There was a recent paper in Nature published by Bahata Ansumali Mukhopadhyay, which has been picked up by many media outlets. In this paper, she has argued that there was a significant presence of Dravidian speakers in the Indus Valley civilization. She bases her argument on the following:
- The word for tooth piru/pilu is of (proto) Dravidian origin
- This word’s derived meanings for elephant & ivory was borrowed into Sumerian (language of the Mesopotamian civilization) & other West Asian languages
- The Mesopotamians were importing elephants & ivory from the Indus Valley Civilization (IVC)
- The derived words from piru/pilu was also borrowed by Sanskrit to refer to an elephant & the toothbrush tree, which was abundant in IVC
- Words for tooth are among what’s known as ultraconserved words, which means that these words don’t get borrowed from other languages & are conserved in their native tongues
- Conclusion 1: All of the above imply that there were Dravidian speakers in large numbers in IVC
- Conclusion 2: Referring to genetics papers, she points out that these proto-Dravidian speakers in IVC migrated south
It’s a good paper with a lot of data & analysis behind it. In this article, let’s examine these claims and see if the evidences support the conclusions.
Before we do so, the reader must take the time to read 2 other research works, which are rigorous & very pertinent to the current discussion. I will refer to these works repeatedly:
- The seminal work by Shrikant Talageri in 2017 titled The Elephant and the Proto-Indo-European Homeland: https://talageri.blogspot.com/2017/06/the-elephant-and-proto-indo-european.html. In this, he makes a solid case that, given the existence of cognates for the elephant in all Indo-European languages, India must have been the Indo-European-Homeland
- Shail Vyas’ pre-print titled Indus Musicians in Mesopotamia: https://osf.io/preprints/socarxiv/kce5x/ . He identifies 90 loan-words that are Sanskritic which have been absorbed into Sumerian, circa 2600 – 2100 BCE. These are words associated with the items that were exported by Harappans to Mesopotamia. The conclusion is that Sanskrit or its proto-version was the lingua franca of the IVC in the mid-late 3rd millennium BCE
Coming to the arguments by Bahata, points (1), (2) & (3) are quite acceptable, though there is some debate about whether the piru/pilu word is of Semitic origin. But Bahata makes a sound case for its Dravidian descent.
It is also true that the Mesopotamians imported a lot of goods from the IVC (like carnelian beads, timber, copper/bronze artifacts, musical instruments, jewellery made from lapus lazuli etc).
Talageri also argues quite thoroughly how the ivory & elephants imported by the Sumerians could have only come from the IVC (& not elsewhere, like say Egypt). Many of these points are made by Bahata as well.
Moreover, piru/pilu, as argued by Talageri, Bahata & others before them, are words borrowed not only into Sumerian, but also into Akkadian, Syrian, post-biblical Hebrew, old Persian & even Norse/Icelandic!
Now we come to point (4): yes, piru/pilu was introduced into Sanskrit too, as Panini states in his Dhātupāṭha (no. 521) and it derived from the root pīl-, “to obstruct” (read Talageri’s essay). Whether it’s a native Sanskrit word or borrowed can be debated. But considering the strength of evidence of its Dravidian origin & a lack of similar etymological depth in Sanskrit, I agree with Bahata that its a loan-word from Dravidian into Sanskrit. However, a couple of things need to be noted here:
- Sanskrit only borrowed the word pilu to refer to the elephant & the Salvadora persica (toothbrush tree), not to tooth
- The borrowing is late: its not in Vedic Sanskrit but in post-Vedic literature
- Even when the word pilu was borrowed, the ancient Rig Vedic word in Sanskrit for elephant hastin continued to be used & is in use even today
We will come back to these points. But moving to point (5), there is no disagreement with words for tooth being ultra-conserved in languages. Neither the Sumerians nor the Vedic-Sanskrit speaking Indians borrowed any word for tooth from elsewhere.
But from the above, is conclusion (6) justified: i.e. were there a substantial number of Dravidian speakers in IVC?
I’d argue that, while its likely that there were Dravidian speakers in the IVC, their presence is not likely to have been substantial. This is because, unlike a substantial number of (proto) Sanskrit words borrowed into Sumerian, only a couple of Dravidian words were borrowed. Moreover, the hydronyms (names of water bodies), toponyms (names of places) & phytonyms (names of plants) in the IVC region are overwhelmingly Indo-Aryan/Sanskritic, not Dravidian. So, for one word like Pilu, dozens of Indo-Aryan words for other trees can be found, with their own, deep etymologies. Hence, it’s a stretch to claim a “substantial Dravidian population” based on its presence.
