Thursday, June 29, 2017

Genetics and the Aryan invasion debate

Pragyata, 24 June 2017

After some fifteen years, we may be witnessing a revival in the debate between the Aryan Invasion Theory (OIT) and the Out-of-India Theory (OIT). The AIT camp has been completely stonewalling any attempt at debate from the OIT camp for years now, though in the late 1990s some of its tenors did briefly try to refute the pro-OIT arguments then doing the rounds on the brand-new internet forums. But it went into smug hibernation, having decided that on second thought, the whole OIT was too flaky and too politically tainted (even more than the colonialist-cum-Nazi-tainted AIT) to dirty its hands on. Meanwhile, save for a handful of busy bees, the OIT camp was equally smug, deluding itself that it had disproven the AIT long ago. 

Now however, one Tony Joseph, veteran business journalist, gate-crashes into the Aryan origins debate via an article in The Hindu: “How genetics is settling the Aryan migration debate” (16 June 2017) This in turn is an interpretation of the impact of a scientific paper by Marina Silva et al.:A genetic chronology for the Indian Subcontinent points to heavily sex-biased dispersals” (Evolutionary Biology BMC, March, 2017; the team was led by Prof. Martin Richards). Joseph claims that the AIT was put in doubt by earlier genetic studies but that this paper and other recent findings have firmly decided the debate in favour of an invasion. (Well, he uses the weasel word “migration”, like most AIT champions who lack the courage of their conviction; but like most “migrationists”, he turns out to mean not a migration of families in wagons, but an entry of all-male bands of warriors who end up fathering plenty of children upon native women; a scenario ordinarily known as an “invasion”.)

The Aryan invasion question is correctly formulated as: “Did IE language speakers, who called themselves Aryans, stream into India some time in 2000-1500 BC?” Joseph’s triumphal answer is: “Genetic research based on an avalanche of new DNA evidence is making scientists around the world converge on an unambiguous answer: yes, they did.”

The strictly genetical parts of his thesis, I will leave to legitimate geneticists to answer. For starters, two excellent rebuttals have come to my notice: by Anil Kumar Suri ( and by AL Chavda ( In the present article, I have mainly tried to put the highly partisan piece by Joseph in the context of the broader Aryan origins debate and of India’s culture wars.

Linguistic or racial invasion?

As a reader of internet media, much lambasted by the mainstream media, I am aware of the genetic studies in recent years that were welcomed as disproving the AIT. I was sceptical: the initial generation of genetic studies had to be taken with a big pinch of salt because it suffered of childhood diseases. More fundamentally, genes, much like archaeological findings, don’t speak. You cannot tell from genes, nor from excavated artefacts, what language the people concerned spoke. I frowned when I saw genetic findings being mustered as proof against the linguistic AIT.

Confusing language movements with demographic movements was a childhood disease of Indo-European linguistics before 1945. Especially after Charles Darwin’s Origin of Species (1859), race thinking came to dominate the Humanities. There were warnings from Indo-Europeanists, including the much-maligned Friedrich Max Müller, to maintain the distinction, but the public and many professionals started speaking of “the Aryan race”, not in the vague sense common earlier (race = any group of hereditary belonging, from family to nation and race to humanity, Sanskrit jāti), but in the biological sense. After 1945, this went completely out of fashion in the West, but in India, not encumbered with the guilt about Nazi racism, time has stood still.

That is why Joseph can assert that a genetic study has disproven a linguistic theory. Strictly speaking, that alone should stamp him incompetent for the Aryan debate. Then again, let us face the fact that among Indians, this confusion is also quite common, as it was among Europeans of an earlier generation. In principle this makes sense: you continue both the genes and the language inherited from your parents. But while your genes can’t be traded in for others, people do change language in the course of their lives, not as a rule but as a sufficiently frequent exception. Before the colonial age, which imparted European languages to a number of colonial populations, white Europe and brown India represented about a half each of the Indo-European-speaking population, thus vividly demonstrating that their language had crossed a racial frontier: either brown to white or white to brown. So, genes and language are different thing, partly related and partly not.

For an invasion as dramatic and rich in consequences as the Aryan influx is deemed to be, with a complete switch in language and culture of a highly civilized and densely populated country, it is nonetheless unlikely that it would leave no traces at all. In those days, there were no internet courses that allowed you to learn a language spoken at a great distance; for imparting a language to foreigners, it was necessary to go to their country. There must have been some kind of sizeable migration demonstrable in the archaeological and genetic record. In this respect, archaeology is completely disappointing the AIT party, for it consistently refuses to throw up any trace of this fabled Aryan invasion. But they have their hopes up anew: perhaps the genetic record can fill in the blanks?

Given that pattern of expectation, studies refuting the invasion theory, for whatever they were worth, were very welcome. They dampened the usual arrogance of the AIT camp,-- at least to the extent that people heard of them. The dominant media largely kept the lid on them. If they reported them, it was in measured, or what Joseph calls “nuanced”, tones; nothing like the triumphalism with which he himself claims that genetics has proven the AIT.

Now that at least one scientific paper has been presented as supporting the AIT, the reaction among activist Hindus was initially one of unpleasant surprise, sometimes panic, for many of them had always harboured a secret fear: “What if the scholars do manage to prove the AIT?” But this was followed by a resolve among some of them to revisit the evidence and get serious about the debate again.

Other activist Hindus, by contrast, feel vindicated in their long-standing support for the AIT. Indeed, there exists a casteist-racist fringe among Brahmins who feel flattered by the claim that they descend from foreign conquerors and that they imposed the caste system as a racial Apartheid system. Outsiders including Joseph may not know about this, but any defence of the AIT plays into the hand of the most regressive elements of Hindu society. In the West too, the AIT is not only an unassailable orthodoxy in academe, it also serves as ideological backbone for some remaining racist ideologues.

