Out-of-India - From Theory to Truth

[Supratik]

Would you be able to say what was path-breaking in that paper other than fancy terminology like ASI/ANI that is not particularly convincing, given their numbers and selection. My comment was in response to your description of that paper being "path breaking".

If you go back I have already stated that it for the first time describes the founder elements of the population of the subcontinent. Those in the field or in related fields are excited by the findings. Now criticisms of findings are common. So you may have a different view and I may have a different one.

[Supratik]

There is a lot more than that they have to clarify. Why on earth are Siddis part of the selection? Heck they have only 130 samples and 5% or more are Siddis?

And unless I am mistaken the paper's diagnosis of ASI/ANI is based on distance from Onge of Andamans. Is that weird or is that weird? Is there any paper that suggests that "ASI genes" are closer to Onge?

Also I discovered today that the standard for "European Genes" (CEU) is the genetic map of several hundred individuals in Utah. What on earth does that mean? The same project seems to use "Gujarati genes" from Texas.

In the ultimate analysis it appears that Indian genes are closer to each other than others while there is a gradient from north west to south east. Given that migration was always from north west for tens of thousands of years the exact meaning of that paper is mysterious because India is a continent sized country.

Siddhis are a negative control i.e. they clearly shouldn't have the genetic constitution of people of the subcontinent. Negative controls are included in biological studies. Your study is limited by the volume of samples you can handle. Of course if you handle samples in the thousands or lakhs or millions your data is going to be more accurate but there are limitations. Onge are pure ASI i.e. there is no ANI mixture. The rest of the Indians in the study are a mixture. They do not talk about pure ANI. They may or may not exist. CEU and ANI form a clade i.e. in simple terms they are close to each other. The ultimate analysis is nothing but that the Indians studied are a mixture of two founder populations, the mixture is to different degrees among North vs South and upper vs lower castes, the tribes became castes and that there has been very little gene flow since the mixing i.e. they have been endogamous.

[shiv]

If you go back I have already stated that it for the first time describes the founder elements of the population of the subcontinent. Those in the field or in related fields are excited by the findings. Now criticisms of findings are common. So you may have a different view and I may have a different one.

Please don't misunderstand. I have not understood. What founder elements have been found? What has been found that causes so much excitement? What are these founder elements of the Indian population? Your post is even more mysterious than the paper.

I want to point out that linguistics caused much excitement, And the finding of R1a cause much excitement. All that excitement died out. What is this new excitement? In fact I have been unable to find what they define as "Ancestral South Indian". Will you be able to explain, or is the excitement related to fitting findings to a model and coming up with a pattern?

Siddhis are a negative control

They don't say that anywhere in the paper. How did you reach that conclusion? Control are better taken from a population that might not have interbred with Indians by living in India for centuries.

Onge are pure ASI

Any cites for this? This is the first time I am reading that the Onge of the Andamans are "pure ancestral South Indian" genes. I actually do not believe this statement. I think you are wrong but I am willing to take back this statement. You will have to come up with a few papers that state this.

[shiv]

, the tribes became castes and that there has been very little gene flow since the mixing i.e. they have been endogamous.

Supratikji this has been well known for a long time so this paper has found nothing new if this is a conclusion. Even that conclusion is not 100% right. I recall reading a paper about a comparison of a small community of Karnataka Brahmins that compared this. I think there is very little gene flow out of tribes, but there is gene flow into tribes meaning that tribal women are being screwed by upper caste men and having babies. But I will speak further only if I find the paper.

[RajeshA]

I think there is very little gene flow out of tribes, but there is gene flow into tribes meaning that tribal women are being screwed by upper caste men and having babies. But I will speak further only if I find the paper.

Some AIT-Nazis and AIT-Sepoys would say that the reason Bhils and other tribes in India have R1a is because of past rapes, and not because R1a really originated among the tribal population of India long time ago. I don't know whether we wish to make that kind of arguments.

[Supratik]

Please don't misunderstand. I have not understood. What founder elements have been found? What has been found that causes so much excitement? What are these founder elements of the Indian population? Your post is even more mysterious than the paper.

I want to point out that linguistics caused much excitement, And the finding of R1a cause much excitement. All that excitement died out. What is this new excitement? In fact I have been unable to find what they define as "Ancestral South Indian". Will you be able to explain, or is the excitement related to fitting findings to a model and coming up with a pattern?

They don't say that anywhere in the paper. How did you reach that conclusion? Control are better taken from a population that might not have interbred with Indians by living in India for centuries.

Any cites for this? This is the first time I am reading that the Onge of the Andamans are "pure ancestral South Indian" genes. I actually do not believe this statement. I think you are wrong but I am willing to take back this statement. You will have to come up with a few papers that state this.

The ANI and ASI are the founder elements. The excitement is that this hasn't been described/detected previously. May be it is difficult for someone not in the field to understand. You can use a tribe from say Arunachal instead of a Siddhi for the purpose as well. You can read the rest of the paper at

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2842210/

[Supratik]

Yes, RajeshA is right. Majumder's group has been publishing such things for a long time. IMO this rubbishes the claims.

[RajeshA]

As I said I wouldn't study Indo-Iranian speaking Pashtuns or Muslim Pathans in order to study the earliest genetic composition of the subcontinent. If they did I would like to know why.

Why not? Has the "earliest genetic composition" imprint vanished within the current Pushun population?

Some admixing with with Turks and Arabs would show up, but the earlier genetic composition would also be very visible!

[shiv]

I think there is very little gene flow out of tribes, but there is gene flow into tribes meaning that tribal women are being screwed by upper caste men and having babies. But I will speak further only if I find the paper.

Some AIT-Nazis and AIT-Sepoys would say that the reason Bhils and other tribes in India have R1a is because of past rapes, and not because R1a really originated among the tribal population of India long time ago. I don't know whether we wish to make that kind of arguments.

That is an argument that can be made. But that argument is a poor excuse for explaining why Indians are more closely related to each other than non Indian population and the genetic connections are deep and old. It is a political argument to explain away a racist theory. The fundamental premise of the Aryan Invasion theory was that the invading Aryans became the upper castes and should be genetically distinct from the lower castes while being genetically close to the European invaders.

