June 11, 2011

The various options for the migration out of Africa

I want to call your attention to the latest entry at M. Petraglia and colleagues' blog Ancient Indian Corridors. The main focus of this short entry (with open access links to the relevant papers) is the Jubbah site in what today is the terrible Nafud Desert but what really called my attention the most is this other article: Trailblazers across Arabia, where Petraglia discusses not just his own work on the matter of the migration out of Africa of Homo sapiens but also the recent one by Armitage, who proposed a coastal migration via southern Arabia some 125,000 years ago.

This map from that article really synthesizes the various options, even if the matter on when the migration happened (125 Ka ago or more like 90-80 Ka ago) remains open:


Sadly the Ice Age sea levels are not reflected, for that reason we cannot appreciate that the Persian Gulf was then above sea level, constituting a fertile riverine and swampland region that surely hosted plenty human settlements.

This was addressed by J.I. Rose in 2010 and discussed by me here, among other key papers on the migration out of Africa into Asia.  This map may help understanding better the geography implicated in the OoA:



Something that I find fascinating about all this is how we are living in such a key moment of discovery of our own prehistory and how in this particular case Archaeology is following Population Genetics. As you probably know, I am adamant of double-checking genetics-based prehistory reconstructions with archaeology when this one is minimally developed and can provide unavoidable key references. But in the case of the migration out of Africa we were until recently very much ignorant of everything: we knew with good likelihood that humankind had formed in Africa and then... well, there are nice decorated caves in Europe and some controversial skulls in Australia. Between Omo II and Cro-Magnon I, so to say, we were totally amiss (and that's like 160,000 years of prehistory, most of our history as distinct species).

But then, because of certain peculiarities of the distribution of lineages in Asia, some geneticists found themselves forced to posit the theory of the rapid coastal migration out of Africa. The theory posited that our ancestors, or at least some of them had migrated, at least partly, by taking advantage of coastal ecosystems even in relatively arid areas. This coastal specialization allowed these beachcombers and boaters to move faster that any inland route would have allowed, largely able to ignore the difficulties posed by deserts, rivers and swamps alike... and even to cross some narrow sea branches as attested by the colonization of Australia and nearby islands, colonization that must have happened (Genetics seem to say) quite early on in the human adventure in Asia. 

This was several years ago and I recall the importance of this discovery in allowing us to understand better the most likely prehistory of our kind across Eurasia. Suddenly all the classical maps, which almost invariably included arrows across Siberia for some odd reason, became obsolete. However there was still no or almost no archaeological evidence. Arabia was not just a desertic region but also seemed an archaeological desert, with very few data, much less reliable one. Even in South Asia, when I tried to find any information on Prehistory, there was not much.

Luckily, in the last years this has changed a lot. And it is still changing: we are learning every other month something new about this key episode of our history as species. So here there go my thanks to Petraglia, Armitage, Rose and all the others who are helping so much in getting us to understand our (pre-)history, what our remote ancestors did, allowing us to be what we are.

28 comments:

  1. I have to agree with Petraglia that the very long intervals between occupations at Jebel Faya, together with what we know about the climate (extremely arid except for relative short wet phases), seems bottlenecks and replacements were likely. That also fits with our genetic discussion about possible back migration of y-DNA haplogroups G and IJ (or F and IJK). The thing is, once a population is established ooA in an area where it can expand and flourish without huge climatic disruptions (Pakistan, India), it is about as likely to reach Arabia under wet conditions as people from Africa.

    A little speculation: It would be interesting if it turns out that layer B corresponds to a second ooA (CF), and layer B to a backmigration (G, IJ), while the first occupation (layer C) corresponds to the not-so-successful D.

    ReplyDelete
  2. Correction: That should be "layer A" for the back-migration, above.

    ReplyDelete
  3. I'd wait till we can clearly correlate the typology of the findings in Arabia with Africa and South Asia. That should be very informative.

    However I can in principle agree that layer A allows for a back-migration from South Asia (in the wider context of the colonization of West Eurasia) and could well be related to mtDNA M1 (and some other rare Ms of SW Asia).

