The debate on Aryan migration has long divided Indian academics into opposing camps. Two recent genetic studies appear to have solved the mystery of where we come from by offering fresh insight into who the people of the Indus Valley Civilisation really were
The Rakhigarhi skeleton at the National Museum, Delhi (Photo: Rohit Chawla)
IN THE mid-2000s, when archaeologist Vasant Shinde was leading an excavation at the Indus Valley Civilisation (IVC) site of Farmana in Haryana, he would often travel, some 25 kilometres away, to another known IVC site, Rakhigarhi. On every visit he would find more mounds being razed and new houses getting built over it. The current-day village of Rakhigarhi of around 5,000 individuals was expanding over and around the old archaeological site. “It was terrible, and I could see it happening in front of my eyes. Here was the most important [archaeological] site in India and it was getting destroyed,” he says.
For a long time, sites such as Rakhigarhi were considered less important compared with its more famous counterparts in Pakistan. The conventional view has been that the IVC originated in Sind in current-day Pakistan, around sites like Harappa and Mohenjodaro, and spread later into distant areas, such as Rakhigarhi and Farmana, in Haryana.
Rakhigarhi was discovered in the 1960s, briefly excavated in the 1990s by the Archaeological Survey of India, before being forgotten about entirely. But we now know, because of the efforts of Shinde and other archaeologists, that Rakhigarhi is much larger than even Mohenjodaro and is throwing up such old dates that some archaeologists have begun to consider the possibility that the civilisation might have even first begun here.
By the time Shinde turned his eyes to Rakhigarhi, he had spent several years exploring IVC sites across northern India. He now reoriented the objectives of his archaeological excavations. He didn’t want to find things, he wanted to answer questions. He was leading what he calls “problem-oriented excavation”.
“There was no point in finding what had already been found at other [IVC] places, like Mohenjodaro. We already knew the IVC was developed. No point finding more signs of development,” he says. Instead, he turned his eyes to the big questions. How did the IVC come about? How did it transition over time? And the biggest of them all: Who were the people at the IVC?
Several have had a go at this question. Some have looked for archaeological clues; others have tried their hand at deciphering its script. All efforts however had so far come to naught. But Shinde was using an entirely new technique—the emerging field of genetic science. There was however one issue. How do you find human specimens from the IVC that had somehow managed to withstand several millennia of the harsh Indian climate and heavy rains and retained sufficient DNA information?
Genetics has in a very short time begun to revolutionise the study of the ancient human past. Harvard geneticist David Reich, who has made some significant discoveries, including about ancient people living in South Asia, compares geneticists to barbarians. ‘Geneticists may be the barbarians coming late to the study of the human past,’ he writes in his 2018 book Who We Are and How We Got Here, ‘but it is always a bad idea to ignore barbarians. We have access to a type of data that no one has had before… .’
Shinde first began collaborating with geneticists and researchers at Hyderabad’s Centre for Cellular and Molecular Biology (CCMB), then forensic scientists from South Korea and later with Reich. But he kept hitting a dead end. Skeletal remains were hard to come by. And those that he found weren’t able to provide sufficient DNA information. The digs at Farmana, where they had found a cemetery and human remains, had led to a spectacular failure. All the samples, because of their inexperience at handling remains for genetic data, had been contaminated. The graves had been left open for far too long and too many curious people had visited the site. At one point, Shinde even began to look inside graves for parasite eggs that might have once existed in the gut of Indus Valley people, in the hope that this could lead to the researchers being able to isolate the DNA of their hosts.
It appeared that genetics too was failing to shine any light on the question of the IVC people’s identity.
The IVC stands at the centre of a large and complex mystery about our origins. It is the missing piece in not just an archaeological but also a highly political jigsaw puzzle. Who were these people living in this large swathe of land along the Indus River and its vicinity? Were they indigenous or outsiders? Were these the same people who later composed the Vedas and started what is known as the Vedic Age? Or did another foreign race come and drive them out? And, most crucially, how do we, modern Indians, relate to this ancient group?
