The Fair-Skinned Sintashta folks did not spread Indo-Aryan languages in India

I haven’t written a post on the blog since July 2020. This has partly been due to a slightly more busy schedule and partly due to the fact that I have become a little ambitious and am trying to write more comprehensive and voluminous articles which take time to prepare. After having written about the dating of the Kushans on my blog, I delved into the linguistics aspect of Indo-European origin. Having been only partly through the subject, I came across a recent paper on the Podolian cattle of southern and eastern Europe, which are distinguished by their significant levels of Indicine admixture.

I decided to write a more lengthy and detailed article on the migration of Indicine cattle westwards from India but as I went about gathering the data, it dawned on me that there was a great amount of archaeological research, most of which has come up in the last few years, that can be marshalled to make a comprehensive case for OIT. So while keeping the focus on the Zebu migration, I am now making a case of how nicely it ties with an OIT scenario. As things stand, the preliminary draft of my article/paper is already quite big with more than 25k words but then I have managed to stumble on some more important data from Bronze Age Europe which I could not neglect. Adding this to the existing draft will likely enlarge the text by a few thousand more words. Once I am able to complete a legible final draft I am planning to get it published somewhere, God willing, on an online platform. Lets see how it goes.

To just give an idea, let us note that cow was a very important animal for Indo-European people. This Anglo-Saxon guy gives you some good understanding on the topic,

Here is a map from that recent Senczuk et al preprint,

You can see that in a large number of native southern and eastern European cattle, there is significant levels of Indicine Zebu admixture. These regions are Indo-European speaking and includes speakers of Slavic, Italian, Greek and Albanian languages. Now if cow was a very important animal for Indo-Europeans, it is likely that they must have taken it along with them in their migrations. Is it not therefore quite noteworthy, that it is the genetic ancestry of the Indian origin Zebu that unites the cattle of all these Indo-European people ? Is it just co-incidence ?

Moving on, recently, Razib came up with the assertion that a relatively fair-skinned group of people with their origins on the steppe, somewhere around Sintashta, are likely the people who spread the Indo-European culture and language in the Indian subcontinent. He also implied that these ‘fair-skinned’ steppe migrants must have perceived the native people of the Saraswati-Sindhu Civilization as dark-skinned. So essentially, the racist colonial trope of fair-skinned Aryans vanquishing the dark-skinned native Dravidians may be true. This is a very fanciful flight of imagination and it is necessary to show just how groundless it is.

The most popular theory for the origins of Indo-Europeans, is the Pontic-Caspian steppe homeland theory. However, being popular has little to do with being true. David Anthony and James Mallory, both American archaeologists, are today the two most prominent proponents of this theory. Here is what James Mallory says, in a book he co-authored with the linguist D Q Adams,

All too often surveys of the Indo-Europeans eventually conclude with something on the order of ‘scholars have concluded that the most likely area of the homeland is . . .X’ with a brief defence of one particular solution (this type of scholarship has been going on since the late nineteenth century). In fact, we not only lack total consensus but where we seem to find something of a major school it is often formed by deference rather than conviction, i.e. linguists or archaeologists indicate agreement with a particular theory that they have not themselves investigated in any depth. This situation means that a small number of advocates—at times, very vigorous advocates—provide an assortment of homeland theories for the rest of their colleagues to comply with passively. The homeland is an interesting question but it is so difficult to resolve (we have over two centuries of dispute to prove that) and requires the application of so many less than robust means of argument that most archaeologists and historical linguists do not find it a worthwhile enterprise, at least for themselves. The last word is, therefore, far from written…

So, one of the major proponents of the Steppe hypothesis is himself admitting that most academics acquiesce passively with the IE origin theory without having come to the conclusion by any significant research on the topic themselves. There are infact weighty reasons to question the steppe origin hypothesis as we shall see.

  • Lack of Indisputable proof of Indo-European languages on the steppe before 1000 BCE.

The earliest evidence of an Indo-European language or of Indo-European speakers on the steppe comes from around the 9th century BCE when the Cimmerians (likely Iranian speakers), inhabited the steppe region north of the Caucasus and the Black Sea.  This too, is based on the record of Herodotus. There is no inscriptional or literary evidence in their own language that directly attests it.  The Balto-Slavic branch of Indo-European languages are today the only IE languages found in and around the proposed Pontic-Caspian homeland of Indo-Europeans. Yet the earliest attestation of a Slavic language on the steppe or anywhere is not earlier than the 6th century CE while the earliest attestation of a Baltic language comes from the 14th century CE. THESE ARE FACTS.

With such a state of affairs, how can you argue that, in a region like the steppe, which is in constant flux, and which has seen language turnovers many times, that Indo-European languages originated around 4000 BCE, when there is no hard evidence of any IE language speaker there before 1000 BCE ? How can we assume that for more than 3000 years from 4000 BCE to 1000 BCE, people on the Pontic-Caspian steppe and its surrounds were speaking Indo-European languages, when there is zero attestation of an IE language during this long period on the steppe ?

It is still understandable if one were to argue that IE groups existed from around 15th century BCE on the steppe, since it is temporally much closer to the first attestation of the Cimmerians a few centuries later. But to suggest that the IE languages were continuously spoken on the Pontic-Caspian steppe from around 4000 BCE right down to Cimmerians more than 3000 years later, without a shred of evidence in terms of attestation of an IE language there is quite incredible. I find it hard that people are so happy and eager to lap up and propogate this shoddy theory.

  • Earlier attestation of Indo-Iranian languages in South Asia and Near East than on the steppe

As we saw, the earliest evidence of an IE group on the steppe comes around 9th century BCE. This is infact, several centuries later than the arbitrary dating of the Rigveda around 1200 BCE and the presence of Indo-Aryan words among the Mitanni in the 14th century BCE. And mind you, the Mitanni Indo-Aryan elites, land up in Syria with Indian elephants and Indian humped cattle.

According to Chakirlar & Ikram,

In Southwest Asia, the earliest representations of elephants appear in art and mythological literature, originating from eastern Lower Mesopotamia, and date to the end of the 3rd millennium BC (Potts 1997: 260–61). The style of depiction, though, seems to derive from that of the Indus Valley (Salonen 1976: 146–47). This strongly suggests a second-hand knowledge of elephants, rather than first-hand, real-life experience. From Greece to Arabia, no single reference to, or depiction of, an elephant or elephant parts, ante-dates these first finds from the end of the 3rd millennium BC… the Holocene elephants of Southwest Asia were not endemic to the region and that the Early Bronze Age peoples of the region knew about them only through their contact with India, or possibly Egypt. The latter is less likely as these animals were no longer indigenous there by that time, although remembered… Secondly, ancient accounts indicate that live elephants roamed and were hunted in the Orontes Valley, the Upper Euphrates Valley and the Middle Euphrates Valley around modern Ana in Iraq, at least between the end of the 16th and 9th centuries BC, possibly into the 8th century BC (Breasted 1906–07; Gardiner 1964: 179, 201; Moorey 1994: 117; Scullard 1974: 28). The core of this region comprises the area of influence of the Mitanni Kingdom, the main local political player in LBA northern Syria… Based on all the evidence reviewed above, and in the absence of fossil evidence, we also support the hypothesis that the Syrian elephant was not endemic, but arrived in Southwest Asia later in the mid-Holocene as an import from Southeast Asia that took hold locally.

Is it also just another co-incidence that the Mitanni also happen to be Indo-Aryan speakers, a language group dominant across much of North India where these Bronze Age Elephants in Mitannian Syria came from ?

The actual date of the Rigveda, on the other hand, is likely to be much earlier to 1200 BCE and probably before 2000 BCE, when the mightly river Sarasvati of the Rigveda had already dried up.  The Rigvedic geography extends from Eastern Afghanistan to Western UP.

Even the Avesta, is older than 1000 BCE, and its geography does not extend further north than the Bactrian region of North Afghanistan. So, we have the attested presence of Indo-Iranians in South Asia more than a millenia before their attestation on the steppe.

This is not all. The linguistic diversity of the Indo-Aryan languages from the Middle Indo-Aryan period is such that the linguists admit, that these languages do not directly descend from Rigvedic Sanskrit but from its sister dialects. According to Thomas Oberlies,

The problem of the linguistic affinity of Pali and the other Middle Indo-Aryan (= MIA) languages is well-known and is undisputed: These languages are by no means straightforward continuants of the Old Indo-Aryan (= OIA) of the Vedic corpus, as in all of them words and forms turn up which cannot be the (regular) outcome of any atte- sted OIA ones…There are a number of words where Pali/Prakrit does not continue what we expect as the regular outcome of OIA. applying the MIA. sound laws. These words point either to the pre-Vedic language or (more probably) to (a) Vedic dialect(s) different from the dominant one.

OIA or Old Indo-Aryan refers to the language of early Vedic texts including that of Rigveda which is Vedic Sanskrit.

According to Claus Peter Zoller,

In the textbooks dealing with the history of Indo-Aryan, a linguistic family tree is drawn with OIA on top and ever new ramifications down to the modern languages, even though every specialist can tell that the different MIA languages (e.g. Pali) are not direct descendants of the Vedic corpusthere is no doubt that at the time of the immigration of Old Indo-Aryan into South Asia a whole bunch of Indo-Aryan dialects/variants existed.

Infact, it has been a long standing theory, going back to the 19th century, that the Indo-Aryan languages can be classed into an Inner Indo-Aryan (represented by Sanskrit and its descendents) and an Outer Indo-Aryan, where the peripheral Indo-Aryan languages in the IA language geography of South Asia are said to have several archaic features in common with other IE languages but not preserved in Sanskrit.

To account for the more southern and eastern geographical presence of these Outer Indo-Aryan in South Asia, the linguists have even proposed that these Outer Indo-Aryan speakers must have migrated into South Asia before the Inner Indo-Aryan speakers who composed the Rigveda.

In other words, when the Rigveda was composed, sometime around or before 2000 BCE, before the drying up of Sarasvati, there were already multiple Indo-Aryan languages present in South Asia, as per the admission of these linguists. Yet, on the steppe, you have no evidence of any IE language until the 9th century BCE and the single parent language of the Slavic languages cannot go further in time than the 6th century CE. So there is a gap of more than two millenium between the earliest attestation of Indo-Aryan languages in South Asia and the earliest attestation of Balto-Slavic languages on the steppe.

So on what grounds can you claim that the Sintashta folks, whose language, we have no clue about and who existed more than a millenia before the earliest attestation of any IE language on the steppe, are actually the precursors of Indo-Iranians in the Indian subcontinent ? Isn’t this stretching the bounds of credulity ? Even if, hypothetically, the Sintashta folks admixed with the people of the Indian subcontinent, it cannot in any way prove that it is these Sintashta folks who spread the Indo-Iranian languages in the region. You simply have no solid proof of what language they spoke. Presence of chariots (questionable) and weapons is not enough. Chariots were also used heavily by the Egyptians and the Minoans and even by ancient Chinese. Yet that does not make them Indo-European.

