From newshub.ccs.yorku.ca!torn!utcsri!rpi!zaphod.mps.ohio-state.edu!cis.ohio-state.edu!news.sei.cmu.edu!fs7.ece.cmu.edu!crabapple.srv.cs.cmu.edu!nickh Tue Jul 28 09:41:37 EDT 1992
Article 6474 of comp.ai.philosophy:
Xref: newshub.ccs.yorku.ca rec.arts.sf.science:2784 comp.ai.philosophy:6474
Newsgroups: rec.arts.sf.science,comp.ai.philosophy
Path: newshub.ccs.yorku.ca!torn!utcsri!rpi!zaphod.mps.ohio-state.edu!cis.ohio-state.edu!news.sei.cmu.edu!fs7.ece.cmu.edu!crabapple.srv.cs.cmu.edu!nickh
>From: nickh@CS.CMU.EDU (Nick Haines)
Subject: Re: How do computers fare on scholastic achievement tests?
Message-ID: <NICKH.92Jul17110340@VOILA.VENARI.CS.CMU.EDU>
Date: Fri, 17 Jul 92 16:03:40 GMT
Organization: School of Computer Science, Carnegie Mellon University
Nntp-Posting-Host: voila.venari.cs.cmu.edu
In-Reply-To: szabo@techbook.com's message of Thu, 16 Jul 1992 09:30:57 GMT
References: <NICKH.92Jul13151811@VOILA.VENARI.CS.CMU.EDU>
	<1992Jul14.014658.4921@newstand.syr.edu>
	<NICKH.92Jul14141610@VOILA.VENARI.CS.CMU.EDU>
	<1992Jul16.093057.8880@techbook.com>
Originator: nickh@VOILA.VENARI.CS.CMU.EDU
Lines: 39

In article <1992Jul16.093057.8880@techbook.com> szabo@techbook.com (Nick Szabo) writes:

   In article <NICKH.92Jul14141610@VOILA.VENARI.CS.CMU.EDU> nickh@CS.CMU.EDU (Nick Haines) writes:

   >(for the sake of discussion, let's specify `intelligent' = `can pass 
   >Turing test') are possible...

   I've never liked the Turing test.  "Eliza" passed the Turing test
   with some people, and it was just a babbling fake Rogerian therapist.
   How about let's consider a practical measure of intelligence, eg the
   SAT tests?  These are used not only as a practical measure of scholastic 
   achievement, but also for membership into MENSA as an alternative to IQ 
   tests.  This constitutes a pragmatic, if not philosophically rigorous,
   definition of intelligence.

But on the other hand SATs (maybe someone should explain what those
are to the non-American readers of this group) and all other
`standard' IQ tests seem much easier to pass than a rigorous Turing
test. If a machine gets a score of 200 on a MENSA test but still can't
carry on a conversation with me about (hmmm) the usefulness of the
Turing test, I don't think it should be considered as of high
intelligence. It has been long understood that IQ tests are
measurements of particular limited aspects of human intelligence.
Given that there is a general correlation _in_human_beings_ between IQ
(as measured by such a test) and what we refer to as `intelligence',
they are accepted as some measure of the latter. But that correlation
breaks down sometimes for people (*) and we have no reason to think
that it holds for machines. Indeed, we have reasons to believe that it
_doesn't_ hold for machines.

Nick

(*): I used to know a mathematician who was _very_ smart (did
considerably better than me in the Cambridge University Maths Tripos),
played excellent saxaphone and had papers published on fungi when he
was 16. When at school he went to the International Maths Olympiads
two years running, and as part of the goings-on there he was given an
IQ test. He scored 85.



