A new paper in Nature by Heng Li and Richard Durbin contains estimates of the “effective population size” of our ancestors at different points in evolutionary time. (Effective population size isn’t the same as census size, as it reflects things like unequal sex ratios—unlikely in our ancestors—or variation in family size, but it’s probably not too far off.) These are the best estimates of our demographic history to date, as they rest on fewer assumptions than previous methods, and have been validated by computer simulation studies. They bear not only on what happened when early humans were in Africa and then left Africa, but also on our recurrent discussion of the scientific evidence that absolutely rebuts the Adam and Eve story.
The data come from “coalescent” models—estimates of the time in the past at which two copies of genes from different people, or from the same person, last shared a common ancestral gene form. From these models, data on genetic differences within individuals (i.e., between the two copies of each non-sex-chromosome gene that everyone carries), and estimates of ancestral generation times (25 years in this paper) and mutation rates (rates of change at individual DNA bases), you can work out what the population size of our ancestors was at different times in the past.
The authors made their model from complete genome sequences of six individuals: two Europeans, a Korean, a Chinese, and two Yoruba (a west African group). The figure below pretty much tells the tale: it gives effective population size (on the Y axis) at different times in our species’ history (the X axis shows time before the present on a log scale). The estimates from each individual are represented by different-colored lines, and the key gives the ethnicity of each person. These data are from “autosomes” (those chromosomes that are not sex chromosomes), but DNA data from sex chromosomes gives pretty much the same result.
The first thing you see is obvious: our ancestors went through two different phases of population “bottlenecking” (constriction): one occurred about three million years ago, when a large population declined to around 10,000 individuals. The authors note that while this may reflect population size decline associated with the origin of hominins after our split with the lineage that produced modern chimps, they also say that this could be an artifact of ancient genetic polymorphisms maintained by natural selection.
The second bottleneck is the one of interest, for it’s the one associated with a reduced population size as humans left Africa. For the Chinese, Korean, and European genomes, effective population size fell from about 13,500 (at 150,000 years ago) to about 1200 between 20,000 and 40,000 years ago. Now this is the effective population size, almost certainly an underestimate of census size, but that only makes the problem worse: we never went through a bottleneck of anything near two individuals, as the Biblical Adam-and-Eve story suggests. This, of course, means that theologians have to scramble to save that story, turning it, as always, into a “metaphor”. (In science, a falsified hypothesis gets tossed on the scrap heap; in religion, a falsified hypothesis becomes a metaphor.) And it also suggests that Jesus died for that metaphor.
But enough of Biblical exegesis. While the bottleneck for non-European populations was probably associated with a group leaving Africa and subsequently colonizing the world, we also see a somewhat less severe bottleneck in the African samples: from about 16,100 people about 100,000-150,000 years ago to 5,700 about 50,000 years ago. It’s not clear why the populations in Africa bottlenecked as well.
Finally, we also see the population recover in size, with a huge increase in all populations beginning roughly 20,000 years ago. This clearly reflects population growth in both Africa and in areas colonized from Africa as humans expanded around the globe.
There are two other interesting points:
- All the data clearly show that all modern humans, African and non-African alike, descend from one “homogeneous ancestral population in the last 100,000 years, with subsequent minor admixture out of Africa from Neanderthals.” This goes against earlier theories that there is a much older divide separating West African from non-African populations.
- Also contrary to earlier assumptions that after Homo sapiens left Africa (ca. 60,000 years ago) there was little interbreeding between African and non-African populations, the new data show that genetic interchange between these populations continued up until 20,0000-40,000 years ago. This conclusion, though, is provisional because it depends on estimates of mutation rates which are necessarily indirect.
The upshot and the lesson for the science/religion debates: we now have a pretty good estimate of how many ancestors our own species had at various times in the past, all the way back to near when we diverged from the lineage leading to modern chimps. And the lesson for theology is the usual: science has shown that scripture is wrong, so yet another Biblical “truth” becomes a metaphor.
Science continues to invalidate the claims of faith. First special creation went by the board, so theologians—at least the rational ones—were forced to show that of course God would have used evolution to fulfill his Big Plan to Produce Humans. Now Adam and Eve have also become metaphors, leading to all kinds of humorous theological speculations about who were humanity’s parents and what, exactly, was the nature of their Original Sin. Next on the agenda are morality and free will, staples of religious doctrine but items that are starting to be explained purely by science. We now see morality as having a purely secular origin, perhaps involving evolution; and we don’t really have the freedom of choice envisioned by many faiths.
All of this shows that science is dominant to religion, for when they clash, as they inevitably must, “sophisticated” theologians must frantically revise their doctrines to comport with scientific truth. And they hate that. This innate recognition of the precedence of science over faith is, I think, one reason why so many religious people and faitheists are picking at science, claiming that the methodology of science is based just as strongly on faith as are the “truths” of religion. But if that were the case, why does religion inevitably bow before science?
Li, H. and R. Durbin. 2011. Inference of human population history from individual whole-genome sequences. Nature 475:493-497.