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

From: Ben Goertzel (ben@goertzel.org)
Date: Wed May 09 2001 - 21:01:20 MDT


Eliezer said:
> Okay. I accept this result. I still maintain that almost all the social
> research that shows up on Eurekalert appears to be politically correct
> crap,

Eurekalert, as I understand it, presents research for which the researchers
have released formal press releases.
Naturally, in the social science domain, this is going to tend to be biased
toward research with political content or implications. It's by no means a
fair sample of social science research. New theorems discovered in
mathematical psychology don't tend to merit press releases, for example.

>but I suppose it's barely possible that the soft sciences aren't
> complete intellectual wastelands and that some of its members are slowly
> straggling into the light of the twenty-first century.

I do think that social science has a far higher ratio of crap than hard
science. But I'm not sure this is due to political biases, I think it's due
to the fact that the data constrains the theories so much less.

> If you want any
> more than that out of me, you'll have to actually show me a social
> scientist doing something right <grin>.

Some EXTREMELY USEFUL social science research has been done on the topic of
welfare reform, recently. See my father's papers on the topic, and
references given therein:

http://crab.rutgers.edu/~goertzel/welfarepage.htm

The following page contains links to socioeconomic analyses of the recent
changes in Brazilian economic policy, effected by President Cardoso, which
have been rather successful [Cardoso is a sociologist and attributes his
successes in government in part to his understanding of various aspects of
sociological and economic theory]:

http://crab.rutgers.edu/~goertzel/fhc.htm

On a less pragmatic and more theoretical level, one fairly interesting
subfield of social science is the study of "social networks." This subfield
uses some nice statistics and graph theory, and relates to complex system
science in various ways. A bunch of links to resources in this area is
given in:

http://www.econ.iastate.edu/tesfatsi/netgroup.htm

I think this area of research has a lot of potential, but remains moderately
boring due to its static rather than dynamical focus.

This is just a random sampling based on my own knowledge and interests.
There is a lot more out there, believe me.

In terms of classic anthropology, try reading Colin Turnbull's original
books on the Mbuti pygmies. There were some errors there, but also a lot of
deep observations, regarding such things as the different views of time that
Stone Age people tend to have, as opposed to industrially or
post-industrially acculturated people. I find this stuff quite fascinating,
more so than Margaret Mead's work by far.

I won't bother to link to all the high-quality research that exists in
cognitive, perceptual and social psychology.

The frustrating thing about social science, to me, isn't that there's no
high-quality work. There is lots of it. It's that the high-quality work
that's there very rarely addresses the core issues one would like to
understand about people or societies or economies.

This is similar to the situation in AI. It's not that the papers in the
journal Artificial Intelligence are *bad*. Most of them are actually quite
interesting, in themselves. They just don't tell you much about how to make
thinking machines!!

-- Ben



This archive was generated by hypermail 2b30 : Mon May 28 2001 - 10:00:04 MDT