Pramāṇa · Search · About · API

The Religion and Philosophy of the Veda and Upanishads v2.c41.s7: G. The Dravidian Element in Indian Thought

The Religion and Philosophy of the Veda and Upanishads v2.c41.s7

Language
English

G. The Dravidian Element in Indian Thought

v2.c41.s7.1

The desire to prove that Dravidian or pre-Dravidian peoples affected deeply Vedic thought and life is natural, and is supported by the generally accepted view that the population of India is predominantly Dravidian or at least of non-Indo-European origin. It is true, however, that in the absence of any real certainty of the physical characteristics of the Indo-Europeans this belief cannot be made the subject of strict proof. Assuming, however, the validity of the doctrine, there remains the question whether the invaders were essentially responsible for the Vedic religion and philosophy, imposing a superior culture on inferior races, or whether the process as regards culture was reversed; instances of analogical cases can be easily adduced.³ Here again, however, we must remember that we have no satisfactory evidence of the relative numbers of the invaders and the earlier settlers; we are left in this matter wholly to conjecture.

v2.c41.s7.2

³ See, e.g., Leaf, *Homer and History*, pp. 44 ff., following E. Meyer, *Gesch. des Alt.*² ii. 52.

v2.c41.s7.3

The best case that can be made out for the Indo-European character of Vedic religion or philosophy depends on comparison with the achievements of other Indo-European peoples, and, of course, in this case we are met with the problem to what extent these religions and philosophies were Indo-European in character, or on the contrary represent elements borrowed by the invaders from the countries they occupied. Thus, for instance, we have Sir A. Evans' suggestion⁴ that the Homeric epic embodies largely traditions of the pre-Hellenic Minoans, though the language is Indo-European, and, even if we dismiss the suggestion as not very plausible, we are not in a position to disprove it definitely by the evidence available. More important is the close similarity of many Indian and Iranian ideas, as developed by Julius von Negelein in his *Weltanschauung des indogermanischen Asiens*, but it may be noted even here that, as that work shows, many of the important ideas of Aryan belief can be paralleled among other races.

v2.c41.s7.4

⁴ JHS. xxxii. 281 ff., 293; on Minoan religion, in correction of Evans' views, see R. Dussaud, *Les civilisations préhelléniques*² (Paris, 1914), pp. 327–413.

v2.c41.s7.5

On the other hand, we must admit that when we come to definite attempts to prove Dravidian influence on Vedic religion or philosophy we are in the region of conjecture. The absence of any really early Dravidian evidence as to culture deprives us of any assured knowledge of pre-Indo-European conditions such as would enable us effectively to gauge Dravidian influences in Vedic religion or philosophy. This leads to the necessity of relying on conjectures, of which many may be easily shown to have no sound foundation, or at most to be mere possibilities. We may, of course, accept such possibilities if we like, but in doing so we cease to be judicial and arrive merely at subjective judgements which have no lasting value.

v2.c41.s7.6

It must be noted that the Dravidians are not allowed the credit of originality by such writers as Elliot Smith,¹ and Fleure, who insist instead on oversea and overland influences from Mesopotamia and Egypt as settling the course of Dravidian culture, though the evidence for this influence in the third millennium B.C. and much later is entirely speculative, depending on the existence of megaliths alleged to be connected with metallic deposits, and on the certainly unproved theory of the dissemination of this mode of building by a particular race, moved by peculiar religious views, who sought life-giving substances. The structures raised on this basis by Elliot Smith and Perry seems to me entirely unfounded and in every respect implausible, nor is there much probability of their receiving general endorsement by competent critics. It is, however, needless here to discuss their views, and Slater,² quite justly, lays more stress than his authorities on the independent power of the Dravidians to develop their culture. Moreover he rejects,³ very plausibly, the doctrine that the Dravidians represent the admixture of Mediterranean bearers of heliolithic culture with pre-Dravidians to form the Dravidian race, and his position treats the Dravidians as immigrants from the north-west who passed through Babylonia before the Sumerians were practising agriculture there.

v2.c41.s7.7

¹ *Migrations of Early Culture*; cf. Perry, *Children of the Sun*; Fleure in Slater, *The Dravidian Element in Indian Culture*, pp. 35 ff.; Peake, *The Bronze Age and the Celtic World*, chap. iv. But see CAH. i. 94 ff. On alleged Dravidian remains (Mohen-jo-Daro, Harappa), see IHQ. i. 176 ff.

