Pramāṇa · Search · About · API

Age of the Nandas and Mauryas c12.s12: Language

Age of the Nandas and Mauryas c12.s12

Language
English

Language

c12.s12.1

By 600 B.C., a little before Buddha, the Aryan speech would appear to have spread from Gandhāra to Videha and Campā in Eastern India, and to have been the common language of Indian Aryandom of the times that embraced the Mahājanapadas. The hill and forest tracts of Central India just to the south of the tracts watered by the Ganges were unquestionably Austric and Dravidian in speech; also Bengal and Assam and Orissa; and within the Aryandom of the upper Gangetic area and the Punjab, particularly within the former, there were still large areas, or small pockets, of non-aryan speech which were fast becoming smaller and smaller. In the Jātakas for instance, we read of Caṇḍāla villages where they spoke the Caṇḍāla speech, and we are told of an incident in which a Caṇḍāla masquerading as a Brahmin was discovered when he unconsciously cried out giligili in the Caṇḍāla language when he put some very hot rice-milk within his mouth in a Brahman feast which he had joined.

c12.s12.2

For the linguistic situation in India during the Nanda and Mauryan periods, we have literary evidence only for the Nanda period, and both literary and epigraphical for the Mauryan period. The Brāhmaṇas, Āraṇyakas and Upanishads, the dates of which cannot be satisfactorily determined, cover at least 600 years, from 800 B.C. to 200 B.C.; and the Buddhist and Jaina canonical literatures also, in their substance, refer to the period immediately before the Nandas; and conditions during the Nanda age not being very different from those obtaining a few centuries earlier, the evidence from the Brāhmaṇas and other works mentioned above can be regarded as equally applicable to it. The Brahmanical Sūtras, Yāska, Pāṇini, Kātyāyana, Patañjali, Kauṭilya, Vātsyāyana, probably Bharata, and above all, the Mahābhārata and the Rāmāyaṇa—all these belong either wholly or partly (as in the case of the two epics) to the Nanda and Mauryan periods. On the epigraphic side, we have to take note of the earliest Brāhmī inscriptions which are few and very small, and Brāhmī legends on the early coins and seals, some of which may be pre-Maurya; and the inscriptions of Aśoka and his successors. Post-Maurya inscriptions for a couple of centuries after the extinction of the Maurya dynasty also have some value for these periods.

c12.s12.3

The Aryan speech during the Nanda and Maurya periods would appear to have been current, in its various local or dialectal forms, as the language of the land, from the Punjab to the eastern boundary of Bihar, which thus became the true home of the Aryan speech in India, the land of the great Aryan states: and from this area, where Brahmanism was developing as the result of a synthesis of the worlds of the Aryan and the non-Aryan, the Aryan speech was spreading south, mainly along the west, through Rajputana and Malwa and Sindh; and it had already been established in Gujarat; and colonies of Aryan speakers would seem to have been formed in what is now northern Mahārāṣṭra, as far down as the Godāvarī. The belt of forest land in the east, corresponding to present day Eastern Madhya Pradesh and Chota Nagpur, harboured rather backward groups of Non-Aryan tribes, the ancestors of the present day Kol (Munda) peoples and of Dravidian tribes like the Gonds, the Oraons and the Maler, and effectively checked, though only for a time, the penetration of the Aryan language into these tracts. Although the conquest of Kalinga (corresponding to Orissa of the present day) by Aśoka in the third century B.C. opened up this area to the Aryan language, it was some time before the latter could establish itself in Eastern India—in Bengal proper, and then (by a double current from Bengal and from Kosala or Eastern U. P. through Mahākosala or Eastern Madhya Pradesh) into Orissa. The main line of North Indian Aryan linguistic expansion into South India thus from the second half of the first millennium B.C. has always been along the west—from the Midland through Rajputana and Malwa; and subsequently North Indian Muslim expansion into the Deccan carrying with it the Hindi speech followed the same route, in both pre-Mogul and Mogul times.