Genetic studies have so far pointed towards the coming of a new group of people into the Indian subcontinent towards the end of the IVC about 4,000 years ago. According to these studies, modern Indians are the descendants of two highly distinct ancestral groups. One of them, categorised as the Ancestral South Indians (ASI), who probably spoke an early Dravidian language, had been living in the subcontinent for thousands of years. The other, consisting of people who had come to the Indian subcontinent from the Yamnaya region of Central Asia—the Eurasian Steppes—and who probably spoke an Indo-European language, termed Ancestral North Indians (ANI), had formed more recently. (There is an interesting catch here. It has been discovered just a few years back that both the ANI and ASI groups, while being distinct from one another, also possess large amounts of Iranian-related ancestry.) All of us, except for the people who remained secluded in the Andaman Islands, both from the highest to the lowest castes, including non-Hindu tribal populations living outside the caste system, are affected by the mixing of these two groups. The percentages of the mixture however vary depending upon geography, caste and language groups. There is higher ancestry among upper castes, Indo-European language speakers and those who originally hailed from north India.
If one were able to decode the genetic history of the people at the IVC and find either the presence or absence of ANI ancestry in their samples, it would go a long way in confirming whether or not this ancestral group was indigenous to this region.
This hadn’t happened so far.
But last week, as one of two studies published in the Science and Cell journals (published by the same group of researchers and on the same date) describes, geneticists managed to squeeze out, with immense difficulty, just enough genetic data from the petrous bone in the inner ear region of the skull of a 4,000-plus-year-old woman from the IVC.
Genetic studies hint at the higher social status of ANI ancestry than the ASI. Those in upper castes tend to have a higher proportion of ANI ancestry compared those below them
While the study in Science is more expansive, using the genes of over 500 people who lived over the past 8,000 years, to track migrations and admixtures of various ancient people across South and Central Asia, the second study, published in Cell, focuses entirely on the analysis of a single genome, the first ever, from the IVC.
This woman, titled I6113, was excavated by Shinde and a team of archaeologists from a cemetery in Rakhigarhi in 2016. IVC cemeteries usually throw up three types of burials. Either the graves contain the remains of the entire body along with possessions such as pots, beads and bangles, or just a few bones and possessions (leading some experts to believe that some people in the IVC also left dead bodies in the open, for vultures and other scavengers, and buried whatever remained), or empty graves where only possessions are found (possibly because the person’s corpse could not be traced).
Not only were all of I6113’s remains found in her grave, her head, like typical IVC graves, was to the north and the feet to the south. There were also a very large number of ceramic vessels and beads, leading Shinde to believe that she probably enjoyed a high social status during her lifetime.
The remains of this woman, along with 60 more people from the IVC, were collected. In the lab, the list of the most promising samples came down to just four. Among them, only a tiny bone from the inner ear region of I6113 managed to provide sufficient data.
“India is a very challenging place for a geneticist trying to extract ancient DNA,” says Vagheesh M Narasimhan, a geneticist working at the Reich Lab in Harvard Medical School’s Department of Genetics who was the lead author of the Science paper. “The weather is so harsh. You have a very hot sun and then heavy rains which degrades the genetic information in ancient samples.” Niraj Rai, the head of the Ancient DNA Lab at Lucknow’s Birbal Sahni Institute of Palaeosciences, who was also a co-author of this paper, claims that the work put into I6113 is the most extensive genetic research ever conducted on any single sample. “This was going to be such an important sample. And we were very meticulous that there should be no questions over it,” he says.
I6113 carried no ancestry from people of the Steppe region. This means the movement of people from this region into South Asia happened only after the decline of the IVC. I6113 instead contained a mixture of two groups: an ancient Southeast Asian hunter-gatherer group and another group with an extremely old Iranian-related ancestry. (This ancient Iranian component in the IVC people dates to at least around 10,000 BCE, long before farming emerged in the Fertile Crescent region in Middle East, not just overturning the theory that Iranians brought farming into South Asia, but also establishing that the split in the common ancestral line between the IVC people and ancient Iranian farmers happened a very long time ago.)
