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!qt.cs.utexas.edu!yale.edu!cs.yale.edu!mcdermott-drew Thu Feb 20 15:20:25 EST 1992
Article 3694 of comp.ai.philosophy:
Xref: newshub.ccs.yorku.ca comp.ai.philosophy:3694 sci.philosophy.tech:2129
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!qt.cs.utexas.edu!yale.edu!cs.yale.edu!mcdermott-drew
>From: mcdermott-drew@CS.YALE.EDU (Drew McDermott)
Newsgroups: comp.ai.philosophy,sci.philosophy.tech
Subject: Re: Reference (was re: Multiple Personality Disorder and Strong AI)
Summary: Meaning is objectively real
Keywords: consciousness,functionalism,meaning
Message-ID: <1992Feb13.045721.29805@cs.yale.edu>
Date: 13 Feb 92 04:57:21 GMT
References: <1992Feb7.232150.8611@husc3.harvard.edu> <1992Feb12.040025.14716@cs.yale.edu> <1992Feb12.063035.15857@organpipe.uug.arizona.edu>
Sender: news@cs.yale.edu (Usenet News)
Organization: Yale University Computer Science Dept., New Haven, CT 06520-2158
Lines: 79
Nntp-Posting-Host: aden.ai.cs.yale.edu


  In article <1992Feb12.063035.15857@organpipe.uug.arizona.edu> bill@NSMA.AriZonA.EdU (Bill Skaggs) writes:

  >  You
  >can replace [3] with the weaker (functional) requirement
  >that M be *used* as a model by P.
  > [...]
  >  For example, consider a thermostat, in which the length of a
  >piece of metal serves as a model of a room (or, at least, of
  >the temperature of the room, which is the only aspect the
  >thermostat cares about).  Even if the thermostat is malfunctioning
  >(i.e. behaving inappropriately), the length of the piece is
  >still a model of the temperature, because it was *designed* to
  >play that role.

Thanks for stating so clearly the (very tempting) position that I am
trying to avoid taking.  Presumably animals aren't "designed," so we
have to find some other approach to finding the parts of them that
"mean."  I also tried to avoid saying that P "uses" M; I think it
does use M, but "using something for a purpose" smacks too much of
intentionality to take for granted as a primitive notion.

In a nutshell, the problem I see is how to spell out how meaning
actually works, and not just under what circumstances we can
usefully attribute meaning to a system.  I think if we are careful
enough the problem can be minimized, but it always seems to lurk in
the background.

However: I don't this problem is specific to the issue of meaning or
intentionality.  Let me illustrate with a fable.  Suppose God created
the universe as a humongous wave function.  After he has admired it
for a few billion years, Satan strolls in, chuckling, and says, "Have
you noticed that the universe seems to contain atoms?"  God asks what
these are, and Satan tells him.  But God objects that atoms aren't
real; you have to tilt your head sideways and squint to see atoms
instead of what's really there, wave functions.  "Is it truly
reasonable to say that this piece of reality is a 'metal' consisting
of a lattice of these so-called 'atoms'?  The electrons don't really
live in the atoms; they actually swim around in a sort of soup (one of
my better creations, I thought).  Why neglect all that and carve the
world into these monstrosities that fit the laws of physics so
poorly?  The bottom line is that they are real only if their creator
chooses to view the universe that way, and I do not.  End of this
discourse." 

Satan says, "Ah, but did you know that the humans find it convenient
to see the world as consisting of atoms?"

God replies: "Humans?  What are 'humans'?"

Satan says: "Well, they're large collections of atoms, basically, who
....," at which point God throws him out.

The point of this fable is that we feel that Satan has a point (he had
better!), but how do we refute God's argument?  It is not enough to
say that atoms are real with respect to the "atomic stance," because
the only creatures who might consider taking the atomic stance are
themselves made out of atoms.  [Actually, of course, I am neglecting
all the other levels of reality, chemical, biological, etc., as
Chandrasekaran has pointed out, but I want to keep the fable simple.]

Clearly, there has to be a sense in which atoms are real regardless of
whether anyone is taking any stance toward them.  And information
processing is just as real, in spite of the attempts by people like
Searle (and Gemar, I think) to argue that whether a system is an
information processor is purely up to human observers (who confer
derived intentionality on it).

If you think God has won the argument with Satan hands down, then you
must feel that dualism is still a strong contender (because it
provides portholes for observers from outside the universe to impose
whatever structures they like on it).  But I think that dualism must
be avoided at all costs, for the usual reasons.

[By the way, the fable does not depend on quantum mechanics.  It would
work just as well in a Newtonian world, if such a thing made any sense.
But I do endorse Chalmers's reply to Baxter on quantum mechanics.]

                                             -- Drew McDermott


