Saturday, February 13, 2010

311. Alternatives

Beginning way back in Post 225, I've been exploring a particular historical/ evolutionary scenario, starting with certain inferences about the nature -- and culture -- of our "most recent common ancestors," and proceeding with a series of increasingly speculative speculations centered on the adventures, misadventures, trials, tribulations and triumphs of their descendants, the Out of Africa migrants, as they and their progeny (allegedly) made their way through Asia along "the southern route," all the way to what is now New Guinea and Australia. It's important to remember that this particular attempt at historical reconstruction was made possible, first, by the musical evidence, which is in certain ways unique; and, second, by the use of the musical evidence to help establish a baseline, from which we were able to proceed step by step, in an orderly and logical manner. As I see it, two of the most glaring omissions in the anthropological and archaeological literature have been the neglect of both these areas: the neglect of musical style as an essential element of core culture, clearly on a par with language in importance, but far simpler and thus far more amenable to cross-cultural comparison; and the assumption that one could reconstruct important aspects of human history and/or evolution without first establishing a clear baseline from which to begin.

I have all along been referring to this overview as an "exploration," meaning that, as far as I'm concerned, it is tentative, incomplete and possibly incorrect -- but imo a useful exercise nevertheless. It also amounts, I would think, to a testable hypothesis, or set of hypotheses, with the ultimate tests most likely stemming from the field of population genetics, which is only now beginning to realize its enormous potential, and still has a long way to go. Nevertheless, I've been accused of being selective in my use of evidence and neglecting to consider alternative explanations, with the implication being that what I'm calling an exploration is in truth nothing more than a pet theory, or worse, a crackpot theory, which must be defended at all costs. I've denied that this is the case and have often asserted that I'd be happy to accept any alternative explanation that both accounted for the evidence and made sense. But at the same time I refused to drop my line of thought for side excursions to examine evidence that didn't seem to fit, and consider alternative explanations. Now that my overview is complete, it's time to do just that.

According to one of the most active commenters on this blog, Maju, the most up-to-date genetic evidence does not support the gap I see in South Asia, nor does it support the notion that there was ever a significant large-scale bottleneck or series of bottlenecks in South Asia, consistent with either the Toba eruption or any other major disaster, such as a Tsunami, drought, etc. that might have occurred during the Out of Africa expansion along the southern route. Maju has argued that the discontinuities between India and Southeast Asia that I've pointed to, clearly apparent in some of the phylogenetic trees and related maps, are either due to sloppy, simplistic research or represent methodological artifacts rather than logical inferences from the genetic evidence. As I see it, it's simply too soon to tell with any degree of confidence which of the many interpretations of the data are correct and to what extent future research with refined methods and expanded samples will either confirm or challenge the findings of any one group.

Assuming, however, that Maju's objections are in fact legitimate, then we must consider alternative scenarios that could explain the apparently inexplicable gap that exists in the cultural evidence, especially the musical gap, where we see no sign of the "African signature," either vocally or instrumentally, anywhere in South Asia and in fact anywhere from Yemen all the way to the eastern borders of India, and yet find it in abundance among a great many isolated indigenous groups to the east and southeast of India, including southeast Asia, island southeast Asia and Melanesia. Maju's explanation is based on a recent finding to the effect that the larger the population, the greater the degree of innovation, which for him means that a society with more opportunities for interaction among various members of its population is a society that is more likely to undergo change. And since there is in fact excellent evidence for a tremendous population expansion centered in South Asia shortly after the Out of Africa migration, that could explain the various cultural changes, including musical changes, that would have taken place there. I have grave doubts about this scenario, which strikes me as overly simplistic and not fully explanatory, but it's an example of an alternative hypothesis and certainly worth considering.

Another angle to consider is the series of extremely complex developments centered in South Asia from the earliest beginnings of the Neolithic to the rapid evolution of civilization(s), a dramatic series of events that transformed the entire region and would certainly have presented a challenge to the various indigenous peoples seeking to maintain age-old traditions in the face of enormous military, political and social pressure. While a great many tribal groups did in fact wind up as castes, under the thumb of more powerful groups who sought to control every aspect of their lives, a great many did apparently maintain their independence, largely by retiring to refuge areas, minding their own business and avoiding conflict wherever possible. There remains the question, however, of the degree to which they were able to remain fully independent, as was the case with so many "relic" peoples farther to the east, or whether there was a certain amount of encroachment in each and every case, over thousands of years, that could have led to a certain degree of cultural homogeneity, despite the isolation of so many of the Indian "tribals." One does get the impression, when one explores the various musics of so many Indian and Pakistani groups, of a certain uniformity of musical style and practice throughout the subcontinent, a somewhat disturbing phenomenon that is especially apparent in the "folk" music of the villages, which never seems to depart very far from the "classical" raga style cultivated by the upper castes. Whether the same sort of pressure to conform has been felt by the ostensibly more independent tribal groups is difficult to assess. But if that has been the case, then it's possible that the absence of any trace of P/B in India or Pakistan could be due to external pressures from more powerful groups, either in recent or ancient times, which could have wiped out just about all trace of more archaic cultural practices. Why such a thoroughgoing process of cultural assimilation would have occurred in South Asia and not Southeast or Island Southeast Asia isn't clear but again, such a possibility is worth considering.

(to be continued . . .)

22 comments:

Maju said...