Can conclusion (7) be justified: i.e. did proto-Dravidian speakers from IVC move to South India?
Though Bahata’s article starts off by saying that the IVC is multi-lingual, the paper veers into the narrative of “the IVC was populated by a Dravidian speaking population, which was driven south by the influx of Aryans from the Steppes”. Though she doesn’t say this in so many words, the genetics papers she quotes (by Narasimhan et al) certainly argue this line of thought. See the section in her paper titled “Proto-Dravidian possibly migrated from IVC to South-India: genetic evidences and linguistic missing links”.
However, there is nothing in her own data that leads to this conclusion. In order to show such a movement, she must be able to show linguistic/literary memories in Dravidian of such a North-to-South migration, substrate words in Dravidian that point to a prior North Indian presence etc. None of these can be found in Dravidian languages – they are firmly rooted in the South of India. Interestingly, the situation is the same in Sanskrit, i.e. there is no linguistic/literary memory of a non-Indian homeland nor of a migration from outside India.
We can rebut the genetic arguments of Narasimhan separately, but it will distract from this article. Point (7) is clearly a weak conclusion in her paper.
So, what is the overall picture?
Let’s split this into time-windows:
Circa 4000 BCE: The thrust of Talageri’s article is that there is a root-word for elephant in Proto-Indo-European (PIE) which is “rbha/lbha” & cognates for the word “elephant” are present in nearly all Indo-European languages. Combined with the fact that, during the time of the Indo-European language expansion, India was the only IE-speaking homeland which had elephants, the inevitable conclusion is that India was the homeland for Indo-European languages! Read his entire article to get an appreciation of the rigor in his argument. This line of reasoning falls under the category of linguistic palaeontology.
Now, this initial IE expansion is supposed to have happened starting ~4000 BCE and continued in waves. This timeframe is more than one millennia before West Asia was importing elephants & ivory from IVC. So, this implies that, during the first wave of Out of India migrations, the Proto-Indo-European (PIE) word for the elephant “rbha/lbha” was making its way across both Asia & Europe.
Circa 2600 – 1600 BCE: Now, we travel in time to circa 2600 – 1600 BCE, which starts from the Mature/Urban Harappan period. During this time, there were substantial exports from IVC to West Asia, along-with emigration of people from India both for trade & re-settlement to setup colonies. By this time, Talageri argues that the rbha/lbha root was already becoming archaic, though the Rig Vedic words Hastin & Vaarana were still being used. Please note that the word hastin is very ancient, making its appearance as early as the Old Rig Vedic Family Book 4. During this time, a new word for elephant “piru/pilu” was possibly also introduced.
However, there is a question that one can legitimately ask: “Why did the Sumerians absorb the pilu word for ivory/elephants rather than ibha/rbha or hastin”?
There is a quick answer & a detailed answer to this:
Quick answer: the Sumerians also borrowed 90 words related to music, musical instruments, timber etc from the Harappans, as Shail Vyas demonstrates. All these are Sanskrit or proto-Sanskrit words, not Dravidian. So, the fact that they also borrowed the word pilu means that the trade in elephants/ivory was probably dominated by Dravidian speaking people/jatis in the IVC & the trade in timber, beads, music etc were dominated by Sanskrit speaking people/jatis! So yes, there were likely Dravidian speakers in the IVC. They probably even dominated certain trades like ivory. But it’s a stretch to claim that they were present in substantial numbers, without evidence.
Detailed answer: if you look at the trend of Bahata’s paper, especially towards the end, it reverts to the Aryan Invasion theme. It’s the old argument that the IVC was Dravidian speaking (or at the very least not Indo-Aryan/Sanskrit speaking). And that Sanskrit was introduced from the outside by invading/immigrating Aryans from the Steppes.
There are several worldview differences between the Aryan Invasionists and the Out of India theorists. One of the main ones is that of “linguistic/racial purity”. It’s a prerequisite for AIT that the IVC inhabitants are racially & linguistically pure (i.e. Dravidian) – at the very least, not Indo-Aryan! If they’re proved to be, the entire theory falls apart. They also assign language to genes and claim that the Dravidian stock is somewhat different from the invading Aryan genetic stock.
