaven-derived religion from other dialects; we cannot, however, receive such opinions without caution, and even distrust; for we perceive from the writings of the Brahmans themselves, both unwilling and prejudiced witnesses in this respect,—that the so-called “common acceptation” of the term, does not accord with facts, and the particular opinions expressed by the learned. For, if the Pràkrit is “equal” (or similar) to the Sanskrit; and, if moreover the former is “an ocean to gems of lofty expressions;” with what justice or propriety can it be said that the Pràkrit is an “out-cast”—that it is “a vulgar dialect”—or that it is (apabhransa), a “jargon” destitute of grammar? A comparison of the Sanskrit with the Pali, clearly disproves the assertion of the Shàstras; and the result is not different when even the Pràkrit passages in the dramas are compared with the “speech of the gods.”

4. Sir William Jones describes the Pràkrit in the plays to be little more than the language of the Brahmans, melted down by a distinct articulation, to the softness of the Italian.¹⁵⁷

5. Professor H. H. Wilson from the testimony contained in the Dramas, written by Kàlidàsa, and in several ‘finished modern imitations’, treats the Pali under the designation of Pràkrit (under which appellation he includes ‘three varieties more or less refined’,) and adds: “the words are essentially the same in all, and all are essentially the same with Sanskrit, the difference affecting the pronunciation and spelling, rather than the radical structure, and tending generally to shorten the words, and subtitute a soft for a hard, and a slurred for an emphatic articulation.”¹⁵⁸

6. The Kàvya darsha defines the Sanskrit to be ‘the speech of the devas,’ and the Pràkrita to consist of several dialects, of which some are born of the Sanskrit; and some which stand in equal relationship with it; and others which are native or vernacular—peculiar to certain countries.

Obs.—I am aware that some writers¹⁵⁹ have put a different construction upon this authority: and have interpreted it to mean that the Prakrits ‘are composed of a three-fold element.’ It is however submitted with much deference, that the differences noticed by Dandi are differences of entire dialects, or, as the commentator says, entirely ’pure dialects and not differences of words in the same dialect. The context sets this at rest; for the writer enumerates ‘several orders’, or the ‘various’ dialects of the Pràkrit, and not the words of which they are composed.

But to return from this digression:

7. “In reference to the meaning of the word Pràkrit,” says Dr. Stevenson, “it may be observed that, among the Maràthi Brahmans, the term is often taken in its widest sense to signify the natural or vernacular language of any province in India. In a more restricted sense, it means any of the ancient dialects of the different provinces, and which as most of their books used till lately to be written in it, obtains, in the South of India, the appellation Grantha.¹⁶⁰

8. The Sinhalese also give the name of Grantha to ancient languages, and doubtless it is identical with the Indian Prakrita as may be seen from the enumeration of the qualifications of King Parakkramabahu III., 1267, A.D., in an extract from History given in my Sidath Sangara p. clxix.

9. “There are” says the Rig Veda, “four measured grades of language: with these intelligent Brahmans are acquainted. Three hidden in secret make no sign. The fourth grade of speech is uttered by men.” Dr. Muir in his Sanskrit Texts p. 163, et seq., after quoting the above, and its comment, which concludes thus:—“the Brahmans speak two sorts of language; both that of gods and that of men—” says, “three of the ancient schools which are quoted, assert the current language (vyavaharika vak) to be the fourth kind of speech alluded to in the Vedic text, as being spoken by men;” and, after expressing a conjecture, that this was the Sanskrit (?) he adds:—“It is true that in the Brahmana which the author of the Parisishta cites, a remark is made (connected with what precedes) that the Brahmans speak two languages, that of the gods, and that of men; and this might seem to prove that, as in later times, a distinction was drawn at the time when the Brahmana was composed, between Sanskrit, the language of the gods, and the Prakrit the language of men.”—p. 165.

Obs.—From the foregoing facts, deductions, and inferences it may be concluded that an “ancient,” “cultivated,” “natural or vernacular dialect,” “similar to the Sanskrit” has continued to retain the name of Pràkrit, along with several provincial dialects, which are usually treated under the same name.

IV Like all the above grounds of evidence, the testimony derivable from the primary signification of the word pràkrita is against the so-called “common acceptation” of that word; and invests the language, to which it was originally given as a name, with a character for originality and independence equal to, if not higher than, the Sanskrit.

I am glad to find that Professor Lassen agrees with me in believing that prakriti is the “source” from whence Pràkrita is derived; but I am equally sorry that I am constrained to differ from him as to the meaning which he assigns to that word. In his Institutiones Linguæ Prakriticæ pp 25, 26, the learned Professor says:—“Notatio vocabuli est a Prakriti, i. e. procreatrix, genetrix natura, unde Pràkrita, generatus, derivatus; derivatur enim hae linguae a Sanskritica aut directo aut una per aliam, quo sensu dicitur prakriti cujusvis linguae esse ea, a qua deducitur: Prakritih Sanskritam, tatra-bhavam tata àgatam và Pràkritam. Hemach: Cap. viii. § 1. ‘Prakriti sive origo in Sanskrita est; in ea lingua ortum vel ex ea profectum quod est, id pràkrita s. derivatum est.’ Opponuntur praeterea in aliis juncturis sibi Sanskrita et Prâkrita, ut de hominibus, qui quum justa cultioris vitae institutione imbuti sint, Sanskritici, sin minus, Pràkritici dicuntur, unde fit, ut pràkrita etiam notet hominem vulgarem. Pràkriticæ igitur linguæ etiam sunt vulgares, rusticae, provinciales.”

Now, it is clear that if Pràkrit comes from prakriti (procreatrix, genetrix natura) ‘nature’, the former cannot mean, as stated by the learned Professor, “derived,” but its very opposite—“the radical, or the root,” and, when applied to a language, “the orignal language from which another springs.” That “the several Pràkrit dialects are regarded” [by modern Brahmans¹⁶¹] “as derivatives of the Sanskrit either directly or mediately” signifies nothing; and cannot, I apprehend, affect the question any more than the dictum of the Buddhists, ‘that the Pali is the [mûla bàsà] radical language’ The only mode by which we may ascertain the correct meaning of Pràkrita is, I submit, as Professor Lassen himself has done, by tracing the word to its radical, primary, signification, in which sense, we may, considering the usages of the East, reasonably believe it was originally used.

By any other mode of determining the sense of this word we shall fail to obtain any information beyond the secondary sense, which, in the arbitrary usage of the modern Brahmans of the dramatic age, is assigned to this word—viz., that it is a name for “vulgar, rustic, and provincial forms of speech.” The modern acceptation of this word, is indeed inadmissible in an inquiry as to what language it was ab initios a name? Words are like men. They grow into various shapes. They gradually lose their original forms. They undergo in process of time, so many changes both in body and sense, that we cannot often determine their original meaning by their “common acceptation” at the present day. If, for instance, desirous of ascertaining how the old Romans preserved their writings, we referred to the modern acceptation of the term volumen, it would give us an idea merely ’of writing preserved on paper folded, or bound like our books (volume) with a number of distinct leaves above one another. But, if we traced the word to its original (prakriti) source, we should find that it meant ‘a folding,’ ‘a roll’—and, therefore, that the Romans ‘rolled up their writings’ as a ‘scroll,’ or like ‘the folds of a snake.’ Suppose again, for instance, we were engaged in an inquiry as to whether the Kandians and the maritime Sinhalese were originally of the same stock; and we were in the course of our investigations furnished with two epithets, ‘opposed to each other,’ and by which the maritime natives were distinguished from the Kandians; viz ‘the high Sinhalese’ and ‘the low Sinhalese.’ By the adoption of the meanings which either ignorance or prejudice has assigned to them among some people, we must conclude that the first meant ’the Sinhalese strictly so called¹⁶² and the second, ‘the low’ ‘the inferior,’ or (as the Brahmans designate the Prakrita) ‘the vulgar.’ But nothing could be more distant from the original signification of these terms as applied to the Sinhalese. The first meant ‘those who occupied the high-lands of the Kandian country,’ and the second ‘those who inhabited the flat levels of the maritime provinces.’

In considering, therefore, the question as to the comparative claims of the Pràkrita and the Sanskrita, we should take their primary, not their secondary, sense—that which they radically import, not that which has been assigned by usage. In a primary sense also, be it remarked, those words are “opposed to each other”; Sanskrita conveying ‘adornment’ and Pràkrita ‘the natural’: and this it would seem is also proof of the Pràkrita being the ‘original’ form of language, and therefore, in a secondary sense, (to adopt the contemptuous expressions of the Hindus) ‘the uncultivated savage,’ ‘the rustic,’ or ‘the vulgar’ idiom from which the language of the Brahmans has arisen to the development of the Sanskrit, ‘the highly polished’, or ‘the civilized.’¹⁶³

I have already shown that the word from whence Pràkrita is derived, was prakriti. It is thus defined by several writers:—

1. According to the Amarakosha (see pp. 32. 55) prakriti means in the Sanskrit ‘nature’ ‘the natural state.’

2. Bopp defines it also to be ‘natura r. kri facere præf Pra s. ti’—Glossarium Sanscritum p. 225.

3. Professor H. H. Wilson gives to it among other significations, the following; ’Nature, in philosophy, the passive or material cause of the world, as opposed to the active or spiritual;—the natural state, or condition of any thing; a radical form or predicament of being; cause, origin, a mother, the radical form of a word before the affixes forming cases, &c. are subjoined. E. pra implying ‘priority’ or ‘precedence,’ kri ‘to make’ aff: ti.’

4. The same meanings are assigned to its kindred expressions (pakati) in the Pali, (see Abhidanapadìpika p. 11.) and in the Sinhalese (see Clough and the Sidath Sangara).

Obs.—Prakriti is therefore, that which is natural, or the nature itself of a thing—that which is pre-eminent—that which is the natural and quiescent state of any thing—‘not made.’ Hence it is clear that the correct and primary sense of the word Pràkrita,—indeed that which was originally assigned to it, despite the so-called ‘common acceptation,’—was ‘original,’ ‘root,’ ‘natural.’ By the Pràkrit was therefore, at first meant the original Indian language, as distinguishable from the Apabransa ‘the ungrammatical,’ and the Sanskrit, signifying [from sam ‘altogether’ or ‘together,’ and krita ‘done’= ‘altogether’ or ‘completely made, done, or formed’] that ‘which has been composed or formed by art adorned, embellished, purified, highly cultivated or polished, and regularly inflected as a language.’

V. What then was this original Pràkrit language which was “similar to the Sanskrit?” It would be in vain to look for it in any of the living languages of the world; for it is an established fact in philology that all languages change¹⁶⁴ in course of time, even without the introduction of foreign elements. It can only be discovered, I presume, by the examination of the earliest writings of those languages which have ceased to be spoken.

Let us first examine the dramatic dialects with reference to the precepts of the Grammarians and Commentators.

1. It has been found by Professor Lassen and others who have examined these writings, (says Mr. Stenzler)¹⁶⁵ that, making great allowances for the errors of ignorant copyists, and the unauthorized alterations of learned transcribers, there is still a difficulty in reconciling the doctrines of the grammarians with the language of the dramas.

2. ‘Much discrepancy’ says Mr. Cowell¹⁶⁶ exists between the Pràkrit of the grammarians, and that which we find in the plays.’

3. Professor Wilson, who was intimately acquainted with the language of the dramas, from his having translated them, says that the term Pràkrit, applied by grammarians ‘to a variety of forms, agrees only in name with the spoken dialects.’ ‘The Màgadhi by which name may be considered that dialect which is more ordinarily understood by Prákrit, is very different from the vernacular language of Magadh or Behar. The Saurasenì is by no means the same with the dialect of Mathura, and Vrindavan, and the Maharashtri would be of little avail in communicating with the Maharattas, or people of Maharastra. The other species enumerated, are equally inapplicable for identification with the dialects to which they might be supposed to refer.’¹⁶⁷

4. ‘Highly finished specimens are to be found in plays which are modern productions. The Vedagdha Madhura, for instance, consists more than half of high Pràkrit, and it was written less than three centuries ago.’—ib.

Obs.—This discrepancy may be accounted for by the following conjecture. The dramatic writers, whose age is fixed by Professor Lassen at 400—100 B. C, and, who wrote the great bulk of their plays in Sanskrit, were eminent Sanskrit Scholars. They indeed studied the Pràkrit languages; and, even if they did not, from the affinity of the Pràkrits to the Sanskrit, they found no difficulty for a display of their learning by improving upon several then existing Indian dialects. In doing so, according to the original aphorism of Bhárata, it was necessary “to employ choice and harmonious terms, and an elevated and polished style, embellished with the ornaments of Rhetoric and rythm.” It may be then believed that they imitated the best writers with whom they were acquainted; that some authors gave preference to some writers as models for imitation, and others to others; and that some selected old, and others modern authors for their guide;—whilst, perhaps, the same writer in different plays adopted the language of both ancient and modern books. Hence ‘the exceeding richness’ of the Pràkrita in some of the plays, and its comparative inferiority in others,—and, perhaps, the same differences exhibited in the works of one and the same person. The Grammarians, on the other hand, especially Vararuchi, whom we regard, according to Hindu traditions, as one of the ‘nine gems’ of Vikramaditya’s Court, and therefore a writer of comparatively after times;¹⁶⁸ were led by different lights. The Pràkrita passages scattered throughout the Hindu dramas, did not afford them sufficient means to obtain an acquaintance with the whole extent of those languages. They found, probably, that one dramatic writer made one class of persons speak a particular dialect, when another put that into the mouth of other classes—thus rendering it difficult even to assign a name to the dialects used. It is also probable that the Grammarians sought in India for, and having found but few traces of, the principal Pràkrit, which tradition represented as a language which “abounded in gems of lofty expressions,” and which, as stated in the Pràkìritadìpikà, “is the most excellent form of speech” took for their guide a then existing dialect of Maharastra.

This was probably a modification of the Pali. For, from the account of Fa Hiam, it would seem that Buddhism had flourished in Mutra from a very early period; and, consequently, its language, the Pali had been early introduced into that country. Yet, from its having been amalgamated with other Pràkrit dialects, and not become fixed as a dead language, as was the case with Pali in Ceylon, the Maharashtrì presents a great many corruptions, of which the Pàli is comparatively free.