FYI, Victor: I posted yesterday a new version of my exercise on Eurasian mtDNA timing (an earlier version of which you already know), this time only based on coding region mutations, which are in principle more reliable than HVS ones. I still fail to see any clear evidence or indication of Toba but whatever the case I comment because I believe you were interested in this variant and even asked for it.

DIRECT LINK.

Regards,
Maju.

DocG said...

Thanks for posting this very interesting information on your blog, Maju -- and thanks for the link. I'm wondering, however, why you don't see any possibility of connecting the huge proliferation of basal M sublineages with a bottleneck event. You yourself have insisted that each such branch probably represents a bottleneck. And since so many of these branchings seem to have occurred at roughly the same time, why isn't it at least possible that they were all produced by a single large-scale event, of the sort that could have had a widespread effect on many different groups in the same general region.

Maju said...

"I'm wondering, however, why you don't see any possibility of connecting the huge proliferation of basal M sublineages with a bottleneck event. You yourself have insisted that each such branch probably represents a bottleneck".

I have not "insisted" in such thing, just considered the possibility as part of my open intellectual (research, understanding) process, a good deal of which is reflected Leherensuge.

Whatever the casem it's all a matter of checking the options one by one. In the case of M, it seems not just the largest but also the oldest branch of all. Even if you wish to consider M and N expansions simultaneous or place N before M, it doesn't matter, because there's nothing happening in Eurasia before M and N in any case: the previous step was in Africa and not that long before.

We lack of any sign of "pre-bottleneck" expansion in Eurasia and, if they were there at all they must have left some signature of some sort. Not in mtDNA, it seems.

"And since so many of these branchings seem to have occurred at roughly the same time, why isn't it at least possible that they were all produced by a single large-scale event, of the sort that could have had a widespread effect on many different groups in the same general region".

That's why I initially tended to place the M event right after Toba. I could still do now: it probably matches reasonably well.

But this would mean that people was not only wiped out in South Asia but also in the East. And that the most dynamic survivor was still in South Asia, re-expanding right after the catastrophe.

An alternative explanation is that this expansion signature was produced upon arrival to South Asia by the OoA migrants. Both make equal sense and even could be argued that both are the same thing.

But 2/7 L3 sublineages seem enough survivors for the OoA alone, so there's no clear signature in mtDNA to argue for a bottleneck (other than the mere migrational founder effect). hence scrapping a brutal bottleneck as the one proposed by Oppenheimer makes total sense.

Just try to think all without cantometrics for a moment, because I feel that your assumptions on musical theory, affect your interpretations of the other data in ways that are excessive.

I think first in terms of genetics and then would see how cantometrics can fit. Some stuff fits nicely, as with the P/B nexus but other stuff does not so easily. That's why I think you're putting the cart (cantometric theorization) before the horse (genetic evidence).

You should dedicate some time to think the genetic evidence on its own merits, regardless of whatever other issues. Then see how can you fit cantometrics in it.

DocG said...

Maju: "We lack of any sign of "pre-bottleneck" expansion in Eurasia and, if they were there at all they must have left some signature of some sort. Not in mtDNA, it seems."

There are four mutations separating L3 from M, right? It seems unlikely that all four could have appeared in the same individual, no? So there must have been at some point some people with only one of these mutations, then at a later time people with two, etc. Or is there something I don't understand?

There has been no trace, thus far, of any of these pre-M haplotypes, but it's possible some could turn up as the DNA samples get larger, no? It's also possible that none of these lineages survived. It's possible that the small band of Out of Africa migrants could have had one, two or three of these mutations, but not all four, no? And it's also possible that the fourth could have appeared in a member of one of the colonies already living east of India, who would have been the founder of all the M lineages.

Assuming the Toba eruption occurred when there were colonies in both South Asia and east of India, then the pre-M groups in S. Asia could have been completely wiped out, while, after the effects of the eruption wore off, this woman's descendants expanded from SE Asia to the east, the south, and also the west, to repopulate S. Asia. Those living closest to India during the eruption would have suffered the most severely of all the surviving groups. And since they are the ones most likely to have re-populated India at a later date, it would be the many bottlenecks produced in these groups that could have led to the great proliferation of M derived clades we now see in India.

Do you see this as a possible scenario? Or have I missed something important?

DocG said...

Maju: "Just try to think all without cantometrics for a moment, because I feel that your assumptions on musical theory, affect your interpretations of the other data in ways that are excessive."

First of all, the patterns revealed by Cantometrics are only one part of the musical evidence. There is a great deal more that Cantometrics doesn't cover. And some of the musical evidence is so compelling that it would make a huge difference even for someone who had never heard of Cantometrics, so long as they had a sufficient grounding in world music and especially the music of indigenous peoples (which admittedly very few ethnomusicologists now have).

While the importance I place on the musical evidence may seem excessive from your viewpoint, I am fighting as hard as I can to make anthropologists (and ethnomusicologists also) aware of the importance of this evidence so, from my viewpoint, when the musical evidence tells us something of importance, this is something I certainly want to emphasize.

Maju: "You should dedicate some time to think the genetic evidence on its own merits, regardless of whatever other issues. Then see how can you fit cantometrics in it."

By the same token, I could complain that you are too focused only on the genetic evidence, and should dedicate more time to the cultural evidence. I am trying to take ALL the evidence into consideration, and I don't feel that I've neglected the genetic evidence -- just that I need to understand it better.

