Here, for example, is what Richard Cordaux et al. have to say, in their paper, Mitochondrial DNA analysis reveals diverse histories of tribal populations from India, 2003. And by the way, Mark Stoneking, one of the leading figures of population genetics, is one of the co-authors:
In summary, although the data support a recent India– Australia connection, we could not find in Indian tribals any unquestionable genetic signature of the ~60,000 year old migration from Africa to Sahul following the postulated southern route. A possible explanation would be that such migration never occurred along that route. Alternatively, the early migrants from Africa may have made their way to Sahul following the southern route without settling in India. Another possibility, which is probably the most reasonable one, is that in India the genetic traces of early migrations along the southern route were erased by the subsequent migrations which shaped the present-day mtDNA gene pool of India (p. 262 -- my emphasis).Whether the subsequent migrations came in the direct aftermath of the bottleneck I think I'm seeing, or at a much later time, is not clear. What is clear is that, at least as far as these authors are concerned, there is a genetic gap, at exactly the same place we find all the other gaps.
[Added at 3:52 PM: Here is what Kivisild et al had to say on this matter, in 1999:
Both western and eastern Eurasian-specific mtDNA haplogroups can be found in India together with strictly Indian-specific ones. However, in India the structure of the haplogroups shared either with western or eastern Eurasian populations is profoundly different. This indicates a local independent development over a very long time period ("The Place of the Indian mtDNA Variants in the Global Network of Maternal Lineages and the Peopling of the Old World," p. 14 -- emphasis mine).]
Nevertheless, Maju could still be right in questioning my interpretation of the gap, because it's possible that it's the result of more recent historical developments having little or nothing to do with the original Out of Africa migration. And it may or may not be supported by the genetic evidence, depending on whose research you prefer to follow. I want to consider all possibilities and to make clear that I'm not in love with the interpretation I've been presenting -- for me it seems like the simplest and most reasonable interpretation, but I'll be happy to accept any explanation that makes sense, even if the events producing the gap can be traced to some other event(s), such as, possibly, the LGM (latest glacial maximum), or migrations dating to the Neolithic, or even relatively modern events due to colonialism. If Maju or anyone else would like to put together a coherent alternative explanation for all these gaps, I invite them to do so as a comment, and I'll bring it over here into the posts. What I will not accept is a dismissal or denial that any such gap exists. At the risk of repeating myself: the gap is there -- it is real -- it must be accounted for.
Posts
56 comments:
"If Maju or anyone else would like to put together a coherent alternative explanation for all these gaps, I invite them to do so as a comment, and I'll bring it over here into the posts. What I will not accept is a dismissal or denial that any such gap exists. At the risk of repeating myself: the gap is there -- it is real -- it must be accounted for."
I've already outlined two alternative treatments of the "gap" situation.
1) The movement of haplogroups went from east to west (from East Asia to North Africa), with M and N lineages differentiating along the way as populations filled in various large geographic pockets/ecological niches (India), Middle East, North/East Africa, Western Europe (and smaller refugia such as the Caucasus and Andaman islands. This happened around 45-40K. This would leave us with several easternmost haplogroups (C, D, A, X, S, B, etc.) and several westernmost haplogroups (African L0, L1, L2, L4, L5, L6), the origin of which would still need to be explained. The route this major migration took was not southern coastal and not circumpolar but somewhere in the middle, along the Eurasian steppe south of the Urals, North India, central Asia, Iran...
There're no gaps under this scenario. What I need from you, Victor, is to identify the corresponding musical signature.
2. There are several gaps along the "southern" route, not just one: East/North Africa, India and Australia. These gaps are explained as resulting from a speciation event in East Africa with a subsequent bottleneck into Central and South Africa (monophony lost, L0, L1, L2, L4, L5, L6 lineages emerge in Africa) and no bottleneck out of Africa (monophony and polyphony, M and N haplogroups, carried along a southern and along a northerly route all the way to Australia and America). India was colonized directly from East Africa-Middle East but the population(s) that ended up in Sahul cut India on a tangent in the north and then descended down south into Southeast Asia.
I like the first scenario more.
German: we don't believe your inverse phylogeny plus super-bottlenecks impossible hypothesis. It just makes no sense. Give up, please.
...
Victor: I reckon the musical gap but I don't see any correlated gap in any other field (genetics or linguistics, as you have mentioned).
In fact, what we have are a few scattered survivals of the P/B tradition, not a widespread survival in one region (Eastern Eurasia for example) and the opposite in another region (South/West Eurasia). This South/SE Asia gap does not exist, hence Oppenheimer's theory can hardly help your analysis or hypothesis.
Alternative explanations? As I have said before, for me it's a complex puzzle and often I rather prefer to just remain perplex before it. But let's try, very tentatively.
What about the hypothetical "bottleneck" affecting all early Eurasians (like in the migration phase or in the Toba-caused global climate disruption)? This would explain why most Eurasians would have lost P/B everywhere.
However how to explain the partial survivals? Independent development is a possibility, of course but maybe unsatisfactory. Another possibility would be that these minor groups would have managed to keep these traditions, even if "mutilated" after the catastrophe or just journey (founder effect: pseudo-bottleneck).
From what I see at your map, they might be related (??) to mtDNA N. This could be a hint, specially if N coalesced in SE Asia, as I understand it did. If so, the post-Toba (??) expansion from South Asia (mtDNA M, Y-DNA F) would have been the "eraser", with only scattered groups keeping some elements of the P/B tradition.
I understand that, at some point (but in the early colonization process, i.e. Middle Plaeolithic), N back-migrated to South Asia (maybe along Y-DNA MNOPS->P), producing R mainly (though also N1'5->N1->I and N2->W). Some of these mtDNA R peoples might have kept some of these traditions, brought from SE Asia, where Toba would have had less impact.
This could be an alternative explanation. As I say, it's just very tentative but fits better with the factual data.
German, I find your scenarios extremely confusing and also beside the point, because neither one accounts for the gap I've been describing. What seems clear to me is that the gap would be as much of a problem for the sort of west to east migration you want to see as the generally accepted east to west migration.
Maju: "I reckon the musical gap but I don't see any correlated gap in any other field (genetics or linguistics, as you have mentioned)."
I'm really puzzled, Maju, by your reluctance to see the point I'm making, which seems quite clear to me. If you look at the map of tone lanuage distribution in the Old World, you'll see it's almost identical to my minimap 2, which depicts the distribution of P/B in the wake of a hypothetical bottleneck event.
The musical distribution you see there reflects the 20th century distribution as documented by thousands of recordings and countless field studies and other ethnographic sources, regarding both vocal and instrumental music.
The distributions are almost identical so how can you fail to see the correlation and the gap between Africa and SE Asia that's so apparent in both maps? I've also pointed to carving traditions and negrito populations, which reveal a very similar gap in essentially the same region, West to South Asia.
Ive also presented additional evidence that at least some geneticists see a genetic gap in precisely the same region, though you prefer to dismiss or ignore that evidence.
"In fact, what we have are a few scattered survivals of the P/B tradition, not a widespread survival in one region"
The survivals of P/B in SE Asia and beyond, to the east and south, may be scattered, but they are not few and when we include instrumental styles featuring hocket and interlock, such as gong ensembles, pipe and panpipe ensembles, hocketed flute ensembles, trumpet and horn ensembles, stamping tubes, slit drums, and even the Indonesian gamelan tradition, which as I have argued, has to be a development from at least some of the above, then as I've said, we have literally a saturation of the African signature in this area.
Maju: "What about the hypothetical "bottleneck" affecting all early Eurasians (like in the migration phase or in the Toba-caused global climate disruption)? This would explain why most Eurasians would have lost P/B everywhere."
I'm not sure what you mean by "the migration phase." But a bottleneck of the sort that would have been produced by the Toba event, would in fact explain a great deal, as I argue in my Echoes essay. I plan to discuss such a possibility in future posts. It would explain the gap in South Asia, which would have borne the brunt of the effects of such a disaster. AND it would explain why the region immediately to the north, east and south would not have been affected anywhere nearly as much. But I was asking for an alternative to this theory, which is the one that makes the most sense to me.
"From what I see at your map, they might be related (??) to mtDNA N. This could be a hint, specially if N coalesced in SE Asia, as I understand it did."
A correlation with N would be a very interesing possibility to look into. But don't forget Galton's problem. A correlation doesn't necessarily imply a cause and effect relation. We would still have to explain why N is distributed where it is.
"If so, the post-Toba (??) expansion from South Asia (mtDNA M, Y-DNA F) would have been the "eraser", with only scattered groups keeping some elements of the P/B tradition."
The above is interesting and makes sense. I plan to discuss various possibilities of this sort in future posts -- but this is not an alternative to what I've proposed but more of an extension of it.
I was wondering whether you had an alternative explanation that wouldn't require the kind of bottleneck or other source of a founder effect that I've proposed and that you are so reluctant to accept. If you see simple migrations as the source of these distributions, then how do you explain the striking contrasts from one region to another?
"This could be an alternative explanation. As I say, it's just very tentative but fits better with the factual data."
I don't see it as an alternative so much as an extension of what I've proposed.
"German, I find your scenarios extremely confusing and also beside the point, because neither one accounts for the gap I've been describing. What seems clear to me is that the gap would be as much of a problem for the sort of west to east migration you want to see as the generally accepted east to west migration."