v2.c41.s7.8

² *Op. cit.*, pp. 8, 22 ff.

v2.c41.s7.9

³ *Op. cit.*, p. 158.

v2.c41.s7.10

Slater⁴ contends that there is evidence to prove the superiority of the Dravidians in culture, ascribing to them the possession of castles, cities, wealth, luxury, the use of magic, superior architectural skill, and ability to restore the dead to life. He holds that they possessed a priest-magician clan or caste such as did not exist among the Aryans. Unfortunately he adduces no evidence for these assertions beyond the unscientific opinions of Mr. C. F. Oldham,⁵ and a couple of even less plausible conjectures, and it is sufficient to observe that there is nothing whatever to induce us to believe in the superiority of the culture of the Dravidians or the Muṇḍā-speaking tribes, and that on the contrary the evidence points rather in the opposite direction.

v2.c41.s7.11

⁴ *Op. cit.*, chap. ii.

v2.c41.s7.12

⁵ *The Sun and the Serpent*.

v2.c41.s7.13

It is further held⁶ that caste is Dravidian, the Brahmans being the product of the mingling of the bringers of the heliolithic culture from Egypt with the Dravidians. They achieved domination over the Dravidians, and, when the Aryans entered India, showing superior prowess, as users of horses, the Brahmans saw the wisdom of adopting the allegedly easier language which the masters spoke, and became the guardians and exponents of the Vedas, until finally they succeeded in imposing themselves as leaders of the Aryans who were Dravidized in culture. It is important to note that on this theory the Rigveda is left to the Aryans during the period when they were in the full flush of their conquests. It is difficult to reconcile this with the assertion that Viṣṇu is certainly not an Aryan god, since he most unmistakably exists in the Rigveda, which also knows Rudra, who is in part the prototype of Çiva, also asserted to be not Aryan.

v2.c41.s7.14

⁶ Slater, pp. 48 ff., 157 ff. But see Macdonell and Keith, *Vedic Index*, ii. 247 ff.

v2.c41.s7.15

The doctrine of the Egypto-Dravidian origin of caste is supported by the following arguments. (1) It is admitted that the Brahmans of the south claim Aryan descent and that they are differentiated from the bulk of the Dravidian population in facial appearance, complexion, and intellectual habit; but the justice of the claim to Aryan connexion is discounted on the score that it may be explained on the Egyptian hypothesis. This contention proves nothing, and leaves the traditional view unaffected. (2) It is alleged that the bearers of the heliolithic culture claimed divinity and established in Indonesia and elsewhere ruling classes claiming divine descent, while the traditional Brahman theory is that every Brahman is a god. The alleged facts are wholly problematical; the divinity of rulers when it exists can be traced to various ethnic causes, and the claims of the Brahmans are easily explicable without going beyond the Aryans.¹ (3) The carriers of the heliolithic culture are asserted to have been worshippers of the sun and the serpent; the Nambudiris, who are specially conservative, worship the cobras in Nayar households, and Brahman was a solar deity. The Brahmans of the Rigveda, however, do not worship cobras, and do not know a god Brahman. (4) The sacred cord of the Brahman is of cotton² which indicates an original association of the caste with cotton spinning, which certainly was no art of the Vedic Aryans. This, it is held, is a clear indication of the descent of Brahmans from foreigners who earned the gratitude and homage of the people of India by teaching them to spin and weave. The argument is clearly without value, and Slater himself admits that India itself was the home of the art of spinning and weaving cotton—primarily tree cotton; so that it is wholly needless to look to Egyptians of problematic reality. Moreover, it will be seen that the account of the origin of caste accepted by Slater is in effect merely a transfer to an Egyptian element of the tendency to stress racial distinctions, which together with occupation he believes to explain caste. A final complication introduced by Slater is his view that when they achieved supremacy the Brahmans became largely Aryanized in blood, because they aimed at securing the fairest wives.

v2.c41.s7.16

¹ Cf. Frazer, *The Magic Art and the Evolution of Kings*; Warde Fowler, *Roman Ideas of Deity*, chap. iv; von Negelein, *Weltanschauung des indogermanischen Asiens*, pp. 127 ff.