In fact the admixture of Indians was recognized in the 19th century, even before genetics, and that served as an excuse for racist AIT Nazis to explain why they were superior. The previously superior Aryan invaders had interbred with the inferior Dravidians making themselves inferior. This is on record and has been linked in earlier pages of the thread.

Like linguistics, every finding has a special excuse.

No one has yet done a genetic study for example of Anglo Indians. Will the findings be a uniform mix of 50% Indian males with Europeans and 50% European women with Indians? In fact the likely finding is that the Y chromosomes will show British markers and the female mitochondrial DNA will show Indian markers. That could be explained as saying that the Brits only raped Indian women.

In fact it was more likely because a preponderance of British males with few women make them take Indian wives. So "rape by dominant class" is a political statement, and not fact.

[Supratik]

Why not? Has the "earliest genetic composition" imprint vanished within the current Pushun population?

Some admixing with with Turks and Arabs would show up, but the earlier genetic composition would also be very visible!

I wouldn't because due to the large scale historically recorded migrations into the region the modern Pashtun may not have anything much to do with the earliest Pathans.

[shiv]

The ANI and ASI are the founder elements. The excitement is that this hasn't been described/detected previously. May be it is difficult for someone not in the field to understand. You can use a tribe from say Arunachal instead of a Siddhi for the purpose as well. You can read the rest of the paper at

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2842210/

Sorry. The statement that a person outside the field cannot understand is the same one used by linguists to pass off fake theories that they are unable to explain with any degree of coherence.

The Reich paper is here
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2842210/

What that paper has done is to take CEU at one end as "European" and at the other end as Indian and have plotted the percentages of the genes from these two groups in different groups of Indians and have judged that some Indian are closer to CEU and some Indians are further away. Not Onge Onge is a separate branch. CEU, ANI and ASI are marked as branches separate from Onge. Perhaps you have not followed that? ASI is not Onge.

The paper does nothing to show any ancestral populations other than saying that some people are closer to European genes and calls them as having "ANI" and some people further and says they are "ASI"

In actual fact no genetic markers have been named. The paper means very little other than validating some model as a possible route of comparing populations. In actual fact European genes like R1B and I2 are virtually absent in India although this paper says nothing about that.

[RajeshA]

The proto-Slovenes and ANI really seem to have a historical bond! They are the two ends and between them one sees a host of other "Indo-European" populations.

One reason Slovenes are interesting is because they hardly have much haplogroup N3. I had written about this earlier as well.

So at two ends (India and Slovenia) one sees two R1a1a populations without the N3 and in the middle all the R1a1a populations (Slavs) have N3 in them.

This means that the Out-of-India migration of the proto-Slovenes happened before the Finno-Ugric N3 clade carrying population spread throughout Eastern Europe.

So basically the Slovenes set up a relative chronology with the Finno-Ugrics, when the first Slavs (proto-Slovenes, para-Indics) left India, crossed the Eurasia and settled down in the heart of Europe.

Now some may say, why do we want to think that it is the proto-Slovenes that migrated from India to Europe and not proto-Vedics who migrated to India from Europe!

Well if that was the case, then Indians would have gotten a lot more of the European clades from the Balkan region, which are also present in the Slovenian population, like the Haplogroup I would have also been found in India, which is not the case. I had spoken about this earlier also. And Hg I is considered older than Hg R1a1.

So to remember is

1. Slovenians have around 39% R1a1.
2. Thus Slovenians are tied to Indians.
3. Slovenians went to Europe before the Finno-Ugrics (Hg N3) spread into Europe.
4. Indians do not show any Hg I, and as such the direction of migration was from India to Europe.

[Supratik]

@Shiv

Boss, I don't like your condescending attitude. I have a Pchadee from a reputable US institute.

If you think it is all rubbish good luck to you. I will rather discuss with someone more interested.

[RajeshA]

In fact it was more likely because a preponderance of British males with few women make them take Indian wives. So "rape by dominant class" is a political statement, and not fact.

I would use "wife" and "raped" depending on whether the woman and her offspring later on are integrated into the man's "tribal" group or they remain in the female's "tribal" group respectively.

[shiv]

@Shiv

Boss, I don't like your condescending attitude. I have a Pchadee from a reputable US institute.

If you think it is all rubbish good luck to you. I will rather discuss with someone more interested.

Sorry if I hurt you boss. But I am unable to see what is "exciting" about that paper other than postulating an "ANI/ASI based on genetic distance from CEU. And Onge are not ASI. Either in that Reich paper or in any other paper that I know of. I have no problem if you don't want to discuss the issue. But I will dispute something if I disagree. Anyone is welcome to call my dispute as ignorance, but I am not afraid to be wrong.

That paper works on a hunch that assumes that India is made up of two ancestral populations possibly on the lines of Aryans and Dravidians. And the findings are adjusted to fit that hunch without any specifics. No. There is no specific evidence of an ancestral North Indian or Ancestral South Indian population even in the Reich paper. There is just a gradient, and evidence of great genetic intermixing in India with no pointer towards any specific ancestor out of Europe.

It will not be possible for you or anyone else to glean anything more than that from the Reich paper.

[Anantha]

I think there is very little gene flow out of tribes, but there is gene flow into tribes meaning that tribal women are being screwed by upper caste men and having babies. But I will speak further only if I find the paper.

Some AIT-Nazis and AIT-Sepoys would say that the reason Bhils and other tribes in India have R1a is because of past rapes, and not because R1a really originated among the tribal population of India long time ago. I don't know whether we wish to make that kind of arguments.

That could also mean upward social mobility (via marriage/courtship) for tribals and lower varnas based on their karmas, the way the varnas were intended.
Do not think that it was a one man-one woman relationship in older times.

[RajeshA]

Just a note:

When we discuss Ancestral North Indians (ANI) and Ancestral South Indians (ASI) we need to take a few things under consideration.

1. We are talking about populations, ANI and ASI, that do not exist today.
2. Today's North Indians are not ANI and they do not nor can represent the culture of the ANI as sole representatives. Today's South Indians also have a genetic and cultural stake in this.
3. Today's South Indians are not ASI and they do not nor can represent the culture of the ASI as sole representatives. Today's North Indians also have a genetic and cultural stake in this.
4. The model ANI-ASI is important in order to explain how the Europeans originate in India but DO NOT carry many markers present in the current Indian population. These would have been the ASI markers. This model in fact strengthens OIT by providing an explanation - that the Indo-Europeans left India from a region at a time when in that region there was hardly any mixture of ANI and ASI. In other regions this may have started.
5. ANI is NOT Indo-Aryans. The genetics just do not add up. ANI were in India tens of thousands of years ago (~40,000 YBP).
6. ASI were also in India tens of thousands of years ago (~50,000 YBP).