The paper you cited and the map you posted showed declining dates for M lineages from East Asia through South India to North Africa. What's so confusing about it? And it solves your gap situation very easily. And there's no need for a postulated pairwise connection between India and Australia, as the two regions were peopled through different waves coming out of East Asia. Monophony came to Australia and to India in two independent waves from East Asia. No gaps here either.
"German: we don't believe your inverse phylogeny plus super-bottlenecks impossible hypothesis. It just makes no sense. Give up, please."
The wealth of linguistic data and kinship data point to an east to west migration. Archaeology is consistent with it. Current genetic phylogenies are flawed and in need of revisions. There have always been phylogeographic alternatives in which geographically widely spread lineages are basal and geographically restricted lineages are terminal. African lineages are terminal lineages, that's why they aren't found outside of Africa. Please stop relying on belief when it comes to data analysis.
Victor:
1. I don not have a problem with the musical gap but with the supposed parallel with tonal languages. SE Asia (Indochina + Sundaland + Wallacea, and often also South China) is not the same as mainland SE Asia (Indochina and maybe South China). Tonality is a feature of mainland SE Asia and middle East Asia (China), while P/B survivals are more of the southern area (Sundaland and Wallacea). There is no strong overlap but is quite weak instead. And if we accept the theories of population replacement in Indochina since Neolithic then the overlap is virtually nil.
2. Not sure if my little speculation of association with mtDNA N is the same of what you have been proposing somehow with other words and other emphasis. I understand that Oppenheimer proposes a total desertification of South Asia at the Toba catastrophe and then recolonization from SE Asia. That is very different of what I think, which is: (1) colonization (M) of SE Asia from South Asia, (2) back-migration of some N from SE Asia to South Asia and (3) re-expansion (R) to both West and East. In my understanding South Asia is still a more important center than SE Asia, while for Oppenheimer it is the other way around, disregarding the greater diversity for some crucial macro-haplogroups (mtDNA M and R, Y-DNA F) in South Asia.
I was wondering whether you had an alternative explanation that wouldn't require the kind of bottleneck or other source of a founder effect that I've proposed and that you are so reluctant to accept. If you see simple migrations as the source of these distributions, then how do you explain the striking contrasts from one region to another? -
I really don't have a model for the transition from P/B to non-P/B music. My only explanation is that somehow it happened, how?, I can't tell for sure nor is really that important at the time of analyzing possible correlations with genetic or other anthropological patterns. Of course it'd be nice to know but I understand that in such a huge migration or rather series of migrations, a lot of opportunities must have existed for culture to change quite radically, as peoples settled new lands almost generation after generation. It's possible that they just lost contact with their elders as they moved on or it's possible that there was a catastrophe like Toba that did not cause a bottleneck as such but caused extreme emotional and social stress.
I can really imagine the poor people of Jawalpuram or anywhere else in South Asia watching ashes fall for months on their beloved homeland and suffering of a sudden climatic worsening that lasted decades. Probably many survived but the stress must have been brutal. Maybe they thought their music and dance were cursed or maybe they just did not feel like singing and dancing much after that. I don't see why it should be any "bottleneck" (otherwise not very apparent in the genetic record anyhow).
But what doesn't kill you makes you stronger... and so the South Asian population was in disposition to expand maybe earlier and more emphatically than their Eastern relatives (less affected).
I don't really know. I know I can't know for sure. But this other version sounds much better to me.
The wealth of linguistic data and kinship data point to an east to west migration.
False. Only your interpretation of it does.
Archaeology is consistent with it.
Absolutely false.
Current genetic phylogenies are flawed and in need of revisions.
I am very certain that this is also false. just a convenient pseudo-reasoning that meets your desires. Mere wishful thinking at best, self-deception at worst.
There have always been phylogeographic alternatives in which geographically widely spread lineages are basal and geographically restricted lineages are terminal.
In your dreams. Phylogeny is structured hierarchically with nearly no doubts as a matriushka doll of sets and subsets. Every single sublineage shares the defining SNPs of its ancestor necessarily. You know that so you are lying to yourself and the rest.
Please stop relying on belief when it comes to data analysis.
Apply that to yourself, please!
Maju: "Tonality is a feature of mainland SE Asia and middle East Asia (China), while P/B survivals are more of the southern area (Sundaland and Wallacea)."
First, as I stressed in an earlier response, I am not trying to make a case for a correlation between P/B and tone language. It would be interesting to see if that were so, but it would be an almost impossible task, due to the many language changes that have taken place in this region over the last several thousand years, as of course you know, changes that may have obliterated tone languages that once existed.
Second, you are wrong about the distribution of P/B. It can be found not only in island SE Asia but also on the mainland, among Vietnamese indigenous peoples (the Ede) and also certain South Chinese minority groups. And historically there is strong evidence of P/B in ancient China, as I mentioned earlier. Additionally we see many examples of hocketing by two or more instruments sometimes called Khene, or Sho, reed-based "mouth organs," clearly derived from panpipes, which are widely found in Thailand, Laos and Cambodia.
What's important in all this is NOT any possible correlations with tone languages, but the contrast between this picture and what we see in India, Pakistan and the Near East, where there are almost no traces at all of P/B in any form. And where there are also no instances of tone language (except in the upper Indus valley as I mentioned). It is this gap that I find to be a potentially very important historical clue and NOT any possible direct association between P/B and tone language.
Maju: "I understand that Oppenheimer proposes a total desertification of South Asia at the Toba catastrophe and then recolonization from SE Asia."
Yes, but as we now know from Petragia, there were at least some survivors, either homo erectus or homo sapiens. According to Petragia, homo sapiens. I don't see a serious contradiction here with what Oppenheimer proposed because the Toba blast would certainly have devastated most of South Asia, no question, even if there were some survivors.
"That is very different of what I think"
The main difference between you and Oppenheimer is in the interpretation of the genetic evidence -- he sees evidence of a huge bottleneck and you don't. And this puzzles me. What is the source of the disagreement? What does he see that you do not?
"(1) colonization (M) of SE Asia from South Asia, (2) back-migration of some N from SE Asia to South Asia and (3) re-expansion (R) to both West and East. In my understanding South Asia is still a more important center than SE Asia, while for Oppenheimer it is the other way around, disregarding the greater diversity for some crucial macro-haplogroups (mtDNA M and R, Y-DNA F) in South Asia."
As I recall, Oppenheimer also sees a huge expansion from South Asia, but for him this is what happens in the wake of Toba, and is a result of the bottleneck. As I see it (and I am planning to discuss all this in more detail in futur posts), the survivors of the disaster could have been more agressive and also more ruthless than their predecessors, and less content than they to remain within one particular area. They could have been less able to co-exist with their neighbors and more inclined to expand into new territories. In other words a new type of human with a very different type of culture than HBC or HMC could have emerged.
These people could have spread out into all areas from India, including SE Asia and ultimately East Asia (China), which is why all of East and SE Asia is now so complex. The original SE Asian groups (largely unaffected by Toba) would have taken to the hills, where many still live now.
Very speculative, of course. But it would explain a lot.
Maju: "I really don't have a model for the transition from P/B to non-P/B music."
But that's not the only transition that needs to be accounted for. We also need to explain the transition from tonal to non-tonal language; the transition from elaborate wood carving and mask making traditions to something much more modest; the disappearance of pygmy-type populations in Asia; and then the reappearance of all of the above (including P/B) when the wave of migration gets beyond the borders of India.
Doesn't it make more sense to conclude there were no such transitions, but that the colonies left in South Asia must have been either destroyed or decimated, while those further East were left relatively untouched and survived with their original culture more or less intact?
Yes, but as we now know from Petragia, there were at least some survivors, either homo erectus or homo sapiens. According to Petragia, homo sapiens. I don't see a serious contradiction here with what Oppenheimer proposed because the Toba blast would certainly have devastated most of South Asia, no question, even if there were some survivors.
Instead what I see is that there was no bottleneck, because in maybe the only site of that age in all Asia, we see continuity and survival.
Also I do not see in the genetic data a colonization from SE Asia into South Asia at that early moment but rather the opposite. I have several times emphasized that all of the most widespread lineages (mtDNA M and R, Y-DNA F) seem rooted in South Asia. I do think that there was some backflow from SE Asia into South Asia (possibly at a second moment) but this did not contribute significatively to the population of the subcontinent (very low levels of mtDNA N and Y-DNA C, none of D) and had more impact in the western frontier, where mtDNA N (specially R) and Y-DNA K had a greater impact.
In fact, this discussion has helped me to understand better what happened at the early phases of the Eurasian expansion and I plan to develop it and post on the matter at Leherensuge in due time.
It is not what Oppenheimer says.
The main difference between you and Oppenheimer is in the interpretation of the genetic evidence -- he sees evidence of a huge bottleneck and you don't. And this puzzles me. What is the source of the disagreement? What does he see that you do not?.
Good question but I think that it'd be best to ask Oppenheimer himself. Why does he see such a bottleneck?
Whatever the case, I don't think his opinion is widely accepted. I have never before come through anyone who defended such thesis so emphatically. Actually the mainstream model is nowadays that the OoA and colonization of Asia happened some time after Toba, between 70 and 50 Kya.
