Pramāṇa · Search · About · API

The Origin and Development of the Bengali Language v1.c16: Introduction

The Origin and Development of the Bengali Language v1.c16

Language
English

Contents

Introduction

v1.c16.2

1. Bengali is a member of the Indic group of the Indo-Iranian or Aryan branch of the Indo-European family of languages. With its sister-speech Assamese, Bengali forms the easternmost language in the IE. linguistic area, just as the Celtic Irish and the Germanic Icelandic are the westernmost. It has been in existence as an independent and characterised language, or, rather, as a distinct dialect group, for nearly ten centuries.

v1.c16.3

2. Among the languages and dialects of India, Bengali is the speech of the largest number of people,¹ 48,367,915 persons having returned it as their mother-tongue during the census of 1911. Bengali is spoken by 92 per cent. of the population of the province of Bengal; and portions of Assam and of Bihar and Orissa linguistically form parts of Bengal. Bengali shades off into its sister-languages Oṛiyā, Magahī and Maithilī in the west, and into Assamese in the north-east. Apart from other Indo-Aryan speeches, notably Hindōstānī (which is spoken with varying degrees of purity), which are brought into the province by immigrants from the north-west and the west, Bengali has, within recent years, come in contact with Khas-kurā or Parbatiyā (the so-called Nēpālī) at Darjeeling in the north.

v1.c16.4

¹ Of course, a modified form of Western Hindī (Hindōstānī, Hindī or Urdū) is the *lingua franca* of all Aryan-speaking India, and is the established language of literature, of education, of the law-court and of public life in the Bihārī, Eastern Hindī, Panjābī and Lahndī, Central and Western Pahāṛī, and Rājasthānī tracts. From this, Hindī or Hindōstānī is often loosely regarded as the language of the people of all Aryan India excluding Sindh, Gujarat, the Mahratta country, Orissa, Bengal and Assam, Nepal, Kashmir, and to some extent the Panjab,—of a tract with a population of over 112 (including the Panjab, over 133) millions. Hindī or Hindōstānī is unquestionably the most important language of India, and the only speech which can be said to be truly national for all India; but it, together with other forms of Western Hindī, like Braj-bhākhā, Kanaujī, Bundēlī, etc., is the *mother-tongue*, the *home-language* of a little over 41½ millions only (according to the census of 1911). Taking into consideration the number of people speaking it as their mother-tongue, Bengali is the seventh language of the world, coming after Northern Chinese, English, Russian, German, Spanish, and Japanese; although as a great world-speech, Hindōstānī alone of Indian languages can rank with English, French, Spanish and Arabic.

v1.c16.5

On its borders, it meets with several aboriginal languages and dialects. Within the western boundary of Bengali is found Santali (Sāōtālī), a dialect of the Kōl¹ (Muṇḍā) group (of the Austro-Asiatic branch of the Austric family of speeches); and Hō and Muṇḍārī, also Kōl speeches closely related to Santali, are found to the west of Bengali. Besides, two Dravidian dialects, intimately connected with each other, are found to the west of Bengali: Malto, which is spoken in the Rāj-mahāl Hills, and Kurukh (Kūrukh) or Oraon (Orāō), which just touches Bengali at its extreme western fringe. In the north and the east, Bengali comes in touch with a number of speeches which are members of some six different groups of the Tibeto-Burman branch of the Tibeto-Chinese family. To the north, we have Lepca or Róng, a dialect of the Tibeto-Himalayan group; Dhīmāl, Limbu and Khambu, which are ‘pronominalised’ speeches belonging to the same group, and are spoken by small numbers in the extreme north; Dänjong-kä or Sikkimese, and Lho-ke or Bhotanese, which are closely related forms of Tibetan. To the north-east and east, Bengali meets dialects of the Boḍo group: Boḍo (Båṛå) or Kacārī (also known as Kōc, Mēc and Rābhā), Garo, and Dīmā-sā, as well as Mrung or Tipurā; it touches the area of the dialects of the Naga group; and dialects of the Kuki-Cin and Burma groups, like Meithei (or Maṇipurī) and Lušai, and Aracanese. Another aboriginal language, not related to the Tibeto-Burman dialects mentioned above, is spoken on the eastern frontier of Bengali, namely, Khasi, belonging to the Mōn-Khmēr group of the Austro-Asiatic languages, and thus connected with the Kōl speeches of West Bengal.

v1.c16.6

¹ I have a preference for the old, familiar and accurate ‘Kōl,’ used by the Aryan-speaking neighbours of the Kōl peoples, as an ethnic and linguistic term, to ‘Muṇḍā.’ ‘Kōl’ comes from a Middle Indo-Aryan ‘Kōlla,’ which is found in Sanskrit (also as ‘Kōla’); and obviously in Sanskrit it referred to the aboriginal people of Central India. (Cf. New Indo-Aryan ‘Bhīl’ from earlier ‘Bhilla,’ found in Sanskrit and Prakrit.) The word ‘Kōlla, Kōl’ itself seems to be of Kōl origin, and in all probability it is an early Aryan transcription of the source-form, in what may be called ‘Old Kōl,’ of the modern Kōl words for *man*—Santali [hɔṛ], Muṇḍārī [hɔṛɔ], Hō [ho:], Korwā [hoṛ], Kūrkū [ko:ro:]. There is no contempt implied in the use of the word among those who employ it, whatever the Sanskrit ‘kōla,’ never a popular word, might mean. A tract in Singbhum District in Chota Nagpur is known as Kolhan (<* Kōllānām) = *(the land) of the Kōls*: cf. Bhōṭān = *(the land) of the Bhōṭs or Tibetans*. The term ‘Kōl’ has an extensive employ, whereas ‘Muṇḍā’ is the designation of only one section of the Kōl people. Besides, it was used by Hodgson to denote this particular group of speeches, which, however, he classed as ‘Tamulian,’ *i.e.*, Dravidian. (A. Nottrott and P. Wagner, ‘A Grammar of the Kol Language,’ Ranchi, 1905, pp. 2, iii; LSI., IV, pp. 7, 8.) ‘Muṇḍā’ has not been able to drive out ‘Kolarian’ entirely; ‘Kōl’ seems to be least objectionable, and it is near enough the unmeaning but popular ‘Kolarian.’

v1.c16.7

Bengali, like other Aryan languages of India, has spread, and is still spreading, at the expense of the aboriginal tongues.