    As you can see I prefer to think in terms of mtDNA. But anyhow the only Eurasian Y-DNA lineage with traces of having so southernly roots in Arabia is IJ, so maybe there was some IJ or already evolved J over there. A 40-38 Ka date is relatively late in the context of the colonization of West Eurasia so it may allow for even J* (J3?) of the kind we still find in Socotra (but has obvious roots in the peninsula) to be the Y-DNA lineage in play there.

    That of course would imply that I, J1 and J2 were already coalesced or at least coalescing, which is no problem for me at all (but may be for others).

    I don't see G so far south on light of the available data, even if it's not impossible it went through the Persian Gulf Oasis.

    Armitage et al. anyhow argue for a continuous occupation, what Petraglia questions. It is not impossible that some of the mtDNA L(xM,N) that we see today in Arabia and are not really the kind that can be easily thought as product of the slave trade (L0, L4, L6...) may actually have been there in this continuous occupation, regardless that there was some back-migration, as we know almost for a fact (on genetic grounds).

    It's difficult to discern the fine detail, really, but it the newly (re-)discovered archaeological wealth of Arabia truly offers for a host of possibilities.

    ReplyDelete
  4. "A 40-38 Ka date is relatively late in the context of the colonization of West Eurasia"

    The way I see it, the more inland Indian lines had much more wide-spread and also larger populations. So, it only seems at the surface like these Arabian haplogroups appear too late. They -- just like T and L -- come from a somewhat marginal region in SW Pakistan, whereas the other, more successful lineages are the ones that took full advantage of the region but are not to be found at the "port of entry," because they quickly moved to the Pakistani/Indian fringes. So, R and other lineages appear younger in comparison because they come from a much, much larger population group. In reality, they are all about the same age.

    ReplyDelete
  5. You are right on Y-DNA T being a good candidate for such a secondary coastal spread (maybe together with J?) T is also present in East Africa, almost like M1. They could well have journeyed together.

    J spread may be different: though I'm still puzzled by the South Arabian J*, the J1/J2 split and the J/I one rather suggest a more northernly, Fertile Crescent origin.

    "R and other lineages appear younger in comparison because they come from a much, much larger population group. In reality, they are all about the same age".

    I do not understand this well but I think you must be right. The spread of P and then Q and R must be contemporary with the backflow of mtDNA N-derived lineages from (probably) SE Asia in westward direction across North India, culminating in the expansion of R (possibly the same as Y-DNA P) maybe in Bengal (or not too far away).

    P is itself a backflow from MNOPS, so that fits well. What is curious is that instead of taking (mostly) South Asian mtDNA, this "population" (??) took South Asian (or Persian Gulf Oasis) Y-DNA (notably IJ and G) in their push westward, and only very few South Asian mtDNA lineages (M leading to M1 and maybe other rare M subclades found in Arabia).

    But guess that's ok in the world of genetic founder effects, which are bound to be somewhat random.

    But totally agreed that they must be of closer age than they appear, whatever the reason.

    Are you planning in elaborating at your blog on this matter of larger and smaller founder populations? It may well make good sense and I'd love to read such an explanation.

    ReplyDelete
  6. "This map may help understanding better the geography implicated in the OoA"

    And note how wide the expanse of exposed shelf is along the Yemen coast. And obviously I think the red or green arrows in the first map are much more likely to reflect reality than does the purple.

    "The way I see it, the more inland Indian lines had much more wide-spread and also larger populations".

    And are probably the only meaningful ones to consider in any OoA scenario.

    "So, R and other lineages appear younger in comparison because they come from a much, much larger population group. In reality, they are all about the same age".

    I don't think 'they are all about the same age'. R is derived from P, which is derived from MNOPS, which almost certainly coalesced somewhere in SE Asia. So R's ancestors had moved a very long way before R reached the Arabian peninsular. They had left Africa and gone all the way to SE Asia through India, then turned around and moved all the way back, again through India.

    "The spread of P and then Q and R must be contemporary with the backflow of mtDNA N-derived lineages from (probably) SE Asia in westward direction across North India, culminating in the expansion of R (possibly the same as Y-DNA P) maybe in Bengal (or not too far away)".