[RajeshA]

I think there is very little gene flow out of tribes, but there is gene flow into tribes meaning that tribal women are being screwed by upper caste men and having babies. But I will speak further only if I find the paper.

Some AIT-Nazis and AIT-Sepoys would say that the reason Bhils and other tribes in India have R1a is because of past rapes, and not because R1a really originated among the tribal population of India long time ago. I don't know whether we wish to make that kind of arguments.

That could also mean upward social mobility (via marriage/courtship) for tribals and lower varnas based on their karmas, the way the varnas were intended.
Do not think that it was a one man-one woman relationship in older times.

I can't understand what karma has to do here! When women marry, they go and live with their husbands, and do not remain in their own family/group/tribe.

[shiv]

I can't understand what karma has to do here! When women marry, they go and live with their husbands, and do not remain in their own family/group/tribe.

Rajsh I think this is OT for the thread. But as pointer there was a famous story in Kannada called "Samskara" made into a movie. Here is the storyline
http://en.wikipedia.org/wiki/Samskara_% ... #Storyline
Caste relations can be quite complex.

[RajeshA]

Caste relations can be quite complex.

Thanks. I read the story-line. Complex indeed! Too many rules!

[shiv]

[*] The model ANI-ASI is important in order to show how the Europeans originate in India but DO NOTcarry many markers today present in the Indian population. These would have been the ASI markers. This model in fact provides OIT a chance.

Rajesh there may be no ANI ASI at all. Even the Reich paper does not explain how Dravidian speaking tribals have 40% "ANI" while upper caste Indo European speakng have only a little more at 60%

But my biggest objection to what you write is as follows. Please think about it:

The Reich study was done specifically to prove that ANI/ASI do exist.
North Indian upper caste genes of Indo European speakers were specifically chosen to represent North India.
South Indian tribals who speak Dravidian languages were specifically to represent South Indians.

In other words there was a selection bias from the start to try and prove something. The suggestion was that Northern Upper caste speaking an IE language was of Aryan descent and South tribal speaking Dravidian would represent the two extremes postulated by the Aryan invasion theory.

But what actually happened was that there was not a big difference between the two. The findings do not support the idea that a migrating population of Indo Aryan speakers displaced Dravidian speakers to create the caste system. The highest and lowest castes and the IE and Dravidian speakers all share a broadly similar percentage of genes.

What the authors have done is to call that proportion of genes that are similar to the European CEU as "ANI" and the dissimilar part as ASI. In other words as per the Reich paper if you have a slightly higher proportion of genes smilar to CEU (European), then you are "Ancestral North Indian". In other words North Indian==European similarity. Upper Caste==European. Therefore Aryan invasion may be true. That is what caused all the excitement despite the ambiguous findings.

I would reject the ANI/ASI terminology, I have been unable to find any other papers that use that terminolgy or have markers for ASI or ANI

[RajeshA]

I have been unable to find any other papers that use that terminolgy or have markers for ASI or ANI

Metspalu et al. (2012) uses the terminology ANI as well. They are a different group than Reich's.

[A_Gupta]

In other words, if you were an alien from outer space, and sampled Y chromosomes of the hominids of modern India and surrounds the first thing you would find is geographic gradients of both genes, AND, genetic diversity. The genetic diversity gradient tends to show the peninsula and the area north of it as the area with the most diversity. The genetic gradients run both nort south as well as east west, which indicates that mostly people were sedentary, in an area most of the people are descended from mostly the Paleolithic people who lived there, who spread out in the Paleolithic, establishing these gradients. Of course, super posed on this is some evidence of movement and mixing.

If you did not know India's history, you would not be able to reconstruct the tribes, castes, language groups from genetics.. After all, the methods of the discussed papers starts with such groups and seeks something that differentiates them. Genes that do not differentiate the predetermined groups are uninteresting and are dropped. The groups overlap more than their "means" differ from each other.

[shiv]

If you did not know India's history, you would not be able to reconstruct the tribes, castes, language groups from genetics.. After all, the methods of the discussed papers starts with such groups and seeks something that differentiates them. Genes that do not differentiate the predetermined groups are uninteresting and are dropped. The groups overlap more than their "means" differ from each other.

Precisely random "Indians" are never selected. It is always about caste and Aryans versus Dravidians. The assumption is that there were "racial" differences in the first place and that those racial differnces can be easily found by the selecting caste groups. The Reich paper does not show any major differences. It shows a gradient. However the gradient is shown only in a graph but in the text that is passed over under the blanket description of ANI for one end of the gradient and ASI for the other end. So the conclusion is that ANI and ASI have been proven.

[A_Gupta]

People may find the maps available from here to be interesting:
http://wals.info/chapter

[Vayutuvan]

Conveniently quoted in gol-gol-kola Telugu text so no one can tell.

Again "chakra" seems to have an IE cognate in kyklos/cycle/circle. Different from gol, ratha and mandala.

One more thing - Chakram is a direct unchanged descendant of Sanskrit Chakra and the same in Hindi as well. What about other dravidian languages? My guess is that they would have a word almost unchanged from Sanskrit.

[Anantha]

[/quote]
I can't understand what karma has to do here! When women marry, they go and live with their husbands, and do not remain in their own family/group/tribe.[/quote]

RajeshA,
What I meant by Karma in this context is, upward mobility of people based on what they had flair for, not necessarily based on birth. This would explain the genetic distances between Brahmins and tribals. What this means is over time, several groups of people intermixed showing the genetic markers within India as we see it now.
Ananth

[A_Gupta]

Mario Alinei, The Celtic Origin of Lat. rota and Its Implications for the Prehistory of Europe, "Studi Celtici" 3, 2004, pp. 13-29.

http://www.continuitas.org/texts/alinei_rota.pdf

Emphasis added.