v2.c41.s7.17

² Cotton is unknown from early texts; cf. on the thread, BDS. i. 5. 5; GDS. i. 36; VDS. xii. 14; ApDS. ii. 4. 22; Manu, ii. 44.

v2.c41.s7.18

Finally may be noted the argument which ascribes special potency in magic to the Dravidians, including the power of raising the dead, on the strength of ‘the Vedas’.³ No such statement, of course, has any Vedic authority.⁴

v2.c41.s7.19

³ Slater, p. 117.

v2.c41.s7.20

⁴ A passage like TS. ii. 1. 1. 3 neither refers to revival as real nor to Dravidians.

v2.c41.s7.21

Nothing more substantial is adduced by Mr. G. W. Brown in his examination of the sources of Indian philosophy.⁵ He claims that animism is Dravidian, and that in it Jainism and Buddhism show closer adherence to Dravidian ideas, rejecting the new Aryan gods. The Aryans must have learned in India to worship new trees like the peepal and banyan, the peafowl, the serpent, the monkey, various species of grass, and sacred places innumerable, from their predecessors, to whom also is due the doctrine of transmigration with its corollary of release. But here again we have nothing of substantial importance; we have no right to hold that animistic or animatistic ideas were un-Aryan; Roman religion for instance is permeated by them, and Greek has relics of them.¹ Transmigration again is recorded as a Druid belief,² and was adopted in Greece, apparently from Thrace, and independently of India. The more concrete suggestion that the terms Sāṁkhya and Yoga, and even Upaniṣad are really derived from Dravidian names is clearly illegitimate in the absence of any difficulty of explaining satisfactorily the terms as Aryan contributions. For asceticism and caste as Dravidian no proof is even attempted.³

v2.c41.s7.22

⁵ *Studies in honor of Bloomfield*, pp. 75 ff.

v2.c41.s7.23

¹ See, e.g., Farnell, *Cults of the Greek States*, iii. 2 ff.; iv. 302 f.; v. 361 ff.

v2.c41.s7.24

² See also Dieterich, *Mutter Erde*, pp. 33, 56; von Negelein, *op. cit.*, pp. 54 f.

v2.c41.s7.25

³ On Sanskrit and Dravidian cf. J. Bloch, *Bull. Soc. Ling.* xxv. 1 ff.; J. Przyluski (*ibid.* pp. 66 f.; xxiv. 118 f., 255 f.) finds Austro-Asiatic words borrowed (including the term *karpāsa*, cotton), and strengthens the argument for borrowing of the phallic cult by the Aryans from the aborigines by the fact that this worship is prominent in Indo-China. It may be noted that here the paucity of phallic worship in the case of other Indo-European peoples strengthens the argument for borrowing. None of the instances alluded to by Farnell, *op. cit.* v. 8, is clearly or probably Indo-European.

