[Footnote: Reprint from the Journal of Philosophy, July 18,1907.]

My failure in making converts to my conception of truth seems, if I may judge by what I hear in conversation, almost complete. An ordinary philosopher would feel disheartened, and a common choleric sinner would curse God and die, after such a reception. But instead of taking counsel of despair, I make bold to vary my statements, in the faint hope that repeated droppings may wear upon the stone, and that my formulas may seem less obscure if surrounded by something more of a 'mass' whereby to apperceive them.

For fear of compromising other pragmatists, whoe'er they be, I will speak of the conception which I am trying to make intelligible, as my own conception. I first published it in the year 1885, in the first article reprinted in the present book. Essential theses of this article were independently supported in 1893 and 1895 by Professor D. S. Miller [Footnote: Philosophical Review, vol. ii, p. 408, and Psychological Review, vol. ii, p. 533.] and were repeated by me in a presidential address on 'The knowing of things together' [Footnote: The relevant parts of which are printed above, p. 43.] in 1895. Professor Strong, in an article in the Journal of Philosophy, etc., [Footnote: Vol. i, p. 253.] entitled 'A naturalistic theory of the reference of thought to reality,' called our account 'the James-Miller theory of cognition,' and, as I understood him, gave it his adhesion. Yet, such is the difficulty of writing clearly in these penetralia of philosophy, that each of these revered colleagues informs me privately that the account of truth I now give—which to me is but that earlier statement more completely set forth—is to him inadequate, and seems to leave the gist of real cognition out. If such near friends disagree, what can I hope from remoter ones, and what from unfriendly critics?

Yet I feel so sure that the fault must lie in my lame forms of statement and not in my doctrine, that I am fain to try once more to express myself.

Are there not some general distinctions which it may help us to agree about in advance? Professor Strong distinguishes between what he calls 'saltatory' and what he calls 'ambulatory' relations. 'Difference,' for example, is saltatory, jumping as it were immediately from one term to another, but 'distance' in time or space is made out of intervening parts of experience through which we ambulate in succession. Years ago, when T. H. Green's ideas were most influential, I was much troubled by his criticisms of english sensationalism. One of his disciples in particular would always say to me, 'Yes! TERMS may indeed be possibly sensational in origin; but RELATIONS, what are they but pure acts of the intellect coming upon the sensations from above, and of a higher nature?' I well remember the sudden relief it gave me to perceive one day that SPACE-relations at any rate were homogeneous with the terms between which they mediated. The terms were spaces, and the relations were other intervening spaces. [Footnote: See my Principles of Psychology, vol. ii, pp. 148-153.] For the Greenites space-relations had been saltatory, for me they became thenceforward ambulatory.

Now the most general way of contrasting my view of knowledge with the popular view (which is also the view of most epistemologists) is to call my view ambulatory, and the other view saltatory; and the most general way of characterizing the two views is by saying that my view describes knowing as it exists concretely, while the other view only describes its results abstractly taken.

I fear that most of my recalcitrant readers fail to recognize that what is ambulatory in the concrete may be taken so abstractly as to appear saltatory. Distance, for example, is made abstract by emptying out whatever is particular in the concrete intervals—it is reduced thus to a sole 'difference,' a difference of 'place,' which is a logical or saltatory distinction, a so-called 'pure relation.'

The same is true of the relation called 'knowing,' which may connect an idea with a reality. My own account of this relation is ambulatory through and through. I say that we know an object by means of an idea, whenever we ambulate towards the object under the impulse which the idea communicates. If we believe in so-called 'sensible' realities, the idea may not only send us towards its object, but may put the latter into our very hand, make it our immediate sensation. But, if, as most reflective people opine, sensible realities are not 'real' realities, but only their appearances, our idea brings us at least so far, puts us in touch with reality's most authentic appearances and substitutes. In any case our idea brings us into the object's neighborhood, practical or ideal, gets us into commerce with it, helps us towards its closer acquaintance, enables us to foresee it, class it, compare it, deduce it,—in short, to deal with it as we could not were the idea not in our possession.