Two words for ‘wheel’ in Indo-European languages: the solid and the spoked wheel

As is known, there are two word families for the ‘wheel’ in IE languages: the Gr. kýklos type and the Lat. rota type. The most plausible explanation for this remarkable synonymy was already advanced by Heyne (1901), and later resumed by Weijnen (1974) and by the writer (Alinei 1996ab): the kýklos type might designate the more primitive and earlier, disc wheel, and the rota type the more advanced and recent, spoked wheel. The two names would thus correspond to two completely different wheel technologies, and to two entirely different periods of prehistory.

Let us first review, and in part elaborate, the main arguments for the interpretation of the rota family as the name of the ‘spoked wheel’, and let us begin by recalling its membership (cf. Pokorny (IEW 866) and Buck (1949 §10.76):

Lat. rota (It. ruota, Fr. roue, Sp. rueda, Rum. roată ecc.); Germ. Rad, Du. rad; Lith. rãtas m. ‘wheel, circle’ (pl. rãtai ‘carts, vehicle’), Lett. rats (Plur. rati ‘vehicle’), (> Finn. ratas ‘wheel’); Ir. roth (m.), Welsh rhod (f.), Bret. rod. Gaul. Roto-magus (Rouen); Skr. rátha- (m.) ‘chariot’, Av. raθa ‘idem’.

On the semantic level, note that the meaning of the Indo-Iranian word is not ‘wheel’ but ‘chariot’, namely a vehicle used by chieftains and their noble entourage of warriors (besides as ritual vehicle).

This has two implications, one technological and the other linguistic one: on the one hand the ‘chariot’ as such necessarily implies a light construction, and thus horse traction and a spoked wheel (Piggott 1992, 56). On the other the change of the word meaning from the original ‘wheel’ to ‘chariot’ – while the Indo-Iranian wheel’s name remained the original PIE one – points to a loanword, rather than to an inner development (cp. the Baltic development).

Note also that the semantic development from OHG rad ‘wheel’ to ON rođull ‘Strahlenkranz, Sonne’ can be understood only if one recalls that the spoked wheel was often associated with a solar cult, and that in the Romance area the name of the ‘spoke’ - which continues Lat. radius - is the same as that of the sun-ray (Alinei 1974).

[ukumar]

[*] The model ANI-ASI is important in order to show how the Europeans originate in India but DO NOTcarry many markers today present in the Indian population. These would have been the ASI markers. This model in fact provides OIT a chance.

I would reject the ANI/ASI terminology, I have been unable to find any other papers that use that terminolgy or have markers for ASI or ANI

Shivji, Reich used ANI/ASI terms for the first time. They were the first to find a way to identify different ancestral components. It is just a symbolic name of the two ancestral genetic components found in Indian population. They were able to do by comparing against Onge. You will not find any reference prior to them. Their main achievement/innovation is to develop such method.

What the authors have done is to call that proportion of genes that are similar to the European CEU as "ANI" and the dissimilar part as ASI. In other words as per the Reich paper if you have a slightly higher proportion of genes similar to CEU (European), then you are "Ancestral North Indian". In other words North Indian==European similarity. Upper Caste==European.

I think you are misreading the paper. ANI refers to the part of the genes which seems to be derived from some ancient population. It doesn't refer to the person. In other word person can't be "Ancestral North Indian". There is no such person exist today.

People have used the ANI to equate with Arya but they are misinterpreting the data. Riech expected to find clear signal to differentiate Brahmins from cast but they didn't find it. Since present north Indian population is decended from IVC (population replacement can't be justified based on archaeological continuty), it is given that they had ANI components. We can only speculate how much ASI components they had.

There was also a problem in Reich's method. He was wrong in showing clade between ANI and CEU. The actual clade is between ANI and west Asian. ANI seems related to neolithic people who populated much of the Asia in last 10000 years.

Anyways Reich paper is now outdated and one should look at the Metspalu paper. They broke the genetic components in further ancestral components and analyzed major Eurasian population.

His K6 corresponds to Reich's ASI and K5 corresponds to ANI. You would see that K5 is found in India, central asia, europe and caucaus. It is clearly a component associated with Neolithic population growth started from Mehrgarh in IVC since that is where it is found most.

Some people argue that minor european K4 component found in India is signal of Arya migration. But I think they are wrong. It is too small to cause language shift in vast area. It is more likely result of the invasion/migration in historic times. If we assume that it is signal of vedic Aryas, it is most likely that they also had majority K5 component in them. So at most you can put them to central asia near BMAC but not in Steppe. Note that Metspalu explicitly rejected intrusion from Steppe (kurgan area) to India due to lack of K7 in India.

[shiv]

Shivji, Reich used ANI/ASI terms for the first time. They were the first to find a way to identify different ancestral components. It is just a symbolic name of the two ancestral genetic components found in Indian population. They were able to do by comparing against Onge. You will not find any reference prior to them. Their main achievement/innovation is to develop such method.

What the authors have done is to call that proportion of genes that are similar to the European CEU as "ANI" and the dissimilar part as ASI. In other words as per the Reich paper if you have a slightly higher proportion of genes similar to CEU (European), then you are "Ancestral North Indian". In other words North Indian==European similarity. Upper Caste==European.

I think you are misreading the paper. ANI refers to the part of the genes which seems to be derived from some ancient population. It doesn't refer to the person. In other word person can't be "Ancestral North Indian". There is no such person exist today.

Thank you for a credible explanation. I have actually plotted the data on a separate graph (which I will put up in due course - its on another ffline machine)

I wondered how Reich and co could take some North Indians and describe 40-50% of their genetic make-up as "Ancestral South Indian" and take some South Indians and choose to describe 40 to 50% of their genetic make up as "Ancestral North Indian". What is South Indian about 40-50% of all North Indian genes? What is North Indian about the remaining 50-60%?

Clearly it was a politically incorrect label that I believe was derived from a desire to prove the hypothesis (as the paper says) that any population can be divided up into two ancestral components.This is what the paper says:

We developed a model to study the historical relationship of Indian groups to those worldwide, based on the hypothesis that most groups can be approximated as a mixture of two ancestral populations followed by group-specific drift

The selection of samples was divided into Upper caste and lower caste, IE and Dravdian speaking. No explanation is given as to why this was done other than the introductory sentences that reveal a suspicion based on some papers that India consists of two ancestral populations and that those ancestral populations are surmised on the Aryan Invasion theory that says that Indo European speaking high caste form one ancestral population and the others form the rest

The paper claims that it has been written to answer the following questions

We analyzed these data to address five questions. Does India harbor more substructure than Europe? Has endogamy been long-standing in Indian groups? Do nearly all Indians descend from a mixture of populations? Is the ancestry of tribal groups systematically different from castes? What is the origin of the indigenous Andaman Islanders?