v1.c16.8

3. The living Indo-European languages can be arranged and classified under eight branches, which are as follows:

v1.c16.9

(I) The Indo-Iranian, or Aryan, falling into three ¹ groups:

v1.c16.10

¹ I accept Grierson’s division of Indo-Iranian into three groups, although this is not admitted by all. (Sten Konow, ‘Notes on the Classification of Bashgali,’ JRAS., 1911, p. 1, ff., where Dardic is relegated to the Iranian group; Jules Bloch thinks it possible that the source-dialects of present day Dardic, together with the Indian Prakritic speeches of the north-west [and the source-forms of the Gipsy dialects of Armenia and Europe?], formed a distinct Indian dialect-group; JA., 1912, i, p. 336. Another view is that the Dardic speeches are ‘ancient Āryan (Vedic) dialects which have been overlaid with Irānian as the result of later invasion’ in the time of the Achæmenids, or earlier: E. J. Rapson in the Cambridge History of India, 1922, p. 52.) See below, § 25.

v1.c16.11

(i) The Indic, Indian or Indo-Aryan group, under which come Vedic, classical Sanskrit, the old Prakrits of the early inscriptions, Pali, and the various Prakrits and Apabhraṁśas of old documentary remains and of extant literature; the modern Aryan languages (‘vernaculars’) of India; Eḷu, or Old Sinhalese, and modern Sinhalese; and the Gipsy speeches of Armenia, Syria and Turkey, and of Europe.

v1.c16.12

The inter-relation of the various Aryan languages, so far as it seems likely, is given in the Table under § 5.

v1.c16.13

(ii) Dardic, or Piśāca: embracing the languages of the extreme north-western frontier of India, and falling into these subgroups: (a) Kāfir—Bašgalī; Wai-alā; Wasī-veri or Prēsun; and Kalaša, Gawar-batī and Pašai; (b) Khō-wār, or Citrālī; and (c) Šiṇā—Šiṇā proper (7 dialects); Kōhistānī (3 dialects); and Kaśmīrī.

v1.c16.14

(iii) The Iranian group, which is represented by a number of speeches, ancient and modern, beginning with Avestic (from c. 600 B. C., the generally accepted period for Zaraθuštra, to whom are attributed the oldest portions of the Avesta, the Gāθās), and Old Persian of the Achæmenian cuneiform inscriptions (about B.C. 520—350) and spreading from the Black Sea region to Central Asia. The relationship among the various Iranian speeches seems to be as in the Table below (A. Meillet, ‘Grammaire du Vieux Perse,’ Paris, 1914, § 5; Sten Konow in ‘A. W. R. Hoernle’s ‘MS. Remains of Buddhist Literature found in Eastern Turkestan,’ Oxford, 1916, pp. 237-238; LSI., X).

v1.c16.15

[Figure: Indo-Iranian (Aryan) Iranian Dardic Indic Persic Group Avestic (= Old Bactrian? Old Medic?) Scythic Group East Iranian so-called ‘North Aryan’ Old Khotanese (? Kuṣāṇa or Śaka speech) ? Old Persian Middle Persian (Pahlavi) Chaldeo-Pahlavi ? Sogdian Ghalca (Pamir) Dialects (Wakhī, Sighnī, Sarīqōlī, Iškašamī, Munjānī, Yūdghā) Kurd Dialects New Persian with Tājikī Persian Dialects of the North and Centre Ormuṛī or Bargista Ossetic Yaghnōbī Paṣtō ? Balōcī]

v1.c16.16

The subgroups other than Persic are sometimes classed together as ‘Medic,’ in the sense of ‘non-Persic.’

v1.c16.17

(II) The Armenic Branch; (III) The Baltic-Slavic Branch; (IV) The Albanian Branch; (V) The Hellenic Branch; (VI) The Italic Branch; (VII) The Celtic Branch, and (VIII) The Germanic or Teutonic Branch.

v1.c16.18

4. Besides the above eight branches, members of which are living languages at the present day, there were a number of speeches both in Europe and in Asia, now extinct, which were members of the Indo-European family; *e.g.*, the languages of the Ligurians, the Iapygians and Messapians, and the Veneti of Italy; the languages of the Illyrians and Dacians, and of the Thracians; Phrygian, spoken in Ancient Asia Minor, a speech related to Thracian, and connected by some with the Armenian; the remains in the above are extremely scanty, and it is impossible to find out their proper relationship within the family. Within recent years have been discovered, from Chinese Turkistan, Buddhist and other documents, in Indian Brāhmī characters, of an Indo-European speech, spoken up to the end of the first millennium A.C. in the Tarim Valley, in the cities of Kucha (Dialect B) and Qara-shahr and Turfan (? Dialect A); and this speech, to which the names Kuchean and Tokharian have been given, and which has some unique points of interest, agreeing more with the Indo-European languages of the west (Celtic, Italic, and Slav and Armenian) than of the east (its neighbour speeches of the Aryan group), has been relegated to a branch by itself. (A. Meillet, ‘Le Tokharien,’ Indogermanisches Jahrbuch for 1913; Sylvain Lévi, ‘Le “Tokharien B,” Langue de Koutcha,’ JA., 1913, ii, pp. 311-380). The language of the Hittites of Asia Minor has been declared to be Indo-European by Friedrich Hrozny’ (‘Die Sprache der Hethiter,’ Leipzig, 1917) and J. S. Marstrander (‘Caractère indo-européen de la Langue hittite,’ Christiana, 1919); but although this has been disputed or regarded as not established with certitude by some, there can be no question about the Indo-European character of the language. (Cf. also ZDMG. for 1922: Johannes Friedrich, ‘Die hethitische Sprache,’ pp. 153-173; E. Forrer, ‘Die Inschriften und Sprachen des Hatti-Reiches,’ pp. 174-269; both these writers agree as to the Indo-European character of the speech, which is named ‘Hethitisch,’ or ‘Kanisisch’ by Forrer, as the language of an ‘Indo-European’ ruling class, in contradistinction to the non-IE. speech of the people of the Hittite kingdom, called ‘Chattisch’ or ‘Proto-hattisch.’) Also see below, § 25.

v1.c16.19

5. The inter-relation between the various languages and dialects of the Indo-Aryan group may be indicated by a Table as in the opposite page, following mainly Grierson in the ‘Linguistic Survey of India.’