Not that I am with the mainstream "rapid coastal migration" model either. I do consider the possibility that the migration was earlier, possibly simultaneous to (or even the same one as) the c. 100 Kya migration to North Africa and Palestine. Some archaeological indications would support that (but they are not conclusive nor widely considered). But I do consider the RCM as a valid possibility from the viewpoint of genetics. The RCM totally ignores Toba or any other catastrophic scenario: it does not need it at all. The only pseudo-bottleneck is a founder effect at the journey through South Arabia, when population figures must have been low.
This model, for which I'm quite sure I provided a link or two in the past, is naturally also upheld by very respectable geneticists, probably most geneticists in fact. We can say it is the current paradigm in the field of human historical genetics, even if it's not written on stone either.
Eurasian lineages include 2/7 of an African lineage (L3) and 3/4 of another presumably African lineage (Y(xA,B), CF'DE). This aboundance of top-tier sublineages is not a signature of a bottleneck at all, just the normal result of a mild founder effect.
However some modelings on "wholesome" (autosomal) genetic data, using HapMap generic samples, has suggested that East and West Eurasians might have experienced some sort of bottleneck in comparison with the YRI sample of Nigeria. Nobody has ever researched this possibility for South Asians nor SE Asians nor any other population, so the implications are far from clear. It is also unclear if the signature would be the same for the funneling produce of founder effects, characteristic of ancient migration events.
(cont.)
(cont.)
As I recall, Oppenheimer also sees a huge expansion from South Asia, but for him this is what happens in the wake of Toba, and is a result of the bottleneck. As I see it (and I am planning to discuss all this in more detail in futur posts), the survivors of the disaster could have been more agressive and also more ruthless than their predecessors, and less content than they to remain within one particular area.
This approaches better what I see in the genetic registry. But I would rather skip all the alleged drama (bottleneck, ruthlessness) unless there is more clear evidence.
Macro-haplogroup M, the star of this expansion, shows a extremely large star-like structure, with dozens of sublineages spawning from that node to present day. This is not the signature of a bottleneck but a golden age of rapid expansion. In the rapid coastal migration model, it would happen soon after arrival to the larger and much more fertile South Asia, after the limitations of the South Arabian hypothetical phase.
However Oppenheimer may be guided by the curious and ill-explained fact that some lineages appear only or at least centered in SE Asia. These "SE Asian" or "Oriental" lineages are Y_DNA D and C and mtDNA N (but not its derivative R). I agree that this genetic puzzle, where SE Asia appears as important as (or maybe even more than) South Asia in the scene of early Eurasian expansion, asks for a good explanation and Oppenheimer has developed HIS own explanation. Fair enough but it's not the ultimate factual truth, just a theory, one among many possibilities.
But that's not the only transition that needs to be accounted for. We also need to explain the transition from tonal to non-tonal language; the transition from elaborate wood carving and mask making traditions to something much more modest; the disappearance of pygmy-type populations in Asia; and then the reappearance of all of the above (including P/B) when the wave of migration gets beyond the borders of India.
Doesn't it make more sense to conclude there were no such transitions, but that the colonies left in South Asia must have been either destroyed or decimated, while those further East were left relatively untouched and survived with their original culture more or less intact? -
I think that you are in a nice exploratory path with all this but I'm not really convinced that it is the best possible explanation. In particular I dispute the bottleneck hypothesis because I don't see clear uncontroversial genetic data supporting it (plus the already mentioned archaeological data).
If there would have been such a bottleneck, i.e. the near-extinction of all people in South Asia, then we would not see such a huge diversity of M in the subcontinent but all deried from some sublineage, like maybe M2 or M2a or even M2a5 if such lineage exists. Something several mutations downstream. The stem of M is too short (just 4 mutations, 3 if only coding region) and actually shorter than that of N (5 mutations, all in the coding region). We should see a much greater diversity (within the Eurasian context) in SE Asia and other less affected areas (Australia, Mid-East Asia) than in South Asia. But we don't see it.
So I have the feeling that Oppeneheimer is forcing things a bit. I also think that people could have been affected by such a calamity without going through a bottleneck as such, at most a very mild one. Finally I also think that the effects of Toba were specially climatic, this is true for the mid and long run, and as such affecting the whole Earth.
Raining ashes may cause a major disruption, including social and emotional disruption, while still allowing for life to follow after a year or two of such hardship. Grass would grow again on the ashes, animals would recover... and so would humans.
But I confess myself unable to recreate the fine detail of such episode.
The wealth of linguistic data and kinship data point to an east to west migration.
False. Only your interpretation of it does.
140 language families in America vs. 20 in Africa is the result of the work of hundreds of scholars.
Students of kinship could never find any old types in Africa. Again, dozens of scholars couldn't, from Morgan on.
Archaeology is consistent with it.
Absolutely false.
Absolutely true: the emergence of modern human behavior in Eastern Eurasia and in West Asia are well-dated. Africa (Dabban) and Europe (Aurignacian) was the recipient of those, not the inventor.
Current genetic phylogenies are flawed and in need of revisions.
I am very certain that this is also false. just a convenient pseudo-reasoning that meets your desires. Mere wishful thinking at best, self-deception at worst.
You consider molecular clock to be pseudo-science but phylogenies as true science. You just can't think straight, Luis.
There have always been phylogeographic alternatives in which geographically widely spread lineages are basal and geographically restricted lineages are terminal.
In your dreams.
Read: Johnson et al. 1983; Excoffer and Langaney 1987, 1989. Old stuff but good. All revisions involve going back to the root of the problem and resetting the outcome.
Phylogeny is structured hierarchically with nearly no doubts as a matriushka doll of sets and subsets. Every single sublineage shares the defining SNPs of its ancestor necessarily. You know that so you are lying to yourself and the rest.
There are different ways of organizing mutations in sets. Read some phylogeny manuals, or something.
Please stop relying on belief when it comes to data analysis.
Apply that to yourself, please!
When you learn the basics of scientific methodology and develop critical thinking, Luis, I very well may. So far I haven't encountered a challenge that would make me change my mind.
Maju: "If there would have been such a bottleneck, i.e. the near-extinction of all people in South Asia, then we would not see such a huge diversity of M in the subcontinent"
What I see (not sure if Oppenheimer would agree) is not one single bottleneck, but a set of bottenecks, all caused by the same event, representing a situation where more than one (possibly several) of the Out of Africa colonies was very hard hit, but nevertheless was able to survive, only with a greatly reduced population.
This theory has great explanatory power, because 1. the resulting founder effects could have produced at least some of the most important morphological distinctions we now see and 2. the greatly reduced populations would have had to re-invent themselves culturally, each in a somewhat different way, which could account for much of the cultural variation we now see among indigenous peoples, including linguistic differences.
If there is no Toba, we might have to invent one. :-)
But founder effects in the context of a rapid coastal migration would look exactly the same and the evidence for a bottleneck as such is rather non-existent or at least very weak, as I have argued above.
You "need" (or you think you need) of a bottleneck in order to explain the loss of P/B but I wonder if you have considered seriously other possibilities.
What possibilities?, you will ask. IMO there are several. A non-comprehensive list:
- A calamity such as Toba ash "rain" without causing a true bottleneck but with a deep psycho-cultural impact possibly (mentioned above by me)
- Mere cultural founder effects. I know you are highly reluctant to this explanation because you emphasize a lot cultural conservatism but I am not really persuaded that you are so totally correct in this, specially in the event of a long migration and such diverse founder effects as happened in southern Eurasia after the OoA.
- That the vanishing of P/B traits in most areas is recent, Neolithic or post-Neolithic, corresponding to cultural expansions such as Indoeuropean, Semitic and Chinese (and possibly others like Dravidian), Neolithic or civilized or highly patriarchal cultures that do not allow for the cooperative nature of ancient societies to persist easily, affecting indirectly the quality of music, though in a somewhat irregular process.
There are other possibilities, like maybe that the catastrophically disruptive events happened in the migration process, like maybe with the desertization of West Asia (if they did not follow the coastal route), influences from Eurasian hominins like Neanderthal or Erectus. Etc.
Or any combination of them, or something we haven't even considered.
Maju: "the evidence for a bottleneck as such is rather non-existent or at least very weak, as I have argued above."
Whether you see it explicitly as evidence of a bottleneck or something else, there is clearly something different about the South Asian genetic picture. As reflected in the two articles I quote from above and also in the molecular clock article that you disagree with so much. The first article says specifically that they can find no evidence of the Out of Africa migration in the current genetic picture for India. Why might that be?
They suggest that "in India the genetic traces of early migrations along the southern route were erased by the subsequent migrations which shaped the present-day mtDNA gene pool of India." Maybe that's what I see, and whether this was preceded by bottlenecks may be something we'll never be able to determine, one way or the other. Regardless, there is definitely a gap and it's the gap that interests me the most. Because it makes so much sense.
And no, it isn't I who needs a bottleneck but I just happen to be the first (as far as I know) to explore the cultural ramifications of Out of Africa in this particular way and anyone else trying to follow the migration would notice the same gap and need to account for it. It's not just a musical gap as I keep stressing.
The alternatives you suggest are wll worth exploring, but keep in mind that any meaningful "solution" would have to be consistent with the overall distributional patterns that now exist. And some of the very large scale distributions are NOT consistent with normal evolutionary "drift."
"It's not just a musical gap as I keep stressing."