Actually the genetic evidence, as I see it (and admittedly you disagree), appears to correlate quite strongly with the musical evidence. Don't forget, the Toba hypothesis did not originate with me, but with Oppenheimer, in a book based largely on the genetic evidence. And while it's true that Oppenheimer is not a geneticist (though he does have credentials in this area), the genetic evidence he draws upon comes from the research of very respected people.

As far as I can tell, and correct me if I'm wrong, you are the ONLY one who has disputed the Toba hypothesis on the basis of genetics. No geneticist has rejected this possibility, apparently. There have been some serious controversies in the genetic literature, but not over Toba. Everyone seems to be waiting for decisive evidence to come from the archaeological research, not the genetic research. And if in fact Petraglia were to find a "modern" human skeleton buried under Toba ash tomorrow, I very much doubt that any pop. geneticists would dispute his finding on the basis of the genetic evidence. If such a find were made (and that is certainly a possibility) then the genetic evidence would be re-interpreted to accommodate it. It would have to be.

So at this point I think both of us need to keep an open mind on this matter and not jump to conclusions.

Maju said...

Doc: That there's a number of mutations between L3 and M/N does indeed seem to mean that these Eurasian clades did not expand right after the L3 expansion. But this just makes total sense if we consider that the population must have remained small between Africa and Southern Asia before they found conditions appropriate for expansion, surely at their arrival to Pakistan/India.

Whatever the case, what I want to emphasize is that there is no sign of expansion before the Great Eurasian Expansion, clearly marked by the huge M starlike structure (and to lesser extent by that of N). If there would have been a pre-bottleneck expansion, we'd see an array of M-derived (or pre-M-derived) clades, maybe just 3 or 5, expanding at the post-Toba scenery. We don't see that: instead we see a huge single expansion at the M node, sustained at the downstream nodes without any signal of disruption.

This applies to South and East Asia. There's no sign of lineages (other than N arguably) migrating from East Asia to South Asia either. There is no sign of South Asian groups suffering more in any way or the population comparatively thriving in the East in relation to South Asia.

If you look at it for a moment from the viewpoint of genetics (and only genetics) you'll see what I mean. If you insist in putting the cart (music or whatever other cultural elements) before the horses (genetics - archaeology too but it's less informative for this period), then you'll see nothing but what you want to see.

...

"... it would be the many bottlenecks produced in these groups that could have led to the great proliferation of M derived clades we now see in India".

These lineages don't show signs of bottlenecks either. A bottleneck (or similar) should act on a phylogenetic stem as clipping all branches, making the stems look clearly longer. This can of course also happen for any random reasons, like a lineage remaining stable in demographical terms. In general branches are sing of expansion, specially if these are numerous, and long stems are signs of stability or even demic contraction maybe (as in a bottleneck). There are few long stems in macro-haplogroup M.

"By the same token, I could complain that you are too focused only on the genetic evidence, and should dedicate more time to the cultural evidence".

They don't have the same weight as evidence. Cultural elements are never as conclusive as hard facts. This is the problem German has and I fear you are sliding down the same dangerous slope.

"Don't forget, the Toba hypothesis did not originate with me, but with Oppenheimer"...

And don't forget that the hypothesis has many detractors. You are choosing it, in spite of all, because it fits your pre-conceived hypothesis and only because of that reason.

I don't have any such preconceptions and hence I don't feel any need to adopt it for such reasons of mere convenience. The same should happen with any other neutral observer, who would consider first the genetic evidence, draw his/her conclusions and only then check the cultural data, if at all, and see how it fits.

But well, whatever...

Maju said...

"As far as I can tell, and correct me if I'm wrong, you are the ONLY one who has disputed the Toba hypothesis on the basis of genetics".

Metspalu 2004 (which you know but misinterpreted at whim):

"Since the initial peopling of South and West Asia by anatomically modern humans, when this region may well have provided the initial settlers who colonized much of the rest of Eurasia, the gene flow in and out of India of the maternally transmitted mtDNA has been surprisingly limited".

"The East Eurasian-specific mtDNA haplogroups are less common in India and more sharply geographically segregated than the haplogroups of western Eurasian ancestry".

Chandrasekar 2009:

"Recent mtDNA evidence on modern human out of Africa migration route suggests a single dispersal by a southern coastal route to India and further, to East Asia and Australia [17], [20], [22], [23], [66], [69]".

"... in the present study, the basal diversity (37 nodes) and founder ages (57,000–75,000 years) of macrohaplogroup M in India reveals initial settlement of African exodus in India".

Maji 2009:

"L3 radiated out of Africa in the form of macrohaplogroups M and N around ∼60, 000 ybp (Quintana-Murci et al. 1999;
Mishmar et al. 2003) and entered South Asia or Indian subcontinent, and later drifted to Southeast Asia and Australia".

"... the next highest mtDNA diversity [after Africa] (in the form of M and N distribution) is traced to South Asia, in particular to Indian subcontinent..."

"The wide mtDNA diversity observed in Indian subcontinent is yet another dimension of biological and cultural
diversity, that has been a unique feature of the Indian populations (Ratnagar 1995; Thapar 1995)".

Mcaulay 2005:

"This evidence suggests that this coastal trail was likely the only route taken during the Pleistocene settlement of Eurasia by the ancestors of modern humans, and that the primary dispersal process, at least from India to Australasia, was very rapid. A founder analysis of western, southern, and eastern Eurasian and Australasian complete mtDNAs suggested a shallow gradient of arrival times, from ~66,000 years ago in India to ~63,000 years ago in Australasia (table S4)".