The following comparative Table of the first thirty words in Vararuchi’s Pràkrit Prakàsa, will render this manifiest.

samriddhi	samiddhi	samiddhì	‘prosperity’
prakata	pàkata	paadam	‘manifest’
abhijàti	abhijàti	ahijàì	‘family’
manasvinì	manassinì	manansinì	‘wise woman’
pratipad	pàtipada	pàdivaà	’1st day l½ mon[?]
sadrikksha	sarikkha	sarichchham	‘like’
prasupta	pasutta	pasuttam	‘asleep’
prasiddhi	pasiddhi	pàsiddhi	‘fame’
asva	assa	asso-àso	‘horse’
ì.hat[?]	ìsan	isi	‘little’
pakva	pakka	pikkam	‘cooked’
svapna	soppa, supina	sivina	‘sleep’, ‘dream’
vetasa	vètasa	vediso	‘ratan’
vyajana	vìjanì	viano	‘fan’
mridanga	mutinga	muingo	‘drum’
angàra	angàra	ingàlo	‘charcoal’
aranya	arañña	raññam	‘forest’
sayyà	seyyà	sejjà	‘bed’
saundarya	sundara	sunderam	‘beauty’
trayodasa	telasa	teraho	‘thirteenth’
àscharya	achchhariya, achchhera	achchheram	‘wonderful’
paryanta	pariyanta	perantam	‘limit’
valli	valli	vellì	‘creeper’
badara	badara	voram	‘jujube’
lavana	lona	lonam	‘salt’
navamallikà	navamallikà	nò mallià	‘jasmine’
mayùra	mayùra, mora	mora, mauro	‘peacock’
mayúkha	mayùkha	moho, maùho	‘ray’
chaturthì	chatutthì	chotthì	‘fourth’
chaturdasì	chatuddasì, chuddasì, choddasì	chuddahì, chaddahì	‘fourteenth’

Having once identified the principal Pràkrit, with that which approached nearest to it, viz. a dialect of the Maharattas, different from the language which, like the Pali, was once greatly admired, and different also from the language now prevailing in that country, the Grammarians had no alternative but to seek for the Màgadhì in the speech of men living in Magadha. This too, was different from that which is in use there, and also greatly at variance with the original Màgadhì (the Pali) which was only preserved in Ceylon. Hence the discrepancy between the Grammarians and the dramatists—and hence also the difference between the Ceylon and Indian writers in respect of Màgadhì—the Sinhalese treating it as the Pàli (or the language of the text books of Buddhism) which found an early retreat in Ceylon, and the Brahmans identifying it with a modification of the same dialect greatly deflected from its original construction.

It would also seem from the foregoing observations that the dialect, which originally received the name of the Pràkrit, has in course of time, undergone a vast change in India, and that the patois of the dramas, does not furnish us with sufficient materials for its identification with any known dialect. Probably the Pràkrit of the Plays is a modification of the Pàli; and it is very probable that the principal Pràkrit of Vararuchi is a still greater modification of the Pràkrit of the plays. But of this we may be sure—that both have many traces of corruption, and that both have less claims for originality than the Pali. [See my comparative Tables infra.]

Another language which may demand attention here, is the so-called Gàthà dialect of the Nepal Buddhists. A solution of the difficulty as to its origin, will be given hereafter. In the meantime it is sufficient to notice what we gather as to its style, from the writings of a learned Hindu gentleman¹⁶⁹ and of M. Burnouf, viz. ’that the Buddhist literature of Nepal, from which the Sacred Scriptures of Tibet, Tartary, and China have been compiled, is in an ugly Sanskrit dialect, destitute of the niceties of the Sanskrit grammatical forms of declension and conjugation, &c.; that the authors have sacrificed grammar to the exigencies of metre; that it is in a mixed style of prose and Gàthàs; that it bears a strong resemblance to the Tantras of the 4th to the 7th Century of the Christian era—and that it appears to be the production of men to whom the task of compilation was assigned without sufficient materials at their disposal. In view of these peculiarities, Mons. Burnouf has pronounced the Nepal sacred scriptures to be a ‘barbarous Sanskrit, in which the forms of all ages, Sanskrita, Pali, and Pràkrita appear to be confounded.’¹⁷⁰

These peculiarities establish its inferiority to the Pali; and the dialect of the Pillar Inscriptions (which were recorded in the third century of the christian era,) being decidedly posterior to the language of Gotama, I shall, in search of this original Pràkrit, next betake myself to the Ceylon Pali, a dead language of antiquity, which came from Magadha, and has been preserved in Ceylon from the time almost of the Buddhistical era. Its identity with the Màgadhì of remote antiquity, is not only established by the history of Buddhism, and by the promiscuous use of the terms Pali and Magadh in Ceylon; but also from the extract above given, of the enumeration of the shad bhàshà, ante p. lxxxiii. Its great antiquity, and high state of cultivation in Ceylon, are made to appear from a variety of circumstances.

It is indeed a remarkable fact that all oriental scholars, who have made Pali the subject of study according to the different, though imperfect, opportunities they have had out of this island, have not failed to perceive something peculiar in the Pali, which distinguishes it from every variety of Sanskrit-idioms.

1. Professor Benfey in his Ersch and Gruber’s German Encyclopædia p 194, characterises the Pali as ‘the sacred language of the Buddhist writings found in Ceylon and Transgangetic India, …………… which is shown both by internal and external indications, to have been the vernacular dialect of central India, and which was diffused along with the Buddhist religion in the countries above-named, where it soon acquired the same sacredness in the eyes of the Buddhists which Sanskrit possessed, and still possesses, for the Brahmans. This language,’ he continues (‘though distinct proof cannot yet be adduced of the assertion,) is one of the very oldest of the Indian vernaculars, and was already in popular use at the period of the rise of Buddhism.’

2. Dr. Muir after citing the last authority, states. “But it matters little in what particular province we suppose the Pali to have originated, whether in Magadha, or in some country further to the westward: as the fact remains in any case indubitable, that it represents one of the oldest Prakatic dialects of northern India.”—Sanskrit Texts p. 79.

Obs.—I may here remark in passing, that if the Pali represents, as it undoubtedly does, the oldest Pràkrit; and, moreover, if the Maharashtrì dialect is, as stated by the writer of Prâkrita-kalpataru, “the root of the other [Pràkrits];” i. e. those of which the Indians had any knowledge, or which have not been lost in India;—the inference is inevitable that the Pali, which had found an early retreat in Ceylon, is the parent of all Pràkrits, including the Maharashtrì. But to proceed: M. M. Burnouf and Lassen, who had but few Pali works within their reach, have not however, failed to perceive the nearest relationship of the Pali to the Sanskrit. Though I reluctantly, but respectfully, differ from them in the belief that the former is immediately derived from the latter; yet the facts which those learned writers record, are not without importance or value.

3. ‘When the Pali, [say those learned writers in their Essai sur le Pali, p. 138] as a derivative from Sanskrit, is compared with other dialects having the same origin, it is found to approach far more closely than any of those others to that common source. It stands, so to speak, on the first step of the ladder of departure from Sanskrit, and is the first of the series of dialects which break up that rich and fertile language.’

4. Dr. Muir, after subjecting the Pali to a comparison with the Sanskrit and Prakrit, concludes by saying, ‘from this comparison it will result that the Pali stands nearer to the Sanskrit, and represents a more ancient phase of the vernacular speech of Northern India than is exhibited in the Pràkrit’

And he adds ‘It has been demonstrated at length that it (the Pàli), in its turn, is more ancient in its grammatical forms than the Pràkrits are, and departs less widely than they do from the Sanskrit.’—p. 137.

I may also remark that entire sentences may be given of, and that whole passages may be composed in, the Pali wherein every word, every grammatical form, and every philological development most closely accords with even the Sanskritized idiom of the Brahmans. The following extracts, for example, from the Bhatti Kàvya, which abounds with specimens of the kind, authorize the above statement.

Chàru samìrana ramane
Harina-kalanka kiranàvalì savilàsá
Abaddha Ràma mohà
Velà mùle vibhàvarì parihìnà—p. 77.

‘The evening, radiant with masses of moon-beams, and which filled (bound) Ràmà with affliction, was spent on the confines of the shore which was rendered delightful by reason of the (sweet) gentle breeze.’

Gantum Lankà tìran
Baddha mahà salila sancharena sahelan
Taruharinà giri jàlan
Vahantu giri bhàra sansahà gurudehan—p. 88.

‘Let the monkeys, able to sustain the weight of mountains, convey the heavy-bodied rocky chain, so that we may in one line, get to the shores of Lankà, on a bridge constructed on the vast profound.’

To the above facts, all which are important in the consideration of the general question, as to the relationship of the Pali to Sanskrit, I shall add the testimony of eminent Oriental scholars, who notice the difference of the Pali from every dialect which is supposed to have arisen from the ancient Magadhì;—its difference

1. From the dialect of the rock inscriptions;¹⁷¹

“Now it is curious enough that some of the distinguishing traits of the pillar dialect are just such as are pointed out by the Grammarians of a later day as constituting the differences between Màgadhì and Pali.” “The [same] language,” Mr. Prinsep adds in another paper¹⁷² “differs essentially from every written idiom: it is as it were intermediate between the Sanskrit and Pali.”

2. From the Buddhist writings of Nipal;

‘They are’ says M. Burnouf, “intermediate between the regular Sanskrit, and Pali—a dialect entirely derived, and manifestly posterior to the Sanskrit.¹⁷³

3. From the Pràkrit dialects;

‘The Pràkrits do not represent the derivative form of speech which stands nearest to the Sanskrit; and we are in a position to point out a dialect which approaches yet more closely to the latter than the Pràkrits do. I mean the Pali, or sacred language of the Buddhists; a language which is extinct in India, but in which numerous canonical books of the Buddha religion, still extant in Burmah and Ceylon are written.’¹⁷⁴

4. From the Màgadhì dialect in which the works of the Jains are written;

‘On comparing the Mahawanso (says Dr. Stevenson) one of the sacred books of the Ceylonese with the Jain writings, I find considerable dissimilarity between the two dialects; the Pali approaching much nearer to the standard of the general Pràkrit, and having few, if any, of the peculiarities of the Màgadhì dialect, while the Jain works exhibit them by no means in a slight degree.’¹⁷⁵

5. From the present language of Magadha or Behar;

‘The Pali’ says a writer in the Asiatic Journal, xviii. p. 764, ‘being generally known by the name Màgadhì, was compared with the modern dialect of Magadh or Behar, and the comparison shews that they essentially differ. In those respects in which it differs from the Pali, it approaches the Prakrit or the sacred language of the Jains.’

6. And from the Màgadhì of the Indian grammarians:—

The following comparative view of the Pali, with the rules given by Vararuchi under the head of Màgadhì, will render their difference evident.

(i.) The first rule of Vararuchi is SHA SOH SAH. In the Pali there is no S. It has only the dental sibilant S. The inapplicability of the rule which states that in the peculiar dialect of Pràkrit termed Màgadhì, ‘S is substituted for sh or S,’ is therefore, self-evident.

(ii.) JO VAH. The occasional substitution of y for j is no more a peculiarity of the Pali than of the Sanskrit or Sinhalese; e. g. yàminì or jàminì in Sanskrit, yama or jáma Sinhalese ‘night.’ The usual Pali nija is written in the Suttans with a y, as niyan puttan ‘own son.’ Instances like these are exceptions, not the rule, in those two languages. But neither in the instance given by Vararuchi, nor in the great majority of Sanskrit words with a j, is it changed into a y in the Pali. The reverse of what is given by Vararuchi may be regarded as the rule. Thus, jàyate ‘he is born,’ is the same in the Pali, and is not changed into yàyade. So likewise, raja is rája, and not ràya ‘king’; gaja is gaja but not gaya ‘elephant;’ vajra is vajira, but not vayara ‘diamond.’ It is true that in words like paryùshana, the Pàli form is payyùsana, and not pajjausana, as in the common Prakrit. This peculiarity in the Pàli, however, does not indicate a change from j to y. It is simply the reduplication of y.

(iii.) The next rule, CHVARGASYASPRISHTATA TATHOCHCHARANAH: seems to refer to a nicety in the pronunciation of the palatal letters, which we do not perceive in the Pali; and therefore proceed to the next;

(iv.) HRIDAYASYA HADAKKAH. This is equally inapplicable to the Pali. Hridaya ‘heart’ never becomes hadakka, but hadaya, in Pali. So likewise hrisva ‘short’ is not hadasva, but rassa; and hri ‘shame’ is not hida but hiri.

(v.) RYARJAYOR YYAH. The substitution here spoken, of yy for ry and rj may be regarded as the exception (and that of very rare occurrence) rather than the rule in the Pali. Thus kàryan ‘to be done,’ is not kayye but kàriyan; and durjana ‘wicked’ is not duyyana, but dujjana, in the Pali. So likewise vìrya ‘exertion’ becomes viriya; bhàrya, bhariya ‘wife’; aisvarya, issariya ‘prosperity’; and also garjana becomes gajjana ‘noise’.

(vi.) KSHASYA SKAH. This is again different in the Pali. Thus ràkshasah ‘demon’ does not become laskase but rakkhaso; nor dakshah ‘clever’ daske, but dakkho. So likewise, vriksha ‘tree’ becomes rukkha in the Pali; kshamà kamà ‘forgiveness’; dakshina, dakkhina ‘south’; kshùra khùra ‘razor’; kshetra, khetta ‘field’. This peculiarity will be found explained in another part of our observations, vide infra, § x.

(vii.) ASMADAS SAU HAKE HAGE AHAKE. The Sanskrit ahan bhanàmi ‘I speak’, is the same in the Pali; and does not become as stated here hake, hage, or ahake banami.

(viii.) ATA IDETAU LUKCHA. The Sanskrit etad (root) eshah (nom:) is said to be changed in the Màgadhì into esa, and su being added to it=esà-su; and the latter affix being elided, the a in esà is changed into i, or e. This is not a peculiarity of the Pali, in which eta (root eso—nom:) becomes esa ràja (which is esha ràjà in Sans:) ‘this king’, and not as in the so-called Màgadhì Esi laa; (!!) and similarly esha purushah, Sanskrit, becomes eso puriso in the Pali, but not, as in the Pràkrit Màgadhì, esà pulise ‘this man.’

(ix.) KTANTAD USCHA: which is rendered by Dr. Cowell into English as follows: U is substituted when the affix su follows a word ending with the affix kta; and also (as we infer from the cha of the Sût:) we may optionally use the i or e of the preceding Sút, or even elide the affix; as hasidu or hasidi, haside, hasida, for hasitah ‘smiling.’ It is only sufficient to state here that the Pali knows no such thing, and that the Sanskrit hasitah is in the former simply hasito.

(x.) NASO HO VA DIRGHATWAMCHA: That is to say ’ha is optionally substituted for nas, the affix of the genitive singular, and at the same time the preceding vowel is lengthened, as pulisàha or pulisassa dhane for purushasya dhanam ‘the man’s-wealth.’ The Pali form of this is purisassa dhanam wherein the Sanskrit inflexion sya assumes ssa, for the simple reason that the Pali dislikes the union of two consonants of different classes. It is further remarkable here that dhane of the Pràkrit-Màgadhì becomes dhanam (neuter) both in the Sanskrit and Pali, in which moreover the cerebral n is not used.

(xi.) ADIRGHAS SAMBUDDHAU. It is to be inferred from the examples given under this rule that in the Prakrit Màgadhì dialect the vocative inflexion a both in the singular and plural number is long. In the Pali, however, the termination of the vocative singular may be either long or short. (See Clough’s Bàlàvatàra p. 19,); as purisa àgachchha or purisá àgachchha—‘O! man come.’

(xii.) CHITTHASYA CHISHTHAH. In shewing the difference of the Pali from the Pràkrit-Màgadhì, it is here sufficient simply to exhibit the Pali forms of the given examples.

Purushah tishthati¹⁷⁶—Sanskrit.
Puriso titthati—Pali.
Pulise chishthadi—Mag: Prakrit.

