Tuesday, January 12, 2010

286. Aftermath

I've already considered some of the genetic evidence to support the notion of a Toba-induced bottleneck, but let's dig a bit deeper this time. It's important to understand that the usual signs of a genetic bottleneck, in the form of reduced haplotype diversity, won't apply in this case, because the initial Out of Africa exodus would itself have produced a very significant bottleneck, making detection of a second bottleneck only a few thousand years later virtually impossible. There are, nevertheless, other unusual aspects of the Asian genetic picture that are very difficult to explain unless such a bottleneck in fact occured. Whether we are content to accept Toba as the cause, or feel forced to reject that theory, it's very difficult to explain away all the many signs that, regardless of the cause, a major bottleneck, centered in India, did in fact occur at a very early, and crucial, period in human history.

Some of the key problems are discussed in an unusually thorough and critical study by Richard Cordaux et al, Mitochondrial DNA analysis reveals diverse histories of tribal populations from India, 2003. While it is often assumed that the tribal peoples of India are directly descended from the initial wave of Out of Africa migrants, Cordaux et al found little sign of that:
Our analyses of mtDNA variation in tribal populations of India indicate that groups in different geographic regions have different demographic histories. In general, southern tribes have reduced mtDNA diversity and mismatch distributions strongly indicative of recent bottlenecks. The distinctiveness of southern groups is also emphasized by the MDS analyses and AMOVA. However, it is difficult to distinguish from these data between old and severe bottlenecks or more recent and less severe bottlenecks (my emphasis).
The picture they found is, in fact, more consistent with -- you guessed it -- a gap.
three typical east-Asian mtDNA haplogroups (A, B and F) are absent or virtually absent from non-northeast India . . . Furthermore, the fourth typical east Asian mtDNA haplogroup M has a different structure in India as compared to other Asian areas. This suggests that, although they show close affinities, the east Asian and Indian mtDNA gene pools are fairly distinct. This result is consistent with the suggestion that the east Asian and Indian mtDNA pools have been separated from each other for about 30 000 years (262 -- my emphasis).
To extend the context of their research, the authors refer to another study of special interest for our purposes, Phylogenetic Star Contraction Applied to Asian and Papuan mtDNA Evolution, by Peter Forster, et al (2001), in collaboration with the noted archaeologist, Colin Renfrew. These authors consider a very interesting question that we haven't yet discussed,
the question of whether modern European, Asian, Papuan, and Australian mtDNA types derive from an uninterrupted demographic expansion of the out-of-Africa founders (strong Garden of Eden model), or whether an initial expansion was followed by the formation of regional gene pools, which, after a period of isolation and drift, expanded demographically and geographically to form the present mtDNA variation in different continents and regions (weak Garden of Eden model) (1864).
These authors also see a gap, which for them suggests a weaker version of Out of Africa, in which a "period of isolation and drift" is followed by migrations in various directions to produce the differentiations, genetic and otherwise, with which we are now aware. While their notion of a "weak Garden of Eden model" strikes me as rather vague, the gap they see makes considerable sense, especially since they are among the few to notice the importance of Melanesia, and in particular, New Guinea, for the Out of Africa model: "The oldest expansion in Eurasia occurred 65,000, + or - 23,000 years ago . . . and is witnessed by mitochondrial descendants preserved in Papua New Guinea" (1875 -- my emphasis).