Thangaraj 2005:

"Our data indicate that two ancient maternal lineages, M31 and M32 in the Onge and the Great Andamanese, have evolved in the Andaman Islands independently from other South and Southeast Asian populations. These lineages have likely been isolated since the initial penetration of the northern coastal areas of the Indian Ocean by anatomically modern humans, in their out-of-Africa migration ~50 to 70 thousand years ago".

These are just a few examples. None of them seems to bother "debunking" the Toba bottleneck hypothesis (why, if nobody seems to anymore believe in it anyhow?) but all think without it, with a timeline that is always "after Toba" (70-60 Kya).

DocG said...

First of all, Maju, before I say anything more, I want to make it clear how much I appreciate your contributions to this blog and the very interesting and useful information you've provided. If it seems at times as though I would prefer to ignore or dismiss your evidence or your ideas, that is not the case. I value them and I am always learning from you, despite our differences.

Maju: "Whatever the case, what I want to emphasize is that there is no sign of expansion before the Great Eurasian Expansion, clearly marked by the huge M starlike structure (and to lesser extent by that of N). If there would have been a pre-bottleneck expansion, we'd see an array of M-derived (or pre-M-derived) clades, maybe just 3 or 5, expanding at the post-Toba scenery. We don't see that:"

A few years ago, I discussed this question with Sarah Tishkoff, asking her specifically about Toba and she was noncommittal. She is, of course, like many geneticists, extremely conservative in her judgements and unwilling to jump to conclusions. What she said basically is that there is nothing in the genetic evidence to help us decide this issue, one way or the other.

When I asked her associate, Floyd Reed, if there was anything in the genetic evidence that might point to a major bottleneck that could have taken place shortly after the Out of Africa exodus, he said there was no way to tell the difference between the Out of Africa bottleneck itself and any bottleneck that might have occurred in its wake, because any evidence of the latter would be too easily confused with evidence of the former.

"These lineages don't show signs of bottlenecks either. A bottleneck (or similar) should act on a phylogenetic stem as clipping all branches, making the stems look clearly longer."

Reference?

"Cultural elements are never as conclusive as hard facts. This is the problem German has and I fear you are sliding down the same dangerous slope."

While the genetic evidence is based on "hard facts" as you say, it is also complex, subtle and not at all easy to interpret. I'm not questioning the hard facts, I'm questioning your interpretation of these facts. Please don't compare me with German, who questions the interpretations of literally every authority in the field of genetics, anthropology, archaeology and even linguistics. The only interpretations I've been questioning are yours.

"And don't forget that the hypothesis has many detractors. You are choosing it, in spite of all, because it fits your pre-conceived hypothesis and only because of that reason."

Aside from you, I know of only one detractor and that is John Hawkes, who is a multi-regionalist. If you can supply me with the names of any geneticists who support OoA and have gone on record as disputing the possibility of a bottleneck along the same lines that you do, I'd be very interested to read what that person has to say. It's true that there have been claims regarding the coalescence times of certain haplogroups that are inconsistent with the Toba theory, but you yourself have dismissed that line of research as unreliable.

DocG said...

Maju, the quotations you've provided don't contain a single reference to either Toba or the possibility of a major bottleneck event, so I'm not sure what your point is. You assume that the information provided in these quotations needs no further comment, but sorry, maybe I'm too dense, but I don't get it. I understand that this is the sort of evidence on which you are basing your interpretation, but as I see it, that's all it is: an interpretation by one person, NOT a fact.

There are many ways to interpret the same evidence, in case you haven't noticed. Which is why there are so many disputes in literally every scientific field. I'm not dismissing your interpretation -- it appears to make sense, as far as I can tell. But as I see it, it is just one interpretation of many that could be offered. For you it is obvious, but not to me.

It's also important to understand that there are many bottleneck scenarios that would be consistent with the hypothesis I've been exploring. For example, the event in question could have left survivors in India as well as east of India. Or it could have left no survivors in India, which would then have been re-populated from both west and east as Oppenheimer suggests. And his interpretation is more than just empty speculation. He presents genetic evidence that would be difficult to explain without invoking a major disruptive event of some sort. For example, he notes the relative lack of N and the relative pervasiveness of M in eastern India. Other sources also note the presence of a very clear cline for M, which diminishes considerably as we go from east to west in South Asia. These distributions may not necessarily be signs of a bottleneck, but they do seem to be signs of something. How would you explain this?

The event in question could have been the Toba eruption. But it could also have been some other equally disastrous event, such as a Tsunami, drought, flood, hurricane, famine, etc. Toba is generally disputed because of the timing, not because there is no evidence consistent with a bottleneck event. But a bottleneck or bottlenecks could have been produced by some event other than Toba, which would make the coalescence estimates irrelevant.

Finally, as you can see, I am willing to consider alternative interpretations. I doubt that you'll be happy with any of them, but they are the only ones I can think of. The most likely is based on the multiregional model, which I'll probably be discussing in today's installment of the blog. If a bottleneck must indeed be ruled out, then as I see it, some variant of the multiregional model might be our only reasonable choice for explaining the musical picture as well as some other puzzles, e.g., the origin of the Australoid "race."

German Dziebel said...

"They don't have the same weight as evidence. Cultural elements are never as conclusive as hard facts. This is the problem German has and I fear you are sliding down the same dangerous slope."