Apart from those fringes, the healthy reaction among more sober Hindu activists is the resolve that they, or at least those among them who are equipped for it, should return to the Aryan debate and win it once and for all. This promises to be an interesting cup final, for the opposing camp is self-assured and in no mind to give in.  

How does a language invade?

The origin of the Indo-European languages in India, starring Sanskrit, is principally a linguistic scenario, which must have come about on the waves of historical developments, and these may have left traces in the genetic record. Joseph is neither a linguist nor a historian nor even a geneticist, and in my quarter-century in the thick of the Aryan debate, I have never encountered his name. That need not be an obstacle, for by their own effort, people can become self-taught experts in a specialism in which they have no degree, even after a career in a different field, including business journalism. But they still have to satisfy the same criteria as the certified scholars or scientists whose equal they aspire to be. This, then, is what is missing in this article. Joseph doesn’t have a grasp of some basic issues in this debate.

This is not so exceptional. Many geneticists themselves don’t properly understand the Aryan debate, already two centuries old before genetics became a useful instrument in reconstructing migrations in history. The first studies in this field, finding e.g. that some genes were strikingly common between North India and Eastern Europe, contained conclusions that were at best not in conflict with the Aryan invasion scenario but did not prove it at all. In casu, they may at that stage have shown up grounds for either an India-to-Europe movement or a Europe-to-India movement without being able to decide between the two. Yet, they concluded in favour of the AIT because they retrofitted their own newfound data into the theory that, they heard, had already been proven by the linguists.

This is the so-called “circular argument of authority”: first you feed an expert a story, then he himself comes out with that same story, and then you can claim that your own little story has been confirmed by a world expert, thus giving it more authority. For example, when Indian “secularists” invoke the positions of Western India-watchers (about Kashmir, the supposedly secular Constitution, Hindutva etc.), the latter invariably turn out to have first swallowed the biases of their Indian “secularist” sources. So the Indian secularists are really only quoting themselves, though embellished with the academic jargon that Western graduates are so good at.

In the Aryan debate, the same thing has happened: the AIT viewpoint is first fed to experts from other fields, such as genetics, and then these view the data in their own field through the glasses which their partisan friends have put on their noses. The geneticists quoted here, both in their original papers and in the quotes from interviews, clearly have no independent grasp of the Aryan debate.

Thus, the quoted paper by Silva et al. states: “Indo-European has been frequently connected to the so-called ‘Indo-Aryan invasion’ from Central Asia ca. 3,5 ka, and the establishment of the caste system.” There never follows a critical second thought about this received wisdom. Not having studied this debate, they fall back on hearsay from their Orientalist colleagues, who all teach the AIT. What other influence would they have? Those capable of situating and defending the OIT can literally be counted on your fingers; none of them has functioned as an information source for geneticists. Though the geneticists certainly live up to the scientific method when it comes to handling genetic data; when they approach the AIT as possible explanation, they become mere followers of convention. 

Nevertheless, we should not use this circumstance as an escape clause from the debate. For meanwhile, Joseph does report on a development among geneticists that may well prove as important as he asserts.


Joseph claims he is part of a bold dislodging of an established orthodoxy: “The dominant narrative in recent years has been that genetics research had thoroughly disproved the Aryan migration theory.”

Oh yes? Where are the schools and institutes that teach the demise of the AIT? The truth is that these studies never made a dent in the pro-AIT orthodoxy. The mainstream media chose to talk about them only minimally, the textbooks never mention them at all. The only people excited about these findings were the OIT adherents, and if you listened in on them, you might have gottenthe impression that the time was ripe for the acceptance of the OIT.

What has suddenly happened, according to Joseph, is this: Until recently, only data on mtDNA (or matrilineal DNA, transmitted only from mother to daughter) were available and that seemed to suggest there was little external infusion into the Indian gene pool over the last 12,500 years or so. New Y-DNA data has turned that conclusion upside down, with strong evidence of external infusion of genes into the Indian male lineage during the period in question.”

In itself, it is no big deal that Joseph misinterprets the “mt” as “matrilineal” instead of “mitochondrial”, but again it suggests he is unfamiliar with his chosen topic. In journalism you often have to deal with unfamiliar subjects, but relative to the tall claims he makes in this article, he ought to have done his homework better. Nonetheless, he whets our curiosity for his evidence on the Y-chromosome data.

“The reason for the difference in mtDNA and Y-DNA data is obvious in hindsight: there was strong sex bias in Bronze Age migrations. In other words, those who migrated were predominantly male and, therefore, those gene flows do not really show up in the mtDNA data. On the other hand, they do show up in the Y-DNA data: specifically, about 17.5% of Indian male lineage has been found to belong to haplogroup R1a (haplogroups identify a single line of descent), which is today spread across Central Asia, Europe and South Asia. Pontic-Caspian Steppe is seen as the region from where R1a spread both west and east, splitting into different sub-branches along the way.”

Note the AIT-serving claim, appearing out of the blue, that the Pontic-Caspian steppe “is seen as” the region of origin. This expression is a sure give-away of hearsay, of a borrowed opinion. It is not based on the geneticists’ R1a research.

Frankly, I had understood that the jury is still out concerning the origin of R1A, and that it is strikingly in evidence among some “aboriginal” Indian tribes. But not being a geneticist, I will not argue too forcefully about that. At any rate, the paper by Silva et al. does not say that it found earlier studies to have been mistaken about this, and is altogether more nuanced and temperate than the tall and abrasive claims by Joseph.