(xiii.) KRINMRINGAMAM KTASYA DAH. Here again we cannot exhibit the difference of the Pàli from the Màgadhì Pràkrit better, than by placing the given examples in juxtaposition with their Pali forms.

Kritah ‘done’ mritah ‘dead’ gatah ‘gone.’ Sans.
Kato — mato — gato. Pali
Kade — made — gade. Mag-Pràk.

(xiv.) KTVODANIH. The following comparative view of the examples given under this rule, shews the relationship of the Pali to the Sanskrit to be far nearer than that of the Prakrit-Màgadhì.

Sodvàgatah¹⁷⁷ —kritvàgatah¹⁷⁸ San.
Sahitvá gato —katvágato Pali
Sahidani gade —karidani aade. Prak M.

(xv.) SRIGALASYA SIALASIALESIALAKAH. The difference between the Sanskrit Srigàla and the Pali sigala is, simply that occasioned by the absence of the Sanskrit r in the latter language. But Vararuchi gives the three following forms into which that word is changed in Màgadhì-Pràkrit; viz. siálá, siale, siàlake.

Obs.—From the above, and many a fact in the history of Buddhism, it may be inferred that there are few traces of the genuine Pràkrit in the existing Indian dialects; that the principal Prakrit, which distantly approaches to the Pali, is a corruption of the latter; that the patois of the dramas is an exhibition of it; and that the Grammarians, who subsequently framed rules for the formation of this corrupt idiom, had not the Pali, which had been banished the Magadha country along with the Buddhist religion. It may also be thence inferred that in very early times the Pali became mixed up with pure Sanskrit, as in the Nepal version; that it next assumed the form of the pillar dialect, and that at last it was reduced to the Màgadhì of the Jains, which distantly resembles the dialect of the grammarians. Buddhism, in a very early period of its history, doubtless, brought the Pali language to Ceylon; where, having soon become a dead language, its use was confined to the priesthood; and from the homage the Ceylonese paid to it as the language of the founder of their religion, it remained in the Island unaffected by those changes to which, as a spoken language, it was subjected in its migrations in India. Thus, its philological peculiarities, which will be noticed hereafter—its great age in this Island—its still higher antiquity in Asia—and the absence of any other Indian dialect which bears traces of so close an affinity to the Sanskrít as the Pali, when taken into consideration with the facts, that ‘the Brahmans and Aryas’ had for their vyavahàrika vàk, a dialect similar to the Sanskrit; and that that dialect is declared by Buddhagosa, himself a Brahman, to be the Pàli [Pràkrit]; we may indeed discover a few at least of the grounds upon which the tradition of the Buddhists is based, viz., that ‘the Pàli was the mûla bàsà.’

VI. This leads us then to the consideration of the theory with which we have set out—that the Pali is “a sister-dialect of the Sanskrit, being probably derived from one and the same source.”

In considering this subject we notice that the Brahmans regard the Sanskrit to be of divine origin, and as a direct revelation from their Creator; and that the Buddhists claim for the Pali an antiquity so remote that they affirm it to be a language ‘the root of all dialects, which was spoken by Brahmàs, by men before the present kappa, by those who had neither heard nor uttered human accent, and also by supreme Buddhas’—

Sà Màgadhì mùla bhàsà
Nàrà yà yàdi kappikà;
Brahmànochassutàlàpà
Sambuddhà chàpi bàsare.

The above is found quoted by Mr. Turnour from the Payogasiddhi; and the following, to the same effect,¹⁷⁹ occurs in the

VIBHANGA ATTHAKATHA.

Tissadatta thero kira Bodhi mande suvanna salàkan gahetvà ‘atthàrasasu bhàsàsu katara bhàsàya katemi—iti’ pavàresi. Tan pana tena atthato uggahetvà pavàritan; na-patisambhidàya thitena; sohi mahàpaññatàya tan tan bhàsan kathàpetvà ugganhi: Tato uggahethatvà evan pavàresi. Bhàsan nàma sattà ugganhantìti vatvàcha panettha idan kathitan. Màtàpitarohi dahara kàle kumàrake manchevà pìthevà nipajàjpetva tan tan kathaya mànà tàni tàni kichchàni karonti; dàrakà tesan tan tan bhàsan vavatthàpentu ’iminà idan vuttan iminà idan vuttan’ti gachchante kàle sabbampi bhàsan jànanti. Màtà Damilì pità Andhako tesan jàto dàrako sache màtu-kathan pathaman sunàti, Damila bhàsan bhàsissati, sache pitu kathan pathaman sunàti Andhaka bhàsan bhàsissati. Ubhinnampi pana kathan asunanto Màgadhikan bhàsissati. Yopi agàmake mahà raññe kathento nàma natthi sopi attano dhammatàya vachanan samuttha pento Màgadha bhàsa meva bhàsissati. Niraye tirachchhàna Yonian pettivisaye manussa loke deva loke-ti-sabbattha Màgadhika bhàsà eva ussannà; tattha sesa Ottà Kiràthà Andhaka Yonaka Damila bhàsàdikà attharasa bhàsà parivattanti. Aya’meva ekà yathàbhuchcha-Brahma-vohàra-Ariya-vohàra-sankhàtà Màgadhika bhàsà eva na-parivattati. Sammà Sambuddhopi tepitakan buddha vachanan tantin aropento Màgadha bhàsàya eva àropesi; Kasmà evañhi atthan àharitun sukhan hoti. Màgadha bhàsàyahi tanti arûlhassa buddha vachanassa patisambhidappattànan sotapathagamane-neva yanche sotena sanghattita matte yeva naya satena naya sahassena attho upatthàti; aññàya pana bhàsàya tanti àrulhakan pothetvà uggahetabban hoti.

’Tissadatta thèra took up the gold broomstick in the Bô compound, and requested to know in which of the eighteen bhàsas he should speak? He so (spake) from (a knowledge of those languages) acquired not through inspiration,¹⁸⁰ but, by actual study; for being a very wise personage he knew those several dialects by learning: wherefore, being one of (such) acquirements he so inquired. This is said here (to illustrate) that men acquire a bhàsà (by study.)

’Parents place their children, when young, either on a cot or a chair, and speak different things and perfom different actions. Their words are thus distinctly fixed by the children (on their minds, thinking) ‘that such was said by him and such by the other’; and in process of time they learn the entire language. If a child, born of a Damila¹⁸¹ mother, and an Andhaka¹⁸² father, should first hear his mother speak, he would speak the Damila language; but if he should first hear his father speak, he would speak the Andhaka language. If however he should not hear them both, he would speak the Màgadhì. If, again, a person in an uninhabited forest, in which no speech (is heard), should intuitively attempt to articulate words, he would speak the very Màgadhì. It predominates in all regions (such as) Hell; the animal kingdom; the Petta sphere; the human world; and the world of the devas. The rest of the eighteen languages—Ottà, Kiràthà, Andhaka, Yonaka, Damila, &c., undergo changes;—but the Màgadhì does not, which alone is unchangeable, and is said to be the speech of Brahmans and Ariyas. Even Buddha, who rendered his tepitaka words into texts, did so by means of the very Màgadhì; and why? Because by doing so it (was) easy to acquire their (true) significations. Moreover, the sense of the words of Buddha which are rendered into doctrines by means of the Màgadhì language, is conceived in hundreds and thousands of ways by those who have attained the pati-sambhidà so soon as they reach the ear, or the instant the ear comes in contact with them; but discourses rendered into other languages are acquired with much difficulty.’

Now it is a fact that ‘all rude nations are distinguished by a boastful and turgid vanity.’ They invent fables to exalt their nationality, and leave records behind them to abuse the credulity of after ages.’ They cannot speak of their race, or of their sacred languages without assigning to them an origin the remotest in the world. In ‘a spirit of adulation and hyperbole’ they exalt them as high as the object of their adoration and worship. This is peculiarly the case with Eastern nations.

Although such extravagantly high pretensions, are by themselves of no value; yet when some of these traditions are partially supported by the concurrence of other testimony, we may by a judicious exercise of our judgments in separating fact from fable, and reality from fiction, receive them, I apprehend, to the extent to which they are confirmed. Let us examine this confirmatory proof.

The term Pràkrita, as we have already seen, means ‘root’ or ‘original;’ and the Pali is the earliest exhibition of the Pràkrit. In this point of view, therefore, the Pali may claim greater originality, if not antiquity, than the Sanskrit, which is confessedly a dialect ‘made’ or ‘done.’ In other words, if the Pali, may be regarded as the prakriti, or an exhibition of the aboriginal tongue, there is nothing in the signification of the term Sanskrita to entitle the language for which it is a name, to be considered the source from whence the former is derived.

The facts, too, which we glean from history or find from natural causes, accord wonderfully with the import of the terms which we have above given, and with the belief that Pali is a Vyavahàrika idiom of the Sanskrit. For, whilst both the languages are fundamentally the same, the Pàli is simpler in its formation, and is more adapted to the vocal organs of men in a rude state of society, who, like children, avoid the sharpness of a union of heterogeneous consonants by the elision of the first, and the reduplication of the second.

These can scarcely be pronounced to be peculiar characteristics of a derivative tongue. For, we know that many nations both of the North and South-Indian class, in their attempts to beautify language, draw largely from the Sanskrit. The Tamils and the Hindus use a dialect full of Sanskrit words; and the modern Sinhalese with a view to beautify language, do not assimilate sounds, and shorten expressions, but Sanskritize our ancient simple language.¹⁸³ This was probably the case with the Sanskrit itself, which has no claims to originality.

The simplicity of the grammatical system in the Pali, as is indicated by the non-use of the dual number, the absence of certain elaborations of simple tenses,¹⁸⁴ the small number of verbal classes,¹⁸⁵ &c. ‘look like the spontaneous substitution of practical to theoretic perfection in actual speech’.¹⁸⁶ For, it is a fact consistent with natural events, that the less finished and elaborate system is usually anterior to that which is more so.¹⁸⁷ The presumption therefore is—not that the Sanskrit had an origin anterior to the Pràkrita, by which I here mean the Pali—not that it received the name of Sanskrit when in a rude state—not that the Pali-Pràkrit is a derivative of the Sanskrit;—but that the latter is only a more finished exhibition of the Pali, or of some unknown idiom from whence both have sprung. I may, in justice to the theory of the Buddhists, add, that many of the laws by which certain derived languages may be distinguished, do not govern the Pàli, e. g. The Pali has not rejected case-terminations for particles; and has not adopted auxiliary verbs in conjugation. It is indeed the Sanskrit which may be charged with using auxiliary forms altogether unknown to the Pali.¹⁸⁸ For, “the Sankrit verbs of the tenth class, and all derivative verbs, periphrastically express the reduplicated præterite by one of the auxiliary verbs,—kri ‘to make’, as and bhû ‘to be.’¹⁸⁹”

The fact, as remarked by M.M. Burnouf and Lassen, that ’no grammatical form is to be found in the Pali, of which the origin [why not say, some traces] may not be discovered in the Sanskrit,“¹⁹⁰ far from proving the Pali to be a daughter, establishes to my mind that she is, like the Zend, a sister of the Sanskrit. It would also seem that no inferences can be drawn by comparisons between the Pali and the Sanskrit¹⁹¹ on the one hand, and, for instance, the Italian and the Latin on the other; because the Sanskrit itself has undergone a great change, and the various influences which contribute to the corruption of languages are not the same both in Europe and Asia. And the differences must be great indeed between the languages (e. g. Sanskrit and Pali) of tribes who had continued together for several thousand years in the same country, subject to the same influences of literature, religion, and clime,¹⁹² and who upon separation have changed their religious faith, and have ceased to speak their respective languages; and of those (e. g. Latin and Italian) of other tribes who have been ‘separated for as many thousands of years, living in regions far apart from each other, under different physical conditions, and whose vernacular dialects are subject to the modifying action of different social, political, and religious institutions.’¹⁹³

Yet it is a singular fact that, in some particulars in which the grammatical forms of the Pali differ from the Sanskrit, they agree with the structure of the Indo-European languages, and of the Pràkrit dialects. e. g. There is ‘a concurrence of the Pràkrit with the old High German and the Latin of the 2d conj. in this point—that it in like manner has contracted the affix aya to ê.’¹⁹⁴ Compare Sanskrit mànayâmi ‘I honor’, Pali mànêmi Pràkrit mànèmi, Old-High German var-manêm ‘I despise’, Latin moneo:“¹⁹⁵

Sanskrit.	Pali.	Prakrit.	Old High Ger.	Latin.
mànayàmi	mànemi	mànemi	var-manêm	moneo
mànayasi	mànesi	mànesi	manês	monês
mànayati	màneti	mànedi	manêt	monet
mànayàmas	mànema	mànemha	manêmes	monêmus
mànayatha	mànetha	mànedha	manêt	monêtis
manayanti	mànenti	mànenti	manênt	monent

In regard to these weak verbs, (adds Professor Bopp) which have suppressed the first vowel of the Sanskrit aya, and give therefore ya as affix, we will here further recall attention to the forms iga (ige), which occasionally occur in Old High German and Anglo Saxon, whose connection with aya is to be traced thus, that the semi-vowel y has become hardened to g (comp. § 19.) and the preceding a weakened to i.’ Hence the Pràkrit padhijjai ‘is read’, gamijjai ‘is gone.’

If the Pali was immediately derived from the Sanskrit, it is sure to have those forms only which the Sanskrit adopts; and cannot possibly know any other forms which her Sisters had taken away, at their separation from the Indo-Ariyan speech. The existence however in the Pali of both forms known to the Sanskrit, and forms which her European-Sisters adopt, as in the instances cited under the causal form of the verb,¹⁹⁶—establishes the belief of Dr. Weber, to which I shall hereafter call attention;—‘the contemporaneous development of both the Sanskrit and the Pràkrit dialects from one common source, viz., the Indo-Arian speech.’

I may also here observe that among the more ancient Brahmans, none, notwithstanding their partiality to the Sanskrit, have expressly stated the locality¹⁹⁷ or the source of the dialect called by way of eminence, the Pràkrit. Comparatively later writers do indeed point at Mahàrashtrì as that principal Pràkrit;¹⁹⁸ but neither Vararuchi nor Bahàmaha has so stated it. The former who treats of four dialects, says that Paisàchi and Màgadhì are derived from the Saurasenì; and the Saurasenì had its source in Sanskrit; but he is silent as to the origin of the Pràkrit, to the elucidation of which he devotes the largest portion of his work.

It is however stated by those who maintain that the Pali, or the principal Pràkrit is a derivative, that the Sûtra (18) at the end of Cap. ix, in which Vararuchi refers the student to “the Sanskrit” for “the rest” or the remaining grammatical forms,¹⁹⁹ implies that the principal Pràkrit had its origin in the Sanskrit. This, however, is not the only inference. The fact from which that inference is drawn, is also consistent with the belief of the Buddhists—that the Pàli or the principal Pràkrit is a sister of the Sanskrit. For, if they were, like the Sanskrit and the vyava-hârika vàk of the Brahmans, two dialects which had a simultaneous origin, and merely differ from each other in some respects like the Attic and the Ionic; there was no necessity whatever to treat of, twice, the grammatical forms which were identical in both. Having a full and complete grammar of one (the Sanskrit), it was surely sufficient to shew the differences only by which the Pràkrit grammatical forms were distinguishable from those of the Sanskrit. Hence the simple reference—Seshah Sanskritàt—to ‘the Sanskrit for the rest.’

