COMMENT ⊗   VALID 00002 PAGES
C REC  PAGE   DESCRIPTION
C00001 00001
C00002 00002	DISCUSSION WITH BOB MOORE (RCM) APRIL 21, 1980
C00009 ENDMK
C⊗;
DISCUSSION WITH BOB MOORE (RCM) APRIL 21, 1980

Major Question

Suppose that a word is used in two senses correctly.  Must there be an internal
distinction in the brain or program between the senses.  JMC intuits no and
attempts to support it by looking for an example where a word can be used in
two ways or in a way that is non-committal between the two.  Some person uses
the word in all three ways.  When asked if the word is ambiguous, he says no.
When the ambiguous readings are given, he recognizes that it is ambiguous,
but says the ambiguity didn't occur to him previously.

There are two possibilities:

  1.  There are structures in his brain that make the distinction, but he wasn't
      aware of them.

  2.  The distinction is not made in any structure until the issue of ambiguity
      arises.

JMC intuits that 2 occurs often, but RCM doubts that a system that didn't some-
how make the distinction would work.

JMC looks for confirmation that 2 is possible from two sources.

      (a)  We try to find a concept that splits further, indeed indefinitely or 
           beyond the point that it is plausible that structural distinctions
           exist.

      (b)  We try to %2design%* a formalism for use by a program which
           
           (1)  Doesn't distinguish the cases by different terms or conditional
                usage.

           (2)  Works appropriately in each of the three cases.

           (3)  Can accomodate the distinction when pointed out without changing
                any existing sentences.  The new sentences making the distinction
                are confirmed by thought only without accepting as true sentences
                from the outside.  Thus, the sentences from the outside all start,
                "Think about the following situation."  In fact,if I could make
                it precise, I would prefer to make the condition that the distinc-
                tion is one that the system could make for itself.
 
Examples:

We discovered the two senses of "write", (1) compose a text and (2) physically
put the words on paper.  JMC proposed that a person or program could use the
word correctly in contexts admitting three interpretations without having thought
of the distinction.  RCM said maybe but pointed out that in most present usage (2)
entails (1).

RCM brought up an example due to Parsons (?) of the crime of attempting to
bribe a public official.  Convictions have occurred in which (1) A tried
to bribe B not knowing that B was a public official [%2de re%*] and (2) A
tried to bribe B under the mistaken belief that B was a public official
[%2de dicto%*].

Most likely the legislators never thought about the problem.  JMC proposes
the slogan "Ceteris paribus de re = de dicto" with the application that
when lawmakers or others discuss a situation in which people perform
actions, they presume, unless the contrary possibility is a prima facie
part of the problem under discussion, that all the relevant facts are
known to the participants.  Formalizing this presumption can be done
only in a system permitting non-monotonic reasoning.

We imagined further the Gettier defense, supposing (apparently
counterfactually) that conviction required knowledge that the bribee was a
public official.  The defendant tried to bribe him and he was a public
official, but the defendant's belief tthat he was a public official was
based on a misapprehension.  Hence, according to Gettier, the defendant
didn't know he was a public official.

[There are cases where a search was made with a warrant,
found the contraband searched for, but was later found by the court to be
without probable cause.  In one case, a warrant was based on an
affadavit that someone had been told by a person alleged to have the
stolen plans that the plans were in his basement.  The house was searched
and the plans found, but the search was ruled "without probable cause" on
the grounds that the admission was made a year before the search, and
it was considered improbable that the plans would be there a year later].