v1.c16.20

This Table seeks to indicate merely the general lines of development of Aryan in India. There has been a very large amount of mutual influence among the various local speeches, and above all, the dominant influence of the literary languages, or *koinēs*, which has frequently overlaid the normal line of development of a local dialect, and in this way has obscured to a considerable extent its original character.

v1.c16.21

6. Of the various modern Indo-Aryan languages or groups of dialects, some have been quite important and dominant as literary languages or as languages of inter-provincial intercourse from very early times, while others have been nothing but forms of local *patois*. Short notes on their respective extent, importance and antiquity are given below. For maps, exact figures, and details, Grierson’s ‘Linguistic Survey of India’ is our great authority.

v1.c16.22

7. Beginning from the extreme north-west, we have the Dardic languages, which, although they do not belong to the Indo-Aryan group, may be noticed in this connection. These languages are spoken in the Kashmir valley, and in the region to the north and north-west of Kashmir, namely, Dardistan (Gilgit etc.), Chitral and Kafiristan, with the Hindu Kush as its north-western boundary. The Dardic languages, or, rather, ancient dialects of the same group, influenced Indo-Aryan dialects of the north-west and the west, of which Western Panjābī (Lahndī) and Sindhī are typical examples. (See §§ 11, 12, 25.) Excepting Kaśmīrī, the Dardic languages, although philologically important, have no exalted position. The number of people speaking them does not exceed 2 millions, of which Kaśmīrī alone takes up over 1 million. The Dardic speeches of the present day, barring Kaśmīrī, were never cultivated, and were written down only in the 19th century, so that no specimens of an earlier period are available. From the earliest times Kashmir was a [The provided image is blank.]

v1.c16.23

part of the Indo-Aryan world in culture and religion as well as in politics, whereas the other Dardic tracts, because of their inaccessibility, were never completely brought under Indian influence and organisation, so that the people retained more or less their primitive ways. Kaśmīrī consequently has always been exposed to the influence of the Indo-Aryan speeches, Sanskrit and the Prakrit vernaculars, from which the other Dardic speeches are free. There was literary cultivation of Old Kaśmīrī, and probably the lost ‘Bṛhat-kathā’ of Guṇāḍhya, written in the ‘Paiśācī’ language, was in Old Kaśmīrī, if it was in any Dardic speech at all. We may be quite certain that Kaśmīrī was cultivated before 1000 A.C. There is some Middle Kaśmīrī literature. The earliest specimens of Kaśmīrī that have been made accessible to us are the poems of Lallā, a mystic poetess of the Śaiva Tāntrika cult, of the 14th century (‘Lallā-vākyāni,’ ed. by Grierson and Barnett, RAS., London, 1920). Kaśmīrī used to be written formerly in the Śaradā character, allied to the Deva-nāgarī, but at the present day, since over 90% of its speakers are Moslems, the Perso-Arabic character is used, although a modified Deva-nāgarī has been devised and employed, for philological purposes mainly. Kaśmīrī was at first regarded as a Sanskritic or Indo-Aryan language, owing to the large Indo-Aryan element in it, but its Dardic affinities have been fully established (Grierson, ‘Linguistic Classification of Kashmīrī,’ IAnt., 1915, p. 270).

v1.c16.24

8. Western Panjābī or Lahndī, also known under various other names (Hindkō, Jaṭkī, Mūltānī, Cibhālī, Pōṭhwārī etc.), is a group of dialects current among nearly 5 millions of people in Western Panjab, who employ for literary purposes Urdū and to a slight extent Hindī and Eastern Panjābī. Western Panjābī does not possess much literature, except some Sikh prose narratives like the ‘Janam-sākhī,’ and some popular ballads and songs, of which the language is often mixed with eastern forms of speech. The native system of writing for Western Panjābī, with the ‘Laṇḍā’ characters, a variety of Śaradā, is in comparatively little use now, Persian characters being commonly employed when the language is put in writing.

v1.c16.25

9. Eastern Panjābī, or Panjābī, is the language of nearly 16 millions (according to the census of 1911). This speech is only one form of a Common Panjābī, extending from the west of the Western Hindī to the region of Paṣṭō; only it has from early times come under the influence of the Midland speech of the west Gangetic region. There are various dialects of Eastern Panjābī, a noteworthy form of it being Ḍōgrī, which is spoken in the state of Jammu and in the district of Kangra. There is a slight literary culture in it, the oldest extant specimens being a few Sikh hymns dating from the 16th century. Sikhs at the present day use E. Panjābī to some extent for literary purposes, employing the Gur-mukhī character, which is a reformed kind of Laṇḍā; but Hindōstānī (Urdū or Hindī) has always been the dominant language among E. Panjābī speakers. The Persian character is also sometimes used to write E. Panjābī.

v1.c16.26

10. Sindhī is the language of the lower Indus valley and of Kachh, and is spoken by a little over 3½ millions. It has 5 dialects—Vicōlī, Siraikī, Lāṛī, Tharēlī and Kacchī. Sindhī is written in an elaborated form of the Perso-Arabic alphabet, but the Laṇḍā character, proper to it, is in common use among merchants, and Gur-mukhī is occasionally employed. In its grammatical forms, Sindhī retains many archaic features, and in its phonetics it is remarkable in possessing four peculiar sounds not found in any other Indian language, Aryan, Dravidian, Kōl, or Tibeto-Chinese, *viz.*, [g’], [j’ or ɟ’], [d’] and [b’], which are only [g, ɟ, ḍ, b] pronounced with simultaneous closure of the glottis. In phonetics and morphology there are points of agreement between Sindhī and Panjābī (W. and E.). Sindhī has a little literature of ballads and prose tales, and of compositions in the Persian style.