In that paper, “16 scientists led by Prof. Martin P. Richards of the University of Huddersfield, U.K., concluded: ‘Genetic influx from Central Asia in the Bronze Age was strongly male-driven, consistent with the patriarchal, patrilocal and patrilineal social structure attributed to the inferred pastoralist early Indo-European society. This was part of a much wider process of Indo-European expansion, with an ultimate source in the Pontic-Caspian region, which carried closely related Y-chromosome lineages… across a vast swathe of Eurasia between 5,000 and 3,500 years ago’.”

That is something else altogether. Their evidence concerning an origin in “the Pontic-Caspian region” will have to be studied, of course. But note that the scientists admit that they have not studied the link between their genetic data and the identification of the purported migrants as Indo-European: this is only “attributed” and “inferred”, meaning “borrowed on trust from our Indo-Europeanist informers”, all of them wedded to the AIT. They do not make a professional claim for the AIT, that is only a speculative afterthought, merely for genetically attested movements.

“In an email exchange”, of which doesn’t tell us the leading questions, “Prof. Richards said the prevalence of R1a in India was ‘very powerful evidence for a substantial Bronze Age migration from central Asia that most likely brought Indo-European speakers to India’.”

Their identification as Indo-European is, in his mouth, not more than hearsay, by definition not corroborated by genetic data, nor by the archaeological data, which show up no such dramatic discontinuity, as even AIT partisans admit. “Bronze Age” is also but a vague name for a long and ill-defined period. But we retain that, if rendered correctly, he does insist on a real influx from Central Asia still showing up in 17,5% of the Indian Y-chromosomes.

Central-Asian invaders in Hindu historiography

If this claim proves correct, there are a number of other explanations. An influx from Central Asia is roundly admitted by Indian tradition. Shakas, Kushanas and Hunas came in the historical period, just before and just after the time of Christ. Long before that, since the onset of the Ice Age, many northerners escaped the barren north and migrated south, hence e.g. the many blondes described by the Egyptians among their western neighbours; India too must have been among their destinations. And between those chronological extremes, there must have been more. 

Though their incursion was considered dramatic at the time, with Vikramaditya of Ujjain hailed by Indians as the defeater of the Shaka invaders, it remained inconsequential culturally, as these invaders dissolved in the Indian population and assimilated the language and religion of their hosts. Only their genes remained, till recently in hiding, but now accessible. Today they must form part of those 17.5%.

A second consideration is that an inflow does not prevent a movement in the other direction. If the Silva paper is sensational, it might be because of passages like this one, about the Bronze Age: “Gene flow at this time was clearly bi-directional, as seen in the expansion west of lineages M5a2a4, U2c1b + 146 and M3a1b + 13105. This is reflected in the genome-wide ADMIXTURE analysis (…), where the autochthonous South Asian component (…) appears at low levels in Iran.”

For readers of The Hindu (please note, foreigners, that this name is a historical relic, and that since decades it is a left-wing anti-Hindu paper), it may come as a surprise that a paper claimed to prove the AIT actually documents a bit of OIT. Others will be less surprised by this confirmation of the emigrations from India, in line with extant archaeological findings and with the general drift of earlier genetic studies,-- which Joseph’s article was meant to counter.

Another possibility, tailor-made for the Hindu reader, is a compromise. Come to think of it, it is surprising that none of those synthesis-enamoured Gandhians raised this possibility before: a scenario that approximately satisfies both the AIT and the OIT camps. Here, there was an influx from Central Asia, which with the help of linguistics we could pinpoint to ca. 4000 BC, the farthest estimate of the fragmentation of Proto-Indo-European. Only, this influx did not bring Indo-Aryan, the exclusively Indian branch of Indo-European (distinct from other branches such as Greek, Celtic, Slavic etc., that “remained” in the west), as the AIT holds, but all of the branches, still united in their foremother, Proto-Indo-European.

As Shrikant Talageri has shown, Hindu tradition envisages all of “Aryan” history as the life and times of the descendents of Manu, the founder-king who landed with his Flood-survival ship in Manali and set up his court in Ayodhya. A descendent of his was Yayati, whose sons Druhyu and Anu must have been the ancestors of the non-Indian branches of Indo-European, destined to leave India in the pre-Vedic period c.q. in the middle Vedic period. Greek ambassadors after Alexander reported a Hindu tradition that Manu had come into India from the west. Like their estimate that Manu was crowned in precisely -6776, this may be a garbled reconstruction of hazy traditions; but for what it is worth, it might correspond to the alleged genetic findings of an early influx. After all, though India may be the cradle of all branches of Indo-European (exactly the OIT), nothing has been claimed about where the Proto-Indo-Europeans originate. It could be outside India, all while leaving the OIT untouched.

As even Silva et al. acknowledge, “genetic data have provided no clear evidence for the ‘Indo-Aryan invasions’ so far, and their very existence is challenged by many archaeologists.” Tony Joseph has a long way to go before the AIT can be accepted as proven.


In spite of the foregoing, let us appreciate why got Joseph so enthused about recent developments in genetics: “The robust conclusions of Professor Richards and his team rest on their own substantive research as well as a vast trove of new data and findings that have become available in recent years, through the work of genetic scientists around the world.”

That is not true for the identification as Indo-European, which would at best be plausible, but otherwise I acknowledge Richards must be a pioneering scientist in a fast-progressing research field. The argument of authority implied in “around the world” is also exaggerated, for the really pioneering genetic research is confined to a few countries. China has recently joined the front group, and India is catching up, but if you see exotic names among the authors, they are mostly at American or European universities; witness all the Portuguese names under the quoted research paper produced in the UK. But this much is true: whereas the Indo-European origins question holds a certain fascination for people of European descent, the new research yields very similar new insights in the demographic history of other cultures.   