    Yes.

    "T is also present in East Africa, almost like M1. They could well have journeyed together".

    I agree with that.

    "I'm still puzzled by the South Arabian J*, the J1/J2 split and the J/I one rather suggest a more northernly, Fertile Crescent origin".

    I also think that is very likely. Which gives rise to the problem of J* (or J3) on Socotra. The only real solution is that the J1/J2 split is no more than about 12,000 years old, about the time that J* reached Socotra.

    "What is curious is that instead of taking (mostly) South Asian mtDNA, this 'population' (??) took South Asian (or Persian Gulf Oasis) Y-DNA (notably IJ and G) in their push westward, and only very few South Asian mtDNA lineages (M leading to M1 and maybe other rare M subclades found in Arabia)".

    Implies very strongly that there were already mtDNA lineages in the region. They didn't have to bring in South Asian ones. And I am reasonably sure that IJ and G were already present in the 'more northernly, Fertile Crescent' region. But as you say G is not really involved in any Arabian colonisation.

    ReplyDelete
  7. Re-read your comments, Terry: they are highly substandard, sadly enough. You just seem to think that by insisting in your prejudices and preconceptions in repeated form, without backing such claims with even the slightest bit of evidence, you somehow make them real.

    Sadly for you, Reality is much more stubborn and cares little about what we can say about it. So we better try to describe reality for what it is not for what we wish it would be.

    I do not know the rest, but I am less interested on what you may subjectively think and more interested on what you may contribute (as in facts and well pondered theories). I do not care if Terry agrees or not with that particular bit, unless that is for a good reason - and in that case I would like to know the reason much more than the yea/nay "vote".

    "Implies very strongly that there were already mtDNA lineages in the region".

    That does not work because neither M nor N appear to have coalesced in West Asia but further East. In fact not even Y-DNA F looks like coalescing West of Pakistan, so G and IJ are probably back-migrants as well.

    But while the line of N (and therefore that of R, N1'5 and others, its descendants) appears to have been in SE Asia (where N coalesced probably), the line of F (and therefore that of G, IJ and other descendants, excepted MNOPS) never went that far East it seems.

    Yet in West Asia, we see not just descendants of myDNA N and Y-DNA MNOPS but also other Y-DNA lineages: IJ and G, which must have been picked up in Pakistan (as F* and IJK*) or in the Persian Gulf (as finished G and IJ).

    ReplyDelete
  8. Maju, I´m writing this from a ciber café. You ow me a laptop !

    ReplyDelete
  9. "That does not work because neither M nor N appear to have coalesced in West Asia but further East".

    Almost certainly wrong. The comment is an example of what 'you may subjectively think'.

    "In fact not even Y-DNA F looks like coalescing West of Pakistan, so G and IJ are probably back-migrants as well".

    I find that almost impossible. G is virtually absent in India, a region where a great variety of haplogroups survive. And any J in India are downstream clades, and almost certainly immigrants from further west. If G and IJ coalesced west of India they must have had some accompanying mtDNA or they would not have survived. So your comments are 'are highly substandard, sadly enough'.

    "in SE Asia (where N coalesced probably)"

    Unlikely. Simply another demonstration of 'your prejudices and preconceptions'.

    "the line of F (and therefore that of G, IJ and other descendants, excepted MNOPS) never went that far East it seems".

    I'd almost agree with that except that F(xKMNOPS) is found in the Philippines and in parts of China. Probably F2.

    "IJ and G, which must have been picked up in Pakistan (as F* and IJK*) or in the Persian Gulf (as finished G and IJ)".

    Why so? surely that is just another example of your 'insisting in your prejudices and preconceptions'.

    ReplyDelete
  10. I have discussed these matters (where M and N coalesced, etc.) in previous entries, here or in Leherensuge. Unlike you, I have made some previous work, which has been fully transparent and is available for review and discussion.

    In fact I have not reached to such conclusions because of any previous idea I might have got but because real data forced me to.

