Last year when discussing the possible admixture of Neandertals with the ancestors of modern non-Africans I joked that Sub-Saharan Africans were “pure humans.” This was tongue-in-cheek in part because the results from the Neandertal genome shifted my assessment of the probability of archaic admixture within Africa as well. In other words, there may never have been a pure “human” type which expanded and assimilated archaic ancestry on the margins of its range. Species Platonism may be very misleading for our particular lineage. Rather, what it means to be human has always been in flux, a compromise between extremely different ancestral components.
For years some groups of researchers have been arguing that there is population structure within Africa itself which hints at admixture events before (or after?) the “Out of Africa” event. Genome blogger Dienekes Pontikos has been discussing this possibility for several years as well. With the possibility of archaic admixture outside of Africa it was inevitable that people would revisit their earlier exploration of ancient African admixture and the modern patterns of variation which that might explain. Finally one of the groups working on this has come out with something in PNAS, Genetic Evidence for Archaic Admixture in Africa. Unfortunately it’s not on the website, and I’m not privy to the embargoed copy, so I can’t say much. ScienceDaily and Nature have lengthy write-ups. The details are pretty straightforward. The authors infer using computational methods that there is a 1-2% admixture in Africans of a population which diverged from the mainline of the human ancestral tree ~700,000 years ago. The hybridization occurred on the order of ~40,000 years before the present. The proportions are highest in Central Africans. I assume that this means Pygmies. And I would further bet that the admixture is highest in the Eastern Pygmy populations, such as the Mbuti. The lead author also cautions that this may not be the last word on admixture. No doubt. There are other groups breathing down his neck.
If this is true then a assimilation model of the expansion of H. sapiens sapiens looks more and more plausible. The time period of admixture is pretty much what other scholars are estimating for Neandertals, and presumably Denisovans. I’m not smart enough to figure out how this could be a statistical artifact, but perhaps that explains the congruence? Otherwise, if this is true then you had several repeated events of expansion of one particular lineage (what I term “Neo-Africans”) which demographically swamped the indigenous populations, but still retained a faint, but discernible stamp of their distinctive genetic content. But this may not be exceptional. It may have happened before the emergence of Neo-Africans, and I believe it happened after them (e.g., the rise of agriculturalists). It’s possibly one instance of a rather banal dynamic in the evolution of Homo.