c12.s12.4

From the Brāhmaṇas we find that probably a century or two before Buddha, North Indian Aryandom included the following ten states—Gandhāra, Kekaya, Madra, Uśīnara, Matsya, Kuru, Pañcāla, Kāśī, Kosala and Videha. These included practically the entire Aryan-speaking world of say 700 B.C., and these states fell under three groups—Udīcya or Northern (including Gandhāra or the northern part of the present N. W. Frontier province and probably also the continuous parts of Eastern Afghanistan, Kekaya or North-Western Punjab east of the Gandhāra country including part of the Sind Sagar Doab, and the Jeep and Rechna Doabs and Madra in two groups—the Uttara or Northern Madras probably in Kashmir and the Dakṣiṇa or Southern Madras in North Central Punjab consisting of part of the Rechna and Bari Doabs), Madhyadeśa or the Midland (comprising Uśīnara in the west and north, corresponding to N. E. Punjab and N. W. Uttar Pradesh, Matsya or N. E. Rajputana, and Kuru and Pañcāla or Western U. P.) and Prācya or the Eastern (Kosala or Oudh, Kāśī or Eastern U. P., and Videha or Northern Bihar). Other States within this Aryan tract also came to be established quickly enough, Sālva, connected with Matsya, and Magadha and Anga in Bihar to the South of the Ganges. It would appear that the division of Aryan India into these three tracts, Udīcya, Madhyadeśa and Prācya, had some basis in dialectal differences. They correspond roughly to the three-fold division of the Indus valley and the upper Gangetic valley which is still prevalent—Punjab, Pachāhā, and Pūrab, corresponding roughly to the Hindki or Lahnda or Western Punjabi and Eastern Punjabi tracts, to the Western Hindi area, and to the tract of the country in the east where Kosali or Eastern Hindi and the Bihari speeches are spoken. Northern, or North-Western, Midland or Central, and Eastern—these formed the three dialect areas of the Aryan tract by 500 B. C., and to these three we shall have probably to add a fourth—the Dākṣiṇātya or Southern, which at this ancient period was probably not much different from the dialect area of the Midland from where the Aryan language was spreading through Rajputana and Malwa into Gujarat and the trans-Vindhyan tracts.

c12.s12.5

About the speech of the Udīcya tract, the popular opinion in the Midland in the age of Brāhmaṇas has been thus expressed in the Kauṣītaki Brāhmaṇa (vii 6) : tasmād udīcyāṃ prajñā- tatarā vag udyate : udañcau eva yanti vācam śikṣitum ; yo vā tata āgacchati, tasya vā śuśrūṣante ‘therefore speech is uttered in the North with mere discernment ; they go to the northern lands to learn speech ; and people like to listen to him who comes from there.’ Thus the people of the other tracts acknowledged the Aryan language as spoken in the north-west as being the best and purest form of it. About Prācya or the east some stray and rather obscure references in the Brāhmaṇas would suggest that the Aryan speech was becoming altered or debased there : it was the land inhabited by the Vrātyas, who did not follow the Vedic religion—they were adikṣitas or uninitiated people, but spoke the language of those who were dikṣitas or followers of Vedic rites and customs, and at the same time declared words or sentences not articulated with difficulty as being articulated with difficulty (a-durukta-vākyam duruktam āhuḥ, a-dikṣitā dikṣita-vācam vadanti). This statement about the speech of the Vrātyas of the East would appear to suggest the presence already of Middle Indo-Aryan or Prākṛitic habits of speech, which found the characteristic consonant combinations of the old Aryan speech difficult, and brought in consonant assimilation and cerebralisation on a large scale. In the Brāhmaṇas there is no hint about a Dākṣiṇātya or southern land as being largely inhabited by Aryan speakers, and nothing about any dialectal or linguistic speciality there.

c12.s12.6

The assumption is quite allowable that by the time of the Buddha, the spoken Aryan language had deviated considerably from the Old Indo-Aryan norm as presented by the speech of the Rigveda, and had developed at least three distinct dialects—a Northern or North-Western, a Central, and an Eastern. This last was already fairly in the Middle Indo-Aryan or Prākṛit stage ; but the North-Western was conservative, and was regarded as the purest form of the Aryan speech, the well of Aryan undefiled—and it was exceedingly likely that in the Udicya as the nidus of the Aryan people in India there was the largest settled Aryan population, and consequently the language was better preserved ; the more Aryan speakers were penetrating further into the East, among masses of non-Aryans, the more they were getting to be smaller in numbers compared with the surrounding non-Aryans, and the Aryan language, in a non-Aryan environment that was growing stronger and stronger, was susceptible to a more rapid change and to change along new lines than the North-west.