v1.c16.27

11. The dialects of the Rājasthānī group (Mārwāṛī, Jaipurī, Mēwāṭī, Mālavī and others) are spoken by over 14 millions, and they seem to form, with Gujarātī, a separate branch of the Indo-Aryan family, which has as its basis the early IA. dialect (or dialects) current in Malwa and Gujarat, strongly modified by the neighbouring Śaurasēnī speech of the Midland (§ 13), and in times *post* 500 A.C., also to some extent influenced by the language of the Gurjara tribes (possibly Dardic in origin), who came from the north-west and settled in Rajputana and Gujarat and became the rulers there. The Western or Mārwāṛī form of Rājasthānī is in reality the immediate sister of Gujarātī, the eastern dialects agreeing more with Western Hindī. The exact affiliation of the Eastern Rājasthānī dialects (Mēwātī, Jaipurī and Hārauti etc., Mālavī and Nimāṛī), whether they are more intimately connected, in their origin, with Western Hindī or with W. Rājasthānī-Gujarātī, cannot be determined, in the absence of genuine ancient remains in them; but it has been found out that W. Rājasthānī and Gujarātī are derived from the one and same source-dialect, to which the name ‘Old Western Rājasthānī’ has been given (L.P. Tessitori, ‘Notes on the Grammar of Old Western Rājasthānī,’ IAnt., 1914-1916: Introduction). This OWR. is represented by a respectable literature, mostly by Jain authors, and this literature dates from the 14th century and earlier to the end of the 15th century. Gujarātī must have differentiated from OWR. in the 16th century into a separate language, which is now spoken by over 10 millions. The first great poet of Gujarat, Narasiṅh Mēhtā, belongs to the 15th century, but the language of his poems, which are very popular, has been modernised in the course of centuries. The dialects of Rājasthānī are not much used for literary purposes now. The language of the Midland has always been dominant in Rajputana. In the earlier days, the Śaurasēnī Prakrit and Śaurasēnī Apabhraṁśa had spread their influence over the original Indo-Aryan dialects spoken in Rajputana and Gujarat; and the literary dialect of the Old Western Hindī period, Avahaṭṭha or ‘Piṅgala,’ was as much cultivated by the bards of Rajputana as ‘Ḍiṅgala’ or the local Rājasthānī dialects, especially Mārwāṛī; and at the present day, Hindī is almost the sole literary language. There is, however, a rich literature in Rājasthānī, mostly in Mārwāṛī, and to some extent in Jaipurī, consisting mainly of bardic poems and chronicles. This literature has been explored, and its nature made known, by Tessitori, who edited, before his lamented and untimely death, some fine Mārwāṛī poems (Journal and Proceedings of the ASB., 1914, 1916-17-19-20; Bib. Ind., Rājasthānī Series).

v1.c16.28

12. The Pahāṛī or Khaśa dialects present a linguistic complication. According to Grierson, the original speech of the Khaśa tribes, who spread from Western Himalayas into the eastern montane tracts, was of Dardic origin, and like the Darūs, the Khaśas were Aryans outside the pale of Hindu society. Indo-Aryan speakers from the plains, mostly from Rajputana, migrated north into the Himalayas among the Khaśas, and Hinduised them, from the early centuries of the Christian era; and the Indo-Aryan dialects they brought completely killed off the original speech of the Khaśas, and became transformed into the present-day Pahāṛī dialects; which are thus forms of south-western (Rājasthānī) IA., carried to the Himalayas at a late period, and modified more or less by Dardic whose place they took, the traces of Dardic being stronger in the west. A nearly analogous case is that of Kaśmīrī, which is Dardic profoundly modified by Indo-Aryan (from the Panjab), only it did not give up its native character and become absorbed by IA.

v1.c16.29

Khas-kurā (called also Parbatiyā, Gōrkhālī or Nēpālī) is the most important Pahāṛī speech. The other Pahāṛī dialects are interesting philologically, but otherwise they have not much importance. Excluding Khas-kurā, for which exact figures are wanting, the Pahāṛī dialects are spoken by less than 2 millions. Khas-kurā seems to be ousting rapidly the Tibeto-Burman speeches of Nepal. It originally spread from Western Nepal, and its oldest remains do not go beyond the latter part of the 18th century. Maithilī seems to have been current in South-eastern Nepal before the advent of Khas-kurā; in any case, Awadhī, Maithilī and Bengalī were used as languages of culture in the court of the (Tibeto-Burman speaking) Nēwārī kings, who ruled before the Gurkhas, as is evidenced by a number of dramas written in the above languages in Nepal, right down to the middle of the 18th century (e.g., A. Conrady, ‘Hariścandra-nṛtyam,’ Leipzig, 1891; Nonī-Gōpāl Banerjee, ‘Nēpālē Bāngālā Nāṭak,’ VSPd., 1324 San, Introduction; C. Bendall, ‘Cat. of Buddhist Skt. MSS. in the Univ. Libr. of Cambridge,’ 1883, pp. 83-84, 183; ‘Kat. der Bib. der DMG.’, referred to by Bendall and Conrady). Kumāūnī and Gaṛhwālī, and other forms of Central Pahāṛī, and the various dialects grouped together as Western Pahāṛī, have no literature worth mentioning. Hindī is the established language of literature in the Central Pahāṛī region, 13. The Central Indo-Aryan language, Western Hindī, is spoken by over 41½ millions. Its chief forms are (i) Braj-bhākhā, the dialect spoken round about Bareilly, Aligarh, Agra, Mathura, Dholpur and Kerauli; (ii) Kanaujī, in the upper Doab, east of the Braj-bhākhā area; (iii) Bundēlī, in Bundelkhand and part of Central India; (iv) Bāngarū or Hariānī in South-eastern Panjab; and (v) the dialects to the north of Braj-bhākhā, from Ambala to Rampur, to which the name ‘Vernacular Hindōstānī’ has been given, as they approach nearest Hindōstānī as now written (Hindī or High Hindī, and Urdū). The oldest remains of W. Hindī extant are in the ‘Prithīrāja-Rāsau’ of Canda Bardāī (12th-13th centuries). The language there is much under the influence of literary Prakrit and Apabhrańśa of the earlier periods. The ‘Prākṛta-Paingala,’ which is a treatise on Apabhrańśa versification, compiled, in its present form, towards the end of the 14th century, gives in illustration of the metrical rules, a number of poems and couplets, most of which are in a dialect which is essentially Old Western Hindī. The Apabhrańśa verses quoted in the Prakrit Grammar of Hēma-candra (1088-1172 A.C.) are in a Śaurasēnī speech which is archaic for the time of Hēma-candra, and which represents the pre-modern stage of Western Hindī.