  • No proof of cultural intrusion from the steppe into South Asia

As I have already shown here before, there is no archaeological evidence of any steppe cultural marker penetrating into South Asia. As per James Mallory,

This is indeed the problem for both the Near Eastern and the Pontic-Caspian models and, following the logic of this analysis, the Bouckaert model appears to be in the same boat. All of these models apparently require the Indo European languages (including their attendant agricultural vocabulary) to be superimposed/adopted by at least several major complex societies of Central Asia and the Indus… In any event, all three models require some form of major language shift despite there being no credible archaeological evidence to demonstrate, through elite dominance or any other mechanism, the type of language shift required to explain, for example, the arrival and dominance of the Indo-Aryans in Indiaall theories must still explain why relatively advanced agrarian societies in greater Iran and India abandoned their own languages for those of later Neolithic or Bronze Age Indo-Iranian intruders.

According to Kristiansen et al. (Supplement A of Damgard et al),

contacts between Bronze Age steppe populations and NMG V and BMAC populations appears to have been one in which the dynamic of cultural influence was stronger on the side of the well-established sedentary food producing populations, and this resulted in the partial assimilation of these initial newcomers to the region both culturally and, to a lesser degree, biologically as well. Third, not all of those who emigrated from the north turned to farming but may have continued a semi-nomadic existence in the highlands, which were unsuitable for the kind of intensive farming practiced in the BMAC homelands or in the regions of Khorezm. Fourth, if there was any Central Asian influence on South Asian populations, that influence likely long predated any development of Iranian, let alone Indo-Aryan, languages, and most likely occurred during the late NMG IV to early NMG V period (ca. 2800–2300 BCE) and even earlier during the Eneolithic from Kelteminar culture groups (4000–3500 BCE).

In other words, not only is there no archaeological evidence of cultural intrusion from the steppe in South Asia but the steppe groups actually ended up getting assimilated into the sedentary agriculturalist society of the Oxus civilization. This is a very significant and fatal blow to the theory of steppe origins of Indo-Europeans. How do groups who get assimilated culturally, get to spread their language, culture and religion ?

In the absence of any solid linguistic attestation of an IE language on the steppe before 1000 BCE, it is merely the presence of the steppe cultural markers attested via archaeology which is taken as evidence of Indo-European presence or spread, as is the case in Europe with the corded ware culture. Now, if this steppe cultural marker did not intrude even in the settled agricultural settlements of Central Asia, what to speak of South Asia, how does the theory of steppe groups spreading Indo-Iranian languages into South and Central Asia, still hold ?

And this is not a minor slip for the PIE on steppe theory. Indo-Iranian languages makes up 311 of the 445 extant Indo-European languages spoken today. In other words, 70 % of all Indo-European languages spoken today are Indo-Iranian. They also constitute 50 % of the population of all IE speakers. Yet, the steppe theory cannot explain how these Indo-Iranians spread from the steppe into these present southern homelands.

  • Evidence of bidirectional exchange between the steppe and the Oxus

It should also be noted that, not only did steppe groups migrate southwards and got partially assimilated into the settled societies of the Oxus, but the Oxus or BMAC influence also spread further north into the heart of the steppeland. As per David Anthony, the major advocate of the steppe theory,

Stepped pyramids or crenellations appeared on the pottery of Sintashta, Potapovka, and Petrovka. The stepped pyramid was the basic element in the decorative artwork on Namazga, Sarazm, and BMAC pottery, jewelry, metalwork…This motif had not appeared in any earlier pottery in the steppes, neither in the Bronze Age nor the Eneolithic…Stepped pyramids appeared for the first time on northern steppe pottery just when northern steppe pottery first showed up in BMAC sites… later it became a standard design element in Petrovka and Andronovo pottery…A lapis lazuli bead from Afghanistan was found at Sintashta. A Bactrian handled bronze mirror was found in a Sintashta grave at Krasnoe Znamya. Finally, the technique of lost-wax metal casting first appeared in the north during the Sintashta period…Lost-wax casting was familiar to BMAC metalsmiths. Southern decorative motifs (stepped pyramids), raw materials (lead and lapis lazuli), one mirror, and metal-working techniques (lost-wax casting) appeared in the north just when northern pottery, chariot-driving cheekpieces, bit wear, and horse bones appeared in the south.

Besides this material cultural influence from the Oxus on the steppe groups far into the North, there was a gene flow, likely from Oxus into the steppe populations. As per Narasimhan et al.,

In the Central Steppe (present-day Kazakhstan), an individual from one site dated to between 2800 and 2500 BCE, and individuals from three sites dated to between ~1600 and 1500 BCE, show significant admixture from Iranian farmer–related populations that is well-fitted by the main BMAC cluster, demonstrating northward gene flow from Turan into the Steppe at approximately the same time as the southward movement of Central_ Steppe_MLBA-related ancestry through Turan to South Asia. 

As per Krzewinska et al,

The Bronze Age Srubnaya-Alakulskaya individuals from Kazburun
1/Muradym 8 presented genetic similarities to the previously published Srubnaya individuals. However, in f4 statistics, they shared
more drift with representatives of the Andronovo and Afanasievo
populations compared to the published Srubnaya individuals. Those
apparently West Eurasian people lacked significant Siberian components (NEA and SEA) in ADMIXTURE analyses but carried traces of the SA component that could represent an earlier connection to ancient Bactria.

Complementing these datasets is the recent Jeong et al paper on ancient samples from Eastern Eurasian steppe, which also documented extensive Iranian farmer related ancestry most likely from BMAC/Oxus from the Bronze Age onward.

Thus, the data clearly indicates a bidirectional cultural and genetic exchange between the steppe groups and people of the Oxus civilization, where those steppe groups coming down south got more influenced by rather than influencing the settled urbanised agriculturalists of the south.

In this milieu, how can one argue that it is the steppe groups that spread their language and culture on the southern populations all the way into India ? On the contrary, since the presence of Indo-Iranian languages is attested much earlier in South Asia than the earliest attestation of any IE language on the steppe, it is more parsimonious to argue that it is these southern agriculturalists who spread their language and culture onto the steppe.  Such a proposition, though quite logical, will make people very uncomfortable.

It may well be that it was the dark-skinned southern urbanised agriculturalists that assimilated and acculturated the fair-skinned nomadic steppe groups who then spread the cultural toolkit onto the steppe.

  • What about the genetic evidence of steppe admixture in South Asia ?

The genetic evidence is also not in favour of a steppe origin of Indo-European people. Whether it is David Reich or Willerslev, both agree that it is likely that the Proto-Indo-Europeans originated within a population with a largely Iranian farmer type ancestry, south of the steppe. Such a population already existed in South & Central Asia since the Neolithic.

So how can we be sure that those early urbanised populations of South & Central Asia were not already Indo-European speaking ? Why should one insist that it is only the steppe ancestry that could have brought the Indo-Aryans into South Asia when we don’t even have any proof of an IE language on the steppe before 1000 BCE ?

The steppe ancestry was clearly absent from Bronze Age Anatolia when we know that Hittite and Luwian groups lived in the region. Nor was R1a found among the Mycenaeans who were clearly Indo-European. Among the Western Europeans, R1a is hardly present and their y-dna profile is dominated by R1b, which is hardly present in South Asia.

When the steppe ancestry is recorded in the ancient Iron Age samples from NW South Asia, we do not find any R1a but find plenty of steppe maternal lineages. This rather supports in favour of the steppe ancestry entering South Asia via the steppe females. It also fits in perfectly with the lack of any steppe cultural marker in South Asian archaeological record. A case can be made that the steppe ancestry spread among South Asians after 1900 BC, when the South Asians took brides from the steppe groups who were also reaching out South. Thus the South Asian people got steppe admixture without any cultural influence of the steppe people by taking the steppe brides.

One final roll of the dice could be that why is it that the steppe ancestry is highly correlated with the high caste individuals in India ? Why do Brahmins have the highest steppe ancestry if steppe ancestry had nothing to do with the spread of Indo-Aryans in South Asia ?

There is also an easy answer for this – the spread of Indo-European or Indo-Aryan languages across much of northern half of South Asia, was followed by what is referred to as the Sanskritization process by the likes of Witzel. It is this Sanskritization which spread the steppe ancestry. A good example of this process spreading the steppe ancestry is the higher steppe ancestry found among Brahmins in South India. There was no language change in South India. There was only heavy Sanskritic cultural influence in South India, yet the Brahmins in this region have a higher steppe component than the rest of the surrounding population. This is what likely happened throughout North India as well,the difference merely being that people across much of North India who got Sanskritized were not Dravidian speakers but speakers of different Indo-Aryan languages that were related to Sanskrit.

As per Witzel, the process of Sanskritization occurred in North India, in the aftermath of the Mahabharata war, when the Kuru state re-organised the Vedic religion and ritual and spread it across the expanse of North India which was already speaking various varieties of Indo-Aryan languages. It is a known fact that the older layer of Indo-Aryan languages across much of North India has been overlaid by a subsequent process of heavy Sanskritic superstratal influence. This was a major political and cultural event according to Witzel who summarizes it thus,

It can be said that the Bh¯arata/Kaurava/P¯ariks.ita dynasty of the Kurus sucessfully carried out and institutionalized a large scale re-organization of the old R°gvedic society. Many aspects of the
new ritual, of the learned speech, of the texts and their formation reflect the wish of the royal Kuru lineage and their Brahmins to be more archaic than much of the texts and rites they inherited. In this fashion, the new P¯ariks.ita kings of the Kurus betray themselves as typical newcomers and upstarts who wanted to enhance their position in society through the well-known process of “Sanskritization.” …The new orthopraxy (and its accompanying belief system, “Kuru orthodoxy”) quickly expanded all over Northern India, and subsequently, across the Vindhya, to South India and later to S.E. Asia, up to Bali…the new dynasty was effective in re-shaping society and its structure by stratification into the four classes (varn. a), with an internal opposition between ¯arya and ´s¯udra which effectively camouflaged the really existing social conflict between brahma-ks. atra and the rest, the vai´sya and ´s¯udra; further, the Bh¯arata/P¯ariks.ita dynasty was successful in reorganizing much of the traditional ritual and the texts concerned with it….The small tribal chieftainships of the R°gvedic period with their shifting alliances and their history of constant warfare, though often not more than cattle rustling expeditions, were united
in the single “large chiefdom” of the Kuru realm. With some justification, we may now call the great chief (r¯aj¯a) of the Kurus “the Kuru king”…we are, I believe, entitled to call the Kuru realm the first state in India…It must be underlined, again, that the developments which brought about the the Kuru realm were lasting and not transient ones as those under the R°gvedic P¯uru or Bharata.
In effect, many of the changes in religion and society then carried out shape Indian society even today.

