by Greg Mayer
In a recent paper in Nature (abstract only), Tom Higham of Oxford and several colleagues report on their effort to determine by radiocarbon dating when Neanderthals went extinct. Higham et al. conclude that it was about 40,000 years ago. It’s gotten a fair amount of media coverage—more on this below—but let’s look at the science first. What’s most interesting is that they strove very hard to get accurate dates not biased by contamination of their samples by younger carbon (developing new and refined methods along the way), and that they sampled a large number of sites across (mostly Western) Europe. Here’s the basic result.
You can see that latest dates range from about 49 to 40 KYA, with a joint estimate of the end of the Mousterian culture at about 40 KYA. There are a few caveats. First, there’s no explanation in the paper for why there are no dates in panel (b) for the 7 southern Iberian localities. These sites are of special interest, because it has been argued in the past that the latest survival of Neanderthals (ca. 35 KYA) was in southern Spain. The Spanish localities are mentioned in the 160+ page supplement, and some are said to not have produced reliable data, but others did, and, maybe the answer’s buried somewhere in the enormous supplement, but I could not readily locate it. (This by the way, is yet another example of the bad practice, characteristic of Science and Nature, of having extremely short papers with monographic online supplements that contain not just the details, but critical parts of the work. If your work is that substantial, then you should publish a monograph, not a tiny summary in Science or Nature.)
Second, many of the sites have little in the way of human remains, so the datings are of a particular cultural style, and the associated type of human is assumed (Neanderthal in the case of the Mousterian), although on fairly robust empirical grounds.
And third, the geographic sampling is sparse outside Western Europe. A claimed late refuge in the Arctic, for example, was not sampled. (There was a very late refuge, until historic times, for mammoths in the Arctic.)
What about the media coverage? It’s been very confused– see examples here and here. Media reports hail the work as showing that Neanderthals went extinct earlier than previously thought; and that we now know Neanderthals and anatomically modern humans overlapped significantly in time, thus allowing opportunities for the genetic mixing that has been now well documented (and much discussed here at WEIT). But these two claims are contradictory– earlier extinction means less temporal overlap; and the second thing is something we’ve known for quite awhile.
So what are we to make of the media claims? Well, the new work doesn’t say much at all about genetic mixing– it occurred whether Neanderthals were all gone by 40 KYA (as this latest work proposes), or survived in Spain till 35 KYA (as some earlier authors had claimed). Higham et al. estimate that there was an overlap of several thousand years of Neanderthals and anatomically modern humans, plenty of time for interbreeding. If the Spanish localities really are a late survival of Neanderthals, that would just add a few thousand more years of opportunity. Now, it will be of great interest to learn (if we can) exactly when and where the interbreeding occurred, but the new paper just adds constraints to the timing– it doesn’t suddenly make interbreeding now seem plausible.
I normally go see what John Hawks has to say about paleoanthropological matters, especially as in this case, since I felt perhaps I was missing something. I looked, but he hasn’t posted in a few weeks– he must be on vacation. I expect he’ll have something to say when he returns.
Higham, T et al. 2014. The timing and spatiotemporal patterning of Neanderthal disappearance. Nature 512:306-309. (abstract only)
h/t Barry Lyons