For them, this original expansion (consistent with the Toba date of ca 74,000 ya, if we take the + or - 23,000 seriously) is "about 20,000 years older than any mainland Asian cluster, although both the Papuans and the Asians are derived from the same two Eurasian founders [M and N -- VG]." On this basis, they propose a remarkable scenario "to account for the obvious phenotypic differences between Papuans and Asians despite their sharing a common mitochondrial ancestry":
The M and N founders derive from a single African migration but split at an early stage (possibly before reaching Europe, which lacks M) into proto-Papuan and proto-Eurasian. The proto Papuan M and N immediately expanded demographically and geographically along a southern route until reaching Papua New
Guinea, thus allowing Papuans to retain their overall genetic similarity to Africans (Stoneking et al. 1997). Meanwhile, proto-Eurasians spent 20 or more millennia genetically drifting to their present distinct European, Indian, and east Asian M and N types, as well as phenotypes (compare the common Papuan/Eurasian melanocortin receptor variants in table 1 of Harding et al. [2000]), long before expanding (1875).
Aside from the authors' fixation on New Guinea, by no means the only place in what could be called "greater Southeast Asia" to perpetuate both very old mtDNA haplotypes and very African cultural practices (not only P/B but also remarkably sophisticated visual art traditions), their scenario makes a good deal of sense. The "weak Garden of Eden" scenario is rarely discussed anymore, probably because it's so vaguely defined and so difficult to test. (What would it mean for proto-Eurasians to "genetically drift" for 20,000 years?) But the Toba bottleneck scenario might well make it unecessary, because the equivalent of 20,000 years of drift might have occured overnight.

In the aftermath of a disaster such as the Toba eruption, many groups in the path of the huge, thick ash cloud would not have survived. For those who did, life would have drastically changed. For one thing many if not most, if not almost all, of their population may have been killed outright, simply suffocated in a sea of ash. Some might have been in a position to retreat to the depths of certain caves, where the worst effects of the ash cloud might not have penetrated. When emerging, after a few days, weeks, months or even years, they would have been faced with a world largely depleted of both vegetation and wildlife. To get a sense of what life would have been like in an environment suddenly deprived of almost all the usual sources of food, water and communal support, we can consider the fate of the group Colin Turnbull, in his book The Mountain People, called the "Ik," whom I wrote about as follows (in Post 176):
Ik society was, at that time, undergoing severe stress, and the stress had very definite and very dire effects on a people who had become increasingly desperate with hunger and other forms of deprivation, to the point that their cultural values were disappearing into a mode of existence based, as one might expect, on the philosophy of "every man for himself" aka "dog eat dog."

(to be continued . . . )

21 comments:

Maju said...

three typical east-Asian mtDNA haplogroups (A, B and F) are absent or virtually absent from non-northeast India . . . Furthermore, the fourth typical east Asian mtDNA haplogroup M...

Apples and oranges.

A, B and F are part of N (which is the only comparable thing to macro-haplogroup M, even if N shows a basal diversity that is about 1/4 of that of M). This is the kind of problems you meet when using obsolete materials.

Also all the three mentioned lineages are now known to exist in South Asia, albeit in small amounts (ref. for mtDNA A, ref. for B, F and other N).

Additionally there is some evidence that the "bottleneck", if anywhere (more like a simpler and less dramatic "funnel"), is at the origin of expansion into Eastern Eurasia, for example the latest HUGO consortium research. There are many papers that have not detected any "bottleneck" in South Asia whatsoever (for example).

Maju said...

Europe, which lacks M...

Another understatement. There is M in Europe, specially in the North (C, D and Z specially).

... allowing Papuans to retain their overall genetic similarity to Africans (Stoneking et al. 1997).

This is totally misleading, fundamentally wrong. Papuans are dominated by mtDNA P (a subclade of R) and Y-DNA M (a subclade of K). In this sense they would be closest to peoples such West Eurasians or East Asians, which largely share such patterns (IMO indication of "late" colonization). But, well, what can you expect from a paper of 1997!

Manju Edangam said...

ref. for mtDNA A,
Maju:
Is it observed among SE Asians and Australian aborigines?

DocG said...

Thanks for the references, Maju. But the first one is also "obsolete" by your standards, as it was published in 2002. And the presence of A as reported there is miniscule. Also, in the same report they refer to
"genomic evidence that the Dravidian-speaking tribals
may have arrived in India after the Austro-Asiatic speaking
tribals." And since the AA speakers are probably immigrants from E Asia, where does that leave Tribal India?

