Section 5

The question arises: If one denies that a formula involving an added determinant is a definition, how can there be a definition of terms which are not simple but coupled? Because they can only be explained by adding a determinant.

I mean, e.g., there is nose and concavity and snubness, the term compounded of the two, because the one is present in the other. Neither concavity nor snubness is an accidental, but a per se affection of the nose. 1 Nor are they attributes in the sense that white is of Callias or a man, because Callias is white and is by accident a man; but in the sense that male is an attribute of animal, and equality of quantity, and all other attributes which we say belong per se.

That is, all things which involve the formula or name of the subject of the affection, and cannot be explained apart from it. Thus white can be explained apart from man, but not female apart from animal. Thus either these terms have no essence or definition, or else they have it in a different sense, as we have said.

But there is also another difficulty about them. If snub nose is the same as concave nose, snub will be the same as concave. But if not, since it is impossible to speak of snub apart from the thing of which it is a per se affection (because snub means a concavity in the nose), either it is impossible to call the nose snub, or it will be a tautology, concave-nose nose because snub nose will equal concave-nose nose.

Hence it is absurd that such terms as these should have an essence. Otherwise there will be an infinite regression; for in snub-nose nose there will be yet another nose. Clearly, then, there is definition of substance alone. If there were definition of the other categories also, it would have to involve an added determinant, as in the case of the qualitative; and of the odd, for this cannot be defined apart from number; nor can female apart from animal.

By involving an added determinant I mean descriptions which involve a tautology, as in the above examples. Now if this is true, there will be no definition of compound expressions either; e.g., odd number. We fail to realize this because our terms are not used accurately. If on the other hand there are definitions of these too, either they are defined in a different way, or, as we have said, definition and essence must be used in more than one sense;

thus in one sense there will be no definition of anything, and nothing will have an essence, except substances; and in another those other things will have a definition and essence. It is obvious, then, that the definition is the formula of the essence, and that the essence belongs either only to substances, or especially and primarily and simply.