Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!mhuxr!mhuxn!ihnp4!qantel!hplabs!sri-unix!DAM%MIT-OZ@MIT-MC.ARPA
From: DAM%MIT-OZ@MIT-MC.ARPA
Newsgroups: net.physics
Subject: metaphysics
Message-ID: <432@sri-arpa.ARPA>
Date: Sun, 21-Jul-85 15:11:00 EDT
Article-I.D.: sri-arpa.432
Posted: Sun Jul 21 15:11:00 1985
Date-Received: Thu, 25-Jul-85 04:14:58 EDT
Lines: 46


	I have studied a fair amount of old fashioned metaphysics; it
has little to do with physics but some of it seems relevant to
artificial intelligence (my field).

	There is no clear-cut definition of old fasioned
metaphysics.  It is often equated with ontology.  In modern AI and
cognitive science AN ONTOLOGY is a collection of TYPES or
KINDS like the types of animals or the types of computer systems, or
the types of data structures used in a program (or mind), or even the
types of topological spaces.  An ONTOLOGICAL ASSUMPTION is an assumption
that things of a certain kind really exist.  For example many people
assume that feelings and thoughts realy exist, but B. F. Skinner is
unwilling to make this ontological (or metaphysical) assumption.


>It would be nice to have a term for speculation on questions like
>
>	1.  Does there exist a finite theory that completely describes the
>	fundamental behavior of matter in the universe?
>
>	2.  Is there a simple and elegant mathematical formulation of this
>	theory?  Why should there be?
>
>	3.  Can we guess constraints on what this theory must be like?
>	(Many general relativity enthusiasts would say yes.)

>Meta-physics would seem the appropriate term, but it has already been taken.

	I like the above suggestion for new field which studies the
collection of all possible mathematical physical theories.  I would
call this field "mathematical metaphysics".  The first step is to
define the general notion of a physical theory, or perhaps the general
notion of "a physics".  Any definition of "a physics" should include
things like Conway's game of life; the game of life seems to define a
certain universe with a certain set of physical laws.  We should also
include Newtonian mechanics; one can certainly imagine a universe of
Newtonian point masses where the only laws of physics are classical
gravitaional and E&M forces.

Any proposals for the general definition of "a physics"?

Mathematical metaphysics could be a branch of pure mathematics (just
as metamathematics is now a respectable branch of mathematics).
Mathematical metaphysics would not be constrained by real physics, but
it might eventually provide insights.