SOME REMARKS ON MEMORY AND INFERENCE. 153 least represents what we aim at and seek to find in inference. It may help us to perceive this if we suppose that the type is modified. Let us assume no longer that b is c simply, but admit that b is c only by the help of x. The premise must now be written b(x) is c, and the old conclusion will not stand. We cannot any longer assert that A.b must be A.e. It only may be so, and, so far as it is so, it is so because of x. The A.b that is e is now not the A6 with which we started. We can no longer assert that the subject has been qualified throughout further without becoming something else. The subject of the conclusion is A6 together with a foreign condition x, and the conclusion is therefore con- ditioned, and, if you assert it of mere A.b, it is conditional or faulty. It is a defect of this kind which vitiates the result of mere imagination. That result we should agree has no necessity. In my mind's wandering the subject A6 may have actually now become A.-e, but we cannot add that the thing is so really and of itself, for A6, also and just as actually, may become something incompatible and may appear as A6-not-e. In mere imagination, because the thing may be otherwise, it is not really what it is. Necessity is not present, and necessity is absent because there is a breach of identity. The subject A.b becomes Ae, but you cannot add " of itself ". Something extraneous has at some point entered in and has vitiated the process, and you have passed from 6 to c not because b is c, but only because the passage has happened. An element has intervened not belonging directly to the pure essence of b, but attached to b merely as b is now present in psychical fact ; and it is this unknown addition, this x, which by a chance association has carried A.b to e. Such is the defect in identity which distinguishes mere imagination from inference, 1 and where this defect is remedied imagination becomes at once the strictest thinking. It may be instructive to notice here the superstition to which I referred. The distinction of mere imagination from thought consists in the absence or presence of logical control, and that control lies, as we have seen, in the preservation of ideal identity. But where this principle has not been grasped most incredible doctrines have found favour. Thought is abstract, we may be assured, while imagination is concrete. 2 1 Compare my Principles of Logic, p. 410. 2 See for example Prof. Sully, Human Mind, I., p. 384. He finds himself later in conflict with fact, and admits (p. 395, note) that the de- marcation is " not to be taken absolutely ". But the real question surely is whether the very principle of distinction is not false and contrary to fact,
