All our knowledge about the world is acquired via experience. A priori knowledge about the world – knowledge based on an appeal to reason, independently of experience – is thus impossible. Reason is not some kind of intellectual searchlight which can, independently of observation and experiment, illuminate the world and provide us with infallible knowledge about it. All propositions which can be known to be true with certainty, in an a priori way, independently of experience – propositions like “either it is raining or it is not raining” or “all bachelors are unmarried” – are empty of factual content. And all propositions which have some factual content, which say something about the world, can only be known with some degree of uncertainty as a result of our experience of or interactions with the world – as a result of observation or experiment. As Einstein once put it “as far as propositions refer to reality, they are not certain; and as far as they are certain, they do not refer to reality”.[i] This unquestionably represents the orthodox view among scientists and philosophers of science. Not all philosophers would, however, be convinced by this bald statement of the case for the non-existence of a priori knowledge in science. Some might claim that the case for a priori knowledge is supported by Kripke’s (1981) argument that identity statements with rigid designators are necessary – a statement such as that water is identical to H2O being, according to Kripke, despite its factual appearance, nevertheless necessary. Kripke would not himself hold this; he makes it quite clear that necessity and the a priori must be sharply distinguished. And in any case, Kripke’s case for the necessity of identity statements with rigid designators is, in my view, not valid, as I have argued in some detail elsewhere: see (Maxwell, 2001, appendix 2).