Originally posted by googlefudge
If I want answers to the kinds of metaphysical questions posed in this forum then science really isn't any use as there's no empirical evidence and no sound justification for extrapolating from nomological certainties.
People keep saying things like this, but when pressed cannot actually provide
questions that actually have answers th ...[text shortened]... or otherwise] that have any cogent meaning that cannot be addressed via
the scientific method.
When you say "Questions that actually have answers" do you mean questions where the answer is known - where justified and true are taken to be within normal levels of certainty. Or do you mean questions which are answerable, an answer is believed to exist, but no such answer is known.
If there is no empirical evidence for some question then it's difficult to see what science can say. Empirical science doesn't help much with proving something like the internal consistency of elementary geometry [1], it's an exercise in logic. Now we come to a problem, if you define logic as part of Science then any answer I give can end up being part of science. If one does not then I've provided an example.
I think you're conflating science and rationalism. I'm under the impression that you're a rationalist, which may or may not be true, but for the rest of the post I'm going to assume it is. The two have things in common, but they aren't the same thing. So rationalism may well be able to provide answers to things that science can't as science is bound by empiricism, which rationalism isn't automatically, as I understand it.
So if your argument is that these things do not require one to turn to religion to find answers then sure. But that is a different statement to the one I made which is that science cannot answer some questions, even in principle.
[1] Where by elementary geometry I mean the Tarski axiomatization in first order logic. It's carefully designed to avoid Gödel's incompleteness theorem and so is both consistent, complete, but only semi-decidable. Meaning that a statement made within that system is provably true. It counts as infallible knowledge. The answer is unarguably exists.
I think the Wikipedia page
could have a mistake. One of the writers seemed to think one could define distances by simply defining a distance to be 1. Since Tarski didn't include this I'm guessing that it could do some damage to the provability of the system. But I don't understand Gödel's theorem well enough to tell.
http://en.wikipedia.org/wiki/Tarski's_axioms