The idea is thus, when functionally considered, an instrument for enabling us the better to HAVE TO DO with the object and to act about it. But it and the object are both of them bits of the general sheet and tissue of reality at large; and when we say that the idea leads us towards the object, that only means that it carries us forward through intervening tracts of that reality into the object's closer neighborhood, into the midst of its associates at least, be these its physical neighbors, or be they its logical congeners only. Thus carried into closer quarters, we are in an improved situation as regards acquaintance and conduct; and we say that through the idea we now KNOW the object better or more truly.

My thesis is that the knowing here is MADE by the ambulation through the intervening experiences. If the idea led us nowhere, or FROM that object instead of towards it, could we talk at all of its having any cognitive quality? Surely not, for it is only when taken in conjunction with the intermediate experiences that it gets related to THAT PARTICULAR OBJECT rather than to any other part of nature. Those intermediaries determine what particular knowing function it exerts. The terminus they guide us to tells us what object it 'means,' the results they enrich us with 'verify' or 'refute' it. Intervening experiences are thus as indispensable foundations for a concrete relation of cognition as intervening space is for a relation of distance. Cognition, whenever we take it concretely, means determinate 'ambulation,' through intermediaries, from a terminus a quo to, or towards, a terminus ad quem. As the intermediaries are other than the termini, and connected with them by the usual associative bonds (be these 'external' or be they logical, i.e., classificatory, in character), there would appear to be nothing especially unique about the processes of knowing. They fall wholly within experience; and we need use, in describing them, no other categories than those which we employ in describing other natural processes.

But there exist no processes which we cannot also consider abstractly, eviscerating them down to their essential skeletons or outlines; and when we have treated the processes of knowing thus, we are easily led to regard them as something altogether unparalleled in nature. For we first empty idea, object and intermediaries of all their particularities, in order to retain only a general scheme, and then we consider the latter only in its function of giving a result, and not in its character of being a process. In this treatment the intermediaries shrivel into the form of a mere space of separation, while the idea and object retain only the logical distinctness of being the end-terms that are separated. In other words, the intermediaries which in their concrete particularity form a bridge, evaporate ideally into an empty interval to cross, and then, the relation of the end-terms having become saltatory, the whole hocus-pocus of Erkenntnistheorie begins, and goes on unrestrained by further concrete considerations. The idea, in 'meaning' an object separated by an 'epistemological chasm' from itself, now executes what Professor Ladd calls a 'salto mortale'; in knowing the object's nature, it now 'transcends' its own. The object in turn becomes 'present' where it is really absent, etc.; until a scheme remains upon our hands, the sublime paradoxes of which some of us think that nothing short of an 'absolute' can explain.

The relation between idea and object, thus made abstract and saltatory, is thenceforward opposed, as being more essential and previous, to its own ambulatory self, and the more concrete description is branded as either false or insufficient. The bridge of intermediaries, actual or possible, which in every real case is what carries and defines the knowing, gets treated as an episodic complication which need not even potentially be there. I believe that this vulgar fallacy of opposing abstractions to the concretes from which they are abstracted, is the main reason why my account of knowing is deemed so unsatisfactory, and I will therefore say a word more on that general point.

Any vehicle of conjunction, if all its particularities are abstracted from it, will leave us with nothing on our hands but the original disjunction which it bridged over. But to escape treating the resultant self-contradiction as an achievement of dialectical profundity, all we need is to restore some part, no matter how small, of what we have taken away. In the case of the epistemological chasm the first reasonable step is to remember that the chasm was filled with SOME empirical material, whether ideational or sensational, which performed SOME bridging function and saved us from the mortal leap. Restoring thus the indispensable modicum of reality to the matter of our discussion, we find our abstract treatment genuinely useful. We escape entanglement with special cases without at the same time falling into gratuitous paradoxes. We can now describe the general features of cognition, tell what on the whole it DOES FOR US, in a universal way.

