Intuition check
I don't think this topic is especially important, but I was thinking about it the other day. What's worse:
(i) Having blatantly contradictory beliefs
or
(ii) Having obviously false beliefs?
So, someone could probably have lots and lots of false beliefs, but have no contradictory beliefs (at least none that we would expect her to recognize as contradictory). So, here's a made-up case that doesn't even sound so plausible: Imagine someone has some fixed set of beliefs, at least one of which is false - call it B. The problem is, she cannot change her beliefs. She comes across some new information that should make her form belief C. It is obvious, however, that C implies ~B. So, if she forms belief C, she will also form the belief that C implies ~B. So, what should she do? Would we think it was worse if she went ahead and formed belief C (and the belief that C implies ~B), even though it would easily lead to a contradiction, or if she formed belief ~C, even though it's false?
(i) Having blatantly contradictory beliefs
or
(ii) Having obviously false beliefs?
So, someone could probably have lots and lots of false beliefs, but have no contradictory beliefs (at least none that we would expect her to recognize as contradictory). So, here's a made-up case that doesn't even sound so plausible: Imagine someone has some fixed set of beliefs, at least one of which is false - call it B. The problem is, she cannot change her beliefs. She comes across some new information that should make her form belief C. It is obvious, however, that C implies ~B. So, if she forms belief C, she will also form the belief that C implies ~B. So, what should she do? Would we think it was worse if she went ahead and formed belief C (and the belief that C implies ~B), even though it would easily lead to a contradiction, or if she formed belief ~C, even though it's false?
Posted by Justin Snedegar on
07.16.2008 at 3:29pm
It's a lot better to have an obviously false belief.
However, I suspect that whatever evidence makes it obviously false will be evidence in support of the beliefs negation and hence 1 collapses into 2. So, neither seems better or worse than the other.
While this may seem stretched at one level, recall Tarski's distinction between formal correctness (avoid paradoxes in a formal language) and material adequacy (make sure the formal concept of truth maps onto ordinary concept).
There is a lot of work to be done on the contours of this sort of normative question.