Although the analysis is based on just a single sample found in the IVC, when we know that the IVC was a cosmopolitan place with various people living there, it does appear that her genetic profile would match what was found among the people of the IVC. This claim is supported by the findings of 11 ‘outliers’, individuals living between 2500 and 2000 BCE and found in modern-day Iran and Turkmenistan, whose genetic ancestry and burial sites appear to suggest they were from the IVC. We know from archaeological evidence that the people of the IVC were in touch with people in these parts. These 11 outliers, the researchers believe, were migrants or the children of migrants from the IVC.
According to the two studies, after the decline of the IVC, some of the people from the civilisation mixed with Southeast Asian hunter-gatherers to form the ancestral group termed ASI. These people probably spoke a proto-Dravidian language. The others from the IVC mixed with with the incoming migrants from the Steppe, who probably brought the Indo-European languages, to form the ANI group. These two ancestral groups, ANI and ASI, then mixed and gave rise to the great diversity of India. A few years ago, some of the researchers associated with these two papers also found that the mixing between the two groups took place mostly between 2000 and 1000 BCE (also the time the IVC was collapsing and the Rig Veda was being composed). Around 2,000 years ago this admixture stopped, pointing probably to the emergence of the caste system and the practice of marriages happening only within a subcaste.
“We are suggesting [through our findings] an out-of-India migration theory,” Rai says over the phone.
According to the conventional understanding of this theory, the so-called Aryans were an indigenous race of people, some of who moved out of the subcontinent towards Central Asia and Europe, thus explaining the genetic and linguistic similarities we find among the people in those areas and some Indian groups today. But Rai uses the same terminology to drive at an entirely different theory, one that doesn’t seem to be supported by what is mentioned in the study. The discovery of 11 people from the IVC at sites in Iran and Turkmenistan indicates strongly, he says, that the common ‘Iranian’ ancestor could be from South Asia. “The IVC people shared some kind of ancestry with the Iranians but the ancestral population could be South Asian. More results are required but there is a strong possibility of an out-of-India migration,” he says.
The paper in Cell however says just the opposite: ‘Multiple lines of evidence suggest that the genetic similarity of I6113 to the Indus Periphery Cline individuals [the 11 outliers] is due to gene flow from South Asia rather than in the reverse direction.”
This is part of an unusual pattern that has emerged after the publication of the studies. Some of the Indian researchers associated with the study have made claims that are unsupported and sometimes run counter to what the studies say. Several media outlets interpreted the findings in such a way so as to claim that it debunks the Aryan migration theory, when in fact the studies now make it clear that the IVC people had no genetic connection with the Steppe pastoralists and whatever Steppe ancestry we now find among modern Indians points clearly towards the arrival of a new set of people after the IVC.
At one point during the press conference while reporting these findings, Vasant Shinde told journalists, “Harappan people are the same as Vedic people.” This is not what the two papers find. The Vedas were composed in Sanskrit, which falls in the Indo-Iranian branch of the family of Indo-European languages. According to the Science study, ‘The Steppe ancestry in South Asia has the same profile as that in Bronze Age Eastern Europe, tracking a movement of people that affected both regions and that likely spread the unique features shared between Indo-Iranian and Balto-Slavic languages.’
When asked, Rai agrees that there is no evidence to back Shinde’s claim of the Harappans being the same who started the Vedic Age. He however also says that he does not agree with the suggestion in the Science study that Indo-European languages probably came from outside the subcontinent. “I don’t think we have enough proof to say this,” he says.
