Nature published a paper recently, New species from Ethiopia further expands Middle Pliocene hominin diversity, which seems to complicate the deep history of the hominin lineage. More precisely, it gets curiouser and curiouser, as the number of human(ish) groups proliferates. Honestly I don’t know well this will hold up, a lot of science seems to fade out. Remember the article in Science, A New Kind of Ancestor: Ardipithecus Unveiled? I’m not totally clear on whether the controversy about this find has resolved. At least we’re at the stage where most people seem to accept that Homo floresiensis was a true hominin lineage, rather than a pathology. (by the way, Carl Zimmer has a good write up on the most recent addition to the human family tree)
Genes are something that is more concrete to me. Nature Reviews Genetics has two pieces of interest in this domain. First, Evidence for archaic adaptive introgression in humans. A close reader of this weblog will find little of surprise; the authors do an excellent job of reducing down the key results of the past five years or so that have issued from the discovery that the ancient whole genome of the Neandertal bears all the hallmarks of having been carried over into some lineages of modern humans. In particular, the authors focus on adaptive alleles and regions through a statistical genomic lens. Second, Svante Paabo has a comment in the same edition, The diverse origins of the human gene pool (ungated), which leans heavily on the previous piece.
Paabo to me seems to finally put to bed some old conflicts which have been roiling the field of paleoanthropogy for decades. I think we’re going to finally see the abating of the rhetorical war between those who promote “Out of Africa” vs. “Multi-regionalism.” The fact is in some ways both viewpoints are worth taking into account. As Paabo, and others, have noted, the former model turns out to likely be correct about the provenance of most recent human ancestry. It is mostly derived from an African or near-to-African (e.g., in the Middle East) population on the time scale of ~100,000 years. But, the latter model’s emphasis on regional evolution and adaptation and gene flow across a meta-population system is also a critical insight in understanding the dynamics of how modern humans came to be.
I do like the fact that Paabo also seems to moving past the idea that genomics will yield the one allele or set of alleles that define what made modern humans so biologically special. I don’t have an opposition to biology as being determinative in our cultural flexibility. But, if anything has been clear in what genomics is telling us about recent human evolutionary history, it’s that it is rather more complicated than we might have imagined. I doubt the uniqueness of the surviving lineage of hominins is going to be any more, or less, subtle in the difficulties of resolution.
I do have one bone to pick with Paabo. He states that:
Adaptation through the acquisition of new mutations is generally a slow process: it is rare for favourable alleles to appear, and these are often lost by chance when they first occur in a single individual or in very few individuals. By contrast, if favourable alleles have emerged in one group, they can spread to other groups relatively rapidly by gene flow. This process, called ‘adaptive introgression’, is well documented in bacteria and plants, and described in some cases in animals but it has not previously been considered an important factor in human adaptation.
The idea of “adaptive introgession” has been something I’ve been thinking about, and talking about, in relation to human evolution since 2006 (Google it). That’s because of the focus that Greg Cochran, Henry Harpending, and John Hawks, put on the topic (all these years later I am also professionally interested in this topic, but that’s for a later post!). Now, it is true that Svante Paabo does not seem to have thought of the issue in much detail. His book Neanderthal Man has no references to “adaptive introgression” according to Google Books. In contrast, 10,000 Year Explosion has 9 mentions of the term.
I will note that the authors of the review paper that Paabo leans on heavily for his comment don’t make this omission where credit is due. They cite both Greg Cochran and John Hawks, with a special laudatory note.
Finally, I want to suggest that to a great extent, Multi-regionalists excepted, the previous consensus in human evolutionary studies tended to overestimate the extinction rate of “archaic” lineages. But, it also underestimated the extinction rate of modern lineages. That is, archaic lineages rooted outside of Africa before ~100,000 years ago may play more of a role in the evolution of our modern lineage than we may have guessed when it comes to both genotype and phenotype. Paabo quotes the standard figures of a few percent for Neandertal ancestry, and ~5 percent of Denisovan among Oceanians. From all I have heard and know this seems about right, but I do wonder if this is actually just a floor. Without ancient genomes I suspect we’d still be debating the possibility of archaic admixture from inferences which only statistical genomicists would have a good grasp of. In 2006 Jeff Wall and Michael Hammer stated that “Neanderthals and an as yet unidentified archaic African population contributed to at least 5% of the modern European and West African gene pools.” They were basically dead right. Second, the ancient DNA is also yielding the conclusion that many local populations which flourished during the Pleistocene outside of Africa seem not have to left much genetic legacy today in the same regions. We’ll get more clarity on the topology of the human phylogeny in the near future, but it strikes me that it’ll exhibit features which are somewhat at variance with what we’d have expected 10 years ago.