The Pith: The Bushmen branch of the human family tree diverged ~130,000 years ago. The non-Africans branched off from the Africans ~50,000 years ago. The Europeans and East Asians diverged ~35,000 years ago.
One of the terms in paleoanthropology which can confuse is that of archaic Homo sapiens (AHS). This is in contrast to anatomically modern humans (AMH). A simple Out of Africa "recent-origin-with-replacement" model allowed to sidestep the semantic imprecision in tossing disparate populations into a generic category such as AHS (similarly, the term "animal" as opposed to "human" has some colloquial utility, but it's not scientifically useful). But the possibility of admixture from archaic lineagesin modern human populations forces us to grapple with the dichotomy between AHS and AMH, as modern humans may be a compound of these two categories (not to mention the idea of behaviorally modern humans, who are a subset of AMH). I assume that fleshing out the details of a new paradigm which is both precise and accurate will be a project for the coming years. But before we move on we need to fix more sturdily our understanding of the genealogical relationships of contemporary human populations. Over the past few years there have been major strides in this domain, confirming the broad outline of a dominant African heritage for modern humans. Geneticists have moved from classical markers to SNP data, focusing on hundreds of thousands of genetic variants. But now they're shifting to whole genome sequences, which with errors excepted encapsulate the totality of the lowest order aspect of human genetic variation.* Earlier this summer I reviewed a paper in Nature which was a foretaste, Inference of human population history from individual whole-genome sequences. Today Nature has published another, Bayesian inference of ancient human demography from individual genome sequences. The authors analyzed six human genomes, a Northern European, a Nigerian Yoruba, a Han Chinese, a South African Bantu of the Xhosa tribe, and a Nambian Bushmen. Some of these samples are familiar. The Northern European is Craig Venter, while the southern African samples are the same as in the paper on Khoisan genomics. Chimpanzees were the outgroup. Since this is a computational heavy paper, note that it is somewhat sensitive to parameters such as the time of divergence between the chimpanzee and hominin lineage and human mutation rate when it comes to assessing the likelihood of any given final number. What the authors propose to do is construct a tree of human relationships using a coalescent model. All things being equal human populations will start to exhibit genetic differences due to varying mutations and the effects of random genetic drift. An estimation of difference or distinction can give one a rough sense of the time since separation. But obviously this is a very simple model, and does not take into account factors such as gene flow which can reduce differences between populations. But when they included a gene flow parameter they found that the phylogenetic tree and estimates to last common ancestor were relatively robust. Note that of the populations above it does seem reasonable that there was only minimal gene flow in most pairwise cases (e.g., there hasn't been too much gene flow between Northern Europeans and Han Chinese since separation). The exception is the South African Bantu, as that individual is from a tribe with known Khoisan admixture. Additionally, many Khoisan also have some Bantu (or even European in South Africa) admixture. Here are their top line results: - San - Non-San divergence of ~130,000 years B.P. - Eurasian - African divergence of ~50,000 years B.P. - European - East Asian divergence of ~35,000 years B.P. First, the authors note that their method generates a divergence between Europeans and East Asians which is more archaeologically plausible than the ~20,000 year divergence using just allele frequencies. This may be due to the fact that allele frequencies are a subset of the genetic variation, and for various reasons they may be biased (in particular "missing" many variants common amongst non-Europeans). As for the second, I'll let the authors speak:
Notably, our point estimate of ~130 kya suggests that the San divergence occurred ~2.5 times as long ago as the African-Eurasian divergence, that major human population groups diverged at least ~80,000 years before the out-of-Africa migration and that the San divergence is more than one-third as ancient as the human-Neanderthal divergence...Still, human effective population sizes are sufficiently large that these divergence times are small relative to the time required for lineages to find common ancestors in ancestral populations. Indeed, of the mutations differentiating a San individual from a Eurasian individual, only about 25% are expected to have arisen since the San divergence. Thus, the ancient divergence of the San population does not alter the essential fact that far more human variation occurs within population groups than between them
There's now a fair amount of data and results which indicate a deep divergence between the traditionally hunter-gatherer populations of Sub-Saharan Africa and the farmers and agro-pastoralists of the continent. If these results are correct (and they seem to be in line with the earlier genomic analysis of the Bantu and Khoisan from South Africa) then we have here a rather straightforward refutation of the very simple Out of Africa model, where one East African group replaces everyone else. The Bushmen seem to be of another lineage. Secondly, these results suggest that we should be cautious about inferring a qualitative difference between Neandertals and AMH, where the former is bracketed as "archaic". The difference between Bushmen and non-Bushmen is far less than between Neandertal and AMH, but it is by a factor of multiples, not order of magnitude. Finally, the authors brush aside the impact of admixture on the margins to their broader story. What they purport to capture is the primary human narrative, the Africans who marginalized and assimilated other human populations over the past 100,000 years. Citation:Bayesian inference of ancient human demography from individual genome sequences. * I assume that gene regulation and what not may be effected by biophysical features not captured in a raw sequence.