Victor, I think, in the flurry of arguments, I missed your explanation as to why monophony couldn't have been lost in Africa after an out of Africa migration. What prevents you from sticking to Lomax's two-root idea and then adapting it to new genetic evidence and hypothesizing the co-existence of two different musical traditions at 60,000 YBP, with the monophonic one eventually disappearing in Africa but surviving in migrant populations elsewhere. In this case, you won't have a gap in India or in Europe or in North America or in Australia. Same for tones: you can have both tonal and toneless languages at 60,000K in Africa, with both "traditions" experiencing differential survival in and out of Africa.
I think you're steeped in unilinear evolutionism that you modernize by putting a unilinear devolutionist twist on it. (smiley face.)
The first article says specifically that they can find no evidence of the Out of Africa migration in the current genetic picture for India. Why might that be? -
I presume that you mean this closing comment:
In summary, although the data support a recent India–Australia connection, we could not find in Indian tribals any unquestionable genetic signature of the 60 000 year old
migration from Africa to Sahul following the postulated southern route. A possible explanation would be that such migration never occurred along that route. Alternatively, the early migrants from Africa may have made their way to Sahul following the southern route without settling in India. Another possibility, which is probably the most reasonable one, is that in India the genetic traces of early migrations along the southern route were erased by the subsequent migrations which shaped the present-day
mtDNA gene pool of India.3
The answer is obvious for anyone familiar with the human genetic landscape: Australian Aborigins are mostly mtDNA N and Indians are mostly mtDNA M. Logically there is no obvious connection.
However if you look at the Fst graphs, it is very apparent that South and East Asia occupy a central position in the Eurasian distribution and that South Asian tribes scatter along very different axis and to very distant corners, almost as much as Negritos do (and certainly more than West Eurasians). While the methods are limited (HVS-I) and the paper is old (most of mtDNA M has been only researched after that 2003 date, which is rather "ancient" for such a thriving discipline as population genetics), I think that it does reflect anyhow reasonably well for its date the general pattern of Eurasian expansion, with long (old and isolated) branches leading to Negritos and some South Asian tribes.
The Australian and other N branches must have got some distinct node in SE Asia (probably). The origins of this node are (and will surely be for many years) controversial but the simplest answer is Founder Effect.
If you need a bottleneck (near extinction) to explain every single founder effect, probably Humankind would not have survived to this day. Founder effects are caused by the small initial colonizer population that may perfectly have markers that are rare at their place of origin and may eventually even disappear totally there (by mere drift). There are dominant founder effects everywhere where an ancient colonization happened on more or less virgin land and there are minor founder effects for nearly every other migration, specially those that did not imply a continuous gene flow for centuries.
Whatever the case, the paper does not seem to mention any bottleneck in South Asia nor is apparent to my critical eye from the large diversity they detected.
Regardless, there is definitely a gap and it's the gap that interests me the most. Because it makes so much sense.
There is a musical gap (I trust you on that) and there is some genetic gap (the difference between the M and N nodes, for example, which may be related). However, beware, West Eurasians are mostly mtDNA N and yet the closest relatives (austosomal DNA: all lineages, not just a few random ones) are South Asians, who are mostly mtDNA M. And this is not probably just the product of post-Neolithic backmigrations but from the fact that West Eurasians (and most of their N lineages) are rooted in South Asia, even if N is rooted maybe in SE Asia.
It's damn complex, as everything in real life. You can simplify, you must simplify to think about the fractal complexity of reality but it's very difficult, a true challenge, to do it well, in an integrative and efficient manner.
The tree and the forest, sure. But ask a Pygmy about that, I'm sure he/she will agree that not all forests or parts of the forest are the same. And they are certainly different to a large extent because of the trees that make them up.
And some of the very large scale distributions are NOT consistent with normal evolutionary "drift".
Founder effects. I'm pretty sure about that, at least in the aspect of genetics.
German: "Victor, I think, in the flurry of arguments, I missed your explanation as to why monophony couldn't have been lost in Africa after an out of Africa migration."
Anything is possible, but I see no evidence for that.
"What prevents you from sticking to Lomax's two-root idea and then adapting it to new genetic evidence and hypothesizing the co-existence of two different musical traditions at 60,000 YBP"
This was also my idea. I was the one who first recognized what is most distinctive in Paleosiberian singing, which I named "breathless" style. And you can see the two-root idea in the earlier phylogetic tree reproduced on this blog, a few posts after the new one.
The two-root theory is consistent with the multiregional model, not with any Out of America model that I can think of, so I'm not sure why it interests you. Breathless style is not typical for the Americas, though it has been found, as I recall among the Guarano and the Ona (Selknam). Also among the Vedda.
Multiregionism never made much sense to me and now that we have so much genetic evidence supporting Out of Africa and the Pygmy-Bushmen relationship I dropped the double-root idea. A single root in Africa makes far more sense in terms of all the other evidence. However, I still see Paleosiberian music and the circum-Arctic complex as of great importance and have written about it in "Echoes." It may even be 60,000 years old, or older. But I can't see making a case for it as an origin point for music. It now seems to me to be a derivation from P/B, in which yodel was retained as glottal shake and hocket was retained, but only in the throat singing traditions. The shift from complex polyphony to monophony must have been due to a drastic population shift, a bottleneck or extreme founder effect of some kind. That's how I now see it in any case.
I know you would love for me to completely change my approach to suit you, but sorry, I see no reason to do that.
Maju: "There are dominant founder effects everywhere where an ancient colonization happened on more or less virgin land and there are minor founder effects for nearly every other migration, specially those that did not imply a continuous gene flow for centuries."
Then neither of these will do. I need an explanation for something special that seems to have been centered in South Asia, and produced the gap I've been writing about (in case you haven't noticed).
If the evidence pointed toward typical founder effects then that would be that and it would be easy to explain.
You think I have a theory that I'm trying to defend, but on this matter trust me I am simply trying to understand the evidence and the evidence points to the sort of bottleneck described by Oppenheimer. You are clearly not bothered by the same evidence I see so I understand why you are so skeptical.
If an ordinary series of founder effects could explain the gap I'd be happy to accept such an explanation.
"Anything is possible, but I see no evidence for that."
Do you see evidence for polyphony in Australia? Do you see any transitional forms between polyphony to monophony that could illustrate this major change? I thought polyphony survives in segregated "pockets," not in transitional forms.
The two-root idea interests me not only for its possible relevance to the out of America idea but also as a way to understand the evolution and distribution of musical styles vs. gene trees better. What I meant is that the two styles of music could have been carried out of Africa, no? This would make India and Australia look less aberrant, no?
I understand you're probably tired of going back and forth on things. But I'm still battling with musical evidence.
"I know you would love for me to completely change my approach to suit you, but sorry, I see no reason to do that."
But you do see evidence of a back-migration (pre-Semitic) into Africa, as reflected by M1 and U6 mtDNA lineages and Y-DNA E lineages? All genetic trees attest to it. Again, both macrohaplogroups (M and N) are represented in this migration. Dates are 25-40K. This must have some kind of reflection in music. If there's no reflection, then music signatures may get lost, so you could postulate two traditions in Africa at 60K and still not find any modern traces of monophony.
"The shift from complex polyphony to monophony must have been due to a drastic population shift, a bottleneck or extreme founder effect of some kind."
But at least in some places along a southern route P/B style is well and alive, right? Slightly simplified but still well and alive. So why would a bottleneck hit India, Australia and North America but spare Southeast Asia, Papua New Guinea and South America?
People who have no sympathy for my out of America idea will still be questioning your results and methodology because certain questions remain unanswered. If you postulate both traditions in Africa at 60K, you may save yourself the headache of fighting off typecasting as a pre-Lyellian catastrophist.
"The shift from complex polyphony to monophony must have been due to a drastic population shift, a bottleneck or extreme founder effect of some kind."
One more question/observation: Nettl, in his analysis of North American Indian music, contrasts simple monophony (Atahbascans) with complex monophony (Plains, Southwest), with Great Basin being in between. In your theory, polyphony devolves into simple monophony and then simple monophony evolves into complex monophony. Is it right? Or complex polyphony devolves into complex monophony? We discussed it already, and I know your take on it, but again, for the record, Nettl describes the music of Eastern Woodlands as "rudimentary imitative polyphony" and as "call and response" antiphony. At least at face value rudimentary polyphony seems to be a continuation (at least geographically speaking) of complex monophony.
German: "What I meant is that the two styles of music could have been carried out of Africa, no? This would make India and Australia look less aberrant, no?"
That's a very interesting point, German. “Monophony” is an ambiguous term, that can stand for either solo or unison performance. Solo vocal traditions are not uncommon among Pygmies and Bushmen though in some cases limited to lullabies. However, unison vocalizing doesn’t seem to be very common and may in fact be consciously avoided. We do find it, however, in certain African traditions, though polyphonic singing seems much more common in SS Africa generally.
What I point to, however, in my “Echoes” paper and in this blog is not so much an evolution from polyphony to monophony as a loss of polyphony. It is the loss of interlocking polyphony that is most evident in India, since there are some Indian tribal people who sing polyphonically, but in a very simplified manner (most sing in unison, though, as do the Australians).
The evolution of solo song is a somewhat different problem that I haven’t said all that much about yest. However, what I see as most important is the loss, first of interlocking polyphony (both vocal and instrumental) in South India, and then of polyphony altogether in most of the rest of Asia, except for SE Asia and southern China. It is this LOSS rather than a simple evolution that is so puzzling.
I could go on about this, but won’t since I’ll be writing soon about this same issue in upcoming blog posts. I do appreciate your question, which does make sense and deserves a more complete answer, which you will hopefully have from me soon.
