Well, I don't see how you can make progress without formulating theories (more correctly, hypotheses). I think the problem is when people become attached to a

Well, I don't see how you can make progress without formulating theories (more correctly, hypotheses). I think the problem is when people become attached to a particular hypothesis. Many people seem unable to accept that they only have sufficient information to assign probabilities to different hypotheses and instead always assign probability 1 to their favorite hypothesis and 0 to all the others.