"Please don't compare me with German, who questions the interpretations of literally every authority in the field of genetics, anthropology, archaeology and even linguistics."

This is so funny, guys. Victor, I don't question every authority in genetics. I don't think I ever questioned a linguist or an anthropologists in our exchanges. You, on the contrary, dismissed all linguists ever writing on the topic of such typological feature as tones. You dismissed kinship studies as too technical. (If you don't want to study my book, there're plenty of other sources.) You latch on genetics for some immutable aspects of your thinking (out of Africa, bottleneck, Toba, etc.) but at the same time admit that you don't understand this science. You even dismiss your own musicological findings (iterative one beat is found in Australia and North America but they were brought there within the past 5-10,000 years but mysteriously disappeared from Asia), when they don't fit your bigger ideas. You seem to be ignorant of so many things, including the ideas of one of your most active commentators.

I only question the dominant interpretations of the Amerindian-Asian and African-non-African nodes in phylogenetic trees. And I do think that the "peopling of the America" angle is the survival from a pre-scientific worldview. This is a historical fact, BTW. I'm wary of "authorities," true, because I know how much narrow-mindedness comes with being an authority. You can think of your hedge fund manager as an example.

Luis, gone are the days when people accepted some kind of hypothesis as a revelation. People used to treat books like the Bible as naturally and clearly showing the workings of divinity. In this day and age, all hypotheses need to make sense. The current version of genetic phylogenies and the underlying order of mutations doesn't make sense when you look at other sciences. Unless you put some thinking into explaining how come 140 language families ended up forming in the past 12K years on the other side of the world from our beloved African cradle, you'll always look to me like a quasi-religious fundamentalist. When you write "A bottleneck (or similar) should act on a phylogenetic stem as clipping all branches, making the stems look clearly longer" you describe Pygmy and Bushmen lineages. Some of those African lineages that geneticists claim are 140K old are in fact post-bottleneck versions of other lineages. They didn't re-expand until some 5,000K ago not because they were vegetating in some kind of "private" state (or if they did you should prove it by, say, looking at linguistic diversification) but because up until 5,000 years ago they didn't exist.

Maju said...

What Tishkoff and colleague are telling you makes sense only if the "Toba bottleneck" does not imply a backmigration from Eastern Eurasia into India. That is: if it happened just after the arrival of people to South Asia and caused great mortality among them, extending de facto the period of migration and low population levels a little bit more.

But what you and Oppenheimer are claiming is different: you claim that people poured from East Asia into South Asia and that's not in accordance with the genetic evidence.

"Reference?"

Common sense, logic: "bottleneck" means narrowing of the population, loss of diversity. You can't have all: tons of rapidly expanding diversity as with the M starlike structure and a massive bottleneck that should have culled most of that diversity.

Ask Tishkoff if you don't believe me. What she and her colleague told you is precisely that: that a migrational founder effect and a true bottleneck have similar effects, clipping diversity.

"It's true that there have been claims regarding the coalescence times of certain haplogroups that are inconsistent with the Toba theory, but you yourself have dismissed that line of research as unreliable".

But all is relative. I question the "worship" of such estimates, specially in "populist" versions of the field but I don't question that there is some logic to the molecular clock hypothesis overall. If something looks very old, it's probably very old, even if the age estimates are wrong by some 20 or 40,000 years.

But anyhow, that's not my point nor the point all those geneticists are making either. The main issue on which everybody insists is diversity. And this is more true in the last five years or so, as the huge diversity of M, and in particular South Asian M (and R too to a lesser extent) has been researched, what was not something geneticists paid much attention to initially.

In 2005 and 2006 we, the genetic/prehistory aficionados, were still wondering what was hidden behind that half of Eurasian mtDNA that was called M. The first researches were starting to appear but it was yet a very obscure area, even for well informed geneticists probably a couple of years before that.

Much of the same happened with African L, only slowly being resolved into specific sublineages with some precision. At first it was everything much Eurocentric, really, and that was quite confusing. Gradually we have learned to appreciate and understand the vastness of non-European genetics, specially those of South Asia and Africa.

"Please don't compare me with German, who questions the interpretations of literally every authority in the field of genetics, anthropology, archaeology and even linguistics. The only interpretations I've been questioning are yours".

It pains me too to make such comparison, believe me. German is of course a very extreme case and sure, you don't go to such extremes, not at all. But the basic pattern is too similar: wishful thinking in order to find support for a pre-conceived hypothesis based only on analysis of cultural elements.

I seriously think you should consider alternative possibilities. I told you that long ago, when you began again writing on this blog after a "sabbatical year" and did so with some papers that were very much focused on Oppenheimer's hypothesis. I told you that you were putting all the eggs in one basket and that, IMO, was not a very reliable basket. That you should consider other possibilities.

(continued below)

Maju said...

(continued from above)

I think yourself realize that not all the loss of P/B can anyhow be attributed to such single "Toba bottleneck" event. You have loss of P/B in Africa, in Australia, etc.

I don't know how to explain that. It's not my specialty (though I have tried to suggest possible alternatives). But I have tried to warn you of such inconsistencies. Hopefully this will serve you in due time. I want to believe that, after you chew on the issue for some time (maybe months or years) you will come to realize the importance of these issues and will come with an improved theory, a much more solid one.

Meanwhile I don't have much more to say because it's all the same: you insist in following Oppenheimer's model and I remain extremely skeptic of it. We could discuss it once and again till the proverbial Hell freezes but I think that, after some riff raff the debate becomes exhausted and little more can be said.