This Sanskritization, like in South India, is likely the reason for the spread of steppe ancestry across the non-Kuru Indo-Aryan states via the Brahmins from the Kuru realm, who would have been primarily responsible for the spread of the re-organised Vedic religion.

What is most interesting is that the Kuru kingdom was centred in the region of Haryana and Western UP, and it is the Jats of Haryana and Western UP, who have the highest steppe ancestry of any South Asian group, higher than the Jats of Punjab or even the Pashtuns further west. It therefore makes perfect sense, that if Sanskritization was spread from a region with a relatively very high steppe ancestry, its genetic influence on those regions it Sanskritized would also result in the spread of steppe ancestry in those regions. And if that spread was mostly related to the spread of Sanskritized Vedic ritual and religion, the group primarily responsible for such a transmission would be Brahmins from the Haryana and Western UP region, where steppe ancestry was quite high. It is therefore hardly surprising that the Brahmins over a period, all across India, began to show higher levels of steppe ancestry than their surrounding non-Brahmin populations.

The only necessary situation for this is that the Sanskritization process must have taken place after the steppe ancestry had admixed into the NW populations including among the Kurus. In other words, if steppe ancestry spread into North or NW India around 1500 BCE, the spread of Sanskritization from the Kuru realm across much of North India, must have been after 1500 BCE.




104 Replies to “The Fair-Skinned Sintashta folks did not spread Indo-Aryan languages in India”

  1. Great concise summary! The acculturation and spread of Sanskrit within India is an equally important riddle.

    The lack of a script or evidence of language on the entire Steppes in the BA has never handicapped AMT theorists. There are only three verified instances of script development in BA Eurasia – Mesopotamia, India and China. One needs to possess competence before being compelled to spread it!!

      1. @Razib

        You know, AIT Vs OIT is the great epistemological battle of this century. I have never had any qualms about the side I have taken.

        AIT has always been a pretender to empirical solidity while all it has in its bag are rational deductions. OIT had to begin with empiricism and the evidence has only been growing in trickles. Deduction travels a mile before empiricism even puts on its shoes.

        AIT always had its pick of deductive horses – linguistics in the last century and genetics now. Both of them were/are very agile in galloping away to conclusions (predestination) without ever stopping to drink at the streams of empiricism.

        I do not deny the inductive power of genetics in proving that people from the Steppes migrated to India. But where is the empiric of their prior knowledge of a language (any language)? This is beyond the ken of genetics.

        In the end, I predict, empiricism will prevail and AIT will be left quibbling over the definition of “things” such as a IE language, dispersal etc. They are already quibbling over the definition of a horse (Bokonyi) and a chariot ( Sinauli).

        1. Ugra

          But where is the empiric of their prior knowledge of a language (any language)?

          This is a dishonest way of framing the question. Ask yourself instead: given the dispersal of an ancestral set of dialect (called PIE) throughout Eurasia in particular eras, what theory explains it best? OIT advocates seem to have good convincing power online primarily because their audience is mostly Indian, who either don’t know this question at all or know little to no history of places outside India.

          1. There are actually empirical evidences. This will be our discussion topic after: previous – genetics, current – linguistics, next – mythology and finally – toponyms. I already presented several hundreds of toponyms (out of thousands) from South Asia on the link bellow. I translated some and, when I get some free time, I will do with others which are or easily translatable or identical with toponyms in Europe (e.g. Serbia) for example. Some translations directly point on the language of origin (so as the word ‘veda’, for example) and I think that is the main reason why this topic is still taboo.


          2. @Numinous

            You just called the Vienna school dishonest. Read up on it. Don’t overlook the verification principles. They are central to every scientific field. Right now, geneticists want to play historian, so they overlook verification. Steppes component = Indo Aryan language. Show me a reasonable hypothesis test with a good confidence interval. Being smug about it doesn’t help either.

          3. Awesome!!!! Welcome to 18th century @Milan Todorovic. With the discovery of Sanskrit by Europeans, Sanskrit toponyms were found in every single IE language. If @Milan Todorovic is just able to fast forward to 20th century, the person will discover that linguists now consider isoglosses in constructing PIE. The startling find was that Slavic shares the most isoglosses, in the decreasing order, with Baltic, Iranian, German, and then Sanskrit.

            The isoglosses with Sanskrit are a subset of the ones with Iranian. This suggests continuous inhabitation with Iranians at some point of time. For this reason, a subset of mainstream theories about Slavic origin consider them to have descended from Iranians. The most popular theory is that they originated from Baltics as they share the most commonalities with them. Furthermore, Baltic is more conservative than Slavics as they have preserved more PIE words.

            Seriously, how can somebody claim that Slavics have anything to do with PIE when they are a daughter language of Baltics (in the sense that they contain less PIE words than Baltic)? Alas! when will motivated reasoning end?

          4. So-called ‘Vienna-Berlin historical school’ was the main fabrication factory of falsified history. They especially falsified Serbian history (because others did not have much to be falsified) and promoted falsification about so-called Slavic (i.e. Serbian) migration to Balkan in the 7th cAC. (Maybe they provided this ‘Slavic’ attest from the 6h cAC?). I will provide some specific examples of their falsifications. In cooperation with British they have been the key reason why we still don’t know the truth about Aryans and why we are chasing our tails for 200 years. One of their main products is also the falsified Greek ancient history, invented and given as a present to Greeks in 1829. AC, when the FIRST ever Greek state in history was created by British/Austrians. I will write about the reasons why this was done.

            It is fair to say, the German’s historical school officially rejected their own past falsifications and started presenting old German’s history. Very significant is their publishing of old maps of present German’s territories with original toponyms where we can see that the most of them are Serbian toponyms (e.g. Berlin, Dresden, Leipzig, Brandenburg, Prussia, entire former East Germany, etc), they established the Serbian Museum in Germany which presents several centuries of systemic genocides against Serbs and where currently live about 60 K indigenous Serbs (Nazis killed 90% of remnants which survived previous genocides).

          5. @Numinous, @Mohan and everyone else on BP

            I just realized that there is a vast amount of ignorance on epistemological methods. Guys, you need to start knowing your epistemes inside out before getting convinced on theories. I will use an example from the astrophysical world.

            Batygin and Brown (astronomers) studied the motions of a clutch of eTNOs (Trans Neptunian Objects) and proposed that there was a giant Planet out there (beyond Pluto) that was perturbing the orbits of the eTNOs and pushing them into strangely aligned orbits. They set up deep orbital resonance decay studies of all the eTNOs and proposed that only a large mass sufficiently dislocated out of the solar orbital plane could achieve these effects. The mathematics and construction is quite convincing.

            The IAU (International Astronomical Union) hasn’t rushed to the press with a new planet name. Why? It is waiting for an empirical piece of evidence . The planet has to spotted in the heavens. Batygin and Brown’s work are fine examples of deductive and inductive reasoning. But in the end, empiricism is king. Repeat it to yourself, guys. This is the bedrock of the the modern scientific world. Albert Einstein lived by it – this is the the Vienna School. Yes, there have been solid critics of it (Popper, for instance). But it is enormously valid.

            Vagheesh Narasimhan’s 2019 paper linking Steppes component with IE dispersal is an example of inductive reasoning. The Steppes aDNA comparison with modern Indian DNA is empiricism. But the link with IE languages is all induction, zero empiricism. Something has to turn up on the Steppes to show that these people spoke a language, composed poems, wrote ballads etc. So far, zilch. Zero evidence of Steppes people speaking or writing any language.

            I will also show you what an example of powerful empiricism looks like. Readers of the Rgveda have long known that Indians have always described the river Saraswati as a mighty one flowing from the Himalayas to the Ocean. The 2019 Chatterjee paper proved this by finding muscovites from the upper reaches of the Himalayas in the Thar and Rann of Kutch. They also estimated the periods of perennialism of the Saraswati – (80-20 kya) and (9-4.5 kya).

            We have three independent sources being verified in proximate empiricism – textual tradition (Rgveda), archaeological excavations (Harappan settlements on the banks of Ghaggar) and fluviology (muscovite dating).

            A similar set of empirical proofs for the claim that (Steppes component = IE languages) is lacking. Induction/deduction will only go so far. Kneel before empiricism!! This is why AIT vs OIT is nothing but induction/deduction vs empiricism in disguise.

        2. “AIT Vs OIT is the great epistemological battle of this century.”

          LoL. AIT vs OIT is NOT the great epistemological battle of this century. England vs India cricket match series is.

          couple of years ago, on the eve of the famous rakhigarhi paper’s publication, razib predicted somewhere on this blog that the publication of the paper will be a seismological event in the cultural history of india. it will shatter the dearly held believes of many in India, and reshape their understanding of their very existence. i predicted that nothing of that sort will happen, and indians will remain busy in the more worthy happenings of cricket and bollywood and hindu rashtra politics.

          two years down the line, i can legitimately claim that I have been proved right. rakhigarhi paper has come and gone, and hardly anyone beyond the academia took notice of it. as RK narayan famously said “india will go on”, india is going on – nonchalantly.

          debates like AMT vs OIT reminds me of big endian vs little endian wars of computer architecture of bygone era. an almighty intellectual fracas over something stupidly trivial.

          personally, i have a very irreverent “marxist” take on the whole conundrum. if you go sufficiently back in history, the lineages of all of us will terminate on some gorilla who lived in the savannah of africa thousands of years back. what stops the descendants of our revered gorilla forefathers took before they finally pitched their tents in india is really immaterial.

          1. razib predicted somewhere on this blog that the publication of the paper will be a seismological event in the cultural history of india. it will shatter the dearly held believes of many in India, and reshape their understanding of their very existence.

            i remember you saying this. but i’m pretty sure i didn’t say that

            1) i’m pretty clear on the record that i think culture influences science more than the other way around in interpretation

            2) that sort of bombastic phrasing is more likely to be deployed by an indian 😉

            i’m 99% you have a false memory about me, but you can prove me wrong. find the exchange.

          2. @ Scorpion Eater

            If I remember my zoology correctly, gorillas are our cousins on the hominin tree, not ancestors. You are probably the target audience for Rakhi Sawant’s shows.

  2. So there is a gap of more than two millenium between the earliest attestation of Indo-Aryan languages in South Asia and the earliest attestation of Balto-Slavic languages on the steppe.

    Even if we accept that IA languages were spoken in India before the Saraswati dried up (still debatable IMO), I’m not sure what the above statement proves. This two millenium gap would be consistent with OIT if Balto-Slavic were proved to be a descendant of Rigvedic IA. But does anyone actually claim that? That fact that Balto-Slavic was attested to relatively recently seems irrelevant to the homeland theory in light of what linguists believe about the relationships between the various IE branches. If AIT is correct, Balto-Slavic is a northwestern remnant of PIE, and if OIT is correct, the ancestors of Balto-Slavic speakers left India long before the Saraswati dried up.