We must remember that this whole inquiry into knowing grows up on a reflective level. In any real moment of knowing, what we are thinking of is our object, not the way in which we ourselves are momentarily knowing it. We at this moment, as it happens, have knowing itself for our object; but I think that the reader will agree that his present knowing of that object is included only abstractly, and by anticipation, in the results he may reach. What he concretely has before his mind, as he reasons, is some supposed objective instance of knowing, as he conceives it to go on in some other person, or recalls it from his own past. As such, he, the critic, sees it to contain both an idea and an object, and processes by which the knower is guided from the one towards the other. He sees that the idea is remote from the object, and that, whether through intermediaries or not, it genuinely HAS TO DO with it. He sees that it thus works beyond its immediate being, and lays hold of a remote reality; it jumps across, transcends itself. It does all this by extraneous aid, to be sure, but when the aid has come, it HAS done it and the result is secure. Why not talk of results by themselves, then, without considering means? Why not treat the idea as simply grasping or intuiting the reality, of its having the faculty anyhow, of shooting over nature behind the scenes and knowing things immediately and directly? Why need we always lug in the bridging?—it only retards our discourse to do so.

Such abstract talk about cognition's results is surely convenient; and it is surely as legitimate as it is convenient, SO LONG AS WE DO NOT FORGET OR POSITIVELY DENY, WHAT IT IGNORES. We may on occasion say that our idea meant ALWAYS that particular object, that it led us there because it was OF it intrinsically and essentially. We may insist that its verification follows upon that original cognitive virtue in it—and all the rest—and we shall do no harm so long as we know that these are only short cuts in our thinking. They are positively true accounts of fact AS FAR AS THEY GO, only they leave vast tracts of fact out of the account, tracts of tact that have to be reinstated to make the accounts literally true of any real case. But if, not merely passively ignoring the intermediaries, you actively deny them [Footnote: This is the fallacy which I have called 'vicious intellectualism' in my book A Pluralistic Universe, Longmans, Green & Co., 1909.] to be even potential requisites for the results you are so struck by, your epistemology goes to irremediable smash. You are as far off the track as an historian would be, if, lost in admiration of Napoleon's personal power, he were to ignore his marshals and his armies, and were to accuse you of error in describing his conquests as effected by their means. Of such abstractness and one-sidedness I accuse most of the critics of my own account.

In the second lecture of the book Pragmatism, I used the illustration of a squirrel scrambling round a tree-trunk to keep out of sight of a pursuing man: both go round the tree, but does the man go round the squirrel? It all depends, I said, on what you mean by going round.' In one sense of the word the man 'goes round,' in another sense he does not. I settled the dispute by pragmatically distinguishing the senses. But I told how some disputants had called my distinction a shuffling evasion and taken their stand on what they called 'plain honest English going-round.'

In such a simple case few people would object to letting the term in dispute be translated into its concreter equivalents. But in the case of a complex function like our knowing they act differently. I give full concrete particular value for the ideas of knowing in every case I can think of, yet my critics insist that 'plain honest English knowing' is left out of my account. They write as if the minus were on my side and the plus on theirs.

The essence of the matter for me is that altho knowing can be both abstractly and concretely described, and altho the abstract descriptions are often useful enough, yet they are all sucked up and absorbed without residuum into the concreter ones, and contain nothing of any essentially other or higher nature, which the concrete descriptions can be justly accused of leaving behind. Knowing is just a natural process like any other. There is no ambulatory process whatsoever, the results of which we may not describe, if we prefer to, in saltatory terms, or represent in static formulation. Suppose, e.g., that we say a man is 'prudent.' Concretely, that means that he takes out insurance, hedges in betting, looks before he leaps. Do such acts CONSTITUTE the prudence? ARE they the man qua prudent?

Or is the prudence something by itself and independent of them? As a constant habit in him, a permanent tone of character, it is convenient to call him prudent in abstraction from any one of his acts, prudent in general and without specification, and to say the acts follow from the pre-existing prudence. There are peculiarities in his psycho-physical system that make him act prudently; and there are tendencies to association in our thoughts that prompt some of them to make for truth and others for error. But would the man be prudent in the absence of each and all of the acts? Or would the thoughts be true if they had no associative or impulsive tendencies? Surely we have no right to oppose static essences in this way to the moving processes in which they live embedded.