Thus: “(…) many studies mentioned in this piece are global in scale, both in terms of the questions they address and in terms of the sampling and research methodology. For example, the Poznik study [David Poznik, cfr. infra] that arrived at 4,000-4,500 years ago as the dating for the splintering of the R1a Z93 lineage, looked at major Y-DNA expansions not just in India, but in four other continental populations. In the Americas, the study proved the expansion of haplogrop Q1a-M3 around 15,000 years ago, which fits in with the generally accepted time for the initial colonisation of the continent. So the pieces that are falling in place are not merely in India, but all across the globe. The more the global migration picture gets filled in, the more difficult it will be to overturn the consensus that is forming on how the world got populated.”

About that, we can only rejoice and join Joseph in his enthusiasm. And we may at once point out that a vaguely similar though more gradual development is taking place in the highly relevant field of linguistics, with the increasing sérieux of the once-flaky subdiscipline of megacomparatism. So far, Indo-European studies including the Aryan origins debate focused on what happened after Proto-Indo-European, once the fragmentation had set in. But now we can look beyond that, and explain the long-existing suspicions of a kinship with Semitic, or the newly-realized kinship with Dravidian. This too is a heady evolution, though encumbered by the 19th-century impression of bookishness when genetics has all the glitter of futuristic science.


More controversial is another big name he invokes: “Peter Underhill, scientist at the Department of Genetics at the Stanford University School of Medicine, is one of those at the centre of the action. Three years ago, a team of 32 scientists he led published a massive study mapping the distribution and linkages of R1a. It used a panel of 16,244 male subjects from 126 populations across Eurasia. Dr. Underhill’s research found that R1a had two sub-haplogroups, one found primarily in Europe and the other confined to Central and South Asia. Ninety-six per cent of the R1a samples in Europe belonged to sub-haplogroup Z282, while 98.4% of the Central and South Asian R1a lineages belonged to sub-haplogroup Z93. The two groups diverged from each other only about 5,800 years ago.”

Note that 5,800 years ago is about the time estimated by legitimate experts in linguistics for the initial fragmentation of Proto-Indo-European. It could for instance be the time when an Indo-European-speaking population left its Indian homeland.

“Dr. Underhill’s research showed that within the Z93 that is predominant in India, there is a further splintering into multiple branches. The paper found this ‘star-like branching’ indicative of rapid growth and dispersal. So if you want to know the approximate period when Indo-European language speakers came and rapidly spread across India, you need to discover the date when Z93 splintered into its own various subgroups or lineages.”

From these data about Z93, it does not follow that “Indo-European language speakers came” to India. Here you see a vivid illustration of how the interpretative framework in terms of an Aryan invasion is merely borrowed on trust from people whose bias the geneticists don’t even realize.

Nonetheless, Joseph takes this as a sufficiently strong launching-pad for an unwarranted jump: “This clear picture of the distribution of R1a has finally put paid to an earlier hypothesis that this haplogroup perhaps originated in India and then spread outwards.”

This is not the expert speaking. Note indeed that in spite of Joseph wanting Dr. Underhill to say that, and no doubt trying to make him say it, the professor is not quoted as saying it. All that the quoted text says, is that these genes have a common origin, and that they split up when spread over sizable distances. The OIT has no quarrel with that. Then, he adds the AIT as possibly explaining his findings, but that part is not his expertise as a geneticist speaking. His findings are just as compatible with the OIT, but that is just not part of his mental horizon.

According to Joseph: “This hypothesis was based on the erroneous assumption that R1a lineages in India had huge diversity compared to other regions, which could be indicative of its origin here. As Prof. Richards puts it, ‘the idea that R1a is very diverse in India, which was largely based on fuzzy microsatellite data, has been laid to rest’ thanks to the arrival of large numbers of genomic Y-chromosome data.”

But even with the data now claimed, R1a is still more diverse in India than in Central Asia and Europe. If the former is therefore incapable of being the homeland, so are the latter. Nonetheless, our AIT activist goes full steam ahead:

“Now that we know that there WAS indeed a significant inflow of genes from Central Asia into India in the Bronze Age, can we get a better fix on the timing, especially the splintering of Z93 into its own sub-lineages? Yes, we can; the research paper that answers this question was published just last year, in April 2016, titled: ‘Punctuated bursts in human male demography inferred from 1,244 worldwide Y-chromosome sequences.’ This paper, which looked at major expansions of Y-DNA haplogroups within five continental populations, was lead-authored by David Poznik of the Stanford University, with Dr. Underhill as one of the 42 co-authors. The study found ‘the most striking expansions within Z93 occurring approximately 4,000 to 4,500 years ago’.”

Here we see no link with an immigration, perfectly unnecessary for a genetic differentiation. And if there was one, it apparently took place 4,500 years ago, still a thousand years before the established Aryan invasion scenario.

But at least we note with satisfaction that Joseph has belatedly accepted that nothing points to the once-orthodox scenario of the Aryan plunderers ruining the Harappan civilization: “This is remarkable, because roughly 4,000 years ago is when the Indus Valley civilization began falling apart. (There is no evidence so far, archaeologically or otherwise, to suggest that one caused the other; it is quite possible that the two events happened to coincide.)”

Then, Joseph makes another jump: “The avalanche of new data has been so overwhelming that many scientists who were either sceptical or neutral about significant Bronze Age migrations into India have changed their opinions. Dr. Underhill himself is one of them.”

If true, such conversion among experts would count as prized secondary evidence. Thus, leading archaeology professor BB Lal’s and Sanskrit professor Nicholas Kazanas’ conversion from AIT to OIT have been functioning as serious arguments of authority, especially in the absence of such conversions in the opposite direction. But let us look more closely:

“In a 2010 paper, for example, he had written that there was evidence ‘against substantial patrilineal gene flow from East Europe to Asia, including to India’ in the last five or six millennia. Today, Dr. Underhill says there is no comparison between the kind of data available in 2010 and now. ‘Then, it was like looking into a darkened room from the outside through a keyhole with a little torch in hand; you could see some corners but not all, and not the whole picture. With whole genome sequencing, we can now see nearly the entire room, in clearer light.’”

