Page:Mind (New Series) Volume 8.djvu/176

162 F. H. BRADLEY : conclusion drawn quite mistaken. The argument should prove, it seems to me, that memory is not fallible at all. Hence, when a particular memory is shown by reasoning to be false, we are left, it would appear, in hopeless confusion. For we must either accept both contradictories at once, or, if we select, we select on no principle, and surely this must be admitted to amount to scepticism. What we are to do when memory is thus divided against itself, and how mere memory is to sit in judgment on itself, are matters not explained. In short, that argument for the supremacy of reason which holds good against scepticism, becomes, if you transfer it to memory, wholly and entirely sceptical. 1 Prof. Ladd's conclusion then is really sceptical, but the foundation of his argument, to return to that, consists in a mistake. It is not the case that reasoning depends on memory, and such an idea implies a wrong view about in- ference. In the first place in inference there need be no premises drawn out and put before the mind, and a very large tract of our reasoning must in this sense be called intuitive. Prof. Ladd has seen this, but without more ado he drives the evidence bodily out of court. Everything of this kind is "a merely mechanical movement of the ideas," a con- clusion which I venture to regard as quite monstrous and a sufficient disproof of its foundation. That foundation is, however, in itself untenable. To assume that in an inference, where I go from premises to a conclusion, I depend upon memory, is to maintain that in inference I am necessitated en route not to know what I am about, and arrived at the end must have forgotten, and so be forced to remember, the start- 1 How is mere memory to be a ruler and judge of itself ? I cannot see how this is to be possible. If, on the other hand, memory is to subject itself to the judgment of reason, I cannot see how anywhere it is to claim independent authority, and to be treated as infallible or as more than de facto not mistaken . These are points on which I seek enlightenment so far in vain. If, for instance, it is urged that, in order to make the world intelligible, I must postulate that memory is right, unless so far as I have some special reason to think it anywhere wrong, I entirely agree. Certainly, I reply, and without doubt, we must make this assumption. But if, on this, I am told that, if so, we have an independent and ultimate postulate, I am forced to demur. Most evidently not so, I answer, if the assumption is made in order to make the world intelligible. If you leave out that, then, I agree, the postulate becomes ultimate, but it becomes at the same tune arbitrary and, so far as I see, quite indefensible. If we are to think at all, we must postulate that reason is in principle in- fallible, and is the ultimate judge of its own errors. But to postulate that memory is in principle infallible seems to me to be, on the one hand, wholly unnecessary and, for any legitimate purpose, quite useless ; and, on the other hand, it appears to me to be in the end really quite devoid of meaning.