German: "But at least in some places along a southern route P/B style is well and alive, right? Slightly simplified but still well and alive. So why would a bottleneck hit India, Australia and North America but spare Southeast Asia, Papua New Guinea and South America?"
If you look at a map you'll see that Mt. Toba is located in northwest Sumatra. The prevailing winds are generally thought to have been northwesterly. Keeping that in mind it's easy to see that exactly the region where I find the gap is the region that would have been most affected by the Toba explosion, because it would have been downwind from it.
The region where I have found survivals of the "African signature" in music but also other areas, such as wood carving, masks, etc., is the region that would not have been in the path of all the soot and ash from the eruption -- either off to the side or upwind or in a completely different area.
Australia would not have been directly affected by Toba, no. But the strong morphological similarity between Australians and Australoid tribals in the south of India suggests that the Australians could be descended from a population that was located in South India at the time of the Toba blast. This doesn't prove anything, of course, but it does provide a possible explanation for the absence of polyphony in Australia as well.
As for the Amerindians, aside from a possible first wave that could have been directly descended from HMP, most Amerindians are descended from "Mongoloid" peoples who would, according to this model, be descendents of peoples originally affected by the Toba explosion.
The above model is probably just as if not more controversial and hard to swallow than your own Out of America model, I admit. But it is also the only model I've been able to come up with that can explain the large-scale diversity we see, not only in the music but in many other aspects of culture, genetics and morphology.
But the strong morphological similarity between Australians and Australoid tribals in the south of India ...
Weren't you a few posts ago arguing that Australia was not settled from India? Something I agree with but that you threw against my objections anyhow.
Victor, your last two responses are very interesting and informative. Thank you. I would like to learn more about "monophony" (I realize it's a blanket term) on a worldwide scale. Maybe even to see a hypothetical tree connecting all forms of monophony, from more to less complex. The pattern that I see is that polyphony and monophony coexist on all continents and subcontinents. Sometimes we see the sharp decline of complex monophony (Africa), sometimes the sharp decline of complex polyphony. (I see the same pattern in kinship structures, with some types going back to exogamous moieties and others to endogamous bands.) Most polyphonic traditions are found in lower latitudes, most monophonic traditions are found in higher latitudes. But then suddenly you have strange overlaps: polyphony in the Subarctic and monophony in Australia.
"As for the Amerindians, aside from a possible first wave that could have been directly descended from HMP, most Amerindians are descended from "Mongoloid" peoples who would, according to this model, be descendents of peoples originally affected by the Toba explosion."
Paleoasiatic peoples, among which we find throat singing (with ties to polyphony, as you described) are very similar to (North) American Indians genetically (mtDNA and Y-DNA) and culturally. Typical Amerindian mtDNA lineage D1 was found in Jomon remains, which are likely ancestral to the modern Ainu who also have throat singing. I'm very surprised that you don't see a tighter link between Paleoasiatic and North American Indian music.
"The above model is probably just as if not more controversial and hard to swallow than your own Out of America model, I admit."
I guess people are divided into two kinds: those who believe in catastrophes, and those who believe in things hidden from the foundation of the world. To me, the Toba explosion explanation is just as fantastic, as the out-of-America idea is to you. But both are attempts to account for what you call "gaps" in the distribution of genetic and cultural traits. I think we both agree that those gaps are real.
Maju: "Weren't you a few posts ago arguing that Australia was not settled from India?"
Depends what you mean by "settled from India." I've been arguing for a possible direct link between the Australoids of India and those of Australia, both stemming from the effects of a Toba or Tsunami or famine or whatever-induced bottleneck centered in southern India.
And by the way I was rummaging around in some old pdf files and found this:
"In sum, we found that 50% of the Y chromosomes sampled from aboriginal Australians share common ancestry with a set of Y chromosomes that represent less than 2% of the sampled Indian subcontinent paternal gene pool. The similarity among C* chromosomes is unlikely to have been caused by chance convergence because we genotyped ten independent STRs. The observed pattern is not specific to central Australians, since our sample also included individuals from the Great Sandy Desert and from Western Australia, and our estimate of the frequency of C* chromosomes agrees remarkably well with other studies of greater numbers of aboriginal Australian Y chromosomes in Arnhem Land, the Great Sandy Desert, the Kimberleys, and the Northern Territory." "These results provide strong evidence for an influx of Y chromosomes from the Indian subcontinent to Australia that may have occurred during the Holocene."
From "Gene Flow from the Indian Subcontinent to Australia: Evidence from the Y Chromosome," by Redd et al., Current Biology, 2002.
I'm not saying this is the last word on this topic, but you have to admit that there IS an argument to be made for an India-Australia link.
The argument is there but it is always very limited: only 2% of Indian Y-DNA, unable to justify the alleged morphological similitudes (too shallow in any case - IMO a matter of shared preservation of some archaisms in the past much more widespread, as shown in the archaeological record, not any special relation). Plus it's a 2002 paper, from a time when Indian C was mush less researched (Y-DNA C5 was only described in the last two years). Plus it is generally accepted that India could have been at the origin of all Eurasians, including Australian Aborigines. Plus some have argued that there is a more recent, albeit very minor, flow between India and Australia.
I also know those who defend a direct relation between Papuans and West Eurasians, again based on morphology and some selective genetics. But in that case too I am extremely reluctant to accept it. I'd rather think that an "Euro-Australoid" morphotype (or at least morphological tendencies) was dominant among early Eurasians.
German: "One more question/observation: Nettl, in his analysis of North American Indian music, contrasts simple monophony (Atahbascans) with complex monophony (Plains, Southwest), with Great Basin being in between. In your theory, polyphony devolves into simple monophony and then simple monophony evolves into complex monophony. Is it right? Or complex polyphony devolves into complex monophony?"
I have no theory of musical evolution per se, German. You clearly want me to have one and feel that I should have one, but I don't. The evidence I see tells me that there have been many instances where polyphonic vocal traditions have been lost. In some cases we see very simple monophonic traditions instead. In other cases, we see more complex traditions. The difference in complexity between Plains and Athabascan (e.g., Navaho) has nothing to do with polyphony as far as I can see. And I see no way of telling which came first, though the differences are very interesting. (But the similarities are imo much more important.)
In Asia, complexity with respect to monophonic music is usually seen in the highly embellished and otherwise elaborated virtuosity and subtlety of solo singers, and also solo instrumenalists, which is probably an expression of the exreme specializations thought to be characteristic of the Neolithic.
But again, this complexity has no relation to polyphony that I can see and did not really "evolve" from it, though it probably did evolve from a much simpler type of monophony that is likely to have appeared just after the polyphonic tradition was lost.
As far as Nettl's take on Eastern Woodland music, his book on The Western Traditions of world music doesn't mention polyphony, just their tendency to sing antiphonally. Since these were the first Indians in N. America to be exposed to Western culture and music (probably hymns) it's hard to evaluate what it would mean if the occasional song were to contain some traces of harmony. In my experience, polyphony among N. American Indians is very rare and when we hear it it's usually an accident (inexperienced singer producing a wrong note). The only exception would be some California Indians who sing in a kind of shouted hocket and it's been reported for NW Coast, though most of what I've heard from there is unison singing.
"I have no theory of musical evolution per se, German. You clearly want me to have one and feel that I should have one, but I don't."
Yes, this is one of our recurrent points of contention. I can't possibly imagine how a student of prehistory, biological or cultural, can work without an evolutionary model. You model your data on population genetic phylogenies but population geneticists again don't have an evolutionary model (social organization, geography, demographic structure, population scenario) in place. This was the very early criticism of the out of African model that came out of the pre-mtDNA "school" of genetics. See, e.g., Neel JV (1991) Estrutura populacional dos amerindios e algumas interpretaqoes sobre evoluqao humana. In WA Neves (ed.) Origens, adaptacoes e diversidade biologica de homem nativo de Amazonia. Pp. 25-38. Belem.
An absence of an evolutionary model naturally leads you down the road of a catastrophic thinking.
Maju: "The argument is there but it is always very limited: only 2% of Indian Y-DNA, unable to justify the alleged morphological similitudes (too shallow in any case - IMO a matter of shared preservation of some archaisms in the past much more widespread, as shown in the archaeological record, not any special relation). Plus it's a 2002 paper, from a time when Indian C was mush less researched (Y-DNA C5 was only described in the last two years)."
Maju Maju Maju! First you dismiss the correlation I’ve found between music and tone language on the basis of a complete misreading of my musical maps. Then, after I correct you, instead of apologizing and accepting that I could be right after all, you go on to find other objections based on the fact that non-tonal languages are also important in SE Asia. When I explain that this is beside the point, you then make a beeline to the Andamans to point out that their language isn’t tonal either. And when I counter by pointing out the many instances of indigenous tone language in New Guinea, you find some other reason to dismiss the whole idea.
Same with the notion of a gap centered in South Asia, which you originally claimed there was no genetic evidence for. Then, when I provided you with genetic evidence from prominent researchers, you dismissed it as obviously wrong. And after I then went on to provide additional evidence, you dismissed that as well.
