Where did our an­ces­tors come from? This can be a highly emo­tional—and po­lit­i­cal—is­sue. It is clearly an im­por­tant ques­tion for many peo­ple and one that mod­ern ge­net­ics has an­swered to a great ex­tent. In re­cent months, a sci­en­tific pa­per pub­lished in the jour­nal BMC Evo­lu­tion­ary Bi­ol­ogy has sparked a heated con­tro­versy in the In­dian me­dia by out­lin­ing an ‘Indo-Euro­pean ex­pan­sion, with an ul­ti­mate source in the Pon­tic-Caspian re­gion’ into the In­dian sub­con­ti­nent.

Be­hind the gen­tly ar­cane schol­arly lan­guage, the pa­per ar­gues that the ge­netic ances­try of all mod­ern In­di­ans dis­plays ev­i­dence of sig­nif­i­cant mix­ing with pop­u­la­tions that moved to the sub­con­ti­nent from north­ern Iran and the Caspian re­gion some 4,000-5,000 years ago. Tem­pers are fray­ing be­cause these find­ings have reignited a long-sim­mer­ing and highly politi­cised de­bate about the an­cient ori­gins of the Vedic ‘Aryans’. For decades, his­to­ri­ans, lin­guists and ar­chae­ol­o­gists have de­bated the re­la­tion­ship of the Aryans to In­dia, with lit­tle res­o­lu­tion—in­deed the ar­gu­ment has been sub­sumed by its im­pli­ca­tions for the con­tem­po­rary bat­tles be­tween the hyper-na­tion­al­ist pol­i­tics as­so­ci­ated with the Sangh pari­var and the sec­u­lar lib­eral op­po­si­tion. Many Hindu na­tion­al­ists are un­com­fort­able with the idea that the an­cient roots of the In­dian peo­ples may lie out­side the sa­cred geog­ra­phy of the sub­con­ti­nent. Mean­while, there is of­ten an el­e­ment of lib­eral schaden­freude in em­brac­ing the nar­ra­tive that sug­gests a par­al­lel be­tween the Vedic Aryans as con­quer­ing ‘in­vaders’, not un­like the later vis­i­ta­tions of Is­lamic and Euro­pean em­pires.

Over the last few years, ge­net­ics has be­gun to of­fer its own find­ings, which much more defini­tively in­di­cate that a peo­ple—who may have been Aryans—moved into the sub­con­ti­nent ap­prox­i­mately 4,000-5,000 years ago. Soon, the field of an­cient DNA re­search may close the case. The BMC pa­per that has drawn so much in­ter­est and ire is just the lat­est in a line of re­search that goes back decades, and grows more pre­cise and in­sight­ful with ev­ery new tech­no­log­i­cal ad­vance. By look­ing at pat­terns of ge­netic mark­ers in mod­ern hu­mans, ge­neti­cists have been able to sketch the fam­ily tree of our species. Re­searchers are also test­ing ge­netic ma­te­rial from re­mains tens of thou­sands of years old. It has al­ready high­lighted the Ne­an­derthal her­itage ex­tant to­day in all hu­mans out­side of Africa.

Be­cause ge­netic sci­ence has been driven by US-based re­searchers, bi­ases have crept into the sort of ques­tions asked. But the democrati­sa­tion of the field, due to a sur­feit of data, is now en­abling ex­plo­ration of more di­verse top­ics, in­clud­ing ques­tions re­lated to the In­dian sub­con­ti­nent.

Euro­cen­tric ide­olo­gies have spawned their own counter-the­ses. While, in 1903, the In­dian na­tion­al­ist Bal Gan­gad­har Ti­lak could write The Arc­tic Home in the Vedas, in keep­ing with the mi­gra­tionist be­liefs of the era, to­day, In­dophilic western­ers such as Koen­raad Elst are pro­mot­ing an ‘Out of In­dia’ the­ory, an in­ver­sion of the older route.