You have a tendency to dismiss any results you don't like that are a few years old as "obsolete" but you uncritically accept recently published results (that you like), although clearly they too are destined to become obsolete in a few years.

The bottom line is that the entire field of pop. genetics is at an early stage and in flux, with the latest results routinely questioning those that came before. What most seem to have in common are samples that are probably too small -- but that will improve over time.

Meanwhile, based on Oppenheimer's work I am exploring a hypothesis to see whether or not there is evidence to support it and I am indeed finding some. Whether this evidence will hold up in the long run is an open question, but I think we are at too early a stage in pop. genetics to come to any hard and fast conclusions regarding any of this. At this point, I am pleased to see that there are many signs in the literature pointing to something problematic about the status of the Indian tribals as OOA survivors, signs that seem consistent with the Toba scenario.

DocG said...

Maju: "There are many papers that have not detected any "bottleneck" in South Asia whatsoever (for example)."

The paper you cite is interesting. Thanks for tracking down that reference. But take a closer look. Fig. 1 lumps all of Eurasia together, so any potential differences between South and East Asia are going to be lost. Another problem with this type of comparison is that many of the East Asian populations may also have been affected by the same bottleneck. It's only among those groups now marginalized and living in remote refuge areas that we are likely to find haplotypes that would theoretically cluster closer to Africa (in other words, not so much affected by the bottleneck), and most of those groups are probably not represented at all in surveys of this kind.

I don't see meaningful results coming from this type of research until we have larger and also more finely tuned samples, especially from smaller, marginalized indigenous groups.

DocG said...

Maju: "This is totally misleading, fundamentally wrong. Papuans are dominated by mtDNA P (a subclade of R) and Y-DNA M (a subclade of K). In this sense they would be closest to peoples such West Eurasians or East Asians, which largely share such patterns (IMO indication of "late" colonization). But, well, what can you expect from a paper of 1997!"

If you read my post again you'll see that I placed this paper in context, making no secret of its relatively early date and the even earlier date of the Stoneking reference. And I did not uncritically accept their findings. My main interest in this paper was its presentation of the "weak Out of Africa" model, which as I see it reflects certain problems due to the gap I've been discussing.

You seem very sure of yourself in characterizing "Papuans," but to my knowledge there has never been an adequate study of Papuan genetics, something that is badly needed. Typically we find either all N. Guinea natives lumped together as one, which makes little sense, or we find samples divided into "Highland" and "Coastal," which would be better if the samples were large enough, which they never are. Even when specific groups are sampled, which is extremely rare, the samples are usually very small and not enough groups are included. The vast majority of N. Guinea populations have never been studied at all by geneticists.

More thorough sampling has been done by Friedlaender for New Britain and New Ireland and some smaller Melanesian Islands but to my knowledge nothing of comparable completeness and detail has yet been done for New Guinea. If you know of such a study, please share that info, as I've been looking for one for some time.

Nevertheless, just about all researchers in this area, geneticists and ethnographers alike, agree that the Highland peoples of New Guinea represent some of the longest established populations in the world outside of Africa. The consensus is that they are the direct descendants of the original settlers of the island, as reflected not only in their genetic makeup, which can be distinguished from that of the Austronesian speaking coastal people, but their culture generally, including their languages, many of which are isolates.

The status of the Indian tribals in this respect is very different. While they are sometimes referred to as descended from the original OOA migrants, the evidence for that is much less clear. And linguistically there is no comparison, since all of India contains only 4 different language families, with maybe one or two isolates. And each of the 4 families appears to have its origin outside of South Asia. Even Dravidian is thought to have migrated into this region along with agriculture, from points west. And many of the tribals may have migrated there along the same routes.

Eventually the genetic picture for India is going to have to be reconciled with all the other evidence, regardless of how clear things might seem to you today.

I don't know what to make of the association you've found between West Eurasia and N. Guinea, which seems bizarre and is probably an artifact. If it's for real, then it can only reflect a very old and very deep association, and is unlikely to be a sign of "late colonization." That seems a huge stretch, Maju.

