COMMENT ⊗   VALID 00002 PAGES
C REC  PAGE   DESCRIPTION
C00001 00001
C00002 00002	pedant[e84,jmc]		Why philosophy is often uselessly pedantic.
C00005 ENDMK
C⊗;
pedant[e84,jmc]		Why philosophy is often uselessly pedantic.

	Ignorant people often accuse philosophers of being hairsplitting
pedants.  The object of this paper is to argue that these ignorant
people are often right about this for epistemologically fundamental
reasons.  Moreover, I want to advocate a way out via formalized
non-monotonic reasoning.  The following proposals are rather technical.

	Ordinary human thought uses concepts that are only approximately
understood.  Philosophers try to understand these concepts better by
making them precise.  Sometimes this succeeds, but often the attempts
fail because our world is such that inherently approximate concepts
must be used in order to understand it.  Attempts to replace inherently
approximate concepts by precise concepts are bound to fail.  The
definitions founder on exotic cases that do not arise in the ordinary
human use of the concepts.

	We give some examples and then some proposals for using
circumscription (McCarthy 1980 and 1984) or other non-monotonic
logical formalisms to express what we know about these concepts
while preserving their approximate character.

1. Ski instructor example.  Causality and counterfactuals.

2. Hopefully a well known philosophical puzzle involving whether
a person is guilty of a crime.

3. The car load of salt.  Ann Gardner heard a voice telling her
not to look back at Stanford, but she did and was turned into
a pillar of salt.

4. A formalization of intentional action.