>
> And thanks to Chuck Grimes for the mathametical proof demonstrating the issue of
> proving a negative, since all logic, by definition, can be reduced to a math
> proof.
>
or the other way around!!!
> Folks can look up the raven proof paradox using a search engine. It actually is
> quite clever and easy reading. The raven proof paradox was part of "Studies in
> the Logic of Confirmation" by Carl G. Hempel. The raven proof leads to the idea
> that all proofs in science are actually arguments for what is more likely rather
> than less likely, not anything that can be considered definitive.
at the risk of making further errors (jks, did you say you studied under hempel?), may i add: hempel's raven paradox was specifically used to highlight a problem with the form of induction we use when we take each occurence of a black raven as evidence that favours the proposition "all ravens are black". logically this is equivalent to "anything that is not black is not a raven" and so we have that each non-black non-raven object we observe is also further evidence that all ravens are black! of course most of these non-black non-raven objects are also evidence, in the same sense, for the proposition "all ravens are white".
the reason i thought of it was that the problem of proving or seeking evidence for these kinds of universal statements (all ravens are black) is its impossible to be empirically exhaustive in verifying them and we have to reason using probability and induction... otoh the negation of the proposition i.e., "not all ravens are black" stands a better chance at verification by producing a white raven. in the case of proving or not proving someone is a spy, it seems to me, the empirical verification is finite and after a few days examination, armed with a clear definition of what it takes to designate someone a spy, we can conclude one way or the other...
anyway, forgive my ramblings,
--ravi
doug, i hope it was clear that i was joking about "selling my lbo quota". i fully understand and respect your policies (and thanks for running the list), though i might have gone over the limit a couple of times myself.