    1. Numinous,

      I am not sure if you are deliberately trying to be obtuse because having seen your line of thinking it does appear that you have already made up your mind.

      What you are clearly missing is that I was highlighting the stark contrast between the earliest evidence of Balto-Slavic which is today the only IE language group close to the steppe vis-a-vis the earliest evidence of Indo-Aryan in India. Now, if steppe is the homeland, Balto-Slavic should directly evolve from PIE parental language in that same region over a period of several millenia. But you have zero proof of it or its parent language on the steppe and how it evolved from that to reach the single Slavic language that is attested from 6th century CE. In sharp contrast, it is already accepted by linguists that even during Rigvedic period around 1500 BCE or earlier, there were multiple Indo-Aryan languages in South Asia. By the way, Balto-Slavic and Indo-Iranian are considered to have evolved in close geographical proximity.

      And as I also mentioned, there is no evidence of any IE language speaker on the steppe before the 9th century BCE. Maybe you should answer why that is also irrelevant.

      1. JR,

        Read Fraxinicus’ comment below for a better articulation of my position (though it was more question than position).

        Now, if steppe is the homeland, Balto-Slavic should directly evolve from PIE parental language in that same region over a period of several millenia.

        No, this is not necessarily true. You are confusing present domination of Balto-Slavic languages (actually, just Russian) in the steppe with historic dominance of the same language family. Today, Russian is spoken in Vladisvostok too, but no one seeks to find a PIE homeland along the Amur river.

        By the way, Balto-Slavic and Indo-Iranian are considered to have evolved in close geographical proximity.

        I’ll take your word for this. My sense was that linguists don’t care about the centum-satem issue anymore, but perhaps there’s more evidence of co-evolution between these two language families. In any case, this would seem to be neither here nor there because AIT posits Indo-Iranian evolution in the steppe. With Balto-Slavic lying somewhat to the northwest, it wouldn’t be surprising if both language families had some shared history.

      2. Mr Rathod, your article doesn’t fit any logic. This is well documented that a lot of migration happened in last 2000 yrs from North West

    2. @Numinous
      You should study linguistics in more detail. You muddled up the relationship among IE languages. For a fact, Sanskrit is closest to Iranian and then Slavic. But did you ever consider which language is closest to Slavic? It is Baltic, Iranian, German, and then Sanskrit. The isoglosses shared between Slavic and Sanskrit are a subset of those between Iranian and Slavic. Furthermore, it is well known that Slavics lived side by side with Iranians at some point of time.

      My recommendation is that you should do an in-depth study before forming opinions on the IE debate.

      1. Read my above response to JR for why this is not an argument for (or against) OIT. The relationships between the languages as you point out are likely correct, but I found the same relationships in David Anthony’s book a long time ago. And Anthony posited a version of the AIT from the exact same evidence.

  3. I didn’t follow the interplay between Brahmins, Sanskritization, and steppe ancestry. Are you saying that ancestors of Brahmins were originally migrants from the steppe who adopted Sanskrit and then spread around the subcontinent Sanskritizing both IA and Dravidian speakers, and in the process, spreading steppe genes as well?

    1. No. If Brahmins have 30 % steppe, 50 % IVC farmer, 20 % AASI, how can you say that ancestors of Brahmins came from the steppe ? Is there any sense in such an interpretation when majority of their ancestry is still Indian ?

      My point is – Rigveda and the presence of Indo-Aryan or Vedic culture and language can very well predate the intrusion of steppe ancestry in India. The steppe ancestry cannot be directly connected with coming of Indo-Aryans since you have no proof of the languages spoken by these steppe groups in the 1st place.

      It could be that as the Harappans were already culturally connected with people of the Oxus, there was inter mixing between these 2 civilizations. When the steppe groups came down South and started assimilating into the urbanized Oxus culture, steppe ancestry must have started spreading.

      Through intermarriages and other social processes, the steppe ancestry could have also entered South Asia. A proof of such a process is the steppe maternal ancestry in the Swat Iron Age samples. After the steppe ancestry levels increased through such social intermixings as far as the Kuru homeland, the process of Sanskritization would have spread the steppe ancestry further inland, but with a skew towards the Brahmins having greater steppe ancestry in those Sanskritized regions than the non-Brahmins. This could be because Sanskritization may have been a upper caste phenomenon with Brahmins from steppe rich Kuru homeland, moving out and settling into the inner regions of India thereby spreading the reorganized Vedic religion and the Sanskrit language along with it. But the people of non-Kuru regions, where Sanskritization occurred were already speaking non-Sanskritic Indo-Aryan languages whose later descendants we know as Magadhi, Sauraseni, Maharashtri, Paisachi, Pali etc

      Another intriguing possibility that I did not mention in my post – it is possible that with the spread of Sankritization and the power of the Kuru state, the fame and renown of the Kuru land as the centre of Vedic learning and as a holy land, was perhaps also prevalent in Central Asia, which could have led some of these Central Asians, including the steppe origin ones, to migrate to the Kuru holy land to acquire knowledge and maybe also settle down for good. There are many ways how steppe ancestry could have come there. A simplistic invasion scenario whereby the steppe groups were able to impose their language and religion is not the only way the steppe ancestry spread into India, and such a process is highly unlikely given the lack of any steppe material culture south of Hindu Kush.

      1. how can you say that ancestors of Brahmins came from the steppe

        Just to be clear, I wasn’t saying that, but rather wondering if you did.

        But I understand your position now. You are suggesting that steppe admixture was much higher in the Kurus (across all castes), and it was Kuru (or Kuru-region?) Brahmins who spread steppe ancestry across India.

        I think it sounds plausible, but I’m still skeptical. Unless things were radically different in the Kuru era, Brahmins don’t seem like the gene-spreading types. Kshatriyas would have been more likely to do that, no? But, given the percentages, maybe there’s a case for it.

        1. Numinous, non-arya kings were possibly more likely to invite brahmins to their regions than kshatriyas. There’s a case to be made that brahmins were more convincing at being high priests than arya kshatriyas were at being soldiers. These brahmins were served up local lasses to settle down with in agraharas, and then over the course of generations, politically elite shudra males take the occasional brahmin wife, hence the modest enrichment for steppe in these populations.

          1. r1a is found in all indian groups to my knowledge except one: munda. i think that suggests matrilineal assimilation of indo-aryan men. this happened with basques. their Y chromosomes are from yamnaya but genome-wide still farmer enriched

          2. “There’s a case to be made that brahmins were more convincing at being high priests than arya kshatriyas were at being soldiers”

            Dont think Arya Kshatriyas were that bad at being soldiers, I mean Shivaji invented a Rajput lineage to appear more Kshatriya than any of the Deccan Kshatriya group. You also have a bunch of Mundas and Kshatriya principality in Central-Eastern India all claiming to be of Rajput descent. I mean even muslims of Pakistan claim Rajput descent.

          3. Saurav, maybe kshatriya prestige comes from the brahmin prestige. It’s shudra post-facto trying to demonstrate legitimacy to brahmins. If deshastha brahmins acknowledged shivajis sacred right to kingship, he wouldn’t have had to cook up a lineage. Anyways, the brahmin impact on southern culture is immense. The arya kshatriya one is specious, if anything the deccan lineages have had more impact in the north, via the rashtrakuta and chalukyas.

          4. I would contest that, The Deccan dynasty outside of Marathas had very limited impact on the North. The Mughals and Sultanates had more in the South. I get all this Chalukyas and Rastrakutas thing, but most of it seemed a bit like Kashmir’s Lalitaditya thing where he supposedly ruled from Iran to Bengal aka exaggerated. The most impact these dynasty had is at best Maharashtra and surrounding regions. I would argue that Bengali Pala dynasty had similar impact on North considering they ruled approximate the same amount.

            Of course till the big boys Pratiharas showed up and made North their realm.

          5. @Saurav “You also have a bunch of Mundas and Kshatriya principality in Central-Eastern India all claiming to be of Rajput descent. I mean even muslims of Pakistan claim Rajput descent”
            What do you mean by claiming descent ? There are Rajputs in Pakistan Punjab, similarly small numbers of Rajputs moved to Central- Eastern India. No Munda claim any relationship, only those counted in census are accepted as Rajputs. You still have some Gond rulers alongside Rajput rulers in Chattisgarh, Jharkhand etc, none of them are related to the neighboring Rajputs.

          6. “Saurav, maybe kshatriya prestige comes from the brahmin prestige. It’s shudra post-facto trying to demonstrate legitimacy to brahmins. If deshastha brahmins acknowledged shivajis sacred right to kingship, he wouldn’t have had to cook up a lineage.”
            How ? When kshatriya/ Rajput is a community like Brahmins is a community. There was no way for Chh Shivaji Maharaj to become kshatriya, this is why he had to claim a Sisodia descent for himself and it didn’t changed anything for him.

          7. This idea about southern population getting the steppe component through females is laughable. R1a is widespread in the UCs there, R2 too. You just need to look at the men.

        2. Brahmins were definitely the gene spreading type down the konkan and malabar coast, and in south east asia. Just check out the brahmins genes (south indian subgroup on 23andme) and related r1a1 frequencies in a myriad of non-brahmin groups (nairs, kodava, south east asians etc). I would not underestimate it.

        3. @Numinous
          F.E Pargiter says that the lunar line kshatriyas (The one which Kuru dynasty belonged to) were responsible for conquering and sanskritising the subcontinent.

          He also says many of the lunar line kshatriyas were Kshatriyas-Brahmins and many also renounced Kshatriyahood to become Brahmins. The most famous of them being Vishvamitra.

      2. F.E Pargiter says that the lunar line kshatriyas were responsible for conquering and sanskritising the subcontinent.

        He also says many of the lunar line kshatriyas were Kshatriyas-Brahmins and many also renounced Kshatriyahood to become Brahmins. The most famous of them being Vishvamitra.

        His theory fits very well with yours.

  4. I wonder if you could summarize the arguments in favor of OIT and against AIT in a few short points. People can then refer to your article for supporting evidence. Thanks!

  5. After one excellent article about ancient numismatics, one very opposite writing. To repeat my first impression which I mentioned couple years ago – a scrambled text, jumping up and down, back and forth in space and time. What a waste of time and energy which could be much better invested. It is a logical result of a fixation with one foggy idea and a futile quest to find elusive supporting evidence. This idea does not deserve the letter T in its name (OIT). It has neither the beginning, nor the end but not the middle, either. It cannot stay standalone, 90% is just a negation of AIT/AMT which, at least, has very solid and bold T. It cannot sustain couple self-sufficient sentences without mentioning A*T.

    What I have heard so far from ALL its proponents – (1) elephants in Syria, (2) zebu now in Balkan (???), (3) Saraswathi exists (or not?) in Rg Veda (4) one ancient astronomical thing in Rg Veda which I cannot remember. And that’s it, plus – all mouses in the world have Indian genetics(?). Just to say for zebu I can say that 99.9% of people in Balkan would not know what the zebu is (90+% probably never heard for this name).