I have already alluded to the fact that the Sanskrit had been in a state of transition until it became fixed as the classical language of the Brahmans. ‘It shews clearly to be’ says a late writer, ‘the adaptation of some vernacular dialect to the state in which we find it, in order to form a characteristic language.’ ‘Its style exhibits all the traces of transition from the first efforts of expression to the highest refinements of grace and inflection, and its literature all the gradations from barbarism to sublimity, and from sublimity to refinement.’ Some of the older compositions, such as a portion of the Vedas, prove this; for thier style, unlike the poetry of Kàlidàsa is rustic and irregular, and ’they are written in an ancient form of the Sanskrit, so different from that now in use, that none but the more learned of the Brahmins themselves can understand them.²⁰⁰

That the Pali, if it were not the vyavahàrika vàk of the Brahmans, had a contemporaneous existence with an old form of language, which has been cultivated to the development of the Sanskrit, appears not only from a comparison of the Pali with the oldest available Sanskrit, viz., the Vedas, but also from a careful examination of the oldest Pali alphabet.

To treat of the latter first: The earliest records are by Pali-speaking²⁰¹ Buddhists in an old type of the Nàgarì alphabet; and, judging by its internal evidence, ‘it bears every impress of indigenous organization and local maturation.’²⁰² Although the age of this character is identical with that of Asoka (235 B. C.); yet that the same character had been in use for a considerable time before that date, may be easily believed. Indeed the following description given by King Pukkusàti of the characters in which the letter of his royal friend Bimbisàra was written, clearly shews that the same was the Màgadhì alphabet used in the time of Gotama Buddha.²⁰³ So tan pasàritvà ’manàpàni vata akkharàni samasìsàni samapantìni chaturassànì’ti àdito patthàya vàchetun àrabhi. That is, ‘when he had unfolded [the gold plate 6 feet × ½, on which the epistle was written] he (observed) that the letters²⁰⁴ were indeed pretty—exact in (the formation of) their heads—and quadrangular (in shape),—and that the lines were of even tenor; and he commenced to read it from the beginning.’—Papanchasûdanìya.

In alluding to the specimens from the Buddhist caves of western India, Mr. Prinsep remarks, “The old alphabet²⁰⁵ appears to be the very prototype of all the Deva Nagari and Dakshini alphabets; and nothing in the pure Sanskrit tongue has yet been discovered, preserved in this character; indeed it would be impossible that it should, because, still more than the Pali, the alphabet is deficient in many letters absolutely necessary to Sanskrit Syntax.”²⁰⁶

Col. Sykes, that enthusiastic advocate for the superior claims of the Pàli, has drawn all the inferences which may be drawn from the above facts; and I prefer to set them before the reader in his own expressive language: ‘It is incredible,’ he states, ‘to suppose that the modern Sanskrit could have existed without symbols or a character to express its present richness, force, and beauty. How, then, are we to account for the fact of the modern Deva Nagari resolving itself into the ancient Pali letters, and those letters expressing only, not the Sanskrit language, but, the ancient Pali? I cannot see any other way of solving the question, than in the supposition, that at the period the primitive Deva Nagari was expressing the old Pali language * * * * the Sanskrit itself, if it existed independently from the old Pali, was in the same rude state with the Pali.’²⁰⁷

I return to the language: It is indeed a very significant fact that many grammatical forms of the Pali, which may be distinguished from those of the modern Sanskrit, are identical with some of the Vedic peculiarities. This may be rendered evident by a few examples contained in the following notes, taken by me in the course of my studies: and I have no doubt that, with a more intimate acquaintance with these two languages than I possess, the list may be greatly enlarged.

(i.) The Vedas—and I shall here take an example from a portion which is not in metre—contain exceptions to the general rule in Sanskrit, by which a word ending in e or o when not combining with a following a, should cause its elision; as Vasishteadhi ‘over Vasishtha.’ This is frequently the case in the Pali. Thus, in the Dhamma-Pada, dussìlo-asamàhito ‘a reprobate free from meditation.’ It is remarkable that owing to this peculiarity Pali Grammarians do not make elision imperative in this case.

(ii.) I believe short vowels are rendered long in the Vedas. See Wilson’s Sanskrit Grammar, p. 453. This is also the case in the Pali. Thus, san rajjati becomes sà rajjati ‘greatly attached.’ See Bàlavatàra, p. 14; khantì paraman tapo (instead of khantĭ short) ‘Forbearance is the highest religious austerity’—Kachchàyana. Evan game munì chare. ‘Thus, may the muni dwell in the village’—Kachchàyana. A long i is frequently rendered short in the Pali. Thus in the Attanagaluvansa, Chap. i. § 2.

‘Yo bodhisatta gunavà sirĭ Sanghabodhi.’ Again in the well known ‘Ye dhammà hetuppabhavà’ stanza of the Buddhists, vàdì is written vàdĭ; thus,
‘Evan vàdi mahà Samano’
Thus also in Kachchàyana’s Pali Grammar Bhó²⁰⁸ vàdi nàma so hoti ‘He is named Bho vàdĭ.’²⁰⁹

(iii.) In the Pali as in the Vedas the cerebral d is very frequently changed into the Vedic l which is also found in the Pàli. Thus gûlha ‘concealed’; dalha ‘hard’; and the common Sanskrit àrudha becomes àrùlha—See Bàlavatàra, p. 110.

(iv.) Adukshat, the Vedic form of adhukshat is in consonance with the Pali, wherein the aspirate dental is changed into the unaspirate; as idha bhikkhave or ida bhikkhave ‘here, O priests’—See Bàlavatàra, § 24.

(v.) Ange ange, would be quite correct in the Pali, and so it is in the Vedic; although in modern Sanskrit the following short vowel should be elided, as angenge ’member, member.²¹⁰

(vi.) The Pali ayan so aggi, which in modern Sanskrit should be written ayansognih, is found to agree with the Vedic ayan so agnih.

(vii.) Even the phrase Hari Hari yàhi in the Gìtagovinda, is more in accordance with the Pali than the modern Sanskrit.

(viii.) As in the Vedic Sanskrit, wherein the person plural bhis is not unfrequently retained instead of the substitute ais (aih) which is enjoined after nouns in a, the Pali invariably takes the former; as devebhi ‘with god’ Buddhebi ‘by Buddha’; rukkhebhi ‘with tree.’²¹¹

(ix.) ’The Pràkrit (says Bopp) has fully followed out the path commenced by the Veda dialect, and changed into e the â of asmâ-bhis, yushmâ-bhis, as also, in the locative plural, that of asmâ-su, yushmâsu; hence amhe-hi(n), tumhe-hi (n)²¹² amhesu, tumhesu. Moreover, in Pràkrit, all other a bases, as well pronouns as substantives and adjectives, terminate the instrumental plural with ehi (n); and thus kusmehi (n) ‘floribus’, (from kusma,) answers to the Veda kusume-bhis.’ Comp. Gram. § 220.

(x.) The substitution of ya for n before (a) the sign of the instrumental case singular, is to be found in the Pali as in the Vedas, but with this difference—that the substitution is confined in the Pali to feminine nouns; as dhànuyà ‘by a cow’; yàguyà ‘with gruel.’

(xi.) In the Rig Veda, p. 60, narà is given for the dual vocative. The Pali, which does not recognize the dual, adopts this in the plural.

(xii.) Again mitra varunà (see Rig Veda, p. 63) which in the modern Sanskrit is changed into mitra varunau, is in accordance with the Pali.

(xiii.) The gender is found changed in the Vedas, as madhos triptàh ‘satisfied with nectar.’ Here madhu is masculine; and similarly it may be either masculine or neuter in the Pali.—See Bàlavatàra, p. 51.

(xiv.) The curtailment of the neuter plural of nouns in a, by the omission of ni, is as frequent in the Pali as in the Vedas. Thus khettà for khettàni ‘fields’; chittà for chittàni ‘minds.’—See Bàlavatàra, p. 44.

(xv.) In the Vedas mri ‘to die’, of the sixth conjugation is inflected as if belonging to the same class, (the first) to which it belongs in the Pàli.

(xvi.) One voice is used for another in the Vedas; as brahmachàrina michchate (for ti) ‘he wishes for the religious student.’ As the distinction of àtmane pada and parasmai pada is not strictly observed in the Pali, the above is equally admissible in that language, and the same sentence will serve as an example.

(xvii.) In the àtmane pada, the initial tà of a termination is rejected in the Vedas, as dakshina tah saye (sete) ‘he sleeps on his right side.’ So likewise in the Pali the above sentence may be correctly rendered thus; dakkhina to saye (sayeyya or sayetha.)—See Bàlavatàra, p. 104.

(xviii.) In the modern Sanskrit the infinitive is tum; but the Vedic shows different forms, amongst which we have tave, which, as well as tum, is found in the Pali. Thus the Vedic kàrtave ‘to do’ becomes kàtave in the Pali.—See Bàlavatàra, p. 121.

(xix.) The Pali past participle pìtvàna ‘having drunk’ is nearer the Vedic pìtvànan.—See Wilson’s Sanskrit Grammar, p. 477.

(xx.) “From the researches of M.M. Kuhn²¹³ and Benfey,”²¹⁴ observes Dr. Muir in his Sanskrit Texts, p. 168,, ‘it appears that many words, which in modern Sanskrit are only of one, two, or three, &c., syllables, have in the Veda to be read as of two, three, or four, &c. syllables, i. e. as of one syllable longer, in order to make up the full length of the lines required by the metre employed by the Vedic poets. Thus tvam has to be read tuam; vyushtan as viushtan; turyam as turiyam; martyàya as martiàya; varenyam as vareniam; amàtyam as amàtiam; svadhvaram as suadvaram; and svastibhih as suastibhih. Now as this mode of lengthening words is common in Pràkrit, it would appear that the Prakrit pronunciation agrees in this respect with that of the old Sanskrit, in contradistinction to the more recent.’

Such are the relations which the oldest Sanskrit now accessible to us, bears to the Pali; and it must be borne in mind that the former is (1) a modification of two sorts of language, the Vedic or the sacred Sanskrit, and the vyavahàrika or the Vernacular;—that (2) the vernacular, or the ‘current’ language of the Brahmans was in course of time assimilated to their Vedic Sanskrit;—and (3) that the Vedas themselves have been tampered with, so that whilst they received additions and mutations in point of substance, the language itself has indeed undergone a considerable change in point of form. We have thus no truthful evidence of the normal development of the Vedic Sanskrit, and which, if we had, might, perhaps, exhibit that many other forms of the Pali, which are distinguishable from their corresponding forms in the Sanskrit, were at one time as much identical with the Anti-Vedic, as several remnants of the Vedic forms are decidedly the same in the Pàli. A few words may however be deemed necessary in support of the above propositions; and

1. That the Brahmans had two kinds of language. It is a well known fact in the East generally, that nearly every nation has a book-dialect and a vernacular speech. Take, for instance, the Sinhalese. Our vernacular language is generally without the contrivance of Sandhi and compounds; whereas the dialect in which our books are written cannot be understood without much reflection, and, in some cases, without a Commentary. The case was doubtless the same with the Sanskrit. Its refinement and development are such that no one can reasonably conclude that it was ever the spoken language (vyavahàrikavàk) of the Brahmans; yet from several passages in the Sanskrit literature, the colloquial use of a modified form of the Sanskrit may be concluded. It would thence seem that their language was two-fold. A passage that we have already quoted (ante p. lxxxviii.) from the earliest Veda, sets this beyond doubt.

2. The development of the existing Sanskrit, other than the Vedic, indeed proves, that the Vyavahàrika vàk has been so assimilated to their sacred language, that—vires acquirit eundo—it has received additional refinement in its progress.

“The language of the Vedas (says Pr. Whitney) is an older dialect, varying very considerably, both in its grammatical and lexical character from the classical Sanskrit. Its grammatical peculiarities run through all departments: euphonic rules, word-formation and composition, declension, conjugation, syntax. Without entering into any specification of them, which would extend this paper beyond its proper limits, it will be enough to say here that they are partly such as characterize an older language, consisting in a greater originality of forms and the like, and partly such as characterize a language which is still in the bloom and vigor of life, its freedom untrammelled by other rules than those of common usage, and which has not, like the Sanskrit, passed into oblivion as a native spoken dialect, become merely a conventional medium of communication among the learned, been forced, as it were, into a mould of regularity by long and exhausting grammatical treatment, and received a development which is in some respects foreign and unnatural.”²¹⁵ If it may be established that the Vedas were altered, it may indeed be concluded that the anti Vedic forms had been greatly different from the Vedic. This leads to the consideration,

3. Were the Vedas altered by Brahmans? Professor Wilson, in his Review of Professor Max Muller’s valuable work on Sanskrit Literature, thus notices the subject.

“The first and most obvious conclusion to be drawn from the hymns of the Vedas, whatever may be their relative antiquity, whether twenty or twelve Centuries B. C., is that the religion which they inculcate is not that of the Hindus of the present day. The Brahman, who from the time of the code of Munu as we have it, had arrogated to himself the attributes of a god upon earth, is in the Veda only among seven, or even of sixteen priests, acting as a sort of master of the ceremonies, but not invested with any superior rank or authority. Of the distinction of caste, all the indications are faint and uncertain, with one exception—that of a remarkable hymn in the 10th Mandala, the tenor and style of which place it indisputably in a comparatively recent stage, and bring it at least to the Brahmanic period, by which time we know that the Brahmanical system had been organized. There is no mention of temples nor of public worship; the ceremonial is entirely domestic, and so far the formulæ, the language of the Suktas, still constitute the liturgy of the domestic rites of the Hindus. It is very doubtful if images were known, although mention of personal peculiarities, as of the handsome jaws of Indra, might be suggested by a sculptured representation of him. Something else may, however, be meant; but the great feature of difference is the total absence of the divinities, both nomina and numina, who have for ages engaged, and, to a great degree engrossed the adoration of the Hindus. We have no indications of a Triad, the creative, preserving, and destroying power; Brahma does not appear as a deity, and Vishnu, although named, has nothing in common with the Vishnu of the Purànas; no allusion occurs to his Avataras. His manifestation as Krishna, the favorite divinity of the lower classes, for some centuries at least, does not appear. As a divinity Siva is not named, nor is his type the Linga ever adverted to. Durgà and her triumphs, and Kali whom the ‘blood of man delights a thousand years,’ have no place whatever in the hymns of the Vedas. These differences are palpable, and so far from the Vedas being the basis of the existing system, they completely overturn it. It would be an interesting subject of inquiry to discover when and by what means the vast mass of the modern mythology of the Hindus sprang into existence and attained a circulation throughout India.”²¹⁶

The Buddhists enlighten us on this subject: and the following extracts show that not only were additions made to the Vedas, but that the Vedas themselves which are said to have been originally composed in accordance with the Buddhist doctrines of Kassapa, a so-called predecessor of Gotama, were in after times altered by the Brahmans.

