From newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.ecf!utgpu!csd.unb.ca!morgan.ucs.mun.ca!nstn.ns.ca!aunro!ukma!asuvax!cs.utexas.edu!swrinde!network.ucsd.edu!orion.oac.uci.edu!cerritos.edu!arizona.edu!arizona!amethyst!organpipe.uug.ari Thu Feb 20 15:20:40 EST 1992
Article 3722 of comp.ai.philosophy:
Path: newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.ecf!utgpu!csd.unb.ca!morgan.ucs.mun.ca!nstn.ns.ca!aunro!ukma!asuvax!cs.utexas.edu!swrinde!network.ucsd.edu!orion.oac.uci.edu!cerritos.edu!arizona.edu!arizona!amethyst!organpipe.uug.ari
zona.edu!NSMA.AriZonA.EdU!bill
>From: bill@NSMA.AriZonA.EdU (Bill Skaggs)
Newsgroups: comp.ai.philosophy
Subject: Re: Searle Agrees with Strong AI?
Message-ID: <1992Feb14.021750.585@organpipe.uug.arizona.edu>
Date: 14 Feb 92 02:17:50 GMT
References: <1992Jan30.143453.19988@oracorp.com> <6186@skye.ed.ac.uk>
Sender: news@organpipe.uug.arizona.edu
Reply-To: bill@NSMA.AriZonA.EdU (Bill Skaggs)
Organization: Center for Neural Systems, Memory, and Aging
Lines: 52

In article <6186@skye.ed.ac.uk> jeff@aiai.ed.ac.uk (Jeff Dalton) writes:
>
>And that's one of the things that's so bad about such discussions.
>If you accept the burden of providing a definition, the other side
>can just sit back and pick holes in it forever.  If real holes are
>hard to find, they can just invent bogus ones, and get you to spend
>all your time trying to correct their errors.
>
  There are better reasons for being cautious about definitions.

  Important words are important because they correspond to
"natural concepts" -- regularities or patterns in experience
which are present independently of whether there are words
to represent them.  To take a trivial example, the word "cat"
is a natural concept because it represents a large group of
objects similar in many important ways.  

  Words that correspond to natural concepts must not be
arbitrarily defined, because bad definitions can make it
impossible to speak comprehensibly about the original natural
concept.  If somebody decides that "cat" should mean "a box
with four handles", then the natural concept previously
represented by "cat" becomes much more difficult to talk
about.

  One of the most important philosophical questions about a
term such as "consciousness" is whether it does in fact
correspond to a natural concept, as opposed to being a sort
of linguistic hallucination.  I take it that most of us believe
that "consciousness" *is* a natural concept.  (Or, at least,
a *set* of natural concepts.) If this is so, we are
not free to arbitrarily define the word.  A good definition
would indeed be extremely valuable, but it must come as
a result of philosophical investigation, rather than
prior to investigation.

  Unfortunately, one of the universal properties of natural
concepts is that they are fuzzy around the edges.  Definitions,
on the other hand, are as sharp at the edges as anything can
be.  Therefore definitions can never correspond exactly to
natural concepts -- nor should they.  An optimal definition
is nevertheless an extremely valuable thing to have.

  Sometimes it is necessary to have at least a rough definition
in order to make progress.  Many times, it seems to me, 
philosophical arguments arise because two groups are using
words in different ways without realizing it -- i.e. are
using the same word to refer to two different natural
concepts.  In such a situation, attempting to come up
with a definition may be the only way out of the impasse.

	-- Bill