My bedroom is above my library. Does the 'aboveness' here mean aught that is different from the concrete spaces which have to be moved-through in getting from the one to the other? It means, you may say, a pure topographic relation, a sort of architect's plan among the eternal essences. But that is not the full aboveness, it is only an abbreviated substitute that on occasion may lead my mind towards truer, i.e., fuller, dealings with the real aboveness. It is not an aboveness ante rem, it is a post rem extract from the aboveness in rebus. We may indeed talk, for certain conveniences, as if the abstract scheme preceded, we may say 'I must go up stairs because of the essential aboveness,' just as we may say that the man 'does prudent acts because of his ingrained prudence,' or that our ideas 'lead us truly because of their intrinsic truth.' But this should not debar us on other occasions from using completer forms of description. A concrete matter of fact always remains identical under any form of description, as when we say of a line, now that it runs from left to right, and now that it runs from right to left. These are but names of one and the same fact, one more expedient to use at one time, one at another. The full facts of cognition, whatever be the way in which we talk about them, even when we talk most abstractly, stand inalterably given in the actualities and possibilities of the experience-continuum. [Footnote 1: The ultimate object or terminus of a cognitive process may in certain instances lie beyond the direct experience of the particular cognizer, but it, of course, must exist as part of the total universe of experience whose constitution, with cognition in it, the critic is discussing.] But my critics treat my own more concrete talk as if IT were the kind that sinned by its inadequacy, and as if the full continuum left something out.

A favorite way of opposing the more abstract to the more concrete account is to accuse those who favor the latter of 'confounding psychology with logic.' Our critics say that when we are asked what truth MEANS, we reply by telling only how it is ARRIVED-AT. But since a meaning is a logical relation, static, independent of time, how can it possibly be identified, they say, with any concrete man's experience, perishing as this does at the instant of its production? This, indeed, sounds profound, but I challenge the profundity. I defy any one to show any difference between logic and psychology here. The logical relation stands to the psychological relation between idea and object only as saltatory abstractness stands to ambulatory concreteness. Both relations need a psychological vehicle; and the 'logical' one is simply the 'psychological' one disemboweled of its fulness, and reduced to a bare abstractional scheme.

A while ago a prisoner, on being released, tried to assassinate the judge who had sentenced him. He had apparently succeeded in conceiving the judge timelessly, had reduced him to a bare logical meaning, that of being his 'enemy and persecutor,' by stripping off all the concrete conditions (as jury's verdict, official obligation, absence of personal spite, possibly sympathy) that gave its full psychological character to the sentence as a particular man's act in time. Truly the sentence WAS inimical to the culprit; but which idea of it is the truer one, that bare logical definition of it, or its full psychological specification? The anti-pragmatists ought in consistency to stand up for the criminal's view of the case, treat the judge as the latter's logical enemy, and bar out the other conditions as so much inessential psychological stuff.

II

A still further obstacle, I suspect, stands in the way of my account's acceptance. Like Dewey and like Schiller, I have had to say that the truth of an idea is determined by its satisfactoriness. But satisfactoriness is a subjective term, just as idea is; and truth is generally regarded as 'objective.' Readers who admit that satisfactoriness is our only MARK of truth, the only sign that we possess the precious article, will still say that the objective relation between idea and object which the word 'truth' points to is left out of my account altogether. I fear also that the association of my poor name with the 'will to believe' (which 'will,' it seems to me, ought to play no part in this discussion) works against my credit in some quarters. I fornicate with that unclean thing, my adversaries may think, whereas your genuine truth-lover must discourse in huxleyan heroics, and feel as if truth, to be real truth, ought to bring eventual messages of death to all our satisfactions. Such divergences certainly prove the complexity of the area of our discussion; but to my mind they also are based on misunderstandings, which (tho with but little hope of success) I will try to diminish by a further word of explanation.

First, then, I will ask my objectors to define exactly what SORT of thing it is they have in mind when they speak of a truth that shall be absolute, complete and objective; and then I will defy them to show me any conceivable standing-room for such a kind of truth outside the terms of my own description. It will fall, as I contend, entirely within the field of my analysis.

To begin with, it must obtain between an idea and a reality that is the idea's object; and, as a predicate, it must apply to the idea and not to the object, for objective realities are not TRUE, at least not in the universe of discourse to which we are now confining ourselves, for there they are taken as simply BEING, while the ideas are true OF them. But we can suppose a series of ideas to be successively more and more true of the same object, and can ask what is the extreme approach to being absolutely true that the last idea might attain to.