German Dziebel said...

"A, B and F are part of N (which is the only comparable thing to macro-haplogroup M, even if N shows a basal diversity that is about 1/4 of that of M). This is the kind of problems you meet when using obsolete materials."

In addition, as I pointed out (with references) earlier, South India is now known to have haplotype D, which is another East Asian lineage in a sisterly position to all other "M" lineages. http://www.phylotree.org/tree/main.htm.

"Additionally there is some evidence that the "bottleneck", if anywhere (more like a simpler and less dramatic "funnel"), is at the origin of expansion into Eastern Eurasia, for example the latest HUGO consortium research."

This study's conclusion are based on higher diversity in SE Asia vs. EAsia. Diversity measures are eff pop size dependent, hence SE Asia can be a site of a major differentiation of mtDNA lineages but not their source.

Maju said...

I told you before, Victor that the paper of Cordaux is not too helpful. You choose to follow that line because it fits better with your preconception (Toba catastrophe hypothesis) but that's a risk you assume and that I have been saying since some time ago is not good for your research because it makes it, quite artificially, dependent on some hypothesis or conjectures that are not really helpful.

Anyhow, minuscle or not, the haplogroup, which is not normally found out of America and NE Asia (and hence the Austroasiaic hypothesis is feeble), it is in fact found in South Asia. And so are B and F. You just prefer to subestimate the genetic diversity of the subcontinent because of your preconceptions at your own peril anyhow.

The bottom line is that the entire field of pop. genetics is at an early stage and in flux...

A little less every year and even every month but certainly there is still a lot to learn (specially in South Asia) and hence I prefer to err in the cautious side than in the partisan one. And, as you are not too knowledgeable on the matter, I would advise you also to be cautious and not take sides without clear evidence.

It's only among those groups now marginalized and living in remote refuge areas that we are likely to find haplotypes that would theoretically cluster closer to Africa.

There is no such thing. I have seen already a gazillion materials and there is nothing "closer to Africa" in any East Asian or Australo-Melanesian population. They are just their own kinds of Eurasians. If you keep reading on genetics, in one or two years from now, you will acknowledge that I am right (at least largely) but you first need to improve your knowledge on the matter because you are really too easily misled by the occasional paper or original researcher who says whatever approaches your Toba catastrophe fetish model.

Even if Oppenheimer would be right, the evidence is far from clearly supporting such a model. Therefore the theory is not too popular, in contrast to the post-Toba rapid coastal migration model, much more mainstream.

I don't marry either model, not even other alternative models of my own. I just say: watch the bumps.

I don't see meaningful results coming from this type of research...

Was just meant as a counter-example. Most papers don't see any bottleneck, certainly not one affecting specifically South Asia. Many see South Asia at the origin of Eurasians as a whole.

Maju said...

You seem very sure of yourself in characterizing "Papuans," but to my knowledge there has never been an adequate study of Papuan genetics, something that is badly needed.

I'll see what I can find in terms of papers but AFAIK Papuans are the best studied aboriginal ethnicity (or macro-ethnicity) in the whole region. They are a recurrent reference in all kind of genetic studies of global, Eurasian and Oceanian scope. Their Y-DNA, mtDNA and autosomal DNA is pretty well known.

So I am rather surprised of your claim in this aspect.

The status of the Indian tribals in this respect is very different.

They have never been so extremely isolated, indeed.

Even Dravidian is thought to have migrated into this region along with agriculture, from points west. And many of the tribals may have migrated there along the same routes

Language migration is not the same as peoples' migrations. Languages are like clothes that we use and throw away, genes are like our hands or hearts, which we can't ever replace. Whatever the case most of South Asian mtDNA is clearly autochtonous and a good deal of Y-DNA is as well. Even Y-DNA R1a is now argued whether it could have originated in South Asia, where some of the apparent oldest clades exist.