These ar­gu­ments ex­ist out­side of pol­i­tics and na­tion­al­ism. The au­thor San­jeev Sanyal con­tends that “the idea of a uni­di­rec­tional ‘Aryan’ in­va­sion or mi­gra­tion around 1500 BC is now con­clu­sively proven to be wrong”. As out­lined in his book, The Ocean of Churn, Sanyal em­pha­sises the re­cip­ro­cal move­ments of peo­ples. He notes that ar­chae­ol­ogy does not tie any In­dus Val­ley civil­i­sa­tion site to Cen­tral Asia, and the Vedas them­selves seem ig­no­rant of geog­ra­phy out­side of South Asia.

Ten years ago, one could rea­son­ably sup­port Sanyal’s sup­po­si­tions from a ge­netic per­spec­tive. The ex­plo­ration of mi­to­chon­drial lin­eages, the di­rect ma­ter­nal ances­try out of Africa, re­mained the dom­i­nant method of in­fer­ence. That line of ev­i­dence strongly sug­gests that South Asian pop­u­la­tions are deeply rooted in the sub­con­ti­nent.

Other re­searchers were look­ing at the de­scent of males, the di­rect pa­ter­nal lineage as recorded by mu­ta­tions of the Y chro­mo­some. The ev­i­dence from these re­sults was more equiv­o­cal. One of the

more com­mon South Asian Y lin­eages, R1a1a, is also very com­mon in East­ern Europe and Cen­tral Asia. The dis­cov­erer of this lineage, the ge­neti­cist Spencer Wells, says that his work “in the late 1990s strongly sup­ported a sig­nif­i­cant mi­gra­tion from the steppes of East­ern Europe and Cen­tral Asia into In­dia in the past 5,000 years”. Wells con­nected this to Indo-Euro­pean speak­ing no­mads, and be­lieves the lat­est re­sults have borne that out. Ob­vi­ously, this is in con­flict with the mi­to­chon­drial re­sults. Be­cause R1a1a is not very di­verse, it was dif­fi­cult to get a sense of where or when it may have orig­i­nated. Many re­searchers, con­tra Wells, con­tended that R1a1a may have been indige­nous to South Asia.

To­day, we know more about R1a1a than we did in the 2000s. Whereas then re­searchers looked at a few hun­dred mark­ers on the Y chro­mo­some, or per­haps some re­gions with very high di­ver­sity, to­day they can se­quence most of the Y chro­mo­some.

In line with Wells’ orig­i­nal sus­pi­cion, af­ter look­ing at whole genomes, many schol­ars now sur­mise that R1a1a en­tered South Asia within the last 4,000-5,000 years from the Eurasian steppe. The rea­son R1a1a is not di­verse is that it un­der­went a mas­sive, re­cent ex­pan­sion; not much time has elapsed for mu­ta­tions to ac­cu­mu­late. With whole genome anal­y­sis, one can see that East Euro­pean R1a1a is one lineage, while Cen­tral Asian and South Asian R1a1a strains form an­other. Martin Richards, co-au­thor of the pa­per in BMC, ex­plains, “This high res­o­lu­tion al­lows for both very de­tailed ge­nealog­i­cal in­for­ma­tion and quite pre­cise ge­netic dat­ing—so we can see where and when lin­eages branch off into a new ter­ri­tory.”

An­cient DNA has also shed light on the re­la­tion­ship of the var­i­ous branches of R1a1a. Ex­tinct Cen­tral Asian steppe pas­toral­ists, the Scythi­ans, and their ge­o­graphic kin, the Srubna peo­ple, who dwelt north of the Caspian Sea 3,750 years ago,


also carry this Y lineage. It is no­table that the R1a1a lin­eages of Scythi­ans and Srubna are the same as Cen­tral Asians and South Asians, not Euro­peans.

