[Avodah] tannur shel achnai

[Micha Berger]

On Thu, Jun 18, 2009 at 09:45:05AM -0400, David Riceman wrote:
: No, it's a principle of psychology: it says that if you have no reason
: to treat species differently then assign them the same probability.

Actually, if you have no reason to treat them differently, it IS good
probability theory. That's what I was saying -- if there is no reason
for heads to come up more often or tail to come up more often, the odds
of either is 1/n. As is true for honest dice.

Occam's Razor is based on not having a notion of which factors are more
likely, and therefore the most likely theory is the one that requires
positing the fewest elements. But this is after that level of
sophistication. Bayes already produced his theorem, and the math of how
to compute conditional probabilities already existed. They knew of more
complex notions than the odds of each outcome is always equal.

Philosophers debate what probability means. Both Bayes and Laplace
held of the epistomic theory, which is now usually called Bayesian
statistics. So I don't think you can draw a firm line between statistical
and psychological principles when discussing their "shitos".

: Kavua simply says you have two species, permitted and forbidden, so
: they're equally likely. Not good probability theory, but it was a long
: time ago.

But you have reasons to treat the species differently. One has a larger
population than the other. The only difference from kol deparish is that
there was some bar chiyuvah who once knew what the state of cheftzah was.


I think that the whole issue is epistemic, because I believe that all of
halakhah can be explained in terms of how the actions impact people. I'm
not sure our mental models of uncertainty are as precise as statistics,
and I'm not sure the rules of safeiq use statistics either. (Although
one could model safeiq, rov, mi'ut and sefeiq sefeiqa with a 5 valued
logic that matches statistics of variables in the domain { 0, .25, .5,
.75, 1 }, or vadai assur through mi'ut delo shechichah, mi'ut, safeiq,
rov, vadai mutar.)

My explanation of qavu'ah, though, is based on R' Aqiva Eiger's notion
that qavu'ah is a safeiq in the din, as opposed to parish which is a
safeiq in the metzi'us. Once the mtzi'us was once known, there is a
din assigned to that known state. Doubt about the din will always be
lechumrah or lequlah depending upon whether the din is deOraisa --
you don't play russian roulette with an issur once it exists.

Tir'u baTov!
-Micha

-- 
Micha Berger             It is harder to eat the day before Yom Kippur
micha at aishdas.org        with the proper intent than to fast on Yom
http://www.aishdas.org   Kippur with that intent.
Fax: (270) 514-1507                       - Rav Yisrael Salanter