For this purpose they have selected those groups who they think will reveal significant differences Upper and Lower Caste, Indo European and Dravidian speakers. The study never had the intention of actually looking for South and North genetic differnces. This is how the selection was done. North and South are not even mentioned.

We genotyped 132 Indian samples from 25 groups. To survey a wide range of ancestries, we sampled 15 states and 6 language families (including 2 language families from the Andaman Islands12) (Table 1 and Figure 1). To compare traditionally “upper” and “lower” castes after controlling for geography, we focused on castes from two states: Uttar Pradesh and Andhra Pradesh.

But yet, at the end of it all, the paper comes out with the astounding conclusion hat the have "found" two ancestral population genetic lines called "Ancestral North Indian" and "Ancestral South Indian". I will upload the graph of the data. No such conclusion is tenable to differentiate North and South. Why the name ASO and ANI? It is a politically loaded name that differentiates North and South after studying caste and language. You cannot look at the findings and call them South Indian or North Indian ancestry. At best one can say there were likely to have been two separate parent populations

And in the discussion of the paper at the end none of the five questions that the study was claimed to have sought to address are even mentioned. See the questions above and the discussion below

We have documented a high level of population substructure in India, and have shown that the model of mixture between two ancestral populations ASI and ANI provides an excellent description of genetic variation in many Indian groups. A priority for future work should be to estimate a date for the mixture, which may be possible by studying the length of stretches of ANI ancestry in Indian samples45,46, and will shed light on the process leading to the present structure of Indian groups. A second priority should be to discern the details of the history of the ANI and ASI before they mixed, including the date of their separation and their history of expansion and contraction; this may be possible by analyzing allele frequency spectrum47 and linkage disequilibrium data45,48,49. Our findings finally have medical implications. By showing that a large proportion of Indian groups descend from strong founder events, these results highlight the importance of identifying recessive diseases in these groups and mapping causal genes.

What kind of paper is this? Is this really a reputable study from a reputable genetics journal?

[RajeshA]

[*] The model ANI-ASI is important in order to show how the Europeans originate in India but DO NOTcarry many markers today present in the Indian population. These would have been the ASI markers. This model in fact provides OIT a chance.

Rajesh there may be no ANI ASI at all. Even the Reich paper does not explain how Dravidian speaking tribals have 40% "ANI" while upper caste Indo European speakng have only a little more at 60%

But my biggest objection to what you write is as follows. Please think about it:

The Reich study was done specifically to prove that ANI/ASI do exist.

I agree. They selected certain markers which they collected under ANI and another set of markers which they collected under ASI.

By dividing a set into two parts, one would get two sets. They got ANI and ASI. These were the two sets they set out to get.

North Indian upper caste genes of Indo European speakers were specifically chosen to represent North India.
South Indian tribals who speak Dravidian languages were specifically to represent South Indians.

In other words there was a selection bias from the start to try and prove something.

I agree (in a qualified sort of way).

If I were to look at their choices of groups and populations abstractly, I would say, well that is their choice. It depends on what theory they wish to prove, what assumptions they make, etc. If they are upfront and honest, then they are free to prove what they wish, and we are free to criticize their methods, their data and their interpretations.

I assume they are using castes as a unit of study, because of their endogamous nature, and as ready-made social constructs for study of the structure of Indian population.

Reich et al. do dabble a lot in caste and region to make their points, and to find a "cline". But that is the exact point they are trying to make. That there is indeed a cline to see, as one moves from North to South, from upper caste to lower caste.

What I do find a little disconcerting is when they say on page 493

The 18 Indian Cline groups all have between 39% and 77% ANI ancestry on the basis of f₃ Ancestry Estimates (Methods), which we quote because it has the smallest standard errors (Table 2). ANI ancestry is significantly higher in Indo-European than Dravidian speakers (P 50.013 by a one-sided test), suggesting that the ancestral ASI may have spoken a Dravidian language before mixing with the ANI. We also find significantly more ANI ancestry in traditionally upper than in lower or middle caste groups (P50.0025), and find that traditional caste level is significantly correlated to ANI ancestry even after controlling for language (P50.0048), suggesting a relationship between the history of caste formation in India and ANI–ASI mixture

The ANI-ASI cline does not really disturb me on the issue of language. I would expect that those who have more Ancestral North Indian ancestry would be speaking North Indian languages.

What however disturbs me is that those with higher ANI ancestry were found to be generally of higher castes. Now they say that the effect of language (region) has been compensated for. I don't know, but a lot more testing needs to be done to know whether this really is the case.

The suggestion was that Northern Upper caste speaking an IE language was of Aryan descent and South tribal speaking Dravidian would represent the two extremes postulated by the Aryan invasion theory.

We really have two ways to define our identities in this context - genetically and/or linguistically.

1) If each of us defines himself/herself through his genetic profile, then one would have to accept that we are mixed, and that there is really no way to make a clean split - only ANI or only ASI.

2) If however one identifies himself over one's language and culture, then it can be the case that one would feel the inclination to tend more towards ANI or ASI.

Through this linguistic and cultural identification, the additional association of ANI with higher castes, and ASI with lower castes, as has been postulated by this paper, can cause certain discomfiture, because the identification what had started out with language and culture ends up being an identification with caste.

But what actually happened was that there was not a big difference between the two. The findings do not support the idea that a migrating population of Indo Aryan speakers displaced Dravidian speakers to create the caste system. The highest and lowest castes and the IE and Dravidian speakers all share a broadly similar percentage of genes.

I agree. However ANI is not being necesarily considered as a population which moved in into the Indian Subcontinent recently as is the case with Aryans according to the AIT/AMT model, and displaced any ASI population. Both ANI and ASI are considered as two past populations, resident in India since a long time. This paper does not speak of any displacement - on the contrary the paper says, there has been a lot of mixing.