    FYI:
    http://leherensuge.blogspot.com/2009/04/mtdna-tree-version-11.html
    http://leherensuge.blogspot.com/2009/04/eurasian-mtdna-expansion.html
    http://leherensuge.blogspot.com/2010/02/reconstruction-of-mtdna-spread-in.html

    But you were present and actively debating when I posted all that so I'd think you would already know that.

    It is always the data which points to the likely origins and not "my subjective ideas" on the matter which do. My subjective ideas emanate from the data and I never before finding out had any bias or preconception about them.

    ReplyDelete
  11. "I find that almost impossible. G is virtually absent in India"...

    Notice that I did never say that G originated in South Asia. I understand that G coalesced in West Asia BUT is derived from F, which has a clear South Asian origin.

    So, for children: some F migrated westward where it coalesced into G.

    "If G and IJ coalesced west of India they must have had some accompanying mtDNA or they would not have survived"...

    Indeed and that is why I think that they migrated together with R1 (leading to R1b) and with mtDNA R (leading to R0 and U) as well as with N1'5 (leading to N1). R2'JT (leading to JT) and the other small N and M sublineages found in West Eurasia, may have also arrived back then or slightly later.

    ReplyDelete
  12. While I agree that Y-DNA T is likely a backmigration to Africa, just like mtDNA M1, the distribution is not right for them to have travelled together, and the distribution of Y-DNA T within India, which is very localized, is not consistent with a hg of great antiquity there. A more likely point of origin for Y-DNA T would be Mesopotamia.

    ReplyDelete
  13. I don't know why you say that Andrew: M1 is clearly a West Eurasian lineage that has penetrated Africa but is found as far East as Tibet (cf. González 2007). Obviously M1 must have migrated with some West or South Asian Y lineage, at least initially, and that one in Africa can only be T.

    As for the origin of T, I do not know of any work which dwells on the diversity (a critical factor to understand any haplogroup) but being most closely related to L, a Pakistani origin is within the logical possibility, if not for T at least for LT.

    ReplyDelete
  14. "But you were present and actively debating when I posted all that so I'd think you would already know that".

    I do already know what you think, and in spite of carefully following the arguments you have presented I'm still sure you are wrong.

    "I understand that G coalesced in West Asia BUT is derived from F, which has a clear South Asian origin".

    It is obvious that F, or its immediate ancestor came from Africa originally. So surely the default position is that some of its descendants should remain along the route, unless something major has obliterated them. When we see possible candidates precisely where we would expect to find them surely we should regard these haplogroups as lying along the route, rather than proposing all sorts of theories about 'private lineages' moving huge distances backwards and forwards around the earth. So G is as likely to have formed from F in SW Asia before F moved into India as it is to have formed from some sort of F that had first moved into India, and then out again. In fact far more likely to have done so.

    "that is why I think that they migrated together with R1 (leading to R1b) and with mtDNA R (leading to R0 and U) as well as with N1'5 (leading to N1)".

    Surely those mtDNA R haplogroups are as likely to have migrated through India into SW Asia with Y-haps P and R. And N1'5 has no ancestry east of Northern India, so probably remained somewhere in Northern India or SW Asia from when N first coalesced.

    "Obviously M1 must have migrated with some West or South Asian Y lineage, at least initially"

    Not necessarily. To me it seems most likely that M1 coalesced in SW Asia even though it 'is found as far East as Tibet'. M1 is a subclade of M1'20,51. The three clades are scattered from SW Asia across India to SE Asia, most likely along the route the haplogroup originally took. M1 may have entered Africa with Y-hap T but it's unlikely to have accompanied T all the way from India.

    ReplyDelete
  15. "I do already know what you think, and in spite of carefully following the arguments you have presented I'm still sure you are wrong".

    So please, now you go an in that nice blog of yours write one or several articles dismantling my theory (if you can). With one line comments here and there (and no graphics!) you are not going anywhere.

    Until you do that I will not consider your opinion to have any merit.

    "... the default position is that some of its descendants should remain along the route, unless something major has obliterated them".

    F as such did not coalesce before South Asia, so this idea is ridiculous. Similarly you can't claim that mtDNA N had descendants before it began its existence in (probably) SE Asia.

    First the parent, then the child. The opposite is not real.