v2.c41.s7.26

An effort to render probable pre-Dravidian influence on Vedic civilization and religion has recently been made by Professor S. Lévi,⁴ who revives the hint of Barth⁵ that the Atharvan *tābuva* may really be derived from the taboo of the Australians and Polynesians, and refers also to Mr. J. Hornell's conclusions,⁶ from the facts regarding Indian boats, of Polynesian influence, and even of a Polynesian invasion, the latter in Dravidian times, which introduced the coco-nut palm into India. Lévi's evidence is based on the existence of pairs—or triads—of ethnic names, applied to neighbouring peoples, which are differentiated by an initial preformant, a usage neither Indo-European nor Dravidian. Thus the Atharvaveda Pariçiṣṭa knows of Tosala as well as Kosala, a union found later in the Purāṇas and textbooks of rhetoric, while Açoka records Tosali, the name of which Lévi would even identify with Dhauli. Aṅga is found in the Atharvaveda, while with Vaṅga it appears from the epic onwards, usually in close association with Kaliṅga. For the latter again we have a parallel in the quite late⁷ name of the Telegu country Tiliṅga, Triliṅga, Tailiṅga, &c., whose antiquity, however, may be supported by Trilingon or Triglypton in Ptolemy's geography.⁸ Lévi compares also the Bhuliṅgas, known to Pliny and Ptolemy and the Gaṇapāṭha, who appear to have been a part of the Sālvas,⁹ though the epic ignores them. Both epics, however, have the combination Utkala with Mekala (with the variant Melaka). We have also Uḍradeça, Orissa, which goes back to an Uḍa to be recognized in Manu,¹⁰ for which Uṇḍa is a variant; comparable are the Puṇḍras, and even the Muṇḍas. The Pulindas are known from the Aitareya Brāhmaṇa onwards; the Bṛhatkathāçlokasaṁgraha¹¹ represents them in terms which suggest that they were small and black, like the pre-Dravidians,¹² and practised totemism. The epic mentions with them the Kulindas who may more properly have been Kalindas, as the name, Kālindī, of the Yamunā suggests, and another variant is Kuṇindas.¹³ The Muṇḍā languages permit the use of consonants as prefixes and the infixation of a nasal in certain cases between the prefix and the root, facts which would explain these variants, and which suggest that in the ultimate issue Utkala is Uḍra, differentiated by the infixation of a *k*, and Mekala Muṇḍa. There may be noted also the parallelism of Takkola as a place-name and the plant-name Kakkola, and of *kṣumā* (*kṣauma*) and *umā* (*aumīna*, *umya*) as the name of linen. The element *-liṅga* in some of these names suggests comparison with Laṅkā as denoting ‘island’, applied in both the Mahanadi and Godavari regions,¹ though P. Schmidt² has suggested for Kaliṅga comparison with the Nicobar word *kalāṅ*, sea-eagle, which would suggest that the ethnic names were of totemic origin. In the Indian archipelago names such as Karmaraṅga, Kāmaraṅga, Tāmraliṅga, suggest that in Kāmarūpa and Tāmralipti in its varied forms we have variants of a pre-Dravidian word. Kam again reappears probably in Kamboja, which is in its turn equated with Kapiça, both imperfect representations of an old form. Without a prefix we have Bhoja, the aspirate of which is due to contamination with the root *bhuj* of Sanskrit. The pairs of names suggest certain conclusions as to the social and political importance of a people who must have created political unities of considerable extent, so firmly connected with the real life of the country that they have persisted through thousands of years to the present day.

v2.c41.s7.27

⁴ *JA.* cciii. 1–56.

v2.c41.s7.28

⁵ *Œuvres*, ii. 254.

v2.c41.s7.29

⁶ *Memoirs of the Asiatic Society of Bengal*, vii. iii (1920).

v2.c41.s7.30

⁷ Mārkaṇḍeya Purāṇa, lviii. 28, &c.

v2.c41.s7.31

⁸ vii. 2. 23.

v2.c41.s7.32

⁹ Kāçikā Vṛtti, iv. 1. 173.

v2.c41.s7.33

¹⁰ x. 44.

v2.c41.s7.34

¹¹ viii. 31 ff.; cf. for colour, Nāṭyaçāstra, xxi. 89.

v2.c41.s7.35

¹² Thurston, *The Madras Presidency*, p. 124; above, p. 11.

v2.c41.s7.36

¹³ Varāhamihira, Bṛhatsaṁhitā, iv. 24; xiv. 30, 33.

v2.c41.s7.37

¹ Tibetan *gliṅ*; Pliny, *N.H.* vi. 18, has Modogalinga in the Ganges. Cf. the Laṅkā of the Rāmāyaṇa, which is not Ceylon (Jacobi, pp. 90 ff.).

v2.c41.s7.38

² *BEFEO.* vii. 261.

v2.c41.s7.39

This evidence, unfortunately, hardly carries us any farther towards ascertaining the amount of pre-Dravidian influence on Vedic civilization, and it must be added that much of it seems of very dubious value. The evidence that Tiliṅga was an early form is negligible; Ptolemy has already Triliṅga before him and he localizes the place in Arakan. Utkala may be Uḍra, but obviously, if we can operate with such names, we can establish anything we like; similarly with Kāmarūpa and Tāmralipti. The attempt to deduce from Yāska³ that he classed the Bhojas, of whom the Kambojas were a branch, as different from the Āryas appears hardly justifiable, *bhoja* in the passage cited having probably the sense ‘enjoying’, and not an ethnic significance, and the one thing certain about that people is that they were reckoned by Yāska as using a language closely akin to that of the Āryas. Clearly a very necessary proof of the thesis is still lacking, plausible etymologies from Austric languages, and these are clearly by no means easy to find, for Lévi rejects the one suggested, *prima facie* with plausibility, for Kaliṅga. If it were true that *tābuva* was taboo, it would seem natural to expect that even the modern languages which are our chief means of investigating Austric language construction should show in other words similar conservatism of form. But in point of form nothing very satisfactory can be made out of the comparisons already, very ingeniously, made by Przyluski.⁴ The conditions under which the investigation has to be conducted perhaps fatally preclude the adduction of really convincing cases, at any rate where Vedic words are concerned. We may readily believe in pre-Dravidian language and religion or even political organization as affecting the same phenomena among the Vedic peoples, but we still lack strict proof.

