Results Diver Hornle, further specialized the research by his Comparative Grammar of the Gaudian Languages (1880), with particular reference to the Hindí. The same scholar and Mr. George Grierson, of the Civil Service, have, during the present year (1885), jointly brought out the first part of a Comparative Dictionary of the Bihari Language, which will enable every European inquirer to study the structure and framework of a modern Aryan vernacular for himself. These and other cognate works have accumulated a mass of new evidence, which settles the relationship of the present Aryan vernaculars to the languages of ancient India. They prove that those vernaculars do not descend directly disclosed from Sanskrit. They indicate the existence of an Aryan speech by the vernaculars. older than Sanskrit, older, perhaps, than the Vedic hymns; from which the Sanskrit, the Prákrits or ancient spoken dialects of India, and the modern vernaculars were alike derived. Passing beyond the Vedic period, they show that ancient Aryan speech diverged into two channels. The one channel poured its stream into the ocean of Sanskrit, a language at once archaic and artificial,' elaborated by the Bráhmanical schools.1 The other channel branched out into the Prákrits or ancient spoken vernaculars. The artificial Sanskrit (Samskrita, i.e. the perfected language) attained its complete development in the grammar of Pánini (circ. 350 B.C.).2 The Prákrits (ie. naturally evolved dialects) found their earliest extant exposition in the Pánini and grammar of Vararuchi, about the 1st century B.C.3 But the Vararuchi. 4000 algebraic aphorisms of Pánini mark the climax of the labours of probably a long antecedent series of Sanskrit elaborators, while Vararuchi stands at the head of a long series of subsequent Prákrit grammarians. gence of Sanskrit and Prákrit. The spread The spread of the Aryans from Northern India is best marked by the southern advance of their languages. The three great routes of Prákrit speech to the southward were— down the Indus valley on the west; along the Ganges valley to the east; and through certain historical passes of the 1 Hornle and Grierson's Comparative Dictionary of the Bihárí Language, pp. 33 and 34. Secretariat Press, Calcutta, 1885. It should be remembered that Indian grammarians, when speaking of the Vedic language technically, do not call it Sanskrit, but Chhandas. They restrict the technical application of Sanskrit to the scholastic language of the Bráhmans, elaborated on the lines of the earlier Vedic. 2 Vide ante, pp. 100 et seq. 3 Hornle's Comparative Grammar of the Gaudian Languages, p. xviii. et seq., ed. 1880. ROUTES OF PRAKRIT SPEECH. 337 During the three lines of march. Vindhyas in the centre. Between 500 B.C. and 500 A.D., the Their Prákrits. By degrees the main Prákrits, or spoken Aryan dialects, Classificadifferentiated themselves into local vernaculars, each occupying tion of a more contracted area. A series of maps has been compiled showing the stages of this process between 500 B.C. and 1800 A.D. Various classifications have been framed, both of the modern vernaculars and of the ancient Prákrits. Vararuchi, Varathe earliest Prákrit grammarian extant, enumerates four classes in the 1st century B.C.,-Maháráshtrí, now Maráthí; 2 Saurasení, now the Braj of the North-Western Provinces; Magadhí, now Bihárí; and Paisáchí, loosely applied to outlying nonAryan dialects from Nepál to Cape Comorin. ruchi's four classes. Prákrits. Apart from the last-named Paisáchí, the literary Prákrits The two really divide themselves between two great linguistic areas. main Saurasení, with the so-called Maháráshtrí, occupied the upper part of the North-Western Provinces, and sent forth offshoots through the Vindhya passes as far south as Goa. Magadhi spread itself across the middle valley of the Ganges, with its brightest literary centre in Behar. These were the two parents of the most highly developed of the Aryan vernaculars of modern India. The Apabhramsa, or 'broken' dialects of the Indus region, may for the moment be left out of sight. The Prakrits, or spoken Aryan dialects of ancient India, Prakrits received their first literary impulse from Buddhism. As the developed Bráhmans elaborated Sanskrit into the written vehicle for their ists, 1 Prefixed to Hornle and Grierson's Comparative Dictionary of the Bihárí Language. See also the Language Map appended to Hornle's Comparative Grammar of the Gaudian Languages. 