APPENDIX I THE PRAGMATIC METHOD, THE WILL-TO-BELIEVE, AND HUMANISM, IMMEDIATISM In almost all expositions of pragmatism that have received wide attention, a foremost place has been given to the so-called 'pragmatic method.' In spite of this prominence, the method has been, of all parts of the pragmatists' program, the most generally misunderstood both by the larger public and by the technical reviewers. How far the expositors have been to blame, and how far the incautious readers, we need not determine. But we shall try to profit by the experience of others, by putting into italics a warning against the commonest misunderstanding. The pragmatic method is a method of explanation, not of proof. It is used to determine the meaning of propositions; but, except in cases where it turns out that the proposition has no meaning at all, the truth or falsity of the proposition is not brought into question. It is true that the results of the exposition may be seized upon by the 'will-to-believe,' and the alternative of truth or falsity may be thus settled; but that is a further distinct step. The method is based upon the following assumption: that every distinction in meaning between ideas is a distinction between possibly desirable modes of conduct. It is inferred, that the meaning of a proposition may be determined by showing the differences in conduct which its truth or falsity would call for; while a proposition whose truth or falsity can under no conceivable circumstances affect the conduct of anyone is meaningless. Thus the meaning of the existence of God is that a man should persevere in right conduct, despite the apparent triumph of evil; and the meaning of the freedom of the will is that a man should not commit suicide from fear of ennui, but live in the expectation of continual novelties. Suppose the fundamental assumption to be correct.1 The method is nevertheless defective. It prescribes no means of determining whether the differences in conduct that are pointed out are the only ones that can arise from the truth or falsity of the proposition, or even that they are the sole important differences. A meaning is discovered; but no assurance is given that this is the whole meaning, or even the principal meaning of the proposition. Hence, even though the instrumental theory of meaning be correct, the pragmatic method is intrinsically fallacious. A possible exception to this general fallaciousness remains. If it is indeed demonstrable that the truth or falsity of a given proposition could in no way affect the advisability of conduct, the proposition must, upon the instrumental theory, be meaningless. But when we examine the illustrations of this contingency that are given by pragmatists, it becomes clear that the pragmatic method is entirely non-essential to them. What is invariably proved is that the proposition in question cannot be confirmed or contradicted by any conceivable experience; that is to say, whether the proposition is true or false, no possible experience would be different. In the words of the well-known formula (already quoted), there is no difference in the "sensations to be expected," and hence no difference in the "reactions that are to be prepared." But in such a case the proposition is meaningless, not only upon specifically pragmatist grounds, but on the basis of a pre-evolutionary empiricism. In fact, Berkeley's proof of the meaninglessness of the assumption of material substance that it is incapable of verification or disproof—is hailed by the pragmatists as an admirable application of their method. But the reference to conduct is altogether lacking. Now it is true that where there is no difference in the phenomenon there can be none in the behavior which it calls for; so that the practical reference can be freely supplied if one wishes. But it 1It should be noted, however, that this involves an isolation of 'import' from 'content,' which we can by no means admit. is a mere addendum, which contributes nothing to the force of the argument. The pragmatic method, then, is either fallacious or superfluous. In current philosophical literature the name 'pragmatism' has been used to cover any sort of attempt to eliminate ambiguity in the use of terms—perhaps from the conviction that any other mode of thought is at bottom mere verbalism. Thus the distinction of various senses in which the world may be said to be 'one' or 'many' is called pragmatic, though it is carried on as the veriest scholastic would require. To 'go around' an animal may mean to go north and east and south and west of him; or it may mean to go in front, on one side, in the rear, and on the other side of him; and to note the two-fold usage-though without the remotest suggestion of any practical difference to the animal or his satellite-is called pragmatism. But this simply robs the term of any controversial importance; and it has no warrant in the formal descriptions of the method, given by its advocates. A second feature of pragmatism, which we believe to be foreign to its deeper spirit, but which is popularly regarded as constituting its very essence, is the theory of the 'will-to-believe.' It may be formulated as follows: Where alternative hypotheses are presented, whose probability, so far as determined by existing evidence, seems fairly equal; and where the belief in the one alternative, were it verified by the event, would produce a satisfaction so far greater than would in any case follow either from uncertainty or from the acceptance of the other alternative, that any relative deficiency of happiness which might arise from the acceptance of the former, in case it were not verified, would be negligible in comparison; there a belief in the former hypothesis is warranted—that is to say, the former hypothesis may rightly be regarded as indefinitely the more probable. It must be noted that the above statement of the theory differs in one important respect from Professor James's enunciation. We have omitted the proviso that a choice of alternatives be necessary; for we cannot see that this is ever the case. Where a question is possible, doubt is always possible. But it is said that to doubt may be practically the same as to accept one or other of the given alternatives. This is true; and if the theory in question referred to the wisdom of action instead of the validity of belief, we should have no quarrel with it. But we must not confuse the acting on a chance with a confidence in the outcome. Hence in our statement of the will-to-believe principle, we have included the state of doubt as a third real possibility; grouping it, however, with the acceptance of one of the alternatives in such a way as to leave the pragmatist position virtually unchanged. It must next be noted that, according to the premises laid down, the happiness consequent upon belief is supposed to be directly thus consequent-not an after-effect of conduct dictated by the belief, but the immediate effect of the belief itself. For if the happiness were supposed to flow from a course of conduct, then that same course of conduct would be equally dictated by an uncertainty in the matter. That is to say, of two conflicting courses of conduct, having apparently equal chances of success or failure, a man would wisely choose the one which promised the greater gain in proportion to the risk involved, even though he had not the least confidence that a favorable rather than an unfavorable issue would result. In other words, an absolute uncertainty as to the result would logically warrant the same course of conduct as would be warranted by an entire conviction as to the certainty of a favorable outcome. The happiness proposed must, therefore, be conceived to be a direct fruit of the belief as such. How far this is removed from the spirit of pragmatism need not be emphasized. In order to escape this interpretation a new premise must be added to those above specified; namely, that even though the same conduct might be dictated by belief and by doubt, yet only the belief in the particular outcome could so strengthen a man as to enable him to act in the manner necessary for success. Now this is by no means an impossible supposition, and it is |