v1.c16.30

The dialect of Braj is the most important and in a sense the most faithful representative of the old Śaurasēnī speech, the source of the W. Hindī dialects, which was current in the Midland (Madhya-dēśa) of Aryan India, corresponding roughly to South-eastern Panjab, the western districts of the United Provinces (Rohilkhand, Agra and Meerut Divisions), and the tract immediately to the south. Braj-bhākhā has a rich literature, mainly in verse and partly in prose; and it and Awadhī, an Eastern Hindī dialect, formed the common literary vehicles for poetical composition in the Upper Ganges Valley (Hindostan proper), until recent times, when standard Hindōstānī came into being. Hindōstānī is in its origin based on the Western Hindī dialects spoken in and around Delhi, dialects which were strongly influenced by the contiguous Panjābī and Rājasthānī; and as the speech of the capital, it gradually came to be adopted by the Turkī, Persian and Paṣtō speaking nobility of the Moslem court. Originally a mixed *patois* of the bazaar and the camp («urdū» < Turkī «ordu» *camp*), it came to have a prestige as the language of the capital city and of the Moslem conquerors who settled down in India; and under the successive Moslem dynasties, officials and others from Delhi helped its spread into the provinces, east and west and south, as a convenient *lingua franca*, a natural successor to the Śaurasēnī Avahaṭṭha of the Rajput courts of Northern India immediately before the Musalman conquest. For some time, when it was in a fluid state, with its grammar not yet fixed, and its vocabulary mainly native Indian, it was only a spoken language, the Moslems of foreign origin who spoke it at home using Persian for literary and epistolary purposes, and the Hindus and converted Moslems employing the vernaculars, Braj-bhākhā and Awadhī. It was first used seriously for literary composition only towards the end of the 17th century, by North Indian Musalmans sojourning in the Deccan; although, it is said, it was taken up earlier for poetical composition in Delhi itself by Amīr Khusrau in the 14th century (Bāl-mukund Gupta, ‘Hindī-bhāṣā,’ Calcutta, Saṃvat 1964, pp. 9 ff). From the 18th century onwards, Hindōstānī became a serious rival of Braj-bhākhā and Awadhī as a literary language. Persian words were being adopted by all Indian dialects from after the Musalman conquest, and poems of Kabīr (15th century) and others in the dialects, with a large Persian vocabulary, have been found long before the establishment of Hindōstānī. Hindōstānī was already the home language of an influential Moslem nobility and gentry in Northern India, and as such, it had gradually taken up a large number of Persian words; it was already the accepted standard speech, «khaṛī-bōlī», for all Northern India; and by the beginning of the 19th century, it was well established as a literary language; and as most of its writers were Musalmans, it was written in the Persian character, and a large Persian element became its great characteristic. Hindu writers (beginning with Lallū-lāl and Sadaḷ Miśra) then took it up, at first at the suggestion of some English scholars in Calcutta, and used Hindōstānī with as few Persian words as possible, and wrote it in the national script, the Dēva-nāgarī. The resultant style of Hindōstānī, known as Hindī or High Hindī, has become established in all Upper Gangetic India, in the Panjab (among many Hindus), in Rajputana, in Central India, in the Central Provinces, as it admirably met the necessity for a convenient dialect for prose. It has been experimented for poetry in the old Hindu style as well, although Braj-bhākhā still holds its own. The Musalman style of Hindōstānī, known as Urdū, has developed greatly, and in versification and other matters has completely shaken off the Indian tradition and adopted those of Persian and Arabic.

v1.c16.31

Hindī and Urdū have their common form in the colloquial Hindōstānī: and this form of Western Hindī, as the dominant language of modern Aryan India, has exerted a tremendous influence on all the Aryan languages of the country; the Kōl languages have been touched and influenced by it, and the Dravidian languages of the south have not been free from the influence of this great northern tongue. The Upper Gangetic Valley has been the centre of culture and political life in India since the dawn of history, and it is in the nature of things that its language, successively as Sanskrit, as Pali, as Śaurasēnī Prakrit and Apabhrańśa, as Avahaṭṭha, as Braj-bhākhā and as Hindōstānī, should be a force throughout the history of India.

v1.c16.32

14. Eastern Hindī (called ‘Purabiyā’ by the speakers of W. Hindī) is the name given to a group of three dialects, Awadhī (called also Kōsalī and Baiswāṛī), Bagḥēlī and Chattis-gaṛhī, spoken by over 22½ millions in the United Provinces, Central India and the Central Provinces, to the east of the W. Hindī area. Bagḥēlī is almost identical with Awadhī, but Chattis-gaṛhī is different in some respects. Awadhī has a rich literature. The earliest great work in it is the ‘Padumāwatī’ of Mālik Muhammad Jaisī (middle of the 16th century), but its literary cultivation undoubtedly goes back several centuries earlier. Tulasī-dāsa wrote mostly in Awadhī. Bagḥēlī and Chattis-gaṛhī have a little literature, some of which has been printed. Hindī and Urdū, however, as successors to Braj-bhākhā, which was formerly much cultivated in the Eastern Hindī area, have been adopted by the speakers of Eastern Hindī as their literary language.

v1.c16.33

15. Marāṭhī is the language of nearly 20 millions in the Deccan, along the Bombay coast and in Berar, Haidarabad and the Central Provinces. Marāṭhī has 3 dialects—Dēśasth or Dēśī, the central and standard dialect spoken in the Deccan tract; Koṅkaṇī or coast dialect; and Varhāḍī-Nāgpurī or eastern dialect. The language round about Goa, called also Koṅkaṇī, is a sister-dialect of Marāṭhī, and has some peculiarities of its own. The oldest specimens of Marāṭhī are epigraphical, consisting of a number of short inscriptions, the earliest of which dates from 1118 A.C. (J. Bloch, ‘Langue Marathe,’ pp. 279 ff; Vināyak Lakṣmaṇ Bhāvē, ‘Mahārāṣṭra-Sārasvat²,’ Poona 1919, pp. 9 ff.). The earliest writers of Marāṭhī whose works are extant are Mukunda-rāja (end of the 12th century: cf. Bhāvē, op. cit., pp. 25-26); Jñāna-dēva, whose ‘Jñānēśvarī,’ a translation with commentary of the Bhagavad-gītā, was written c. 1290; and Nāma-dēva, a contemporary of Jñāna-dēva, a few of whose poems are preserved in the Sikh ‘Ādi-Granth.’

