Saul Kripke deserves great credit for fastening the minds of the philosophical community on the existence of different types of possibility and, correlatively, of necessity. The type of necessity upon which analytical philosophers had been accustomed to concentrate was analyticity; for clarity, I will replace this term by “a priori knowability.” We need not stop to argue over a precise characterisation of this familiar notion: I will say that a statement, considered as made by the utterance, actual or hypothetical, of a meaningful assertoric sentence by a particular person at a specific time, is a priori knowable if the speaker is or would be able to recognise its truth simply from the meanings of the words composing the sentence, together, when needed, with some deductive argument which he devised or with which he was presented. When it is not knowable a priori that a given statement is not true, let us say that that statement is possible a priori. And I shall say that a statement is knowable only a posteriori if the speaker is able to recognise its truth only on the basis of some observations he has made or of reports made to him by others of observations they have made. No doubt there are many holes in these formulations, considered as definitions, but they will serve present purposes.