COMMENT ⊗   VALID 00002 PAGES
C REC  PAGE   DESCRIPTION
C00001 00001
C00002 00002	.require "let.pub[let,jmc]" source
C00006 ENDMK
C⊗;
.require "let.pub[let,jmc]" source;
∂AIL ∞
Dear Colleagues:

	Here is another draft of %2Ascribing Mental Qualities
to Machines%1.  It seems to me that I did not succeed in making
my ideas entirely clear in the previous draft and in the
Philosophy Colloquium, but the exchanges there were very helpful
in pointing out areas of incompleteness and confusion.
I hope the present draft is better.

	In particular, I hope I have succeeded in meeting the
following points raised by Professor Moravcsik and others:

	1. Why doesn't the formalism allow ascribing beliefs to
stones?  Ans: Any ascription of beliefs to stones contains a
subascription that assigns no beliefs and accounts for just as
much internal state, behavior, and is just as good at distinguishing
this stone from other stones.

	2. What is the distinction between a definition relative
to an approximate theory and a definition based on assumptions?
Ans: The approximate theory is not assumed to be true; it is assumed
to be an approximation, and it is argued that a definition relative
to such a theory may be useful %3even if the concept cannot be defined
in the world or in a more refined theory%1.

	I have not yet formulated why I use mental terminology
rather than calling them pseudo-beliefs or talking about the presence
of information.

	Let me express my appreciation of your tolerance of this
venture into a field that is new for me.  I intend one more round
of revision before I submit the paper for publication, and any
comments, oral or written, will be much appreciated.  I especially
want to remove some redundancy that has crept in, remove some
extraneous observations, and revise or eliminate the polemical
section.

	It now seems to me that I should eventually try to take
my own advice and devise an approximate theory within which the
main content of the paper would be technical and uncontroversial,
leaving the controversy to the overall fit of the ideas to reality.
Unfortunately, I don't yet see how to do it.  Perhaps the present
paper should be regarded as a heuristic introduction to the more
technical paper that I don't yet know how to write.

.sgn