2 Mr. Beames thinks that there is as much of the Magadhi and Saurasení type in the modern Maráthí as there is of the Maháráshtrí Prákrit, Comparative Grammar of the Modern Aryan Languages, vol. i. p. 34, ed. 1872. He holds that Maráthí reproduces the name rather than the substance of Maháráshtrí. by Buddh VOL. VI. Y for their orthodox religion, so the teachers of the new faith appealed to the people by works in the popular tongues. The Buddhist missionaries to Ceylon, circ. 307 B.C., carried with them scriptures; the spoken Prákrit of the Gangetic kingdom of Magadha. This dialect of Northern Indian became Páli, literally the series or catena of holy scripture in Ceylon. While the early Buddhists thus raised the Eastern or Magadhí Prákrit of and by the Behar to a sacred language, the Jains made use of the MahárJains. áshtrí Prákrit of Western India for their religious treatises. In this way, the two most characteristic of the spoken Aryan dialects of ancient India obtained a literary fixity, during the centuries shortly before and after the commencement of our The Prakrits also era. The Prákrits also remained the speech of the people, and underwent those processes of development, decay, and reremained generation to which all spoken languages are subject. On spoken the one hand, therefore, we have the literary Magadhí and languages. Maháráshtrí Prákrits of the beginning of the Christian era, the former embalmed in the Buddhist scriptures of Ceylon, the latter in the Jain sacred books of Western India. On the other hand, we have the spoken representatives of these two ancient Prákrits in the modern vernaculars of Behar and of the Maráthá country.1 Evolution of modern vernacu Prakrits. The evolution of the modern vernaculars from the ancient Prákrits is involved in deep obscurity. The curtain falls on lars from the era of Prákrit speech within a few hundred years after the birth of Christ, and does not again draw up until the 10th century. When it rises, Prákrit dialects have receded from the stage, and their place has been taken by the modern vernaculars. During the dark interval, linguistic changes had taken place in the old Prákrits not less important than those which transformed Latin into Italian and Anglo-Saxon into English. Those changes are now being elucidated by the series of comparative grammars and dictionaries mentioned on pp. 335-36. It is only practicable here to state the most important of the results. Obscure interval, 400-1000 A.D. The old Prakrits were synthetical in structure. The 1 This statement leaves untouched the question how far Maráthí is the direct representative of Maháráshtrí, or how far it is derived from the Saurasení Prákrit. As already mentioned, both the Saurasení and Maháráshtri poured through the Vindhya passes into South-Western India, and combined to form the second of the two main Prákrits referred to in the classification on a previous page. PRAKRITS AND VERNACULARS. 339 Prákrits modern Aryan vernaculars of India are essentially analytical. The During the eight centuries while the curtain hangs down synthetic before the stage, the synthetic inflections of the Prakrits had worn out. The terminals of their nouns and verbs had given place to post-positions, and to the disjointed modern particles to indicate time, place, or relation. The function performed in the European languages by prepositions. for the nouns are discharged, as a rule, by post-positions in the modern Indian vernaculars. The process was spontaneous, become analytical and it represents the natural course of the human mind. 'The flower of synthesis,' to use the words at once eloquent lars. and accurate of Mr. Beames, 'budded and opened; and when full-blown began, like all other flowers, to fade. Its petals, that is its inflections, dropped off one by one; and in due course the fruit of analytical structure sprung up beneath it, and grew and ripened in its stead.'1 vernacu As regards their vocabularies, the Aryan vernaculars of Three elements modern India are made up of three elements. One class of in vernatheir words is named Tatsama, 'the same as ' the corresponding culars; words in Sanskrit. A second class is termed Tadbhava, ‘similar Sanskrit in nature or origin' to