Studies conducted on nuclear DNA shows there is on average less difference between any two living humans than there is within most other species. This presumably shows we have a more recent common ancestry than do most other species. How long ago and in what form is a matter of debate. Beliefs range between just one couple created as recently as 6000 years ago to a population of reasonable size as long ago as two million years.
Research on the human Y-chromosome suggests the male common ancestry dates back to a single Y-chromosome Adam who lived in Africa more recently than 140,000 years ago (Jobling et al 2004). The Y-chromosome types found outside Africa today all diversified even more recently at just 89,000 to 35,000 years ago (Ke et al 2001).
Evidence from the study of mitochondrial DNA indicates we all descend from a woman who also lived in Africa some time before 100,000 years ago. We could refer to this woman as mtEve. The defence mentioned in “First Humans” that she was probably in no way different to her brothers, sisters, parents, cousins and neighbours though. The three people who originally constructed the mtDNA tree (Rebecca Cann, Mark Stoneking and Allan Wilson, a New Zealander) gave the date of their mtEve common ancestor as 290,000 to 140,000 years. So there is a margin of error. But she could have lived twice as long ago as Y-chromosome Adam did.
In “Pedigrees” [Selection] the jury saw that the survival of a single female’s mitochondrial DNA line may be due to no more than the loss of other lines around at the same time. In spite of the popular perception, therefore, the existence of this single ancestor doesn’t necessarily prove we are all descended from just her and her immediate family.
Perhaps there was some selection for her line. But any advantage her descendants had is unlikely to have been simply because of their mitochondrial DNA. Perhaps there was some selection for her genes, her nuclear DNA. There is fossil evidence of fairly modern-looking humans in Africa from more than 100,000 years ago (Olson 2002) and we’ll come back to this in “Out of Africa”. But there is no sudden change in human fossils we can use to follow mtEve’s descendants, and only her descendants, in any expansion around the world. Of course, as we hear at times, “absence of evidence is not evidence of absence” and we may just not have identified the genes yet. Anyway I pointed out in “Pedigrees” [Ancestry] her daughters would have had only half her nuclear DNA, her granddaughters quarter of it and so on. And the jury has also seen that mtDNA can be fairly independent of both Y-chromosome and nuclear DNA. We’ll look at “Culture” in Part V but I mentioned in “Polynesian Origins” [Societies] it is mainly women who transmit culture from generation to generation. Only women transmit mtDNA. I suggest any advantage mtEve’s line may have had is much more likely to have been through a stage in the evolution of culture rather than being genetic.
Agriculture’s development is accepted as being responsible for the modern distribution of many Y-chromosome and mtDNA lines (Jobling et al 2004). Surely similar selection through cultural advantage would be true for their earlier history? As far back as “Conception” [Me] I pointed out that culture, including language, is able to spread independently of gene flow. The defence will explain shortly [Interpretation] how an expanding mtDNA line need not necessarily be closely correlated with a more general genetic expansion.
What cultural advantage might mtEve’s descendants have had? It could have been any one of a number of things; social behaviour or organisation, use of fire, improved language or hunting techniques, religion, the invention of music. But in most regions there is in fact no sudden change in culture or technology we can use to identify her descendants’ earliest presence either. The jury will see that the big change in culture with the development of the Upper Paleolithic, about 40,000 years ago, is actually much too recent to fit the mtDNA evidence. Besides, as the defence said in “Technology”, the Upper Paleolithic was very complex in its development and didn’t spring up overnight fully formed in a single group of people. Any evolutionary advantage mtEve’s line had culturally is almost certain to have been slight. We’ll look at what this advantage may have been in Part V (“Culture” [Families]) after we’ve examined evidence from our early evolution and expansion.
The Trees
The following diagrams are based on Jobling et al (2004), along with updated versions freely available on the Internet (and see Behar et al 2008). The splits in the lines show regional variation, just like the ducks we met in “Species” [Kinds]. The jury will see there is no need to assume only members of these lines, or even only descendants of mtEve or Y-chromosome Adam moved. We can be sure that parent lines broke into offspring lines after they’d spread out. And many parent lines moved on with one or more of their offspring lines. Offspring lines have replaced parent lines in some regions. Lines have become extinct.
You will see as we continue with our family history that all branches in the trees make sense if we take it that mtEve lived around 190,000 years ago and Y-chromosome Adam lived about 70,000 years ago. The branches then fit with the periods of cooling and lowered sea level that have occurred at various times over the last 200,000 years. Of course we now have a gap of more than 100,000 years between mtEve and Y-chromosome Adam. This raises the question of what people had mtEve’s line been breeding with to sustain her mtDNA line all those years?
The Branches
The Sahara’s aridity has several times been extreme enough to eliminate human populations from the whole North African coast except for what is now Morocco. This part of Africa has been wet enough for human occupation to be continuous for the last one and a half million years. A drying Sahara has almost certainly actually pushed various migrant waves out of Africa several times during our evolution. In fact Stringer and McKie (1996) mention research that suggests the drying Sahara separated Neanderthals from mtEve’s ancestors about 400,000 years ago. They also write that by about 150,000 years ago the Sahara Desert had once more dried and expanded, separating humans in North Africa from those in Southern Africa, again just like the ducks. The Kalahari Desert in Southwest Africa also expanded. A date of 190,000 years ago for mtEve coincides with the climate cooling beginning around 185,000 years ago that led to this drying (“Neanderthals et al" [Climate]).
Members of the first branch on the mtDNA tree (L0) are concentrated in the southern half of Africa (Jobling et al 2004, and Behar et al 2008). The defence has said many times that selection works most effectively on small isolated populations. It is reasonable to suppose, therefore, that selection for any new culture acted on a population isolated for a long time in a relatively small region. Whatever the new culture was it helped the population survive (see “Culture” [Families]). The mtDNA evidence can be interpreted as showing that by 150,000 years ago a change in culture had spread through the South and East African point of the human star from the southern end.
The next splits in the mtEve line (L1, L5, L2, L6 and L4) seem to simply indicate a slow female expansion from East Africa. So with the global warming and presumably increased precipitation beginning about 130,000 years ago the inbred group with the new culture had expanded north. They interacted with other humans. The cultural innovation and genes spread around Africa. By 90,000 years ago modern-looking humans even began to appear outside Africa (“Out of Africa” [The Middle East]).
The haplogroup diagrams show that perhaps around that time mtEve’s line forked into yet more branches (L3, M and N). In spite of any propaganda M and N didn’t come out of Africa themselves. The lines were almost certainly selected for outside Africa. Their ancestors had come out sometime after the mutation that gave rise to L4, perhaps as long ago as 85,000 years. As I said, modern humans had certainly made it into the Middle East by 90,000 years ago. Neanderthals replaced them there about 70,000 years ago but it’s entirely possible Neanderthals absorbed some of their mtDNA lines. Like chimpanzees and gorillas, females may have been kidnapped or wandered off to join neighbouring groups (“Polynesian Origins” [Societies]). Lines M and N may simply have then been separated by the first obstacle to their expansion, the Zagros and Taurus Mountains (“The Human Star” [Geography]). The Himalayas separated them even more, sending N across the Central Asian plains and M into India. By then Y-chromosome Adam had probably appeared, but before then mtEve’s descendants must have had children with other men as her line had spread. A few of M and N’s descendants moved back into Africa.
After an initial expansion a series of mutations in the M and N mtDNA lines gave rise to regional varieties throughout the world, especially in the east. Line N reached Australia, probably more than 50,000 years ago. A later migration brought M’s line, and N’s descendant line P, to New Guinea (“Into Australia” [Explanation]). Others of M’s descendants reached Central Asia and eventually moved to America (C and D). N’s descendants A and B also eventually reached America, either with or after M’s descendants C and D. We’ll return to America in “North to Alaska”. In Southeast Asia mtDNA line R evolved from mtDNA N. Her descendants, and R herself, spread north and west from there. One of these lines gave rise to H, the most common mtDNA line in Europe.
Now a look at the Y-chromosome line. The first fork in the Y-chromosome Adam line seems to be when a group moved out from the South and East African point of the human star, perhaps a little more than 60,000 years ago. The new Y-chromosome line, B, is especially common in modern day Pygmies. From here on we begin to find non-African Y-chromosome lines.
Line B then split into two: one gave rise to D and E, and the other led to C and F. The expansion of Y-chromosome lines D and E must be more recent than the YAP Y-chromosome mutation that defines them, most likely by 50,000 years ago. Lines D and E were possibly simply separated by D’s departure from Africa, or E’s return. D moved across Central Asia, the middle of the human star, and reached the East Asian point. E expanded around Africa in force. The evidence suggests that there E replaced many A and B Y-chromosome lines.
