COMMENT ⊗   VALID 00002 PAGES
C REC  PAGE   DESCRIPTION
C00001 00001
C00002 00002	.require "memo.pub[let,jmc]" source
C00013 ENDMK
C⊗;
.require "memo.pub[let,jmc]" source

Notes on %2Descriptive Epistemology: Psychological, Sociological,
and Evolutionary%1 - (Preliminary Draft of the William James Lectures,
Harvard University, Spring 1977) by Donald T. Campbell

1 - This is the first time I have seen reference to Quine's "epistemology
naturalized" and Shimony's "Copernican epistemology", and I entirely
agree with the approach and have advocated it in various papers.

1 - agreed that it will be more like science than philosophy.  In fact,
why not call it "scientific epistemology"?
I suspect that I would be inclined to find that logic restricts
epistemology more than Campbell would and would prefer a more formal
theory.

2 - l.3 "knowing community"?  Why not "knowing individual"?

2 - agree with "Longings for certainty, for a firm anchoring of
beliefs, or even for complete explicitness in the evidence
supporting beliefs, must be relinquished".  On  the other hand,
AI can eventually be totally explicit about the belief-seeking
system, and such a system might be proved to work in worlds with
certain characteristics.  (It might also be developed purely
experimentally even in artificial worlds).  However, its application
to the real world suffers all the limitations Campbell mentions.

2 - l.-8 I would hope we could avoid having psychology and a fortiori
sociology play a role in scientific epistemology.

12 - "But I also want descriptive epistemology to include the theory
of how these processes could produce truth or useful approximations
to it: In what possible worlds, in what hypothetical ontologies,
would which knowledge-seeking processes work".  This is my proposal
for metaepistemology.

16 - l.6 - "Illusions are illusions only relative to a prior acceptance
of genuine bodies with which to contrast them". - yes.

16 - l.-7 - The quote from Quine is new to me.  "Epistemology is best
looked upon, then, as an enterprise within natural science.  Cartesian
doubt is not the way to begin.  Retaining our present beliefs about
nature, we can still ask how we can have arrived at them.
Science tells us that our only source of information about the external
world is through the impact of light rays and molecules upon our
sensory surfaces.  Stimulated in these ways, we somehow evolve an
elaborate and useful science.  How do we do this, and why does the
resulting science work so well?  These are genuine questions, and no
feigning of doubt is needed to appreciate them.  They are scientific
questions about a species of primates, and they are open to investigation
in natural science, the very science whose acquisition is being
investigated" (Quine 1975, p. 78).

	This quote raises the following questions: (11) minor point.  One
shouldn't make too much of "surfaces".  Kinaesthetic sensations occur
in our interiors.  The theoretical surface between the physical world
and the intentional ascriptions is more to thhe point.  However, this
issue is not important to Quine's points.  (2) Quine seems here to
neglect the question of how we learn ordinary particular facts about
the world far short of scientific theories.  This just as much a part
of epistemology as science is, and it is probably more basic, and it
has many peculiarities, because it isn't as unified and public as are
the paradigm sciences.  (3) I don't think that epistemology is mainly
about a species of primates.  Most of its facts will concern any causal
system learning about a larger system in which it is imbedded.  Thus
epistemology will be quite distinct from psychology and sociology.

19 - "Arne Ness ...".  I don't understand this paragraph.  Maybe
Dennett does.

19 - "epistemology of the other one".  Yet another precursor I didn't
know about.  I have proposed that epistemology should study "How does
it know?" rather than "How do I know?".

20 - I concur with his bet on the correspondence theory of truth.  It
will be hard to do meta-epistemology without it.

25 - I agree with his treating the frame illusion as a contingent fact
that could be otherwise were it important.

29 - It seems to me that Campbell misses the fact that exact knowledge
is sometimes possible provided one isn't interested in unnecessary
detail and one the knowledge isn't required to be certain.  A computer
usually works precisely according to its order code, and one computer
can simulate another precisely, and people often get exactly the right
answer to problems in arithmetic and logic.

31 - "Now I share ...".  Maybe Dennett understands this paragraph.
AI tends to emphasize mechanisms that get good answers with surprisingly
little blind search.

35, l.4 - Ascribing knowledge to the New England fruit tree may indeed
be worthwhile (or it may not).  He doesn't mention that the value of
an ascription of intentionality may depend on our state of knowledge
and not merely on the nature and state of the system being analyzed.

36, l.6 - There is grist for my approximate theories here, but the
evolutionary neurophysiologists won't be ready to play their part in
this game for quite some time.  AI people are the collaborators he
needs for now.

36, l. 16 - Perhaps there is some relation here with the "epistemological
adequacy" - "heuristic adequacy" distinction.

37 - The ghost of the 1970 Aaron Sloman walks again.


38,l.8 - Trying out complete theories against the world is too
expensive computationally.  Indeed most theories are never really
completed.  Perhaps I miss his point.

38 - He lacks the concreteness that working with programs would supply.
I doubt that the rest of this chapter will amount to much.

49,l4 - Finally, a mention of computers.

	As I exclusively predicted, the rest of the chapter didn't amount
to much.  The material on levels of cybernetic hierarchy, while mainly
true, is peripheral to epistemology.

59- can't make out footnote.

67 - agree about abandoning definition in favor of ostension for natural
kind words.  agree about black swans

80 - par 2 - Here is the first hint that some epistemological (or in
this case heuristic) facts might apply to Martians and machines as well.

94 - Reminds me of my statement, "Digital physics is very implausible,
but the scientific community can surely afford to invest a Fredkin or
two in it".

97 - How to build firm structures on quicksand.

100+ - Let the story of Tolman be a lesson to us all.

120 - agree about science

128 - I don't see that evolutionary epistemology has to give up
"literal truth".  Maybe this has to do with Campbell's earlier error
about exactness.

128 - "They (random rites to determine new hunting site) had the further
social role of making no one group member to blame for the frustration
of such exploration, most of which was destined to be unsuccessful.