Header image

john hawks weblog

paleoanthropology, genetics and evolution

Photo Credit: Excavation at Denisova Cave, Russia. John Hawks CC-BY-NC-ND

Naming archaic human populations

Ancient human populations like the Neandertals and Denisovans were not separate biological species.

I’ve written about that before, including in the months after the publication of the first Denisovan genome (“Is the Biological Species Concept a ‘minority view’?”).

The data now are even clearer. We are not talking about the hypothesis of rare hybridizing. Every hominin ancient DNA we have is part of some story of population mixture.

Now a slightly different problem arises. What do we call these ancient populations?

At the time the first Denisova genome was announced by David Reich and colleagues, in 2010, I thought that the idea to coin the name, “Denisovans” was a very good one. The name gave us a way to talk about this population without presupposing that they were a different species from modern humans or Neandertals.

But Denisovans were not alone in Denisova.

So now we face a Denisova Denisovan genome and a Denisova Neandertal genome. The Denisova Denisovan genome is part Denisova Neandertal. There are two additional Denisova Denisovan mtDNA genomes. For all we know, the nuclear genomes of those apparent Denisovans may be different mixtures of the Denisova Denisovan and Denisova Neandertal populations.

Of course, talking about “Denisova Neandertals” can do nothing but confuse everybody. Prüfer and colleagues avoided this by using the term “Altai Neandertals”.

They additionally invented the term, “Siberian Denisovans”, to account for the reality that there were once very different populations sharing some – but not all – the genetic makeup of the Denisova Denisovan genome. The “Denisovans” that mixed into the modern human population of Australia and nearby regions probably lived nowhere near the Altai. Such modern-human-mixing Denisovans were more different from the Denisova Denisovans than any but the most divergent groups of living humans.

I wonder: Why should we call them Denisovans at all?

Likewise, the Altai Neandertal was apparently more different from other Neandertals than present-day peoples of Asia are from any Europeans or Native Americans. Why should we call the Denisova “Neandertal” a Neandertal?

At the moment, there is only reason I can suss out: mitochondrial DNA. The known “Neandertals” are on one mtDNA clade, the known “Denisovans” on another. But that distinction will begin to fail as more ancient genomes are recovered. The Sima de los Huesos mtDNA already shows where this is going, as this Spanish specimen lies deep on a clade that includes the Denisova Denisovan mtDNA sequences. If the Denisova Denisovans got their mtDNA from a much more divergent archaic human population, as hinted by Prüfer and colleagues, then many closely related populations may have belonged to a very different mtDNA clade.

Living populations are complicated. Some populations have relatively close relationships from common ancestor populations, others are the products of recent population mixture, and still others have both patterns of ancestry. We don’t worry too much about this when we talk about groups like Brazilians, or Pima, or Wolof, because these groups have names already. They may be complicated, but we still can talk about them.

We do not yet have an adequate representation of the genomic variety of ancient people. What we see so far suggests they will turn out to be as complicated as recent populations. They don’t come with names attached.

The names we’re attaching now don’t seem like they’re quite up to the task. We need a new approach to lend consistency to naming without presupposing a species-level taxonomic solution.


Prüfer, K. et al. (2013). The complete genome sequence of a Neanderthal from the Altai Mountains. Nature (in press) doi:10.1038/nature12886