UK United Kingdom

Sex with our evolutionary ancestors? Proceed with caution

A paper published last week in the Proceedings of the National Academy of Sciences provides a compelling alternative to the idea that we Homo sapiens interbred with Neandertals or Denisovans as had previously…

New evidence has cast doubt on one of the accepted truths of archeaogenetics.

A paper published last week in the Proceedings of the National Academy of Sciences provides a compelling alternative to the idea that we Homo sapiens interbred with Neandertals or Denisovans as had previously been thought.

As such, it’s a clarion call to anthropologists to be a lot more sceptical about the claims made in the fledgling field known as archaeogenetics.

It’s an uneasy time to be practising anthropology. Our scientific understanding of human evolution is in a wild state of flux. Many long-held notions are being rapidly discarded for radical new ones that were unimaginable merely a decade ago. The winds of change seem to be sweeping through our science so rapidly they’ve caught many of the old guard off-guard.

But what has caused this change? Are we really in the midst of a revolution in our understanding of hominin evolution?

Or does the seemingly radical nature of recent discoveries simply reflect the kind of evidence we rely upon in anthropology, as well as our ongoing, imperfect knowledge of human origins?

Old evidence

Much of this radical new understanding comes from fossil discoveries. The human fossil record has always held surprises but some of the finds have been so challenging – inconceivable after 150 years of human evolutionary understanding – they’ve been met with cavernous scepticism.

Science is conservative by design, and scepticism is vital. But there comes a point in the timeline of any discovery where the weight of evidence can no longer be ignored.

Take the example of Homo floresiensis (or the “Hobbit”), announced in 2004. This is a chimp-sized human, with a brain the size of a grapefruit, and a skeleton so primitive it could have been found in the African savannah 3m years ago.

Yet the bones of the Hobbit could be as young as 17,000 years, the species thriving on the remote Indonesia island of Flores, thousands of kilometres from the Great African Rift.

Should Homo floresienis really have been such a big surprise?

Anthropologists have always expressed great and apparently genuine surprise when something unusual has been found. Take the discovery of the “Black Skull” (fossil KNM-WT 17000) from the western shores of Lake Turkana in northwest Kenya in 1985. It was apparently unimaginable on the basis of current knowledge at the time, as with the Hobbit.

The details of why this was the case are not so important here, other than to say it forced a major reconfiguring of the human evolutionary tree. The fact it delivered a fatal blow to several long-held scenarios also meant it ruffled the feathers of quite a few anthropologists who held such scenarios dear to their hearts.

In an article published in Science by British science writer and journalist Roger Lewin at the time anthropologists were quoted as saying: “whichever way you look at it, it’s back to the drawing board” and that its implications “may be a lot for some people to swallow”.

Revolutionary Hobbit

At the time the Black Skull was announced, Graham Clark, an archaeologist at Arizona State University, wrote an article in the journal American Anthropologist stating that, if the Black Skull was really so unexpected and so radically changed our view of human evolution, “it would indicate an exceptionally low level of conceptual and theoretical development in that field".

He made a compelling point. But it’s not one that resonates with the discovery of the Hobbit.

The former was geographically, temporally and anatomically predictable at the time of discovery, as Clark emphasised in his article. The hype surrounding the discovery of the Black Skull, Clark contended, had more to do with personal desire for publicity and the “never-ending quest for funding” than the reality of the discovery itself.

But Homo floresiensis seems to me to be a genuine example of American historian and philosopher Thomas Kuhn’s thesis about the revolutionary nature of scientific change.

An awful lot of science, including anthropology, doesn’t lead to rapid, radical revision. But the Hobbit did.

New kinds of evidence

Two other radical discoveries in the past five years have made similar waves in the science of human origins. But this time, they came from the fledgling field of archaeogenetics: the recovery and analysis of DNA from ancient skeletons.

The sequencing of the Neandertal and Denisovan genomes led to the conclusion that our early modern human ancestors interbred with these non-modern hominins when they colonised Eurasia after 100,000 years ago.

Even more radical, the genes we inherited from these archaic hominins were claimed to have been a key to the success of our modern ancestors in the harsh Ice Age conditions of Eurasia.

Back to the study

The Proceedings of the National Academy of Sciences paper suggests we need to subject claims made by archaeogeneticists to more scrutiny in the future.

The study, by Anders Eriksson and Andrea Manica, offers a simpler “null hypothesis” in place of the idea Homo sapiens, Neandertals and Denisovans interbred. It’s a hypothesis that can’t, at present, be rejected.

In short, their idea is not only intuitively appealing, it fits with emerging understanding of the human fossil record, as well as knowledge from many other species.

They suggest:

  • the earliest modern humans in Africa would not have been genetically homogenous
  • that some populations probably retained many more “old” DNA segments than others who lost them through chance events during their history
  • that some of those old DNA segments would have been shared with the Neandertals and Denisovans, because we share an evolutionary ancestor
  • that this perhaps creates the impression of interbreeding when in fact it might simply be that some populations retain more of these old DNA stretches than others.