In the Ambatta Suttan, Gotama declares that the mantas of the Brahmans (which are identified with the Vedas by the Commentators) were compiled by Attaka (a) Vàmaka, Vàmadèva, (a) Vessàmitta, (a) Yamataggi, Angìrasa, (b) Bhàradvàja, (a) Vàsettha, (a) Kassapa (b) and Bhagu. (b)²¹⁷

Te kira dibbena chakkhunà oloketwà parùpaghàtan akatvà Kassapa sammà sambuddhassa pàvachanèna sahà sansandhetvà mante ganthesun. Aparà’ paran pana Brahmanà panàti pàtàdìni pakkhi pitvà tayo vede bhinditvà Buddha vachane saddhin viruddhe akansu.—Sumangala Vilàsinì.

‘Those sages after obtaining²¹⁸ the supreme discourses of Kassapa Buddha, through the medium of their divine eyes, compiled the Manta conformably to those discourses, without the mention of tormenting (sacrifices.) The Brahmans in course of subsequent times, however, set aside the three Vedas, and made (a different compilation) by departing from the words of Buddhism, and introducing (an authority for) life-slaughter &c.’

Tìsu vedesù’-ti àdisu Irubbeda Yajubbeda Sàma-veda sankhàtesu tesu vedesu: tayo eva kira vedà Attakàdìhi dhammikehi isìhi lòkassa sagga magga bhàvan’ atthàya kathà; tenevahi tayotì vuchchanti. Athabbana vedo pana pachchhà adhammikehi Bràhmanehi pàna vàdhàdi atthàya kato, purimesucha tìsu vedesu teheva dhammika yàjayo apanetvà yàga vadàdi dìpikà adhammika sàkhà pakkhittà-ti veditabbà.

‘Tìsu Vedesu i. e. the three Vedas called Irubbedda (Rig. Veda.)’ Yajubbèda (Yajur), and Sàma Veda. These three Vedas were made by Attaka, and other religious sages for the attainment of the path of heaven by mankind: they are therefore called ‘three.’ The Athabbana (Atharvan) Veda, however, was since made²¹⁹ by irreligious (wicked) Brahmans with a view to introduce life-slaughter (animal sacrifices). Moreover, be it known, that they themselves introduced into the first mentioned Vedas, irreligious branches which proclaim sacrificial torments &c., after expelling virtuous promoters of sacrifices.—Vimata Vinodana Tìkà.

I have thus glanced over the peculiarities which distinguish the oldest from the modern Sanskrit—the structure of the latter being that which may be regarded as the most expressive and harmonious that has ever been attuned to human utterance.”

An examination of Pali Grammars establishes beyond all doubt the affinity which it bears to the Sanskrit (i.) “from which it differs only in such modifications as are exhibited by those European dialects which are most immediately derived from the unknown idioms of antiquity.”²²⁰

(ii.) In a review of the Revd. B. Clough’s Pali Grammar (see Asiatic Journal for 1827, p. 663) the writer bears similar testimony to the identity of Sanskrit with the Pali. He says, “every essential part of it (the Pali) is found in the Sanskrit. The vocabularies of its nouns and of its verbal roots are nearly the same. The Grammar is also formed on the same model, but is much more simple.”

(iii.) Professor Bopp also thinks that “the relation between these two idioms (the Sanskrit and Pali) is nearer than that which subsists between most of the distinct branches of the Indo-European system, and that it may be compared to the degree of affinity which the Latin bears to the Greek.”²²¹

(iv.) Fausboll in his Introduction to the Dhammapada (p 6.) states the relationship of the Pali or Màgadhì to the Sanskrit, thus: “Inter literas buddhisticas sacras hunc librum antiquiorum in numero habendum esse ex sermone, quo utitur, elucere videtur, qui multis in rebus a sanscritico et quidem antiquissimo prope abest, et multum ab eo dicendi genere differt, quo utuntur Sutta prosaica et scholia Buddhaghosæ. Huc pertinent: nom. præs. participii in—am, ut ganayam, rodam; a. pers. plur. præs. medii in—are, ut socare, upapajjare; dat. gerund., ut netave pahàtave; præterea formæ, quales sunt: karoto et kubbato (karontassa) káhiti (karissati) al., arahatam (arahantànam), sabbhi (santehi), vaddha (vuddha), klesa (kilesa), cetya (cetiya) etc.

The weightiest and most conclusive authority on the subject, however, to which I wish to call attention, is contained in the following observations of Dr. Weber, and which I have only seen through the medium of the valuable translations in Dr. Muir’s Sanskrit Texts, since these sheets were prepared for the press.

Speaking of the way in which he conceives the Prakrits²²² to have risen, Dr. Weber says:—

(v.) “I take this opportunity of declaring myself distinctly against a commonly received error. It has been concluded from the existence (in inscriptions) of Prakrit dialects in the centuries immediately preceding our era, that the Sanskrit language had died out before these dialects were formed; whereas we must, on the contrary, regard the development of both the Sanskrit and the Prakrit dialects from one common source, viz. the Indo-Arian speech, as entirely contemporaneous. . . . . For a fuller statement of this view I refer to my ‘Vajasaneyi Sanhitæ specimen,’ ii. 204-6; and, in proof of what I have urged there, I adduce here the fact that the principal laws of Prakrit speech, viz. assimilation, hiatus, and a fondness for cerebrals and aspirates are prominent in the Vedas, of which the following are examples: kuta=krita, R. V. i. 46, 4.; kāta=karta, (above, p. 30); geha=griha, (above, p. 40); guggulu=gungulu, Kātyāy., 5, 4, 17; vivittyai=vivishtyai, Taitt. Arany. x. 58; krikalāsa, Vrih. Ar. Mā. i. 3. 22.=krikadāsu, Rik. i. 29. 7; purodāsa,=purolāsa (comp. dasru=lacryma); padbhih=padbhih; kshullaka=kshudraka; bhallāksha=bhadrāksha, Chhandogya, 6. 1. (gloss); vikirida=vikiridra (above p. 31); gabhasti=grabhasti, or garbhasti; nighantu=nigranthu; ghas=gras; bhanj=bhranj=bhuj=bhruj; bhas=bras. . . . . . Comparative philology exhibits similar phonetic prakritizings within the circle of the Indo-germanic languages as compared the one with the other.” The same writer says in his Yajas. Sanh. specimen ii. 203. ff.;²²³ “I incline to the opinion of those who deny that the Sanskrit Bhāshā, properly so called, was ever the common spoken language of the whole Arian people, and assign it to the learned alone. Just as our modern high German, arising out of the ancient dialects of the Germans, reduced what was common to all to universal rules and laws, and by the power of analogy obliterated all recollection of varieties; and just as, on the other hand these dialects, while they gradually degenerated, often preserved at the same time fuller and more ancient forms; so also the Vedic dialects, became partly combined in one stream, in which their individual existence was lost, and so formed the regular Sanskrit Bhāshā, and partly flowed on individually in their own original (Prākrita) irregular force, and continued to be the idioms of different provinces, in the corruption of which they participated. The Sanskrit language and the Prakrit dialects had, therefore, a common and a simultaneous origin: the latter did not spring out of the former, but rather, being connected by a natural bond with the ancient language, have often a more antique fashion than the Sanskrit, which, being shaped and circumscribed by the rules of grammarians, has sacrificed the truth of analogy for the sake of regularity. The Prakrit tongues are nothing else than ancient Vedic dialects in a state of degeneracy; while the Sanskrit (or Epic) bhāshā is the sum of the Vedic dialects constructed by the labour and zeal of grammarians, and polished by the skill of learned men. In this way we obtain an explanation of two facts: 1st, That the very same exceptions which are conceded by grammarians to the Vedic language (chhandas) are often found in the Prakrit dialects, being in fact nothing but original forms; and 2nd, That in the Vedic writings, forms and words occur which are more irregular than any Sanskrit word could ever be; for as yet no fixed rules of euphony, orthography, or formation existed,—rules which were eventually deducted in part from those very irregularities. All the irregular forms which prevail in the Prakrit tongues are to be found throughout the Vedas. In the latter, the faculty which creates language is seen exuberant in its early power, while in the former (the Prakrits) it is seen in the degeneracy of full blown license, luxuriating wantonness, and at last of senile weakness. Assimilation, the hiatus, and a fondness for cerebrals and aspirates, play an important part in the Vedas, not so much in those portions which are peculiar to the Yajur-veda (which, as forming a transition from the Vedic to the Epic period, or rather itself initiating the Epic period, has also a style of language of a more modern cast, and adapted to a grammatical rule), as in the older forms and words of the Rig-veda, many of which were difficult to understand in the age of the Aitareya and Satapatha Brāhmanas (paroxavrittayah: comp. Roth. p. li. Nighantavah.) There occur moreover in the Epic poems many words which, however corrupted, have been received into the Sanskrit sometimes with no change, sometimes with very little, from the Prakrit languages in use among the greater part of the people.”

I have thus laid before the reader, the facts stated by eminent philologists,—men who enter quite as fully into a comparison of the Grammars, as of the system of words in the two languages,—to shew the difference between the Pali and the Sanskrit; and those statements, it would seem, go the length merely of establishing the theory with which we have set out, viz., that the Pali, like the Sanskrit, is the offspring of an unknown language; and, whether the cultivation of both commenced at one and the same time or not, that “the former stands in fraternal connection with the latter—not in the relation of descent from it.”

Before however dismissing this part of the subject, it may not be out of place here to advert to a few circumstances which point out that the Màgadhì (Pali) had its origin in the Punjab, or Bactria.

Dr. Stevenson remarks that “it seems highly probable that the ruder dialect from which the present Sanskrit has been formed was the spoken tongue of the tribe, who, under Bharat, as they themselves relate, settled in upper India, and afterwards gave the name of their Sovereign to the whole country, which extends from Cape Comorin to the Himalaya mountain.”²²⁴ The Magadhas, whose language was the Pali, also trace their origin to Bharat. It is hence probable (for, in questions relating to languages we can only deal with probabilities²²⁵,) that both the Pali and Sanskrit branched off from the same parent stem,—the latter taking a lead in a comparatively civilized country, attaining its present high refinement, so as to tempt men to mix it with their non-Sanskrit tongues on the South of the Vindhya; whilst the former was banished from the land from whence it arose, to different Indian countries, where it assumed the various forms exhibited by the Jaina dialects and the Maharàstri²²⁶ of the Grammarians, and to our own “utmost Indian Isle, Taprobane,” where alone it has become a dead language.

Another circumstance is deserving of attention. Tradition, which is after all the best evidence on these matters, says that the Màgadhì was the language of Saurasena,²²⁷ on the banks of the Yamunà, close upon Mathura.²²⁸ The language of Saurasena²²⁹ is also called Pràkrit by the Brahmans; and they treat Màgadhì under the same name, and place it in the same class with Panchàla, or the language of Punjab,²³⁰ by which we understand the Zend. It is also clear that the Sanskrit Brahmans had also come from Bactria,²³¹ and that several languages found in that locality, for instance the Persian, the Phelevi, and the Zend,²³² are more nearly related to the Sanskrit than her Indo-European sisters. On this subject, the erudite editor of the Bengal Asiatic Society’s Journal (see vii. p. x.) remarks: “The history of India had been traced back to the period before the invasion of Alexander, and had been verified at each step by coins and by inscriptions; but the language of Bactria and of Persia, at the period of that Conquest, was still insufficiently ascertained. The Bactrian alphabet was already more than half discovered through the comparison of letters upon coins with bilingual superscriptions. Several inscriptions, as obtained from the Topes excavated, or as forwarded by travellers from within the ancient limits of Bactria, were nearly deciphered, so that very little remained to perfect the discovery also [by Mr. Prinsep], and to establish, that the ancient Pali language, or something very closely resembling it, prevailed over all those countries.”²³³ It would hence seem that the Pali, which approaches closely to the Sanskrit even in some of its oldest forms, had originally started from the same country in which the cradle of the Sanskrit Brahmans is placed, “in or near Bactria.”

VII. That the Pali was, at least, in the time of Gotama Buddha, 628-543. B. C., a highly cultivated language of Magadha and several adjacent countries, does not admit of reasonable doubt. We find from numerous works extant in Ceylon, that in grammatical structure and precision at least, it is but very little inferior to the Sanskrit. A language too, which is capable of enunciating discourses so varied and abstruse as the Pitakattàya and the voluminous Commentaries thereon, cannot but be deemed copious. It is rich in expression; and its force and harmony are but one degree removed from the idiom that has been Sanskritized.

Professor Wilson in his Introduction to Vikrama and Urvasi, thus speaks of the Pràkrit, regarding it as an exhibition of the Màgadhì. “The richness of the Prakrit in this play, both in structure and in its metrical code, is very remarkable. A very great portion, especially of the fourth act, is in this language, and in that act a considerable variety of metre is introduced. It is clear therefore that this form of Sanskrit must have been highly cultivated long before the play was written.”

If such is the case as respects the Pràkrit of a period nearly three centuries after the Buddhist era, a higher refinement ought certainly to be accorded to the Pàli, the language of the time of Gotama.²³⁴ The presumption is therefore irresistible, that it had been highly cultivated very long before the age of Gotama.

I have already adverted to the dynasty of the Màgadha kings, which commenced from the war of the “Maha Baharata,” an event involved in the Mythological obscurity of the past; and it is also therefore to be presumed, that their language is of as great an antiquity as their kingdom is certainly older than the written Vedas.

“But in very truth,” says Hodgson, “the extant records of Buddhism, whether Sanskrit or Pràkrit, exhibit both languages in a high state of refinement.”²³⁵ If this be true of the Pràkrit, it is undoubtedly so of the Pali,—“a rich, refined, and poetical language of the land in which Buddhism as promulgated by Sàkya or Gotama had its origin, at which period it was a highly refined and classical language.”²³⁶

When therefore we consider the high state of refinement, to which the Pali had in very early times attained as a language,—its copiousness, elegance, and harmony, combined with its high antiquity, and its comparative simplicity, both verbally and grammatically,—its relationship to the oldest language of the Brahmans, from which their present dialect has been Sanskritized,—its claims to be considered the Vyavahàrika vàk of the Brahmans to which the Rig Veda refers,—its concurrence with some of the Indo-European languages, in some forms which differ from the Sanskrit,—its identification with the only original Pràkrita dialect, which was “similar to the Sanskrit”,—the absence of any statement in old Brahman writers to the effect that that Pràkrita dialect was a derivative of the Sanskrit,—the great improbability of a derivative being denominated the [prakriti] Pràkrita,—the palpable inaccuracy of the definition by which in modern times, it is called “the derived, the vulgar, or the ungrammatical”—the absence in it of many a peculiarity which distinguishes derivative tongues,—and the probability that it had issued from the same ancient seat (Bactria or Punjab) from whence the Sanskrit itself had taken an easterly direction,—I believe it may be concluded that the Pali and the Sanskrit are, at least, two dialects of high antiquity, contemporaneously derived from a source, of which few, if any, traces can be discovered at the present day.