I am not going to buy on the Toba catastrophe hypothesis just because you find it convenient. I definitively need good evidence (genetic evidence preferably) and I see it nowhere.

You also seem unable/unwilling to reconsider your construct under alternative scenarios. So there's not much more to say. I could review the whole Internet in search of more and more papers but you're going to dismiss them anyhow, as you have done with the previous ones (which are already a good bunch). As you say, geneticists tend to be cautious on the big picture, on the interpretation of the data beyond some point and with good reason, after all it's a young science and those who draw too pretentious conclusions are often reckless, like Oppenheimer. And they have an academic prestige to protect.

It happens also in other disciplines. For instance I think now of Colin Renfrew and his Anatolian origin hypothesis for Indoeuropean. For some time he was quite popular, the theory was catchy for the public, specially in Europe (Indoeuropeans could be perceived as civilizing instead of just barbaric conquerors)... but eventually common sense prevailed and the old good Kurgan model is the one that has the backing of the hard data, archaeology mainly in this case.

In the case of the OoA and the Eurasian expansion the most solid model at the moment is the rapid coastal migration. There are others but not so parsimonious. Myself I have got some doubts, specially considering some archaeological evidence like Petraglia's. I try to perfect my understanding of such important process in the genesis of Humankind (and share this private research at my blog) but no matter how I look at it, I see no sign of any major back-migration from East Asia to South Asia. Instead I see many signs of migrations in the opposite direction (two basically: M initially and R at a later stage).

So I can't back you within the Oppenheimer model because I can't see how it could be right. Of course, I'm always evaluating it from the perspective of population genetics only (or at least primarily) and I have nothing to gain or lose with either option, just to find out what is more realistic, closer to what really happened to our great-great...great-grandparents at that remote time.

Well, that's what I had to say. I appreciate your research but I have to say I believe you have deviated from the more likely scenario out of wishful thinking, out of convenience with a reasoning that you like.

By the way, the "Ik" people that you have mentioned several times as example backing your catastrophe scenario... are they any real? Who are they? Where can we find them in reality? It sounds too much like a fiction story.

Maju said...

"Maju, the quotations you've provided don't contain a single reference to either Toba or the possibility of a major bottleneck event, so I'm not sure what your point is".

They all are assuming the rapid coastal migration model and the quotes are quite clear about this. The first of the papers is actually the one that proposed the rapid coastal migration first of all, though the model was refined later, suppressing the West Eurasian branch from the initial migration.

Nobody who matters considers Toba anymore. Maybe they are wrong but they don't seem to think that Toba has any relevance when they estimate the migration to c. 70-60 Kya (after Toba) and when the diversity is clearly larger in South Asia than in the East.

I understand that because I've been the last many years reading and discussing and thinking in depth on these matters. I can see why it's too much for you but I can't help you to understand further because I don't know of any paper that explicitly tries to debunk Oppenheimer's hypothesis. They may propose other hypothesis or just adhere flexibly to some other hypothesis (normally the rapid coastal migration) but that's it.

"For you it is obvious, but not to me".

How long have you been chewing on this matter? One year, a few months? When you have dug on the issue for several years, then it'll be obvious to you too. I trust that you are clever enough and that the only barrier is imperfect understanding (and the bias of your musicological model).

If you don't ever see it. Well, know that I won't mind. I did what I could.

"It's also important to understand that there are many bottleneck scenarios that would be consistent with the hypothesis I've been exploring. For example, the event in question could have left survivors in India as well as east of India".

This could make better sense, IMO. But it's not anymore Oppenheimer's model. The key for me is not whether the Eurasian fraction of humankind underwent Toba or not, but that there is no signal of any meaningful flow from East Asia to South Asia, as Oppenheimer proposes.

I have been considering (very shallowly by the moment) that possibility in fact, specially after you mentioned that there are some indications of a relatively late colonization of the eastern coasts of India (or part of them).

The same I have been working a bit on mtDNA I have also briefly revisited Y-DNA and I have the impression that NW South Asia might have been particularly important in the events that unfolded after the OoA. However this line of thought is a bit on hold until the issue of a newly proposed lineage (MNOPS, K-derived) is clarified.

So this kind of Toba bottleneck alternative model might make some sense. Not sure anyhow because the mtDNA evidence rather doesn't suggest it and, as I say, the Y-DNA data is in wait of consolidation.

It would require a lot of detailed and methodical work to confirm or deny this "new" hypothesis and I don't know of anyone who has proposed it so far.

(cont.)

Maju said...

(continued)

"He presents genetic evidence that would be difficult to explain without invoking a major disruptive event of some sort. For example, he notes the relative lack of N and the relative pervasiveness of M in eastern India".

N looks to me like secondary minor arrival to South Asia and having an expansion center in SE Asia. Only a few N sublineages are found in South Asia and all are the same than those found in West Eurasia (R specially but also N1'5, ancestor of I, and N2, ancestor of W).

But the key is mtDNA R, which has clear highest diversity in South Asia again, indicating that it expanded from the subcontinent. So, as I see it, some N migrated from SE Asia to South Asia without really altering the already existing dominance of M there (so they were few people) but somehow they managed to get involved (notably mtDNA R) in the last episode of the Great Eurasian Expansion, with a branch heading eastwards (maybe with the proposed Y-DNA MNOPS) and another more diverse branch heading westwards (with Y-DNA IJK in general, plus minor lineages such as Y-DNA G and mtDNA N1, M1 and probably X, though this one is not found in South Asia at meaningful levels).

