Sunday, August 19, 2012

A Question for Anthropologists

This year's special issue of Scientific American is "Beyond the Limits of Science." One of the articles is about human evolution. The title is Super Humanity in the print issue but the online title is Aspiration Makes Us Human.

The author is Robert M. Sapolsky, a professor of biology and neurology at Stanford University in California (USA). Stanford is a pretty good school so he probably knows his stuff.

Here's how Sapolsky starts off ...
Sit down with an anthropologist to talk about the nature of humans, and you are likely to hear this chestnut: “Well, you have to remember that 99 percent of human history was spent on the open savanna in small hunter-gatherer bands.” It's a classic cliché of science, and it's true. Indeed, those millions of ancestral years produced many of our hallmark traits—upright walking and big brains, for instance.
This doesn't make sense.

Let's assume that our ancestors left Africa only 50,000 years ago. If that represents 1% of our evolutionary history then it means that our species and it's immediate direct ancestors lived on the African open savannah for 4,950,000 years.

Could that possibly be true even if you only count the main line of descent? What is the evidence that supports these claims? How much of the early history of Homo sapiens was influenced by adaptation to open savannah? Does anyone have a scientific answer to this question?

Setting aside the "main line," we now have good evidence that modern Homo sapiens acquired alleles from Neanderthals, Denisovans, and, perhaps more ancient Homo erectus. All three spent substantial time evolving in places that looked nothing like the open savannah in Africa. The proportion of the "invading" alleles may be only 10% or less but that's still significant.

Do we know for sure that all of the important features of modern humans came from alleles that were fixed by adaptation on the savannah? What if some of the more important behavioral alleles came from Neanderthals and became fixed because they were so much fitter than the savannah alleles?

Take the alleles that make women like to shop, for example. Maybe they arose in the Denisovans because they have access to better trade routes in central China? Maybe the women on the savannah preferred to store their cash in elephant tusks?

The evolutionary psychologists have developed awesome explanations for human behavior based on their detailed understanding of the social structure of hunter-gatherer groups living on the savannah for millions of years. What if our genetic ancestors lived elsewhere? The bad news is that all those just-so stories will be wrong. The good news is that they can publish a completely different set of stories and get twice as many publications.


  1. Paleontologist (not an anthropologist) here. You are correct Larry. The African savanna statement is an oversimplification of a pre-settlement and pre-agricultural past. Even if we restrict our view to Africa, there is evidence of hominins in woodlands and more forested environments.

    (Savanna misspelled as "savannah" in places.)


    1. Savannah isn't a misspelling:

    2. It is a misspelling if you start off "savanna" and then drift into "savannah".


    3. There is also evidence for coastal dwelling in along the beaches and in caves in South Africa. This could have been the location for the allopatric speciation of sapiens from an archaic Homo species

    4. Thanks for pointing out the inconsistency. I corrected everything I said to "savannah," which is the correct Oxford English dictionary spelling.

  2. If important genes came from Neanderthals and Denisovans, how many of those genes got back into african populations? Somebody who had bad will would make the argument that you are being racist, that you are saying people of african descent are inferior. Obviously you aren't, but that conversation is poised waiting to be started like a beartrap in your path. How can we discuss this sort of thing without triggering it?

    Your basic point is a good one. Once humans reached the point that we could exploit lots of different biomes, we did. Pygmies adapted to dense scrub. Lots of people adapted to lots of things. How much of it was savannah? And how long ago did we reach the point we could do that?

    How much of africa was savannah over the last million years? Probably sometimes more, sometimes less.

    Suppose for the sake of argument that we expanded into lots of niches around 100,000 years ago. That might leave 900,000 or so years when humans and prehumans were living in tiny populations trying to adapt to a single biome. Was it savannah? There's a fairly logical argument that it was. Compare humans to chimpanzees and they are adapted to a life we cannot manage. We are not good at brachiating and we lack hands on our feet etc. Why would monkeys come down from the trees and live on land? Why would they develop our extreme adaptation to hot dry weather? If the climate changed and the only thing we had was open savannah, then we had to live there and adapt. It makes a kind of logical sense in the absence of any real evidence.

    That woman who invented the "aquatic ape" theory had evidence about that good. Maybe better. It was real obvious that her evidence was not very good, but the standards should have been pretty low considering the competition.

    So I think this is a topic that has room for some evidence. Anybody who can estimate paleoclimates and give us an idea what africa was really like over the last million years, should be able to get some publications out of it. There's surely fossil evidence about what the climate was like in the times and places that human fossils have been found, and that's worth at least a good review paper.

    If we could get past the PC horror somebody might manage to actually test what sorts of environments different modern people can handle without much technology. It seems obvious that black skin is better for people who spend time in direct sunlight, though I haven't heard it was ever tested. Conversely, would people with white skin have other adaptations for deep forests, the kind of place where it's generally hard for mammals to make a living? I once read a JustSo story that claimed white skin was an adaptation to ice ages, that people who had to dress up in furs all the time could make enough vitamin D from just their exposed faces. I thought a science fiction writer came up with the story, but maybe he was just copying somebody else.

    There ought to be some papers in all this for people who actually collect data and guess at what it means.

  3. I guess I'm missing something here but it was my information that it was the Australopithecines who adapted to the savannah via bipedal locomotion.

  4. The test of this is simple: find high Fst alleles that show evidence of strong selective sweeps. Count them up... you'll find between 30 and 100 of them (maybe as many as 400, if you count non-coding regulatory regions). Not nearly enough to account for the variation in human behavior.

    The narratives of evo psych are so seductive that we should inherently distrust them. The narrative must arise from the data, not the other way around, or we risk confirmation bias.

    1. Nat Genet. 2008 Mar;40(3):340-5.
    "Natural selection has driven population differentiation in modern humans."

    2. Ann Hum Genet. 2008 Jan;72(Pt 1):99-110.
    "Identification and analysis of genomic regions with large between-population differentiation in humans."

    3. Bioinformatics. 2006 Sep 1;22(17):2122-8.
    "A whole genome long-range haplotype (WGLRH) test for detecting imprints of positive selection in human populations."

  5. Also, am I the only one who saw Sapolsky's picture and thought Larry accidentally put an anthropologist's name over the picture of some sort of early hominid recreation? He looks like living evidence of Neanderthal admixture.

  6. Technically...he never said..." ONLY on the open savannah..." If we consider that there may even be small groups of hunter gatherers on the open savannah today...(there certainly were 50 years ago), he could even say for 100% of human history and be correct. Whether or not it's a chestnut might be debatable.

  7. The savanna idea is based on a few misconceptions, but all fossil, paleo-environmental, physiological etc data show that human ancestors during the Ice Ages (Pleistocene Homo) did not walk or run over open plains (sweating water + salt = scarce in savannas!) as often assumed in popular & sometimes even "scientific" views on human evolution, but simply followed the African & Eurasian coasts & rivers (1.8 mill.yrs ago at least as far as Java, Georgia, Algeria & Turkana), beach-combing, diving & wading bipedally for littoral, shallow aquatic & waterside foods (recent info e.g. google researchGate marc verhaegen). Only a few remote human populations today live in dry savannas, and this is a recent innovation: we need lots of water, are much too fat & slow etc. For a disproof of the savanna idea, see e.g. our recent publications, e.g. Marc Verhaegen & Stephen Munro 2011 "Pachyosteosclerosis suggests archaic Homo frequently collected sessile littoral foods" HOMO J.compar.hum.Biol.62:237-247.