Out of Africa: by the skulls?

Update: John Hawks weighs in. Here is the abstract.

Several people have asked about a new paper coming out that uses the diversity in skulls to "prove" the Out of Africa hypothesis. The paper is going to be out in Nature yesterday. Yes, you read that right, it was supposed to be on the site on the 19th, but it still seems embargoed. But here is the headline from ScienceDaily, New Research Proves Single Origin Of Humans In Africa. Press releases are generally a little inflated, so no worries.

The basic gist is that the authors used the variation in skulls to trace population bottlenecks due to migration. Generally a good assumption is that the region which exhibits the most variation is the region from which the population expanded, as expansions by their nature tend to result in loss of genetic information via bottlenecks. Just as much of the genetic data suggests Africa tends to have more variation, implying that the longest lineages in this region, so the variation in skull morphology points in the same direction. Does this "prove" that humans emerged out of Africa? Historical sciences are more about inference than "proof," so I think the terminology is rather strong. One could posit, for example, that there were powerful selection pressures in Eurasia which resulted in the homogenization that we see. But in any case, I tend to lean toward the opinion that the preponderance of evidence does imply an overwhelming demographic expansion Out of Africa in the past 100,000 years. Nevertheless, I must object a bit to the framing of the issue in such a manner:

Competing theories on the origins of anatomically modern humans claim that either humans originated from a single point in Africa and migrated across the world, or different populations independently evolved from homo erectus to home sapiens in different areas.

As I have argued before the "either"/"or" formulation results in the discarding of a more subtle and nuanced explanatory model: though total genome content is overwhelmingly African and the general balance of characters are African-derived, that does not negate the importance of hybridization events leading to the introgression of significant genetic elements from non-African lineages. The whole genome is not created equally, and our weighting of significance is highly contextual (there are a host of necessary & sufficient parameters). Ancestry, roughly the phylogenetic data, is important and critical toward general our sense of the past. But though a gestalt model of the histories of all the genetic elements serves as the essential framework, the individual histories of the genetic lineages must not be elided. I think a strong case can be made that some of the genetic lineages exhibit histories which imply that they are not parsimoniously explained by derivation from a recent African demographic expansion, but rather, were absorbed from local populations which were marginalized in all other ways (i.e., they went extinct as a population).