Meanwhile, at the Asian end of the cline, it seems that, like the mtDNA lines, the Himalayas had separated the Y-chromosome lines, C and F. Line C seems to have originated on the Iranian plateau, or further north. As it expanded a series of mutations again gave rise to regional varieties. Men with Y-chromosome C moved through Central and Eastern Asia, as well as into Northern India. The line is the most common Australian Aboriginal Y-chromosome.
Y-chromosome line F may also have originated around the Iranian Plateau, or in India. Some people believe a volcanic eruption had earlier emptied this subpoint of the human star. F may have moved into India with mtDNA line M. It seems that some of F’s descendants (G, I and J) replaced their own parent line (F) outside India. On the other hand they may have moved back north from there. Y-chromosome H probably evolved in India. Another of F’s descendants (K) seems to have moved east across India. This Y-chromosome K, along with mtDNA line M, made it to New Guinea from Southeast Asia. Y-chromosome K also made it to already occupied Australia.
There seems to have then been an expansion of Y-chromosome K and his descendants, perhaps along the coast and rivers, north to the East Asian point (N and O) and west back through the Indian subpoint, onto the Iranian plateau, and even as far as Africa (L, P and T). This expansion, along with many of the mtDNA ones, was probably associated with the improved boating technology necessary to reach the Australian point (“Into Australia” [Wallace’s Line]). The jury will see (“North to Alaska” [Genetic Evidence]) that P’s descendants were eventually able to expand north beyond 50°. They spread across the whole region from the American subpoint (Q) to the Northwest European point (R). Y-chromosome R also moved back south and spread through Africa, India and Australia.
Interpretation
There are a number of ways to interpret all this evidence. The jury saw that evidence for an African home for both mtEve and Y-chromosome Adam is provided by the fact that Africa has the greatest depth and diversity of lines. We find only more recent lines outside Africa. The main disagreement is between those who say that these modern humans replaced all other human types when they came out of Africa in a single migration, and those who believe modern human variations show a regional continuity going back more than 200,000 years. The evidence is perhaps capable of supporting either hypothesis (Smith et al 2005).
The fossil and genetic evidence is usually interpreted as showing that at some time more recently than 100,000 years ago (perhaps even 50,000) modern humans, presumably descendants of Y-chromosome Adam and mtEve, left Africa together. They are said to have then spread around the rest of the earth in a single wave of migration, rapidly displacing all earlier human populations. Supporters of this single origin theory claim no hybrids were formed between people from this expansion and other contemporary human inhabitants of the earth. However there appear to be many continuities of population difficult to explain away.
Richard Klein (1989) in “The Human Career” quotes a major objection to the single origin theory. Milford Wolpoff has pointed out that in most regions the very earliest modern humans look more like the present inhabitants of those regions rather than like any supposed common African ancestor. In other words regional variation exists right from the beginning of modern human development rather than there being a common ancestral type that later evolves into the regional variations we see today. Perhaps this is a result of “founder effect” (“Hybrid Vigour and Inbreeding” [Survival]). But in many cases the present inhabitants in each region also look a fair bit like the people who were there even before the modern humans. Of course this has been explained away as being parallel evolution (“Species” [Labels]) but is this likely to account for all the continuities? Chris Stringer and Clive Gamble (1993) even admit that, unlike the spread origin supporters, the single origin “camp has not so far produced a comparable theoretical dogma to account for evolutionary change”.
Another problem for the idea of a single sudden migration out of Africa is that although the mtDNA and Y-chromosome lines both come from Africa the expansions seem to be fairly independent of each other, they even originate as much as 100,000 years apart (Jobling et al 2004).
So how could various genes, including Y-chromosome and mtDNA lines, move independently through a population? The jury saw that the mutation that gave rise to Y-chromosome line C, for example, probably happened on the Iranian plateau. But the mutation must have occurred in a single individual. He must have had male children but they would have only half his genes, or nuclear DNA. His grandsons in turn would have on average just one quarter of his genes, his great-grandsons one eighth, and so on (“Pedigrees” [Ancestry]). If he had children with close relations the children would have shared many genes. But suppose members of this Y-chromosome line developed some new technology and moved a short distance away. He and his sons would then have had children with women who were more different. The original ancestor’s Y-chromosome would survive but his other genes would rapidly become diluted. We know Y-chromosome C’s descendants eventually reached Australia.
We can presume that neither the individual with the original Y-chromosome mutation, nor even his grandchildren, took part in the last stages of that movement though. It would have been a slow migration across Asia, picking up resident genes as it moved. Therefore, in effect, we can say the Y-chromosome itself moved independently through our species’ nuclear DNA or genes. Resident elements of these genes could easlily remain in spite of the expansion of Y-chromosome C. The same thing with mtDNA. In fact with any individual gene (“Hybrid Vigour and Inbreeding” [Wave theory of Evolution]). But mtDNA is replaced less readily than Y-chromosome; therefore, as you saw, mtDNA’s root is more ancient than the Y-chromosome root. Although the root of both is in Africa there is no reason at all why they must both go back to just a single population. And of course it is quite possible that either or both Y-chromosome Adam and mtEve descend from earlier migrations into Africa. There is no reason at all why they must descend from some branch of Homo erectus that had never left Africa.
If at times men carrying any particular Y-chromosome are able to move into an unoccupied region, or at least an environment that is not being exploited (“Species” [Ecology]), they must have some accompanying mtDNA lines. Otherwise there will be no descendants. However the jury saw in Part II (“Polynesian Origins” [Genes] and “Pacific Population” [Mixing]) that a Y-chromosome expansion is quite capable of picking up mtDNA lines not previously associated with it.
The reasons why so many people are so ready to accept a complete genetic replacement are many. The main reason is that it closely fits widespread cultural beliefs of how species, including the human species, originate. It is also easy to explain, as it is simple, much simpler than reality. And it gives us a defined point of origin, a beginning we can easily understand. It is possibly also influenced by some people’s attitude to the desirability or otherwise of maintaining the “purity” of their own racial or religious group.
These reasons are all based on assumptions, preconceptions and prejudices, Chinese drover’s clever dog syndrome. Unfortunately any simple single origin theory requires selectively ignoring much relevant evidence. The existence of a mtEve and a Y-chromosome Adam merely indicates that both these lines reduce to separate single male and female African lines dated between 50,000 and 200,000 years ago, more recently than the original Homo erectus expansion.
Tests on mtDNA in Neanderthal remains show that they too separated from modern human lines more recently than the original Homo erectus expansion, this time around 500,000 years ago (Wade 2001). According to the literal interpretation of mitochondrial DNA evidence, as used by single origin supporters, some Africans have been developing separately from Europeans for nearly half that long but these two groups have no trouble at all producing fertile hybrids. Culture has tended to separate them to some extent though. To a varying extent culture has also separated Europeans from the many different people they have encountered as they made their way around the world. Perhaps culture was all that separated Neanderthals from modern humans.
Mungo Man
Mitochondrial DNA evidence from prehistoric skeletons in Australia suggests that at least one of the earliest humans there was not a member of mtEve’s line. It parted from her line about 250,000 years ago but was totally modern looking, if not ultra-modern looking (Adcock et al 2001). The skeleton is dated to at least 40,000 years ago and possibly 60,000 years. We’ll assume this Australian skeleton represents the last survivor of modern human mtDNA lines contemporary with mtEve. If mtEve lived even just 150,000 years ago it took roughly 100,000 years, or 5000 generations, for the lines to be reduced to one. If mitochondrial lines are reduced at the rate of one per generation (“Pedigrees” [Populations]) this means there were at least 5000 females in the original mtEve population, i.e. well over 10,000 people. In fact by looking at chromosomes and DNA Dr. Sarah Tishkoff and her colleagues have suggested the original population involved was in fact at least twice this large and possibly consisted of up to half a million individuals (Wade 2001). It would have been a widespread population.
Some single origin supporters may believe there was indeed a mass exodus of more than 10,000 people from Africa some time more recently than 100,000 years ago. Led by a Moses, perhaps? I would bet the population was fairly well spread out and a great deal of gene flow occurred.
