But actually, the story is a little more complicated, and a lot less contradictory. Taken together, the two studies push us to make better social sense of what usually gets reduced to a purely biological issue, raising the question, "who would a neanderthal/human love child date?"
Back in August, a paper by researchers at Cambridge University poked a big hole in arguments that early modern humans and neanderthals shared genetic material due to sexual liaisons.
The challenge: to explain why analysts find 1% to 4% of DNA is shared between these two groups. There are really only two options: inheritance from a common ancestor; or inheritance from parents engaged in sexual exchanges across species boundaries, or hybridization.
The Cambridge study, published in Proceedings of the National Academy of Sciences, didn't actually reject the possibility of hybridization between early modern humans and Neanderthals. Instead, the authors urged researchers to be explicit about "spatial population structure". Instead of just comparing two species (or subspecies) to each other, we should construct a model of the distribution across space of the local populations that could have been engaged in sex.
To quote the press release describing the Cambridge research:
Neanderthals and modern humans once shared a common ancestor who is thought to have spanned Africa and Europe about half a million years ago. Just as there are very different populations across Europe today, populations of that common ancestor would not have been completely mixed across continents, but rather closer populations would have been more genetically similar to each other than populations further apart.
In the demographic model the Cambridge research team constructed, each small population could only interbreed with its closest neighbors; so if hybridization were occurring, it would be in localized networks or clumps, not uniformly across the entire spatial distribution of the two species.
The researchers found that their model could produce the results seen in genetic studies of early modern human/neanderthal DNA, meaning that inheritance from a distant ancestor could be responsible for the observed patterns.
Fast forward to a new study publicized this past Friday.
In Mexico, there are two populations of howler monkeys that are genetically distinct species (Alouatta palliata and Alouatta pigra) that can interbreed and hybridize. Researchers assessed genetic relationships among howler monkeys in the contact zone -- precisely the kind of "spatialized population" that the Cambridge University study says we need to think about for ancient human lineages.
The paper, published in the American Journal of Physical Anthropology, reports that hybrid offspring of cross-species sexual pairings in turn had sexual relations with individuals of one of their parent groups, producing second- and further generation offspring closely resembling the species that their hybrid ancestors lived among and bred with:
Hybrid zones could comprise a large number of multigenerational backcrossed hybrids that are indistinguishable from the parental species.
Hybrid babies grew up to mate with members of the population they lived among, "backcrossing" so much that eventually, their offspring were "indistinguishable" from the local population.
To return to the question I posed above: who could the neanderthal/human love child date?
Growing up among relatives from one of the parent populations, these rarities would leave behind a stronger trace of interbreeding than their descendants, whose other ancestors belonged to the species of the local population.
Our basic challenge is that we do not, in studies of fossil populations, get to analyze all the individuals who once lived; we just have the ones that happen to be preserved and recovered archaeologically.
As Mary Kelaita and Liliana Cortés-Ortiz, co-authors of the new study, put it, the fossil record "represents only a snapshot in prehistory". We don't get to choose who is in the picture. How many first-generation hybrids have we analyzed; how many multi-generation backcrossed descendants of hybrids?
Kelaita concludes that
"the lack of strong evidence for hybridization in the fossil record does not negate the role it could have played in shaping early human lineage diversity."
But oh, that double negative: "does not negate" proved a bit confusing for the broader press.
Science Recorder had to change its original headline from "Another study finds ancient human interbreeding is more myth than fact" to "Study finds ancient human interbreeding may be 'underestimated' in the human fossil record". Quite a whiplash factor there.
Other media tried make the story into something simpler, emphasizing morphology as the supposed source of the problem: "the bones left behind by the two species [sapiens and neanderthal] don't bear any obvious traces of interbreeding".
This line of argument does start with statements by the researchers. In a press release, Cortés-Ortiz is quoted as saying
"The implications of these results are that physical features are not always reliable for identifying individuals of hybrid ancestry. Therefore, it is possible that hybridization has been underestimated in the human fossil record."
In the abstract for their article Kelaita and Cortés-Ortiz emphasize that their results
suggest that primate natural hybridization studies should incorporate molecular methods for determining ancestry, because morphology may not always be a reliable indicator of hybrid status.
But of course, contemporary arguments for and against interbreeding between early human lineages rely less on morphology and more on DNA evidence.
The Cambridge University team called for modeling local populations rather than merging continent-wide variability. The howler monkey study suggests we need to include a wide sample in such localized models. and remember that they are only snapshots of developments unfolding over time.
Not quite throwing a monkey wrench into the works, then: maybe providing us a flexible tool to help with a complex problem. A monkey wrench, in fact.