Earlier attempts to show that simpler theories always have higher prior probabilities have failed, but there is a restricted circumstance in which the claim is right.