operations such as comparison, which presuppose in the compared qualities already existing, could in no case prove that these qualities depend on no relations at all. But I cannot believe that this is a matter to be decided by analogy, for the whole case is briefly this. There is an operation which, removing one part of what is given, presents the other part in abstraction. This result is never to be found anywhere apart from a persisting abstraction. And, if we have no further information, I can find no excuse for setting up the result as being fact without the process. The burden lies wholly on the assertor, and he fails entirely to support it. The argument that in perception one quality must be given first and before others, and therefore cannot be relative, is hardly worth mentioning. What is more natural than for qualities always to have come to us in some conjunction, and never alone?
We may go further. Not only is the ignoring of the process a thing quite indefensible—even if it blundered into truth—but there is evidence that it gives falsehood. For the result bears internally the character of the process. The manyness of the qualities cannot, in short, be reconciled with their simplicity. Their plurality depends on relation, and, without that relation, they are not distinct. But, if not distinct, then not different, and therefore not qualities.
I am not urging that quality without difference is in every sense impossible. For all I know, creatures may exist whose life consists, for themselves, in one unbroken simple feeling; and the arguments urged against such a possibility in my judgment come short. And, if you want to call this feeling a quality, by all means gratify your desire. But then remember that the whole point is quite irrelevant. For no one is contending whether the universe is or is not a quality in this sense; but the question is entirely as to qualities. And a universe con-