Missing Links

At a seminar in Delhi in the first week of July, Vasant Shinde, an archaeologist and the vice chancellor of Pune’s Deccan College, said that in late September or early October, we should expect a paper that will significantly contribute to our understanding of ancient India. The paper will reveal the results of DNA tests on human skeletons uncovered at Rakhigarhi, Haryana, the most extensive of all Harappan civilisation sites, where excavations supervised by Shinde are still underway.

The DNA results should go some way towards settling the question of the genetic relationship between the people of the Harappan civilisation and the current population of the subcontinent. What the DNA tests reveal—say, if they show that the Harappans had a close genetic affinity to the Indo-European language speakers who composed the Vedas, or to the Dravidian language speakers of the south—will not only shed light on the Harappans. It will also affect almost every question regarding the evolution of Indian civilisation after the decline of the Harappans—including the question of the possible migration of Indo-European speakers from outside the subcontinent into north-west India: the famous Aryan migration or invasion theory.

Some background is needed to understand the current status of the problem. We know that by 3,500 years ago, the main sites of the Harappan civilisation (whose script is still undeciphered and whose spoken language remains a mystery) were more or less abandoned, while numerous smaller settlements in the upper reaches of the drying course of the Ghaggar-Hakra river (which, depending on preconceptions, may or may not be the Saraswati of the Vedas) were coming up. We also know that 2,600 years ago, the Indo-Gangetic region of north India was divided into 16 republics populated by Indo-European language speakers. But there is little clarity on what transpired during the 1,000 years between the decline of the Harappans and the rise of urban life in the Indo-Gangetic region. We call it the Vedic Age, but that only suggests that the Vedas were composed sometime during this long period, and became the most important part of an oral tradition. Archaeological remains, such as pottery, at the upper reaches of the Ghaggar-Hakra and the Indo-Gangetic plain indicate settlements that in their geographical spread and antiquity lead up to the republics. But we have no clear knowledge of the evolution of culture and institutions from the end of the Harappans to the beginning of the republics.

The natural question is: how do we explain the vast extent of Indo-European speakers spread over much of north and east India some 2,600 years ago? Two possibilities suggest themselves, both of which rely on the fact that, as scholars have noted, the Vedas contain descriptions of geography that seem to match Punjab and north-west India, and were thus probably composed by people who were familiar with that region. The first possibility is, of course, that the Harappans were not just themselves speakers of an Indo-European language, but also the composers of the Vedas, and as the Harappan civilisation declined, the population moved up along the Ghaggar-Hakra and down into the Indo-Gangetic plain. The second is that nomadic Indo-European speakers, who were the composers of the Vedas, intermingled with the descendants of the Harappans to give rise to the later urban civilisation of the Indo-Gangetic plain.

The problem with the first scenario is that the Vedas include extensive details, such as of wealth measured in herds of cattle, which indicate a nomadic pastoral culture; they reflect little or nothing of the sophisticated urban civilisation of the Harappan people. Moreover, the repeated mention of horse-drawn chariots in the Vedas suggests a culture well versed with the horse; but no horse remains have been found at Harappan sites, even though other animal remains have been found in plenty. To argue that the Harappans composed the Vedas would suggest that the Vedas are the first works of fiction in known human history—successful attempts by residents of a sophisticated urban culture at passing themselves off as hard-fighting, soma-drinking pastoral nomads.

This leads to the second scenario—that the Indo-European speakers who composed the Vedas came from the region that is now Punjab. It also prompts further questions. How long did these people live in Punjab? Were they migrants into Punjab, perhaps even invaders? Did they mingle with the Harappans, and, if so, when? Though there is much debate about the date of the composition of the Vedas, if they do indeed reflect the geography of Punjab, where some Harappan settlements were situated, it is reasonable to assume they were composed after the decline of the Harappans—otherwise, one would need to find convoluted explanations for the absence of any mention of Harappan cities and towns. This dates the composition of the Vedas to a period after 3,500 or so years ago. The intermingling of the Vedic people and the descendants of the Harappans seems to have happened after this point. The subsequent centrality of the Vedic corpus and the evolution of language from Vedic Sanskrit onwards suggest this happened in a manner where the power equation was weighted in favour of the Vedic people.

So who were the Vedic people? Right-wingers who want to claim that nothing of worth came to India from outside it have enormous trouble with the notion that the Vedic people were migrants to Punjab (though even they cannot debate that modern humans are migrants to the subcontinent from Africa). The Harappans, as we have seen, could not have composed the Vedas. This leaves the possibility, remote but not ruled out, that a section of the Harappans abandoned their cities for nomadic life as their civilisation declined. This unlikely scenario would require that the Harappans spoke Vedic Sanskrit.

