COMMENT ⊗   VALID 00002 PAGES
C REC  PAGE   DESCRIPTION
C00001 00001
C00002 00002	.require "memo.pub[let,jmc]" source
C00007 ENDMK
C⊗;
.require "memo.pub[let,jmc]" source
.cb 1979-80 STUDY OF ARTIFICIAL INTELLIGENCE AND PHILOSOPHY


	An increasing number of philosophers and researchers in
artificial intelligence have come to believe that these subjects
have relations closer than the usual ones that philosophy has
with other scientific disciplines.  On the one hand, an intelligent
computer program must be equipped with a formalized general view of the
world, and general views of the world have been the province of
philosophy.  On the other hand, an increasingly attractive way
of verifying a theory of how knowledge about the world can be
obtained is to program a computer to seek knowledge in accordance
with the tenets of the theory - or at least to discuss how a computer
might be programmed to seek knowledge.  Beyond that, both fields
share an interest in philosophical logic - the study of formal
theories of knowledge, causality, ability, wants, etc.

	From the AI side, the connection has been explicitly
recognized by John McCarthy (Stanford), Patrick Hayes (Essex),
Herbert Simon (Carnegie-Mellon), and Robert Moore (SRI), among
others.  From the philosophical side the connection has been
recognized by Daniel Dennett (Tufts), Margaret Boden (Sussex),
Patrick Suppes (Stanford), Zenon Pylyshyn (U. of Western Ontario?),
Martin Ringle (SUNY, New Paltz), and others, and less explicitly
by Hilary Putnam among others.

	Therefore, the idea of organizing a joint study by philosophers
and artificial intelligence researchers in the academic year 1979-80
at the Center for Advanced Study in the Behavioral Sciences has met
with an enthusiastic reception among many of the people mentioned
above and some others in addition.

	At the request of Gardner Lindzey, John McCarthy has discussed
with several people their interest in participating.  There will be
no problem in attracting more than enough highly qualified people,
but there hasn't been time to contact enough people to come up with
a tentative list.  The number originally mentioned is four to seven,
and a good study is possible with a number of attendees in that range.

	A tentative list of the interests to be covered is the following:

1. The conditions for ascribing specific mental qualities
to machines.

2. What an intelligent machine must know about the world, what modes
of reasoning must it use, and how can all these be realized in its
structure.

3. Formalisms for expressing facts about the world and about a person's
or a machine's attitudes.

It will be an advantage if some participants' interests include several
of the topics.