Now, after you’ve insisted there were no genetic signs of a connection between India and Australia, and I’ve provided you with two separate studies based on exactly that sort of connection, you refuse to admit you were wrong but find other reasons to dismiss the whole idea. This is the definition of a nit-picker, Maju, and I warn you, unless you change your ways, you will become a permanent member of that sad tribe. :-)
Science is not about establishing definitive proof. It is about exploring, formulating testable hypotheses, and testing them to see whether there is some reasonable fit with the evidence. There will always be holes to be found in any argument, and they need to be taken seriously, but science would grind to a halt if we let the nit pickers have the last word.
Maju Maju Maju! First you dismiss the correlation I’ve found between music and tone language on the basis of a complete misreading of my musical maps. Then, after I correct you, instead of apologizing and accepting that I could be right after all, you go on to find other objections based on the fact that non-tonal languages are also important in SE Asia.
1. Why should I apologize? If you make "bad" or hard to understand maps is not my fault but, if anybody's, yours. You have not "apologized" either for misreading my words on China (that you insisted to read as "Chinese language") or not knowing what Indochina is or in general oversimplifying the East Eurasian picture and never clarifying your arguments on this matter enough but yet insisting that you were right. I won't ask for an apology (I almost never do) but I do think that you are as much in fault as I am, if not more.
2. The more I look into the matter the more confusing and unclear I see it, specially in what regards to your claims of correlation between P/B survivals and tonal languages.
3. I am highly allergic to being patronized and strongly inclined by nature to slam the door and head elsewhere when I feel insulted that way.
When I explain that this is beside the point, you then make a beeline to the Andamans to point out that their language isn’t tonal either. And when I counter by pointing out the many instances of indigenous tone language in New Guinea, you find some other reason to dismiss the whole idea.
Because your arguments are not convincing on this point. It's not your fault that SE Asia (south of the Chinese border) is largely lacking any linguistic evidence from the Paleolithic, with only Andamanese (and arguably Austronesian) language families (both non-tonal) but I do think it is lack of rigor on your side to insist on such correlation when the evidence is so weak and contradictory.
Same with the notion of a gap centered in South Asia, which you originally claimed there was no genetic evidence for. Then, when I provided you with genetic evidence from prominent researchers, you dismissed it as obviously wrong. And after I then went on to provide additional evidence, you dismissed that as well.
Now, after you’ve insisted there were no genetic signs of a connection between India and Australia, and I’ve provided you with two separate studies based on exactly that sort of connection.
Well, may I remind you that the paper of Soares was provided by me to you and not the other way around. I did so you could have different viewpoints on the matter with a recent paper on age estimates that are different than my own. I do not fear different opinions but I am ready to discuss them openly.
The paper of Cordaux is highly irrelevant and obsolete anyhow and I have already explained why to you and whoever reads this discussion. Feel free to discuss the matter objectively but stop making appeals to authority to counter good arguments because that is a quite poor way of doing science, rather a method to mislead by selectively picking the authors that better fit you and selectively interpreting their conclusions also at your own convenience.
I have in the past provided other materials that you have conveniently forgotten. When I engage in debates and collaborations like these, I expect honesty from the other side, not tendentiousness and sloppiness.
If you have some critical neurones still left, and I believe you do, you must realize that all the supposed Indo-Australian connections you have mentioned are ridiculously trivial (all refer to minor lineages and in some cases the phylogeny is not even clearly established - and all or most of them claim this alleged minor connection is recent, not ancient).
This is the definition of a nit-picker, Maju, and I warn you, unless you change your ways, you will become a permanent member of that sad tribe.
Sincerely I prefer to be a nit-picker that to be self-deluded as you seem to be right now.
... but science would grind to a halt if we let the nit pickers have the last word.
Science would not be science if we would stop paying attention to the details. The Devil is in the details, goes the English saying, and has lots of reason. If we ignore them, we just get lost in a world of imprecise fantasy and speculation and science dies.
Rigor is important.
And by the way, please notice this counter-point to Soares' work just published at PLoS ONE by Loogväli et al. (Kivisild's team).
Notice in particular table 6, that gives a significative older date to mtDNA M (75 Kya) than to N (68 Kya). They specifically mention that they do not get the same substitution rates as Soares.
Needless to say that I also have my reserves regarding this paper, specially on the recent ages attributed for Pan-Homo and intra-Homo divergence that IMO should be at least 23% older and maybe as much as double.
Maju: "I do think it is lack of rigor on your side to insist on such correlation when the evidence is so weak and contradictory."
The problem, Maju, is that you are the one who is continually insisting, not me. I am sorry to have offended you, especially because your contribution to this blog has been so great and so frequently positive. I would rather have you here nitpicking than have you go off in a huff because even when you nitpick I learn something of value.
But you (and German also) continually assume I am taking certain positions when I am not. I see the correlation between tone language and music as a clue that could be extremely fruitful and if I insist on anything, it is on the importance of pursuing such a possibility.
You seem to feel that if it's not 100% then it can't be right and can safely be dismissed -- and to me that is nitpicking, sorry. I see no need to prove something at a point when I am simply exploring it as a possibility.
Face it. There are significant instances of tone language in greater SE Asia. There is no denying it. And while it is true that there are many non-tonal languages as well, we know that the most widespread of these, Austronesian, is recent and must have supplanted many of the languages that had been in place earlier. I'll be having more to say about this issue in future posts and you can snipe at me then if you like, but I see no point in using Austronesian as a counter-example when we have no way of knowing what the picture was before its expansion in this region. When you use it as a counter-example anyway, then you give yourself away, Maju, as someone who is simply looking for whatever couter-example you can find, instead of seriously thinking through the problem. And by the way this is not just my problem but a problem for anyone genuinely interested in understanding why there is such a huge cultural (not only musical) gap between Africa and greater South East Asia.
Maju: "Science would not be science if we would stop paying attention to the details. The Devil is in the details, goes the English saying, and has lots of reason. If we ignore them, we just get lost in a world of imprecise fantasy and speculation and science dies."
It's because I am paying close attention to the details that I see problems you do not see. And I see clues to the possible solution to these problems that you can't see either, probably because you are insentive to the problems in the first place.
If we cannot account for the gap, then we will be unable to respond to multiregionalists who will argue that all the many similarities between Africa and SE Asia/Melanesia, etc. must be due simply to convergent evolution, and we will be back at the crudest stage of the multiregional position, which even Wolpoff and Templeton no longer try to defend.
Even if you can’t accept the musical argument I’ve presented, based on P/B, then at least acknowledge the problem posed by the distribution of musical instruments, which even the archaeologists can’t ignore.
The gap is real. If we can’t account for it within the context of the Out of Africa model then we might as well give up that model and go back to the original multiregional dogma, that humans are who they are due to convergent evolution taking place in parallel in various parts of the world. So not only music, but language too would have to be understood as somehow magically “evolving” in the same direction at roughly the same time due to some intrinsic, built-in, magical faculty of the human mind.
I’m not saying I have the solution, and if you have a better idea on this matter I’d love to hear it. But don’t deny the problem exists – it is very real.
"First you dismiss the correlation I’ve found between music and tone language."
Allow me to interject, Victor. You haven't found a correlation between tones and music. First, you acknowledged that there's no functional dependencies between the two (as there's between yodel and glottalization). Second, you dismissed a cognitive science monograph that suggested that such a connection may exist, specifically between a certain type of tonal languages and musical pitch in general but you don't even have a map for the distribution of register vs. contour tones and you don't seem to care about this distinction. Third, you dismissed in-depth linguistic studies (by Blevins and others) that argued for parallel evolution and diffusion of tones in different parts of the world. (You tend to latch on any genetic study that seems authoritative to you, while dismissing all studies of languages produced by authoritative linguists.) Fourth, you dismissed a serious implication of these studies for your mapping correlations, namely that 10K years ago (50K after the proposed out of African migration) there probably was no similarities in the distribution of music styles and tones.
The genetic study you brought up is interesting, though, and deserves a further look. It's also a bit oldish from the point of view of haplogroup C research. Do Redd et al. refer to what would later become subhaplogroup C5? In this case, we have three remote "pockets" of C lineages - North America, Australia and South India - plus, of course, Asian regions in between. This triangulation is different from your southern route signature.
"There is no denying it. And while it is true that there are many non-tonal languages as well, we know that the most widespread of these, Austronesian, is recent and must have supplanted many of the languages that had been in place earlier."
Blevins again made a convincing case that Ongan (Andaman islands) and Austronesian are related languages. Neither of them has tones. Some of the younger languages and some of the older languages in Asia have tones. This tells me again that tones flicker in and out of the picture. It's just so simple, Victor.
"So not only music, but language too would have to be understood as somehow magically “evolving” in the same direction at roughly the same time due to some intrinsic, built-in, magical faculty of the human mind."
Linguists don't care about out of Africa or Multiregional one way or another. They work with facts. Linguistic theory, which is built on two hundred years of observing these facts says that phonetic and grammatical similarities are most likely due to convergent evolution. You do seem to be biased in favor of the out of Africa model vs. Multiregional. Out of America is also built on the recognition that certain cultural and genetic "mutations" are homoplasies, and if you want to get to the root of the issue you have to go beyond those.
German: "Allow me to interject, Victor. You haven't found a correlation between tones and music."
The correlation is there, you can see it if you compare my musical maps with the WALS tone language map. The question is: is it meaningful -- and on that question I have an open mind, though you and Maju seem convinced that I think I've proven it. I don't. I think it worth exploring and I intend to.
"First, you acknowledged that there's no functional dependencies between the two (as there's between yodel and glottalization)."
