RE: Chunking intelligence functions (was Re: [Fwd: com-mensa-rate digest#1] )

From: Ben Goertzel (ben@goertzel.org)
Date: Wed May 09 2001 - 18:53:38 MDT


> Ben Goertzel wrote:
> >
> > This strikes me as a rather absurd statement. I grew up around
> academics
> > trained in the social sciences and I can assure you that they
> hold a very
> > wide variety of views -- much wider, for example, than the set of views
> > represented on this e-mail list.
>
> Why are we discussing this issue in general terms?
>
> Let's take a specific example: Margaret Mead. How many of the social
> scientists of your acquaintance regard her as (A) an important
> anthropologist whose work on Samoan customs helped to reveal Western
> customs such as sexual jealousy as cultural artefacts rather than inherent
> properties of human nature, or (B) one of the all-time rose-tinted
> screwups of cultural anthropology, whose work was disproved in toto by
> Derek Freeman (who spent six years on the project rather than twelve
> weeks)? If most social scientists you know think of Margaret Mead as a
> screwup and believe that sexual jealousy is an evolved human universal,
> then either my statement was unfair and stereotypical and I'm attacking a
> straw man in an already-won war, or there's a very strong selection bias
> in the sample group of "social scientists who Ben Goertzel feels like
> talking to".
>
> -- -- -- -- --
> Eliezer S. Yudkowsky http://singinst.org/
> Research Fellow, Singularity Institute for Artificial Intelligence
>

I don't actually know ANY social scientists who believe Margaret Mead's work
on Samoan sexual customs was accurate. And I know a LOT of social
scientists. Even radical feminist commies who worship Margarat Mead now
accept that this work was largely wrong.

Most social scientists don't 100% accept Derek Freeman's portrayal of Samoan
sexual life either -- although it's generally accepted as closer to the
truth than Margaret Mead's original account.

Many social scientists believe Mead was an important anthropologist even
though some of her work was inaccurate.

In contemporary anthropology courses, the "classic" anthropologists like
Mead, Colin Turnbull ("The Forest People") and so forth are generally looked
down upon as having been pre-scientific, and given short shrift. They come
up in the intro course but appear nowhere in advanced coursework.

So I really do believe that, in this particular example at least, you're
attacking a straw man.

Similarly, in modern psych courses, Freud and Jung are mentioned in Intro to
Psych and in Theory of Personality or a couple other courses, but they
rarely show up in advanced work -- it's all about experimental design, data
analysis, and so forth. I taught some Jung in an advanced psych course once
and the students were flabbergasted that I had the guts to bring up such
unscientific stuff in a senior-year course (I just introduced his theory as
a metaphor for some dynamical systems theory based data analysis we were
doing).

There certainly IS a lot of narrow-mindedness and foolish dogma in the
social sciences, I wouldn't deny that. But it's not nearly as simplistic as
you're implying. These are smart, well-informed people making much subtler
mistakes than you seem to realize.

A very common mistake made in sociology nowadays, for example, is applying
statistical methods when there are too few data points for them to be
meaningful.... And somehow the outcome of the half-bogus statistical
analysis tends to reflect the political views of the scientist involved.
For instance, conservatives do half-bogus statistics to prove that gun
control increases violence. Then liberals do half-bogus statistics to prove
the opposite. Ultimately, a lot of qualitative analysis is needed to guide
the use of fancy statistics. But a lot of contemporary social scientists
don't want to realize this, because qualitative analysis has a bad name
because of work like Mead's in the past. Yet, what recent experience proves
is that fancy statistical work is not devoid of bias either, rather it's
strongly guided by it, in complex & subtle ways.

To me, your view of social scientists is similar to the view I used to have
that "rich people are greedy, selfish bastards." Well, now I know a lot of
rich people, though I'm far from rich myself. Some of them are greedy,
selfish bastards. A lot of them aren't. And when the rich people I know DO
make what I view as selfish moral errors, biased by their wealth, they don't
do so nearly as crassly or stupidly as I had naively thought before.

ben



This archive was generated by hypermail 2.1.5 : Sat Nov 02 2002 - 08:07:34 MST