the corresponding words in Sanskrit. Prákrit The third class is called Desaja, or country-born.' This tadbhavas. classification is an ancient one of the Indian grammarians, and Nonit is so far artificial that it refers the modern vernaculars to Aryan tatsamas. Sanskrit standards; while we know that the modern vernaculars desajas. were derived not from the Sanskrit, but from the Prákrits. It suffices, however, for practical purposes. Prakrit The great body of modern Indian speech belongs to the Their second or Tadbhava class of words, and may be taken loosely frameto represent its inheritance from the old spoken dialects or work; Prákrits. But the vernaculars have enriched themselves for literary purposes by many terms imported directly from the Sanskrit; to represent religious, philosophical, or abstract ideas. and SansThese are the Tatsamas, 'the same as ' in Sanskrit. The dif- krit enrichferent vernaculars borrow such 'identical' words from Sanskrit in widely varying proportions. The strongest of the vernaculars, such as Hindi and Maráthí, trust most to their own Tadbhava or Prakrit element; while the more artificial of them, like the Bengalí and Uriyá, are most largely indebted to direct importations of Sanskrit words. The third element in modern vernacular speech is the Desaja, or 'country-born.' This represents the non-Aryan and 1 Mr. Beames' Comparative Grammar of the Modern Aryan Languages of India, vol. i. p. 45 (ed. 1872). ments. Non- element in the ver other words not derived either from the Sanskrit or the Prákrits. At one time it was supposed, indeed, that the modern vernaculars of India were simply made up of the Sanskrit of the naculars; Aryan settlers, modified by, and amalgamated with, the speech of the ruder non-Aryan races whom they subdued. Modern philology renders this theory no longer tenable. It has proved that Sanskrit played a comparatively unimportant function in the formation of those vernaculars. It also tends to show that the non-Aryan element is less influential than was supposed. Both in structure and in vocabulary the modern vernaculars formerly of India are the descendants neither of the written Sanskrit, supposed, nor of the aboriginal tongues, but of the Prakrits or spoken dialects of the ancient Aryans. less important than of nonAryan words; Proportion In regard to grammatical structure, this position is now firmly established. But the proportion of aboriginal or nonAryan words in the modern Indian vernaculars still remains undetermined. The non-Aryan scholars, with Brian Hodgson and Bishop Caldwell at their head, assign a considerable influence to the non-Aryan element in the modern vernaculars.1 Dr. Ernest Trumpp believes that nearly three-fourths of the in Sindhí, Sindhí words commencing with a cerebral are taken from some non-Aryan or Scythic language, which he would prefer to call Tátár. He thinks, indeed, that there is very strong proof to show that the cerebral letters themselves were borrowed, by the Prákrits and modern Indian vernaculars, from some idiom in Gangetic anterior to the introduction of the Aryan languages into India. Bishop Caldwell states that the non-Aryan element, even in the Northern Indian languages, has been estimated at onein Maráthí. tenth of the whole, and in the Maráthí at one-fifth.2 vernacu lars, The real still Such generalizations are not accepted by the most eminent. proportion students of the Indo-Aryan vernaculars. Mr. Beames strongly unknown. expresses his view that the speech of the conquering Aryans completely overmastered that of the aboriginal tribes. The early grammarians were wont to regard as Desaja, or nonAryan, all words for which they could not discover a Tatsama 1 See Mr. Brian Houghton Hodgson's Aborigines of India, Calcutta, 1849; and pp. 1-152 of vol. ii. of his Miscellaneous Essays (Trübner, 1880). Also the Rev. Dr. Stevenson's paper in the Journal of the Asiatic Society of Bombay. 2 Bishop Caldwell's Comparative Grammar of the Dravidian Languages, introd. p. 57 (ed. 1875). Lassen held that the aboriginal tribes not only introduced peculiar varieties into the Prákrit dialects,' but also ‘occasioned very great corruptions of sound and form in the Indo-Aryan languages' (Indische Alterthumskunde, ii. 1149). But the more recent investigations of Beames, Hornle, and Grierson render these dicta doubtful, |