Impressive progress, yet that does not make Underhill find genes moving into India. Though certainly trying hard, Joseph has not managed to extract a statement from Underhill that his earlier research was wrong.

Joseph is very good at making the most of what comes under his hand, and of shading over nuanced expert findings into his own blatantly partisan narrative. However, our interest is not in finding fault with Joseph; indeed we thank him for drawing our attention to this new scientific development. Our interest is in what genetics really has to say on the Aryan origins question.

North-South divide

Joseph claims another convert from OIT to AIT: “David Reich, geneticist and professor in the Department of Genetics at the Harvard Medical School, is another one, even though he was very cautious in his older papers. The best example is a study lead-authored by Reich in 2009, titled ‘Reconstructing Indian Population History’ and published in Nature. This study used the theoretical construct of ‘Ancestral North Indians’ (ANI) and ‘Ancestral South Indians’ (ASI) to discover the genetic substructure of the Indian population. The study proved that ANI are ‘genetically close to Middle Easterners, Central Asians, and Europeans’, while the ASI were unique to India. The study also proved that most groups in India today can be approximated as a mixture of these two populations, (…)”

Westerners will only be mildly excited by this, but in India, the racial interpretation of the linguistic division between Indo-Aryans and Dravidians is so vividly kept alive by Dravidianists, Ambedkarites and their missionary shepherds, that others there will heave a sigh of relief at news of the “admixture” between the two populations.

In the same vein, Indian patriots rejoiced that the study found “the ANI ancestry higher in traditionally upper caste and Indo-European speakers”.

Well of course: speakers of Indo-Aryans live predominantly in the north, so they have a higher incidence of northern genes. As for the higher castes, it doesn’t take any invasion scenario to understand that they must be more from the north, as per the written Hindu tradition. Especially in Tamil Nadu, hotbed of Dravidianism, this is striking, with Bania castes originating in Rajasthan and Gujarat, and Brahmins imported from the north by kings who wanted to endow their dynasties with Vedic prestige. So obviously, in Dravidian areas, upper castes will stand out by their more northern genes. But only gradually, for even here there is an “admixture”.

Hence this comment quoted from a prominent Indian researcher involved in this study, geneticist Lalji Singh: “But at some point in time, the ancient north and the ancient south mixed, giving birth to a different set of population. (…) This paper rewrites history (…) there is no north-south divide”! A newspaper headline made it even more sensational: “Aryan-Dravidian divide a myth: Study”.

(The linguistic state of the art on this is that Dravidian is related to Elamite in Southwestern Iran, a reasonable theory insufficiently verified but largely based on trust in the very few scholars who have investigated these data. An anti-India bias predictably deduces from this that Dravidian must have come from Elam, not the other way around. And then, after having entered India and developed in Gujarat, Dravidian expanded to the south, where a more primitive native population adopted the language. To the dismay of Dravidian chauvinists and separatists, this means that Dravidian is not native to South India, but that it is native to India. The political unit they ought to be defending is not “Dravidasthan”, but India. Moreover, Dravidian is related to Indo-European on a par with Uralic, Semitic et al., descending from a common grandmother-language some 15,000 years ago, “Nostratic”, as distinct from more distant languages like Munda, Tibetan, Andamanese etc.)

ANI and populations to the west

Joseph’s misunderstanding of the whole Aryan issue is again clear from this claim: “By itself, the study didn’t disprove the arrival of Indo-European language speakers; if anything, it suggested the opposite, by pointing to the genetic linkage of ANI to Central Asians.”

That North-Indians and Central-Asians are related, as the genetic linkage between them implies, is true in any serious scenario, certainly also in the OIT. Being closer to Central Asia, it is perfectly normal for the ANI to have more in common with it than the ASI. The only narrative that throws up a problem here is not the OIT, with speakers of Indo-Europeans streaming out of India into Central Asia and thence into Europe, but a purely reactive alternative to the AIT, viz. the non-migration scenario. When Europeans first thought up the AIT around 1820 (after having espoused the OIT for half a century) and then took it to India, many Indians simply denied that anyone had come from anywhere, and pointed out correctly that nothing amounting to what is known as the Aryan invasion (with the importation of Sanskrit from outside) is described in scripture.

No, but the Rg-Veda and the Puranas do describe a hoary emigration from India, viz. of the Druhyu and Anu tribes. Therefore, knowledgeable Indians have adopted the OIT, which necessitates commonalities between India, Central Asia and Europe as much as the AIT does.

But I have no doubt that Joseph’s dialogue partners in Aryan origins discussions included many who saw no need to explain the linguistic commonalities between Indo-Aryan and the languages further west, and simply settled for the static and unimaginative non-migration narrative. Those people reject the very concept of an Indo-European language: to them, Greek, Germanic and other European languages are not related to Sanskrit, and so there is no need for any migration either way to explain how they came to be in their present habitat. If at all you confront them with similarities, they make bold that these are just loanwords.  

This is part of a larger problem, that explains why the OIT has made no headway at all in terms of international acceptance. Most Indians don’t have their heart in the OIT, a non-Indian term launched by the American Hare Krishna follower Dr. Edwin Bryant ca. 1996. In describing the Aryan debate, he took ample account of a vocal Indian vanguard that understands the basics of the Indo-European question, viz. that one language straddles the Hindu Kush, and that therefore half its speakers must have crossed this border either in eastern or in western direction (AIT c.q. OIT). But far more weighty and numerous in India are those who merely want to be left alone, who don’t want to be bothered with the AIT and all its concomitant politics (colonialist, Nazi, Dravidianist, Dalit). They simply have no specific theory about the non-Indian Indo-Europeans: these are simply not within their mental horizon, which stops at the Khyber Pass. The question whether they migrated, and whether they really speak a cognate language, is simply of no interest to these AIT-rejecting laymen.