Not just tribals (except Austroasiatic ones probably but mostly on the Y-DNA side) are surely autochtonous but most of South Asian ancestry looks native and pretty much unique.

I don't know what to make of the association you've found between West Eurasia and N. Guinea, which seems bizarre and is probably an artifact.

MtDNA-wise is something quite clear. Y-DNA-wise too at least at the level of macro-haplogroup K (right now it is debated whether there is a sublineage in K, MNOPS, that emphasizes that link or not). This can hardly be any artifact because it is phylogeny.

Also, morphologically Papuans look like "caricatures" of the Caucasoid phenotype and I know someone who was persuaded that they had a strong direct link with West Asians specifically for that reason (I could not agree with that though but I do think that there is a Caucasoid-Australoid continuum that represents best the original Eurasian phenotype range, with Mongoloids being the outliers and Papuans representing one of the aspects of this range).

It is anyhow tiresome to give lessons on such basic materials for free. I'm not asking for a pay, of course, but I really feel that my efforts are being wasted here (or that they could only bear fruit after too much of a titanic effort to be worth it).

DocG said...

Maju: I told you before, Victor that the paper of Cordaux is not too helpful."

Yes, and if I saw you as the ultimate authority on such matters, I'd take your evaluation more seriously.

A little knowledge is dangerous, Maju. You may know more than I do about genetics, but I have a feeling people like Cordaux and the many others whose work you dismiss (some of the leading people in the field) know a lot more than you. And if you want to dispute that, then please, send me a link to your list of publications, because I haven't noticed anything with your name on it or seen any example of original research you've done.

Why should I take the word of an amateur over that of a professional with real credentials in this field?

And sure, since I am looking for evidence that might support my hypothesis I pay special attention to certain research and less to others. If you have any experience with real science and real scientists you'll notice that we all do that. Sure it can be dangerous to ignore findings that don't support your own theory, which is why I go out of my way to read whatever I can regardless, and I take everything seriously. But when I am developing a theory, I naturally refer to the literature that tends to support it, why wouldn't I?

If you can give me an example in the scientific literature of a theory based on evidence that contradicts it, I'd be really curious to read that. Which doesn't mean that one should ignore contradictory evidence, not at all. And I most certainly don't ignore it. I don't ignore you, either.

But I don't see you as the ultimate authority either and will continue to be guided by my own sense of what seems logical and whatever I can cull from the literature that makes sense to me.

German Dziebel said...

"There is no such thing. I have seen already a gazillion materials and there is nothing "closer to Africa" in any East Asian or Australo-Melanesian population. They are just their own kinds of Eurasians."

Agree. I probably went through the very same pile. Hence, Victor's "African signature" in music is not well supported by genetic data.

"Anyhow, minuscle or not, the haplogroup, which is not normally found out of America and NE Asia (and hence the Austroasiaic hypothesis is feeble), it is in fact found in South Asia. And so are B and F."

In addition, M was detected where it's normally absent, namely in human remains from British Columbia. (Mitochondrial haplogroup M discovered in prehistoric North Americans, by Ripan S. Malhi et al.) They assigned it to a "modal" type, which may mean an American-specific lineage of M. This weakens the SE Asian hypothesis for the origin of M but strengthens the overall "Oriental" (again, not African) profile of South Asians.

Maju said...

A little knowledge is dangerous, Maju.

Apply that to yourself, Doc.

I am certainly not the "ultimate authority" on genetics nor anything else but I know certainly much more than you do and I have all the time trying to be eclectic and help you to expand your horizons.

Why should I take the word of an amateur over that of a professional with real credentials in this field?.

Why should you take the word of one professional over another one, as you are doing all the time?

Plus Oppenheimer is also an amateur: he is a pediatrist.

But when I am developing a theory, I naturally refer to the literature that tends to support it, why wouldn't I? -

Because you are leaning to one interpretation among many possible. If this interpretation would happen to be wrong all your theory would crumble like a castle of cards.