A res­o­lu­tion to the para­dox of mtDNA and Y chro­mo­so­mal lin­eages point­ing in dif­fer­ent di­rec­tions has now pre­sented it­self. The an­swer: mi­gra­tion into In­dia was not sex bal­anced. This is a ma­jor point in the BMC pa­per. Richards states, “There’s a very marked sex bias in the ar­rival of new peo­ples from, ul­ti­mately, the steppe zone of east­ern Europe, in the Bronze Age.” (This is the case in Europe too, with steppe mi­grants over­whelm­ingly male.)

Our un­der­stand­ing to­day does not rest on Y chro­mo­some and mtDNA alone. With whole genomes avail­able for anal­y­sis, sci­en­tists have re­shaped our un­der­stand­ing of the past of South Asia. In 2013, ge­neti­cist Priya Moorjani and col­leagues pub­lished re­search con­clud­ing that at least two very dis­tinct pop­u­la­tions were mix­ing in the In­dian sub­con­ti­nent 2,000-4,000 years ago. Moorjani says that “4,000 years ago, there were unmixed ANI and ASI groups in the In­dian sub­con­ti­nent”.

ANI and ASI are acronyms for An­ces­tral North In­di­ans and An­ces­tral South In­di­ans. The for­mer pop­u­la­tion was ge­net­i­cally very sim­i­lar to Near Eastern­ers and Euro­peans. One of the orig­i­nal re­searchers who de­vel­oped this model of In­dian ori­gins, Nick Pat­ter­son, char­ac­terises the ge­netic dis­tance be­tween ANI and Euro­pean pop­u­la­tions as so small that if you did not know of the prove­nance you might say it was a Euro­pean pop­u­la­tion. The ASI, in con­trast, do not have any close rel­a­tives. Rather, they were dis­tant rel­a­tives of indige­nous An­daman is­lan­ders. Moorjani re­it­er­ates that though there were unmixed pop­u­la­tions rep­re­sen­ta­tive of these groups in the rel­a­tively re­cent past, to­day all na­tive South Asian groups dis­play ances­try from both.

In Moorjani’s 2013 pa­per, she es­ti­mated that Dal­its from Tamil Nadu are 40 per cent ANI (the re­main­der pre­sum­ably be­ing ASI). Pathans are 70 per cent ANI. Kash­miri Pan­dits are 65 per cent ANI, while Brah­mins and Ksha­triyas from Ut­tar Pradesh are 55-60 per cent ANI. When it comes to the mix of ANI and ASI, there are two rules of thumb one needs to con­sider. The fur­ther northwest you go, the more ANI you will get. Up­per castes have more ANI ances­try as well (Ben­galis and Munda tribes have East Asian her­itage that is nei­ther ANI nor ASI). This is ex­actly the pat­tern you would ex­pect from Y chro­mo­so­mal lin­eages such as R1a1a, which many ge­neti­cists posit have ar­rived from Cen­tral Asia in the last 4,000-5,000 years.

In­dian ob­servers of his­tor­i­cal pop­u­la­tion ge­net­ics have noted these find­ings and in­te­grated them into their own un­der­stand­ing. Sanyal says he be­lieves “In­di­ans are the mix of sev­eral ge­netic streams, par­tic­u­larly the ANI and ASI who have been liv­ing in the In­dian sub­con­ti­nent from the Stone Age”. Moorjani states that her “study did not specif­i­cally look into the di­rec­tion of mi­gra­tion”. But, she also ad­mits that “the di­rec­tion of mi­gra­tion lead­ing to ANI is prob­a­bly into In­dia”.