To the above remarks on the relationship of the Pali to the Sanskrit, I had originally intended to add a brief sketch of the progress of Buddhism in the East, along with the Pali language, with which the former is inseparably connected. But the great difficulties which I have experienced in printing, have compelled me to reserve the subject for a future publication. For the same reason I have given the Text in Sinhalese characters. If, however, I should be permitted to complete the translation of the remaining seven books of Kachchàyana’s Grammar, with which I am now engaged, I hope to be able to forward to my publishers in England, a complete edition, including the matter which has been reserved, together with the Text, in Roman characters.

Of the work now submitted, with great diffidence, to the European public, I have little to say, beyond expressing a hope that they will not severely judge of this my first attempt at translating from the Pali into English, and transferring, what may be termed “algebraic aphorisms” into intelligible phraseology. I am sensible that there are many errors, of omission and commission, in the translation; but they are such as I could not avoid. For, though living at “the very fountains of Pali literature,” I have, nevertheless, been unable to consult a single friend, either as to the choice of my language or the correctness of my renderings into English. I have indeed had much assistance from native Pandits, of whom I shall have occasion to speak hereafter, but none of them possess a sufficient knowledge of the English language to be able to rectify an incorrect translation. The numerous Tìkàs and Comments, again, to which I have had access, being entirely in Sinhalese, could not afford any greater help than I have derived from the Pandits. The only European Pali Scholar in this Island, I mean the Rev. D. J. GOGERLY of the Wesleyan Mission, who had “cheerfully” promised to assist me in my labours, and to revise my translation, was, unfortunately, removed by death, at the very period when I desired to avail myself of his invaluable advice.

As to the deficiencies of my language, I believe it is unnecessary to offer any apologies,—for, I have no doubt, the European reader will make great allowances for the shortcomings of one who cannot claim the English as his native tongue.

The errors of the press are far too numerous to be passed over in silence. I believe few persons in this Island are ignorant of the difficulties which an Author has to contend with in publishing his works on Oriental literature in Ceylon.—Witness the Mahavansa, whose learned translator was obliged to append no less than thirty-five closely printed quarto pages of corrections. Referring to my own work, I may be permitted to remark, that the Compositors in Ceylon entertain an aversion to handling a MS. containing Pali or Sanskrit passages written in Roman characters, and especially with diacritical marks. Indeed it was with great difficulty that the men in one of our printing establishments could be induced to undertake this work, or, when undertaken, to continue it. From the universal inattention in Ceylon to the orthography of Oriental words and names, they sometimes took it upon themselves to set aside my spelling, and to adopt their own: and this, I need hardly remark, has entailed much labour in the correction of the press. It will scarcely be credited, that for the correction of these errors, I have often had to revise six or seven proofs of one and the same sheet. Even with such labour, and with all the vigilance I could bestow, it has been impossible to avoid a great many errors, which have rendered it necessary to add rather a copious list of Errata. My absence too, from Colombo, and from my library, during the whole of the period during which this work was going through the press, has added not a little to my difficulties.

These remarks, however, apply to the Introduction. The Grammar, which is comparatively free from errors, has been printed in the Wesleyan Mission Press, whilst its Translation, and the Appendix, have received the invaluable supervision of Mr. SKEEN, the Government Printer, at whose establishment they were printed.

In the Introduction and the Translation, I have adopted the following scheme of orthography, which will be found to approach closely the system adopted by Sir WILLIAM JONES.

VOWELS.

[Sinhalese] [Sinhalese] [Sinhalese] [Sinhalese] [Sinhalese] [Sinhalese] [Sinhalese] [Sinhalese]
a à; i ì; u ù; e o

CONSONANTS.

Gutturals—[Sinhalese] k [Sinhalese] kh; [Sinhalese] g [Sinhalese] gh; [Sinhalese] ṅ
Palatals—[Sinhalese] ch [Sinhalese] chh; [Sinhalese] j [Sinhalese] jh; [Sinhalese] ñ
Linguals—[Sinhalese] ṭ [Sinhalese] ṭh; [Sinhalese] ḍ [Sinhalese] ḍh; [Sinhalese] ṇ
Dentals—[Sinhalese] t [Sinhalese] th; [Sinhalese] d [Sinhalese] dh; [Sinhalese] n
Labials—[Sinhalese] p [Sinhalese] ph; [Sinhalese] b [Sinhalese] bh; [Sinhalese] m
[Sinhalese] y, [Sinhalese] r, [Sinhalese] l, [Sinhalese] v, [Sinhalese] s, [Sinhalese] h, [Sinhalese] ḷ, [Sinhalese] ṃ

Owing however to the absence of some of the accented letters, such as ṭ, ḍ, ḷ, &c., I have been obliged in printing this work to deviate from the above system; and to adopt italics in their stead. This substitution again, has not been uniformly attended to by the printer; whilst I myself have failed to pay any attention to the difference of ṅ ṇ n and ṃ, all which will be found expressed by an unaccented n.

It only remains for me to acknowledge the assistance which I have received in the course of this publication; and here I cannot adequately express the sense of my obligations to my Pandit, BATUVANTUDAVE, who has assisted me during a considerable period of time which has been devoted to the translation of the following sheets, and the extracts in the Introduction. I have, with his permission, given expression to a few opinions of my own, upon which, from religious differences and the dissimilarity of our education, and other causes, agreement was found impossible. But, whenever we differed, it is but right to state that I did not hastily reject his views without first devoting my best and most serious attention to them.

In the collection of materials for this rather lengthy introduction, the reader will perceive from the Notes and Annotations, the extent of assistance which I have received from the published works of Colebrooke, Wilson, Max Muller, Ballentyn, Muir, Burnouf, Lassen, Weber, Spiegel, Goldstucker, Fausboll, &c.; and I must not omit also to state that I have been greatly assisted by several learned Buddhist priests, especially SUMANGALA of Hickkaduwa, to whom my best thanks are due.

It is impossible to pass over, without due acknowledgement the useful hints and information I have obtained from Mr. J. R. BLAKE, in the course of frequent conferences on the subject of Oriental literature, and also the still more valuable translations which he has made for me from several German and French writers.

I cannot conclude more appropriately than with the words of one²³⁷ in the same field of labour in which I am engaged, and who has less reason than myself to say, “et nunc haec folia non sine justo timore in lucem emitto, quæ, si non omnino displicuerint viris doctis, jam operae pretium factum esse censebo.”

JAMES ALWIS.

Footnotes

1. See my Sidatsangarà, p. 222. ↩

2. Ib. p. xi., et seq. ↩

3. For a specimen, See Appendix. ↩

4. “The high state of cultivation to which the Pali language was carried, and the great attention that has been paid to it in Ceylon, may be inferred from the fact that a list of works in the possession of the Singhalese, that I found during my residence in that Island, includes thirty-five works on Pali Grammar, some of them being of considerable extent.”—Rev. S. Hardy’s Eastern Monachism, pp. 191-2. ↩

5. I find that this is also extant in Burmah. The Rev. F. Mason of the Baptist Union says:—‘The grammar reputed to have been written by Kachchàyana, still exists. I had a copy made from the palm-leaf, on small quarto paper, and the Pali text occupies between two and three hundred pages, while the Burmese interpretation covers more than two thousand. I made a compendium of the whole Pali and English, a few years ago, on the model of European Grammars, which might be printed in one or two hundred pages, and convey all the information contained in the two or three thousand in manuscript.’—Am. Or. Journal, iv. p. 107. ↩

6. Professor Spiegel’s Kammavachà, p. vii. ↩

7. See Sanyut Sangiya. ↩

8. Pràkrit Prakàsa, p. 179. ↩

9. See the Friend, vi. p. 236. ↩

10. Prinsep, Bengal As. J., vii. p. 282. ↩

11. Turnour’s Mahàvansa, p. xxii., where he merely gives the opinion of the Buddhists; and this is no more correct than the Brahmanical opinion, that Pràkrita means ‘the derived.’—Vide post, p. xxxix. ↩

12. See Abhidhànapadìpikà, p. 71. It is indeed not a little curious that Mohammedans, between whom and the Buddhists there was no intercourse at the period when their sacred books were written, call the larger portions of the Koran “Sowar,” (‘Sura,’ Sing.) signifying precisely as the word Pàli does, ‘a row, order, or regular series.’ The Arabic Sùra, whether immediately derived from the Sanskrit ‘Sreni’ or not,—is the same in use and import as the Sura or Tora of the Jews, who also call the fifty-three Sections of the Pentateuch, Sidàrim, a word of the same signification. ↩

13. Itaran pana; Atthànan sùchanato; sùvattato savana totha sùdanato Suttànato sutta sabhà gatocha suttan suttanti akkhàtan. ’The other (which is) the Suttan, is called ‘Suttan’ from its illustrating the properties (of duties); from its exquisite tenor; from its being productive (of much sense); and from its overflowing (tendency) the protection (which it affords); and from its being like a string.’—Buddhagosa’s Atthakathà. ↩

14. Hevan cha hevan cha me pàliyo vadetha: ‘Thus, thus shall ye cause to be read my pàliyo or edicts.’—Prinsep’s Asoka Inscrip. ↩

15. Yattàcha suttena sangahìtàni pupphàni navi kirìyanti naviddhan siyanti eva me thena sangahìta atthà. ‘As flowers strung together with a string are not scattered, so likewise the doctrines which are taken from this (Suttan) are not lost.’—Sumangala Vilàsini. ↩

16. Mahavansa, p. 253. ↩

17. Ib. p. 252. ↩

18. Abhidhànapadìpikà, p. 16. ↩

19. Wilson’s Sanskrit Dictionary. ↩

20. B. A. J., vol. vi. p. 511.—Pathama mahà sangìti nàma esà kinchà-pi vinaya pitake Tanti’n àrulhà. ‘This first great rehearsal was moreover rendered into tanti, (the original discourses or the text) on the Vinaya Pitaka,’—Sumangala Vilàsinì. ↩

21. “Nara and ànara ‘human and non-human.’ ↩

22. The “tivattambunidi.” ‘The ocean, encompassed with three circles, is here used for “metempsychosis;” and the three barriers are Kamma, action which begets merit and demerit; Klesa evil, trouble, pain or sorrow; and “Vipàka” the rewards of merit and demerit.’ ↩

23. Sons—a term applied to disciples. ↩

24. That is, ‘Preserve the Sila or precepts.’ ↩

25. Lit.—“Lamp of Nouns.” ↩

26. “He reformed the religion.”—Upham Vol. 1. p. 299. ↩

27. “Association or Congregation performing the same duties.” ↩

28. Pachchaya—“Objects of maintenance” which are four, viz., chivara ‘garments’; pindapata ‘food’; Senasana ‘sleeping objects’; gilana pachchaya ’that which is necessary for the sick—medicines.” ↩

29. See Ceylon Almanac for 1834. ↩

30. “He built the Viharas in the City of Anuradhapura”—ib. at p. 190. ↩

31. “The King also made several hundreds of houses and many streets arranged with shops.”—Mahawansa. ↩

32. “He formed many pleasant and delightful gardens.”—Mah. C. B. A. S. J. p. 148. ↩

33. “He built three more Cities.”—Upham’s Mahawansa, p. 277. ↩

34. “He formed Paddy fields.”—Mah. C. B. A. S. J., Vol. VII., p. 141. ↩

35. “The King also repaired many ancient Tanks.”—Mahawansa, ib. p. 149. ↩

36. This is a Pali translation of the Sinhalese proper name Velgam, ↩

37. Inscription in Ceylon Almanac for 1831. ↩

38. Mahawansa, p. lxvi. ↩

39. Eka seshah: “one left out” i. e., the omission of one to designate the same by another, which has been mentioned; or, conversely, the expression of one name to designate another omitted name of the same genus or family; as Asvinau “the two Asvin,” in the dual, designate ‘the Physicians of heaven, and twin sons of the sun, or children of the constellation Asvini,’ who are separately named Nasatya and Dasra. ↩

40. See my Sidatsangara p. li. ↩

41. Lit. ‘who has attained to the end of sciences.’ ↩

42. Lit. ‘love for others.’ ↩

43. Lit. ‘Son-pupil.’ ↩

44. ‘Those religious observances which lead to the destruction of keles.’ ↩

45. For a list of Pali Grammars, see Appendix. ↩

46. Satta situttana Sutta cha sata sun pamanato=687 Suttans. ↩

47. Vakkhami “I utter”—The true import of this word taken in connection with the allegation that ‘Kachchayana published (pakasesi) his Grammar in the midst of the priesthood’ may lead to the inference that it had at first only a memorial existence. But, since the same words are found used in works which were doubtless written from the beginning, the phraseology alone does not, I apprehend, warrant that inference. ↩

48. Moggallayana disputes the correctness of this Suttan, and says that the Pali, alphabet contains forty-three characters, including the short e (epsilon) and o (omicron.) The Sinhalese Alphabet, which is nearly as old as the Sinhalese nation also omits these. This is evidence of that language being derived from the Pali. ↩

49. The anusvara. ↩

50. Names or technical terms. ↩

51. Adho-thitam “that which stands below [after such separation.]” The word below must however be understood to mean preceding; for in composition, which Eastern writers regard as a tree from bottom to top, the first-written character is considered as being at the bottom or below the rest. ↩

52. These are doubtless, taken from “the contemporaneous History of Buddha,’ which, as stated by Buddhagosa, “contains records such as these—‘at such a period Bhagava dwells at Savatthi, or at the Jetavana vohara’ . . . . . . ‘he dwells at Rajagahan, or at the Veluvana Vohara’. . . . . ‘he dwells at Vesali’ &c.’ ↩

53. These are said, in the Buddhistical annals, to have been compiled by Attaka, Vessamitta, Yamataggi, Angirasa, Bharadvaja, Vasettha, Kassapa, and Bhagu. ↩

54. Mahabharata is frequently mentioned under the designation of Itihasa. Asvalayana is mentioned by Gotama. Vide extracts infra. ↩ ↩²

55. “Another name for the Rupasiddhi.”—In the above note Turnour identifies Rupasiddhi with Niruttipitaka. But, it would seem that the latter is an original work of Maha Kachchayana, different from his Grammar, and different also from his theological work entitled the Nettipakarana. See Kachchayana Vannana. ↩

56. Studies such as abstract meditation &c., preparatory to the attainment of the paths leading to Nibban. ↩

57. Or rather pondering on what he had observed. ↩

58. Vide extract therefrom infra ↩

59. See Appendix. ↩

60. Sanskrit Literature p. 312. ↩

61. See the introductory Verses of Kachchayana quoted at p. xvi. ↩

62. See Sanyutta Nikaya—as to the Magadhi being the language of Magadha. ↩

63. This word is sometimes written gantha. ↩

64. The commentator explains that by a ‘long letter’ are meant the Sanskrit e and o which become short in some languages, or are each equal to one syllabic instant. ↩