    ReplyDelete
  16. "F as such did not coalesce before South Asia, so this idea is ridiculous".

    Do you actually have any evidence for such a claim?

    "Similarly you can't claim that mtDNA N had descendants before it began its existence in (probably) SE Asia".

    And similarly you have no evidence to support that claim.

    "First the parent, then the child".

    We cannot tell at present where the parent lived. We only know where the children live today. Presumably the children were not all born at the same time. They would have formed and been drifted to single haplogroups as the parent and some offspring haplogroups moved around. And we can be certain that parent, granparent of great grandparent had left Afica at some stage. And presumably spent some time near it when they first moved out.

    "So please, now you go an in that nice blog of yours write one or several articles dismantling my theory (if you can)".

    I can't be bothered and it is more fun reading your apoplectic reactions.

    ReplyDelete
  17. "Do you actually have any evidence for such a claim?"

    YES. We have discussed before, Mr. T.: F's basal diversity is highest in South Asia. Only G seems to be centered West of it.

    Can you keep tabs of our discussions in a better way, so I am not forced to repeat myself once and again?

    Same for N.

    "Presumably the children were not all born at the same time".

    It does not matter to me: they were all born in short notice before (or rather as) the parent mutated into something else. Whatever the case, the fleas' theorem (most fleas end up close to where you left them and not far away) allows us to discern in most cases, with some good approximation where root lineages existed: where most direct descendants are found today.

    "We cannot tell at present where the parent lived".

    Actually we can with high likelihood in most cases. Your negative attitude is just the same you can find among creationists who claim that nothing may be known with 100% certainty (maybe true) and therefore whatever I believe is true (almost for sure a false claim).

    Scientific relativism may shield you from science (somewhat) but won't make your subjective opinions more scientific nor credible.

    "And we can be certain that parent, granparent of great grandparent had left Afica at some stage".

    According to your negative statement we cannot: everything is fuzzy, blurry and unclear. A lineage restricted to Australia could have coalesced in America or Mars or Africa... because according to you no scientific analysis is possible of the extant data that might produce a certain result on where the ancestors of any given lineage lived. Everything is so blurry that in absence of documents (or even with them) we just CANNOT KNOW AT ALL.

    But I'm not of course accepting that conjecture of knowledge impossibility.

    "I can't be bothered"...

    You can't be bothered but you don't mind bothering me and my other readers with endless circular discussions. You cannot be bothered to write a single article but you don't mind coming here night after night (my time) to repeat your feeble nonsense, even in totally unrelated entries, making the comment sections of many of them totally impossible to read.

    Bother, bother yourself a bit. Because you have managed to bother me a lot.

    ReplyDelete
  18. "You can't be bothered but you don't mind bothering me and my other readers with endless circular discussions".

    Of course I don't mind bothering you. I am trying to help you understand what has actually happened during out history. What you presently believe owes a huge amount to your predjudices and preconceptions rather than to an examination of the actual evidence. As demonstrated by:

    "they were all born in short notice before (or rather as) the parent mutated into something else. Whatever the case, the fleas' theorem (most fleas end up close to where you left them and not far away) allows us to discern in most cases, with some good approximation where root lineages existed: where most direct descendants are found today".

    That is exactly the point I was trying to make during our discussion on E1b1b, but is the exact opposite of what you claimed during that discussion. How about you demonstrate some consistency instead of altering your perspective just to make the evidence fit what you already believe?

    "YES. We have discussed before, Mr. T.: F's basal diversity is highest in South Asia. Only G seems to be centered West of it".

    And, as I've many times tried to get you to understand, basal diversity can be the product of a number of factors. Place of origin being just one possibility. It is often, in fact probably more usually, a product of rapid population expansion.

    I'll grant it may be possible to make a case for F having originally coalesced in India, although not by any means a definitive case. But it is much more difficult to argue convincingly that its descendant haplogroup IT originated in India. You conveniently forget haplogroup IJ, presumably on purpose.

    To make a case that IT coalesced in India you have to claim that pre-F traveled from Africa to India as a private lineage. Then (once one of those F-descended lines had diversified into pre-IJ, pre-TL and MNOPS) just pre-IJ moved back west, yet again as a private lineage. And just MNOPS moved east out of India, presumably also as a private lineage.