c12.s12.7

Epigraphical records of the 4th-3rd centuries B.C. bear out the occurrence of the situation as deduced from literary references—only some new developments had in the meanwhile taken place. The oldest Brāhmī inscriptions including of course those of Aśoka give us a sufficiently clear idea as to the linguistic conditions for Aryan India; Aśoka's inscriptions, giving the same texts in as many as three distinct local dialects have been aptly described as the first ‘Linguistic Survey of India’. In the Aśokan documents, we have: (i) a Prākṛit or Aryan speech of the North-West, as in the edicts at Mansehra and Shahbazgarhi. This is based on the Udicya dialect of the earlier period, and its phonetics even in the 3rd century B.C. shows that it had deviated the least from the Old Indo-Aryan norm, and it thus bears out the encomium of the earlier author of the Brāhmaṇas that speech here was *prajñāta-tarā*, more discerning. Northern and North-Western Punjab thus showed a great deal of conservatism in its language, even as late as the 3rd century B.C. We may even say that it was still in the Old Aryan stage (at least in its phonetics, retaining as it did a good many conjunct consonants, and the three sibilants *ś, ṣ, s*), while the speech of the East had deviated most. (ii) There is a Prākṛit of the East, found in Eastern inscriptions of Aśoka and elsewhere. This form of Indo-Aryan speech has deviated exceedingly from Old Indo-Aryan norm, and besides it shows phonetic peculiarities (e.g. use of only *l* and no *r*) and some forms (e.g. *-e* from *-aḥ* rather than *-o* in the case of masculine nouns ending in *-a*) which are not found in the other dialect areas. It is exceedingly likely that this Prākṛit of the East was the language of Aśoka's court at Pāṭaliputra, and the edicts of Aśoka were first written at Pāṭaliputra in this dialect, and sent to the provinces for publication by being engraved on stone at prominent places. Where the local dialect differed so appreciably from this court speech as to make the latter not easy to follow locally, as for example in the North-West (Mansehra and Shahbazgarhi) and in the South-West (Girnār), the edicts were rendered in the local dialect; but this rendering was not very careful, but rather haphazard as a good many forms and expressions from the court dialect were allowed to find a place in the versions in the North-western and South-western dialects. Where the local dialect did not differ so much as to make the Prācya court dialect unintelligible, it would appear that the latter was employed, as much as in the home districts in Eastern India. Thus in Rajputana, in Western U. P., in North-western U. P. (Kalsi), and in Central U. P. (Allahabad), the Eastern dialect is employed as much as in Eastern U. P., Benares (Sarnath) and Bihar (Lauriya, Rummindei, Barabar caves). A few special characteristics may occasionally be noted, e.g., at Kalsi; the exact reason for this is not known. It would appear that the use of Eastern Prākrit, the Court dialect of Bihar and Benares, was like that of the use of Hindi (a form of Western Hindi of Western U. P.) in Eastern U. P. and Bihar. Generally it has been the language of the Midland that has prevailed in the East, but in the Aśoka inscriptions, owing to the political importance of Magadha as the home province of the empire, for the first and last time we have an Eastern speech established as the official language in the Midland.

c12.s12.8

In tracts far away from Arya-land, where Dravidian and also probably Kol (Munda) languages were spoken, the edicts were published in this Eastern official speech, e.g. at Dhauli and Jaugaḍa in the Kalinga country, which was both Dravidian (old Telugu, old Kannada) and Kol in speech, and at Siddhapur, Maski and Yerraguḍi where the language was equally Dravidian (Old Kannada).

c12.s12.9

This Eastern speech was unquestionably the same for the upper classes in Kosala, Kāśi, Videha and Magadha; it was the language of the Buddha, who called himself a Kshatriya of Kosala (Kosala Khaṭṭiya), and of Mahāvira also; it was the language of Aśoka and also of Chandragupta and the Nandas. The oldest Buddhist canon, as Sylvain Lévi and Heinrich Lüders amply demonstrated, was composed in this Eastern Prākrit, and not in Pāli. The Pāli canon appears not yet to have been known—at least, sufficiently known, in Magadha. Aśoka, when he quoted from the Buddhist texts, quotes from a version in this Eastern dialect, and not from the Pāli.

c12.s12.10

Epigraphical evidence from the 4th century B.C. shows that in Magadha had appeared a local form of this Eastern Prākṛit which deviated in two of its sounds from what may be called the standard Prācya or Eastern Prākṛit; it had palatal ś for the dental s of the latter (<ś, sh, s of old Indo-Aryan), and probably it had developed a palatalised ky from k after a palatal vowel. This specialised Magadhan form of Prācya was in all likelihood current among the masses of the people, of less exalted ranks, and the ś pronunciation was evidently looked upon as something vulgar and uneducated, judging from the fact than in later times in the Sanskrit drama the ś dialect was relegated to the least exalted characters.