65. Papanchasudaniya. lib. iii. in my possession. ↩

66. Id. also Maha Vagga. See Chamakkhandaka and a host of other authorities. ↩

67. Id. Sanyut Nikaya—&c. &c. ↩

68. Sumangala Vilasini. ↩

69. Maha Vagga. ↩

70. Sanyut Nikaya; Maha Vagga, and atthakatha to Dhammapada ↩

71. Papancha Sudaniya. ↩

72. That such was the fact may be easily gathered from several authorities. ↩

73. Mukhamatta ‘the very (word of) mouth’, a term which does not necessarily imply the absence of writing. ↩

74. See Sivpilisimbia in Clough’s Dictionary. Turnour has defined this to be ‘the attainment of the four gifts of sanctification.’ ↩

75. In the Chammakkhandaka Section of the Maha Vagga. ↩

76. Dr. Muir in his Sanskrit Texts, says that ‘the people whom Yaska designates Prachyas, or men of the East, must have been the Kikatas or the Magadhas, or the Angas, or the Vangas.’—p. 371. In the Buddhist annals however, the word Pachchanta is used to designate all the countries beyond the Majjhima desa, which is thus defined in the Maha Vagga: ‘Here the Pachchanta are these Countries. On the East [of Majjhima] is the market town called Kajangala, and on the West Maha Sala. Beyond them is the great country of Pachchanta, and this side of it is the Majjha. On the South-east is the river called Salalavati. Beyond it is the Pachchanta country, and this side of it the Majjha. On the South is the town called Setakanni. Beyond it is the Pachchanta country, and this side of it the Majjha. On the West is the Brahman village called Thuna. Beyond it is the Pachchanta country, and this side of it the Majjha. And on the North is the mountain called Usuraddhaja. Beyond it is the Pachchanta country, and this side of it is the Majjha.’ ↩

77. It is stated in the Buddhist annals (see the first Banavara of the Parinibban Suttan) that this city, which in modern times has received the name of Patna, was built during the lifetime of Gotama, for the purpose of checking the Vajjians; and it is also stated that at the time it was built by Sunidha and Vassakara, two ministers of the reigning prince Ajatasatta, Gotama predicted its future opulence and grandeur as well as its partial destruction by fire and water. ↩

78. See Appendix. ↩

79. Dipavansa. ↩

80. See appendix. ↩

81. For an explanation of this, see Hardy on Buddhism, p. 500. ↩

82. Which is explained to be—‘to eat only to live for religion—but not to live to eat only.’ ↩

83. Refraining from much sleep—which the recluse is restricted to the middle watch of the night, or four English hours at midnight. ↩

84. Abstract meditation which leads to the entire destruction of all cleaving to existence. See Gogerly’s Essay C. B., A. S. J. ii. 15. ↩

85. See remarks of Mon. Burnouf in his Histoire du Buddhisme, 1. p. 141. ↩

86. Attanagaluvansa, Cap. ii. § 1. ↩

87. Max Muller’s Sanskrit Literature p. 84. ↩

88. Quoted in the Appendix (p. 382) to Remusat’s translation of the “Foe Koue Ki ou Relation des royaumes Bouddhiques.” See also M. Julien’s Hist: de la vie de Hiouen Thsang, p.p. 102, 165. ↩

89. The common date of the Nirvana of Buddha is B. C. 543 ; but Hiuan Thsang (as quoted in a note to p. 237) gives several different dates as current in India in his time, the latest of which is about B. C. 360. ↩

90. See Turnour’s “Mahawansa”, Introd : p.p. 25—27. ↩

91. Professor Goldstucker says in his work on the Age of Panini: “In general sameness of names, like that of Katyayana, can never prove the identity of persons, [who bore them] ; there is nothing proved by it, except that both belonged to the same family, or (‘resp.’) were followers of the same School, the Katas.”—p.p. 187-8. ↩

92. Vide extract from Dandialankara, infra. ↩

93. See comparative Tables, infra. ↩

94. I have not been able to procure a copy of this, for the purpose of comparison. ↩

95. Colebrooke’s Miscellaneous Essays. Vol. ii. p. 4. ↩

96. “Sagara made the Yavanas shave their heads.”—Vishnu Purana, iv., 3. ↩

97. See Pr. Benfey’s Article on India. ↩

98. Wilson’s Hindu Theatre, II, p. 179. ↩

99. Vide Wilson’s Ariana, p. 230. ↩

100. Chapter x. 43. 44. ↩

101. Anusasana Parva, verses 2103 et seq. ↩

102. See Turnour’s Introd. to Mahavansa, xl., vi. ↩

103. For the Brahmanical definition of this word as well as dasa ; See. Dr. Muir’s Sanskrit Texts ii. pp. 379, 380. ↩

104. Sankrit Literature p. 245. ↩

105. ‘None of the more ancient works seem to be now extant’—Colebrooke’s Essays ii. pp. 5, 6. ↩

106. Indeed this word like the others, has claims to a higher antiquity than the age of Gotama. See Dena in Pr. H. Wilson’s Glossary of Judicial Terms. ↩

107. See Professor Max Muller’s remarks hereon in his Sanskrit Literature p. 304 et seq. ↩

108. For which as well as various other passages from French and German writers my acknowledgements are due to my Teacher J. R. Blake Esq, ↩

109. Introduction to Panini by Bohtlingk—See contra by Weber in his Introduction. ↩

110. The following observations have an important bearing upon the question :— ’This is the only date, the fixing of which is called “impossible,” in Muller’s Ancient Sanskrit Literature ; and as it has hitherto been my fate to differ from this work in all its chronological views, I seem merely to follow a predestined necessity in looking upon the date of Patanjali as the only one which I should venture to determine with anything like certainty. ’I do so, because Patanjali, as if foreseeing the conjectural date which some future Pandit would attach to his life, or the doubt that might lift him out of all historical reach, once took the opportunity of stating a period before which we must not imagine him to have lived, while on another occasion he mentions the time when he actually did live. ’“If a thing,” says Panini, “serves for a livelihood, but is not for sale” (it has not the affix ka). This rule Patanjali illustrates with the words “Siva, Skanda, Visakha,” meaning the idols that represent these divinities and at the same time give a living to the men who possess them,—while they are not for sale. And, “why ?” he asks. “The Mauryas wanted gold, and therefore established religious festivities. Good ; (Panini’s rule) may apply to such (idols, as they sold) ; but as to idols which are hawked about (by common people) for the sake of such worship as brings an immediate profit, their names will have the affix ka.” ‘Whether or not this interesting bit of history was given by Patanjali ironically, to show that even affixes are the obedient servants of kings, and must vanish before the idols which they sell, because they do not take the money at the same time that the bargain is made—as poor people do,—I know not. But, at all events, he tells us distinctly by these words that he did not live before the first king of the Maurya dynasty who was Chandragupta, and who lived 315 B. C. And I believe, too, if we are to give a natural interpretation to his words, that he tells us, on the contrary, that he lived after the last king of this dynasty, or in other words later than 180 before Christ. But he has even been good enough to relieve us from a possibility of this doubt when commenting on another rule of Panini, or rather on a criticism attached to it by Katyayana.’—Goldstucker’s “Panini” p.p. 228—9. ↩

111. Professor Max Muller’s Sanskrit Literature p. 240. ↩

112. Gotama himself says that long before his advent the Brahamans had fallen off from their high sanctity. ↩

113. Goldstucker’s Panini, p 225 et seq. ↩

114. As follows :—’As by the general reception of Goldstucker’s results, a relatively chronological result only concerning Panini’s connection with the work in question has after all been attained ; let us now proceed to the crown with which he has adorned that work, if his which awaits this consummation, viz., the demonstration that Panini must have lived before the time of Buddha. This indeed, which, if true discloses an important discovery, is founded upon two points. First, upon this, that Panini does not mention the name of Sakyamimi. Now we learn from Goldstucker himself (p. 18. vide supra p. 48.) that nothing is to be inferred from that circumstance—“sometimes the words which belong to his (Panini’s) province will be at the same time also of historical and antiquarian interest ; but it does not follow at all, that, because a word of the latter category is omitted in his rules, it is absent from the language also.” The second point is, that Panini mentions indeed the word nirvana, but in the sense of “free from wind, wind-still,” and not in the sense in which the word is held by the Buddhists :—“and since it is not probable that he would have passed over in silence that sense of the word which finally became its only sense, I hold that this sense did not yet exist in his time : in other words that his silence offers a strong probability of his having preceded the origin of the Buddhistic creed,” (p. 227.) It is quite evident from this passage that this exposition stands in direct opposition to the above-cited earlier expressions of Goldstucker’s from p. 18. It is also again to be observed that the word avate in “nirvano ’vate” Pan. 8. 2. 50., in the sense of “free from wind,” as a possessive adjective, is not the “natural interpretation,” but a perfectly arbitrary one, blundering against Panini’s usage of language, as well as against the sense in which the commentary understands it. The word avate rather stands in juxtaposition with the words asparce anapadane, avijigishayam in the sutra immediately preceding, and is to be understood, with them, as Karmadharaya. The sutra subsequently says, “Nirvana [it is the past perf. pass. of the root va] “out of the wind” or, “when there is no wind,” that is to say, nirvana is not from the wind which blows out, it is blown out, it has ceased to blow, but from the regular part p. p. nirvata. From other things on the contrary, things that are blown out are blown away ; for example, according to Patangali for the fire, a light, or as the Calc. Scholiast (how correctly ?) adds that nirvana is used for bhikshu. This last example is, from Panini’s frequent mention of bhikshu, directly such a one as to lead one to suppose that by his rule he had it quite particularly in his eye. But I add that this is a mere conjecture, which may probably be so, but can be of no value anywhere as a proof. I have therefore also in these Studies, IV. 89., where I treat of intimations found in Panini’s vocabulary concerning his time, only very briefly pointed them out in the note on the expression nirvana in VIII. 2. 50. If I had at all believed that that word must of necessity relate to bhikshu, or indeed that it could bear that signification which suits the ’nirvana of the Buddhistic faith,” I would throughout have laid quite another weight upon it. In truth, both words—and it is therefore that Goldstucker’s reciprocal exposition of nirvana has failed—have nothing to do with each other. The nirvana of the Buddhistic faith is by no means a neuter of the part perf. passive, which may have acquired an abstract signification, but it is wholly a noun substantive, as niryana, nirmana in the sense of “the blowing out” “the extinction.” It is so regularly formed that Panini had not the least occasion to make mention of it, while the irregularly formed past. perf. passive nirvana instead of nirvata, required altogether a special rule.“—Weber’s Indische Studien, p. 136 et seq. ↩

115. Vide Supra p. xxxiii. et seq. “The technology of the Buddhists” says Pandit Rajendralal Mittra, “is to a great extent borrowed from the literature of the Brahmans…………… Their metaphysical terms are exclusively Hindu, and the names of most of their divinities are taken from the Hindu Pantheon”—Lalita Vistara p. 3. ↩

116. See extract in proof of this, infra. ↩

117. See Turnour’s exposition of these frauds, in the Bengal Journal of the Royal Asiatic Society for September, 1837, with reference to the Nepaul “amplified [vaipulya] sutras,” says the learned Rajendralal Mittra, “they allude to individuals who lived long after the days of their alleged author, and claim a degree of elaboration and finish, which leave no doubt as to their having been compiled at a much later period.”—Lalita Vistara, p. 16. ↩

118. Professor Wilson’s Sanscrit Dictionary, pp. x: xi. ↩

119. Professor Max Muller himself has fairly stated the weight due to this authority. See pp 242. ↩

120. By ‘different countries’ says the Commentator, ’are meant Anga and Magadha. ↩

121. Assalayana was of the Bhagu (Bhrigu) family, and this, therefore, agrees with the Brahmanical account—that ‘the Bhrigus have their heads quite shaved.’—Grihya-Sangrah parisishta. ↩

122. The above enumeration of Brahmanic sciences may not be unimportant in the identification of the person, who is said to have been accomplished in them. The three Vedas are here unmistakably mentioned. They are, as we learn from the Ambatta Suttan, Irubbeda (Rig. Veda) Yajubbeda (Yajur), and Samaveda, made by Attaka and other religious Sages. The fourth Veda, which is here omitted, is elsewhere stated to be the Attabbana (Atharvana) Veda, ‘made in subsequent times by impious (wicked) Brahmans, introducing life-slaughter, and other irreligious ceremonies, such as sacrificial torments &c.’ The fifth veda is called Itihasa [puravutta pabando Bharatadhiko] ‘compositions of ancient times such as Bharata &c.’ It is also here stated that the three Vedas included ‘Nighandu, Ketubha, and the distinctions of Akkhara &c.’ These are doubtless the supplements to the Vedas, viz. the Vedangas, which are also expressly mentioned in the Buddhist annals. By Nighandu is meant, as Professor Roth says, ’a collection of difficult and obsolete words, which formed a basis for instruction in the mode of expounding the Veda.” Ketubha is explained in the Glossary to be [ketubhanti kiriya-kappa vikappo kavinan upakaraya Sattan] ‘a science which is an anxiliary to poetry.’ As a supplement, however, to the Veda, Ketubha can only be identified with the Sanskrit Nirukta, a science serviceable for the understanding of the Vedic hymns ; Tasmad Vedarthava’-bodhaya upayuktan Niruktam. ’Hence the Nirukta is serviceable for the understanding of the meaning of the ‘Vedas.’ ‘The distinction of Akkhara &c., may also be identified with the Brahmanical Siksha, which Sayana defines to be “the science of the pronunciation of letters, accents, &c.’ ↩

123. Padaka.—This word is not explained in the glossary. From its being however mentioned immediately before Veyyakarana, I am inclined to believe that it is a Pali expression for the Sanskrit Nairuktas, or ‘Etymologists,’ a large class ‘who made the verbal origin of all words the leading principle of all their researches,’ as opposed to another school also mentioned in the text viz. the Vaiyakaranas or ‘Analysers,’ who, according to Professor Max Muller, p. 164., ‘following the lead of Gargyo the etymologist, admitted the verbal origin of those words only for which an adequate grammatical analysis could be given.’ ↩

124. The system of Atheistical philosophy taught by Charvaka. ↩

125. A science which teaches of the temper or fortune of a person, and of the linements of his body. The Commentator on the text, says that there was a work on the subject, consisting of 16,000 heads of instruction. ↩

126. Muller, p. 239. ↩

127. See Appendix. ↩

128. British Asiatic Society’s Journal, vol. xvi, p. 190. ↩

129. See Colebrook’s Miscellaneous Essays ii. p. 178. Also American Oriental Journal iii. p. 314.; and Gotama Buddha’s account of them in Bengal Asiatic Society’s Journal, Vol. vii., p. 698. ↩

130. See my Lecture on Buddhism, p. 5, et seq. ↩

131. Turnour’s Mahavansa—Int. p. xxvii. ↩

132. Prof. Lassen’s Ind. Ant. 1, 489-491. ↩

133. Tadeva vanmayan vidyat—Lassen p. 33, I am indebted for the text to a Sinhalese copy. ↩

134. The Commentator says—‘pure Sanskrit, pure Prakrit, pure Apabhransa, and a mixture of these with the Paisachi &c.’— ↩