Difficult lessons

The prevailing belief that interbreeding had probably occurred in geographic regions for which there are large gaps in the fossil record ought to have been a warning sign I guess, with hindsight. Among other things, it meant anthropologists couldn’t undertake independent testing.

Yet some geneticists contend that if interbreeding had occurred it was at such a low level it wouldn’t show up in the skeletons of early moderns humans or those of late Neandertals anyway. How could we have tested it then?

The main barrier for anthropologists scrutinising scenarios emanating from archaeogenetics is the vast intellectual gulf between our disciplines. Ancient DNA research is so highly specialised and detailed that it takes teams of dozens of geneticists to do the work.

How can we possibly penetrate this opaque intellectual world and cast a critical eye over such claims, even if they seem to fit uneasily with our understanding of human evolution from the fossil record? We can’t.

Some of my colleagues have simply embraced the idea of interbreeding for two reasons:

  • it suits their intellectual pedigree
  • they feel the evidence is so sophisticated, so compelling, that to not embrace it would seem “unscientific".

I guess I would count myself among the second group.

It’s a tough lesson to learn. Just how we proceed remains a huge challenge as science becomes increasingly specialised and technology-driven. But one thing we can’t afford to do is to suspend disbelief even in the face of seductive claims!

Join the conversation

11 Comments sorted by

  1. Lennert Veerman
    Lennert Veerman is a Friend of The Conversation.

    Senior Research Fellow, School of Population Health at University of Queensland

    Intriguing question and interesting observations. In their introduction, Eriksson and Manica write that there is no difference between Europeans and Asians in their similarity to Neanderthals. A good test for the intermixing hypothesis would be to assess the differences between Europeans and Asians in their similarity to Denisovans.

    If Europeans and Asians are equally (dis)similar to Denisovans, the intermixing would be unlikely unless one assumes that the three groups got together in the Middle…

    Read more
    1. Roberto Crispino

      Public Servant

      In reply to Lennert Veerman

      I would have thought that it would be Europeans who would likely have Denisovan traits rather than Asians given the range of the Neanderthals. Would it not be the opposite in that Europeans would have greater traits given that they would have co-existed with the Neanderthals for longer? If this is the case, then this would support the interbreeding hypothesis.

  2. Russell Walton


    Very interesting article. Shouldn't the title refer to "cousins" not "ancestors."

    DNA analysis hasn't settled the question as to how agriculture spread from Anatolia either, I was rather optimistic --must be the "CSI Effect".

    1. David Paxton


      In reply to Russell Walton

      Russell, I think that the original harvesting of grains from wild grasses proximal to human beaten paths would have changed them such that strains with more adherent seeds were naturally selected to predominate locally through natural selection over millennia . Our invention of food production(agriculture) is yet another hubristic myth, started by God in the Garden of Eden when He decided that a man's destiny would henceforth be to scratch in the ground. Please keep an eye out on-line for the proceedings of the Australian Institute of Animal Management conference to be held in Penrith in October 2012.

  3. Michael Bunce

    ARC future fellow researching in the fields of ancient DNA and wildlife forensics at Curtin University

    Hi Darren,

    Thanks for the article.

    You wrote “How can we possibly penetrate this opaque intellectual world and cast a critical eye over such claims, even if they seem to fit uneasily with our understanding of human evolution from the fossil record? We can’t.” – this statement is hard to fathom as it is a criticism that could probably be leveled at any area of science outside of ones own specialty field. You find it “opaque” as you do not have the expertise to analyse genome data or conduct…

    Read more
    1. Emma Anderson

      Artist and Science Junkie

      In reply to Michael Bunce

      Got anything on neanderthal or denisovan introgression of class II HLA allelles?

  4. David Paxton


    Another great conversation, Darren. Thanks. If I ever have the chance of another life I'll enrol in one of your courses. Your point about suspending disbelief is important. A solution for me is to retreat to the abstraction of Darwin's theory that species originate by means of natural selection. Then I don't have to tie myself in knots the way Ussher (and Lighfoot) did. While the idea of interbreeding with unappetising palaeolithiccousins has a certain matinee thrill about it, such inetraction…

    Read more
  5. David Paxton


    Sorry for the typos in my comment, especially to Russell Walton.

  6. emily vicendese


    "Kuhn failed, of course, to recognise that an awful lot of science including anthropology doesn’t lead to rapid, radical revision." Not true, I'm afraid. Kuhn recognised that most science was what he termed "ordinary science": science which carried on operating successfully within the established paradigm. Scientific revolutions are not ordinary science, they are the process by which we move from one paradigm to another.

  7. Roy Niles

    logged in via Facebook

    I've often wondered why science is less concerned with the evolution of intelligent strategies that gave rise to a variety of "human" forms than with the apparently more easily observable changes in those forms that were in general the result of needed changes in intelligent strategies.