But it is our emphasis, perhaps, on the aspect of unity which has caused this confusion. Sugar is, of course, not the mere plurality of its different adjectives; but why should it be more than its properties in relation? When “white,” “hard,” “sweet,” and the rest co-exist in a certain way, that is surely the secret of the thing. The qualities are, and are in relation. But here, as before, when we leave phrases we wander among puzzles. “Sweet,” “white,” and “hard” seem now the subjects about which we are saying something. We certainly do not predicate one of the other; for, if we attempt to identify them, they at once resist. They are in this wholly incompatible, and, so far, quite contrary. Apparently, then, a relation is to be asserted of each. One quality, , is in relation with another quality, . But what are we to understand here by is? We do not mean that “in relation with ” is , and yet we assert that is “in relation with .” In the same way is called “before ,” and is spoken of as being “to the right of .” We say all this, but from the interpretation, then “before ” is , and “to the right of ” is , we recoil in horror. No, we should reply, the relation is not identical with the thing. It is only a sort of attribute which inheres or belongs. The word to use, when we are pressed, should not be is, but only has. But this reply comes to very little. The whole question is evidently as to the meaning of has; and, apart from metaphors not taken seriously, there appears really to be no answer. And we seem unable to clear ourselves from the old dilemma, If you predicate what is different, you ascribe to the subject what it is not; and if you predicate what is not different, you say nothing at all.
Driven forward, we must attempt to modify our statement. We must assert the relation now, not of one term, but of both. and are identical in such a point, and in such another point they differ; or, again, they are so situated in space or in time. And