Western participants in the debate may have debated some OIT sophisticates from a distance, but they have also interacted with Indian-born colleagues from, say, the Medical or Engineering Department of their university, who are bright but have never studied the Aryan question. These far more representative Indians often have strong opinions about the “breaking India” implications of an Aryan invasion but not the necessary knowledge to argue this point, often even rejecting the whole notion of an Indo-European family. They are “under-informed but over-opinionated”. Meeting such people makes Western Indo-Europeanists sceptical of any Indian contribution to this debate, and by extension of any pro-OIT contribution even by Western colleagues. Both Sanskrit professor Nicholas Kazanas (Athens) and myself have many anecdotes up our sleeves of how any Homeland debate in which the OIT figures, is being stonewalled by AIT-leaning linguists.

Genetics and linguistics

Now, Tony Joseph is going to teach us reason: “However, this theoretical structure was stretched beyond reason and was used to argue that these two groups came to India tens of thousands of years ago, long before the migration of Indo-European language speakers that is supposed to have happened only about 4,000 to 3,500 years ago.”

For that, no “stretch beyond reason” is necessary, for the two phenomena, the genetic division in ANI and ASI and the linguistic division in Indo-European, Dravidian and others, are unrelated. Even if the ANI people came from outside, any “reasonable” scenario has them mingle with natives and take in lots of their genetic heritage, rather than commit genocide on them and replace them wholesale with a different genetic type. Then again, this genocide does exist, viz. as part of the Dravidianist mythology, especially in its missionary version, because of overcompensation for the real genocide of natives in the Americas that accompanied their Christianization.

If I have understood anything of these genetic discussions, it is that chronological estimations of genetic events are possible (much as in historical linguistics, though this “glottochronology” is in fact highly controversial). Apparently, the split of the Indian people in a northern ANI and a southern ASI dates a lot deeper into the past than the fragmentation of Indo-European and than the “Bronze Age migration” discussed in Silva et al.’s paper, and need not even result from a migration, as Joseph had assumed. And if there was a migration at all, it may not have been the Aryan invasion, or may have nothing to do (causally nor chronologically) with the division of Indo-European.

The concepts of ANI and ASI were developed by geneticists untrained in linguistics and never visibly bothering about a coordination between their own data and those of historical linguistics. So, it is not so scandalous if the two sets of data are greatly at variance.

Conspiracy by association

In some reactions, Tony Joseph has been attacked on account of the company he keeps. Look whom he quotes: “In his column for Discover magazine, geneticist Razib Khan said this about the media coverage of the study: ‘But in the quotes in the media the other authors (other than Reich that is -- TJ) seem to be leading you to totally different conclusions from this. Instead of leaning toward ANI being proto-Indo-European, they deny that it is.”

Contentswise, we have a repeat of Joseph’s just-discussed mistakes: (1) he identifies the genesis of ANI with an immigration, as if such genetic development could not originate in a vast area like the subcontinent; and (2) he identifies this unsubstantiated migration with the supposed Indo-European migration. But the interesting thing here is not what Khan says, but who he is.

It is rumoured, in some reactions below Joseph’s article as well as elsewhere, that Khan is a right-winger, has published on the “race-realist” website Vdare (after Virginia Dare, the first white person born on American soil), and has been sacked as a columnist by the New York Times. This is “guilt by association”: Joseph’s position is blackened by association with the company he keeps, and Khan’s position on Indian genetics is blackened by association with his writing outlets. This type of gossip is very popular among illiterates who lack the habit of serious debate on issues, and of seasoned debaters who realize that in a given case, no argument ad rem will succeed, and who therefore resort to an argument ad hominem.

A quotation that deserves repetition in almost every debate on the Aryan question, and many other debates besides, is: “Great minds discuss ideas, average minds discuss events, small minds discuss people.” It is often flatteringly ascribed to Eleanore Roosevelt but is traceable farther back, at least to Theology professor Samuel Miller in 1827. It contains part of the reason why AIT skepticism has made no headway.

A few of us have tried to present the arguments for the OIT, but our discourse was drowned out by the shouting and cackling of numerous vocal Hindus against famous upholders of the AIT: “The British concocted the AIT as part of their divide-and-rule policy!”, “William Jones was a colonial agent!”, “Friedrich Max Müller was a missionary agent!”, “Michael Witzel is a professor of Harvard, started as a theology school!” This has greatly put off serious scholars who didn’t know better than the AIT and would have given a fair hearing to the OIT if it had not been associated with this miserable conspiracy discourse (as well as with Hindu Nationalism).

Not that their camp is much better. There too we have “guilt by association” thinking. AIT pamphlets often contain phrases like: “The OIT is defended by people like PN Oak, NS Rajaram, Koenraad Elst…” (This, incidentally, is untrue: the late PN Oak deduced everything everywhere from India or Hinduism, with no special role for Indo-European, and NS Rajaram rejected the AIT but equally pooh-poohed the OIT and any other concern about any non-Indians, including the claim that they came from India.) This is mere juxtaposition, but in the real world it functions as association and achieves the target of putting all the people mentioned outside of polite society.  

All these associators on the anti-AIT side can shout all they want, because they only speak to their own side, never interact with the AIT camp and never suffer any (healthily correcting) consequences. The consequences are for us, who live in the real world and interact with real opponents. I have been disinvited at several conferences (and no doubt silently excluded from speaking there at many more) because, though my abstract was judged interesting enough, someone up there was briefed about my views and associations, and intervened to my detriment.

