@Cyracuz,
Quote:Logic depends on definitions. Definitions depend on our ability to make them. Paradoxes come of bad definitions of premises.
If we can't agree on a few definitions, we can't have a functional dialogue. But you seem to agree that paradoxes are an issue. Encouraging...
I see logic is a sort of innate, often unconscious (yet today partly formalized in mathematics) language linking propositions or representations.
If I say a tautology like 'a cat is a cat', anyone who knows what a cat is will agree with me. Chances are that even those who don't know what a cat is -- some dudes deep in Papua New Guinea perhaps -- will agree that a cat is probably the same thing as a cat. It is possible to deny it, no doubt, even reason ably so eg all cats are different and not two cats are the same... So a cat is not ANY cat but 'a cat is a particular cat'... :/ but still, by and large a cat is a cat... To use your language, it 'works' and it works for all of us humans. I'm sure even cats would partly agree.
Now, one may argue that not all logic axioms or theorems are as universally self evident as tautologies. Necessary conditions, sufficient conditions and the like are perhaps colored by cultural mores but I would expect them to have a biological (instinctive) substrate. We don't have that many languages pretending to universality. I say let's give logic a chance, and a voice. It can only help us understand one another. We should be able to call a tautology when we see one for instance.
Logic is not indeed an absolute criterion, but together with empiric observations and other (relative) criteria, it helps in a multi-criteria assessment.
Quote:And your 2) is just another way of saying "what works".
If "what works" means "what best fit the data", I'm all for it. But isn't "data" just another word for "fact"?