The Prākrits. Before I leave the Aryan languages of India, I must give a brief summary of what Sir George Grierson says of the Prākrits, the spoken speeches which have always, implicitly or explicitly, been distinguished from the artificial and literary Sanskrit. The Primary Prākrits of the Midland and Outer Band (of which latter no record survives) were of the same type as the Latin known to us in literature. They were synthetic and inflected languages. These gradually decayed (or developed) into what Sir G. Grierson calls the Secondary Prākrits. These are still synthetic, but diphthongs and harsh combinations of consonants are avoided, "till in the latest developments we find a condition of almost absolute fluidity, each language becoming an emasculated collection of vowels hanging for support on an occasional consonant." These Secondary Prākrits lasted from the days of the Buddha (550 B.C.) to about 1000 A.D.
One at least of these Secondary Prākrits, Pāli, has obtained world-wide fame as the language of the Buddhist scriptures. Thus crystallised, it underwent the same fate as Sanskrit and became more or less what we call in Europe a "dead" language. In the Midland was a great and famous Prākrit called Sauraseni, after the Sanskrit name, Surasena, of the country round Mathura. In Bihār was Māgadhī; in Oudh and Baghelkhand was Ardha-māgadhī or "half Māgadhī"; south of these was Mahārāshtri, which is best known to students of the ancient Indian drama as the vehicle of the lyrics with which the plays are studded. Kings, sages, heroes and other noble characters speak Sanskrit. Inferior personages use Sauraseni.
The Secondary Prākrits themselves degenerated into what Indian grammarians call Apabhramsas, "corrupt" or "decayed" tongues, which were used for literary purposes and finally became the parents of the great Aryan languages of the present time.
For comparison with the preceding table of the Dravidian languages, I give below the census table of the Aryan languages as recorded in 1901:—
| Number of speakers | |||
| A. Language of the Midland. | |||
| Western Hindi | 40,714,925 | ||
| B. Intermediate languages. | |||
| a. More nearly related to the Midland language: | |||
| Rājasthānī | 10,917,712 | ||
| The Pahārī (or 'mountain') languages of the Himalaya | 3,124,981 | ||
| Gujarāti | 9,439,925 | ||
| Panjābi | 17,070,961 | ||
| b. More nearly related to the Outer languages: | |||
| Eastern Hindi | 22,136,358 | ||
| C. Outer languages. | |||
| a. North-western group: | |||
| Kāshmīrī | 1,007,957 | ||
| Kohistānī | 36 | ||
| Lahndā | 3,337,917 | ||
| Sindhī | 3,494,971 | ||
| b. Southern language: | |||
| Marāthī | 18,237,899 | ||
| c. Eastern group: | |||
| Bihārī | 34,579,844 | ||
| Oriyā | 9,687,429 | ||
| Bengali | 44,624,048 | ||
| Assamese | 1,350,846 | ||
Of all these modern languages, their idioms, their characters, their literature, I do not venture to give even a summarised account. Those who have any curiosity to learn more about them cannot do better than consult Sir George Grierson's work on The Languages of India, until it, in its turn, is superseded by the book he is now writing from the materials collected in his Linguistic Survey. But everyone who has read The Newcomes will want to know what Hindustāni is, especially as it is one of the languages prescribed for the study of probationers for the Indian Civil Service and is taught at the universities of Oxford, Cambridge, London, and Dublin. In the strictest sense Hindustāni is the dialect of western Hindi spoken between Meerut and Delhi. It was much cultivated, as a literary dialect, by both Hindus and Musalmāns. The latter wrote, and write it, in the Persian character, and have added a large number of Persian and Arabic words. In this Persianised form it is known as Urdū, "a name derived from the Urdū-e mu 'alla, or royal military bazaar outside the imperial palace at Delhi, where it is supposed to have had its origin." Under Muhammadan rule Urdū was almost as much the lingua franca of India as English has come to be in modern times.
Another point is worth noting here. The Aryan languages of northern India are, in a very real sense, Hindu languages. Perhaps I shall make myself clearer by asserting that the languages of Western Europe are Christian languages. For historical reasons, their religious phraseology has a Christian connotation and allusiveness. But in the west, the distinction between things secular and things religious has become so familiar that the Christian element in our speech is not recognisable in our ordinary talk. In Hindu India, on the other hand, almost every act of a man's life has some religious or superstitious significance, and hence all the Aryan languages in the mouths of Hindus are markedly different from the shape they assume when spoken by Musalmāns. In the case of western Hindi we have the recognised Muhammadan dialect of Urdū, but in other languages too there is a Muhammadan dialect or patois, even if it has no separate name. A curious exception, however, occurs in eastern Bengal, where the bulk of the population is Musalmān. In this region the Muhammadans are comparatively recent converts from the lower aboriginal or Mongoloid castes, whose Muhammadanism sits very lightly on their habits and consciences, and so far as my own experience goes, there is little difference between the speech of the lower Musalmāns and their friends and cousins the Chandāls and other indigenous castes.
The Indo-Chinese Languages.
Finally, I must say a few words about the Indo-Chinese and Mon-Khmer languages. I spent most of my official life among people speaking these languages, and find, somewhat shamefacedly, that Sir G. A. Grierson makes me responsible for sundry vocabularies compiled in my distant youth. Naturally, I feel a personal interest in the people of the north-eastern border, and am tempted to enlarge on their qualities of speech and character. But I have left myself little space, and the Mongoloid races of the frontier are hardly Indian in any proper sense of the word. Moreover, though their total number is not great, they speak many languages. The Census of 1901 recognises 119 such languages. The most important of them all is, of course, Burmese, which is spoken by about seven and a half millions of people. There are nearly 900,000 Karens in Burma, and about 750,000 Shans. The Meithei (now Manipuris) mentioned above are 272,997 in number. The Boro or Kachari people of the Assam valley, a most attractive and delightful race, number somewhat less than 250,000. The other languages of this type have mostly a much smaller number of speakers than these. But mention should be made of 250,000 Mons, Palungs and Was in Burma, and 177,827 Khāsis in Assam, since these constitute the only members of the Mon-Khmer family still found within the limits of British India.