Another aspect of the backbiting against Razib Khan concerns its relevance. Those who cite the incriminating opinions about him, should tell us what difference such contemporary facts make to the truth or otherwise of the AIT, which pertains to events that took place thousands of years ago. It is irrelevant chronologically. At least if your goal is not to win an ongoing debate with Joseph by making him look bad, but to disprove the AIT. I guess small minds don’t aim that high.

It is also irrelevant logically. The truth of a proposition is independent of its formulator’s record or character. Surely you can find writing-pads from Adolf Hitler’s childhood in which the devil himself writes: “1 + 1 = 2”. Does this suddenly become untrue just because an unsavoury character agreed with it?

As an extra, I would like to learn whether the purported boycot of this alleged brown racist is a policy that our AIT disbelievers advocate. Attempts to shut down debate and impose left-wing control are as much a problem in Berkeley as in Jawaharlal Nehru University. The people who call this Khan a “right-winger” are the same who attack Hindus with the same allegation. Hinduism is deemed a racist religion, with Brahmin Aryan invaders oppressing the native Dalits and Dravidians. American liberals, in this case strange bedfellows with the Evangelists, consider the struggle against caste and untouchability and against its justifying ideology, Hinduism, as a noble and necessary struggle, a new frontline now that slavery has been abolished elsewhere.

So, this Razib Khan may have unsavoury opinions or associations, and he may be downright wrong about the AIT, but he is entitled to having his comments on a genetics debate discussed in an Indian left-wing paper. To that extent, he deserves our support.

 Symmetrical migration

Let us return to genetics and quote Joseph again: “Let’s (…) ask what Reich says now, when so much new data have become available? In an interview with Edge in February last year, while talking about the thesis that Indo-European languages originated in the Steppes and then spread to both Europe and South Asia, he said: “The genetics is tending to support the Steppe hypothesis because in the last year, we have identified a very strong pattern that this ancient North Eurasian ancestry that you see in Europe today, we now know when it arrived in Europe. It arrived 4500 years ago from the East from the Steppe...”

Yes, there was an Aryan invasion, but it was not the one usually meant. In Central Europe in the -3rd millennium, the material culture changed drastically. In some areas, the population was entirely replaced by newcomers, in others too the percentage of newcomers was sizeable. By culture and physical type (now mapped and understood more precisely thanks to genetics), they were the same people later identified as Celts, Germans, Romans. There, in Europe, we really have an invasion of Aryans moving in from the east. And now that we know what a real Aryan invasion looks like, we note that it is completely missing in India.

Speaking of which: “About India, he said: ‘In India, you can see, for example, that there is this profound population mixture event that happens between 2000 to 4000 years ago. It corresponds to the time of the composition of the Rigveda, the oldest Hindu religious text, one of the oldest pieces of literature in the world, which describes a mixed society...’ In essence, according to Reich, in broadly the same time frame, we see Indo-European language speakers spreading out both to Europe and to South Asia, causing major population upheavals.”

Note that Reich doesn’t describe an invasion, at least not in the passage Joseph chooses to quote here. He describes an admixture of different populations, something that can easily be imagined as internal to a large territory such as India, not needing foreign immigrations. Instead, it could have to do with the migrations triggered by the dessiccation of the Saraswati basin, for instance. It is only Joseph himself who interprets Reich’s statement as “in essence” a matter of Aryan invasions.

Also note that Joseph has “the Indo-European language speakers” spread from the steppes to “both Europe and South Asia”. While this is not impossible, it rarely happens. Migrations are mostly very asymmetrical because the reasons why humans migrate are rarely symmetrical.

Thus, the Bantu languages were carried forward by the Neolithic innovation, agriculture, and ended up being dominant in half of Africa, all starting from somewhere in sub-Saharan West-Africa. Why did they only expand in a southeastern direction, and not north? Because you cannot practise agriculture in the desert.

Similarly, Austronesian has spread south from Southeast China, not north, because to their north was a numerous and well-ordered society that would assimilate any immigrants; southwards, they could find a niche in Malaysia and Indonesia, and virgin territory in islands from Madagascar to New Zealand and Hawaii. For the same reason, Russian could expand eastwards all the way to Alaska, but not westwards. In the case of Arabic, no such reason can immediately be identified (Iran could just as well have been arabicized as Egypt), yet fact remains: Arabic spread from Arabia westwards all the way to Morocco, and not eastwards at all.

Also, the steppe does not support large populations, whereas India was a demographic heavyweight, then already. It is not impossible for (all-male) bands of steppe warriors to conquer centres of civilization, but leaving a lasting imprint is rarer. The Huns invaded China, India and Europe and did not leave their language anywhere; in fact, we are not even sure which it was (Attila was a Gothic nickname; after much research, the dominant estimate now is that they spoke Chuvash, a para-Turkic language marginally surviving in Russia). The Mongols invaded Central and Western Asia, Eastern Europe, and China, yet except for marginal Kalmukkia, they did not leave their language anywhere. And such invaders should have completely changed the linguistic and cultural landscape of mighty India?

Parting shot

The final paragraph is merely an exercise in slamming open doors. Or so it seems, for several in-your-face assertions are built into this innocuous piece of journalistic emptiness:

“What is abundantly clear is that we are a multi-source civilization, not a single-source one, drawing its cultural impulses, its tradition and practices from a variety of lineages and migration histories.”

This is a phrase absolutely no one will disagree with; though it conceals the more pressing question how weighty the different contributions are, and the false implication that these are all equal. Yet, it is here for a reason, part of The Hindu’s editorial line: it is meant as a punch in the face of the Hindu Nationalists, who stress unity. Not racial unity, as is here falsely intimated, but still some kind of pan-Indian sense of national unity, translating today in e.g. the conviction that Kashmir belongs with India. Therefore, among secularists, it is always welcomed if an anti-unity statement of any kind is smuggled in.

