Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!yale!quasi-eli!cs.yale.edu!mcdermott-drew
From: mcdermott-drew@cs.yale.edu (Drew McDermott)
Newsgroups: comp.ai.philosophy
Subject: Re: Searle, Strong AI, and Chinese Rooms
Summary: I can't resist a Searle argument
Message-ID: <27320@cs.yale.edu>
Date: 16 Nov 90 19:58:34 GMT
References: <1990Nov15.204949.12075@Solbourne.COM>
Sender: news@cs.yale.edu
Organization: Yale University Computer Science Dept., New Haven, CT 06520-2158
Lines: 70
Nntp-Posting-Host: aden.ai.cs.yale.edu
Originator: dvm@aden.CS.Yale.Edu


   In article <1990Nov15.204949.12075@Solbourne.COM>
   vic@corona.Solbourne.COM (Vic Schoenberg) writes:

   Searle isn't satisfied with an
   >operational definition of intelligence because this doesn't address the
   >issues of subjectivity, qualia, the problem of other minds, and so
   >forth that are central to the human experience and constitute the core
   >unsolved problems of this area of philosophic study.
   >
   >Searle is often accused of dualism or even mysticism, but he doesn't 
   >consider himself as either. If anyone is taking a leap of faith here,
   >it is the AI advocates. I doubt if any of them think a radio
   >understands speech, or a television enjoys sitcoms, or a computer reads
   >the email that passes through it and forms opinions on its quality. 
   >But we think that with the right wiring and right programs it will
   >suddenly become conscious and have beliefs. 

If you delete the word "suddenly," then of course you're right: Our
operating assumption is that with the "right ... programs, it will be
... conscious."  

   >Searle doesn't deny that material entities can have such properties, but
   >he suggests that something in the brain is making possible these
   >subjective experiences which humans have, and the something that does
   >this, whatever it is, is quite beyond anything computer scientists have
   >created or proposed. 

   >I think he has a valid point, and I wish we could address the problems
   >of qualia, other minds, and the subjective experiences of humans and
   >other intelligent agents instead of belittling him and the issues he
   >raises.
   >

I agree entirely (although "quite beyond" seems a little strong), ....

   >--
   >
   >Vic Schoenberg  	vic@Solbourne.COM
   >303/678-4603		...!{uunet,boulder,sun}!stan!vic

.... however, it would be nice to have a decisive refutation of
Searle.  And here it is: Searle's argument takes the form of a
Gedanken experiment.  Such an experiment resembles a real experiment
in that one starts with a prediction and at some point it gets refuted
or confirmed.  Obviously, this makes sense only if the experiment
causes two theories to interact in surprising ways.  E.g., Einstein
imagined what light would look like if you were traveling at the speed
of light, thus exposing basic contradictions among existing physical
theories.  

Now, the question for Searle is: Exactly what prediction would
cognitive science (or "Strong AI") make about the Chinese Room
situation?  He always talks as if the theory would predict that the
squiggle-manipulator would come to understand Chinese.  But I doubt
anyone would agree to abide by that prediction.  Instead, the
prediction would be that (with the "right programs" again), a virtual
person would come into existence that did understand Chinese.  This is
the virtual person you are communicating with via squiggles and
squoggles.  This prediction may seem crazy to scoffers at AI, but
merely seeming crazy is not sufficient for a hypothesis to be refuted
in a Gedanken experiment.  One actually has to arrive at a
contradiction of existing theories.  And of course we're ludicrously
underequipped with useful theories in this area.

Anyway, the key principle here is that those espousing the theory
being critiqued, and not those making the critique, get to say what
the theory predicts.

                                             -- Drew McDermott