Re: Group selection

From: Keith Henson (
Date: Sun 20 Feb 2005 - 16:56:04 GMT

  • Next message: Keith Henson: "Re: Group selection"

    At 12:00 PM 20/02/05 +0100, you wrote:
    >I am splitting this thread (What happened to the journal of memetics?)
    >because Keith Henson introduced a new subject.
    >Keith Henson wrote:
    > >I got this far and then took a look at your web page . . . .
    > >It might look like group selection happens, but it is much better
    > described by selfish genes and Hamilton's inclusive fitness criteria.
    >That's the predominant view among evolutionary biologists today, but I
    >disagree. Some simplified mathematical theories say that group selection
    >doesn't work, but real world observations seem to indicate the opposite.
    >That's why I am doing research on group selection, and I am trying to
    >refine the mathematical models.
    >The non-reproductive castes among ants and bees might possibly be
    >explained by kin selection because they have haplodiploid inheritance,
    >i.e. they have more genes in common with their siblings. But this
    >explanation doesn't work for termites, naked mole rats, and other social
    >animals. They have diploid inheritance, so they share only half their
    >genes with their siblings, and less with their half-siblings.
    >Kin selection theory says that I should help my brother if his gain is
    >more than the double of my costs.

    That's not the usual way it is expressed. You should be willing to *die* if doing so saves more copies of your genes in relatives than are lost in you.

    Now humans don't seem to have evolved relatedness detectors. So we go by proxies, those we grew up with we treat as between sibs and half sibs, those in the little tribe around us as between first and second cousins. For all of our evolution as hunter gatherers, this was a reliable way to estimate relatedness. Summing up all the relatedness in a tribe and setting the number of gene copies against yours indicates that we should be willing to die if (we think) it would save the rest of the tribe.* This is the origin of the psychological traits tapped in suicide bombers.

    >But this doesn't explain why we are sending money to starving children in
    >Africa and tsunami victims in Asia. You may say that this is because of
    >religious memes. This may be true to some extent, but atheists give to
    >charity too.

    We are stuck with psychological traits that were shaped by millions of years of being hunter gatherers where the people around you were mostly related. It is not surprising that we do things for others in the modern world that are not in our genetic interest.

    >Why do birds and many other animals have ritualized fights, and why do
    >they respect the outcome of the fight? A hungry bird that has lost the
    >fight for the best territories would be better off by not respecting the
    >territorial boundaries than starve to death.

    Dawkins expresses this better than I can, but fighting to the death over a breeding territory is fairly rare because doing so would miss the chance that a territory holder will die and an animal that didn't fight to the death can take it over.

    >There are so many behaviors among animals as well as humans that can't be
    >explained by kin selection and reciprocal selection.

    Reciprocal altruism I think you mean. As far as I know, when the EEA and the resultant environmental mismatches are factored in, I don't know of any behavior that can't be explained, even highly maladaptive ones. Perhaps you could provide a few examples?

    >This is why I am doing research on group selection. I have not published
    >very much yet, but you can see the most important results at

    You do some interesting simulations. I think you would find the simulations to be more in line with reality if you tried to model them on inclusive fitness rather than group, especially if you factor in the EEA.

    It would be interesting to do a simulation that captured the average relatedness of the members of small tribes during the EEA. Because they exchanged wives with related tribes, I think you would find that relatedness was relatively high, especially with a small founder stock. You might even get an optimal number for the size of a tribe.

    Keith Henson

    *There are cases in post-agricultural history where suicidal self-sacrifice made a huge difference. For example in 480 BCE Leonidas king of Sparta commanded 300 Greeks in one of the most important events in modern human history: the battle of Thermopylae A small force (eventually wiped out to the last man) held up over 100,000 Persians in a narrow pass for six days while the Greeks mustered the forces that eventually defeated the Persians. The achievements of Greece lie at the root of western culture. Without this sacrifice the world would be very different.

    =============================================================== This was distributed via the memetics list associated with the Journal of Memetics - Evolutionary Models of Information Transmission For information about the journal and the list (e.g. unsubscribing) see:

    This archive was generated by hypermail 2.1.5 : Sun 20 Feb 2005 - 17:22:08 GMT