“The Out of Africa immigrants, the pioneering, fearless explorers who discovered this land originally and settled in it and whose lineages still form the bedrock of our population”.

That “bedrock of the population” phrase may not be so innocent, as it is probably meant to exclude the Brahmins and thus forms part of the dominant anti-Brahmin discourse. But it is not said in so many words, so let that pass. Anyway, Joseph already refrains from saying that Indian indigenousness is confined to the tribals (some 8%), whom the missionaries ca. 1930 have started flattering as Adivasis, a pseudo-traditional Sanskrit coinage meaning “aboriginals”, like the Amerindians in America.

Nevertheless, against the “Aryan invaders”, he still tries to affirm their superiority by calling their entry from Africa more than 50,000 years ago “fearless”. Exchanging lion territory for tiger territory must indeed be courageous and certainly “pioneering”, but peopling an empty land is not braver than invading a human habitat. The Aryan invaders may never have existed, but if you do postulate them, as Joseph does, you have to at least credit them with the bravery required for defeating a vastly more numerous native population. If you care to look at the details, his bias is showing.

“[T]hose who arrived later with a package of farming techniques and built the Indus Valley civilization whose cultural ideas and practices perhaps enrich much of our traditions today”.

Here we get the bulwark part of the secularist view of ancient India: the Harappan population spoke a different language than the Northwest-Indian population today, mostly taken to be Dravidian; and their civilizational innovations starting with agriculture had been borrowed from abroad, viz. from West Asia. This latter point is important to stress, as Hindu Nationalists might get the pretentious idea that some inventions had been done in India and even by Indians; Allah forbid!

But at least Joseph recognizes that the Harappans, deemed non-Aryans, “arrived” from somewhere else. We are talking of what was then the largest demographic concentration on earth, so it is hard to imagine how they all came from the surrounding mountains and deserts. Nevertheless, the core of truth here is that Dravidian, deemed to have been the language of Harappa, arrived from somewhere. The aboriginality of Dravidian, which Dravidian chauvinists take for granted as a logical counterpoint to Aryan foreignness, deserves questioning.

“[T]hose who arrived from East Asia, probably bringing with them the practice of rice cultivation and all that goes with it”.

The Mundas, tribals whose language must have been spoken all around the Bay of Bengal, speak an Austro-Asiatic language related to Vietnamese. However, the jury is still out on whether their homeland (as well as the homeland of rice cultivation) was in India or in Southeast Asia.

“[T]hose who came later with a language called Sanskrit and its associated beliefs and practices and reshaped our society in fundamental ways”.

This phrase, affirming the foreign origin of Sanskrit through the Aryan Invasion Theory, is the raison d’être for this whole paragraph. Tony Joseph may not be a geneticist, nor a historian or linguist, but having been editor of the Business Standard, he is a first-class journalist. The occupational hazard of this vocation is that you have to talk about any topic that may come under your hands, often very much outside your area of expertise; such is the case in this article about the genetic evidence for an Aryan invasion. But a strength of this professional group is their mastery of simple rhetorical devices. Case in point: writing a conciliatory final paragraph full of empty phrases amounting to an all-together-now chumminess, and yet, inside it, burying a dagger aimed at your usual target: “Aryans”, Brahmins, Hindus.

“[A]nd those who came even later for trade or for conquest and chose to stay, all have mingled and contributed to this civilization we call Indian.”

Though not the topic of the article, this phrase is nevertheless highly welcome to drive home the usual concern of any secularist paper: affirming that Muslims and Christians are equally entitled to whatever India has to offer. Note however that in crucial respects, unlike earlier immigrants such as Shakas and Hunas, who have dissolved their identities in the ambient population and culture, Muslims and Christians have not “mingled”.

“We are all migrants.”

What an enlightened phrase, meant to reassure his secularist friends of his own virtue and to exercise moral pressure on status-conscious Hindus that they should shed this backward attachment to the nationalism-infested Indian homeland theory. Yet, do you think that all the anti-Hindu and anti-Brahmin activists who now go around triumphantly quoting Joseph’s article, intend to say: “We too, we are but migrants, as much as the invading Aryans”?


By now, Tony Joseph may wish he had never written this piece. He presents a blatantly partisan interpretation of a recent research paper in a field he visibly doesn’t master. At least he could have had it proofread by a legitimate geneticist. His bias pertains to the Aryan origins question, and that too he hasn’t thought through.

Then again, I understand that he felt the normal journalistic urge to let the world know about a new development that he considered very important. For a definitive account, you do not go to journalists, their role is not that of experts, merely of messengers: “Something is happening over there; now go see for yourself what is really is.” At the present state of the art in genetics, however, developments are not as dramatic as he seems to have thought.

Nonetheless, we had better look into them. The geneticists he has quoted here may not be the publicity-seeking types who like to make bold statements. Nonetheless, their belief in an all-male incursion into India that has left traces in the Y-genes may well have substantial ramifications for Indian history, even by implying an Aryan invasion. And while none of them has been quoted as actually having proven, through his research, an Indo-European invasion, it is still possible that some of them do think so. Or, because of the present commotion over Joseph’s article, they may step back to make up their minds, now with a better grasp of the Aryan origins question, and finally conclude that what they have proven, does imply the Aryan invasion. It will take argumentative acumen and a serious research effort to convince them otherwise.

Hindus have long acted as if it is enough to convince yourself in order to be entitled to declaring yourself as winner of the debate. In fact, it is necessary to convince others before you can rest on your laurels. Therefore, we must thank Joseph for spurring us into more and better research. After two centuries, let’s get this invasion nuisance over and done with.

Read more!