COMMENT ⊗   VALID 00002 PAGES
C REC  PAGE   DESCRIPTION
C00001 00001
C00002 00002	%accept[f88,jmc]		Acceptance speech for Kyoto prize
C00007 ENDMK
C⊗;
%accept[f88,jmc]		Acceptance speech for Kyoto prize

Mr. Inamori, ladies and gentlemen.

	I am very grateful to everyone involved in this award
including Mr. Inamori, the Inamori Foundation organization,
the people who recommended me for the award and the committees
involved in the selection.

	In the middle 1950s, I started my work on the use of
mathematical logic to express common sense knowledge and
reasoning for purposes of artificial intelligence.  Steady
progress was made even though very few people worked on it until
the development of formalized nonmonotonic reasoning in the late
1970s.  Since then it has become a much more active field, and I
will talk about my current work and that of some others in my
lectures.  However, the main problem of understanding how to
represent common sense is still far from solved, and I can't
expect to finish it in the time I have left to work.  Maybe it
will be like genetics --- about 100 years from Mendel's work to
the cracking of the genetic code in the 1960s.  Also like
genetics, the AI basic research has led to applications long
before the main scientific problems have been solved.  My
opinion, however, is that organizations rushing to build AI
applications, i.e. expert systems, will find that they also need
to support basic research in understanding common sense if they
are to achieve some of the goals they have already set for
themselves.

	My contributions to the discovery of the formalisms
embodied in the Lisp programming language took only a few years
from the first ideas in 1956 to the first Lisp system in 1959.  I
call it a discovery, because I believe that the basic functions
on lists and the recursive use of conditional expressions would
have been discovered by someone else if I hadn't done it.  My own
contributions to Lisp have not been large since 1962.  Since that
time I have been essentially a spectator of the development of
Lisp both as a programming language and in hardware.  I never
thought it would outlast so many other approaches and become the
second oldest programming language still in use and would at this
time still have a bright future.

	The work on replacing debugging by machine-aided proofs
that programs meet their specifications is another area that has
had slow progress, perhaps too slow, towards an ambitious goal.
Much of the basic mathematics has been done --- by many people ---
and perhaps the main problem is to get the mathematical talent
applied to specific important classes of programs.

	I think that Inamori Foundation has taken an important
step in including computer science, and especially artificial
intelligence, among the areas for which work can get such major
recognition as the Kyoto Prize.  It will hasten the day when man
understands intelligence well enough to make machines as
intelligent as man.