passes into another state also complete. The several threads of causation seem, that is, always to imply the action of a background. And this background may, if we are judicious, be irrelevant practically. It may be practically irrelevant, not because it is ever idle, but because often it is identical, and so makes no special difference. The separate causes are, therefore, legitimate abstractions, and they contain enough truth to be practically admissible. But it will be added that, if we require truth in any strict sense, we must confine ourselves to one entire state of the world. This will be the cause, and the next entire state will be the effect.
There is much truth in this conclusion, but it remains indefensible. This tendency of the separate cause to pass beyond itself cannot be satisfied, while we retain the relational form essential to causation. And we may easily, I think, convince ourselves of this. For, in the first place, a complete state of existence, as a whole, is at any one moment utterly impossible. Any state is forced by its content to transcend itself backwards in a regress without limit. And the relations and qualities of which it is composed will refer themselves, even if you keep to the moment, for ever away from themselves into endless dissipation. Thus the complete state, which is necessary, cannot be reached. And, in the second place, there is an objection which is equally fatal. Even if we could have one self-comprised condition of the world preceding another, the relation between them would still be irrational. We assert something of something else; we have to predicate of , or else its sequence of , or else the one relation of both. But in these cases, or in any other case, can we defend our assertion? It is the old puzzle, how to justify the attributing to a subject something other than itself, and which the subject is not. If “followed by ” is not the nature of , then justify your predication. If it is essen-
