It looks as if the vast majority (95% or more depending on the population) of the ancestry of non-African humans derives from a population expansion which began around ~60,000 years ago. Before this period some researchers argue there was a non-trivial period of isolation. The “long bottleneck” (David Reich alludes to this in Who We Are and How We Got Here: Ancient DNA and the New Science of the Human Past). For the vast majority of humans then the last 60,000 years is characterized by a branching process, some reticulation (e.g., South Asians merge West and East Eurasian lineages) between these branches from a common ancestor, as well as introgression from archaic lineages like Neanderthals and Denisovans.
Though I do accept that it seems that modern humans probably migrated out of Africa before 60,000 years ago, mostly due to the results from archaeology, I think the genetic evidence is strong that these groups contributed very little genetically to contemporary populations.
The situation within Africa is very different. Being conservative it seems likely that the Khoisan ancestral lineage diverged from some other Africans ~200,000 years ago. I say conservative because there are researchers who want to push the divergence much further back. Additionally, several different research groups are now converging in a result that West Africans are a mixture between eastern Sub-Saharan Africans (think the population ancestral to Mota in Ethiopia) and a lineage basal to all other humans. That means that the Khoisan are not the most basal, so even assuming the conservative 200,000 year divergence point for Khoisan, modern humans share a common ancestor earlier than 200,000 years ago.
The upshot here is that around 75 percent of the history of modern humans is within (greater)* Africa. The distinctive “Out of Africa” bottleneck and expansion defines most humans only in the last 25 percent of the history of our species. And, within Africa, the dynamics were very different. The biggest difference is that African populations are not defined by a large number of lineages emerging and diverging around the same period, because there wasn’t a massive and singular expansion within Africa analogous to what occurred outside of Africa (at least until the recent past, with the Bantu expansion). That’s why there’s deep structure within Africa today between groups as divergent as the Bantu, Mbuti, Hadza, and Khoisan.
The term “Basal Eurasian” kind of makes sense in the non-African context because of the singular importance of divergence between lineages in the first 10,000 years or so after the “Out of Africa” event. I’m not sure “Basal human” makes as much sense because there wasn’t a singular event within Africa that allowed for the emergence of modern humans. Rather, it was a process, and probably quite resembles something like multiregionalism.
* Some wiggle room here for the likelihood that modern humans were long present in the liminal Near East.
“West Africans are a mixture between eastern Sub-Saharan Africans (think the population ancestral to Mota in Ethiopia) and a lineage basal to all other humans.”
Would this basal lineage be as deeply divergent as Neanderthals and Denisovans, or not? I’ve read one preprint looking for that lineage’s genomic signatures (https://www.biorxiv.org/content/early/2018/03/21/285734) but didn’t see their relative chronology spelled out — or, perhaps, I lacked the wit to notice it.
not as deeply.
One way of looking at the divergence between archaeology and genetics on Out of Africa is to see it as more of a “Beringian standstill” phenomena and less like a first Out of Africa that failed, followed by a second that succeeded.
Until ca. 45,000-40,000 years BP, Neanderthal occupation seemed to prevent modern human expansion in Europe, and the jungles of Burma and any archaic hominin population seemed to be a similar barrier to modern human expansion until the Toba erruption (ca. 70,000 years BP). But, there is good archaeological evidence for a modern human presence in South Asia pre-Toba, and the lion’s share of non-African Y-DNA seems to derive from Y-DNA F which seems to have a South Asian epicenter. Y-DNA C seems to have a West Asia or South Asian origin. Y-DNA D seems to be largely an expansion that failed with a sporadic distribution that might indicate that it took a Northern route sometime after Y-DNA C expanded with most of its lineage wiped out in Northern regions by the Last Glacial Maximum.
It could be that there was a swath of Out of Africa modern humans that had a range of something like Arabia to West Asia to South Asia until the gates opened first to the East, and later into Europe from ca. 125,000 years ago. The non-African ancestral population must have been isolated for a while, otherwise clades other than Y-DNA CF, Y-DNA D, mtDNA M and mtDNA N would have seeped in.
A long time period with multiple bottlenecks and a low effective population could have facilitated Neanderthal admixture, the narrowing of the number of source uniparental lineages (which are easily lost through drift in periods of stable and declining population, but are rarely lost during period of population expansion), and would have allowed for cultural divides to develop as well. I am particularly struck by the haplogroup diversity seen in the original settlers of Australia and New Guinea that had to have developed outside Africa, and yet involved lineages that should have taken many tens of thousands of years to diversify. This couldn’t have happened without a long gap between “Out of Africa” and ca. 70,000 to 65,000 years BP when the modern human presence in Southeast Asia, Melanesia, Australia and East Asia suddenly surges.
I don’t think that the evidence bears out Dienekes notion of a non-African origin for Y-DNA E, but it doesn’t at all strain credulity to think that non-Africans may have, out of necessity, made cultural and technological innovations that primed them for rapid expansion out of their Arabia to South Asian range when Toba’s impact on mainland Southeast Asia’s jungles and the disrupted ecology’s harms to the native hominin populations there. As they ballooned with a rapid expansion into virgin territory, it wouldn’t be surprising if some of those cultural and technological innovations seeped back into Africa and led to a major wave of intra-African expansions around the same time (less dramatic because these innovations weren’t tailored for African conditions) without a really major demic impact. The Y-DNA E evidence suggest both a major East to West sweep and a major West to East sweep across the region that is north of the Congo and South of the Sahara, which would have been greener for long intervals in the last 70,000 years. But, the Congo remained a major barrier to North-South expansion and links between communities of people into the Bantu era as it never suffered an indignity comparable to what Toba did to mainland Southeast Asia’s jungles for a brief while.
The only way to distinguish an “out of Africa” ca. 60,000-70,000 years ago, from a long stand still, is the internal haplogroup diversity of non-African clades as of 60,000-70,000 years ago at the time of the expansion. Otherwise, a long period of stasis outside Africa leaves little in the way of a genetic trace.
In Africa, the absence of Khoisan and Pygmy type populations outside refugia is the dog that didn’t bark that points to population replacement pre-Bantu.
“the liminal Near East.”
I don’t quite get this one.
Liminal:
1. relating to a transitional or initial stage of a process.
2. occupying a position at, or on both sides of, a boundary or threshold.
Might you have meant:
Littoral:
adjective: relating to or situated on the shore of the sea or a lake.
noun: noun; a region lying along a shore.
Or perhaps:
Levant: the area of the E Mediterranean now occupied by Lebanon, Syria, and Israel.
mean liminal to africa. threshold of africa.
@ohwilleke, using those y haplogroups as signals of early population divergence seems inconsistent with predicted divergence times of those y-clades though from https://link.springer.com/article/10.1007%2Fs00439-017-1773-z. E.g. DE from CF at 65 kya, C from F split 64 kya there. Unless generation time radically different (double?) seems not possible to get to those split structures at those time depths. But those generation time estimates were validated through ancient dna.