v1.c16.34

16. We now come to the Eastern or Magadhan group of speeches, of which Bengali is a member. The westernmost is Bhōjpuriyā, spoken by nearly 20½ millions. It is spread, roughly, from the east of the towns of Mirzapur, Jaunpur and Faizabad to the Son and the Gandak rivers, and makes a sort of a wedge in the south-east, south of the Magahī area. Magahī is the dialect of over 6½ millions, in Gaya, Patna, Munger and Hazaribagh districts, as well as of some settled communities of South Bihar people in the west of the Maldah district of Bengal. Maithili is spoken by over 10 millions in Bihar to the north of the Ganges, and in the districts of Munger, Bhagalpur and Santal Parganas to the south of the river. In the north, it shades off into Bengali in Eastern Purnia. Oṛiyā, the language of over 10 millions, is current in a corner of South-western Bengal, in Orissa, and in part of Chota Nagpur, the Central Provinces and Madras Presidency. Assamese is spread in the Assam valley, among a population of 1½ millions.

v1.c16.35

The oldest specimens of these languages, and their relationship with each other, are discussed later.

v1.c16.36

17. There are some curious dialects current in Bastar and the Central Provinces, which are a mixture of Chaṭṭis-gaṛhī, Oṛiyā and Marāṭhī. These are spoken by tribes who originally had Goṇḍi and other non-Aryan languages, and came simultaneously under the influence of three slightly different forms of Indo-Aryan. The most important of these dialects is Halabā or Halabī (LSI., VII, pp. 330 ff).

v1.c16.37

18. Sinhalese is another Indo-Aryan language which was cut off from its sisters and cousins from c. 5th century B. C., when, according to tradition, Indian emigrants under Prince Vijaya went to the island and settled there. Vijaya’s homeland was Lāḷa : it is mentioned in connection with Vaṅga (East Bengal) and Magadha; and Suppāraka and Bharukaccha (Sopara and Bharoch or Broach on the Bombay coast) are said to have been visited by Vijaya in course of his wanderings. Lāḷa is identified by some with Rāḍha or West Bengal, by others with Lāṭa or the Gujarat (and Sindh) coast country. The latter identification seems more likely: the first immigrants who carried the Indo-Aryan speech to Ceylon seem to have been from the Western India coast (W. Geiger, ‘Litteratur und Sprache der Singhalesen,’ in the Grundriss der Indo-Arischen Philologie und Altertumskunde, pp. 90, 91). Later, from the 3rd century B.C. downwards, Ceylon seems to have come in touch with Magadha, through Bengal, and traditions of intimate relations between Bengal and Ceylon are preserved in medieval Bengali literature. The Middle Indo-Aryan (‘Prakrit’) period is represented in Ceylon by numerous inscriptions, and there is evidence regarding the existence of a literature in what may be called a ‘Siṃhala Prakrit’ going as far back as the 3rd century B. C. But this literature is entirely lost. The oldest specimens of Sinhalese extant go back to the middle of the 10th century. The older form of Sinhalese is known as Eḷu (=Heḷu, < Hiaḷu < Sīhaḷu = Siṃhala), which is a sort of ‘Apabhraṃśa’ for Sinhalese. Sinhalese, by virtue of its position, has developed along its own lines, unlike the Indo-Aryan speeches of the mainland, which have always influenced each other very strongly, and have not allowed entirely independent progress of any one of these. After its development as a modern Indo-Aryan speech, however, it has been influenced by Pali and Sanskrit. Closely connected with Sinhalese is the language of the Maldive Islands, which is derived from Old Sinhalese of the 9th-10th centuries (W. Geiger, ‘Maldivische Studien, III,’ Sitzungsberichte der philos.-philol. und der histor. Classe der kgl. bayer. Akademie der Wissenschaften, 1902, Heft I, p. 114).

v1.c16.38

19. The Gipsy dialects of Western Asia (Armenia, Turkey, Syria) and of Europe fall into two classes, one European, and the other Armenian. The Gipsy speeches are derived from Prakrit dialects spoken in the north-west of India, and these dialects had some connection with the Dardic speeches. The ancestors of the Gipsies seem to have dispersed from India for the first time as early as the 5th century A.C.; and the earlier bands went to Europe by way of Persia, Armenia and the Byzantine empire, arriving in Eastern Europe in the 12th century, and thence spreading to Western and South-western Europe. A later band stopped in Armenia, where their language is more faithful to the Middle Indo-Aryan forms, but seems to be more impregnated with Armenian. (F. Miklosich, ‘Über die Mundarten und die Wanderungen der Zigeuner Europa’s,’ I-XII, Vienna, 1872-1880; F. N. Finck, ‘Lehrbuch des Dialects der deutschen Zigeuner,’ Marburg, 1903; *ibid.*, ‘Die Sprache der armenischen Zigeuner,’ Memoires de l’Academie imperiale des Sciences, VIII, Petrograd, 1907; Encyclopædia Britannica¹). The connection between the modern Indo-Aryan languages and the Gipsy dialects is, from point of view of origin, very close; but as these have developed entirely on their own lines, they are not usually considered in discussing the history of Modern Indo-Aryan; they throw valuable light on some points of Middle and New Indian phonology and morphology, however.

v1.c16.39

¹ Britannica

v1.c16.40

20. Taking into consideration the main phonetic and morphological trend of the Indo-Aryan speech as a whole, its history has been conveniently divided into 3 broad periods: (1) *Old Indo-Aryan* (OIA.), when the language was most copious in both its sounds and forms; (2) *Middle Indo-Aryan* (MIA.), when there was a movement towards simplification of older consonant groups, and a general curtailment of grammatical forms. The MIA. period may further be subdivided into an *Early*, a *Second* and a *Late* stage, with a *Transitional* stage between the Early and the Second; (3) *New Indo-Aryan* (NIA.), when the old simplifying tendencies inaugurating the second period had worked themselves out; the old inflectional system having been worn down to a few meagre forms, grammar had to be eked out with a number of new help-words, so that the whole character of the language became altered, and the modern IA. ‘vernaculars’ came into being. Vedic and Sanskrit form the typical or representative languages of the first period. For the second period, we have the various Prakrits of the earlier inscriptions beginning with those of Aśoka, Pali, and the Prakrits of literature. At the confluence of the second and the third periods we have the literary Apabhraṁśas; and these Apabhraṁśas of literature are mainly based on hypothetical spoken Apabhraṁśas, in which the earlier Prakrits die and the Bhāṣās or modern Indo-Aryan languages have their birth. The terms ‘Vedic’ or ‘Sanskrit,’ ‘Prakrit’ and ‘Bhāṣā’ may be used as short and convenient, though rather loose, terms for the three periods of Indo-Aryan; and the transitional stage between ‘Prakrit’ and ‘Bhāṣā,’ properly forming a part of the ‘Prakrit’ or MIA. period, can be conveniently called ‘Apabhraṁśa.’