This unusual diversion from what the studies themselves say perhaps has something to do with the intense political scrutiny over the findings. Rai says they removed all political terms from the papers. “The papers are purely scientific,” he says. “Let people make their own inferences and interpretations.” Reich mentions in his book how in 2008 their work on ancient Indian ancestry almost came to a halt when he discovered that modern Indians trace their ancestry from two distinct groups, one being a ‘West Eurasian’ group. His Indian collaborators, who had supplied the samples, Lalji Singh and Kumarasamy Thangaraj of the CCBM in Hyderabad, he claims, appeared to be threatening to nix the project. ‘They did not want to be part of a study that suggested a major West Eurasian incursion into India without being absolutely certain as to how the whole-genome data could be reconciled with their mitochondrial DNA findings. They also implied that the suggestion of a migration from West Eurasia would be politically explosive,’ he writes. The impasse was resolved by coming up with new terms for the two ancestor groups, ASI and ANI (which Reich had earlier termed West Eurasian). Although the ANI’s relation to Europeans and Central Asians was mentioned, the paper made no claims about the location of their homeland or any migrations.
When I ask Shinde about the incongruence between what he says and what the two studies report, he says, “Look, we are presenting the Cell paper [which focuses on the IVC’s I6113 sample] as the base… The [Science] paper was written before the findings of the Cell paper. So [we haven’t] been able to take [the Cell paper] into account.”
He appears guarded when I ask him more questions on this issue. According to him, we all come from the same ancestors. But people use the theory of migration to pull one another down. And he corrects me. “Let’s not say ‘migration’ [of Steppe pastoralists]. There was no migration. Maybe we can say movement.”
Narasimhan points out that if we look at the issue of whether the people of IVC later established the Vedic Age through the prism of languages, then it is clear that this is not the case. “Indo-European languages are now spoken right from Dublin to Delhi. How did this come to be? And language spreads in pre-state society by the significant movement of people,” he says. “But genetically [if we look at whether IVC people established the later Vedic Age], then it is not wrong to say so. The people of the Vedic Age were a mix of Steppe pastoralists and the IVC people. So the IVC did [genetically] contribute to it.”
The science of genetics usually can only tell who one’s ancestors were, or when and where a mixture began. But in the Indian context, this science has also begun to provide an interesting insight into the resulting social structure such a mixture might have provided.
Geneticists, in the past, looking at the mitochondrial DNA (which is transmitted only from mother to daughter) of modern-day Indians found that nearly all of them come from the ASI, even in north India. However, when they looked at the Y-chromosome, which is transmitted from father to son, they could see a large trace of ANI ancestry. About 17.5 per cent of Indian male lineage, for instance, belongs to the haplogroup R1a, which is found today across Europe and Central and South Asia.
How does one explain such incongruence? This probably has to do with the differences in social power between men and women and who among the two ancestral groups was more powerful. According to Reich, similar sex-asymmetric population mixture is seen among African-Americans, where the approximately 20 per cent of ancestry that comes from Europeans derives in an almost four-to-one ratio from the male side, and the Latin American population in Colombia, where the approximately 80 per cent of ancestry that comes from Europeans is derived in an even more unbalanced way from males (a fifty-to-one ratio). ‘[T]he common thread is that males from populations with more power tend to pair with females from populations with less,’ he writes in his book.
‘This pattern is exactly what one would expect from an Indo-European-speaking people taking the reins of political and social power after four thousand years ago and mixing with the local peoples in a stratified society, with males from the groups in power having more success in finding mates than those from the disenfranchised groups.’
Genetic studies also hint at the higher social status of the ANI compared to the ASI. Those in upper castes typically tend to have a higher proportion of ANI ancestry compared to those traditionally below them, even when the two groups are from the same state and speak the same language. Those belonging to the Brahmin caste tend to have the highest ANI ancestry. The Science paper mentions finding incredibly high proportions of Steppe ancestry among Tiwaris and Bhumihars today. This is another line of evidence, the paper writes, for a Steppe origin for South Asia’s Indo-European languages since Brahmin groups are the traditional custodians of liturgical texts in the early Indo-European language Sanskrit.
Shinde, even though being one of the credited researchers of the Science study, disagrees with this view. According to him, if such a group of Indo-European language speakers came from outside the subcontinent, they should have left behind archaeological evidence. “When we go to Europe or America today, don’t we still carry our culture there? The women still wear saris there, right?” he asks. “Why don’t you see such a migration archaeologically then?”