In quite a few recent writings (e.g., Europe before Rome by T. Douglas Price or The Human Species by John Relethford) human biological prehistory is presented as a succession of species—Homo ergaster, Homo antecessor, Homo heidelbergensis, Homo neanderthalensis, Homo sapiens. The first three of these are groups I would lump into the broader category of Homo erectus, a conclusion supported by the recent report on the Dmanisi fossils by David Lordkipanidza.
While this proliferation of paleospecies might be useful for purposes of discussion, it makes no sense biologically and, ultimately, is almost certain to lead to confusion among those who don't have a fairly detailed grasp of evolution and genetics.
Paleospecies are defined on the basis of morphological differences. In other words, they are defined on the basis of features observable in the fossils. But we do not have nearly enough fossils to be able to say whether the range of variation of humans in the past was greater or less than at present.
In contrast, the most widely used biological definition of species states that two species are held to exist if two similar populations are unable to successfully interbreed to produce fertile offspring under natural conditions. Obviously, we can not observe this in prehistoric populations. However, it is possible to infer it. And in the case of all those species in the list above, we simply cannot make any such inference. At most we can argue that they should be considered subspecies of Homo sapiens.
Let me explain why.
The key process in speciation, the dividing of one species into two or more, is that of reproductive isolation. This is a process we do not understand very well at present, though it isn't simply a matter of geographical isolation. Rather it appears to involve an accumulation of genetic differences, some of which make it more difficult for two populations to successfully interbreed. This process may take hundreds, if not thousands, of generations before full speciation occurs. If we are going to argue for speciation among our human ancestors, we must be able to point to evidence for such reproductive isolation. At this point, we are unable to do so through genetic evidence because we really don't know what to look for. We could, however, argue for it on the basis of either clear evidence for geographical isolation over a very long period of time or differences in behavior observable either on the fossils or in the archaeological evidence that make interbreeding highly unlikely. Only in the case of the group called Homo floresiensis does it seem likely that gene flow was completely cut off through geographic isolation for a sufficient period to allow speciation.
In the view of those who accept the progression offered above, Homo ergaster appeared in Africa around 1.8 million years ago and migrated into Eurasia. In Europe, these migrants soon took the form of Homo antecessor, which in turn, through adaptation to the European environment, produced Homo heidelbergensis, which subsequently gave rise to Homo neanderthalensis. We thus have a proposed progression of four separate species. However, nowhere do we have evidence for speciation within Europe. Rather we have a simple accumulation of morphological differences in the fossil remains—and those quite minor until we reach Homo neanderthalensis. But we must also bear in mind that there can never be a point at which one generation belongs to one species while the next belongs to another. We thus have a continuum. And changes accumulate within any species.
It might be argued that speciation had occurred if we were able to show that a typical member of any of these paleospecies would have been unable to interbreed with any of its predecessors or successors. This, however, we cannot do. We can only show that there has been change over time. But the evidence is strongly against speciation.
An alternative approach to the issue of speciation would be to show that any of these stages in human evolution could not have successfully interbred with the humans remaining in Africa. Here we have an opportunity. It is widely agreed that the characteristics marking anatomically modern humans arose first in Africa and spread from there to the rest of the world. Therefore, we can argue for speciation, and the resulting validity of at least one of the species named earlier, if these anatomically modern humans could not interbreed with the surviving European humans, who at the time of the spread of modern characteristics were those peoples known as Neandertals.
For more than a decade, this is exactly the argument that was offered. Modern humans were said to have completely replaced the Neandertal populations, either through direct aggression or through out-competing them for scarce resources (though the low population densities projected for Europe at the time lead one to wonder why resources should have been scarce).
Now, however, it has been shown that most Eurasians carry from 1 to 4 percent Neandertal genes, particularly in our immune system or segments of the genome that affect the skin. Even more recently, it has been shown that people differ widely in what is included in that 1 to 4 percent. It has consequently been estimated that, overall, perhaps 20 percent of the Neandertal genome is preserved in modern humans across Eurasia. This is a very long way from supporting the idea that we belong to different species.
Also added to the mix in recent years is the problem of the Denisovans, offered as yet another species solely on the basis of a couple of bones whose genome does not match that of the handful of Neandertals whose genome has been deciphered nor most modern humans. We are confidently told, however, that the Denisovans are a population from Asia that did contribute to the genome of some modern groups from the western Pacific.
Thus, it appears that these predecessors to anatomically modern humans—both Neandertals and Denisovans--were quite capable of interbreeding with those carrying the modern genome, though fertility may have been somewhat reduced. Therefore, they must be considered the same species. And if they are, then so too must all those populations ancestral to them.
In other words, we know that differences had been accumulating in both the Eurasian and African populations from the time our earliest ancestors left Africa. But, because anatomically modern humans could successfully interbreed with both Neandertals and Denisovans, those changes had not accumulated to the point of speciation. According to an analysis of my DNA carried out by the Genographic Project, I am 2.2 percent Neandertal and 3.4 percent Denisovan.
The only viable conclusion is that, with the exception of Homo floresiensis, who seems to have been isolated from the rest of humanity for upward of 250,000 years, all human populations from the time of Homo erectus 1.8 million years ago must be considered members of the same species. Insofar as we can identify differences between populations, these must be at the subspecies level. So, we cannot speak of Homo neanderthalensis but only of Homo sapiens neanderthalensis. The same applies to all the others extending back to 1.8 mya. At best they are subspecies. and we could even argue for all those before the Neandertals to be included in a larger grouping, Homo sapiens erectus.