What I'm exploring is the possibility that both the musical and the linguistic distributions could have a common cause. I am not attempting to establish a cause and effect relation between music and linguistic tone so functional dependency is beside the point.
"Second, you dismissed a cognitive science monograph that suggested that such a connection may exist, specifically between a certain type of tonal languages and musical pitch in general but you don't even have a map for the distribution of register vs. contour tones and you don't seem to care about this distinction."
Not at this point, no. One thing at a time. Again, you completely misread my intentions. I am NOT presently trying to prove anything, but simply considering some of the possibilities. When I get more time, and if this issue pans out as something worth looking into further, then I would definitely want to look at every aspect of tone language certainly.
"Third, you dismissed in-depth linguistic studies (by Blevins and others) that argued for parallel evolution and diffusion of tones in different parts of the world."
That argument is questionable because we can see that a vast expanse of Europe and Asia exists where no tone language appears on the WALS map. If tone languages were something that could simply evolve due to parallel evolution we would expect to see them scattered across the globe. Nevertheless, this is a matter I certainly intend to look in to, as the WALS map is incomplete and there could be more tone languages in this region that what we now see.
German: "Fourth, you dismissed a serious implication of these studies for your mapping correlations, namely that 10K years ago (50K after the proposed out of African migration) there probably was no similarities in the distribution of music styles and tones."
Are you saying you know this for a fact? What research have YOU done on this matter. Sure, it's possible the correlation I see is simply an artefact. This is what I would like to determine, so this is what I intend to explore. You and Maju, for some reason, would prefer that I simply forget about such a possibility altogether, but I refuse to do that, sorry.
"Are you saying you know this for a fact? What research have YOU done on this matter."
Victor, there are several linguistic studies that argue for the relatively recent (within hundreds and a few thousands of years) origin of tones in Asia and America. Parallel evolution and diffusion are common processes that account for the current distribution of tones. Yes, I did my research back in the mid-1990s, while writing my first Ph.D. thesis, and now just double checked the most recent publications.
If you'd conducted a functional analysis of music styles and tones, then you could argue that, say, in certain parts of the Old World such as Papua New Guinea, Africa and South China tones are functionally dependent on some unique properties of pitch as found in the polyphonic music also concentrated in these areas. Then, these functional complexes straddling music and language would indeed look ancient and would account for the diffusion of tones into Southeast Asia and elsewhere where the musical properties in question aren't found. This is the logic I apply to polyphony and moieties (polyphonic ensembles in South America and Papua New Guinea replicate the division of society into cosmologically and matrimonially significant divisions, while in Africa they don't) and to throat singing and clicks (both based on ingression of the airflow).
I understand that everything you blog about is provisional, but you do seem to be lured away from nitty-gritty structural-functional analyses by the apparent parsimony of out-of-Africa solutions. When you encounter problems with out of Africa ("gaps," as you call them), you let the balloon of imagination even further out into the realm of natural disasters instead of focusing on building solid evolutionary models impervious to the vicissitudes of historical accidents.
"As far as Nettl's take on Eastern Woodland music, his book on The Western Traditions of world music doesn't mention polyphony, just their tendency to sing antiphonally. Since these were the first Indians in N. America to be exposed to Western culture and music (probably hymns) it's hard to evaluate what it would mean if the occasional song were to contain some traces of harmony. In my experience, polyphony among N. American Indians is very rare and when we hear it it's usually an accident (inexperienced singer producing a wrong note). The only exception would be some California Indians who sing in a kind of shouted hocket and it's been reported for NW Coast, though most of what I've heard from there is unison singing."
I looked at Nettl's Music in Primitive Culture. You're right that Woodlands polyphony could result from European influence. Hard to tell, although Nettl would have known, I suppose. But even from what you just said about NW Coast I infer that polyphony and monophony (unison) in some places are very tightly interspersed, which makes a bottleneck idea unlikely, unless of course two populations came together and blended their musical styles.
FYI: the most popular model of phonological evolution (again, as summarized in Blevins) explains changes as resulting from a hearer's misinterpretation of an incoming phonetic sequence. In both music and language, evolution seem to occur as a process by which individual mistakes achieve fixation in society.
The tone of your posts (and comments) don't make some things look as a mere exploratory possibility but as some sort of discovery you have suddenly become fond of (without sufficient research). And that's why I think I'm doing my "job" (so to say: playing my role as critical, yet supportive, reader and commenter) warning you about those.
Obviously I hate as much as you whichever misunderstandings. But, well, they happen - so let's get over them as soon as they are sufficiently clarified.
You seem to feel that if it's not 100% then it can't be right and can safely be dismissed...
No. But if it is just 50% (for instance) then it must be placed on some quarantine clearly. Quarantine doesn't mean it's impossible but that it's clearly not a demonstrated fact. I don't think it should be dismissed totally but neither waved as "evidence" and, IMO, should be pondered along with other equiprobable alternatives.
... but I see no point in using Austronesian as a counter-example...
I haven't used Austronesian as counter-example. I used Austroasiatic, which you argued would not be tonal originally (something I had totally missed, btw). But in any case, Taiwan, the urheimat of Austronesian, is also part of Greater SE Asia, by which I understand you mean including southern China.
... hen we have no way of knowing what the picture was before its expansion in this region.
I am satisfied I have played my role by reading you acknowledging that we really not know what most of SE Asians (Negritos, "Hoabinhians") spoke in the Paleolithic.
It's because I am paying close attention to the details that I see problems you do not see. And I see clues to the possible solution to these problems that you can't see either, probably because you are insentive to the problems in the first place.
Maybe, probably yes, but paying attention to details is what I think I do when you call me nit-picky. So maybe you are also nit-picky in your own way.
If we cannot account for the gap, then we will be unable to respond to multiregionalists who will argue that all the many similarities between Africa and SE Asia/Melanesia, etc. must be due simply to convergent evolution...
As you know, I don't think that continuity can be explained in most cases by mere convergent evolution.
But I do explore other possibilities for the gap or gaps (not sure anymore if it is just one). For example I've been thinking as of late that maybe there are other reasons pushing people to more or less gradually or partially to abandon the HBC. This issue of partial abandonment of the HBC elements is not only seen in Eurasia but also in Africa, even Hadzas seem to have abandoned some elements of HBC (following your reasoning) and certainly have Niger-Kongo peoples and others.
So maybe the process began already in East or NE Africa, becoming increased in the migration to Asia... maybe it is not (or not only) a catastrophic episode what accounts for it but a more gradual and irregular process of cultural "evolution".
I suspect that the main difference between my thought and yours is that you have almost total faith in cultural conservatism while I have more faith in cultural variability and innovation, without totally rejecting the weight of conservatism. However this poses the problem of how did Bushmen and Pygmies managed to be so conservative. The answer maybe is just their isolation (it is a well known phenomenon: isolated communities are almost invariably the most conservative ones).
But of course nothing of this is a definitive theory either, I try to remain open minded - unless I think I have more or less all the evidence in one basket, which is not the case.
German: "Victor, there are several linguistic studies that argue for the relatively recent (within hundreds and a few thousands of years) origin of tones in Asia and America. Parallel evolution and diffusion are common processes that account for the current distribution of tones."
I want to look more deeply into the literature on this issue when I get a chance. However, it's important for you (and Maju) to understand that for me one of the most important factors in evaluating a cultural attribute is its worldwide distribution. Unfortunately, the worldwide distribution of most aspects of culture are not widely known or studied. The WALS maps represent a very exciting new breakthrough, however, and the importance of this publication should not be taken for granted.
From what I've noticed on the WALS maps I've looked over so far is that almost all the traits they've mapped tend to be distributed almost randomly, or at least that's how it looks. Such a distribution is consistent with the pattern one would expect if these traits are largely determined by parallel evolution and diffusion.
The pattern we see for tone language is dramatically different, at least for the Old World. It is highly structured. And for me that must mean something. It's hard for me to see this as simply a coincidence.
There is no question that diffusion has taken place among the languages of greater SE Asia, and that at least some of the wide distribution of tone in this region is due to that. But if that's the whole story, then why don't we see similar patterns of of tone language diffusion in other parts of Eurasia?
There have been a great many studies of every aspect of linguistics by many investigators and every conceivable theory seems to have been promoted at one time or another, usually based on assumptions about how language worked or how the brain worked or what was universal, etc. The WALS maps are different, because, as with Cantometrics (which that whole approach strongly resembles) we see real evidence, and on a worldwide scale. This is new and exciting and because of my experiences with Cantometrics I feel very much at home with it.
I'm not saying that further examination of the details of tone language and its history are not important and I intend, when I get the time, to look into that. The map by itself proves nothing, clearly. But it does look to me at this point as though the correlation I see could be an important historical clue and I will certainly not abandon this line of research simply because some linguists might think otherwise.
German: "you let the balloon of imagination even further out into the realm of natural disasters instead of focusing on building solid evolutionary models impervious to the vicissitudes of historical accidents."
Evolution is largely about historical accidents, German. The notion that there is some sort of destiny behind the "stages" of evolution, based on certain universals was discredited a long time ago.
Maju: "The argument is there but it is always very limited: only 2% of Indian Y-DNA,"
Two percent is what would be expected, since we are talking about an affinity between a very small minority population in South Asia and a large population in Australia."
"unable to justify the alleged morphological similitudes"
Their sample from South Asia included both South India and Sri Lanka, which is the home of the Vedda, an australoid tribal group.
