that particular passages to this effect are to be found in the works of the others. The more important fact is that the twofold conception of meaning—as content and import—is plainly implied in the pragmatist theory of truth; to which we now turn.

Truth is a property which we attribute to our beliefs-so far as we do, indeed, believe in them. Whether the particular beliefs actually possess this property or not,1 the meaning of the property itself, which is thus attributed to them, is of course unchanged. A method is accordingly suggested for analyzing our conception of truth; namely, the genetic method that consists in observing the conditions under which belief changes and the general features of the process of change-how doubt arises, how speculation proceeds, and how belief becomes reëstablished.

As a result of such observation, it is found that truth contains two essential factors, which (we would note in passing) are analogous to the two aspects of meaning already noted. One is consistency with other beliefs (including, by indirection, the beliefs in terms of consequences in life of some formula which has its content, its logical meaning, already fixed; or does he employ it to criticise and revise, and ultimately, to constitute the proper intellectual meaning of that formula?" And below (with reference to the pragmatic determination of the meaning of design in nature, as a 'vague confidence in the future'): "Is this meaning intended to replace the meaning of a 'seeing force which runs things'? Or is it intended to superadd a pragmatic value and validation to that concept of a seeing force? Or does it mean that, irrespective of the existence of any such object, a belief in it has that value? Strict pragmatism would seem to require the first interpretation, but I do not think that is what Mr. James intends." Journal of Philosophy, Psychology, and Scientific

Methods, V, pp. 90, 91.

1Cf. James, The Meaning of Truth, p. 183.

2On account of the one-sidedness of the usual pragmatist account of meaning, the writers of the school are unable to give a very definite account of this consistency, harmony, or agreement. We are told simply that we "feel" that certain ideas are in agreement with other parts of experience, "such feeling being among our potentialities” (Pragmatism, p. 201, cf. Meaning of Truth, p. 101, ll. 1-7). This is the old empiricist faculty of 'comparison' over again, with the important difference, to be sure, that the consciousness of agreement is (or may be) simultaneous with, rather than posterior to, the consciousness of the terms compared. But though the existence of such a faculty, or potentiality, be admitted, the problem certainly remains of determining under what conditions the feeling is felt. Even so, in the case of an externally excited sensation, such as sweet or bitter, we are not

of other men in whose judgment we have confidence); the other is the satisfactory guidance of conduct. The truth of an idea is, then, its workability in combination with our other ideas. Thus the interpretation of a new experience, in such a way as to conflict with a great body of accepted maxims, can hardly ever win our acceptance, no matter how successfully it suggests the conduct suitable to the circumstances. And, contrariwise, however beautifully a theory may harmonize with accepted notions, its persistent failure in practice not only condemns it but casts doubt upon the old notions as well. Change of belief is thus characterized by the continuity which belongs to evolution generally. Existing structures and functions are modified as slightly as possible, in accordance with new demands; and, moreover, such modification as occurs is always more apt to attach to recently acquired, than to older (and thus more deeply involved), features.

The truth-formula is most frequently presented by pragmatists in a form which consolidates the two factors. Recognizing that consistency is itself an important subject of human interest, they declare that the truth of an idea is its satisfactoriness—including the satisfaction of intellectual interests as well as of all others that may be involved. There may be matter for serious criticism here (as we hope hereafter to show); but in fairness it must be said that a mere confusion, in which the specific character wholly satisfied with the statement, that the experience of these sensations is a potentiality of our nature. We desire to know the general characteristics of their respective stimuli. It is a pressing problem of psychophysics. Even so the moralsense school of ethicists, who believed the feeling of approbation to be an original, fundamental endowment of our nature, recognized the problem of determining what the object of this peculiar reflective sense was. Indeed, they differed among themselves upon the matter, Shaftesbury, Hutcheson, and Hume having each his own characteristic theory. Now it is clear that logic has at least an equal interest in determining the general nature of the combinations of ideas (or other forms of experience) which are felt to agree. The mere fact that they are felt to agree is so far from being a solution that it is what sets the logical problem.

If we are correct in our interpretation, Professor James and his more immediate friends have formally deprived themselves of the only means of attacking, much less of solving, this problem. That the deprivation is only formal, and can be amended in full accordance with the general spirit of the pragmatist theory, we freely admit.