Linguistic evidence, by researchers such as the Indologist Edwin Bryant, suggests that Vedic Sanskrit was closely related to the language of the Zend Avesta—the sacred text of Zoroastrianism. It further suggests that this Indo-Iranian language was an offspring of the proto Indo-European language. This, in turn, suggests that Vedic Sanskrit evolved from a language that travelled to Punjab.

This conclusion is being reinforced by the genetic study of human origins and migrations. But while examining genetic evidence, a word of caution is in order: there is not always a link between genetic origins and language. The Bhils of central India, whose genetics indicate they are closely related to most Dravidian speaking people, speak an Indo-European language, while the Brahui of Balochistan, who genetically resemble their Indo-European-language-speaking neighbours, speak a Dravidian language.

But genetic and linguistic evidence, combined with archaeological evidence, allows the piecing together of a more complete and coherent story. This evidence, published in journals such as Nature, seems to suggest that the origins of all Indo-European speakers may well lie in the Yamna culture, found in the Pontic steppe some 5,500 years ago. The study of this population’s dispersal is now making rapid progress owing to recent advances in the study of the genetics of the male line of descent in humans.

Males inherit the Y chromosome from their fathers, so the study of the genetics of the Y chromosome allows an understanding of the paternal lineage (tracing a father’s father’s father, and so on) while the DNA in the mitochondria (mtDNA), the energy-producing body of the human cell, is exclusively inherited from the mother, allowing an understanding of the maternal lineage (a mother’s mother’s mother, and so on).

Studies of the matrilineage of Indian populations have shown that there has been very little external influence on Indian maternal lines of descent over the last 10,000 years. The story with the male lineage is different. Tests on several sections of the Indian population have shown the presence—particularly among speakers of the Indo-European languages—of the pattern of male Y-DNA that is labelled R1a, one of the patterns linked to the Indo-European dispersals from the steppes.

Much of the linguistic, archaeological and textual evidence suggesting an Indo-European migration had been thrown in doubt when early study of the R1a lineage suggested that its occurrence in India traced back to earlier than the last 10,000 years. Subsequent work that has refined our understanding of R1a has virtually turned this conclusion on its head, as has been well documented in a recent article in The Hindu, which, expectedly, aroused the ire of right-wing commentators.

Changes in DNA accumulate sequentially, and almost all the R1a in India is tagged with a change termed Z93, while much of the European R1a is tagged with Z282. DNA studies make it clear that neither is descended from the other, and that the separation between the two seems to have occurred around 6,000 years ago, in all likelihood outside India, on the steppes. Subsequently, the Z93-tagged population underwent rapid expansion beginning some 4,000 years ago.

As David Wesolowski, an Australian blogger who tracks research into genetic origins, has hinted, the ancient DNA from Rakhigarhi could well suggest a genetic linkage between the Harappans and the speakers of the Dravidian languages today, and would begin clarifying the story of the Vedic people as well as provide the first glimpse of who the Harappans really were.

A 2009 paper in Nature had suggested that the Indian population could be considered a mixture of two theoretical constructs, an “Ancestral North Indian”—ANI—population closer to west Asians, central Asians and Europeans, and an Ancestral South Indian—ASI—population, distinct from any other population in the world. Unfortunately, while the two populations were mathematical constructs, the names caused some confusion because they seemed to suggest specific groups of people. This important work has since been expanded and elaborated on, and it is clear today that the ANI component and ASI component in the Indian genome represent a cluster of different groups mixing over a period of time. They certainly do not suggest that the Vedic people and the Harappans correspond to two populations whose mixture completely explains our current population distribution.

Given the long existence and spread of the Harappans, it would certainly seem likely that their genes should show a very varied heritage. What they are unlikely to show is an R1a Z93 component. Recent data from a study in the journal BMC Evolutionary Biology suggest that the inward migration of a largely male population marked by Z93 Y-DNA took place in several diverse but related streams, with the earliest arriving roughly 3,400 years ago.

The most reasonable scenario that fits this evidence is probably a slow decline and end of the mature Harappan civilisation some 4,000 years ago, likely due to environmental reasons, and a movement of people along the Ghaggar-Hakra and down the Yamuna to much smaller settlements that would hardly be in any position to offer resistance to incomers. These are the settlements the migrating or invading Vedic people are likely to have encountered, and they would have had an upper hand in any interaction with them. An elite replacement would dovetail nicely with a later urban expansion in the Indo-Gangetic plain that would take its sacred texts and rituals largely from the Vedic people while incorporating many of the civilisational elements that survived the decline and end of the Harappans.

On the other hand, an unexpected result from the ancient Harappan DNA could send everyone back to the drawing board, which would only make the study of ancient India even more interesting and enigmatic than it already is. What is clear is that over the next few years we should be arriving at a far greater understanding of two populations (if they are indeed two) that have greatly shaped many aspects of our social inheritance. Only those who want the past to be a product of their present ideological beliefs would look to such a possibility with apprehension.