• Newsgroups: comp.ai.fuzzy,sci.stat.math
• From: Stephan Lehmke <Stephan.Lehmke@cs.uni-dortmund.de>
• Subject: Re: Thomas' Fuzziness and Probability
• Date: 15 Aug 2001 09:47:26 GMT
• Organization: CS Department, University of Dortmund, Germany

In article <23af61c2.0108142014.6f210d4f@posting.google.com>, Robert Dodier writes:
>
> Any such definition must ignore the relation between elements in a
> compound: if truth(B')=truth(B), then in any proposition containing
> A and B, I can swap in B' in place of B, and get exactly the same
> truth value for the compound; whether the elements are redundant, 
> contradictory, or completely unrelated doesn't enter the calculation.

It's exactly the same in two-valued logic. As fuzzy logic agrees with
classical logic on the extremal truth values, there is no way the
behaviour you observe can be avoided.

If _you_ have some additional knowledge about how eye colors usually
behave, you have to introduce this knowledge as an additional axiom,
exactly as you would do in two-valued logic.

regards
Stephan