    IJ, the first haplogroup to branch off pre-IT, is the most westerly of the haplogroups. A remarkable coincidence, do you think? Then TL branched off and looks very likely to have coalesced in India, especially in the northwest of the subcontinent. MNOPS, the last of the three, looks to have coalesced in SE Asia. Another remarkable coincidence?

    Rather than conclusively showing an Indian origin, the three haplogroups' distribution is far more likely to lie scattered along the ancestral haplogroup's route east from SW Asia.

    But that becomes a problem for your belief. If IJ coalesced in SW Asia its ancestor F must have lived there at some time. G's presence in the west shows that this is more than just a slight possibility.

    ReplyDelete
  19. "I am trying to help you"...

    I can't know for sure which are your real intentions but I'm sure you are not helping.

    "That [the 'fleas' theorem'] is exactly the point I was trying to make during our discussion on E1b1b, but is the exact opposite of what you claimed during that discussion".

    As you insist in this I can't but be persuaded that senility has taken over you and that you cannot discern anymore what I actually say and what you imagine I am saying.

    Hence all debate is pointless.

    "... basal diversity can be the product of a number of factors".

    So in one line you argue that you support the "fleas' theorem" and then (here) you claim it's worthless.

    It's totally pointless to debate with you: your inconsistence (and not just your stubbornness) is extreme. I'd get better reasonings from a wall.

    ReplyDelete
  20. "It's totally pointless to debate with you: your inconsistence (and not just your stubbornness) is extreme. I'd get better reasonings from a wall".

    Likewise. You seem to have no idea of practical genetics.

    ReplyDelete
  21. Easy fellows...

    @Maju: "I do what ?"

    You do have to read:

    http://johnhawks.net/node/15582, the post that inspired that post

    and also

    http://dienekes.blogspot.com/2011/06/stature-and-robusticity-during.html

    P.s. Sorry for my bad english.

    ReplyDelete
  22. If you think with Razib that present genetic variation has nothing or almost nothing to do with past genetic variation, then all discussion is pointless and all we can do is embrace Nirvana or something.

    Anyhow, Razib (who Hawks follows in this matter) may argue here that the Onge are a "recent arrival" to the Andaman (on what grounds?, I do not think so) but then he also argues elsewhere (or is that his friend Zack?) that the Onge represent the ASI (ancestral south Indian) component but in an imperfect way because they have been separated from Indians for soooo long that it's just like a weak echo. So what's the correct thing? I'd say that both are wrong, but that Zack is closer to the real thing.

    I find that Andamanese lineages arrived to destination very early on. Not the first ones to split but almost. But my approach to mtDNA molecular clock is iconoclast (and produces results that are later vindicated by archaeology, a sign of a good robust theory).

    According to my approach to the matter, M31 coalesced in Andaman in step 5, which should be c. 50 Ka ago. The other Andamanese lineage, M32a'b, coalesced in step 6, which is just a few thousand years later. So the Andamanese arrived to destination when the first West Eurasians were colonizing the Fertile Crescent, not before but not later.

    A different chronology attempt produced very similar results, even if I'm unhappy with its imperfection (it needs a logarithmic approach of some sort, I suspect): 49-46 Ka ago for these Andamanese lineages, and in any case contemporary of H (and H1, H2, H3, etc.), V, M1a, Z, A, X, U6, U2'3'4'7'8'9, and others.

    I just cannot agree with all that which is based (I guess) in the common MC methods, which are clearly wrong (several errors).

    ...

    I have posted a few hours ago on that same finding of Neolithic uselessness. I am, as you can imagine, in total disagreement with Dienekes' wishful thinking: Neolithic was a pain in many aspects, it still is highly problematic and the extreme lifespan extension of the 20th century is clearly not sustainable... unless a much more environmentally-friendly approach becomes the dominant paradigm. But this is a debate more for my other blog or at least for the relevant entry.

    ReplyDelete
  23. I just said: "If you think with Razib that present genetic variation has nothing or almost nothing to do with past genetic variation, then all discussion is pointless and all we can do is embrace Nirvana or something".