Why would Moorjani state this? First, let us take a step back and ad­dress two points of the Out of In­dia frame­work. As Sanyal ob­served, the ar­chae­o­log­i­cal con­nec­tions be­tween South Asia and Cen­tral Asia are ten­u­ous. The Aryans do not seem to rec­ol­lect a time be­fore In­dia. But the past 10 years of dis­cov­er­ies of an­cient DNA have shown us there were mass mi­gra­tions where ar­chae­olo-


gists sus­pected none. The phys­i­cal record is in­com­plete, and it may be dif­fi­cult to con­nect with the broader pat­terns of his­tory. But part of it is that some pop­u­la­tions, such as no­mads, likely do not leave much of an ar­chae­o­log­i­cal foot­print. As for the ar­gu­ment based on In­dian re­li­gious and oral his­tory, it must be ob­served that the Greeks also do not have any mem­ory be­fore Greece, and yet they are just as an­tique an Indo-Euro­pean peo­ple. The ar­gu­ment about cul­tural mem­ory alone can­not be trusted to ad­ju­di­cate on this mat­ter.

Which brings us to what could ul­ti­mately re­solve un­cer­tain­ties: an­cient DNA. The test­ing of sam­ples from the Near East and Europe over the pe­riod be­tween 5,000 and 10,000 years ago has shown that a few pulses of mi­gra­tion mixed to­gether to cre­ate the pre­dom­i­nant ge­netic pat­terns we see around us to­day. We do have a rea­son­able sam­pling of an­cient in­di­vid­u­als from the Near East, Cen­tral Asia and Europe, and what we see are mas­sive pop­u­la­tion changes over the past 10,000 years. Should we ex­pect In­dia to be any dif­fer­ent?

For the pur­poses of an un­der­stand­ing of the South Asian ge­netic land­scape, two an­cient pop­u­la­tions from West­ern Eura­sia share strong affini­ties with peo­ple from the sub­con­ti­nent. First, the ear­li­est farm­ers of West­ern Iran, in the Za­gros, whose her­itage is now found all across Eura­sia, evince high affini­ties with many In­dian pop­u­la­tions. Sec­ond, Cop­per Age pas­toral­ists of the Yamna cul­ture of the Pon­tic steppe, who flour­ished 4,000 to 5,500 years ago, also ex­hibit a strong affin­ity to South Asians, in par­tic­u­lar pop­u­la­tions from the northwest and up­per caste groups such as Brah­mins.

Re­searchers in David Re­ich’s lab at Har­vard have tested what pos­si­ble groups could be com­bined to cre­ate the ANI el­e­ment in South Asians. Af­ter ex­haus­tive com­par­isons, they find ANI is best mod­elled as a com­bi­na­tion of the Pon­tic pas­toral­ists and early Ne­olithic Ira­nian farm­ers!

In The Ocean of Churn, the the­sis is pre­sented that ideas and peo­ple move in a bidi­rec­tional fash­ion. In­dian re­li­gious and philo­soph­i­cal ideas did im­pact the West through Pythago­ras and Plato. Con­versely, many In­dian al­pha­bets quite likely have their ori­gins in the Near East, while Chris­tian­ity and Is­lam have both taken root in the sub­con­ti­nent.

So too with genes. South Asian ge­netic mark­ers are found in South­east Asia, from Thai­land to Bali. Con­versely, Ben­galis, As­samese and Munda peo­ples show their South­east

Asian her­itage on their faces, their genes, and in the case of the Munda, their lan­guages. But this idea of ubiq­ui­tous gene flow has lim­i­ta­tions.

The dis­tinc­tive ge­netic her­itage of In­dia, the ASI com­po­nent, deeply rooted in the sub­con­ti­nent and not closely re­lated to pop­u­la­tions else­where, ex­ists in low pro­por­tions in Iran and Afghanistan. But with the ex­cep­tion of the Roma peo­ple, ASI ances­try is notably ab­sent through­out West­ern Eura­sia aside from In­dia’s near neigh­bours. This sug­gests there has been very lit­tle west­ward move­ment out of In­dia over the past few thou­sand years be­cause, as Moorjani ob­served, all In­dian pop­u­la­tions have ASI ances­try within the past 4,000 years.