    The biggest problem is a totally corrupted taxonomy – outdated, backdated or invented names, what is ‘Indo-Iranian’ at the time when neither Indo nor Iranian terms existed, who are Indo-European, where they came from, which language they spoke, what is Indo-Aryan, etc. The problem is also the logic. Numi found couple examples. Who and where attested ‘Slavic’ (what is this?) language in the 6th cAC. Why use the term Balto-Slavic if Balto is ‘attested’ in the 16th cAC? Balkan Indo-European languages? – Greek (it was NOT IE, whatever is this), Albanian? (who came to Balkan in 1043 AC and not worth mentioning in any context except narco-trafficking). I wrote few comments about words which Greeks, according to Plato, took from indigenous Serbs long time before Jesus. How about the 6th cAC attest?

    Several things, the author was afraid to spill explicitly. E.g. R1a cannot be found just about anywhere but he stopped short of mentioning that R1a in India is indigenous. Otherwise, it should be explained how was spread btw Kamchatka, British Isles, Arctic and Arabia. Numi already mentioned Sanskrit – it is unclear what was the intention in the text. I guess, somehow between the lines to assert that Sanskrit was also indigenous what was even implied in the title. But, it also stopped of saying this because it would need to be explained the similarity btw e.g. the Serbian language (but not surrounding languages!) and Sanskrit and many other things. How Sanskrit dropped by parachute to Serbia?

    There are, of course, not mentioning of “Indo-Europeans” in China, Tibet (Tibeto-Aryans?), Tarim Basin, etc, because there would be a question – how “IE” came everywhere around but not to India? There are many other points, I will just stop here. I wrote in a good spirit expecting, probably vainly, that author will focus on science not on suspiciously political idea. But, I am sure that hubris will not allow this to happen what is really a shame and another wasted opportunity.

  6. The Senczuk preprint is quite voluminous – in addition it references the Verdugo paper from 2019 that nailed Zebu introgression into the Levantine and Near East. This should be clear to every BP reader on what animal genomics indicate.

    Three features of this zebu influx after ~4000 yr B.P. attest that the influx was likely driven by adaptation and/or human agency rather than passive diffusion. First, the extent of indicine introgression does not follow a simple east-to west gradient; for example, it is pronounced in Levantine genomes from the western edge of the Near East. Second, the introgression was widespread and took place in a relatively restricted time interval after four millennia of barely detectable B. indicus influence. Third, it was plausibly driven by bull choice, as we observe up to ~70% autosomal genome change but a retained substratum of B. taurus mtDNA haplotypes (Fig. 2 and table S3). Hybrid B. taurus–B. indicus herds may have enabled the survival of communities under stress and perhaps facilitated expansion of herding into more-peripheral regions. Restocking after herd decline may have also been a factor. Westward human migration has been documented around this time (19, 20) along with archaeological evidence for the appearance of other South Asian taxa such as water buffalo and Asian elephants in the Near East (21), suggesting the movement of large animals by people.

    This is the same archaeological time period of the Mitanni (1700 BC onwards) who worship Indo-Aryan divinities and the king-names are onomastically Vedic.

    From another 2019 paper –

    The spread of zebu cattle (depictions) from the Indus valley to the East Mediterranean region including Syria, Anatolia, Cyprus and Palestine correlates well with the present geographical distribution of human Y-haplogroups L-M11, R2 (R-L266) and R1a Z93 originating from Hindustan and of South Asian human genome K5 and South Indian human genome K16. Both sets of data correlate with maps showing the relative proportions of zebu admixture in the three genomic components (mitochondrial, autosomal and Y-chromosomal) from the present day cattle populations across the Near East. These human population and domestic humped cattle movements are to be connected with the dispersal of Indo-European languages (spoken — Indo-Aryan and Iranian — and extinct — Kassite and Mitanni Aryan and Anatolian) from Hindustan to Asia Minor in 7000–1000 BC.

  7. Though I’m not sure I agree with the conclusions, you’re the most well reasoned and thoughtful of the OITers we’ve come across. Great article and thank you.

  8. JR,

    lots to disagree with. but i really dislike this tendency: Here is what James Mallory says, in a book he co-authored with the linguist D Q Adams,

    this book is from 1997, correct? you know very well a lot of these scholars have changed their views on lots of issues. anthony himself shifted from elite diffusion to mass population replacement in europe in the last 10 years (btwn 2008 and 2014).

    some of the stuff is also out of date. you quote the supplements of Damgard et al., but that paper is OK for its time, but clearly superseded.

    you say things like mycenaeans don’t have R1a (true), but don’t mention that steppe ancestry shows up in samples from greece after 2000 BC.

    1. @Razib,

      Perhaps you can provide a short answer if aryan intrusion in India is driven by “R1a” or by “autosomal steppe”?

      You seem to flip-flop between exactly what counts as aryan intrusion to India from genetics point of view.

      If you are going to get behind the “autosomal steppe”, please explain steppe admixture dating of Vysyas with mean time before 2000BCE. This is from Narasimhan et al. 2019. Hope this hasn’t changed by something even more latest.

      1. Perhaps you can provide a short answer if aryan intrusion in India is driven by “R1a” or by “autosomal steppe”?

        your comments regularly confused underpowered/small sample size with representativeness when you talk R1a.

        both matter. the martin richards paper explains

        am i confused that you don’t understand sex mediated gene flow? i can’t under how you couldn’t after reading me all these years

        If you are going to get behind the “autosomal steppe”, please explain steppe admixture dating of Vysyas with mean time before 2000BCE. This is from Narasimhan et al. 2019. Hope this hasn’t changed by something even more latest.

        not going to respond unless you post exact page number or exact quote.

        i’m pretty sure i wasted time figuring this before when someone got confused about what the supplement says. not going to waste my time again

        1. That’s exactly my question about R1a. How can an underpowered sample be representative of population-wide percentages?

          I am not confusing that R1a came from steppe. I am questioning that R1a being “wide-spread” in Indian populations due to nature of underpowered stats.

          On “male-mediated” dispersals, I am not sure why you are quoting 2017 paper using 1440 samples of mtDNA lineages (which makes assumptions about what is “largely” in maternal gene pool referring to a paper that used 992 samples as a gene pool) while I was wondering about the paper you posted yesterday with 20k samples at 8% of maternal gene pool (their words) being mtDNA of steppe-origin.

          Just to be clear on the usage of different terms. I am concerned about % estimates of population-wide presence of R1a. I am NOT concerned about % contribution of steppe to Indian population (we went over this already).

          Ref of Vysyas:
          Narasimhan et al. 2019, supplement excel Table S5 , row 70, Cols O and P.

          1. I am writing in multiple comments to be as clear as I can be:
            1. Large percentage of population having R1a is indicative of rapid steppe take over. This means, we need large sample size of the population to be sure that R1a is indeed a large percentage of all y-haplogroups within the population (as a whole).

            2. Even small percentage of steppe autosomal DNA indicates arrivals from steppe mixing with native Indians (for lack of better word). Then, the timing of that arrivals is important because of when they arrived earliest is more important than when they arrived at the latest. Again, this needs large number of samples with a population to figure “when” steppe admixture to a population happened. (ref: Razib saying Linkage Disequilibrium needs significant sample size to figure population-wide variance).

            3. Small sample size (i.e., 5 samples) is sufficient to figure out what is autosomal DNA of steppe admixture. In this case, multiple group “representativeness” is important to see if steppe is indeed present in numerous endogamous groups of India. Even though about 200-400 groups from India are sampled, total population representation by them is less than 50% of Indian population.

          2. Also that R1a in Indian Subcontinent is of the older variety then elsewhere. I am not taking about ancient found dna, but current branches. TMRCA of R1a in India-Pakistan is apparently 15,450 +/- 2,900 ybp compared to Eastern Europe’s 12,516 +/- 3,800 ybp. Also the fact that R2 only exists in India confirms that R1 & R2 emerged from R* in India only before going elsewhere. Also remember Native Americans have R1, not R1a, had swastika(meaning Aryans predate the R1a) and in PCA graphs are closure to Indians, look like us too in skull structure.

          3. @ RJ – Wow! Let’s see what does it all mean. That Sanskrit was carried to Europe 12000 years ago? Not any part of Europe then directly to Vinca where 95% of European people lived and where (in Serbia) is the oldest R1a – 12000 years. Considering that Europe was still in Ice Age it would be interested to know how these ancient Indians travelled and how did they know where to go? And the hell were looking for in Europe covered by ice? And probably they remained in Vinca for several thousands of years because migration from Vinca started at about 6000BC. They probably mastered Sanskrit during this period before they diffused it around the Europe and Russia. At that time Russia was still under water, there was a lake after the melting of ice with size of Mediterranean. It is interesting that 12000 years ago they had Sanskrit words for wife’s sister, husband’s mother, husband’s brother’s wife, etc. I only wonder how zebus survived this trip from India to Vinca without any grass. Apparently, elephants could not persist the whole trip and were abandoned in Syria. I also wonder how they selected only R1a guys for this trip but no other haplogroups, maybe because they are more resistant on ice and cold. I wonder so many other things.

          4. @Milan Todorovic
            No R1a/I has ever been found in Vinca. Looks like R1a/I arrived after Vincans left/perished. Vincans aka Anatolians have only the following Y-DNA haplogroups: G2a2a1, G2a2a1a, G2a2a1a2a, G2a2a1a, G2a2b2a1a, H2. They have nothing to do with PIE or Slavics.

            Furthermore, the oldest R1a (~11000-10000 BC) has been discovered in Siberia, Russia you piece of trash. Genetics has found that Europe’s population has been replaced — either partially or wholly — multiple times. All the time talking nonsense, you need to get treated at a mental asylum.

    2. Razib,

      The quote is the from ‘The Oxford Introduction to Proto-Indo-European…’ by Mallory & Adams that came out in 2006. What you have in mind is ‘Encyclopedia of Indo-European Culture’ by the same set of authors that came out in 1997.

      The quote in question points out how a particular homeland theory becomes widely accepted in academia. I am not sure things have changed so much in the last 15 years. Genetics cannot tell you the language spoken by that dead man. If you already know his language, genetics can help you understand what DNA was carried by that particular language speaker and so on.

      The Damgaard supplement was purely about archaeology. The particular quote gives the latest understanding of archaeologists on the nature of interaction between steppe groups and Oxus people. I am not sure there has been a radical shift in the opinion of archaeologists in the last 5 years but I will be happy to be corrected.

      The ‘steppe’ in Mycenae is complicated because it could very well have come from Bronze Age Armenia as the paper clearly showed. Considering how Greek is most closely related to the Armenian among all IE languages, it is very probable. There was no point wading into it as it was outside the scope of my article. But if R1a was a marker of IE spread, it is surely relevant to ask why R1a is not found among known ancient IE groups like Mycenaeans and Anatolians, unlike the hypothetical steppe IE speakers.