You are putting the horses before the cart, what means that your own theory has great risk of being flawed because of your capricious choice of supports.

...

In addition, M was detected where it's normally absent, namely in human remains from British Columbia. (Mitochondrial haplogroup M discovered in prehistoric North Americans, by Ripan S. Malhi et al.).

True. But a lot of people questions the validity of aDNA, so a single finding of this kind is not too strong evidence for anything.

Whatever the case, 2/4 (or 2/5 if we consider X2) Native American lineages are part of M (C and D).

DocG said...

docG: "A little knowledge is dangerous, Maju."

Maju: "Apply that to yourself, Doc."

I do. All the time. But you don't. I'm probably just as arrogant as you, but I don't go around insisting I know more than the experts, because clearly I don't and I know I don't.

I don't dismiss evidence that appears to contradict the hypothesis I'm exploring and I'm willing to admit that I could be wrong. You dismiss evidence that doesn't suit you all the time and you are not willing to admit you could be wrong. So there is an important difference between us, yes.

It's true that I focus on evidence that might tend to support my hypothesis, but sorry I can't think of a single theorist who doesn't. I take all the evidence into consideration, but until I am confronted with a clear falsification I will persist.

It looked for a while as though Toba had been falsified and I took that seriously. For a long time I played down the importance of Toba and tried to find some other type of event that could have a similar effect. But when I read that Petraglia had found artifacts that looked African to him I decided to take Toba seriously again. It could still be wrong, but at this point I see it as the most likely cause of the anomaly I see in the cultural evidence. You simply dismiss Petraglia when his conclusions don't suit you but are happy to point to him when they do.

DocG said...

Maju: I am certainly not the "ultimate authority" on genetics nor anything else but I know certainly much more than you do and I have all the time trying to be eclectic and help you to expand your horizons."

When you are being helpful I appreciate your efforts. When you are being dismissive, especially when you dismiss legitimate research by real experts that doesn't suit you, that's not helpful and I don't appreciate it, no. And when you insist that I accept your judgement, with the implication that you couldn't possibly be wrong, that's just annoying.

"Why should you take the word of one professional over another one, as you are doing all the time?"

I don't take anyone's word for anything. I present evidence in support of the hypothesis I'm exploring because that's what any theorist is expected to do, find supporting evidence. But I also consider any problems I see with that evidence, and I consider contradictory evidence as well. What you want me to do is accept your view of things, but I can't do that because there are essential questions that you have failed to ask and as a result the theory implied in your comments doesn't work.

"If this interpretation would happen to be wrong all your theory would crumble like a castle of cards."

First of all I'm not afraid to be wrong. Second of all, only that aspect of my hypothesis would crumble. Falsification of the bottleneck would have no bearing on the aspects of my thinking pertaining to Africa alone, to HBP/HBC or even HMP/HBC. Nor would it have any bearing on the "African signature" I've found, which would be there bottleneck or no.

Thirdly, it's not MY theory that's threatened, but the entire OOA apparatus, which can work only if an explanation can be found for the gap. I am exploring a hypothesis that has the potential to explain the most serious anomalies in the OOA model. You can't or won't see those anomalies so there's really no point in arguing with you about it. But I see them. And others do as well. And they won't go away on their own.

DocG said...

Maju: "You are putting the horses before the cart, what means that your own theory has great risk of being flawed because of your capricious choice of supports."

If you read my posts again you'll see that there is no one piece of evidence that I depend on. And I'm always willing to admit that any of these reports could be contradicted by new evidence at any time. Some of it already has. But that's OK, because the whole field is now in flux so what is contradicted today may well be supported tomorrow. Meanwhile, I see evidence that's consistent with the hypothesis and for now that's enough for me.

The bottom line is that some explanation has to be found, for the gap I see, but also for the anomalies in the genetic evidence (questions over the meaning of all those M haplogroups, for example) -- and also for the large scale population patterns, based on such obvious morphological differences, differences that cannot be explained on the basis of the smooth continuities you see.