v1.c16.41

Definite dates cannot be laid down in language history, but the period from the time of the composition of the Vedic hymns (? 1500 ? 1200 B.C.) to the times immediately preceding Gautama Buddha (557-477 B.C.) may be regarded as the OIA. period. The MIA. period may be said to have extended from 600 B.C. to about 1000 A.C.; of which 600 B.C. to 200 B.C. would be the Early or First MIA. stage, 200 B.C. to 200 A.C. the Transitional MIA. stage, 200 A.C. to 500 or 600 A.C. the Second MIA. stage, and 600 A.C. to 1000 A.C. the Third or Late MIA. stage. The first few centuries after 1000 A.C. would be an Old NIA. period, during which the NIA. languages enter into life.

v1.c16.42

The main characteristics of IA. during the three periods, affecting most of the dialects, may be briefly noted.

v1.c16.43

[1] OIA. Period : 1500 B.C.—600 B.C. (Vedic as type). Phonetic: ṛ, ḷ, ai, au; consonants retained in full; final consonants (surd stops, visarga, some nasals); consonant groups like kr, kl, kt, gd, tr, sm, hm, rt, rk, etc. in full force; ‘root-sense’ fully present, at least in the early stages; variable pitch accent. Morphological: Declension—complicated systems: root-nouns, and derived nouns ending in vowels and consonants; 3 genders, 3 numbers, 8 cases; special suffixes for masculine and neuter, and feminine, and a special pronominal declension. Conjugation—elaborate systems of tenses (present and imperfect, aorist, perfect and ‘pluperfect,’ future and conditional); moods (indicative, subjunctive, optative, imperative); participles (present, perfect and future); athematic and thematic roots, classified by Indian grammarians into 10 gaṇas; 2 voices—active and reflexive and some special forms for the passive (present tense 3 pers. singular aorist); causative, desiderative and intensive forms; passive participles; verbal nouns (infinitives) and gerunds, and indeclinable participles. Syntactical: wide use of finite verbal forms in the various past tenses, of the subjunctive mood, in the earlier period; position of particles not fixed; word-order free.

v1.c16.44

[2] MIA. Period: 600 B.C.—1000 A.C.

v1.c16.45

(i) Early stage: 600 B.C.—200 A.C. (Aśōkan Prakrit and Pali as types). Phonetic: ṛ, ḷ, lost; āi, āu, also aya, ava > ē, ō; simplification of consonant groups, by assimilation, etc. (kk, dd, tt, mh, etc.); final consonants and visarga dropped; tendency towards cerebralisation of dental stops and aspirates in connection with r, ś ṣ s reduced generally to one sibilant, s or ś; intervocal single stops retained. (In the North-west and the West, and possibly also in the Midland, the cerebralising tendency was resisted. In the early period, but it showed itself as a characteristic of the eastern dialects very early, even as early as the OIA.; the dialects of the North-west preserved a great deal of the phonetics of OIA.,—e.g. occurrence of r in consonant groups, retention of 3 sibilants ś ṣ s—right down to the end of the transitional MIA. period, but gradually through the influence of other dialects, yielded to the pan-Indian tendencies.) The old ‘root-sense’ grew obscured. A fixed stress accent, in the place of the earlier free pitch, seems generally to have been established. Morphological: considerable simplification of the declensional system: move towards standardising the -a declension; the dual lost; the dative merged into the genitive; the forms of the pronominal declension were extended to the noun. Conjugation: imperative and optative remain, but the subjunctive, found in a few rare cases, is lost; perfect system becomes less and less used, and is confined to a few verbs only; aorist and imperfect fall together, and are less and less used; middle voice, preserved as a relic in the earlier period, disappeared, of the derived forms, causative alone a living form; verbal nouns and gerunds become fewer, but more used; wider use of the passive participle for the past tense; fewer prepositional particles.

v1.c16.46

(ii) Transitional stage: 200 B.C.—200 A.C. (The Prakrits of the earlier inscriptions,—Kharōṣṭhī and Brāhmī—as types). Changes mainly phonetic: single intervocal unvoiced stops and aspirates became voiced, and fell together with original voiced stops and aspirates; these, except -ḍ(h)-, became open consonants, and came to be lost entirely in the next period.

v1.c16.47

(iii) Second MIA. stage: A.C. 200—600 A.C. (Dramatic Prakrits—Śaurasēnī, Mahārāṣṭrī and Māgadhī, and Jaina Ardha-māgadhī as types). Phonetic: the process of turning original single stops to zero completed during this period; original intervocal aspirates (except -ṭh-, -ṭh- > -ḍh-) became spirants aspirated, probably, and then weakened to -h-: (the above processes were carried on in all the dialects, but seem to have been completed first in the southern speeches, e g., Mahārāṣṭrī—while in some of the northern dialects—Śaurasēnī and Māgadhī—the voiced [and spirant] stage was maintained longer ; Ardha-māgadhī in this seems to have agreed with its neighbours): -p-, however, commonly found as -v-; intervocal -s- becomes -h- in some cases; sandhi between udvṛtta vowels. *Morphological*: Declension—further advance towards simplification, but masculine and feminine declensions preserved; beginning of the use of help-words in forming cases of nouns and pronouns—kēra < kārya, kaa < kṛta; (in the transitional stage, -santaka and one or two others can be noted). *Conjugation*: reduced to indicative present and future, imperative, optative present, perfect for a few stray verbs, and passive present; use of the passive participle to indicate the past becomes the rule; a number of verbal nouns and participial forms. *Syntactical*: Word-order becomes stereotyped; preference for the nominal rather than the verbal phrase.

