THE AMBIGUITY OF TRUTH. 167 of the positive solutions of the problem, and assumes the conceptions of ' self-contradiction ' or ' incoherence ' as the simplest things in the world. In point of fact neither of them has been adequately analysed by intellectualist logicians, nor is either of them naturally so translucent as to shed a flood of light on any subject. As, however, I cannot now enter upon their obscurities, it must suffice to remark that Capt. Knox's masterly article in the April (1905) number of MIND l contains ample justification for what I have said about the principle of contradiction. If on the other hand the ' negative criterion ' be stated in the form of incoherence, I would inquire merely how intellectualist logic proposes to distinguish the logical coherence, to which it appeals, from the psychological coherence, which it despises. Until this difficult (or impossible?) feat has been achieved, we may safely move on. 2 III. Let us proceed therefore to discard old prejudices and to consider how in point of fact we discriminate between 'claims ' and 'truths,' how the raw material of a science is elaborated into its final structure, how in short, truth is made. Now this question is not intrinsically a hopeless one. It is not even theoretically particularly difficult to answer. For it concerns essentially facts which may be observed, and with care and attention it should be possible to deter- mine whether the procedures of the various sciences have anything in common, and if so what. By such an inductive appeal to the facts, therefore, we greatly simplify our pro- blem, and may possibly discover its solution. Any obstacle which we may encounter will come merely from the difficulty of intelligently observing the special procedures of the many sciences and of seizing their salient points and general import ; we shall not be foredoomed to failure by any in- trinsic absurdity of our enterprise. Now it would be possible, I think, to arrive at our solution by a critical examination of every known science in detail, but it is evident that this procedure would be very long and laborious. It seems better, therefore, merely to state the condensed results of such investigations. They will in this shape stand out more clearly and better exhibit the trend of an argument which runs as follows. It being taken as established that the sphere of logic is that of the antithetical valuations 'true' and 'false,' we observe in the first place that in every science the truth or falsehood of an answer depends on its relevance to the 1 N.S., No. 54. 2 Cf. also Humanism, pp. 52-53.
