From newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.csri.toronto.edu!rpi!think.com!yale.edu!qt.cs.utexas.edu!cs.utexas.edu!uunet!mcsun!uknet!edcastle!aiai!jeff Sun Dec  1 13:06:21 EST 1991
Article 1721 of comp.ai.philosophy:
Path: newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.csri.toronto.edu!rpi!think.com!yale.edu!qt.cs.utexas.edu!cs.utexas.edu!uunet!mcsun!uknet!edcastle!aiai!jeff
>From: jeff@aiai.ed.ac.uk (Jeff Dalton)
Newsgroups: comp.ai.philosophy
Subject: Re: A Behaviorist Approach to AI Philosophy
Message-ID: <5739@skye.ed.ac.uk>
Date: 28 Nov 91 18:09:18 GMT
References: <YAMAUCHI.91Nov24030039@magenta.cs.rochester.edu> <5727@skye.ed.ac.uk> <YAMAUCHI.91Nov27203011@magenta.cs.rochester.edu>
Reply-To: jeff@aiai.UUCP (Jeff Dalton)
Organization: AIAI, University of Edinburgh, Scotland
Lines: 62

In article <YAMAUCHI.91Nov27203011@magenta.cs.rochester.edu> yamauchi@cs.rochester.edu (Brian Yamauchi) writes:
>In article <5727@skye.ed.ac.uk> jeff@aiai.ed.ac.uk (Jeff Dalton) writes:
>>In article <YAMAUCHI.91Nov24030039@magenta.cs.rochester.edu> yamauchi@cs.rochester.edu (Brian Yamauchi) writes:
>
>>>Suppose AI researchers could build a robot that was indistinguishable
>>>from a human being in every way -- sensorimotor behavior, language
>>>abilities, learning and reasoning powers, even physical appearance.
>>>Would you argue that this robot is incapable of consciousness simply
>>>because it was the product of human engineering rather than mutation
>>>and natural selection?

No.  Nor would Searle.  The Searle complaint about computers is
not that they're man-made.

>>The problem here is that you first say "in every way" and then list
>>a small number of things.  If you really do mean in every way, if
>>you cut it it bleeds, etc, then its relevance to such questions as
>>whether computers can think is, at best, obscure.
>
>Does it make any difference whether it bleeds or not? 

It was a literary reference.

>I've included those attributes that I consider most relevant to
>"thinking".  Are you arguing that I have omitted relevant attributes
>or included irrelevant ones?  If the former, what else would you
>suggest?  If the latter, what difference does it make?

Your "simply because it was the product of human engineering rather
than mutation and natural selection" imples that it is like humans
in every other way.  For the answer to that case, see above.

Once you start omitting attributes, it depends on what you omit.
No one knows exactly what attributes are required.  If Searle's
argument is correct, it shows that it must be more than merely
instantiating the right computer program.  So to counter Searle
you need an example that is nothing more than that, yet clearly
does understand.

>For the sake of argument consider three cases:
>
>Case I:  A robot absolutely indistinguishable from human.

See above.

>Case II: A robot behaviorally indistinguishable from human, and
>physically distinguishable only through surgery or dissection.

This might depend on the nature of the insides and on what
processes are involved in producing the behavior.  Can you
be sure it doesn't?

>Case III: A robot behaviorally indistinguishable from human, similar
>in physical capabilities and structure (bipedal, two dextrous
>arms/hands, stereo vision/hearing, etc.), but very different in
>appearance (e.g. kevlar and titanium rather than skin and bones).

Again this might depend on exactly how it works.  Moreover,
if the materials are sufficiently different, the behavior will
also be different, though perhaps not in significant ways.

-- jeff