135. Professor Cowell states in the introduction to his Prakrit Prakasa, that this a ‘rare and ancient Prakrit poem’ the knowledge of which he derives ‘from Dr. Hofer’s interesting article in his Zeitschrift’—p. x. ↩

136. The Commentator understands by ‘such like,’ the languages of Magadhi (Pali) and Panchala (Zend.) ↩

137. ‘Usually’—that is ‘in practice,’ ‘by custom’ or ‘usage.’ ↩

138. Abhiri—a dialect of herdsmen, of people of the lowest tribe. ↩

139. By the Shastras are meant the works of the Brahmans—their scientific books. ↩

140. See Professor H. H. Wilson’s Preface to the Dasakumara Charita p. 4. ↩

141. Colebrook’s Essays ii. p. 61. ↩

142. This last fact accounts for the common acceptation of the term Prakrita, to which I shall hereafter refer. ↩

143. Professor Wilson, speaking of the Pali, says “It is, as Messrs. Burnouf and Lassen remark, still nearer to Sanskrit, and may have prevailed more to the North than Behar, or in the upper part of the Doab, and in the Punjab, being more analogous to the Sauraseni dialect, the dialect of Mathura and Delhi, although not differing from the dialect of Behar to such an extent as not to be intelligible to those to whom Sakya and his successors addressed themselves.”—Journal of the Royal Asiatic Society xii. p. 237[?], ↩

144. Mr. Colebrooke on the authority of Kullaka Bhatta on Menu ii. 19, identifiee Sauraseni as the language of Saurasena, another name for the country of Mathura. Avanti is Oujin, or as the Singalese call, Udeni ; and the language of which is supposed to be the Malava. Prachya is the language on the East of India. It is identified with the Gowdi or Bengali by the Commentator on the Sahityadarpana. Magadh. is the language of Magadh or Behar. It is at present a corruption of the ancient language of that country. The original is alone preserved in Ceylon and Burmah. Sakari, and Sabari have not been identified ; Candali is supposed to be the dialect of herdsmen of the lowest tribes ; and Dhakki is probably Dhakshinatya ; see next note. ↩

145. Ardha-Magadhi is a corrupt dialect of the Pali. It may be identified with the Magadhi of the Prakrit Grammarians. Bhalika is supposed by Mr. Colebrooke to b. the language of Balkh, in the Transoxana, a country famous for its horses and situated on the North of India—Dakshinatya is identified by the Commentator of the Sahitya darpana as the country of Viderbha, which is said to be the modern Berar. ↩

146. See my Sidath Sangara Intr. p. clxxxviii. ↩

147. See Introduction to T’udave’s Selalihini Sandsa. ↩

148. See Prakrit Prakasa by Cowell ; Introduction p. xi. ↩

149. Miscellaneous Essays, ii, p., 1. ↩

150. From the Sanskrit are usually enumerated fifty-six dialects as known in India the principal of which are the Pali, long since the dead and sacred tongue of the; Buddhists ; the Magadhi, a more recent form of Pali, and an ancient dialect of a great part of Behar, also a dead language ; various forms of Prakrit ; besides nine-tenths of Hindi. Bengali, Maharatti, Gujarati, and the rest of the fifty-six dialects”—Journal of Royal Asiatic Society, xvi p. 174. ↩

151. Several inscriptions, as obtained from the Topes excavated, or as forwarded by travellers from within the ancient limits of Bactria, were nearly deciphered, so that very little remained to perfect this discovery also, and to establish that the ancient Pali language, or something very closely resembling it, prevailed over all those countries”—Bengal Asiatic Society’s Journal vii. p. ix. ↩

152. It is a question well worthy the attention of the learned, whether or not every fact stated in respect of the Zend in the following extract is not equally applicable to the Pali ? Professor Max Muller says : “It is clear from his (Burnouf’s) works, and from Bopp’s valuable remarks in his comparative grammar, that Zend in its grammar and dictionary, is nearer to Sanskrit than any other Indo-European language. Many Zend words can be re-translated into Sanskrit simply by changing the Zend into their corresponding forms in Sanskrit…….. Where Sanskrit differs in words or grammatical pecnliarities from the Northern members of the Arian family, it freqnently coincides with Zend. The numerals are the same in all these languages np to 100. The name for thousand, however, (sahasra) is pecnliar to Sanskrit, and does not occur in any of the Indo-European dialects except in Zend, where it becomes hazanra…….. These facts are full of historical meaning ; and with regard to Zend and Sanskrit, they prove that these two languages continned together long after they were separated from the common Indo-Eupean stock”—Professor Muller’s Last Results &c. pp. 112. ↩

153. See extract from Vibhanga Atuwa, post. ↩

154. Introduction to the Pali version of the Daladavansa. ↩

155. See his Miscellaneous Essays, ii. p. 2. ↩

156. See his Prakrit Prakasa xvii. ↩

157. Preface to the Sakuntala. ↩

158. Hindu Theatre 1. p. lxiv. ↩

159. Dr. Muir’s Sanskrit Texts ii.[?] p. 60. ↩

160. Kalpa Sutra, p. 132. ↩

161. “It is in the period with which we are now concerned (says the same writer) viz., that between Vikramaditya and the later Gupta Kings, that the names Sanskrit for the classical language, and Prakrit for the forms of speech springing from it, must have arisen.”—Lassen’s Ind. Ant. ii. p. 1151. ↩

162. See Ceylon Branch Royal Asiatic Society’s Journal, vii. p. 240. ↩

163. Colonel Sykes, after alluding to certain doubts expressed by Professor Wilson as to how the descendants (the Prakrits) could have been so exquisitely refined as he found them in the Plays, if the parent (the Sanskrit) was comparatively rude, which he believes was the fact before the age of the classical language of the Hindu literature, says; “A simple solution of Professor Wilson’s doubts would be to consider the Sanskrit emanating from the Pali, [? I shall rather say—the one original Prakrit language which has assumed the form of Pali]—the perfect from the imperfect, the polished from the rude, and the expressive from the simple; at least such is the natural progress of languages with growing civilization.”—Journal Royal Asiatic Society, vi. p. 422. ↩

164. Subodhalankara on Rhetoric. ↩

165. See his preface to the Mritchakatika. ↩

166. Prakrit Prakasa p vii. ↩

167. Wilson’s Hindu Dramas, 1 pp. lxiii, iv. ↩

168. See Indische Altathumskunde ii., p. 1160. ↩

169. Article by Babu Rajendralal Mittra Esq., in the Bengal R. A. S. Journal for 1854, p. 604. ↩

170. l’Histoire du Buddhisme Indien, p. 105. ↩

171. ‘We have seen (p. 72) that the Pali has some grammatical forms which are older than those of the inscriptions; and vice versa.’—Dr, Muir, p. 137, ↩

172. Bengal Asiatic Society’s Journal, vol. vi, p. 567, ↩

173. L’ Histoire du Buddhisme Indien p. 105. ↩

174. Dr. Muir’s Sanskrit Texts, p. 65. ↩

175. Kalpa Suttra, p. 132. ↩

176. ‘The man stands.’ ↩

177. ‘Having borne went.’ ↩

178. ‘Having done went.’ ↩

179. Parivattesi sabbapi Sihalatthakatha tada, Sabbesan mulabasaya Magadhayaniruttiya.—Mahavansa, p. 253. ↩

180. Patisambhidaya—the four supernatural attainments peculiar to the highest order of Arahanta, including inspired knowledge. ↩

181. Damila (or Tamil) is the Pali form of Dramida, or Dravida. ↩

182. Andhaka is the Pali form of Andhra, the Sanskrit name for the Talugu—see Caldwell’s Dravidian Comp. Gram. p. 5. ↩

183. See specimen in my Sidat Sangara, p. xxxvi. ↩

184. See my Notes at the end of the First Chap. on Verbs. ↩

185. Also my notes at the end of the Chap. ii. ↩

186. Professor Wilson’s Hindu Plays, 1. p. lxv. ↩

187. ‘I feel bound to concede that, by its greater simplicity of construction and superior facility of enunciation, the Prakrit may easily bear away the palm from its rival as a simple, yet polished and harmonious vehicle of human thought, admirably fitted to be the spoken tongue of a great and refined nation.’—Dr. Stevenson’s Kalpa Suttra, p. 137. ↩

188. Panini iii., 1. 35. ff. ↩

189. Bopp’s Comp. Grammar, ii., p. 841. ↩

190. Essai Sur le Pali, p. 138. ↩

191. “From the facts detailed in the preceding paragraphs (says Dr. Muir in his Sanskrit Texts ii. p. 274.) which prove that compound roots have been taken by the Indian grammarians for simple ones, and that old forms have been modified or lost in the modern, or even in the Vedic-Sanskrit, it is clear that that language (especially in its modern form) cannot always be regarded as a fixed standard, according to which the originality of the Latin and Greek [I would also add the Pali] forms could be estimated.” ↩

192. “In general it appears that in warm regions languages, when they have once burst the old grammatical chain, hasten to their downfall with a far more rapid step than under our milder European Sun.”—Bopp’s Comp. Gram. p. 711. ↩

193. “Closely related dialects are known to develope and change at very different rates of progress.”—Pr. Whitney Am. Oriental Journal Vol. v. p. 352. ↩

194. Also see Bopp’s Comp. Grammar, p. 701. ↩

195. Bopp’s Comp. Grammar, pp. 109, 10. The examples of the Pali in the above table, have been introduced by me. ↩

196. See my notes at the end of the Cap. iii. infra. ↩

197. ‘No province is assigned however to the principal Prakrit dialect’—Lassen’s Inst. Prak. § 3. ↩

198. Shadbhasha Chandrika. ↩

199. It is indeed remarkable that the Pali Grammarian, who has even borrowed technical terms from the Sanskrit, does not refer the Student for ‘the remainder’ to the Sanskrit, but to the Pali language, as the same is developed in the discourses of Gotama. See Cap. iv. § 36 a. ↩

200. Elphinstone’s India, vol. i., p. 72. ↩

201. Or, “Magadhi, by which name may be considered that dialect which is more ordinarily understood by Prakrit”—Wilson’s Hindu Plays, i., p. lxxii. ↩

202. Prinsep’s Indian Antiquities, ii. p. 43. ↩

203. ’The alphabet which we possess, as used by the Buddhists of a couple of centuries later, was that in which their sacred works had been written by the contemporaries of Buddha himself, who died in 543. B. C.—Prinsep’s Indian Antiquities, ii. p. 39. ↩

204. This proves that the cursive departure from the square form should be dated after the Buddhist era; and that the latter was not, as supposed by some, confined to Inscriptions, from its being better suited for lapidary purposes. For, the Epistle of Briubisara was written with “pure vermilion”, a material, which, if ‘the rounding of angularities’ was known in his time, ‘presented no difficulties to any series of curves or complicated lines.’ ↩

205. ‘This primitive character may well have proved sufficient for all purposes of record, so long as the language it was called upon to embody remained as simple as that for expression of which we may suppose it to have been originally designed and adapted. On the introduction of the Sanskrit element, it was necessarily subjected to previously-needless combinations, and under this and other process, perhaps, lost some of the stiffness of outline, which it may, nevertheless, have retained together with its original literal simplicity among the vulgar, even in the presence of an improved style of writing, auited for more polished literature; as in the existing Orthography of Hindi, contrasted with the elaboration of Sanskrit alphabetial definitions.’ Prinsep’s Indian Antiquities by Edward Thomas, vol. ii. pp. 43, 4. ↩

206. Beng. Asiatic Society’s Journal, vi. p. 1043. ↩

207. See Journal of the Royal Asiatic Society, vi. p. 411. ↩

208. ‘Sir’—a term in the vocative, used amongst the Brahmans; and Vadi—‘speaker’; thence a name for the Brahman—‘venerable speaker.’ ↩

209. These examples are taken from Kachchayana, lib i. Section 3. ↩

210. The same form of sandhi frequently occurs in the Rig Veda; e. g. deva so apturah, p. 72. ↩

211. The bh is sometimes changed into h in Pali; as devebhi or devehi. ↩

212. The n is lost in the Pali. ↩

213. Zeitschrift fur die Kunde des Morgenlandes, iii., 80. ↩

214. Sama Veda, Introduction, p. liii., ff. ↩

215. American Oriental Journal, iii., pp. 296-7. ↩

216. Edinburgh Review No. 228, pp. 381-2. ↩

217. All these names are also given in the Abhidanapadipika. Some of them are the Rishis of the Rig-Veda.—See Max Muller p. 44 note. Names marked with an (a) may be identified with those given in Muller’s Sanskrit Literature, p. 42; and for those marked with (b) See Max Muller, p. 378. ↩

218. Lit. ‘beholding.’ ↩

219. See also Edinburgh Review, No. 288, p. 376. ↩

220. See Asiatic Journal, xvii. p. 763. ↩

221. Pritchard’s Physical History of Mankind, iv. 22. ↩

222. “I once conjectured” says Mr. Colebrook, “the Prakrit to have been formerly the colloquial dialect of the Sarasvata Brahmans [see his Essays Vol. ii p. 21]; but this conjecture has not been confirmed by further researches. I believe it to be the same language with the Pali of Ceylon.”—Miscellaneous Essays, ii. p. 213. ↩

223. 101 Reprinted in Indische Studien. ii. pp. 110, 111. ↩

224. Kalpa Sutra, pp. 132—3. ↩

225. “There are no data from which the original formation of any one language can be ascertained; and consequently all opinion on the subject must rest entirely on conjecture.”—Asiatic Journal, xxi., p. 653. ↩

226. “There is so close an affinity between the primary dramatic dialect, and the Pali, as to leave scarcely any doubt of their being originally identical.”—Professor Lassen’s Institutiones Prakriticæ. ↩

227. Prakrit Prakasa, Sec. xi. § 2. ↩

228. Lassen’s Int. Prakrit, § 3. ↩

229. “These two (Sauraseni and Maharashtri) dialects stand the nearest to the Pali, though it (the Pali) is decidedly older than they are.”—Lassen’s Indian Antiquities. ↩

230. See Note, ante p. lxxix. ↩

231. “The oldest seats of the Indians, of which we find any mention made, are to be placed in the Punjab”—Spiegel’s Avesta, i. p. 5. “The earliest seat of the Hindus within the confines of Hindustan was undoubtedly the Eastern Confines of Punjab”—Professor H. H. Wilson. ↩

232. See Professor Spiegel’s Discourse referred to in the last note. ↩

233. “We learn that Pali not only pervaded India, but Bactria and Persia; and tha this is no wild theory or hazardous speculation is attested by the very high authority of the Pali Scholar, Professor Lassen, of Bonn, * * * who says, “the legends. upon the Bactrian Coins are in Pali or Pracrit.”—Col. Sykes in the Journal of the Royal Asiatic Society, vi. p. 425. ↩

234. The Rev. F. Mason of Burmah says:—“I do not think that the Pali is quite understood yet. * * Pali is much more copious than the Savans are aware though not to be compared with the Sanskrit,”—Amer. Or. Journal, ii. p. 336. ↩

235. Bengal Asiatic Journal, vi., p. 686, ↩

236. Turnour’s Mahavansa. ↩

237. Professor Spiegel’s Kammavacha. ↩