That's 2 for 2 for Denisova surprising us: its mtDNA, reported earlier this year is discordant with the autosomal evidence: the Denisovans trace their mtDNA ancestry to a much earlier time than the division of humans and Neandertals. And, yet, the Denisova remains are about 50,000 years old, i.e., at a time close to the appearance of modern humans in Asia.
Who would have expected a fairly young specimen in South Siberia to have really archaic morphology and mtDNA, to tie itself to West Eurasian Neandertals on the one hand and to modern Melanesians on the other?
The other big paper of the year was the Neandertal one which tested Neandertal DNA from another cave, Vindija in Croatia (image on the right: Johannes Krause, Max Planck institute of evolutionary anthropology) threw another curveball at theories of human origins. Prior to the appearance of this paper there were two theories about possible Neandertal admixture in humans: one that there was none, and another that it might have affected West Eurasians (where Neandertals lived) and could be detected in either their genes (stories about the "red hair gene" that circulated a few years ago), or the morphology of late Upper Paleolithic anatomically modern humans from Europe.
And, yet, the Neandertal paper gave a startling conclusion: Neandertals were more related to non-Africans in general than they were to Africans. This evidence must be explored further: how is affinity to Neandertals distributed within Africa itself? Within Eurasia? In America?
No one would have expected this outcome: that people in the far east would be at least as related to Neandertals as Europeans are.
Speaking of the far east, it was another cave, Zhirendong that threw the third curveball on mainstream theories of human origins. Here we have palaeoanthropological remains that look especially modern, but with some archaic traits as well.
The story so far had been of a recent African Exodus of anatomically modern humans 40-60 thousand years ago. The presence of anatomically modern humans in other caves in the Levant such as Qafzeh around 100 thousand years ago was acknowledged, but it was usually brushed off, as the Out-of-Africa-that-failed, an early settlement of modern humans outside Africa that contributed little to the later pulse of humans that went on to populate the world.
The fossil evidence of Homo sapiens and pre-existing humans is notoriously difficult to make sense of. But, it certainly becomes more difficult to view Qafzeh as an experiment-that-failed when a similar mix of modern-and-archaic features can be found from roughly the same time all the way to the east in China.
Interpretation of these new findings will take some time, and the fact that we're getting regular DNA reads from samples that are a few myriad years old suggests to me that we should expect more surprises to come. I've voiced my opinions on many of these papers, but I'm ready to be surprised.
It was only a few decades ago that the notion of a common recent African descent of mankind was surprising and revolutionary (thanks to mitochondrial Eve & co.). The pendulum has begun to swing the other way, with all sorts of dead folks that were once considered to be irrelevant side-branches re-asserting themselves as our ancestors or kin.