(Excerpts from the paper read by Iravatham Mahadevan at the Second All-India Conference of Dravidian Linguists, Sri Venkateswara University, Tirupati, June 1972 – full article on Roja Muthiah Library site – http://rmrl.in/wp-content/uploads/2014/02/papers/4.pdf )

I. THE BACKGROUND
During the half-century which has elapsed since the discovery of the Harappan civilization, three developments have taken place which have greatly increased the probability that the civilization
was Dravidian.

2. In the first place, new discoveries have revealed the great extent and duration of the Harappan civilization and have led to a fundamental re-assessment of its character. The spade of the archaeologist has uncovered hundreds of Harappan sites over a vast urea from the eastern borders of Iran to the Gangetic Doab and from the foot-hills of the Himalayas to the estuary of
   the Tapti. In the Gujerat region , late Harappan, sub-Harappan and post-Harappan settlements occur in sequence and demonstrate the survival of the Harappan intluence well after the middle of the Second Millennium B.C. What was thought to be a localised culture of the ‘ Indus Valley’ (which, at least by implication, could have been created by relatively small bands of alien maritime people) has now turned out to be the largest Bronze Age civilization known to the ancient world. It has now become inconceivable that this great and populous civilization, with its continental spread and millennial duration, should have appeared suddenly or utterly perished without a trace. Ethnic continuity overlaid by a linguistic change wrought by the incoming Aryans seems to be the only possible answer to the question, ‘What happened to the Harappans?’
3. Secondly, recent advances in Dravidian studies have led to an increasing realization of the decisive influence of the Dravidian substratum over the evolution of the Indo-Aryan languages
   and Hindu social institutions. It is now well established that the Dravidians were present in North-west India when the Aryans entered the country, most probably sometime around the middle of the Second Millennium B.c. The survival of the Brahui, a Dravidian language, and the presence of words of Dravidian origin in the Rigveda, provide irrefutable evidence for this fact.

While the Aryans imposed their language and established a new social order, they themselves must have been in a small minority and rapidly lost their ethnic identity. So complete is the racial fusion, that the terms ‘Aryan’ and ‘Dravidian’ can now be used legitimately only in a linguistic context. While the Dravidian languages have disappeared over most of North India, their substratum influence on the Indo-Aryan languages is most clearly seen in the latter in phonological changes like the introduction of retroflex sounds, in morphological changes like the switch-over from inflexion to post-fixation, in lexical borrowings, and especially in the near-identical syntactical structures of the modern Indo-Aryan and Dravidian languages. Such changes could have been brought about only after an extended era of bi-lingualism particularly on the part of the subject people (Dravidian speakers, in the present instance) , as always happens in similar circumstances.

4. The racial assimilation of the Aryans by the Dravidian people led in due course to the re-emergence of Dravidian social institutions, although with a Sanskritic veneer. In the field of
   religion, the older Dravidian deities like the Mother Goddess and Dravidian modes of worship pushed the Vedic religion into the background even by the time of the evolution of the middle IndoAryan dialects. Our increasing knowledge of the extent of the non-Aryan and Dravidian substratum influence in North India, has made the earlier view of some historians that, upon the advent of the Aryans, the indigenous population retreated southwards and that those who remained back were made into slaves and serfs, much less probable. There were undoubtedly migrations and subjugation in the earliest phase; but the numerical and cultural superiority of the indigenous population make it unlikely that they were all driven away or relegated to the lowest rungs of the society. The transformation of the Hindu religion in the post-Vedic period would have been impossible but for the fact that the new classes of priests as well as of kings, nobles and merchant-princes who patronised the priests, had risen from the indigenous non-Aryan stock. This circumstance also makes it probable that at least a part of the historical tradition of ancient India recorded in the Epics and the Puranas or handed down as ballads and folk-lore may go back to the pre-Aryan epoch.
5. Thirdly, systematic studies of the Indus Script, using scientific methods of statistical-positional analysis have led the investigators to the conclusion that the Harappan language is tpyoJogically non-lndo-European and resembles the Dravidian languages closely. In particular, it appears that the Harappan language was mono-syllabic and of the suffixing type with a Dravidian-like word-order.

6. When we place these developments side by side and consider objectively (a) the extent and duration of the Harappan civilization, (b) the extent and character of Dravidian substratum
   influence in North India from the Vedic Age, and (c) the Dravidian-like typological features of the Indus Script, we cannot but be struck by the inevitability of their inter-connection. To hold otherwise would be to presume that the extensive Harappan civilization left no discernible traces and that the deep Dravidian substratum influence is totally unmatched by any
   material remains. Neither assumption seems to be reasonable in the light of our present knowledge of the linguistic and social pre-history of India. It has also been argued that the possibility of the Harappan language being typologically similar to, but not necessarily identical with Dravidian, cannot be ruled out. The evidence of the new developments summarised above makes this view altogether too cautious and even somewhat pedantic. Taking the totality of available evidence, the hypothesis of a Dravidian authorship of the Harappan civilization seems to offer the most promising line of investigation. Absolute certainty can however be reached only when an acceptable solution to the riddle of the Indus Script emerges in the fullness of time.

27. We may therefore conclude that the Bharatas in the Vedic and the Epic periods in North India, the Satavahanas in the Puranic and early historical periods in the Deccan and the Cheras (Poraiyar) in the early historical period in the Tamil country, represent different layers and streams of parallelisms, all ultimately going back to the ‘bearer’ concept of the
    Harappan-Dravidian substratum.