Most of us accept gene flow has occurred since mtEve lived (see for example Olson 2002) but some are sceptical when others suggest it was going on long before her time. The defence suggests that the distribution of the mtDNA and Y-chromosome lines even supports the conception modern humans are a product of gene flow backwards and forwards throughout the world. At times cultural and technological selection for various mtEve and Y-chromosome Adam lines has resulted in something between the single origin and spread origin scenarios. Groups often mix with locals as they spread out. In Part II the jury saw that this sort of mixing and mingling is common. In Part V “Conquest” the jury will see other examples that support the interpretation of the evidence the defence has offered (“North to Alaska” [Genetic Evidence] for example).
Both mtDNA and Y-chromosome research indicates that worldwide these each have relatively recent common origins. But this doesn’t eliminate the possibility nuclear DNA from earlier populations survives. Besides, scientists have discovered human nuclear DNA that shows a common origin much earlier in human history (Jobling et al 2004). And studies of X-chromosomes, which are mixed genetically in daughters but not in males as they have only one of them, give age estimates of between half a million and nearly two million years for our common origin (Ke et al 2001). This fact is confusing for single origin supporters but it makes sense to others of us. The defence mentioned in “Technology” [Progress] that each gene travels on its own wave, it has its own evolutionary history. Perhaps mtDNA and Y-chromosome should simply be regarded as just two more separate human genes that have moved through our species.
John Wilford (Wade 2001) has written, “Though the extent of mixing between the modern and Neanderthal populations is not known, Dr. Trinkaus said that the disappearance of Neanderthals could be attributed in part to interbreeding as well as competition”. Just like the grey duck in New Zealand (“Species” [Difference]). In other words perhaps mainly just their Y-chromosome and mtDNA lines have been replaced (“Technology” [Progress]).
Witnesses Called
Adcock et al (2001) Mitochondrial DNA Sequences in Ancient Australians. Proc. Natl. Acad. Sci. Vol. 98 pp. 537-542.
Behar et al (2008) The Dawn of Human Mitochondrial Diversity. Am. J. Hum. Genet. 82: 1-11
Jobling et al (2004) Human Evolutionary Genetics. Garland Science, New York.
Ke et al (2001) African Origin of Modern Humans in East Asia. Science Vol. 292 1151-1152
Klein, Richard G. (1989) The Human Career. University of Chicago Press, Chicago.
Olson, Steve (2002) Mapping Human History. Houghton Mifflin Company, New York.
Smith et al (2005) The Assimilation Model, Modern Human Origins in Europe, and the Extinction of the Neandertals. Quaternary Intern. 137, 7-19.
Stringer, Christopher and Gamble, Clive (1993) In Search of the Neanderthals. Thames
and Hudson, Great Britain.
Stringer, Christopher and McKie, Robin (1996) African Exodus. Random House, UK.
Wade, Nicholas ed. (2001) The New York Times Book of Fossils and Evolution. The Lyon Press, New York.
N.B. this post is a revised version of the first essay on this subject, which appeared on January 10th, 2008
51 comments:
Let's see if I can synthetize my criticisms in few paragaphs (it's a very long post).
1. Genetic age estimates are just that: educated guesses. The variables involved are not well known: effective population size is nothing more than an speculation and mutation rate is based in other estimates of the divergence between the Pan and Homo branches that are probably underestimates. They can and will surely be reviewed in the future.
2. The genetic study of Neanderthals so far has yielded repeated negatives in their hypothesized admixture with modern humans. Their mtDNA is widely different an so is their red hair gene. Cro-Magnon instead is inside modern human and European variation.
Additionally there are no remains that suggest hybridation (except the controversial Lagar Velho kid), Neanderthals do not look at all like modern Europeans or West Eurasians and all Eurasian and Australian ancient human remains are within the timeframe of the OOA model. Still they often do not look as close to their modern descendants as you claim, but much more than Neanderthals for sure.
3. Some of the data you manage is clearly obsolete. Citing works of 1993 in the rapidly evolving field of genetics is like quoting Archimedes in relation with nuclear fussion. C is likely to have migrated via SE Asia (from South Asia, that looks like the main hub of human dispersal in Eurasia) along with other (F and D) clades. It's not sure but very possible that all Sahulian (Australia-New Guinea) lineages arrived there at the same time and that even migrated together. Same for other locations maybe. Equally M and N do not probably represent two different expansions but just one of plural lineages. South Asia is the most likely center of dispersal for all Eurasian lineages.
4. "Chimpanzees or gorillas" are surely not the best proxy for human behaviour. Why do you ignore bonobos or our own modern paleolithics like the !Kung (Bushmen)? Anyhow chimps and gorillas do not kidnap females, at least not normally.
5. The multirregional model is out, sorry. It's not impossible that there was some occasional admixture and the subsequent introgression but overall whatever archaic Homo species that were around when our ancestors arrived have not left any visible genetic (nor morphologic) legacy on us.
6. There is no major problem with Y-DNA and mtDNA having different patterns of dispersal. The founders of the different lineages probably never met at all. We are talking thousands of generations here and maybe multiple small groups crossing their paths to spread out again, founder effects, drift, gender bias... all that must be taken into account. The result is pretty chaotic at times.
7. Adcock's LM3 "evidence" has been criticised for lacking independent control testing and doubts about the likehood of aDNA preservation in such enviroment (among other criticisms that basically suggest it's a stunt). Further testing may not be possible due to those silly laws that assimilate paleolithic remains to aborigine ones. All other aDNA evidence around the world is in good correlation with the OOA model.
8. X chromosome studies are still in their infancy. Still the so much publicited odd presumably old clade shared by Basques and Native Americans is also present in Africa - in small apportions but the continent has not been well sampled yet anyhow. Anyhow a prudent wait-and-see attitude is better than wishful thinking in these cases. It's interesting stuff but it's much like those 80s and 90s papers that nowadays are obsolete for the most part. Only further research will tell if there is something in that speculation or just lack of sufficient knowledge.
In brief: not so fast!
First of all thanks for taking the time to post your comments. I'll try to cover them in order. But perhaps before then I should concede my acceptance of hybridisation between modern humans and Neanderthals stems partly from an unwillingness to believe genocide has a long well established place in our evolution.
1) I agree that "Genetic age estimates are just that" but I suspect the Pan/Homo split is now fairly well dated. Any adjustment will be relatively minor.
2) Regarding genes for complexion and skin colour. This link is relevant:
http://anthropology.net/2008/05/19/a-new-genome-wide-association-study-pinpoints-more-human-skin-color-alleles/
There is no simple, single gene involved. And regarding introgression:
http://johnhawks.net/weblog/reviews/neandertals/neandertal_dna/coop-foxp2-recent-selection-2008.html
Introgression must have happened in one direction or the other.
3) Data from 1993 has not been overturned, just expanded. Therfore completely unlike "quoting Archimedes in relation with nuclear fussion". I'll return to the rest of your comments here in number 6.
4) I include bonobos with chimpanzees. They certainly form fertile hybrids with the other three subspecies if given half a chance. And are you saying "modern paleolithics like the !Kung (Bushmen)" are not humans?
5) I think the jury is still out on that in spite of many peoples' cultural beliefs.
6) "There is no major problem with Y-DNA and mtDNA having different patterns of dispersal". But in number 3 you wrote, "It's ... very possible that all Sahulian ... lineages arrived there at the same time and even migrated together. Same for other locations maybe. Equally M and N do not probably represent two different expansions but just one of plural lineages". Now that's shifting positions more than just a little! Why the different interpretation of the evidence?
7) I haven't heard that criticism and it may be justified. Can you provide a link?
8) "Only further research will tell if there is something in that speculation or just lack of sufficient knowledge". Agree totally, but it's fun to speculate. Again I'll say I really appreciate your comments. Thanks.
Glad you appreciate them. :)
my acceptance of hybridisation between modern humans and Neanderthals stems partly from an unwillingness to believe genocide has a long well established place in our evolution.
You don't really need genocide to explain it, nor minor hybridation would exclude it.
One reasonable possibility is that H. sapiens eventually just became more fit in procuring resources and multiplied a lot faster in about the same territory. I think there's some indications suggesting this pattern of better use of all type of resources, including much larger hunting territories (Gamble) and the explotation of river/marine resources, that were not among Neanderthal preferences.
Whatever the case, many hunter-gatherer groups can certainly be violently hostile to strangers and displace them easily, cornering them in the worse niches. If H. sapiens displaced Neanderthals that way from most of Europe, North Africa and West and Central Asia, not with any genocidal aim but just inside a normal hunter-gatherer behaviour, reducing the area and population of Neanderthals and their genetic variability that way, Neanderthal extinction was just a matter of time. And that is what precisely the arhaeological record seems to suggest.