c12.s12.11

(iii) The third dialect of Aśoka is that of the South-west as in Surāṣṭra or peninsular Gujarat (Girnār). This is well-established there, and if the Gujarat speech of the 3rd century B.C. is derived from that of the Midland, then in Aśoka's Girnār edict we can see a form of the Midland speech, slightly modified perhaps from the genuine Midland dialect of the Mathurā area, given a recognition far away from its own home district—for, as we have just seen, in the Midland proper the Eastern official language was used in inscriptions.

c12.s12.12

This roughly is the situation for the spoken dialects of Aryan India during the Nanda and Maurya periods. Aśoka employed them in his inscriptions as already the Eastern speech appears to have attained a certain literary position through the Buddhist and Jaina canons being redacted in it; and the use of the North-Western and South-Western speeches was just a concession to two distant and important dialects the speakers of which would find difficulty in following the Pāṭaliputra court speech. We know that the Greeks when they first came to India were confined to the area of the Udicya or North-Western dialect, a dialect which Aśoka's officials employed in the inscriptions at Mansehra and Shahbazgarhi. That this North-Western dialect retained some archaic or old Indo-Aryan characters is borne out not only by the evidence from the Brāhmaṇas and from the inscriptions of Aśoka, but also from the Greek transcriptions of Indian names heard from speakers of this dialect. Thus names like Sandrakottos, Sandraphagos, Prasioi, Eronnaboas, Brachmanes, Ottorakorrhas, Amitrokhatés or Amitrokhadés and Palibothra are Greek renderings of *Chandrakupta (a genuine North-Western form for Candragupta, with -k- for -g- characterising the Darada or Piśāca Prākṛit of the North-West), Candrabhāgā, Prācya, Hiraṇyavāha, Brāhmaṇa, Uttarakuru, Amitraghāta and *Pāllibutra for *Pad'liputra = Pāṭaliputra as heard in the north-west, where, as partially indicated by the Mansehra and Shahbazgarhi and later North-western inscriptions, groups like pr, tr, kr, br, dr, gr, did not assimilate the r.

c12.s12.13

The relationship between the Aśokan dialects and the later forms of Indo-Aryan can be tentatively indicated as follows :

c12.s12.14

1. The North-Western dialect—from this originated Hindi, Lahnda, or Western Punjabi, Eastern Punjabi (the latter strongly influenced by the Midlands peech) and Sindhi. This N. W. dialect was taken by Indian settlers to Chinese Turkistan where it was in use for some centuries as an official speech, in the southern part of the country.

c12.s12.15

2. The Midland dialect : Not represented in the Aśoka inscriptions, but the Girnār dialect is probably a form of Midland. From this originated the Western Hindi dialect (partially influenced by the North-Western dialect), and Rājasthānī-Gujarātī.

c12.s12.16

We do not know anything about an Aryan speech being current in the Deccan, but evidently from Mālava and Gujarat and Varad (h)ā-taṭa (Varhāḍ or Berar) Aryan dialects, mostly from the Śauraseni area, were spreading into Mahārāṣṭra.

c12.s12.17

3. The Eastern dialect. The standard form of it, at first current all over Eastern U. P. (Oudh etc.) and Bihar, differentiated into Eastern Prācya (Māgadhi) and Western Prācya (Ardha-Māgadhi). The latter came strongly under Midland influence, and became finally transformed into the Kosali or Eastern Hindi dialects (Awadhi, Bagheli, Chattisgarhi). The former, Māgadhi, spread into Bengal and Assam and Orissa, and it is the source of Bhojpuri, Magahi-Maithili, Bengali-Assamese and Oriya.

c12.s12.18

There is no evidence from the Nanda and Maurya documents about the spread of the Aryan tongue into the Himalayan regions. Probably the Dardic speaking Aryans (Khasas and other tribes) were penetrating into the Central Himalayan areas (the present Western Pahari and Eastern Pahari regions), and their Dardic Khasa speech was later overlaid by Indo-Aryan from the Midland.