Notice, by the way, that, unlike the situation with archaeology, you never see real disputes among the geneticists, despite the very different results they come up with. They too realize that now is the time to explore and not expound.

German Dziebel said...

"True. But a lot of people questions the validity of aDNA, so a single finding of this kind is not too strong evidence for anything.

Whatever the case, 2/4 (or 2/5 if we consider X2) Native American lineages are part of M (C and D)."

Evidence is evidence. Of course we need more evidence.

D is a sister clade of M (see Phylotree), or it can be called as an outlier in the whole clade. "M" is just an empty label. They could have named this higher order macrohaplogroup "D" just as easily. C is indeed "part of M" but again American Indians have several deep-rooting C's. A new one was recently found. Look it up on Dienekes. The authors (the same lab from UC Davis) believe this finding is consistent with the finding of M in the ancient remains of the same general area of the Northwest.

German Dziebel said...

"Thirdly, it's not MY theory that's threatened, but the entire OOA apparatus, which can work only if an explanation can be found for the gap. I am exploring a hypothesis that has the potential to explain the most serious anomalies in the OOA model. You can't or won't see those anomalies so there's really no point in arguing with you about it. But I see them. And others do as well. And they won't go away on their own."

Funny: Luis believes in out of Africa and doesn't care about gaps. Victor sees a "gap" but tries to salvage out of Africa. German sees several gaps that make out of Africa unlikely and offers a new solution. And everyone seems to be happy with his vision.

Maju said...

So saying that mtDNA A, B and F do exist in India with due references is "dismissing evidence", saying that A, B and F, are not comparable to M (N would be) is "dismissing evidence"?

You're taking a more and more one sided viewpoint. It is your problem: your eggs and your single basket, your cards castle with too feeble foundations...

When you are being helpful I appreciate your efforts. When you are being dismissive, especially when you dismiss legitimate research by real experts that doesn't suit you, that's not helpful and I don't appreciate it, no.

I am not dismissing any particular evidence, unless it's contradictory with other data (normally better established). It is just you who decide to judge any criticism of the few and quite questionable papers that seem to back your catastrophist hypothesis as "dismissive"... but at the same time you have no problem dismissing all the rest.

And when you insist that I accept your judgement, with the implication that you couldn't possibly be wrong, that's just annoying.

I do not do that, I just ask you not to kill your own research by putting all your eggs in a single basket - that additionally doesn't look too good anyhow.

I sincerely think that you have a large risk of being self-defeating by insisting on a catastrophism that is anything but proven.

You have decided already which is the solution to the puzzle and you are just trying to mount up a barricade around it with a pretension of exploration. You are not "exploring" anymore, you are just mounting a barricade around your single basket of eggs. But it was you and nobody but you who chose to put all the eggs in that basket (against my most sincere advise).

At this stage, as you grow more and more defensive (narrow-minded), I can't do much more to help you because I know that whatever I do (researching for further materials or warning you of contradictions) that doesn't suit your preconceptions will be thrown to the trash bin (or, as has happened in some cases, grossly misinterpreted).

It's not the kind of job I want to do for free, certainly.

Have fun with your eggs and basket.

DocG said...

Maju: "So saying that mtDNA A, B and F do exist in India with due references is "dismissing evidence", saying that A, B and F, are not comparable to M (N would be) is "dismissing evidence"?"

No. But discounting the entire article as "not useful" on this basis is. Here's what they wrote:

"three typical east-Asian mtDNA haplogroups (A, B and F) are absent OR VIRTUALLY ABSENT from non-northeast India . . ."

The above is essentially the same as what you have said: that they are present but in very small amounts. Note also that they don't deny they can be found in northeast India, but that their presence in the rest of India is negligible. You can quibble if you like, but I see no significant difference between negligible and present, but in small amounts.

