METAPHYSICS Metaphysics D existence is not even mentioned as one of the senses of the word This is surprising, for from time to time in his logical works he seems to have identiWed it as a special sense Thus in Sophistical Refutations he makes the point that ‘to be something is not the same as to be, period’, i.e to be and to be F are not the same (5 167a2) He uses this principle to dissolve fallacious inferences such as ‘What is not is, because what is not is thought of’ or ‘X is not, because X is not a man’ He makes a similar move in connection with the being F of that which has ceased to be: e.g from ‘Homer is a poet’ it does not follow that he is (Int 11 21a25) In a famous passage of Posterior Analytics (11 92b14) Aristotle says ‘to be is not part of the substance (ousia) of anything, because what is (to on) is not a genus’ This can be taken as saying that existence is not part of the essence of anything: i.e that there is such a thing is not what anything is If that is what it means, then it deserves the compliment paid by Schopenhauer when he said that with prophetic insight Aristotle forestalled the Ontological Argument.3 But it is not clear that this is the only sense that can be given to the passage The premiss that to on is not a genus need not mean that there is no such kind of thing as the things that there are, true though that may be Aristotle elsewhere argues that being is not a genus because a genus is diVerentiated into species by diVerences that are distinct from it, whereas any diVerentia is a being of some kind (Metaph B 998b21) The clearest case where ‘be’ must mean ‘exist’ is when it is attached to ‘entia per accidens’: when he says ‘wise Socrates is’ and distinguishes it from ‘Socrates is wise’ he can hardly mean anything else than that wise Socrates exists, and is among the things that there are It is much more diYcult to decide, when Aristotle writes simply ‘Socrates is’, whether this means that Socrates exists or that Socrates is a subject of predication: we cannot pin him down to the distinction that seems so clear to us between the copula ‘is’ and the ‘is’ of existence When ‘is’ does occur as a copula, joining subject and predicate, we may ask what it signiWes Two possible accounts are suggested by the Aristotelian texts One is that it has no signiWcation: it is an incomplete symbol, not to be construed by itself, but to be taken with the predicate-term that follows it, so that ‘ is white’ is to be taken as standing for the accidental form being See G E M Anscombe, in Anscombe and P T Geach, Three Philosophers (Oxford: Blackwell, 1961), 20–1 224