Many ge­neti­cists now be­lieve a ma­jor mi­gra­tion of peo­ple from Cen­tral Eura­sia and West Asia into South Asia dur­ing the Ne­olithic and Cop­per age is the sim­plest and most par­si­mo­nious model to ex­plain the data we have. The Out of In­dia model is not the­o­ret­i­cally im­pos­si­ble, but it strikes many as far-fetched and a stretch to ex­plain the pat­tern of the ac­cu­mu­lated data.

To move beyond prob­a­bil­i­ties, we need to make re­course to what has res­cued us in the past: an­cient DNA. Un­for­tu­nately, there is cur­rently no an­cient DNA data from South Asia proper. Even now, re­searchers are try­ing to get ge­netic re­sults from sam­ples at Rakhi­garhi in Haryana dat­ing to the Harap­pan pe­riod. Sup­po­si­tion can quickly be re­placed with cer­ti­tude when we sam­ple these in­di­vid­u­als. in­dia to­day learns that re­sults of the Rakhi­garhi sam­ples will be an­nounced in early Septem­ber this year. Dr Vas­ant Shinde, an ar­chae­ol­o­gist at Dec­can Col­lege, Pune, which con­ducted this project in col­lab­o­ra­tion with ge­neti­cists from Seoul Na­tional Univer­sity, is un­der­stand­ably re­luc­tant to of­fer any point­ers as to what the Rakhi­garhi sam­ples sug­gest. “It’s very po­lit­i­cally sen­si­tive,” he says. But given the fact that the graves from which DNA was ex­tracted were dated to some­where be­tween “2300 and 2500 BC”—the same pe­riod in which Martin Richards and his col­leagues sug­gest a pulse of mi­gra­tion from the Pon­tic-Caspian re­gion into In­dia—it must re­main a pos­si­bil­ity that Rakhi­garhi will yield R1a1a DNA and not set­tle the de­bate.

And yet this is just one site. There are hun­dreds of sam­ples for Europe and the Near East, and from those hun­dreds we have gleaned star­tling re­sults. At some point, there will be hun­dreds of sam­ples from South Asia, and there is no doubt we will glean some fas­ci­nat­ing re­sults.

The tide in his­tor­i­cal pop­u­la­tion ge­net­ics has turned to­wards mi­gra­tion, but some still hold to the model of con­ti­nu­ity dom­i­nant in the 2000s. Gyaneshwer Chaubey has been pub­lish­ing and re­search­ing hu­man ge­net­ics for over 10 years, with a sub­stan­tial con­tri­bu­tion in the area of In­dian pop­u­la­tion his­tory. He is not per­suaded by the hy­poth­e­sis of a mass Aryan mi­gra­tion. Rather, he ob­serves that pub­lished re­search has shown that “the ge­netic im­print of this mi­gra­tion (if we want to main­tain any) is min­i­mal”.

With­out an­cient DNA, we can only per­ceive broad coarse out­lines from the vari­a­tion of liv­ing hu­man be­ings; frag­ments of the past, re­ar­ranged and re­con­structed us­ing sta­tis­ti­cal frame­works and data from peo­ple alive to­day. Our con­jec­tures have as­sump­tions. Chaubey does not deny the data show­ing Ne­olithic Ira­nian farm­ers and Cop­per Age Pon­tic pas­toral­ists hav­ing ge­netic sim­i­lar­i­ties to mod­ern In­di­ans. He ar­gues for an equal prob­a­bil­ity that groups that “lived in In­dia in the Bronze Age or in Ne­olithic time hav­ing quite sim­i­lar ances­try as the Steppe belt pop­u­la­tions or Ne­olithic Ira­ni­ans…” By re­fram­ing as­sump­tions, he also dis­agrees with the re­vi­sion in re­gard to the his­tory of R1a1a. He ad­mits that R1a1a was the pri­mary rea­son he took him­self off the BMC pa­per. Though un­con­vinc­ing to many, Chaubey’s ra­tio­nale does have a ba­sis in the­ory and data. The dis­agree­ment is on the mat­ter of prob­a­bil­i­ties. This is not un­com­mon in sta­tis­ti­cal ge­net­ics. An­cient DNA should re­solve many of these dis­putes rather soon, but this is­sue is overly politi­cised.