"(too shallow in any case" - IMO a matter of shared preservation of some archaisms in the past much more widespread,"
Certainly a debatable point.
"Plus it's a 2002 paper, from a time when Indian C was mush less researched (Y-DNA C5 was only described in the last two years)."
Has a refutation been published? Is the new data incompatible in some way with that presented in 2002?
"Plus it is generally accepted that India could have been at the origin of all Eurasians, including Australian Aborigines."
They are aware of this and if you read the paper you'll see that they took such a possibility into account -- and saw only a South Asiatic-Australian connection for the particular haplogroup under consideration. More recent studies may have revealed other possibilities, of course.
"Plus some have argued that there is a more recent, albeit very minor, flow between India and Australia."
They see the connection as recent, ca 5,000 years ago, and associate it with the introduction of the Dingo, plus several other indicators of significant change in Australia during the Holocene. This does NOT gibe with the picture I presentd in my "Echoes" essay, which posited a much earlier date for the connection, but it's not incompatible with it either. I'll be looking into this whole issue soon and reporting my thoughts on the blog.
Maju: "The tone of your posts (and comments) don't make some things look as a mere exploratory possibility but as some sort of discovery you have suddenly become fond of (without sufficient research)."
I don't consider myself an academic and have little use for the overly careful and cautious academic tone, so at times, when I am expressing myself informally, and expressing enthusiasm, I admit that I might give the wrong impression.
For example, when I write "it [the gap] is most certainly present in the distribution of tone languages" I now realize that this might sound as though I am already convinced that tone languages can be explained by the same gap I've found in the musical evidence. That is NOT how I feel, I am NOT yet convinced and if you read carefully you'll see that all I'm saying is that there is a gap on the WALS map for tone languages, which is certainly the case. It is A gap, but perhaps not THE gap, and I should have made this clear.
But there are many other places where I go out of my way to explain that I am simply speculating. As when I write, in the same post: "I want to consider all possibilities and to make clear that I'm not in love with the interpretation I've been presenting."
You are right, however, to challenge me when I seem to be going overboard on such correspondances and I hope you'll continue to do so. But you too have a problem with your tone, which often comes across as dismissive and arrogant.
Maju: "This issue of partial abandonment of the HBC elements is not only seen in Eurasia but also in Africa, even Hadzas seem to have abandoned some elements of HBC (following your reasoning) and certainly have Niger-Kongo peoples and others.
So maybe the process began already in East or NE Africa, becoming increased in the migration to Asia... maybe it is not (or not only) a catastrophic episode what accounts for it but a more gradual and irregular process of cultural "evolution"."
The questions you pose are important. And yes, we certainly see partial abandonment of the HBC norms in Africa. But in Africa the process seems closer to what we would expect from "a more gradual and irregular process of cultural "evolution"" because there are many continuities, and also relatively minor discontinuities, that we can use to reconstruct certain histories fairly easily. Whereas what we see along the Out of Africa migration route appears to me as very different, with a huge break in continuity.
For me, the biggest difference between us is that I am extremely fixated on worldwide distribution patterns and you seem willing to discount them and fall back on certain basic principles instead. To me these principles are based on assumptions, whereas the distribution patterns reflect (however indirectly) real evidence.
Two percent is what would be expected, since we are talking about an affinity between a very small minority population in South Asia and a large population in Australia.
Not a large population in Australia. The lineages are always minor in Australia. Australia's C is mostly C4, which does not exist today anywhere else.
By all accounts the closest relatives of Australians by "wholesome" (autosomal or autosomal+haploid) DNA are Papuans and Melanesians. However their relation does not seem much more recent or intense than what they have with other Eurasian peoples separately, hence refers to the early colonization period basically, maybe slightly reinforced by their neighborhood.
Has a refutation been published? Is the new data incompatible in some way with that presented in 2002?.
I don't think Cordaux is making claims that would require refutation. This study was part of the process of understanding Eurasian mtDNA beyond the over-studied European and Native American cases. In 2002 the phylogeny of M and most of N was still a huge mystery (and would be for some years after that, even today researchers are still discovering new rare M sublineages and possibly N ones too).
Cordaux data though has to be read on the light of what we know now, almost 8 years later. And that's what I did when I said that his suggestion that South Asian and Australian mtDNA is different is perfectly consistent with our knowledge of India being strongly dominated by M and Australia by N.
They are aware of this and if you read the paper you'll see that they took such a possibility into account -- and saw only a South Asiatic-Australian connection for the particular haplogroup under consideration.
Again mtDNA N. There's no other link.
They see the connection as recent, ca 5,000 years ago, and associate it with the introduction of the Dingo, plus several other indicators of significant change in Australia during the Holocene.
It's a possibility but the relevant lineage is extremely minor at both sides of the Indian Ocean: it does not "magically" make both peoples more akin, just shows that there was (maybe - not sure if this has been sufficiently confirmed) a contact.
I don't consider myself an academic and have little use for the overly careful and cautious academic tone, so at times, when I am expressing myself informally, and expressing enthusiasm, I admit that I might give the wrong impression.
For example, when I write "it [the gap] is most certainly present in the distribution of tone languages" I now realize that this might sound as though I am already convinced that tone languages can be explained by the same gap I've found in the musical evidence. That is NOT how I feel, I am NOT yet convinced and if you read carefully you'll see that all I'm saying is that there is a gap on the WALS map for tone languages, which is certainly the case. It is A gap, but perhaps not THE gap, and I should have made this clear.
But there are many other places where I go out of my way to explain that I am simply speculating. As when I write, in the same post: "I want to consider all possibilities and to make clear that I'm not in love with the interpretation I've been presenting."
You are right, however, to challenge me when I seem to be going overboard on such correspondances and I hope you'll continue to do so. But you too have a problem with your tone, which often comes across as dismissive and arrogant.
Fair enough. I think this sums it up. I admit I am somewhat arrogant by nature (which can be not such a bad thing if you are also self-critical and open to legitimate criticisms) though sincerely I wouldn't like to sound as dismissive.
Whatever the case, I know that, even if not any academic expert, I am much more familiar with the overall picture of population genetics than you are, because I have been following the matter with great interest for many years now. I am possibly also somewhat more familiar with some archaeological matters, another of my major interests - that can be sumed up as a deep interest for Prehistory, specially in what affects H. sapiens. On the other hand you are infinitely more knowlegeable than I am in what regards to musicology, issue on which I have only paid some attention to through this blog.
So when I perceive that you are becoming one-sided or confused on matters where my knowledge is probably wider, I try to explain it to you. This can of course sound "dismissive", I presume, but I really don't know how to express such necessary criticisms (of you or of some scholars too) without sounding that way.
The questions you pose are important.
I think they are, so I am glad that you acknowledge this.
For me, the biggest difference between us is that I am extremely fixated on worldwide distribution patterns and you seem willing to discount them and fall back on certain basic principles instead. To me these principles are based on assumptions, whereas the distribution patterns reflect (however indirectly) real evidence.
I do not understand well what you mean here. In particular I'd ask which are those "principles" you think are biasing my understanding.
In what regards to global distribution patterns, I think I am aware of the problem they may pose because I have also gone through that kind of what I'd call oversimplifying understanding in the issue of global human genetics and I still see many people, usually those who are relatively "new" or "shallow" on the understanding of the subject who tend to such oversimplifications, for instance over-emphasizing the duality of mtDNA M and N as if they represent two totally distinct populations and even "races".
In fact, and I think that everybody that has a somewhat deep understanding of the matter will agree (even if each one with his/her particular emphasis) this can't be the case but you really need to dive for some time in the "nit-pick" (the devilish details, the "yes, but..." and then again "but...", and so on) of the matter before you really gain a good insight.
So I think I am aware of the problems that such a pattern-based approach on a map can cause if not done with due care.
"Evolution is largely about historical accidents, German. The notion that there is some sort of destiny behind the "stages" of evolution, based on certain universals was discredited a long time ago."
As all non-academics (as you admit you are), you tend to fall into extremes. Human history is not accidental and not providential but systematic.
"The WALS maps are different, because, as with Cantometrics (which that whole approach strongly resembles) we see real evidence, and on a worldwide scale. This is new and exciting and because of my experiences with Cantometrics I feel very much at home with it."
The people behind WALS, however, don't share your assumption that their maps captured traces of out of Africa or Multiregional or any other kind of migration or retention. WALS is a great endeavor, and I corresponded with Comrie at some point (too late in their process) regarding including kinship evidence in it. One recent outcome of this mapping is that there are 24 grammatical and phonological traits that are significantly better represented in the New World vs. the Old World.
"so at times, when I am expressing myself informally, and expressing enthusiasm, I admit that I might give the wrong impression."
I personally like your enthusiasm and I agree that most academics, out of fear to look stupid, to expose their ignorance, to strike a wrong chord with colleagues, etc. choose an overly cautious tone. (Most non-academics naively take this caution as a sign of knowledge which is the opposite from what it is.) However, I wish you did more research (the positive side of academic comportment) and stepping into the shoes of an archeologist, a comparative ethnologist or a linguist before coming up with rather novel ideas inspired by genetic phylogenies. Otherwise, I always find myself in an awkward position of being forced to choose between ignoring your posts altogether and "unblogging" them in the comments section. This may be too much to ask from a blog writer, though. Especially, since this writer is the only one who tries to take a holistic perspective on human prehistory.
Post a Comment