c12.s12.19

As for literary Indo-Aryan of the Nanda and Maurya periods we have to reckon first with classical Sanskrit, which became fully established as the language of Brahmanism and Brahman organisation before the Nanda regime. It was at first confined to the Brahman schools, and as a language, in the 5th century B. C. when Pāṇini flourished in the Udicya region, it was sufficiently near to the spoken dialect of Pāṇini's home districts as to merit from him the name Laukika i.e. ‘popular or current’ as differentiated from the elder Vedic dialect which he called Chāndasa or Chandas i.e., ‘poetic speech’ which was tantamount to ‘archaic speech’. Classical Sanskrit was not the creation of the Udicya people only; just as the modern literary forms of the Hindi or Hindustani language of Delhi, viz. High Hindi and Urdu, are the creation not only of High Hindi and Urdu writers of Delhi and Agra and Meerut, but also of Lahore and Lucknow and Haidarabad and Mathurā and Allahabad and Benares. The śiṣṭas or learned men, i.e. Brahmans, in the Midland, in the Prācya, and also in the Dākṣiṇātya, also helped to build it up; and it came to be closely connected with the Midland as here the Brahman synthesis of Aryan and non-Aryan cultures into Hindu culture and religion seems to have started. Because of its archaic character and the clarity of its forms, it obtained the homage of the Buddhists and the Jains as well, just from the end of the Maurya period.

c12.s12.20

The Eastern Prākṛit became an important vehicle of religious culture when the message of Buddha and Mahāvīra was delivered in it in the 6th-5th centuries B.C. and although it was a very much advanced or degenerate speech from Old Indo-Aryan standards, it obtained some pre-eminence in the Nanda Maurya periods both as the language of the Buddhist and Jaina faiths and official language of the court or of the empire. But this pre-eminence seems to have died out with the passing away of the Mauryan empire.

c12.s12.21

Pāli as a literary language associated with the Theravāda school of Hinayāna Buddhism appears not to have been prominent during the Nanda and Maurya periods, if it had come into existence at all. Buddha gave a great charter to all the languages of mankind when he declared that people were to study his message in their own languages : and we may presume that as a result of this great pronouncement, translations into different dialects were encouraged. There is evidence that the teachings of Buddha were first written in the Eastern Prākṛit. But this form of Aryan speech, in spite of its being the official language of the empire was not a central dialect, but belonged to the easternmost extremity of the empire, and it was most deviated from the norm of the rest of Aryandom. As such it was not much intelligible to the rest of India. The Midland forming the real heart of Āryāvarta, had a dialect which could be understood by the Udicya people, as well as by the Prācyas and the Dākṣiṇātyas. It was the precursor of the Śaurasenī apabrahṁśa of late mediaeval times (c. 600-1200 A.D.) and of Brajbhākhā (c. 1500-1700) and the Khariboli Hindi or Hindustani of the present day. Buddha's discourses were rendered into the Midland dialect, as current in Mathurā (and extended from Mathurā into Ujjain and Malwa) : after the death of the master, some of his disciples who were from Mathurā had a hand in redacting the canon—at least one version of it ; just as they were rendered into the North-western Prākṛit, as fragments in this dialect recovered from Central Asia show. The same process was repeated in later times : Kabir (15th century A.D.) spoke and composed in the Bhojpuri dialect current in his native city of Benares, and yet his writings are found in a mixed variety of Western Hindi, Brajbhākhā and Khariboli of Delhi with plenty of Awadhi (Eastern Hindi) forms and a few Bhojpuri forms occurring as palimpsests. Mahendra, the son of Aśoka born and brought up in his mother's city of Ujjain, according to the Ceylonese tradition, took the Pāli canon to Ceylon. It is likely that Mahendra studied his Buddhist texts not in the Eastern version as his father evidently did, but in the Midland one (which was Pāli) as it was current in Ujjain.

c12.s12.22

Pāli does not at all agree with the Māgadhi and Ardha-Māgadhi dialects, later variants of the Prācya speech—it agrees rather with Śauraseni, which is the Midland speech as we find it in its later Prākṛit form. Pāli can only be looked upon linguistically as a literary form of the Midland speech as it was current in the centuries immediately before Christ. The Midland speech in its literary form thus was taken to Ceylon by Mahendra from Ujjain via Pāṭaliputra and Tāmralipti and from there brought back again into North India with the Theravāda doctrine by Buddhaghosha. In the meanwhile, it was coming into prominence from about the time of Christ, as an important and the best cultured form of a Middle Indo-Aryan speech, as Śauraseni Prākṛit, the earliest use of which we find in the drama fragment from Aśvaghosha discovered in Central Asia, and possibly also in Śūdraka's *Mṛicchakaṭikā*, a Prākṛit which Bharata noted some time during the early centuries after Christ and which Rājaśekhara lauded as the most elegant form of speech in the 8th century A.D.