We do know some things. Ge­neti­cists have con­firmed di­ver­gences of caste and re­gion in the genomes of In­di­ans. A great deal of mix­ing seems to have oc­curred over the past 4,000 years. Be­fore that pe­riod, much of the sub­con­ti­nent was in­hab­ited by peo­ple ge­net­i­cally very dif­fer­ent from those alive to­day. By co­in­ci­dence—or per­haps not—the ex­tremely com­mon R1a1a pa­ter­nal lineage, which binds many In­dian men to Euro­peans and Cen­tral Asians, be­gins to ex­pand rapidly just as the last ma­jor mix­ing event in South Asia be­tween ANI


and ASI lin­eages oc­curred.

Prom­i­nent pop­u­la­tion ge­net­ics lab­o­ra­to­ries that have re­shaped our un­der­stand­ing of the his­tory of Europe and the Near East through an­cient DNA stud­ies are now look­ing to In­dia. It won’t be long be­fore new tools are brought to bear on old con­tentious ques­tions. One can no longer say that the the­sis that Indo-Aryans ar­rived in the sub­con­ti­nent in large num­bers is re­futed. Con­nec­tions to an­cient groups out­side of the sub­con­ti­nent seem highly likely, and many prom­i­nent ge­neti­cists are now pro­mot­ing a view­point pred­i­cated on mi­gra­tion and pop­u­la­tion turnover, which seems to have been the norm in Europe and the Near East.

Yet there re­main cred­i­ble sci­en­tists at prom­i­nent in­sti­tu­tions who are scep­ti­cal of the new mod­els. While a new con­sen­sus may be emerg­ing, it has not crys­tallised. Ge­net­ics can speak in broad strokes with er­rors of the or­der of a thou­sand years here or there, and each new dis­cov­ery re­fines de­tails of the broader pic­ture. We are in a time of sig­nif­i­cant tran­si­tion and great in­tel­lec­tual fer­ment. That fer­ment ex­tends to pol­i­tics. Just as the orig­i­nal Aryan in­va­sion model was pro­moted on a po­lit­i­cal ba­sis, so are Out of In­dia the­o­ries given po­lit­i­cal va­lid­ity. But the re­al­ity is that these mod­els of hu­man his­tory are ei­ther true—or they are not. Whether or not they were dis­cernible, the facts have al­ways been with us. It is our po­lit­i­cal in­ter­pre­ta­tion of them that seems to change. Hu­mans are pro­tean, but na­ture is time­less.

We see through a glass darkly. In a few years, it may be crys­tal clear that a new peo­ple ar­rived in the In­dian sub­con­ti­nent 4,000 years ago. That now seems to be the be­lief among the ma­jor­ity of prom­i­nent re­searchers. A cen­tury of the­o­ris­ing and ide­ol­o­gis­ing has armed us with an­swers and ob­jec­tions, but his­tory as un­veiled by ge­net­ics may hold some brac­ing sur­prises for our rigid grandiose pre­ten­sions. That may be the most ex­cit­ing as­pect of these lines of re­search, not how they align with cen­tury-old ar­gu­ments.

Razib Khan is a ge­neti­cist with an in­ter­est in pop­u­la­tion his­to­ries and per­sonal ge­nomics. He works at In­sit­ome and is a Ph.D can­di­date at UC Davis. He writes the blog ‘Gene Ex­pres­sion’



The 4,500-year-old skele­tons dis­cov­ered in Rakhi­garhi, Haryana

Newspapers in English

Newspapers from India

© PressReader. All rights reserved.