What the authors have done is to call that proportion of genes that are similar to the European CEU as "ANI" and the dissimilar part as ASI. In other words as per the Reich paper if you have a slightly higher proportion of genes smilar to CEU (European), then you are "Ancestral North Indian". In other words North Indian==European similarity. Upper Caste==European. Therefore Aryan invasion may be true. That is what caused all the excitement despite the ambiguous findings.

I think here we need to distance ourselves from any genetic definition of modern Indians as ANI or ASI based on their ANI-ASI ratio. Modern Indians are a combination of both components.

I agree that in order to find out exactly what constitutes ANI component or ASI component per their markers, comparison with CEU on the one hand and Onge of Andamans on the other hand were undertaken.

Now what we do get through this model is that ANI was indigenous to India, and thus Indo-European languages were developed in India, and from India, not only languages but also speakers of these languages moved out through Central Asia into Europe.

So I think that part of AIT, the AIT-Nazis can really forget.

However the equation of higher ANI ratio leads to higher caste does disturb me. Perhaps here one can intervene and say that those groups with the highest ANI percentage - Pushtuns and Sindhis are not Hindus at all, much less be of higher caste. In fact even our history is witness to a certain scorn among Vedics towards those in our Northwest as they did not observe the Vedic rituals, so even then Pushtuns may not have been regarded as "upper castes".

I would reject the ANI/ASI terminology

I have not been able to decide whether to keep it or reject it. On the one hand this helps explain why the Europeans who share some clades with Indians, do not have any genetic markers often observed among South Indian populations. On the other hand, caste hierarchy is being shown to correlate with ANI-ASI cline, which has its own risks.

Also CEU may have similarity with ANI but it doesn't say anything about caste. The CEU in fact lies outside Indian social structures. Furthermore there is also no pure CEU population. That too is mixed with other West Eurasian populations.

Even if one were to accept that ANI-ASI cline reflects caste hierarchy in India, one must be clear that this is only a statistical correlation and should not be treated as higher ANI component as deserving of consideration as higher caste. These are two different things.

[shiv]

What I do find a little disconcerting is when they say on page 493

The 18 Indian Cline groups all have between 39% and 77% ANI ancestry on the basis of f₃ Ancestry Estimates (Methods), which we quote because it has the smallest standard errors (Table 2). ANI ancestry is significantly higher in Indo-European than Dravidian speakers (P 50.013 by a one-sided test), suggesting that the ancestral ASI may have spoken a Dravidian language before mixing with the ANI. We also find significantly more ANI ancestry in traditionally upper than in lower or middle caste groups (P50.0025), and find that traditional caste level is significantly correlated to ANI ancestry even after controlling for language (P50.0048), suggesting a relationship between the history of caste formation in India and ANI–ASI mixture

What however disturbs me is that those with higher ANI ancestry were found to be generally of higher castes.
.

Rajeshji. This is PRECISELY what I am questioning based on the data. I have plotted the data in another less confusing graph. But first a few explanations are required

1. Reich et al had 132 samples, of which they had 15 Paki Pathans and 10 Paki Sindhis (It says so in the paper). None of these 25 have been tabulated in the results, so my graph too excludes them.

2. That leaves 107 samples. Of these 20 are excluded as being either Siddi or Onge. That leaves 87 samples

3. Of these 87, about 10 are rejected in the results as not being significant, leaving a total of 77 samples. I have tabulated those 77 samples' results according to the percentage of "Ancestral North Indian" genes the link below. Please click to open in a separate tab or window.
https://docs.google.com/file/d/0B3JNY4I ... RVRWs/edit

The graph represents 77 samples in 14 groups. The top two are upper caste IE speaking (9 samples). The bottom two are "lower caste Dravidian" - 7 samples. These are the two extremes with 60 to 70% ANI for the former and about 40% for the latter.

The middle 10 groups represent 80 % of all the samples that were used for data analysis Look at this middle 80 % group carefully. There is no clear correlation between caste, language and ASI. Lower caste IE scores higher than upper caste IE. Tribal IE scores higher than upper caste Dravidian. Dravidian tribal has more ANI than lower caste IE speaker. Nobody has anything less than 40% ANI. Not even the tribals, low castes, Dravidian speakers. Nobody. And nobody has anything more than 60% ANI. The ANI/ASI mix for the middle 4/5th ths of that list is approximately 50-50 (40 to 60%)

The data indicates that for 80 % of the significant samples, the ASI and ANI mix is about the same. High caste or low. IE or Dravidian speaking. I am unable to see any caste or linguistic hierarchy in these samples that make up the bulk of the samples.

The selection was from caste groups and IE/Dravidian speakers in Andhra and UP only. The results too are tabulated according to caste. How was the name ASI and ANI given to their genes? That is a complete and unexplained mystery.

How did these geniuses come up with the names "Ancestral North Indian" and "Ancestral South Indian"? Surely there should be at least one graph saying who is north and who is south. You and I are Indians and know which is north and south when we are given place names. But a non Indian would never know. So where did the terminology come from - with no explanation or logic for the names ASI and ANI? The results of the study are sorted according to caste and linguistics, but the conclusions are stated to be Ancestral North Indian or Ancestral South Indian. Even then the data do not fit the conclusions. How did they work this sleight-of-hand? Is it possible that the authors have a hidden bias that makes them attribute "North Indian" to upper caste IE speaking and South Indian to lower caste Dravidian speaking. There are two Indian names among the authors. I would put it as possible that these authors may have that inbuilt AIT based bias in the way they view India's population history because of their Macaulayite education like the rest of us. This paper has made a Freudian slip and is passing off that Freudian slip as population genetics of India.

What kind of pir reviewed paper is this?

[KLP Dubey]

Lat. rota (It. ruota, Fr. roue, Sp. rueda, Rum. roată ecc.); Germ. Rad, Du. rad; Lith. rãtas m. ‘wheel, circle’ (pl. rãtai ‘carts, vehicle’), Lett. rats (Plur. rati ‘vehicle’), (> Finn. ratas ‘wheel’); Ir. roth (m.), Welsh rhod (f.), Bret. rod. Gaul. Roto-magus (Rouen); Skr. rátha- (m.) ‘chariot’, Av. raθa ‘idem’.