  9. Fourth, if there was any Central Asian influence on South Asian populations, that influence likely long predated any development of Iranian, let alone Indo-Aryan, languages, and most likely occurred during the late NMG IV to early NMG V period (ca. 2800–2300 BCE) and even earlier during the Eneolithic from Kelteminar culture groups (4000–3500 BCE).

    to be clear to readers, for various reasons this paper thought the gene flow of steppe came from the yamnaya culture, which is much older than sintashta. so they had to say this by their results. but more samples and better methods since this paper indicates that the gene flow isn’t from yamna but from sintashta, which is dated to much later

    1. I guess this could confirm what I mentioned in several occasions and what is opposite to the mainstream – Yamnaya people (R1b) were NOT Aryans (not even as a reflux) and they did NOT bring ‘Indo-European’ (whatever is this) language to Europe (unless Yamnaya and Aryans, theoretically, spoke the same language). So-called ‘Indo-European’ (why ‘Indo’?) originated in Vinca and this is Sanskrit’s pra-language. Some other time we can even discuss if Yamnaya were white what would be very interesting in contemporary discussions against ‘white supremacy’ in US.

      Also, for those who don’t know, Vinca’s script is the oldest in the world. Swastika is one ‘letter’ in this alphabet. We can hear the scientist who researched Mesopotamian script for 40 years and finally realised where the future research should be directed (2.5 min):

      1. For guttersnipes like @Milan Todorovic:
        1. Vincans aka Anatolians have the following Y-DNA haplogroups:
        G2a2a1, G2a2a1a, G2a2a1a2a, G2a2a1a, G2a2b2a1a, H2. They have nothing to do with PIE or Slavics.
        2. The earliest evidence of Swastika in Madhya Pradesh, India is dated to at least 10,000 B.C. — if not earlier — suggesting indigenous origin of swastika .This predates Vinca by 5000-6000 years, and it is at least contemporary to or predates Ukraine Swastika.
        3. Calling graffiti script can only be done by @Milan Todorovic.

  10. Whether it is David Reich or Willerslev,

    knowing what i know about eske willerslev i’m 99% sure he doesn’t have strong opinions privately, though he will express opinions if asked 😉 eske only cares about it getting published. he’s not super invested in details of theory

  11. It seems you’re saying “the Kurus with high steppe ancestry won the Mahabharata and then Sanskritized everyone”.

    Aside from timing, how is this all that different from “sintashta people mixed with Indus valley remnants and then this ANI spread across India to mix with ASI”?

    1. He is separating Indo-Aryan speakers from Steppe ancestory. (So indeed those two statements could be equivalent, but JR is saying the IE languages existed in India before this steppe component intruded and spread). If I understand it right.

      1. There is so much confusion, especially related to taxonomy. E.g. Indo-Aryans – are they Aryans who came to India or ‘local’ guys who named themselves ‘Aryans’? Also, ‘Indo-European’ languages (plural?) spoken in India before Aryans arrival – why not say – ‘Indo’ languages which migrated to Europe and enforced to all local Euro population (when, how, who?). How is it possible if the Sanskrit is 4000 years old? It seems that OIT realised that the genetic input to SA cannot be denied but there is a strong effort (not only here) that it could not happened after 2800BC and before 1000BC, i.e. during the Vedas and language creation period. Those, who came before 2800BC got assimilated and did not have anything with later Vedas so as those who came after 1000BC, who came when everything was already in the place.

        OIT (I had impression that they recently genuinely synergised into IT but I was wrong) should present their elusive comprehensive ‘theory’ without mentioning A*T. And to qualify for this, they need first to explain the meanings of ‘rg’ and ‘veda’.

        1. The incoherent rumblings of @Milan Todorovic rages on:
          IVC started from 3300BC and only ended ~1300 BC. It was multi-cultural, and hypothesized to be multilingual too. It spanned from South Central Asia to North India (BMAC had a related culture + belief + language according to archaeologists). For such a vast civilization, there is nothing surprising if they spoke IE languages too.
          Vincans aka Anatolians have the following Y-DNA haplogroups:
          G2a2a1, G2a2a1a, G2a2a1a2a, G2a2a1a, G2a2b2a1a, H2

        2. As usual for @Milan Todorovic, whose ignorance reaches the high heavens:
          Aryan is the term Indians gave to themselves. Its cognate exists only in Iranian and no other IE language. Slavics gone to sleep (lol) and never heard the word. For Serbians who claim Greek Alexander, Plato, Aristotle, etc. as their own, without a shred of proof, I guess it must be devastating.

  12. There are many threads through this article. I look forward to the full version if see how much of the details are squared.

    I don’t understand the wholesale narrative of “high R1a” in north Indian Brahmins stuff. When Sharma et al (2009) sampled 31 UP brahmins R1a is 67.74% and when Underhill et al. (2010) report 171 of UP brahmins (supplement Table 1), they report only 85 of them as R1a1a (i.e., 49%). This is clearly showing small sampling bias where % decreases as sample size increases (assuming “UP brahmins” label consistent between both the papers and all R1a reported by Sharma et al 2009 is actually R1a1a).

    There are about 20k samples of mtDNA showing 8% steppe mtDNA sources apparently starting from 2500BCE in Indus region. There are less than 2k y-DNA samples from all over India showing about 15% R1a. I don’t know what nuance I am missing for the AIT narrative to be insisting on this R1a stuff other than that it got stuck during low sample sizes and isn’t being updated.

    1. This is clearly showing small sampling bias where % decreases as sample size increases (assuming “UP brahmins”

      you mean sample variance, not sample bias

      1. No, not sample to sample variance, (ie., from taking multiple samples from same population). I meant bias introduced due to small sample size because estimated mean of the population is biased.

          1. Parameters of a random variable will have a bias and variance. Mean of a population is a point estimate which will have its own bias and variance.

        1. We would have a variance if we had confidence bounds on the given % values (e.g., 15% +/-10% can change to 15%+/-5% with higher sample size). That is reduction in sample variance.
          But change of 15% to 5% implies bias if the +/-10% remains the same.

    2. The 20K mtDNA samples are from northern Pakistan, aren’t you referring to that recent paper? The Indian value on the other hand includes many different Indian groups. So if you want a consistent comparison then compare northern Pakistani mtDNA to the frequency of R1a in northern Pakistan.

      1. Agreed. That’s what I am trying to compile from Narasimhan et al. data and all available R1a and MtDNA data by sub-population (ie, location +caste) along with the 95% confidence intervals.

        Life is beginning to interfere though. But it is nice side project now that covid data tracking is less interesting.

        1. See if you can request data from the UK storage/reserves. Allegedly they have a lot of south Asian DNA so you could find enough from each group of interest from a general region to do that.

  13. violet, you have poor intuitions in how many populations are needed to make assertions on the indian genetic landscape. we clearly have enough at this point. adding more populations is not finding anything new at all genome-wide

    re: not having enough samples for Y estimates. we pretty much due. the european populations have been surveyed at larger numbers and there’s nothing surprising. the percentages are around the early estimates.

    re: bias. there’s no bias the researchers had in whom they selected. it was random.

    i can ask tho what the UK biobank samples come out as. i don’t have access, but others do. there are thousands of indians, bangladeshis, and pakistanis there.

    1. @Razib,

      I am deep diving in data because I want to build those intuitions. I am not expecting anything new with respect to genome-wide admixture.

      What I am looking for is explanation on Figure 4 (c) from Narasimhan et al. (2019). On x-axis is the admixture percent from genome-wide analysis and on y-axis is the percent of R1a in population. It is proportion of R1a in excess of steppe admixture percent that indicates male-mediated migration (if I understood correctly). Each point is a sub-population (or we can say “caste group” for short-hand naming).

      Apriori, I had no problems consider 30 samples per caste-group to represent percent y-haplo. We can find 95% confidence bounds on 30 samples with 67% R1a as 56-86%. This implies additional 30 samples taken from same group should give proportion of R1a between 56% to 86% over 95 times out of 100. But comparison of Sharma et al. (2009) and Underhill et al. (2010) for UP_Brahmins shows that 30 samples gives 67% while 171 samples gives 49% (43-58% at 95% confidence). This is an evidence of non-random sample. I am not accusing any authors of bias. They used what they got which is common in high-cost data acquisition fields. The data sampling process itself is suspect is what the available open-access publications are pointing at.

      So, each y-axis location of any given point in Figure 4 (c) needs to be taken as suspect until multiple independent samples are taken. (This might be already happening as you pointed out). This is where large sample size would resolve this with greater confidence.

      Second point is about, how many points are there on the figure. This is a caste-group representation issue and can be inferred with reasonable number of groups IF they are selected properly (based on caste status and group population – rather than discrepancies like “kamma” labelled as “lower caste” and “reddy_telangana” given as 20 million population). These can be resolved by general public with minor effort in data collation.

      1. Second point is about, how many points are there on the figure. This is a caste-group representation issue and can be inferred with reasonable number of groups IF they are selected properly (based on caste status and group population – rather than discrepancies like “kamma” labelled as “lower caste” and “reddy_telangana” given as 20 million population). These can be resolved by general public with minor effort in data collation.

        something like this is never a minor effort in india.

  14. most admixture estimation methods are much better at pinning “last date” than “first date.” honestly i am cautious until we have more ancient dna

  15. Ref of Vysyas:
    Narasimhan et al. 2019, supplement excel Table S5 , row 70, Cols O and P.

    the answer is in the supplemental page 307.

    i swear to god i answered this before for you, so this is annoying: THEY ARE USING THE ADMIXTURE DATE FOR THE WHOLE WEST EURASIAN-AASI POPULATION.

    this means if steppe is 0 the date is for AASI+indus periphery. if the steppe is 30% then the date is for the *average* of admixtures.

    for the vysa the signal is mostly indus periphery+AASI

    1. @Razib,
      Palliyars are not expected to have any steppe ancestry. Vysyas are modelled to have steppe admixture (Table S5, cols AB-AL) because they are expected to have steppe given caste status. But I see where you are coming from (i.e.., 5% steppe or less shouldn’t count).

      I read the exact text and interpret as this in Page 308 in supplements:
      However, for groups with both Iranian and Steppe ancestry, our dates reflect an average of the two mixture events, complicating interpretation of the inferred dates. We note that here and in what follows, we model admixture as an instantaneous process (a single episode of gene flow), even though in fact it could have occurred at multiple time points or gradually over a long period of time. Thus, one way to interpret the dates reported below is that they reflect the timing of major mixture of the ancestral sources in the history of a tested
      population and represent an upper bound on the time of the last gene flow as well as lower
      bound on the time of the earliest gene flow.