I'm still chewing on all this but it's the conclusions I have been reaching as of late.

I'll make a synthesis one of these weeks probably. Just don't push me.

"Other sources also note the presence of a very clear cline for M, which diminishes considerably as we go from east to west in South Asia".

This may be caused by the secondary R expansion, as nearly all that is not M in South Asia is R. However R is also important in the South and to some extent also along the Ganges route. It's the Orissa area which is the least diverse probably but we can't know if this was just caused by local factors such as a more hostile environment or just greater isolation from the main routes, which makes total sense from the archaeological perspective as well as with simulations of possible migrations through the Indian landscape.

As I say I'm considering that the main core of expansion in South Asia was in the NW (or the West in general). Again the lack of East Asian lineages (mostly) in Eastern India shows that anyhow there was no significative back-migration from the East, as Oppenheimer proposes (the West was empty... or rather full of Neanderthals).

"If a bottleneck must indeed be ruled out, then as I see it, some variant of the multiregional model might be our only reasonable choice for explaining the musical picture as well as some other puzzles, e.g., the origin of the Australoid "race"."

I see no puzzle in the Australoid race (Australian Aborigines and only them). They are just totally normal Eurasians of their own branch. But I don't see any clear "African affinity" for Negritos/Melanesians either. They are black, sure but they are also tropical and marginal (i.e. remained out of the formative processes of the main two races of Eurasia, evolving on their own). They have anyhow traits shared with other groups: blond hair is only found in West Eurasians, Australian Aborigines and island Melanesians, prominent noses are typical of West Eurasians and Papuans, curly hair is again typical of all Eurasians but East Asians and arguably Northern Europeans, dark skin color is common in all Tropical Eurasia (naturally), they are also very variegated for many other traits such as head length, prognathism, etc.

In brief, they are just normal Eurasians of their own kind, who, by remaining isolate have retained diverse archaisms that have bee more diluted elsewhere, where interaction was greater.

German Dziebel said...

"It pains me too to make such comparison, believe me. German is of course a very extreme case and sure, you don't go to such extremes, not at all. But the basic pattern is too similar: wishful thinking in order to find support for a pre-conceived hypothesis based only on analysis of cultural elements."

It's easy to see that out-of-Africa has been percolating in European minds since at least turn of the 20th century. That is long before mtDNA got sequenced. Victor has been believing in it at least since 1965. It's even easier to see that the late peopling of the Americas (first, from Israel, then from Siberia) had been firmly "established" in European minds long before scientific method developed. Europeans wanted to justify colonialism by presenting American Indians as a recent offshoot of a marginal Asian population. These are truly preconceived cultural molds and examples of wishful thinking into which geneticists simply plugged their data. My thinking, on the contrary, is the example of an approach driven only by data and only by interdisciplinary data spanning genetics, kinship studies, linguistics, archaeology and paleobiology, not by pre-conceived ideas.

DocG said...

Maju: "But what you and Oppenheimer are claiming is different: you claim that people poured from East Asia into South Asia and that's not in accordance with the genetic evidence."

Maju, I have great respect for your knowledge of population genetics and your interpretations of the genetic data. The real problem I have with you, and also with German (whom I also respect, by the way) is that you are both continually making unwarranted assumptions about my pov and what it is I'm trying to accomplish. There's no point in going over all that again, since it's been said and resaid. But please understand that I am searching for answers that make sense in terms of ALL the evidence -- I am not asserting anything, I am exploring certain possibilities -- and I am certainly open to scenarios other than the one proposed by Oppenheimer.

"The main issue on which everybody insists is diversity. And this is more true in the last five years or so, as the huge diversity of M, and in particular South Asian M (and R too to a lesser extent) has been researched, what was not something geneticists paid much attention to initially."

Help me out here. Obviously a bottleneck reduces diversity, so a sudden reduction in diversity would be the sign of a bottleneck. And the "huge diversity" of South Asian M would appear to be telling us there could not have been a bottleneck. But there is something wrong here, because the Out of Africa exodus would itself have produced a major bottleneck. So obviously the diversity we see for M must have developed after the Out of Africa exodus. And the question I have for you is: when? How long after? What tools do we have for determining whether this proliferation of M clades began one hundred, one thousand, five thousand or ten thousand years after South Asia was initially colonized?

DocG said...

Maju: "I think yourself realize that not all the loss of P/B can anyhow be attributed to such single "Toba bottleneck" event. You have loss of P/B in Africa, in Australia, etc."

Any tradition can be lost for a variety of reasons, of course, not just due to a Toba-like event or even a bottleneck. I've already written a fair amount on the loss of P/B in Africa, and this certainly has nothing to do with Toba -- but it could be due to a bottleneck event of some sort in Africa. For details, see Post 47: http://music000001.blogspot.com/2007/07/47-call-and-response-and-highland-and.html

For my thoughts on Australia, see the last set of "Aftermath" blog posts.

"But I have tried to warn you of such inconsistencies."

There are no inconsistencies. Certain important cultural changes might be attributable to Toba, or some similar event. Others are clearly attributable to other causes. There are many such examples on this blog. I'm not asserting some universal "Toba bottleneck theory" if that's what's bothering you. However, I do think that the same events that produce genetic bottlenecks may well produce cultural changes of the sort that are usually attributed to a process of gradual "evolution." I see no inconsistency in that.