Take also in account possible epidemics brought by the intruders (our ancestors).
Unless you think with Zilhao that Neanders were already out when H. sapiens arrived, whatever the causes. In any case it doesn't seem a premeditated genocide but just the usual "healthy" competition and the survival of the fittest (taking many milennia in any case).
Regarding genes for complexion and skin colour. (...)
There is no simple, single gene involved.
I only mentioned red hair, that is quite clearly defined by MCR1. The fact is that some Neanderthal aDNA showed up as causing red hair but the code was totally different than that of modern humans. Convergent evolution: different genes for the same function (or lack of it).
Introgression must have happened in one direction or the other.
It's not clearly demonstrated. The evidence for and against seems to add up without any clear conclussion.
But, even if there was introgression, it doesn't mean large scale admixture. Introgression precisely means transfer from one species (or subspecies) to another of an adaptative gene without significative admixture (often mediated by a small hybrid population). It is that adaptativeness of the gene what makes it expand, independently of the degree of admixture. Almost no admixture does not contradict the introgression of one or more adaptative genes.
Introgression does not prove massive hybridation, it actually means the opposite.
Data from 1993 has not been overturned, just expanded.
Brutally expanded. The data is not overturned but the conclussions often are.
I include bonobos with chimpanzees.
They are a different species and specially their social behaviour radically differs from that of chimpanzees (matrilocal and rather matriarchal, do not fight with other groups, low levels of violence, clear compassion - like humans but not chimps).
When discussing the instinctive fundaments of human (and Pan sp.) behaviour one cannot "include bonobos with chimpanzees", as they are nearly the opposite extremes.
And are you saying "modern paleolithics like the !Kung (Bushmen)" are not humans?
Obviously not. I am saying that their behaviour and social structure is likely to represent that of other hunter-gatherer populations of the past much better than chimps.
Of course, humans are strongly cultural and hence it may not be 100% representative. But a much closer example than chimps for sure in any case.
"There is no major problem with Y-DNA and mtDNA having different patterns of dispersal". But in number 3 you wrote, "It's ... very possible that all Sahulian ... lineages arrived there at the same time and even migrated together. Same for other locations maybe. Equally M and N do not probably represent two different expansions but just one of plural lineages". Now that's shifting positions more than just a little! Why the different interpretation of the evidence?
I don't think it's contradictory. If my interpretation is correct, the ancestors of modern Australian and New Guinean aborigines arrived, maybe in different subwaves from SE Asia (and ultimately from South Asia) in a single wave that happened after the differentiation of macrohaplogroup K and C (Y-DNA) and that of N and M (mtDNA). Several of these founder lineages became fixated on the road or upon arrival, and then got internally differentiated somewhat.
Instead it's difficult to explain so simply the genetics of Europe for instance, where some mtDNA clades (U5, U8a) seem as old as Aurignacian and others as recent as Neolithic (with H and V being possibly of Gravettian origin and re-expanded with Magdalenian maybe). Y-DNA doesn't seem to correlate well with them in any case, specially haplogroup P (including Q, R1a, R1b and R2). Haplogrpoup P is, like many Sahulian (specially Melanesian) clades, an offshot of K but its distribution is wildly ubiquitous, correlating wih East Asian mtDNA clades in Siberia and America (Q), with South Asian mtDNA clades in the Indian subcontinent (R1a, R2) and with West Eurasian (West Asian and extensions) in Europe (R1b, R1a). So in this case there is some apparent contradiction between Y-DNA and mtDNA but there is also quite clear archaeological and (to lesser extent) linguistic evidence supporting many different migrations, some of which could have been almost exclusively male-mediated (case of Indo-Europeans with all likehood).
They are two different cases clearly. Now, if you can provide archaeological evidence supporting second migrations into (specially) New Guinea, that could represent the K wave, then I'd be willing to reconsider. But so far the impression I have is that Sahul was colonized only once early on, while Europe withstood at least four different immigration epysodes (Aurignacian, Gravettian, Neolithic and Indo-Europeans) plus one or more internal re-expansions (Magdalenian clearly, maybe in Epipaleolithic and Neolithic too - and also the Bronze/Iron age warrior migrations of mostly Indo-European character).
Very different cases: an isolated corner vs. a much better connected one that was repeatedly the goal of new migrations.
I haven't heard that criticism and it may be justified. Can you provide a link?
http://www-personal.une.edu.au/~pbrown3/LM3.html
I think it's a good read. It doesn't mean that Adcock was necesarily wrong but his study looks rather feeble, specially considering the huge problems that aDNA studies have (contamination, degradation...).
"Only further research will tell if there is something in that speculation or just lack of sufficient knowledge". Agree totally, but it's fun to speculate.
For sure. But caution is strongly advised.
What we know by now of X-chr is a very limited ammount of information and only at the level of haplotypes. As you may know haplotypes can be coincident between only far related humans and only those haplotypes inside a well documented SNP-defined clade (haplogroup) are likely to mean close relatedness. This is somethng you can read in the disclaimer or FAQ of any serious comercial DNA testing site.
Enjoy.
Thanks again Maju. My first point would be that genocide doesn't necessarily need to be planned. I'd doubt if many early American settlers for example actually planned the genocide of the indigenous inhabitants.
Regarding evidence for separate movements to Sahul and Australia check out Kow Swamp fossils. The fashion these days is to downplay the similarity between them and SE Asian H. erectus. Doesn't fit with recent out of Africa? I've already commented on this:
http://remotecentral.blogspot.com/search/label/Human%20Evolution%20on%20Trial%20-%20Into%20Australia%20-%20by%20Terry%20Toohill
Regarding the distinction you make between European haplogroups and worldwide ones you may find it interesting to look at the maps Tim and I included at the end of the first version of "MtEve". The new evidence fits exactly the pattern presented there. The only thing I would change is the numbering of the maps, although I could improve mtDNA line L;
http://remotecentral.blogspot.com/search/label/Human%20Evolution%20on%20Trial%20-%20Mitochondrial%20Eve%201st%20edit
I'd now start with map 5. It certainly provides evidence for an mtDNA expansion from India. But those lines had all evolved from an earlier line, R.
Map 2. N's expansion across Central Asia: The first branches, I and W, separated by the Zagros, Taurus or Caucasus mountains.
Map 1 provides evidence for the expansion of an early cline of Y-chromosome lines: C, D, E and F. I'll concede that mtDNA line N may have accompanied Y-chromosome lines C and D as they moved across Central Asia, but not necessariliy.
Y-chromosome line F moved into India, possibly but again not necessarily, with mtDNA line M: map 3.
I'll also concede that Y-chromosome K's expansion from SE Asia, map 4, may be associated with some mtDNA expansions demonstrated in map 3 and; back to map 5, the most recent of the basal mtDNA lines. At the probable time of this expansion, say 40,000 years, Homo erectus still survived in Java.
You can understand how these haplogroups could basically move independently through a population but not necessarily indicate a more extensive genetic movement. Indigenous genes could remain in many regions.
My first point would be that genocide doesn't necessarily need to be planned. I'd doubt if many early American settlers for example actually planned the genocide of the indigenous inhabitants.
Probably. In any case the situations were completely different. The replacement of Neanderthals by H. sapiens was a process that in any case must have taken one, two or several thousand years - only in what respects to the Franco-Cantabrian region, where we know Neanders tried to adapt by creating the Chatelperronian UP techno-culture, and where there is a very heated argument between those who support local alternation between both species (Mellars) and those who radically reject it (Zilhao). The final extinction in refuges of southern/western Iberia or North Africa may have taken even much longer.
The North American genocide was mostly a matter of one century and was propelled by industralization and incredibly high rates of overseas immigration, something plainly impossible in UP Europe. Still we humans are sometimes intentionately genocidal and the deportation of the Cherokee for instance (who were trying hard to adapt to Western culture) is a good example.
But maybe the basic element is the same: expropiation of the territory by the newcomers. Even if circumstances and timelines are so radically different.
Regarding evidence for separate movements to Sahul and Australia check out Kow Swamp fossils. The fashion these days is to downplay the similarity between them and SE Asian H. erectus. Doesn't fit with recent out of Africa?
In the link I posted yesterday, there are many pictures of H. erectus and archaic H. sapiens. The erectus of Asia look nothing like modern humans certainly (much more simiesque with much lower vaults and all that).
I've already commented on this...
Yeah. Long post though (lazy me!). I wonder where you got that strange genetic map from. Never seen anything like that anywhere.
