A show of "no support"

5 minute read

In an article about the controversy over the Liang Bua fossils, Rex Dalton of Nature inserts this non sequitur:

Somewhat scientifically isolated, the 75-year-old Jacob has continued to adhere to the theory that hominin species from multiple regions around the globe evolved into a single line that produced modern H. sapiens. In this theory, Neanderthals were direct ancestors of modern humans, but most studies have shown that they were not related to </i>H. sapiens</i>. The 'multi-regional' theory has virtually no support among the world's leading anthropologists (Dalton 2005:433).

Those knowledgable about the field will recognize that this comment is startingly false. One may dispute whether this scientist or that one is a "leading anthropologist" but there is a better way to address the question: simply do a citation survey. My own informal survey of citation rates indicates that the top two paleoanthropologists in terms of citations both are public supporters of a multiregional model. The next few have written mainly about australopithecines, with varying opinions about modern human origins. (UPDATE: A later post has provided a top ten list of citation rates. The top two are no longer the same as indicated here -- formerly they were Wolpoff and Trinkaus, who are now 3 and 10 on the list -- and I make no claims about the opinions of the current leaders about modern human origins).

Now suppose we extend the frame to encompass the top 100 or so paleoanthropologists. Does anyone seriously think that this does injury to the representation of the multiregional perspective?

OK, but suppose we consider only those that have their work published in Nature. After all, we can presume safely I think that Dalton reads his own journal. And it is true that scientists that have been asked to comment on new finds within the past five years have been supporters of a more speciose, bushy taxonomy for the hominids. But this is partly because of the recent finds themselves, like the Lomekwi hominids, that invite the splitter-vs.-lumper dialectic. Among research articles we find not only Tim White and colleagues' (2003) Herto paper arguing for a recent African origin, but also Alan Templeton's (2002) paper supporting multiregional evolution from a genetic perspective. In short, there is no single perspective that dominates the field today, and I think that is a very good thing. We do not need comments that appear to stifle honest disagreement or marginalize those who remain skeptics of even widely-held notions.

I have addressed elsewhere the idea that the Liang Bua skeleton somehow addresses the modern human origins problem. I concluded, and still believe, that it makes virtually no difference one way or another. Clearly if a small population of ancient humans became isolated 900,000 or more years ago and no longer exchanged genes with other human populations, its evolution can say nothing about the possibility of genetic exchanges across the continental areas of the Old World. Multiregional evolution simply has nothing to say about this particular problem

But it cannot be denied that people like Teuku Jacob, Alan Thorne, and Maciej Henneberg who have expressed skepticism over the original interpretation of the skeleton have elsewhere written in support of a multiregional model of human evolution. Some, like Dalton, have argued that this shows that the multiregional model is threatened by the possibility that Homo floresiensis was a real dwarf species of hominid.

I disagree. In my opinion, the reason is that some scientists treat variation more seriously than others. The real split in the field is not between lumpers and splitters or between multiregional evolution and a recent African origin. The real split is between populationists and typologists. Not all populationists are multiregionalists, but the hypothesis is more often accepted by those who consider variation within populations as an important product of evolution. A populationist is less likely to accept the idea that a singular fossil is representative of its population (or species), because no single individual can be representative of a range of variation. It is that perspective that leads to skepticism about the Liang Bua fossil.

The variability within species of hominoids is often underestimated. For apes, this is often because researchers have only examined a small number of specimens in detail. Most research studies that include extant apes as comparisons use fewer than 30 specimens. Moreover, even very extensive skeletal collections of apes, like that at the Cleveland Museum of Natural History have biases that may underrepresent variability. Most of the specimens in that collection were wild-shot, and among the gorillas males are highly overrepresented. It is plausible, even likely, that there is a bias among these males for large, healthy specimens that would fetch a good price. Paleopathology is very different in an ape context than in a human context. Among apes, serious pathologies are generally fatal early in life, and we do not find them in skeletal collections. So from that perspective, apes make a poor comparative model for early humans.

For humans, there is not only the problem that collections are often small and geographically represent only a small area, but also that pathologies may have been specifically excluded. For example, the Hamann-Todd collection at the Cleveland Museum of Natural History consists of a large proportion of indigents whose bodies were assigned to the collection by the state. In various ways, these people were not representative of the Ohio population in health and other characteristics, but for our present purposes the most obvious problem is that microcephalics do not tend to die in circumstances that would get them into this collection. Pathological specimens do end up in medical pathology collections, but these are in many instances separated out from the normal variation within their populations.

Thus, few biological anthropologists gain a full understanding of human variation including pathological variants. No one knows to what extent the proportion of congenital or developmental defects may have been in early human populations that differed from ours in environment. For that matter, no one knows how a pathology present in today's humans may necessarily have manifested in a Stone Age human lacking today's medical interventions, nutrition, and lifestyle. These are some of the reasons why paleopathology is such an interesting and problematic research field.

None of this proves that the Liang Bua specimen does not represent a new species of hominid. But they do give reason for cautious skepticism.

More on the Flores hominids



Dalton R. 2005. Looking for the ancestors. Nature 434:432-434. Nature online