v1.c16.48

(iv) Third MIA. stage (Apabhraṁśa): c. 600 A.C.—1000 A.C. (Type—Western or Śaurasēnī Apabhraṁśa). *Phonetic*: in the main, the state of things as in second MIA., but there was further decay: weakening of final vowels— -ā > -a, -ē, -ō > -i, -u; in many dialects, -s- -ss- of earlier periods became -h-; intervocal single -m- > nasalised -v- or -w-; nasalisation of vowels commences. *Morphological*: all declensions practically reduced to one; survivals or relics of the feminine and neuter declensions in some places; typical case forms—nominative, singular -u, -i(?), and plural -a; instrumental— -ēṁ, -hiṁ; ablative — -ahu; genitive, singular — -aha, -āha, -assu, -ā, plural -ṇa, -haṁ; locative— -i, -ahiṁ, -ahu, -asu; establishment of inflected help-words to emphasise case relations—kaṇa, kaṇṇa, kicca, maha, kaccha or kakkha, ṭhāma, sama, anta, antara, etc., generally added to the genitive or some other form; also of verb forms like (a)hanta, sanda, thakkia, dia, etc.; these help-words became the inflections and post-positions of the next period. *Conjugation*: indicative present and future, passive present; optative tending to be less and less used; imperative; the other moods and tenses entirely lost, the use of the passive participle for the past tense establishes the passive and neuter constructions; employ of fortifying affixes, pleonastic, like -ill- or -all-, -ḍ-; compound-verb constructions coming into use more prominently.

v1.c16.49

Use of rime in versification comes into being, wider use of jingles and onomatopoetic forms in ordinary language, noted already in the preceding stage. Loan-words (new tat-samas and semi-tat-samas) from Sanskrit; influence of both Sanskrit and the literary Prakrits of the earlier stage.

v1.c16.50

[3] NIA. Period: after 1000 A.C. Common characteristics in the oldest period: *Phonetic*: reduction of earlier double consonants to single ones, with compensatory lengthening of preceding vowel, except in the North-west and West; existence of inter-vocal euphonic -y- -w- between uncombined udvṛtta vowels, excepting -a- -a- in certain cases. *Morphological*: Declension—distinction of feminine; nominative and oblique bases—the latter an earlier oblique case-form; formation of plural by new methods (agglutination, use of the genitive, etc.); inherited inflections very few; no accusative-dative for inanimate names; use of post-positions; Apabhraṁśa (inherited) forms for the nouns of number (no new combinations). Conjugation: use of the present participle as a tense-form commences; beginning of compound tenses by combination; passive construction, with the passive participle qualifying the object, for the past tense of the transitive verb; adjectival (and neuter) construction for the intransitive verb past; use of inflected passive, present tense, in -i- or -ij-; establishment of analytical passive form with roots jā, paḍ, etc.; causatives in -ā(v)-; some verbal nouns; wide use of the conjunctive participle (indeclinable) in adverbial and other senses. Syntactical: compound verb constructions are established.

v1.c16.51

21. Each of the three stages of IA. forms a subject by itself. The first stage especially, as represented by Vedic and Sanskrit, has been studied with conspicuous success in Europe and in America, thanks to the ancient and medieval Indian grammarians on the one hand and to the historical and comparative methods of modern times on the other. With the help of Avestic, Greek, Latin, Old Irish, Gothic, Old Slavic, and other IE. languages, a great deal of the obscure points in the history of the origin and development of Vedic and Sanskrit has gradually been cleared up. The Aryan came to India, assuredly not as a single, uniform or standardised speech, but rather as a group, or groups, of dialects spoken by the various ‘Aryan’ tribes who entered into the land and settled down there. Only one of these dialects, or dialect groups, has mainly been represented in the language of the Vedas. But there were undoubtedly other dialects, not very different from Vedic, which might be expected to have continued and to have been ultimately transformed into one or the other of the various NIA. languages and dialects. The mutual relationship of these OIA. dialects, their individual traits and number as well as location, will perhaps never be settled; NIA. forms, remote as they are, do not help us here much, and from a study of the MIA. records, and of OIA. (Vedic and Sanskrit) itself, we can only suspect that there were marked dialectal differences in Indo-Aryan of 3000 years ago. (See § 32.) The true significance of the various Prakrits as preserved in literary and other records, their origin and inter-relation, and their true connection with the modern languages, forms one of the most baffling problems of Indo-Aryan linguistics, owing to a three-fold reason, that the Prakrits (and Apabhraṁśas) are literary and to a great extent artificial languages, standing to some extent off from the general current of development of MIA. as spoken; that our other authorities, the Prakrit grammarians, are not reliable guides, themselves influencing as much as being influenced by the texts and theories; and that there has been intermixture among the various dialects to an extent which has completely changed their original appearance, and which makes their affiliation to forms of MIA. as in our records at times rather problematical.

v1.c16.52

22. Common forms and inflections, as well as common habits of phonetics and syntax show that the dialects of Bengali, as well as Assamese and Oṛiyā on the one hand, and the dialects of the Bihārī group on the other, must have originated from some early form of IA. current in the eastern part of Northern India. To this mother-dialect, the name ‘Māgadhī’ has been given; and Māgadhī, with its immediate neighbour ‘Ardhamāgadhī,’ the source of E. Hindī, formed the ‘Prācya’ or eastern group of dialects in the late OIA. and MIA. periods. Some of the phonetic characteristics of Māgadhī, e.g., «ś» for «ś ṣ s», «l» for «r», noticed from a very early period, are preserved or can be traced in its descendants; and some time before its break-up into Bhōjpuriyā, Maithilī and Magahī, and Oṛiyā and Assamese-Bengali, which continued along their own lines since then, it developed certain morphological features and syntactical tendencies (e.g., use of «kaa» and «kēra» or «kara» with the genitive, of «-ill-», «-ell-» or «-all-» with the passive participle, of the verbal noun in «-ebba-» or «-abba-» for the future, of an active construction for the past tense), which were new in its history, and which have been inherited by its descendants. This late form of Māgadhī, when these peculiarities originated, can be described as ‘Māgadhī Apabhraṁśa.’ We have no remains in this immediate source-form of the modern Māgadhī languages preserved for us, but we can form some idea of it from a comparison of the oldest specimens of Bengali, Oṛiyā and Maithilī, and their sister-languages, as well as of Śaurasēnī and other Western Apabhraṁśa dialects of the same period, of which we have records. The Prācya speech, including Māgadhī, originally differed remarkably in its phonetics, and to some extent in its morphology, from the other forms of IA. which prevailed to its west (§ 39).

First 51 of 341 passages. Read on

Linked texts (29)