Regarding the distinction you make between European haplogroups and worldwide ones
Not worldwide ones but specifically Sahulian ones. Each regional situation may require its own asessment. I just used the rather well studied European example as a very distinct counterexample to Sahul, at least apparently.
...you may find it interesting to look at the maps Tim and I included at the end of the first version of "MtEve". The new evidence fits exactly the pattern presented there.
Nice maps, even if I may not agree with them in many aspects. I admit I have also in the past speculated about Y-DNA K being of SE Asian origin but on second thought, rather not.
Of the 10 K subclades 5 are found in Oceania (and that is an argument for a SE Asian origin of K) but three are found (or believed to have split at, case of P) west of India. Of the other two, one is not reported any locality (K4 in the 2008 draft tree) and the other one is NO, which has some reasons to be believed of northern East Asian origin (split of N and O, presence of NO*).
A North Indian urheimat for K actually seems to explain better the actual distribution of all K subclades. If so, it is prefectly integrable with the out-of-South-Asia model of Eurasian dispersal.
For the rest, everything seems to point to South Asia:
Y-DNA:
- C: C* and C5 found in South Asia
- F: F*, F1, F2 and H found in South Asia. Other subclades mostly west of it (IJ, G, and partly K - F4 too, found in Europe recently).
- D is the most problematic as it could arguably account for a Central Asian route and is almost absent in south Asia (some YAP+ reported anyhow) but Andamenese overwhelming D* rather supports again the coastal migration model.
MtDNA:
- M: M has its greatest diversity in South Asia, with many many subclades (often poorly studied, I've counted 16 out of 30) in the subcontinent. The rest are mostly found in East Eurasia and Oceania, exception made of M1.
- N: N* is only reported in Cromagnon aDNA, AFAIK, but the distribution of N(xR) is not really that different from that of M (Australia, East Eurasia and some in West Eurasia too).
- R: is again not substantially differently distributed than M or N. With 7/14 subclades in South Asia, 3 in East Eurasia, 3 in West Eurasia and one in Sahul.
The problem may be that we tend to ignore smaller subclades, focusing only in those that have a large distribution. Check this graph of mtDNA geography for a reality check:
http://img176.imageshack.us/img176/7041/mtdnageographyto7.png
(can't recall the source but it's good stuff)
Overall, with the only possible exception of Y-DNA K (discussed above) and arguably D, they strongly suggest a South Asian urheimat for Eurasians (included Native Oceanians and Americans). Not only South Asia has the highest top-level diversity for almost all clades, it also is the place in the middle from all their destinations (it's just impossible to explain connections between Sahul and West Eurasia without considering South Asia at least as transit).
Also, map 1 seems to suggest an OOA movement across West Asia. But this was probably not the case. Nowadays the mainstream hypothesis is a coastal migration via southern Arabia, where the top-level Eurasian clades (Y-DNA C, D and F and mtDNA M, N and probably R too) may have coalesced (as the population was surely small) before reaching South Asia and spreading around, following largely coastal routes (that are often the easiest ones and also provide with some aboundance of resources).
Enjoy.
Thanks again. And of course there is another "thing I would change". I haven't put in the recently defined Y-chromosome lines S and T. Both part of K, S confined to the extreme southeast and T widespread.
Obviously the displacement of Neanderthals was hindered by primitive weapons. But under a simple recent out of Africa scenario geneocide nevertheless. By the way there are a few cultures besides Gravettian usually thought to be hybrid.
The significance of SE H.erectus is not that they are similar to modern humans (who vary immensely anyway) but to a particular group of Australian fossils. Modern Australian Aborigines also exhibit some of these traits, although greatly reduced. "That strange genetic map" is a direct copy of Cavalli-Sforza's first principal component map from "The History and Geography of Human Genes". Although published in 1994 it is nevertheless a fascinating book. And a significant map. At Anthropology.net one contributor is arguing for an out of America scenario. We could just as validly argue for an out of Australia scenariio.
Sahul is actually just as complicated as Europe. It's just that people of European origin have done most of the genetic studies, therefore the clades in Europe appear to be more complex. It certainly makes sense that the boating technology necessary to cross Wallace's line then spread around the world. This is much more likely than the idea humans moved out of Africa with somehow efficient boats. Strengthens the case for an expansion of Y-chromosomes and mtDNA from Sahul.
Y-chromosomes N and O must have coexisted in reasonably close proximity. East Asia is the only realistic possibility. N/O would therefore represent a clade that moved north from Sahul. M and S moved east to New guinea and Australia and L, T and P moved west through India. Some K moved with all the descendant lines. The most likely scenario to me.
"C* and C5 found in South Asia." Yes but at different places. And what about Y-chrs C1, C2, C3, C4 and the new one C6. They basically spread in a cline from C5 in Pakistan, through Central Asia (C3) via Japan (C1), to Sahul, Australia, SE Asia (the rest) to the opposite end of the cline, C* in SE Asia and parts of coastal India. The distribution speaks volumes to me.
I agree totally with your comments regarding Y-chr F and mtDNA M, definitely India. But mtDNA N? Not found in India at all as far as I know. OK. You can argue for any scenario with bottlenecks and selection sweeps. But that's a bit like saying God did it.
"The problem may be that we tend to ignore smaller subclades." True. But as I pointed out above they often appear to be smaller simply because they are not connected to European haplogroups.
Look forward to hearing more of your ideas. By the way. Check mtDNA line Q2 in the link you provided. Does it or does it not provide evidence for a later migration into Australia? Perhaps others came in with it. And the diagram is in no way incompatible with my maps. Perhaps showing R mtDNA is actually Indian.
I forgot. You wrote, "Andamenese overwhelming D* rather supports again the coastal migration model". I'm afraid you'll find occupation of the Andaman Islands is much too recent to reveal anything about ancient migrations in the region. The Andamanese could well descend from a movement south through Burma from Tibet. After all the Burmese language is related to Tibetan so there is obviously some connection.
I'm afraid you'll find occupation of the Andaman Islands is much too recent to reveal anything about ancient migrations in the region.
I haven't found that. Wikipedia claims that the Andamanese are thought to have arrived from Africa and to have lived on the Andaman Islands for up to 60,000 years. It doesn't look like any very scientific claim (other than losely based in DNA studies, not cited) but it's all I have.
Still their D* is clearly distinct from that of Central/East Asia and very variegated. Still I agree that it's not in itself a very solid argument. The opposite logic could perfectly be applied to: they moved from coastal Burma into East Asia, where the haplogroup is relatively frequent with an apparently old distribution too (Ainu specially). The issue is not settled, as it's not settled the one of NO (that could have also arrived to East Asia via Indochina).
Haplogroup C instead is more clear to my eyes. True that it is not found (except in derived recent forms apparently) in West Eurasia but much of the same can be said of mtDNA M. It could perfectly have participated only in the eastward expansion.
The existence of at least two different emigrations to East and West (maybe more, ok) from South Asia is pretty clear otherwise, specially in the mtDNA and the F macrohaplogroup of Y-DNA. C is not any real problem in itself, only D and K could ask for alternative explanations maybe (but not necesarily). As Y-DNA is more easily subject to change than mtDNA, I think that the overall OOSA model is very valid. Sure that the neighbouring regions of South Asia (West, Central and SE Asia) are not excluded from playing some role but still South Asia is always in the middle of all - and the only alternative route between Eastern and Western Eurasia, not valid for Sahul-West Eurasia) is many thousand Km northwards, in Siberia. It was used certainly but thinking it was used in the UP only by Y-DNA P (Q actually) and rare mtDNA X2 is perfectly valid, I understand. And Occam's Razor suggests to exclude what only adds unnecesary complications.
Of course Occam's Razor can be wrong, specially as we may be missing important data. But generally is a good way to think until that new revolutionary data arrives (if it does).
I haven't put in the recently defined Y-chromosome lines S and T.
Actually they have just been renamed. they were known as K5 and K2 respectively before. So nothing new. There are some other rearrangements going on (specially from Karafet, 2008) but nothing radically new. Except one thing: C and F are now known to be united upstream by a shared SNP, the corresponding macro-haplogroup is called CF. Hence CR (or CT) split into DE and CF first of all.
Obviously the displacement of Neanderthals was hindered by primitive weapons.
Not just weapons, also economy. Unlike in North America (or Argentina or Australia for the case), in early UP Europe there was not the huge economical difference between natives and immigrants: both were hunter-gatherers, not ones industrial and the others hunter-gatherers with some agriculture here and there. The population densities one and the other could support were wildly different and also the newcomers could bring more and more new settlers from the other side of the ocean easily. This situation has no precedent. You can compare for instance with the Bantu migration but this, even if technologically very superior (iron, agriculture) took about 1,000 years to be completed, leaving many pockets of natives and allowing for more time for admixture and mutual assimilation.
