Newsgroups: comp.ai,comp.ai.philosophy,alt.consciousness,comp.ai.alife
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!news.mathworks.com!gatech!swrinde!cs.utexas.edu!utnut!torn!watserv2.uwaterloo.ca!undergrad.math.uwaterloo.ca!watdragon.uwaterloo.ca!bpvanstr
From: bpvanstr@yoho.uwaterloo.ca (Brian Van Straalen)
Subject: Re: Darwinian selection for intelligence
Message-ID: <D93CBK.90K@watdragon.uwaterloo.ca>
Sender: news@watdragon.uwaterloo.ca (USENET News System)
Nntp-Posting-Host: yoho.uwaterloo.ca
Organization: University of Waterloo
References: <3nh78b$o65@nntp.Stanford.EDU> <JASON.95May9125127@wratting.harlqn.co.uk> <1995May12.220833.19334@aw101.iasl.ca.boeing.com> <JASON.95May22133315@wratting.harlqn.co.uk>
Date: Wed, 24 May 1995 16:26:06 GMT
Lines: 55
Xref: glinda.oz.cs.cmu.edu comp.ai:30113 comp.ai.philosophy:28316 comp.ai.alife:3492

In article <JASON.95May22133315@wratting.harlqn.co.uk>,
Jason Trenouth <jason@harlequin.co.uk> wrote:
>
>jason@harlequin.co.uk (Jason Trenouth) writes:
>
>Jason> Would large-scale very-long-term persecution have similar
>Jason> Darwinian effects on intelligence. E.g. If Jewish people are
>Jason> smarter on average, are they smarter because of
>Jason> pogroms/holocausts?
>
>Steven> Let's clearly distinguish between cultural evolution (2-10
>Steven> generations) and biological evolution (minimum 10,000 years).
>
>It sounds like you are quoting these figures from somewhere. Are they
>established statistics? Why are they in two different units? And
>surely different biological features evolve at different rates in the
>same population?
>
>A human generation is generally reckoned to be 30 years, so you are
>saying that it takes a minimum of 300 or so human generations for any
>human biological evolution to take place. This figure sounds a bit
>large.

It does seem too large.  There has been a recent upshoot in the theory
of micro-evolution (in conjunction with gradual evolution theories):
 Where measurable physiological changes occur in just three
or four generations (Experimental evidence circa 1993 from , where else,
Darwin's finch populations).  The type of physical attribute which
can evolve in this fashion has not been pinned down (They were studying
beak length, shape, feather size, toe angles etc.), some attributes seem
resistant to evolution , others quite pliable.  Speculation would
suggest that attributes with large population variance are faster
to evolve (There are doubtless other factors involved). An example for
this, for humans, would be height.

Fruit fly experiments are starting to indicate that the number of
cooperating genes in the genotype that result in a measurable phenotype
might indicate evolutionary pliability.

The Jewish example mmight be extreme to cite, since most the last
persecution only modified a single generation.  But, as a race, they
have attempted to minimize mixing of their gene pool with gentile genes.
That would make their effective population smaller, hence, fewer
generations would be required to establish a genetic norm shift.

Regardless, the two attributes that might effect a fast rate of evolution
(at least, the two being studied), are characteristics associated
with intelligence (wide population variation, many inter-related genetic
factors).  Maybe the long held belief that we could have taught early
homo sapiens to read and sing isn't all that well founded. Maybe we're
all considerably smarter than our ancestors.  Maybe genetic approaches
to AI should consider these points.