      1. if you read that text why did you bother asking the question? the obvious answer is the ‘major admixture’ is IVC+AASI as is true in every group in the south except brahmins jews and parsis

        1. I was assuming that 5% steppe admixture in Vysyas should be treated as “steppe intrusion” to Vysyas group. So, I interpreted it as dates indicating this. The text above states 8 groups of south indian dalit groups but didn’t say the same about trader-varna caste groups.

          But from your explanation, you are indicating that anything less than 5% steppe should be ignored as noise. The dates are more stable for about 30% steppe admixture with sufficient sample size.

          I am still looking in to what to consider about 15-6% steppe admixture all south indian caste groups. Where can the comfortable boundary between “likely noise” vs “this is signal” drawn needs looking to Narasimhan et al. formulation of Eq. 6-1, and comparison component variances.

          1. But from your explanation, you are indicating that anything less than 5% steppe should be ignored as noise. The dates are more stable for about 30% steppe admixture with sufficient sample size.

            it’s not noise…tho some ppl claim that the parameter estimates are off and we should DOWNWEIGHT steppe %. i don’t follow all the details cuz ppl have been saying this for a few years but no one has updated (the steppe % was initially higher with 2015 data).

            basically this is the way you think about it:

            if 5% steppe then about 90% of the west eurasian ancestry in vysa is IVC. so IVC+other (ASI/AASI) mixture events are going to dominate the signal

          2. @Razib,

            Thanks Razib. I get what you are saying. As a whole, the split between West Eurasian and AHG is being used for calculation of dates. So, the dates are giving the right admixture for last event of the mixture between AHG and West Eurasian. But West Eurasian is complicated by IVC+Steppe. Depending on the proportion of IVC vs. Steppe, the date is more attributable to the larger proportion component of the two.

            It is not about ignoring the steppe% as a number but the proportion of IVC vs. steppe split that dominates which event is being dated with the ALDER output.

  16. This isn’t convincing at all. The sheer amount of speculative verbs usage in this article is stark.
    This could be argued, author XYZ cautions hence QED, may well be argued, etc etc. No shit. No one had a camera back then, all of this is post-fact investigation. For all one knows Ashoka could be mongoloid East Asian looking fella. It Could-be.
    This article reminds me of a Joe Rogan classic, It’s Entirely Possible…

  17. Some points in response to JR:

    1. The date of earliest written record means nothing when dealing with a period that overlaps with the birth and spread of literacy. If there’s no writing on the steppe or anywhere close to the steppe, we wouldn’t expect proof of Indo-European languages even if they were there.
    2. If we take this as a valid argument, however, Hittite and Mycenaean Greek are attested from around the same time (or earlier) than Mitanni Indo-Aryan. If anything, this line of reasoning supports the Anatolian hypothesis, not OIT.
    3. Without a time machine, historical narratives are only conclusive to the extent that there is no plausible counter-narrative to explain the facts. But I can think of at least one plausible narrative to explain the spread of Indian fauna to the west that meshes with AIT. If Aryans did invade from the steppe, ties were likely maintained between Aryans in India and Aryans who remained in Central Asia, and probably some back migration as well (see below). Zebu cattle could have flowed out of India via this cultural network of related Aryan peoples. Perhaps they were better adapted to hotter and drier conditions in Central Asia than European cattle, and that’s why they caught on? The European cattle with significant Zebu DNA are from the regions which border and closely interacted with the steppe, not the parts where Indo-European languages dominate (which is virtually the entire continent). It could have been a simple process of diffusion (Aryans bring Zebu out of India > Zebu admixture becomes normal in steppe pastoralist herds > Zebu DNA spreads across the whole steppe region, and into adjacent places like SE Europe).
    4. RE: The Mitanni, they may be a back migration of Indo-Aryans out of India. In the absence of written records or currently available ancient DNA to validate this narrative, why is it plausible? Openness to long-distance migration is culturally inherited – in Migration Period Europe, the Germanic groups most likely to migrate were those who had a recent history of migration. Suppose I’m an Aryan living in the Indus Valley in the first generations after the invasion. I grow up hearing stories of the great deeds of my grandfather and his generation, and want to win that same glory for myself. But the whole Indus Valley is divided up among Aryan chieftains by now, and there’s a wall of jungle and desert to the east. Where do I go to win my own glory? Well, I’ve heard from Aryans who served as mercenaries in the Middle East, and some of those kingdoms sound as weak as they are wealthy. Why not gather up an army and do what my grandfathers did? And then once we’ve set ourselves up on a nice chunk of land along the Euphrates, we import some of the zebu and elephants that we grew up with back in India.

    1. Fraxinicus,

      1. If you dont have proof, you cannot assume that these groups were IE speakers and go about creating an elaborate theory, which has massive holes in it like the lack of archaeological evidence of Indo-Iranian migration from the steppe into South Asia and Iran. You should start looking for alternatives. You cannot be stuck with just one theory for decades.

      2. It does not support the Anatolian model. There is archaeological evidence of any Near Eastern cultural element in South Asia around this time. But we have Indian cattle, Indian Buffaloes, Elephants, spices, even Indian monkeys besides materials such as carnelian beads which land up in the Near East. You may want to ponder on this – Minoans from around 1700 BC develop a very elaborate hydrological system with wells, reservoirs, sewers, toilets, bathrooms etc. A parallel to such a system was nowhere in the Near East but it has close parallels with the Harappan hydrological system. You should wonder how.

      3 & 4. Only valid if you have clear proof that Indo-Iranians came to India from the steppe. In its absence, you reasoning is a case of special pleading. Zebu ancestry spread in Southern and Eastern Europe via the Near east and not the steppe. Iranian cattle are of largely Zebu ancestry. Zebu ancestry is also present in Iraqi, Syrian, Azerbaijani, Turkish, Levantine, Egyptian and some north African cattle. So the spread was certainly via the Near East and we now have aDNA to confirm that Zebu spread rapidly across the Near East in a post-2000 BC scenario. Curiously, Indian buffaloes also spread across this path into Europe and are present in all the above mentioned regions as well as in Italy, Greece, Balkans and the Carpathian basin.

      1. @JR:
        By your own standards, you have no proof that Indo-European languages spread from India. This entire debate operates in an area without definitive proof, because definitive proof would require either a time machine, or deciphered writing from the Bronze Age steppe or India.

        You emphasize archaeological evidence, presumably because you consider hard evidence from dig sites to be more conclusive than others. But even when you establish a valid archaeological connection, you have zero evidence that language spread along with those connections. There are plenty of examples of the spread of material culture without the spread of language. It cannot be assumed.

        If you’re going to reject any narrative without definitive proof, you have to reject your own as well, and conclude that we’ll never have a clear answer to this question. But in the absence of speculation based on archaeology, genes, or historical linguistics (all of which provide hints, and none of which provide “clear proof” about where a given language was spoken at a given time in the absence of written evidence), all we have is the evidence from modern Indo-European languages. And with ~10 primary branches of the family outside of India, and only one sub-branch of one primary family within India, speculation based on the here-and-now is even less kind to OIT.

        1. Good logic thinking frax. I will be making soon my concluding remarks related to the topic. It is unbelievable that a group which insists on origins of so-called ‘IE’ languages, who cites Talageri, which draws key conclusions from Rg Veda (e.g. Saraswathi) cannot answer the meaning of ‘rg’ and ‘veda’.

        2. A present for @Milan Todorovic who does not know Sanskrit, and claims to be an expert.

          Here is the explanantion from @Francesco:
          “Rg is derived from Proto-Indo-European *h₁ŕ̥kʷs, from *h₁erkʷ- (‘to praise’). Cognate with Old Armenian erg ‘song’), Tocharian A yärk, Tocharian B yarke (‘worship, veneration’).”

          Here is a better etymological explanation:

          The meaning of the Proto-Indo-European verb *h1erkw- > Proto-Indo-Iranian *Hark- was not only ‘to praise/sing’ but also ‘to shine’ (compare Old Irish erc- ‘sky’):

          “The word-form ṛg-veda is not the standard usage adopted in this text because only its components are attested: ṛ́c- ‘hymn; verse’, and véda- ‘knowledge’. Ṛ́c- is a feminine root-noun from the verb arc- ‘to shine, glitter; to glorify in song’ (compare masculine noun from the same root – arká- ‘a ray; brilliance; sun; laudatory song’). These etymological connections suggest that the notion of a hymn was associated with the idea of light and brilliance; in other words, IT WAS A CONCEPT CONNECTED WITH VISION” (T.E. Elizarenkova, Language and Style of the Vedic Ṛṣis, Albany, NY 1995, p. 14; emphasis mine).

          However, the meaning ‘to shine, be brilliant’ (cf. arcā́- [f.] `brightness, radiance’, arcí- [m.] ‘ray, flame’, arcín- [adj.] ‘shining’, arcís- [n.] ‘ray’, all attested in the RV) is rare in Vedic, while the noun ṛ́c- is quite well-documented in the RV, in both its singular and plural forms. The noun ṛ́c- is derived from Proto-Indo-Iranian *Hrk-/Hrč- (< Proto-Indo-European *h1rkw- ‘song of praise’, a derivative of the verb *h1erkw-), whose only possible reflexes in Old Iranian are some personal names in *ṛk- attested in external (non-Avestan) sources or “side-traditions”.

          Regarding Veda: It derives from *weid- ‘see, know (as a fact)’.
          There are 2 meanings. Both exist in Sanskrit. Only one (partially) exists in Slavic. It also exists in Iranian from which I suppose Slavics got their word.

  18. To complement some previous information regarding the language spoken by Aryans there is a small fraction of translated Serbian toponyms from Central Asia only. I will also translate toponyms from Tibet, China, Persia, Tarim Basin, etc, and in a play-off final – Iiiiindiaaaaa (and Rrrrrrrrrrrrrrrg Veeeedaaaaaaaaa).

    CENTRAL ASIA TOPONYMS (Afghanistan, Kurdistan, Bukhara, Turkmenistan):

    Sikira (=axe), Комаr (surname), Dub (=oak), Ruj (=red), Djura river (m. name), Kozar (=goatherd), Kuća (=house), Lipa (=linden), Kaljurna (=muddy), Čičak (=burdock), Kožan (=skin), Otrok (=child) river, Mala (=little), Kum (godfather), Guskin (=goose’s), Kučan (surname), Tabor (=camp), Šiban (=flogged), Šiprag (marshland), Seja (=sister), Čika (=uncle), Rosa (=dew, fem. name), Koliba (=hut), Čupan (=twitched), Deraguz (=flayed ass), Guz river (=ass), Jama (=cave), Mirča (, jar (=vehemence), Jaruga (=gully), Malić (surname), Brda (=hills), Blagoje (, Baja (, Tama (=darkness), Ugljan (=coal), Div (=giant), Danić (surname), Pajin (surname), Kula (tower), Laka (=light), Ruma (town in S.), Lisjak (=leaves=surname), MILAN, Komoran (surname), Kotoran (surname), Čir (=ulcer), Čika river (=uncle), Morava (river in S.), Duvno (place in S.), Neman river (=monster), Cer (=oak=mount in S.), Užice (town in S.), Risan (town in S.), Čedan (=chaste), Kozice (=little goats), Kosice (=little hair), Guba (=leprosy), Lija (=fox), Ljudi (=humans), Moma (, Vetar (=vind), Raž (=rye), Žaoka (=sting), Pljeva (=chaff), Sirote (=orphans), Sjenke (=shadows), Kuda (=where), Kiša (=rain), Tor (=sheepfold), Gusle (=music. instr.), Rosa (=dew), Sanjari (=dreamers), Čekić (=hammer), Kaljuga Gučeva (mount. in S.), Kurva (=whore), Gumno (haystack), etc.