c12.s12.23

The Aryan speech was taken out of India by Indian missionaries or military adventurers who went out of India during the time of the Nandas and Mauryas. In Sin-Kiang, the city of Khotan (*Ku-stana* in Sanskrit) was colonised by Indians from Taxila in the 3rd century B.C., and the Indian community in the Khotan territory became quite numerous and strong, and although they lost their separate existence among the surrounding Iranian and Tibeto-Burmese speaking peoples, the Indian North-Western Prākṛit taken by them continued to be used (in a form much influenced by the local languages) as an official language in all state documents. Indian troops were found in the armies of the Achaemenian emperors and in Xerxes' army there were Indian troops. At the battle of Gaugamela or Arbela where Alexander finally overthrew Darius, the last Achaemenian emperor, Indian troops fought very stubbornly. The Indians came in touch with the Greeks through the Persian empire, and this must have taken place by 500 B.C., when the old form of the word *Iones* (Ionians, the Greeks of Asia Minor who came to be best known in the East) viz., *Iawones* or *Iavones* came to India as *Yavana*. Indian elephants with Indian mahouts were in the army of Pyrrhus when he fought the Romans in the 3rd century B.C. and in the Carthaginian army led by Hasdrubal and Hannibal in Italy, Indian elephant drivers also came into prominence. Indian philosophers and learned men, and later Buddhist missionaries sent by Aśoka, found their way into Greece, and we have mention in Greek records of at least one Indian philosopher who came to Athens and with whom Socrates had a conversation (before 400 B.C.). There was a good deal of intellectual and cultural intermixture both through the Achaemenian empire and the Greek empire of Alexander and his successors, and the Indian languages (including the newly formed literary speech, classical Sanskrit) came to acquire a number of Iranian (Persian) and Greek words (e.g. mudrā, dīpi or lipi, nipista. ‘written’, asavāri, kshatrapa, kārshā in kārshā paṇa, tashta-tast, pusta etc., and dramma from Greek drakhmē, suranga from surinks or syrinx, samidā from semidalis, khalina, and astronomical terms which came in later). The languages of the West, Greek specially, similarly obtained a number of Indian words. Connection with China though commerce may have commenced from the 4th century B.C. by way of Assam and South-West China (Yun-nan), and probably before Christ a few Chinese words came to be admitted into Indian languages (e. g. the name China—China, kīchaka—‘a kind of bamboo’, musāra—‘a kind of precious stone’.) There were persons in India speaking Iranian dialects and Greek, and the Persian official language as in the cuneiform inscriptions appears to have exerted some influence on the style of the inscriptions of Aśoka. The presence of different languages side by side in India, Indo-Aryan, Dravidian, Austric and the foreign Persian and Greek, gave rise to what I have called ‘Translation compounds’ in Indo-Aryan, in which words of same or similar meaning from two languages are combined to give a single word in Indo-Aryan (e.g. Iranian Karṣa ‘a monetary unit’ and Indo-Aryan paṇa of non-Aryan Austric origin-‘computation on the basis of four’ gave kārshāpaṇa in Sanskrit, kahāpaṇa in Pāli, meaning ‘a coin’; Austric sāta, sāda>śāli ‘horse’ and non-Aryan of unknown origin*ghutra,*hotra whence we have ghoṭa ‘horse’ gave Sanskrit śāli-hōtra ‘horse’ etc.)