The Out of India hypothesis, on the other hand, neither claims nor requires linguistic or racial purity. The IVC was a sprawling civilization, covering a million sq km in landmass, with millions of inhabitants. India has always been genetically & linguistically diverse, ever since the first Out-of-Africa migrants settled down here. So, its very likely that the IVC region was multi-lingual, multi-cultural, multi-religious, with a high genetic diversity!
Remember that India is home to not two, but four language families: Dravidian, Indo-European, Austro-Asiatic and Tibeto-Burman (not to mention some language isolates, which belong to no family).
So, a rich, powerful, geographically wide civilization like the IVC would have attracted migrants from multiple (possibly all) language families.
Now, over the course of time, as is again the wont in India, different jatis/clans/guilds would have specialized in certain trades and dominated them. Much like how the Chettiyars from Tamilnadu specialized in timber trade from Burma, while the Gujaratis dominate the diamond trade. Similarly, some jatis/guilds in the IVC would have specialized in bead making, some in the working of lapus lazuli (sourced from Afghanistan), some in timber, others in metal-working and some in ivory. Their techniques, their supply chain, the market they sell into, their land-vs-sea routes, their agents etc would have all been different, though there may be some commonalities as well (like weights, seals etc).
So, while a large number of Indo-Aryan speaking merchants would have exported beads, timber etc to Mesopotamia, resulting in a substantial number (90, as per Shail Vyas) of Indo-Aryan loanwords into Sumerian, a relatively-smaller-but-influential group of Dravidian speaking ivory traders could have exported ivory/elephants along-with their Dravidian word to Sumerian & other languages. Many of these people also emigrated, setup colonies and settled down in West Asia. There are Sumerian tablets that bear testimony to such Meluhhan colonies.
There is another possibility to keep in mind: IVC does not just mean North or North-West India. It extended all the way to Lothal (port) & Dholavira in Gujarat. Its quite possible that elephants & ivory were shipped to West Asia from this port. A sea-route would make a lot more sense for these heavy cargo, than a land-based one. Moreover, Gujarat is closer to the Western Ghats, one of the large elephant habitats in India. The Gujarati language also shows close affinity to Dravidian languages. So, Dravidian speakers need not even be present in NorthWest India, but could have been trading with Mesopotamia via Gujarat. There might have been only tenuous links initially between them and the Vedic speaking areas of North & North-West India, which increased over time. That will explain the late entry of the pilu word into Sanskrit.
Conclusion
While Bahata’s paper presents compelling data-points, the conclusions seem like a stretch. In fact, I will make a couple of more arguments that show why an Out-of-India scenario is more likely in the context of this paper:
- Unlike the Mesopotamians, Akkadians etc who had a weak vocabulary for the elephant and therefore imported pilu (because the animal wasn’t native), the Sanskritic people always had multiple words for it. These words had deep word-roots, which reflect their antiquity. This shows that the Sanskritic speakers had “native familiarity” with the elephant. This just reinforces Talageri’s central argument that the IE homeland had elephants, viz. India
- If as AIT’ians claim, the invading Aryans brought Sanskrit with them, how did they have multiple words for the elephant, especially since the Steppes doesn’t have the animal? On the other hand, if they acquired a word for it near West Asia after encountering elephants, why did they not pick up the word pilu? After all, “pilu” was popular in the timeframe when the Aryans were supposed to have come via Central Asia, circa 1500 BCE (per AIT). Alternately, if they saw the elephant only after they entered the IVC, once again, why didn’t they pick-up pilu from the Dravidian natives, since the invaders wouldn’t have had a local word for it?
Post-script: Bahata’s paper interestingly, provides us with a valuable insight. If the Vedic people were co-habiting with the Dravidian speaking ivory traders in the IVC, surely the word would have made its appearance in the Rig Veda, given that it traveled all the way to Mesopotamia. But it’s a post-Rig-Vedic word. Per Bahata’s paper, the word piru’s earliest occurrence belongs to the Old Babylonian period (2000 – 1600 BCE). So, the lower-limit of the timeframe (1600 BCE) provides a terminus ante quem (i.e. latest date) to the completion of the composition of the Rig Veda.
VogonPoem 