By the way there are a few cultures besides Gravettian usually thought to be hybrid.
Hmmm. "Perigordian" is now genrally considered obsolete and meaningless. True that you can see similitudes between Chatelperronian and Gravettian (and other techs of West Asia) but there is a clear time gap that basically forbids them to be related. Same about Aurignacian and Magdalenian, even if these two techs involve only H. sapiens, or what about Solutrean and Clovis? It's a comlex issue but simplistic morphological connections with no fossil record to substatiate them look really feeble.
"That strange genetic map" is a direct copy of Cavalli-Sforza's first principal component map from "The History and Geography of Human Genes".
I see. Without a scale it seems to make no sense, as Europe, Africa and parts of America are shaded the same (meaning surely low PC1). PC graphs and maps anyhow can be misleading, as what is a major PC in the global context is maybe trivial in the regional one and vice versa. When dealing with autosomal DNA I strongly prefer K-means stats, specially when deep enough.
At Anthropology.net one contributor is arguing for an out of America scenario.
Have you read the discussion? He rejects (out of ignorance) the SNP hierarchy, he claims erroneously that N and M have nothing to do with L3. I just abandoned that discussion after a while, saying that it was "like arguing with a creationist". Sorry but the guy is totally wrong and needs a good re-learning of autosomal genetics. He may be very good at kinship studies but he is making a fool of himself when he tries to adress genetics with a shallow and outdated knowledge.
I'd say this is not really important if that is made by an amateur in a forum but a scholar that is authoring a book... well.
We could just as validly argue for an out of Australia scenariio.
I did speculate for an out of New Guinea (or SE Asia) only for the particular case of K - and guess the last word on this issue is yet to be said. But for the overall Eurasian spread it's not likely at all.
Also the archaeology of SE Asia/Sahul does not yield advanced enough implements to justify such a hypothetical "second wave" or its origin there. Instead West and Central Asia (and even India) do provide strong indications and even "evidence" for the developement of blade-based UP techs that could correlate for instance with Aurignacian (and other stuff maybe).
Sahul is actually just as complicated as Europe. It's just that people of European origin have done most of the genetic studies, therefore the clades in Europe appear to be more complex.
You are probably right about that. Eurocentrism (coupled with relative lack of research elsewhere) is a problem in all this field.
It certainly makes sense that the boating technology necessary to cross Wallace's line then spread around the world. This is much more likely than the idea humans moved out of Africa with somehow efficient boats. Strengthens the case for an expansion of Y-chromosomes and mtDNA from Sahul.
I wouldn't claim that but the opposite. People that was already living at the coast in Africa, then Arabia and then India, had surely some knowledge of boat making enough to travel rivers and sail coastal waters (deep water navigation only seeems to appear with Neolithic). This knowledge is certainly necessary to cross the Wallace line, but also to cross Bab el Mandeb and probably the Ganges-Brahmaputra Delta (and other broad rivers probably).
In any case, autosomal DNA proves with too much strength that OOA is real. Thinking otherwise is smashing your head against a too solid wall. It'll bring you nowhere - unless of course you are an able geneticist that can gather the evidence to demonstrate the opposite, what in any case doesn't seem likely to happen at this stage.
"C* and C5 found in South Asia." Yes but at different places. And what about Y-chrs C1, C2, C3, C4 and the new one C6. They basically spread in a cline from C5 in Pakistan, through Central Asia (C3) via Japan (C1), to Sahul, Australia, SE Asia (the rest) to the opposite end of the cline, C* in SE Asia and parts of coastal India. The distribution speaks volumes to me.
The distribution suggests an Indian origin or at most a SE Asian one. Forget about Central Asia/Siberia for the case of C anyhow: the distribution is obviously along the southern and eastern Asian coasts, plus mainland Oceania. The presence of C3 in Central Asia is a recent Turco-Mongol input. Wether the origin is in South or SE Asia, at one extreme or slightly more centered is arguable, I guess. But the presence of C* in South and SE Asia, while not sufficient evidence, does strongly suggest possible origins, origins that are consistent with the distribution of other haploid clades.
But mtDNA N? Not found in India at all as far as I know.
N(xR) is pretty rare in any case (except in East Asian A and the Oceanian S). But N is found certainly in South Asia in form of R in any case, and of U (a major and very old subclade of R that was certainly already divided in Aurignacian times).
N is also thought to be older than M. So basically no or little N* must have arrived to mainland Eurasia but already formed subclades possibly.
You could argue, as some have done, for a dual origin of both clades: one (N) travelling through West Asia in small numbers and the other (M) being the one arriving via the coastal route. Speculative but not impossible. In any case, both clades or their derivates must have mixed at some point and that point was with all likehood South Asia, spcially because it is the only logic route between West Asia and Australia.
OK. You can argue for any scenario with bottlenecks and selection sweeps. But that's a bit like saying God did it.
Drift and founder effects. The result is similar to bottlenecks but they are much more logical (at least for me).
Drift, as you may know, happens at strong rates in small populations (and all Paleolithic populations were that) - you need no epydemics or other catastrophes to justify it, just the usual randomness fixating one or few clades for each of these small populations. In large populations instead it almost doesn't exist.
Founder effects happen every time a small group breaks apart and moves elsewhere. If a given "clan" had two lineages (arbitrarily A and B) and a handful break apart and found a new clan in a new territory, it is very likely that they will be only or almost only one of the two clades, specially as they may be relatives. Drift does the rest.
But still I admit that N(xR) (but not N as a whole) looks like less strongly related with South Asia.
One possiblity (just for the fun of speculating) could be that after crossing what is now the Persian Gulf, different populations estabilished themselves in different spots between Iran and India. The possible local dynamics at this micro-scale level has not really been studied AFAIK but looks very interesting to me.
But it would not substantially alter the South Asian hub model, just refine the detail (something that may not be possible anyhow).
"The problem may be that we tend to ignore smaller subclades." True. But as I pointed out above they often appear to be smaller simply because they are not connected to European haplogroups.
Some are just smaller anyhow. But they reperesent equal sister branches. K1 may be quite rare and cannot compare with the spread of P or NO, but they all descend from the same K ancestor, so when studying K you must be careful to consider K1 as well as other clades.
By the way. Check mtDNA line Q2 in the link you provided. Does it or does it not provide evidence for a later migration into Australia?
No. Q/M29 is exclusively Oceanian. It may provide indications of internal migrations beyond the Wallace line but it does not suggest any immigration from the outside. The non-Austronesian clades of Oceania all seem very old.
Mainland Oceania seems to have been peopled by groups that were somewhat genetically diverse (large, maybe several groups) but the clades that their founding fathers and mothers brought with them are not connected with the exterior except by upstream nodes. There is nothing like Y-DNA P or NO stretching from one end of Eurasia to the other (and even into America), or (less common) mtDNA X (rendered as WEA in the graph but actually also found in Siberia and America), or (very old) U stretching all through West Eurasia, plus parts of Africa and Siberia, or the shared mtDNA haplos between East Asia and America.
Sahul looks like people went there once upon a time and remained almost totally isolated since then. Cavalli-Sforza work itself (I have his 1996 book) suggests the same, and anything else I have read as well.
If that means a single migration (split or not before arriving to Australia and New Guinea respectively) or two I can't say for sure but certainly their were not wildly separated in time and seem to correspond to a single general migrational current.
If a single migration, it must have happened after the split of Y-DNA K in different subclades (what IMO must have happened in Northern India, near the Ganges - but could also be attributed to SE Asia hypothetically). If two, then K would have arrived later but not surely spread from such a remote place as New Guinea.
Enjoy.
These comments are getting really long so I'll try to keep this as short as possible. I agree the arguments for out of America seem totally ridiculous.
You quote: "thought to have arrived from Africa and to have lived on the Andaman Islands for up to 60,000 years". Only because this time frame is the only possible fit with the simple single recent out of Africa scenario. There's absolutly no evidence humans have been on the Andamans much more than half this length of time.
I see at Dienekes you are arguing that modern haplogroups have behaved in exactly the manner I am arguing for ancient ones. Why the difference? It seems very likely that on many occassions two or more haplogroups have moved together but there is no reason at all why they invariably do, now or in the remote past. You remark, "N is also thought to be older than M". Therefore straight away we have evidence for two separate populations.