I am curious to know who are the Latin, Germanic, Finnish, and Welsh grammarians of the ancient times who could tell us what the roots of this word are in their native language. But of course, such grammarians are not to be found, which is why there is an urgent need for a PIE "root" to cover up the fact that these cultures were all using the Indian word in a corrupted form.

On the semantic level, note that the meaning of the Indo-Iranian word is not ‘wheel’ but ‘chariot’, namely a vehicle used by chieftains and their noble entourage of warriors (besides as ritual vehicle).

Stupid and wrong assertion. When there is no "Indo-Iranian" language then how can there be an Indo-Iranian word ? The Rgvedic sounds are "ratha" and "cakra". The Indians have attributed the word "ratha" to the root "ra~nH" (to hasten/speed). The meaning of "ratha" is therefore "that which hastens/speeds". Not necessarily a chariot. The two other roots suggested - but not substantiated - are "rap" and "ras". For "cakra", the suggested roots are "cak" (repel), "car" (move), or "kram" (go).

The Iranians have no grammar or glossary for the Avesta, so where are the Iranian meanings coming from? From Ahura-wazoo ?

It is in the Atharvaveda and Brahmanas that the word "ratha" starts to describe a chariot specifically. These documents are Indian, not "Indo-Iranian".

This has two implications, one technological and the other linguistic one: on the one hand the ‘chariot’ as such necessarily implies a light construction, and thus horse traction and a spoked wheel (Piggott 1992, 56). On the other the change of the word meaning from the original ‘wheel’ to ‘chariot’ – while the Indo-Iranian wheel’s name remained the original PIE one – points to a loanword, rather than to an inner development (cp. the Baltic development).

The Indians assigned the meaning of "chariot" to "ratha", presumably because they invented the chariot itself. Others started to copy the invention and its word. That is the technological as well as linguistic implication. The author has some convoluted and muddle-headed explanation which ends up saying the same thing that has been known for 3000+ years.

This reminds me of the post by ManishH wherein he cited someone that claimed they had "restored the correct pronunciations in the Rgveda" and thought it was a big deal, whereas exactly the same common knowledge is mentioned 3000+ years ago in the Rgveda pratishakhya so many times that even a blind man could read it.

Note also that the semantic development from OHG rad ‘wheel’ to ON rođull ‘Strahlenkranz, Sonne’ can be understood only if one recalls that the spoked wheel was often associated with a solar cult, and that in the Romance area the name of the ‘spoke’ - which continues Lat. radius - is the same as that of the sun-ray (Alinei 1974).

Such associations begin in the Brahmanas, not in Latin.

KL

[shiv]

Here is a link to my earlier graph using Reich et al's data posted again for reference
https://docs.google.com/file/d/0B3JNY4I ... RVRWs/edit

Below is the same data showing state of origin so you can figure out North Indian or South Indian. This data is not specifically tabulated anywhere in the Reich paper. I have done it manually. The percentage represents percentage of ANI (Ancestral North Indian) genes

Once again if you remove the outliers and look at only the middle 10 groups representing 80% of all the samples you find no special North-South distribution.

Code: Select all

Kashmiri Pandit	Kashmir	70.6%
Vaish	 Uttar Pradesh	62.6%
----------------------------		
Meghawal	 Rajasthan	60.3%
Srivastava	Uttar Pradesh	56.4%
Velama	   Andhra Pradesh  54.7%
Tharu	     Uttarkhand	  51%
Naidu	    Andhra Pradesh	50.1%
Lodi	    Uttar Pradesh	49.9%
Kamsali	Andhra Pradesh	44.5%
Chenchu	Andhra Pradesh	43.2%
Satnami	Chhattisgarh	43%
Bhil	Gujarat	42.9%
-------------------------------		
Madiga	Andhra Pradesh	40.6%
Mala	Andhra Pradesh	38.8%

The presence of the 4 outlier groups of two upper caste North Indian (Pandit and Vaish - total 9 samples), and two lower caste South Indian (Madiga and Mala - 7 samples) with 60-70% ANI in the North and 40% ANI in Andhra are the only suggestion of a north-South bias in ANI/ASI. And these two groups constitute 16 out of 77. (about 20%)

80% fall in the middle group which shows no North-South ANI/ASI bias

What's going on? The abstract of the paper that was widely published and quoted by half the planet has little connection with the data.

Someone please tell me where I am wrong without giving me mysterious excuses that I cannot know unless I am in the field of genetics. The data is pretty plain for all to see.

The data says that 80% Indians seem to have "western Eurasian" genes (whatever the fuk that means) and non Western Eurasian genes in a nearly equal mix no matter what caste or region or language group they belong to. Only 20% of Indians seem to have a bias with about 60-70% ANI or 40% ANI.

Other than Kashmiri Pandits no Indian has less than 40% ANI or ASI

How the hell did they give the names ANI and ASI to genes that all Indians have in huge percentages, never less than 40%?

[harbans]

The more i read about this, the more i think and am convinced, the real origins would be placed east of the Indus/ Saraswati..probably on the plains of the Ganges or the Himalayas. The Aryan is nothing but a Noble peoples. Noble..equating vaguely to Dharmic value systems in dealings and ethic. For some reason even Voltaire assumed the same..everything came to us from the banks of the Ganges..

early Indo-Europeanists of the Enlightenment Voltaire,Immanuel Kant, Schlegel believed deeply that India was the 'Urheimat' of all Indo-European languages. In a 1775 letter, Voltaire : “I am convinced that everything has come down to us from the banks of the Ganges, - astronomy, astrology, metempsychosis, etc. It is very important to note that some 2,500 years ago at the least Pythagoras went from Samos to the Ganges to learn geometry...But he would certainly not have undertaken such a strange journey had the reputation of the Brahmins' science not been long established in Europe.”