ΙΟ

of intellectual interests is lost sight of, is not to be attributed to pragmatists generally. The consolidated formula is, however, significant to this extent, that the various interests which may be active summate themselves in the total effect. The acceptance of a truth by no means implies either its perfect accordance with other accepted truths or the unmixed satisfactoriness of its practical working-out. It is "eminently a matter of approximation.” And, as elsewhere in human life, the choice of the best involves a compromise. To insist too rigidly on the theoretical criterion is the part of mere visionaries; to slight it almost entirely for the practical criterion is the part of short-sighted dolts. The average man is content with truth that avoids explicit self-contradiction and saves him from the ruder shocks. In the last resort, however, all this is a matter of individual taste. "We say this theory solves it on the whole more satisfactorily than that theory; but that means more satisfactorily to ourselves, and individuals will emphasize their points of satisfaction differently."1

It is noteworthy that belief, rather than knowledge, is the starting-point of the pragmatist epistemology. This has at least the controversial advantage, that while the very possibility of knowledge has been questioned, no one has dreamed of questioning the possibility of belief. The theory is thus founded upon patent matter of fact. It has, however, this difficulty. Truth is defined as a property attributed to beliefs. It thus remains undetermined whether any belief actually possesses this property; that is to say, is reasonably consistent with all other unquestioned beliefs, and is incapable of serious failure in practice. But the pragmatist, in a genuinely empirical spirit, does not hesitate to take his stand upon the beliefs actually and commonly entertained by men as true. Truths are for him, primarily at least, the truths of actual practice-that is to say, the beliefs that are recognized as true. The distinction between knowledge and belief is then interpreted as one of degree only. Our knowledge is simply the body of our best attested beliefs.

1Pragmatism, p. 61.

What becomes of the conception of an absolute knowledgeof beliefs possessed of absolute truth? It acquires the potent significance of an ideal limit. For the change of human beliefs is by no means altogether a mere fluctuation. In great part, it shows itself to be a gradual convergence; and this is especially true of the history of the sciences. Now a convergence may be conceived as having a finite terminus or as proceeding ad infinitum. In the case of the progress of knowledge, however, it is hard to see how the attainment of a terminus could be sufficiently attested. For it has happened repeatedly, that beliefs which for centuries have been regarded as possessing a certainty which nothing could surpass, are found to require correction. Nevertheless it may be admitted, that if a considerable body of science should remain for a great length of time without modification, men would feel obliged as they have felt under similar circumstances in the past-to regard such knowledge as ultimate. But from the vantage ground of the opening twentieth century, it seems far more natural to regard scientific progress as the convergence upon a goal which will never be definitely reached. The question whether the goal is attainable or not, is a question, which, from the present standpoint of science, leaves the meaning of the goal unaffected; for its attainment is beyond any reasonable expectation. Absolute truth is truth incapable of correction. Whether such truth can be secured, only time can tell.1

The pragmatist theory of reality offers serious difficulty to the expositor, and that for two reasons. In the first place, its most distinguished advocates are also believers in humanism or immediatism or both; and while they generally endeavor to keep these theories apart, human nature forbids that they should invariably succeed. In the second place, there is, we believe, a frequent ambiguity even in the definitely pragmatist usage of the

'This holds as a general statement of the pragmatist position in the matter. We shall hereafter have occasion to call attention to a class of absolutely true beliefs, which Mr. James believes to be even now entertained by us. Our belief that two and one make three is an example.

term 'reality.' That is to say, the term denotes either a belief, qualified as knowledge, or the things and relations which make up the object of the belief. In Mr. James's Pragmatism, these figure as distinct kinds of realities, with which a new idea must 'agree' if it is to be accepted as true. Now it appears to us perfectly clear, that the belief and its object are not kinds of realities (as if 'reality' were a generic term comprehending them both), but realities in different senses of the term. In a later volume the author of Pragmatism assumes that "the only realities we can talk about" are objects-believed-in.1 This we take to be obviously the better statement, and we propose to hold to it in this place.

Reality, then, may be said to have two aspects corresponding to the two factors in truth itself. In the first place, it is that with which our ideas must agree if they are to be true. In the second place, it is that to which our conduct must conform if it is to be satisfactory. More briefly, it is on the one hand the object of knowledge, and on the other hand the condition of success and failure. It is a principal object of the pragmatists to exhibit the essential unity of these two aspects, and they do not consider them separately. We think, however, that for the purposes of the present exposition a brief separate treatment may be helpful.

Reality, as the object of knowledge, is conceived to be relative or absolute, according as the knowledge itself is accepted as relative or absolute. Primarily, reality means the realities of actual experience and expectation. Though, upon sufficient reflection we may admit that these realities have not been definitely ascertained, nevertheless, in so far as we naïvely accept them, we accept them as if they were absolute-that is to say, as perfect standards to which our other beliefs (as well as the beliefs of other men) must, if they are to be true, exactly conform. They are believed in as if their existence were independent of the present belief itself; as if a change of belief would be a change from true to false, leaving the reality itself unchanged. The

1The Meaning of Truth, p. 236. The whole passage is a silent, perhaps unconscious, correction of the looser exposition given in Lecture VI of Pragmatism.