    I must add: ... or speculate wildly, that is where you end when you dismiss all available evidence as irrelevant and useless.

    ReplyDelete
  24. "your inconsistence (and not just your stubbornness) is extreme".

    You are being completely disingenuous in accusing me of 'inconsistency'. Although it is a fact that E1b1b's greatest basal diversity is inthe Upper Nile/Ethiopia region that is just one of many pieces of evidence I used to place the origin of E1b1b's expansion there. Just as it is necessary to look at more than just 'basal diversity' intrying to understand F's expansion. Context is everything.

    "Razib (who Hawks follows in this matter) may argue here that the Onge are a 'recent arrival' to the Andaman (on what grounds?, I do not think so)"

    Because you ignore the actual evidence, as usual.

    "the Onge represent the ASI (ancestral south Indian) component but in an imperfect way because they have been separated from Indians for soooo long that it's just like a weak echo. So what's the correct thing?"

    They are a mixture, just like everybody else on the planet. Although they display some Indian ancestry they also have ancestry from the nearby mainland, ie. Burma.

    "I find that Andamanese lineages arrived to destination very early on".

    You have absolutely no evidence to back up such a claim. Just your rather suspect coalescing steps. I'm sure that Razib and Hawks know more about the subject than you do.

    ReplyDelete
  25. "You have absolutely no evidence"...

    Molecular clock :p

    It is a theory and the evidence is the mtDNA sequences themselves, nothing else. We'd need to do some underwater archaeology to "prove" the matter in absolute terms. So the whole issue is, in practical terms, impossible to "prove" in either direction, so we better move on and stop speculating specifically about the Onge (what's the point other than being annoying?)

    My only point was to explain that while Razib, Hawks and the other speakers of the recentist school may think that, I disagree and I have good reasons to do it.

    ReplyDelete
  26. "Molecular clock :p It is a theory and the evidence is the mtDNA sequences themselves, nothing else".

    A theory that may or may not be correct. Both main mtDNA lineages are also found in India in closely related clades and they may have been more widespread in the past. For someone who has never hesitated to invoke 'private lineages' when the evidence fails to fit a particular theory you are remarkably reluctant to consider the possibility in the present case, where it is much more likely to have occurred than is so for many other examples you have claimed. A minority lineage (or two) from the coast could quite easily have reached the islands and later been drifted out on the mainland, representatives being left at either end of the distribution. Consider the Y-hap evidence. D can hardly be called an 'Indian' haplogroup. It is present through the hill country between South China and Tibet, and it would be a fair bet that it is common in Burma. So it presumably reached the Andamans from the Burmese coast, whether it's still present in Burma or not.

    "We'd need to do some underwater archaeology to 'prove' the matter in absolute terms".

    I disagree. If humans had reached the Andamans all that time ago they would not have confined themselves just to the coast. We would expect to see evidence all through the islands. We don't see such evidence.

    "My only point was to explain that while Razib, Hawks and the other speakers of the recentist school may think that, I disagree and I have good reasons to do it".

    The only 'good reasons' you have are your stubborn clinginging to the 'Great Southern Coastal Migration Theory'. There is actually far more evidence supporting the 'Razib, Hawks and the other speakers of the recentist school'.

    ReplyDelete
  27. Until recently we did not have evidence of colonization of Papua, since a few months ago (but not before) we have and know for a fact that people was there since c. 50 Ka. ago.

    So I'd like to know how hard and intense has the land archaeology of Andaman been because in all that part of the world our archaeological ignorance is much greater than our knowledge.

    We have evidence of people crossing relatively large bodies of water in that region soon after 50 Ka ago, so there's no particular reason why they would not have arrived to Andaman by about the same time, what is exactly what my "molecular clock" theory finds as well.

    ReplyDelete

Please, be reasonably respectful when making comments. I do not tolerate in particular sexism, racism nor homophobia. Personal attacks, manipulation and trolling are also very much unwelcome here.The author reserves the right to delete any abusive comment.

Preliminary comment moderation is... ON (your comment may take some time, maybe days or weeks to appear).