v2.c41.s7.40

³ *Nirukta*, ii. 2. Cf. E. Kuhn, *Avesta, Pahlavi, and Ancient Persian Studies* (1904), pp. 213 ff.; Nariman, *JRAS.* 1912, pp. 255 ff.; L. Sarup (trs., p. 22, n. 5) misunderstands Roth’s view on this point; Yāska himself uses *çu* as ‘go’ (iii. 18; iv. 13), which contradicts his assertion that the Kambojas used *çavati*, the Aryans *çavas* only. Roth does not deny the Aryan use of *çavas*, which Yāska doubtless connected (as *bala*) with *gatikarman*. Yāska is either inconsistent or there is an old interpolation (known to Patañjali, Mahābhāṣya, i. 1. 1).

v2.c41.s7.41

⁴ *MSL.* xxii. 205 ff.; *BSL.* xxiv. 118 ff.

v2.c41.s7.42

An effort to determine the characteristics of the pre-Dravidians has been made by Professor Giuffrida Ruggeri,⁵ who holds that India was successively occupied by Negritos apparently platyrrhine; pre-Dravidians (Australo-Veddahic); Dravidians (akin to Homo Indo-Africanus Aethiopicus); tall dolichocephalic (Mesopotamaic?) elements seen in the Todas; dolichocephalic Aryans; and brachycephalic Leukoderms—presumably of the Alpine type. The pre-Dravidians, on his view, were numerically preponderant, as held by Chanda, but the Dravidians imposed their language on large bodies of them, though the Muṇḍā languages persisted. The Dravidians were mesorrhine as opposed to the pre-Dravidian platyrrhines, the Niṣādas of Vedic texts, the ‘noseless’ enemies of the Rigveda. Similarly he assumes that the pre-Dravidians imposed their language on the Negritos, thus accounting for the fact that the speeches kindred to the Muṇḍā are spoken by tribes of Negrito as well as of pre-Dravidian type.¹ It must be admitted, however, that despite the ingenuity of this reconstruction evidence for it of any decisive character is wholly lacking, as it is based on recent anthropometrical data, whence conjectures are made for events of some four thousand years ago.²

v2.c41.s7.43

⁵ *Outlines of a Systematic Anthropology of Asia* (1921), pp. 43 ff.

v2.c41.s7.44

¹ W. Schmidt, *Die Mon-Khmer-Völker* (1906); *Die Gliederung der australischen Sprachen* (1919).

v2.c41.s7.45

² Cf. the divergent views of Myres, *CAH.* i. chap. i; Haddon, *The Races of Man* (new ed., 1924). It is natural to regard the Dravidians as ultimately not essentially different from the pre-Dravidians or even the Negritos, but all these views rest on insufficient grounds. Giuffrida-Ruggeri adopts the idea of the Aryan ‘Midland’, as excluding the Punjab and Rajputana with the United Provinces, but this view is not in harmony with any ancient authority on the Midland. Chanda (*The Indo-Aryan Races*, i. 74 ff.) regards the men of the Alpine race as Tocharians, a branch of the Indo-Europeans of *centum* speech, and ascribes to their influence the difference between ‘Outer’ and ‘Inner’ languages. But we have no proof of the early existence of these Tocharians in East Turkestan, and the ‘Outer’ languages show no *centum* features. His view of Yadus, Turvaças, &c., as immigrants—partially Semitized—from Mesopotamia, is quite implausible, as is his denial of the aboriginal character of the Çūdras; though that class may, of course, have included Aryan slaves, this is implausible on the evidence (Macdonell and Keith, *Vedic Index*, ii. 380 ff.). Cf. *JRAS.* 1917, pp. 167 ff.

Linked texts (15)