[ExI] Ants for spike! Dawkins and Wilson

[Keith Henson]

On Sat, Nov 14, 2009 at 4:00 AM,  <spike> wrote:

snip
'
>> URL for where you got these numbers please.
>>
>> Keith
>
> No references Keith, I just offered of them as an example.  I should have
> worded it differently, such as "If, for instance, 94% of..."
>
> I pulled these numbers out of the air to suggest perhaps it isn't exactly
> right to say evolution does not work at all on the group level.  I agree it
> works primarily at the gene level, over 90%, but it is important to
> understand whether or not there is a small percentage of genetic change as a
> result of group selection.  Most theorists today agree there is no group
> selection in evolution, but a few say group selection sometimes plays some
> small part.

Group selection for humans cannot logically happen because human
groups swap women around.  Biological evolution is changes in gene
frequencies.  If due to some group selection effects a group has an
accumulation of advantageous genes in their gene pool, it is lost to
nearby competing groups as the women take the advantageous genes with
them when they are swapped.  QED.

> It might be that group selection applies only to those lifeforms in which
> memes dominate behavior, such as in humans.  We are seeing a clear case
> where the meme "breed early and often" is causing dominance of the genome of
> that group which holds these memes, over the group which holds the meme
> "breed seldom if at all, in order to save the earth."

Current conditions are not typical of human evolution in the past or
likely to be typical of the future.

I used to be worried about humans breeding themselves into a state of
stupid.  Then I finally realized that unless a trait is changing due
to current evolution, you expect both sides of the bell curve to be
trimmed equally.  So if the Mensa types have one kid or fewer, the
people that many standard deviations below normal IQ don't have many
surviving kids either.

> The reason why I think this is important is that I perceive we really are on
> the verge of the holy grail of modern evolutionary theory: a working
> computer simulation of evolution.  If we are within a decade or two of that
> breakthrough, then we need to program in that factor of group selection, if
> its influence is non-zero.  I suspect the influence of group selection is
> usually flat zero, but in a few species, such as humans, under a few
> circumstances, such as war, group selection goes non-zero.

You don't need group selection to account for the psychological traits
that cause humans to go to war under some conditions.  Inclusive
fitness is enough.  The basics for such simulations already is clear
from EP.  See  http://lists.extropy.org/pipermail/extropy-chat/2009-July/052083.html

Keith