As for the rest, I think the most solid aspect of the hypothesis I'm exploring is the earlier part, centered in Africa. My approach to the Out of Africa migration is admittedly much more problematic -- more speculative and based on evidence that that remains questionable. You have questioned it and it may surprise you to learn that I too have questions about its validity. Nevertheless, I find it worth exploring. And as you can see, I HAVE considered alternatives, as should be evident in the last few posts.

"By the way, the "Ik" people that you have mentioned several times as example backing your catastrophe scenario... are they any real?"

Yes. But the word "Ik" is not. They were the object of a study by Colin Turnbull that he reported in his book "The Mountain People." His assessment of this group was so negative that a decision was made to keep their identity a secret. The book led to much controversy and Turnbull was accused of distorting his findings by focusing on only one segment of this society.

Maju said...

"Help me out here".

The only thing we can use with mtDNA is the mutation count. We know that from L3 to M there are 3 coding region mutations and one HVS mutation, totaling four. That's the "time" pre-M (L3) spent before exploding in more than 40 surviving sublineages.

Many M sublineages instead show signs of expansion (both in India and further east) one or few mutations downstream of M, so a sizable fraction of the M explosion kept expanding in the following millennia.

Similarly most L3 sublineages in Africa show signs of expansion right after the L3 node. But not M nor N, which go through some "bottleneck" (often interpreted as just the journey through Arabia, when the population did not expand but could also include something else).

You can count the mutations to mtDNA Eve and to present day lineages in order to make a comparison and estimate the possible duration of that 3-5 mutations at the stems of M and N (we mentioned that before and the OoA stem is rather small in comparison: 15-20% or so. But mutations do not happen at exactly regular intervals and different lineages have in fact different numbers of accumulated mutations, so there is huge uncertainty. Also population size should affect the chances of new mutations surviving (more the larger) and being fixated (more the smaller). It's quite complex and no method is too reliable.

If you wish to make your own research on this matter (I hate differential equations), I can suggest you to begin by reading the Wikipedia article and also the article I posted recently at Leherensuge mentioning several major papers on the molecular clock applied to human mtDNA.

There's no exact method, no radiocarbon of genetics, but the comparison with what is before and after should give you at least an impression of how long.

What is most important for me is that, if there was a true bottleneck after some initial migration (not much in any case), it also affected the East necesarily and not just South Asia. And also that the main expansions after it seem centered in South Asia.

German Dziebel said...

"The real problem I have with you, and also with German (whom I also respect, by the way) is that you are both continually making unwarranted assumptions about my pov and what it is I'm trying to accomplish."

Victor, you've let me know, on a number of occasions, that you and I are talking at cross-purposes. This makes me think that you have a purpose, and this purpose does beyond exploration and into trying fit cultural evidence into the out of Africa model. You've softened up a bit once you've tried to walk the out of Africa talk. I hope you'll become open minded enough to give a fair consideration to all alternatives.

Maju said...

Victor: maybe you also want to check this, with new (and also old, reposted) age estimates for human mtDNA lineages after sequencing four full Khoisan genomes. However no dates are provided for after the OoA.

DocG said...

Thanks for your very useful analysis, Maju.

Maju: "What is most important for me is that, if there was a true bottleneck after some initial migration (not much in any case), it also affected the East necesarily and not just South Asia. And also that the main expansions after it seem centered in South Asia."

I have no problem with the above assessment. As I originally saw it, a Toba-type disaster could have affected just about every colony in both South Asia AND east of India, with the exception only of the leading edge of the migration, consisting of groups that had already made their way as far as the eastern edge of the Malay Peninsula and possibly beyond. This could have been followed by a re-expansion of disaster survivors, centered in India but also expanding into all of East and SE Asia. The expanding groups would have been affected by the bottleneck and thus, theoretically at least, more aggressive, and could have, over time, marginalized those groups that had been for the most part spared the effects of the disaster, and thus retained their original African and pacifist cultures, along with their original musical culture (P/B). This would seem to explain the situation we now see.

But there are, as you have pointed out, inconsistencies in the above scenario. For one thing, the greatest expansion of M seems to have been limited to India and there seems to be little evidence of expansion from India to the east or vice-versa either. And there is also the question of whether or not there is any real evidence of a bottleneck at all.

I've considered several alternative explanations in recent blog posts, but none of them seems likely, as far as I can see, though some are more likely than others. Anyone following my scenario is free to draw his or her own conclusions. There is still a lot more research to be done, that's for sure.

Maju said...

".... the greatest expansion of M seems to have been limited to India and there seems to be little evidence of expansion from India to the east or vice-versa either".

This is not really correct. The expansion of M as such must have began in just one place (the haplogroup itself must have begun with just one woman) and this place is logically South Asia, that holds the greatest diversity by far. But a lot of M sublineages also made it to East Asia and Oceania and in fact make up about 50% of the mtDNA over there, the rest belonging to mtDNA N and specially R, which must have also spread from South Asia (for similar diversity reasons).

So "South Asians" colonized East Asia and large parts of Oceania. The only region anywhere dominated by N(xR), the macro-lineage that could have got a SE Asian urheimat, is Australia.

My whole point is that with the mtDNA evidence we can only conclude that "South Asians" colonized all Eurasia, not just once (M) but twice (R as well).