c12.s12.24

The rapprochement between the Indo-Aryan and the Dravidian and Austric speeches was going on intensely during this period when the fusion of the diverse elements of the population into a common Hindu people under the intellectual domination of the Brahmans was in full swing. The Aryan language was being transformed from its purer Indo-European character into something different under the impact on Non-Aryan, through a larger and larger number of non-Aryan speakers turning into speakers of Indo-Aryan. In Middle Indo-Aryan, the Old Indo-Aryan accent had changed from a free pitch to a fixed stress. Vowel length became dependent more on speech rhythm than on etymology, a tendency towards an open rather than closed pronunciation of syllables became established (this led to widespread assimilation of conjunct consonants 'ushering in the Middle Indo-Aryan stage e.g. dhar-ma, sah-ya, bhak-ta, of earlier Old Indo-Aryan became dha-rma, sa-hya, bha-kta, etc. which were soon assimilated to dhamma, sajjha, bhatta), and there was an increase in the cerebralisation of t, th, d, dh, n, to ṭ, ṭh, ḍ, ḍh, ṇ and of l to ḷ; also the voicing of the intervocal unvoiced stops and aspirates started by which loka became loga, aṭavī became aḍavi and aḷavī etc., and in morphology we note a tendency towards reducing all declensions and conjugations of Old Indo-Aryan to a single type, and the commencement of the use of post-positional help words after case inflexions of the noun; besides, the inflected tenses and moods of the verb were reduced, and there was a larger use of participial adjectives, present, past and future, to express the time idea of the finite verb; further, the conjunctive participle or gerund in the -tvā (-tvī) and -ya became an exceedingly popular form. The vocabulary changed its character : a great many Old Aryans words were dropped, and their places were taken either by new Indo-Aryan formations, or by borrowings from the non-Aryan languages which entered the Aryan language through the back-door (i.e. without scholars admitting that they were non-Aryan words) and by the score. The entire spirit of Indo-Aryan was during the second half of the first millennium B. C. being changed fundamentally, and Aryan was more and more approximating the spirit of Dravidian and Kol (Austric).

c12.s12.25

Probably a good deal of the masses, in the Northern Indian plains, particularly among the lower classes, were bilingual, but the fast disappearing non-Aryan was not getting anybody's sympathy. The situation is like what one still sees in certain parts of India like Chota Nagpur and Assam, where the non-Aryan speeches are being steadily pushed out by Aryan.

c12.s12.26

In the Deccan and South India, except possibly in the Western Deccan right down to the early course of Godāvari where Aryan settlements were taking place, the non-Aryan languages were reigning supreme. Vidarbha or Varad(h)ā-taṭa (present day Varhāḍ or Berar), and Aśmaka on the Godā-vari, were Aryan states in the Deccan prior to 400 B.C. The *Aitareya Brāhmaṇa*, anterior to Buddha, mentions the Andhras, the Śavaras, the Pulindas and the Mutibas as *dasyu* or non-Aryan (probably Dravidian) tribes in the Deccan (of these the Śavaras, probably also the Pulindas were Kol tribes), and North Indian Aryans before the Buddha do not seem to have known much about the Southern Dravidian states. Sindh, from the evidence of the *Baudhāyana Dharma-sūtra*, was, like Bengal, still outside the Aryan pale during the closing centuries before the Christian era; Sindh was probably still Dravidian, a language allied to Brahui being current in it : and the Greeks say that a tribe named Arabitai lived in south Sindh. But there cannot be any doubt that throughout the whole of southern and eastern Deccan and South India, the ancestors of the Telugus, the Kannaḍas and the Tamil-Malayalis were flourishing as independent states with their distinctive South Indian or Dravidian culture which is best represented in the ancient Tamil literature ascribed to the centuries immediately after Christ. But unfortunately no authentic specimen of a Dravidian language is available before the early centuries of the Christian era.

c12.s12.27

The Dravidian language-family is now confined to India: but if the original Dravidian speakers were a Mediterranean people, then Dravidian must be relegated to a common stock with the language of the ancient Aegean and Asia Minor peoples who were living in Greece and the Islands and Asia Minor before the Indo-European Hellenes came to Greece. I have suggested that a tribal name of this people was *Dr(a)mil-* or *Dr(a)miẓ*, which is found in one branch of them in the Island of Crete in the Hellenised spelling as *Termilai*, and in another branch of it in Lycia in Southern Asia minor as *Trimmili*. Various tribes with their different names naturally made up these Mediterranean invaders of India, and *Dramiẓ* was evidently one of them. This name was Aryanised as *Dramiḍa* or *Dramiḷa* and then as *Draviḍa*, certainly earlier than the Christian era. Round about the time of Christ, the name became *Damiz* in the mouth of the people who bore the name, and they had by this time been fully settled and had built their culture and founded their states in the extreme south of India. The ancient Sinhalese Aryan-speaking settlers from Gujarat and Sindh, heard this name and wrote it in Pāli and Sinhalese as *Damiḷa*; and Greek and Egyptian traders similarly heard the name as *Damir* and called the country the native name of which was obviously *Damirakam* *Damirikē* in Greek. Then certain wide-spread phonetic changes swept the language of the *Dramiẓ-Damiẓ* (and probably also of the Kannaḍiga) people, among which was the unvoicing of the voiced stops *g ḍ d b* to *k ṭ t p*; and in a few centuries after Christ, the language entered a phase which we find in the oldest Tamil texts now found (the Śaṅgam texts), and the name became *Tamiẓ* or *Tamil* which is the form still obtaining in the Tamil language.

