Several philosophies of mathematics try to reformulate the content of our mathematical beliefs and knowledge. That is, they do not hold that a face-value reading of our mathematical assertions, such as the assertion that 5 + 7 = 12, corresponds to the content of our corresponding beliefs. One alternative reading that is proposed is that the content actually is something along the lines of there is a proof for 5 + 7 = 12. This alternative reading is proposed by several nominalists (in the restricted sense as those who object to the existence of mathematical objects), as well as by constructivists. In this talk I will criticize this type of reformulation on the basis of the fact that most people do not have a concept of PROOF readily available. Given this lack of a concept of PROOF, it would be impossible for these people to have beliefs whose content includes this concept. Thus, a reformulation of this kind would exclude the possibility of mathematical beliefs in the case of the vast majority of people. Furthermore, I will also criticize a related approach to meaning, found in constructivism, where the meaning of mathematical statements is construed as given by proof-conditions. There, the main point of criticism is that in order to know the proof-conditions of a mathematical statement one has to at least know the relevant inference rules, but here too most people fall short of the mark. That is, the vast majority of people does not have a good grasp of the inference rules that  So here a similar point can be made as for belief, namely that saying that proof-conditions give the meaning of mathematical statements makes it impossible for most people to know the meaning of mathematical statements.