The maximal conceivable truth in an idea would seem to be that it should lead to an actual merging of ourselves with the object, to an utter mutual confluence and identification. On the common-sense level of belief this is what is supposed really to take place in sense-perception. My idea of this pen verifies itself through my percept; and my percept is held to BE the pen for the time being—percepts and physical realities being treated by common sense as identical. But the physiology of the senses has criticised common sense out of court, and the pen 'in itself' is now believed to lie beyond my momentary percept. Yet the notion once suggested, of what a completely consummated acquaintance with a reality might be like, remains over for our speculative purposes. TOTAL CONFLUX OF THE MIND WITH THE REALITY would be the absolute limit of truth, there could be no better or more satisfying knowledge than that.

Such total conflux, it is needless to say, is ALREADY EXPLICITLY PROVIDED FOR, AS A POSSIBILITY, IN MY ACCOUNT OF THE MATTER. If an idea should ever lead us not only TOWARDS, or UP TO, or AGAINST, a reality, but so close that we and the reality should MELT TOGETHER, it would be made absolutely true, according to me, by that performance.

In point of fact philosophers doubt that this ever occurs. What happens, they think, is only that we get nearer and nearer to realities, we approximate more and more to the all-satisfying limit; and the definition of actually, as distinguished from imaginably, complete and objective truth, can then only be that it belongs to the idea that will lead us as CLOSE UP AGAINST THE OBJECT as in the nature of our experience is possible, literally NEXT to it, for instance.

Suppose, now, there were an idea that did this for a certain objective reality. Suppose that no further approach were possible, that nothing lay between, that the next step would carry us right INTO the reality; then that result, being the next thing to conflux, would make the idea true in the maximal degree that might be supposed practically attainable in the world which we inhabit.

Well, I need hardly explain that THAT DEGREE OF TRUTH IS ALSO PROVIDED FOR IN MY ACCOUNT OF THE MATTER. And if satisfactions are the marks of truth's presence, we may add that any less true substitute for such a true idea would prove less satisfactory. Following its lead, we should probably find out that we did not quite touch the terminus. We should desiderate a closer approach, and not rest till we had found it.

I am, of course, postulating here a standing reality independent of the idea that knows it. I am also postulating that satisfactions grow pari passu with our approximation to such reality. [Footnote 1: Say, if you prefer to, that DISsatisfactions decrease pari passu with such approximation. The approximation may be of any kind assignable—approximation in time or in space, or approximation in kind, which in common speech means 'copying.'] If my critics challenge this latter assumption, I retort upon them with the former. Our whole notion of a standing reality grows up in the form of an ideal limit to the series of successive termini to which our thoughts have led us and still are leading us. Each terminus proves provisional by leaving us unsatisfied. The truer idea is the one that pushes farther; so we are ever beckoned on by the ideal notion of an ultimate completely satisfactory terminus. I, for one, obey and accept that notion. I can conceive no other objective CONTENT to the notion of ideally perfect truth than that of penetration into such a terminus, nor can I conceive that the notion would ever have grown up, or that true ideas would ever have been sorted out from false or idle ones, save for the greater sum of satisfactions, intellectual or practical, which the truer ones brought with them. Can we imagine a man absolutely satisfied with an idea and with all its relations to his other ideas and to his sensible experiences, who should yet not take its content as a true account of reality? The matter of the true is thus absolutely identical with the matter of the satisfactory. You may put either word first in your ways of talking; but leave out that whole notion of SATISFACTORY WORKING or LEADING (which is the essence of my pragmatistic account) and call truth a static logical relation, independent even of POSSIBLE leadings or satisfactions, and it seems to me you cut all ground from under you.

I fear that I am still very obscure. But I respectfully implore those who reject my doctrine because they can make nothing of my stumbling language, to tell us in their own name—und zwar very concretely and articulately!—just how the real, genuine and absolutely 'objective' truth which they believe in so profoundly, is constituted and established. They mustn't point to the 'reality' itself, for truth is only our subjective relation to realities. What is the nominal essence of this relation, its logical definition, whether or not it be 'objectively' attainable by mortals?

Whatever they may say it is, I have the firmest faith that my account will prove to have allowed for it and included it by anticipation, as one possible case in the total mixture of cases. There is, in short, no ROOM for any grade or sort of truth outside of the framework of the pragmatic system, outside of that jungle of empirical workings and leadings, and their nearer or ulterior terminations, of which I seem to have written so unskilfully.