c12.s12.28

Although both Dravidian and Kol influenced the evolution of the Aryan language in the North—this influencing was at its height during the second half of the first millennium B.C. during the Nanda-Maurya period—and although culturally and politically highly advanced Dravidian states were flourishing in South India, states which had relations with Aśoka Maurya, it is curious and almost inexplicable why any of the Dravidian speeches did not manifest a literary life during the period under review. The finished character of the old Tamil *poruḷ* or *artha* i.e. matter of poetry, and of old Tamil literary ideas and ideals (dividing, for instance, poetic subjects into two categories of *aham* and *puram*, roughly corresponding to love and war, subjective and objective), took some centuries to develop, and it can be reasonably expected that during the Nanda and Maurya periods, the cultivated South Indian languages, particularly old Tamil and old Kannaḍa were essaying their first footsteps in the direction of sophisticated or advanced poetic composition, as opposed to purely popular poetry about love and war which we find orally among all peoples in the earlier stages of their history.

c12.s12.29

The advancement of a language is not possible without writing, and the Aryan speech came to be written down in all likelihood at a time when it made it possible for the Veda books to be compiled ; and this may well have taken place during the 10th century B.C., which is the date of the Mahābhārata battle and of Vyāsa, according to F. E. Pargiter and Hemchandra Raychaudhuri. The discovery of the Mohen-jo-Daro and Harappa script, with characters seemingly the pictorial prototypes of the Brāhmī letters of the 4th-3rd centuries B.C., now enables us to discard the theory of the Phoenician origin of the ancient Indian alphabet. The Brāhmī script in its oldest form—the proto-Brāhmī of the 10th century B.C. showing but an intermediate stage between the Mohen-jo-Daro script of c 2500 B. C. and the finished Brāhmī of 300 B.C.—could not be, as is natural in a similar situation, the finished alphabet with a scientific and etymologically sound orthography which we see in the Maurya and post-Maurya times. It was at the best a makeshift script, acting rather as a mnemonical writing than a proper and a complete alphabet. The orthography of Brāhmī as used for Prākrit in the 3rd century B.C. is not yet complete ; thus, e.g. certain consonant combinations are cumbrously made, and double consonants are not indicated at all (vassa being written as vasa or as vāsa). It was still a rather stumbling medium for the Prākrit dialects; it was much more so for Sanskrit. The Kharoshṭhī script current in the Udicya country during the period 400 B.C.—A.D.400 is admittedly of Semitic origin, a legacy of the Syrian scribes in the employment of the Achaemenian sirkar in India ; and its existence in India was an isolated episode, as isolated as the phenomenon of Gandhāra art. The name would appear to be a folk-etymology in India (whether khara+oshṭha or ‘ass-lip’ character as advocated by Sten Konow, or Khara-ushṭra or ‘ass-and-camel country’ character, as sought to be established by Sylvain Lévi, it is not necessary to take sides) of a Semitic word for ‘writing’ which we find in Hebrew as Xarošeθ (Kharosheth). The discovery of an inscription of the 4th-3rd century B.C. in Aramaic (Syriac) at Taxila, which has been read by Herzfeld, giving the name of ‘our Lord Priyadarśi’ (mr’n prydrś) is a direct proof of the contact of India with Aramaic writing, and helps us to envisage the origin of Kharoshṭhī.

c12.s12.30

The Indian Brāhmī alphabet is in all likelihood a derivative of the Mohen-jo-Daro script. But it is strange that a knowledge of writing should have been obtained by the later Dravidians from the North Indian Aryan speakers round about the time of Christ, if the Mohen-jo-Daro people were really the ancestors of the Dravidians. It is likely that the Mohen-jo-Daro script of *c.* 2500 and later was a very complicated thing, and when the Aryan alphabet was evolved out of it as a simpler system of writing after the Mohen-jo-Daro civilization had become moribund through Aryan impact as well as internal decay, and the people probably scattered, the simpler alphabet associated with a new and a vigorous people of a composite origin such as the ancient Hindus, descendants of both Aryans and non-Aryans, won the day,—making the more complicated, syllabic script of early Mohen-jo-Daro obsolete; and this alphabet, as well as the Aryan language of which it was the vehicle, became a conquering force in the South—the old script being no longer current among a probably dispersed people—during the closing centuries of the first millennium B.C.

Linked texts (21)