  19. @JR

    It’s such fun, isn’t it, to fancy mighty treks of “Vedic” Mitanni Aryans travelling to Mesopotamia/Syria from India via Bactria and the overland route through Iran along with their herds of elephants and zebu cattle. Same as Hannibal’s army! :^)

  20. I am closer to Hindu nat views and in Razib’s camp (soft AMT/AIT whatever)

    Does that make me weird?

  21. This actually reminds me of the Big ‘Attested’ Slavic (i.e. Serbs) Migration to Balkan in the 7th cAC which has never happened. Logistic experts calculated that this procession would be 1500 km long (or 10 in parallel, 150 km each) but in the whole Europe no one noticed it, except of one lonely, (full)moron-stalker, who woke up from winter hibernation as a shameless liar what is not a surprise. It represents the total moral debacle of OIT that are afraid to assert who followed all these zebus, which also never reached the Balkan ranches, and few loose elephants. Why not? Because, this should be supported by evidence that only R1a zebuboys were allowed to ‘go west young men’ and spread their genes (and language) there. Otherwise, we would see some other zebuboys’ haplogroups in Europe or MEast or Scandinavia or Russia or anywhere. It seems that the knowledge of genetics was pretty high at Aryan times to select only R1a zebu-escort travellers.

  22. RK: “this happened with basques. their Y chromosomes are from yamnaya but genome-wide still farmer enriched”


    That is my opinion, too, and with very, very important conclusion which is also against the ‘mainstream’. Yamnaya (R1b) did NOT bring so-called ‘Indo-European’ language neither to Europe nor to SA. Because they were on much lower cultural-technological levels, they adopted the language of indigenous European people which originated in Vinca and was widespread in Europe (this is Sanskrit’s grandpa, too). The remnants of the original Yamnaya language is actually this ‘mysterious’ Basque language which is heavily enriched with Vinca’s language (at that time we can already call it – Serbian – this is actually what we today call – ‘Indo-European’ language, but the real McCoy one).

    I will write in details but just to say that this Basque language, so as Serbian too, has some words taken from newly-arrived Vinca’s G2a Caucasian ‘guest-workers’ who, in small numbers, followed I2a migrations to the west of Europe (Spain, even British Isles). That is, for example, the same name of the rivers in Kosovo and Spain (Ibar or Iber). There is a documented opinion that the name of Bosnia is also of G2a origin (meaning – the forest area).

    Btw, after Yamnaya intrusion to Europe and their genocide conducted on Vinca people, G2a also escaped, some to the west (Spain) others back to the old homeland – Caucasus. They brought Serbian language to Caucasus. For example, the name of the city of Tbilisi is of Serbian origin. As I said, G2a came as a sort of ancient ‘guest-workers’ to Vinca, where they were welcome because in this couple thousands of prosperous years there were neither wars in Europe nor pushing of some groups by other groups. Everything changed when future westerners (Yamnaya) came and started a string of wars and genocides up to today. G2a disappeared from Vinca and today there are no traces of them, neither in central Europe nor in Spain. A strong opinion is that Ossetins are descendants of G2a ancient Vinca’s refugees……..

    So many, exclusive, first time published information in this comment, isn’t it?

  23. I posted this comment earlier, but it is very appropriate here:
    Everybody seems to be talking about genetics. So here are some googlies:

    1. Bustan_BA, Dzharkutan1_BA, Sumbar, and Parkhai_LBA — Indo Aryan regions — have no Sintashta ancestry till about mid 2nd millennium and that too in one sample (I6667, Parkhai_LBA_o, 1497-1413 BCE)! These regions were already Indo Aryan starting 2nd millenium BC before Steppe ancestry shows up. They also don’t show R1a; I suppose AMT must be arguing for female mediated invasion starting from the 15th century BC. Astonishingly, Dzharkutan is the region from where Zoroastrianism descended. How do you explain this?

    2. Kangju that are mentioned to have invaded India can be a source of Steppe ancestry. According to Razib, “admixture models” show earlier ancestry of the Steppe in India. This is perfectly consistent with “ancient DNA” that shows widespread Indian ancestry in BMAC starting from the 3rd millennium BC. The admixture between Steppe and Indian ancestry population happened around 14th century BC and thereafter not in India but in BMAC. This is what the “admixture models” are picking up. So, the mystery is now solved.

    3. Not a single R1a has been found in Middle East from Mittani ruled territories. Historical records show that Mitannis were present in the middle east since the 18th century BC.

    4. While Razib continues on with the debunked “color theory,” he seems to not wade into the historical and linguistic analysis done by Elst and Talageri. This is what we call selective pickings.

    5. Rig Veda is a Bronze age text. And iron age in India started in the 16th century. Furthermore, according to linguistic analysis done by Talageri, Mittanis correspond with late Rig Veda. This pushes the date of Rig Veda to early 3rd millenium BC.

    6. Myceanean Greeks have J2 haplogroup. There is no R1b too. Though there are speculations that they may have R1b too.

    7. Hittites have J2 haplogroups. They also have no Steppe ancestry.

    If Jaydeep is reading this post, what is the best way to contact you? Should I post on your blog? I have other proofs too.

  24. Huh, it was a pretty exhaustive discussion and exchange in a last day or two. Thanks to the author of the intro text for his great effort and all contributors who wrote positively and in a good spirit. We collectively did a great job and pushed the envelope of our discussion by moving to the toponyms, languages and other related topics. Let’s now relax for a minute along some old tunes:

  25. @Arminius

    “What do you mean by claiming descent ? There are Rajputs in Pakistan Punjab, similarly small numbers of Rajputs moved to Central- Eastern India. No Munda claim any relationship, only those counted in census are accepted as Rajputs”

    No Rajputs accepts Pakistani rajputs as real , forget Pakistan, the rajputs dont see even ravana rajputs and Purbias as rajputs.

    “You still have some Gond rulers alongside Rajput rulers in Chattisgarh, Jharkhand etc, none of them are related to the neighboring Rajputs.”

    They Gond rulers claim Kshariya lineage through their marriage to rajputs.
    ” Rajputisation of Adivasi people

    Bhangya Bhukya notes that during the final years of the British Raj, while education introduced Westernisation in the hilly areas of central India, the regions also parallelly underwent the Hinduisation and Rajputisation processes. The Gond people and their chiefs started doing the “caste–Hindu practices” and frequently claimed the “Rajput, and thus kshatriya status”. The British empire used to support these claims as they viewed the adivasi society to be less civilized than the caste society and believed that adivasi peoples’ association with the castes would make the adivasis “more civilized and sober” and “easier for the colonial state to control”. Bhukya also points out that central India’s “Raj Gond families” had already adopted the religious and social traditions of the Rajputs before the British Raj in India, and there were “matrimonial relations” between a number of Gond and Rajput Rajas. However, the British empire’s policies of offering “zamindari rights, village headships and patelships” fueled the process.[24]

    According to Patit Paban Mishra, “the ‘ksatriyaisation’ of tribal rulers and their surroundings, resulted in the Hinduisation of tribal areas”.[25]”

    “How ? When kshatriya/ Rajput is a community like Brahmins is a community. There was no way for Chh Shivaji Maharaj to become kshatriya, this is why he had to claim a Sisodia descent for himself and it didn’t changed anything for him”

    I mean yeah, apart from him being claimed as a King ( and that being the only way he could do that) , it didnt change much 😛

  26. “i’m 99% you have a false memory about me, but you can prove me wrong. find the exchange.”

    @razib. i searched the archives cursorily, but couldn’t find the exact post. may be i do have just a false memory. but my point remains. this blog accords a disproportionately high importance to the questions of genetics of indians, which is not reflected in the public discourse in india.

    this is not to say that indians do not care about the issues of national identity. far from it. they are really obsessed with the identarian questions. it’s just that their intellectual slugfests mostly revolve around the axes of regions and cultural nationalism.

    i remain in touch with indian media. the average twitter hindu understands religion to some extent and can jot some lines on it. but most of them have no idea what y-dna or mt-dna means, so they simply bypass these debates. it’s just too esoteric for them. genetics appears to be a particular obsession of a select group of people on this blog.

    1. this blog accords a disproportionately high importance to the questions of genetics of indians, which is not reflected in the public discourse in india.

      Scorpion, I don’t want to get between you and Razib, but I just want to make the point that this blog was started by (and for) a Bangladeshi-American and a couple of former Pakistanis now resident in the West. There is no reason why it ought to reflect Indian public discourse. Also, most of us who have been around since the early days are shaukeen and not necessarily typical Indians with typical Indian obsessions.

      1. “There is no reason why it ought to reflect Indian public discourse. “

        because india is the 800 pound gorilla of the brownverse. one can’t bypass india and still be an authenticate brown voice.

      2. Ah it should really be brown Sahibs and not pundits? Why appropriate coolie terms then?

    2. “i’m 99% you have a false memory about me, but you can prove me wrong. find the exchange.”

      i found the exchange. (since it is a friday night and i have no life, i have nothing better to do).

      “I believe that the Reich lab will publish results which will falsify the beliefs of a substantial number of Indians about the nature of the origins of the native peoples of the region. This will shatter world-views, undermine mythologies, and rock peoples’ worlds.”

      i believe i remembered the gist of the message quite well even after 2 years.

  27. Tocharians who were there for atleast 7-8 centuries in the IAMC were displaced by Turkics and mongols. Many would have assimilated and taken refuge in India . Similarly kushanas and Scythians would have brought genes from the IAMC into India and formed the modern steppe signal in India. These were people who had some Indian admixture and indic cultural influence starting from the Bronze Age into historical times

    1. Indic cultural influence in the IAMC would ease their assimilation in to greater India during the historic period.

  28. I see the article and all the exchange is about proving or disproving ideas that were definitely first pushed and over time have been challenged and explanations are now retrofitted, and that the ideas stemmed from some sort of racial supermacy or identity crisis. But then why in 21st centry do we debate on such a subject on a website called “brownpundits”? Should the name not be free of the racial connotations and ideas that mostly drove things to us to fiercely debate on?

Comments are closed.