"Different populations estabilished themselves in different spots between Iran and India". And "one (N) travelling through West Asia in small numbers and the other (M) being the one arriving via the coastal route". I believe the evidence shows exactly this, although I suspect the Iranian Plateau was more important, especially early on. A route to the far east through Central Asia is entirely possible during periods of warmer climate.
"NO ... could have also arrived to East Asia via Indochina". But that's exactly what I've been saying! K diversified around Wallace's line and moved in all directions from there.
And "C and F are now known to be united upstream by a shared SNP". But I've always accepted the two had a common ancestry separate from D/E, probably from a separate movement. C and F presumably diversified from B outside of Africa, then replaced their parent line outside that continent.
"Deep water navigation only seeems to appear with Neolithic. Ahh. Excuse me. How did humans cross wallace's line then? Certainly long before the Neolithic. "People ... living at the coast in Africa, then Arabia and then India, had surely some knowledge of boat making enough to travel rivers and sail coastal waters." Why would they? I'd accept the rivers but they would have no ecological advantage in developing anything like functioning boats capable of going even a short distance from the coast. Boats are much more likely to have developed where there was an advantage in reaching previously unoccupied regions.
We fundamentally agree. It's just you believe in one main centre of diversification, a sort of Garden of Eden, where I believe the evidence reveals shifting patterns of diversification and expansion, one of the important ones being further east that where you see your main one. We agree that India was important but I see more movement into there from further east, as well as from the west.
Thanks for your time and I look forward to more stimulating comments.
There's absolutly no evidence humans have been on the Andamans much more than half this length of time.
Nor almost anywhere else in Eurasia. Human fossils dated before 40 BP are very rare. Even in South Asia the oldest known human fossl is only 35 KY old - but with all likehood humans were there before.
I see at Dienekes you are arguing that modern haplogroups have behaved in exactly the manner I am arguing for ancient ones. Why the difference?
Not sure what you mean. But in any case, much smaller populations, as were those of UP, are good terrain for drift. This situation changed totally with Neolithic.
You remark, "N is also thought to be older than M". Therefore straight away we have evidence for two separate populations.
No evidence that would be accepted by any impartial tribunal here. Even the difference of age between N and M is just a good guess. Two different clades do not necesarily mean two different separate populations, not at all. That's a too simplistic way to deal with the genetic data.
If we would have two different peoplings, we would probably find even today some Y-DNA and mtDNA clades coupled as they would have been in the distant past. Instead we find them coupled (threesomed or whatever) in all possible combinations. So the most likely scenario still seems that all migrated together or that, if they did separately, mixed again soon after before spreading through the rest of Eurasia (Sahul included).
I believe the evidence shows exactly this, although I suspect the Iranian Plateau was more important, especially early on. A route to the far east through Central Asia is entirely possible during periods of warmer climate.
But the coastal route seems easier. I'd like to see your evidence anyhow. Genetics is not conclusive (nor strongly suggestive) about that but maybe you are thinking in some archaeological info I'm missing.
It remains a possibility anyhow. I can consider certainly a two stages expansion into Central Asia/Siberia: first D (if it did not go via the coast) and then P. But for Siberia it's been quite reasonably argued that Q represents the oldest layer.
"NO ... could have also arrived to East Asia via Indochina". But that's exactly what I've been saying! K diversified around Wallace's line and moved in all directions from there.
Indochina is not really "around Wallace line". I would rather say "around South Asia". But whatever.
But I've always accepted the two had a common ancestry separate from D/E, probably from a separate movement. C and F presumably diversified from B outside of Africa, then replaced their parent line outside that continent.
More likely they replaced nothing as they were all the BR (rather than B, that is an Africa-only clade) existing in Eurasia (together with D). I see it more as CR migrating out of Africa (possibly with D) and coalescing for some time while F anc C formed before beginning the real expansion after crossing Hormuz. E as such never went OOA (except recently in the form of E3b), much less B. So no high level BR in Eurasia, nor CR. Just CF and D.
C anyhow is almost unknown of west of South Asia, so the expansion of C must have begun there (or farther east). Same about mtDNA M (M1 is now thought to be a back-migration from India).
"Deep water navigation only seeems to appear with Neolithic. Ahh. Excuse me. How did humans cross wallace's line then?
You don't need deep water navigation to cross a strait, do you? You just need a canoe or raft. It's not like sailing across the open ocean, that's what I mean.
Why would they? I'd accept the rivers but they would have no ecological advantage in developing anything like functioning boats capable of going even a short distance from the coast.
How did they then cross Bab el Mandeb?
Besides boating certainly provides good alternatives for food gathering (fishing specially). So yes, knowledge of boats provide advantages re. food supply and provide good mobility options that otherwise would be impossible.
Boats are much more likely to have developed where there was an advantage in reaching previously unoccupied regions.
An advantage in reching to resources. Think like a hunter-gatherer: most of the time you may move around but do not really migrate. Your main concern is to gather resources (food and raw materials), so if you are going to develope a quotidiain tech like boats, you will for quotidain needs like resource gathering and exploration, not for estraordinary needs like migrating to an unknown territory. Once you have it, you can also use for that as well, of course.
Even if you don't go out to the sea to fish, you will certainly move faster on boat in many circumstances, allowing you to return home faster and with more captures/materials than if you just walk, specially if the terrain is difficult.
I deduce from your words that you are not personally familiar with the coast or big rivers, where boats are a great advantage, specially where there are no modern roads. And they are certainly a huge advantage to cross rivers, bays and stuaries specially, as well as to reach off to islands and rough coasts full of seafood.
We fundamentally agree. It's just you believe in one main centre of diversification, a sort of Garden of Eden, where I believe the evidence reveals shifting patterns of diversification and expansion, one of the important ones being further east that where you see your main one. We agree that India was important but I see more movement into there from further east, as well as from the west.
You want to make it unnecesarily complicated, I think. Certainly parts of West Asia and SE Asia may have played important roles in the spread but for each high level clade, you see its subclades west and east of South Asia, what really demands a special role for this part of the world. The Central Asian corridor, cold, arid, remote and once full of potentially hostile Neanderthals, cannot explain that pattern, much less when the connection is between Sahul and the West. Only South Asia (together with SE Asia certainly) can.
Also it's been quite logically reasoned that the people that came out of Africa (or out of South Arabia) was more used both biologically and culturally to tropical conditions (and to coastal ones possibly). And again Central Asia is the opposite of those desirable conditions and requires an special adaptation. This adaptation would eventually happen but it's not like the first way you would choose when other options are still open.
Enjoy.
"Human fossils dated before 40 BP are very rare". But evidence for their presence before then is more common. Even without taking the dating of Mungo Man seriously humans were in Australia by at least 45,000 years ago and probably a fair while before then. Did they get there via the southern coast of Asia? For many reasons I think it's unlikely.
"we find them ... in all possible combinations". Yes. But wouldn't that mean that the most UNLIKELY scenario would be "that all migrated together"? It would be much more likely the haplogroups moved independently through pre-existing populations, as you take for granted happened in Greece in more recent times.
"Indochina is not really 'around Wallace line'". From west to east we have 1 Indochina, 2 Wallacea, 3 Sahul. Close enough for me.
"But the coastal route seems easier". Hardly. What about coastal cliffs, mangrove swamps. How far along any coast have you ever tried to walk? The region I live in is noted for it's sailors, many visiting here after making extremely long trips through the Pacific.
"M1 is now thought to be a back-migration from India". Thanks for that. I didn't know it but you can see I'd readily agree.
"You don't need deep water navigation to cross a strait". Well actually...It's something like 100 km at least to cross Wallace's line at the shortest gap so it's quite a bit "like sailing across the open ocean".
"How did they then cross Bab el Mandeb?" I doubt that they did. They took the land route. I don't think the possibility of fishing from a boat would have developed until boats had been invented. I mean early humans would not have known there were fish in deeper water until they had access to it. However if they could see an island they would assume it was much the same as where they already were, especially once they'd pulled the trick off at least once.
I agree that Central Asia is not tropical. But I'm not talking about Siberia where I'd certainly accept Q was the first Y-chromosome. "The Central Asian corridor, cold, arid, remote and once full of ... Neanderthals". Agreed: "potentially hostile", but not necessarily. How about friendly, in fact mixing and hybridising? The haplogroups moved through pre-existing human populations. Neanderthal genes may be extinct but they have potentially been widely spread throughout the world, not just confined the the remotest regions of Western Europe. And in East Asia we have Pekin Man's descendants and in SE Asia we have H. erectus surviving until long after modern humans (whatever we mean by that expression) reached Australia.
