View header  
So with some think time off line.

I am happy with the notion of "Calibration propositions"
(described in the original question) as a means for establishing a fuzzy
sets membership and domain scale relationships.

The Truth Series, along with a number of other systems, described by Cox
in his book- Fuzzy Systems Handbook validate this method for discovering
a fuzzy sets membership & value pairing.

So is it probabilistic ?  ....  In method I would have say that it is.

BUT

From an intent and outcome stance there are enormous differences.

Fuzzy is not a variation of a probability math because you need to discard a
basic foundation of classical set theory required of probability to accept
the
legitimacy of  Fuzzy Math.

When Voting about the probability that a height is tall in order to
yield a probability distribution curve the voters are working to producing
a membership value (% population Yes  OR  No) that a value is tall  OR
is not tall.......... not both.

When Voting if a height is Fuzzy tall the voters accept the notion
that the height is also not tall. The vote goes to the degree of tallness.
The % of population that vote Tall and that give a fuzzy height value its
membership may also vote that the height is not tall.

The law of non contradiction must be discarded by those accepting
Fuzzy logic as a valid mathematics.

Fuzzy means that nothing is either OR ---  probability needs only either OR.

This is not to say that the mathematics of Probability are not useful.
It is a math method accustomed to processing "degree". In that it doesn't
compromise Fuzzy set contradiction has some useful method.

Rob W

Wise wrote in message <9kia7r$kc9$1@perki.connect.com.au>...

> The original question that launched this stream > talked about > ########## > defining a membership function of a set by using > ``calibrational propositions'' -- the idea is that if > you ask 100 people if John is tall, and 70 of them say ``yes,'' then > mu(tall) = .7. While this seems to do a good job of capturing common > word usage, it's not at all clear to me that it captures the fuzzy > behavior of the ``tall'' set; > > it seems probabilistic rather than fuzzy. > > ########### > > > So can a "fuzzy expert" panel of 100 vote on a particular elements inclusion > in a fuzzy set (ie that a height is tall) and the extent to which they agree > (say 70%) on the inclusion of that element in the set be converted to a > membership value ? > > Then IF the answer is yes why then is this not probabilistic ? > > Rob W > > >> > >