I think that Bayesian reasoning applies to a different category of knowledge; there are things for which we have no predictive theories that we have to reason a

I think that Bayesian reasoning applies to a different category of knowledge; there are things for which we have no predictive theories that we have to reason about probalistically and Bayes is good for that. But thinking about how things are and probability seems a poor duo - after all, we can end up 99.999% sure that something is true, observe one counter example and now know that 100% it is false... Bayes doesn't do that.