Page:Mind (New Series) Volume 8.djvu/169

 SOME REMARKS ON MEMORY AND INFERENCE. 155 tion in various forms, intellectual, practical and aesthetic. It is the special nature of the end and the special nature of the control which makes the difference in principle, and in the case of inference we have seen in what that difference consists. 1 From this our enquiry may return to the subject of memory. The mere imagination of the past, we have seen, is, like in- ference, an ideal construction from the present, and yet it fails to be inference. Memory is also an ideal construction from the present, and thus we are led to ask in what way memory differs from inference and from fancy ; for that there is some difference seems plain. I may, to repeat our instance, infer that on last Monday I must have posted a letter, or I may remember the fact, or again I may merely imagine it, and these three attitudes are not the same. Now, as against fancy, it is clear that memory has necessity. It does not qualify its subject by a predicate the opposite of which can also be remembered, and which for this reason does not qualify the subject itself. Memory, in other words, is a judg- ment and an assertion about its subject. Hence it is often again said to involve belief, a point which I shall consider lower down. Thus memory being a judgment is so far the same as inference, and we must go on to ask if they are the same altogether. If inference is understood in the sense in which we have taken it above, inference and memory certainly differ. For in memory there is a sequence and a continuity which is necessary, but on the other hand the necessity is not wholly intrinsic, or, if wholly intrinsic, is not so visibly. We do not, as in inference, go from A.b to A.bc, because b is c. The se- quence in memory cannot be so stated. The premises are not Ai, be, but must be written as A6, Be. Now certainly b is contained in and is an element in B, but, with only so much, the sequence fails to be logical. For you cannot logi- cally proceed from Ab, Be, to A-c, unless you assume that Be is equivalent, say, to 6-B-c, and not merely to b(x)-c. The essential question is as to how the difference, which turns 1 I do not know whether Wundt (Grundzilge, II., p. 490) really means to say that all imagination involves a plan and an idea which it develops. Such a statement seems to be in collision with the obvious fact of mental wandering. The nature of the different kinds of control over mere wan- dering is, so far as I see, the only ground from which this whole question could be satisfactorily treated. I certainly could not myself attempt that treatment, and I do not myself know where to send the reader for satis- faction. Wundt's exposition seems not only confused in detail but based on no clear principle whatever. Such principles of division as " passive " and " active " are, for instance, much worse than merely useless.