What started this confusion is a simple assumption/ axiom by Latham..the place of origin of a language is closest to the place of it's greatest diversity. I find that hard to digest. That law would not hold in many cases of behaviour and political systems of enforcement/ command/ influence. Thats why latter scholars were searching nearer the points of greatest diversity in Europe than India..then it came closer and closer. An my bet is the earlier scholars were right..Voltaire was right. A group of Brahmins reciting the Veda's by the Ganges at Dawn seems more plausible than even reciting it on the banks of Saraswati..yet the Vedas are a kick start kind of mechanism to Human civilization..is my reading. I am not going to wait all my life for the linguists to come to a firm conclusion about the origins of the Veda's. I believe Krishna in BG, Vyasa and others who said so for no personal profit or religious motif or fame and wealth. There was a reason why the Brahmins of yore shunned wealth, fame and more. A few lwords in the Mittani treaty with the Rig Veda convince me on the contrary even more that the origin was meant to be in warmer climes..the recitation and practice by the banks of a serene river. Not some cold desolate steppe. Are there any major versions XYZ of the Veda's? Nope. There is one and it is to be delivered with the correct stress on each syllable explained to the minutest detail. And that could only have happend East of the Indus..Ganges. JMT and Voltaires of course.

[shiv]

On the subject of a Celtic origin for the word "wheel" here is something about Celtic mythology
http://neuralmodel.net/library/more/hom ... druids.htm

Druidism designates a vast intellectual, technical & spiritual tradition common to all Celtic societies, that was lost, not through Romanization, but because of Christianity. Celtic society is recognizable around the 5th century bc, slowly swallowed up by Ceasar's conquest of Gaul, but surviving into the 6th century ad in Great Britain. Celtic was a single tongue of Indo-European origin, but divided into two branches: Gaelic and Britannic (Gallic, Breton, Cornish, Welsh).

Key to Druid philosophy is the concept of Monism - never did there exist the slightest difference between the sacred and the profane. All action was sacred. Druids made the decisions, the king merely acted. Decisions were made to put the society closer to the masterplan, the way of the Gods, realized in the Otherworld.

<snip>

Beliefs and Ethics & Values. In short:

Life is all and all are responsible for protecting it.
Action has a consequence that must be observed and you must be prepared to compensate for your actions if required.
Uphold the Truth, starting with yourself. Be sure in your convictions, particularly when judging or accusing someone, but also when debating.
Make an honest living. Work with high standards. Serve your community. You live in society and are bound by its rules.
Be a good Host as well as a good Guest.
Maintain a healthy Balance of the Sacred and the Mundane.
etc

The following festival name really tickled me..

July 31 - August Eve
Lugnasa (loo-nus-uh)
* Lugnasa is a symbol for the formation of family, marriage.

lugna-lagna anyone?

[shiv]

I have been unable to find any other papers that use that terminolgy or have markers for ASI or ANI

Metspalu et al. (2012) uses the terminology ANI as well. They are a different group than Reich's.

Thanks. I had seen this paper earlier, and read it in some more detail today. This is a much better constructed paper in terms of aims and conclusions.

The terms ANI/ASI appear twice. Once with reference to the Reich paper and once again where the findings of the Reich paper are not disputed. This paper of course thrown a huge new perspective, and I will need to read it twice more at least to get details that I may have missed.

There is an important supplement to that Metspalu paper with all the dtata which are dead links in the Scribd site. that data is here
http://www.biotech.ebc.ee/Publications/ ... lement.pdf


One things that strikes me on reading this paper is that all these people are held down by established beliefs that genetic spread did actually occur from Central Asia to India. This has not been helped by the fact that (as mentioned in this paper) in most studies until recently "South Asia" was represented by Pakistanis. I suspect that Pakistani authorities have readily allowed researchers to walk in and take tissue/blood samples unlike India which has more strict laws about who and why that type of sampling can be dome. I recall about a decade ago an American company via an Indian American was ready to pay us (doctors) at a particular hospital for cancer tissue samples to be placed into a liquid nitrogen container immediately after harvesting from a patient on the operating table. The conniving hospital owner was ready to allow this (the Indian American contact was a distant relative) - but I (and others) refused because even at that time we expected that any new finding leading to a gene based drug would get patented in America and cost billions for us while we doctors would have cheerfully earned a few peanuts to give away local genetic material to American gene harvesters with no accountability.

Fortunately Indian laws were put in place soon after. But in the meantime researchers have used Pakistani samples to stand in as examples of "South Asia/India" and earned the peanuts paid or got vijjas in return. The Metspalu paper shows how this is wrong

From the paper:

this is powerful
evidence that Pakistan is a poor proxy for South Asian
genetic diversity, despite having often fulfilled this role in
previous publications.

Pakistanis have taken "We are India onlee" to great lengths under our very noses while acting like they are not India.

But as the above paper shows it is the Pakistani population that has very little genetic diversity (and antiquity) compared with India and it is the Paki population that shows the greatest similarity with West Eurasian populations.

Like i said, i could have told anyone that from a reading of history. But nobody asked me.

[Arjun]

Pakistanis have taken "We are India onlee" to great lengths under our very noses while acting like they are not India.

But as the above paper shows it is the Pakistani population that has very little genetic diversity (and antiquity) compared with India and it is the Paki population that shows the greatest similarity with West Eurasian populations.

Like i said, i could have told anyone that from a reading of history. But nobody asked me.

Quote from Metspalu paper:

Genome-wide scans on the Human genome diversity panel (HGDP) data involving 51 global populations have revealed that South Asia, represented by Pakistani populations, shares most signals of recent positive selection with populations from Europe, the Near East, and North Africa.23 Given the environmental differences between Europe and Pakistan and the possible depth of human habitation in South Asia, this result is surprising, but considering the lack of Indian data it remains to be determined whether South Asian-specific signals of positive selection do exist.

So, just to confirm, Metspalu et al affirm that Pakis and Indians don't even have much to do with each other genetically?

Can someone clarify the 'South Asian-specific signals of positive selection' part?

[RajeshA]

Another way Pakistan does not represent India is because even if they use Pakistani genetic data (Pakjabis, Sindhis, Baluchis, Pushtun, etc.) and correlate with European data to look for Indo-European migration dynamics, they cannot change that Pakistan really represents zero culture of any Indo-Aryans who may have migrated into or out of the Indian Subcontinent.

So on the cultural front - language, mythology, original way of life, they would still have to refer to the Indic heritage.

Genetic analysis is only interesting when one can determine the cultural, linguistic, ethnic backgrounds of the population groups studied. Pakistanis do not really make any cultural statements.

[RajeshA]

The following festival name really tickled me..

July 31 - August Eve
Lugnasa (loo-nus-uh)
* Lugnasa is a symbol for the formation of family, marriage.

lugna-lagna anyone?

लग्न, is often used for wedding, विवाह!