More significant is the following:

"Furthermore, the fourth typical east Asian mtDNA haplogroup M has a different structure in India as compared to other Asian areas. This suggests that, although they show close affinities, the east Asian and Indian mtDNA gene pools are fairly distinct. This result is consistent with the suggestion that the east Asian and Indian mtDNA pools have been separated from each other for about 30 000 years."

I'm not an expert and cannot explain exactly what the above means or why they came to the conclusion they did. But the paper was written by experienced geneticists, including Mark Stoneking, who is one of the pioneers in this field, so why should I dismiss their conclusion? Because it doesn't suit you? Because it's consistent with my hypothesis? (For you, this would seem to be the biggest crime.) Sorry, I find it difficult to believe you when you claim neutrality in such matters.

And yes, other investigators have concluded that M in India is consistent with direct descent from HMP -- and I've made no secret of that. What's important to me is that there ARE interpretations of the data consistent with a genetic gap, which means that there IS evidence that seems to support my hypothesis. And as long as such evidence exists then the hypothesis has not been falsified. That's all that interests me at this point, as I continue to develop the principal ideas. It will, of course, be up to others to evaluate the theory when it is complete -- and that's fine with me.

DocG said...

Maju: "I am not dismissing any particular evidence, unless it's contradictory with other data (normally better established)."

There are many contradictory results in the field of pop. genetics, as there are in any active scientific field. You don't discount evidence simply because there is other evidence that seems to contradict it. And who is to be the judge of what is "better established"? The experts debate such issues all the time, but for you there is no debate. In your mind there is only evidence you accept and evidence you reject. Sorry, but I see no reason to follow your lead in this. Which doesn't mean I don't respect your opinion and appreciate your analysis, which I do.

But when you attack me for not going along with your verdict, really Maju, that's too much.

"You have decided already which is the solution to the puzzle and you are just trying to mount up a barricade around it with a pretension of exploration."

This is absolutely not the case. As I've said before I am not in love with the hypothesis I'm exploring and would certainly be happy to explore other alternatives if they are proposed. I cannot remove the gap I see, however. That's not a hypothesis, it's an observation, based on years of careful research. The question is: does the genetic evidence also show a similar gap? If so, then it would be much easier to explain the cultural gap. If the genetic evidence is not consistent with the cultural gap, then I would be forced to consider some other explanation. And that would be fine.

The explanation you offered, on the basis of population size and/or interactivity doesn't really address the problem and so I rejected it, sorry. Interactivity or size in themselves can't explain the distributions we now see and the gap I see. If you can explain how they could account for the well known geographical differentiations, morphological and cultural, i.e., the usual "racial" distinctions, and also explain the gap in the musical evidence, then I would be happy to consider it.

If it turns out that there is no gap -- e.g., let's say that in-depth study of some Indian tribals reveals that they sing in P/B style after all, or have instrumental ensembles consistent with that style, then I would have no problem whatsoever admitting I was wrong and the gap I thought I saw was due simply to lack of evidence. That would please me, because that would make sense, even though it doesn't support my hypothesis, which I would happily discard.

It would nevertheless still leave many questions regarding the distribution of morphological traits, but I would be satisfied to at least not have to worry about the cultural gap.

The bottom line is that I would have no problem if I learned tomorrow that Indian tribals actually do have P/B related traditions and that the musical gap is just an illusion. There would be no collapse of any house of cards and no breaking of eggs. I'd be pleased because such evidence would be as consistent with my thinking on world music as anything I've posted on this blog.

But to date, I have seen no such evidence, and since my policy is to follow the evidence then I must continue to explore other alternatives.

German Dziebel said...

"The bottom line is that I would have no problem if I learned tomorrow that Indian tribals actually do have P/B related traditions and that the musical gap is just an illusion."

There's another alternative that you don't consider. A musical gap would be an illusion if monophony is not derived from polyphony but rather represents its own line of musical evolution dating back to at least pre-dispersal times. South India is then organically connected with East Asia, North America, West Asia, North and East Africa, as well as Australia. For this alternative, you don't need any new evidence. It's right there on your plate.