"Human fossils dated before 40 BP are very rare". But evidence for their presence before then is more common. Even without taking the dating of Mungo Man seriously humans were in Australia by at least 45,000 years ago and probably a fair while before then.
Maybe. It doesn't change a thing, as the coastal migrtion model actually suggest older dates of c. 50 and even 60 KYBP. I just meant to contradict your claim that Andamanese have only been there the last 35,000 years or so. Also remeber that what once were coasts are now the depths of he seas, so evidence in this aspect must be extremly scarce.
But anyhow they could be the last visible survivors (the ones showing clearly the old lineage) of a group that was once more extended and had to retreat to remote areas like those islands. Their haplotype diversity is certainly very high in any case.
Did they get there via the southern coast of Asia? For many reasons I think it's unlikely.
I just hope you don't think they took the plane New York - Sydney. ;)
Coastal (or more or leas along the coasts) travel is the only way they could have arrived to Sahul. Would we be discussing Kostienki or the Pygmies of the African jungle, then I would think otherwise but the ancestors Australian Natives and Papuans can only have arrived there by boat, what strongly suggests familiarity with the coasts and some navigation before their arrival.
wouldn't that mean that the most UNLIKELY scenario would be "that all migrated together"? It would be much more likely the haplogroups moved independently through pre-existing populations, as you take for granted happened in Greece in more recent times.
Could you please quote where I said that in the Greek DNA discussion? I am starting to think that you are misreading me. If you mean the J2b "jump", then I was just saying that J2b is largely absent (very low) in most of Greece (per Dienekes' data). It may mean that the area was already densely populated for them to make an impact on arrival or that they never travelled through Greece, something they did not need to do to reach Rumania (or wherever this clade is found) in fact.
Anyhow, that they all "migrated together" is, agreed, a simplified explanation surely. We just don't have the info to know how they actually did. Southern Arabian archaeology is virtually null and genetic studies are not precisely aboundant either (yet there is one that has found an "Australian" M clade, unknown before, in the Arabian peninsula: http://dienekes.blogspot.com/2008/02/news-on-arabian-mtdna.html).
But, speculating, if the D clan coalesced primarily in Yemen and the C and F clans in different parts of Oman at that time, the difference would be small in any case. Specially as they surely mixed once and again with each other, being so close neighbours. Same for M, N and R. The microscopic detail is for us now impossible to discern in any case.
What doesn't seem likely is that they took widely different routes because in that case we would still see the effects of that intial divergence. And we don't. We only see clear divergence from South Asia (or somewhere nearby). But these differences are not strongly linked to the different high level haplos but to the next level ones. M, N and R are found both west and east of South Asia. Ok, that's not the case of Y-DNA C and D but male lineages are more susceptible to drift and fixation (as men can have much higher diferential rates of reproductive success than women).
"Indochina is not really 'around Wallace line'". From west to east we have 1 Indochina, 2 Wallacea, 3 Sahul. Close enough for me.
Forgot the Malay Archipielago. Indochina peninsula (mainland SE Asia) is between South Asia, East Asia and the Malay archipiealgo (then another large peninsula - but still a different subcontinental mass). Ok that the difference may not look brutal but it's still closer to South Asia than to Sahul.
"But the coastal route seems easier". Hardly. What about coastal cliffs, mangrove swamps. How far along any coast have you ever tried to walk? The region I live in is noted for it's sailors, many visiting here after making extremely long trips through the Pacific.
It's certaily still easier by boat. Anyhow, excluding vegetation, computer models suggest the coastal route as the easiest one (together with some riverine routes, specially the Narmada-Son-Ganges one), at least for India (see Julie S. Field, 2006). This model correlates very well with the actual distribution of middle and upper paleolithic remains in India, except for the coast itself that now is submerged.
"M1 is now thought to be a back-migration from India". Thanks for that. I didn't know it but you can see I'd readily agree.
How does that correlate with your idea of a migration via mainland West Asia? I cannot imagine really. In fact some of the greatest weight for the coastal migra odel seems to come from the fact that some clades (C, D, largely M) are totally or almost totally absent in West Eurasia. This strongly suggests this region as destiny of one or more secondary migrations from South Asia, and not the center of Eurasian dispersal (not at all).
"You don't need deep water navigation to cross a strait". Well actually...It's something like 100 km at least to cross Wallace's line at the shortest gap so it's quite a bit "like sailing across the open ocean".
Gibraltar strait is some 13km across and can be crossed swimming on a surfing board. So I guess the 50-100 km that Wallace line straits had in the Ice Age should be a barier but not an impossible to cross one. In fact they actually made it.
Btw, Lombok strait is only 60 km wide at current sea levels.
"How did they then cross Bab el Mandeb?" I doubt that they did. They took the land route.
Are you saying that Y-DNA clades D and C (and mtDNA M too) were for milennia hunting in Western Eurasia (not in the southern Arabian coastal strip) and have not left a single trace? It sounds incredible to say the least. The genetic evidence strongly indicates an East Africa > South Asia almost direct migration.
I don't think the possibility of fishing from a boat would have developed until boats had been invented. I mean early humans would not have known there were fish in deeper water until they had access to it. However if they could see an island they would assume it was much the same as where they already were, especially once they'd pulled the trick off at least once.
Never mind fishing, what about coastal and riverine travel? If you know that down the cliff there is a lot of seafood and that the only way to reach there is a risky swim or, eventualy, some kind of boat, any half-smart human would have started plotting immediately how to reach that food fast and with minimal risk.
Ok, maybe first they climbed down the cliff... but anyhow boats would be desirable. What about apnea diving? That's another excellent source of food and boats would help to reach difficult but rich areas. There are islands in Bab el Mandeb too, btw, and are visible from the mainland, not like something that is 100 km away.
I can think in a million reasons to design and build boats, specially if you live by the coast. Travelling to far lands would only be a less important motivation certainly.
"The Central Asian corridor, cold, arid, remote and once full of ... Neanderthals". Agreed: "potentially hostile", but not necessarily. How about friendly, in fact mixing and hybridising? The haplogroups moved through pre-existing human populations. Neanderthal genes may be extinct but they have potentially been widely spread throughout the world, not just confined the the remotest regions of Western Europe. And in East Asia we have Pekin Man's descendants and in SE Asia we have H. erectus surviving until long after modern humans (whatever we mean by that expression) reached Australia.
No evidence. We have go through this before.
In fact I am inclined to think that partly it was Neanderthal success what kept ancient humans stuck in Africa until boats were developed. We know that some sapiens arrived to Palestine c. 100,000 years BP but we also know they had no continuity and were replaced by Neanderthals. They also would seem partly responsible to have hindered later human advance in West Eurasia. If our ancestors had problems getting to temperate mainland West Asia, I can only thik it was even more difficult to get estabilished in cold and arid Central Asia.
I can't know how Neanderthals behaved or how open minded and welcoming they were with strangers but I know they were much stronger than us, so unless we could overcome them with brains and organization, they would normally succeed in direct confrontation.
Remember that we were the intruders in their land. The "ugly", dark-skinned, lean, whimpy, short-faced and surely (in their minds) viper-tongued invaders, who brought diseases and did not respect the well known traditionl boundaries of their hunt teritories. I don't think they had any reason to be friendly towards us.
Maybe the reality was more complex but certainly the fast advance of Aurignacians in Europe, for instance, doesn't look like a friendly move on our side.
Hey Maju. "the ancestors Australian Natives and Papuans can only have arrived there by boat". It's quite likely they developed it in the region just prior to crossing Wallace's line. There is absolutely no evidence for the southern coastal route including no genetic evidence. Australian-type genes found there are more likely to arise from a back migration. They're certainly not ancestral to australian ones. Sure, physical evidence may be submerged but that again is arguing like a fundamentalist Christian.
Tim sent me some links recently regarding a conference "Archeology in the Middle East", or something like that. I'm sure he can find the link easier than I can. But several papers deal with sites in Yemen, Hadramawt and the Persian Gulf. Although each site is different not one site examined shows any connection to anywhere in East Africa. In spite of their variation they are all much closer to those in the Levant. NO CROSSING OF THE BAB AL MANDAB. Admit it.
At times of low sea lavel "the Malay Archipielago" is connected to "Indochina". One and the same place for much of our evolution.
Finally I'm sure you'll like this link. Shows ancient human movement around the world is even more complicated than I've argued for!
http://anthropology.net/2008/05/25/how-was-the-world-peopled/#comment-11037