The last chapter closed with some incidental observations on the continuity of material objects, and the grounds on which we believe in it. But I touched only on the outer fringe of a subject which certainly deserves some further mention, if only for the purpose of indicating its difficulty and its importance. Its discussion formed no part of my original plan, and what I now propose to say about it will scarcely extend beyond the limits of a note. This, however incomplete, may serve to indicate a gap in my treatment of perception, though not, I fear, to fill it.
It will be remembered that when dealing with the problem of our familiar beliefs about the external world, we proceeded on the usual assumption that only through the avenues of sense could particular facts about it be immediately known; and doubtless, up to a point, such an assumption is both true and convenient. Yet are not cases of continuity and of change to be counted among particular matters of external fact? And if so, through what avenues of sense are they revealed? Can we see them or feel them; do they emit scent or sound?
If the answer is negative, as I think it should be, it seems to prove that what is immediately apprehended in perception includes much more than can be gathered through the senses. To what an empty husk would perceptual experience be reduced if it made us immediately acquainted neither with continuity nor with change! What value would the most compact series of momentary sensations possess if the message of each had to be considered in isolation, if there was no machinery for intuitively grasping them as successive revelations of an enduring whole!
There is such a machinery—the machinery of memory. And if this be admitted, it seems to follow that memory is, of all our sources of intuitive knowledge, perhaps the most important.
It is the most important, because without it no other sources would be of the smallest value. We might enjoy stretches of continuous perception. But continuous perception is quite a different thing from the perception of continuity, and much less valuable. No amount of industry could construct any sort of creed out of the dust of unremembered experiences. For these would have perished in the very moment of their birth; and even if their former existence could be revealed to us by some outside historian, what we should thereby learn we should learn indirectly, and on his authority, not directly and on theirs.
These general reflections, which I shall proceed immediately to develop, are sufficient to show how all-important is the part played by memory in the creation of our creed, and how desirable it would be to understand its bearing, cognitive and causal, on the general body of our familiar beliefs about the external world. This, I admit, is but a fragment of a great subject. Memory has its psychological and physiological sides; educationists are interested in its improvement, doctors in its pathology; while M. Bergson, in the wide sweep of his admirable speculations in Matiére et Mémoire, has dwelt on its relation to action, and its bearing on the metaphysics of body and mind. The point on which I propose to say a few words is a comparatively narrow one, touching only the relation between memory and perception.
At first sight the problem seems simple enough. Might we not solve it by saying that perception is confined to the present, while memory deals only with the past—that perception tells us of contemporary facts, while memory keeps a copy of the communication? Such a statement seems plausible enough, but is not without its difficulties.
To begin with, if we wish to be precise, we cannot say that perception tells us about the present; for as we saw in a preceding chapter,1 this is one of the things that perception never does. External experience always lags behind fact; sometimes a little, sometimes a great deal; but always enough to falsify the generalisation.
To simplify matters, however, it may be convenient for the moment to assume, in accordance with ordinary usage, that what we perceive exists or happens at the moment when we perceive it, and that when the moment is over, memory takes up and preserves, in some kind of psychological cold storage, what perception has just surrendered.
This seems a fair division of labour; let us consider its consequences. It proceeds on the view, which I think we must accept, that only at the ever-moving present do we come into touch with external reality; that through this point of time flows all the perceptual knowledge which memory retains and reason uses; that here is the sole port of entry for the sensuous material of our familiar creed.
This seems obvious—almost tautological. But how is it possible? The pure present is but a fleeting instant. Extend it by as little as you please, and immediately it embraces either a future which can be known only by conjecture, or a past which can be known only by recollection—leaving us, as before, to do all our perceiving through the infinitesimal fissure, or more accurately at the moving frontier, where past and future meet.
This seems to follow inevitably from the nature of Time. But how is it to be harmonised with any tolerable theory of life in general or perception in particular? Monsieur Bergson, for example, tells us that perceptions always possess some degree of “thickness” (épaisseur). I incline to his opinion. And even those (if any) who think the statement excessive, who suppose themselves able to recall flashes of perception, which in the strictest sense were momentary, will admit such experiences to be rare. How do they propose to treat the “thick” perceptions which are in the vast majority? Evidently there is a serious incongruity between such “thickness” and a merely momentary present. Is there, then, a present which is not momentary—a present which endures? Can time stand still, its pauses be measured, their contents held up for our inspection?
The idea is fantastic. But before we try and find a better, let us consider another difficulty, or (it may be) the same difficulty from another point of view.
This new problem is primarily connected with the psycho-physical machinery of sense perception. Clearly this cannot work within the limits of the passing present. Always it requires temporal “elbow room.” Seeing and hearing, for instance (I take the examples which are at once the most important and the simplest), both depend upon the multitudinous iteration of similar impacts. A single electromagnetic disturbance gives no light, a single air wave gives no sound; yet each of these, however brief it be, outlasts the moment in which it began. How much more must the number needful to produce a sensible experience involve the lapse of time?
From this it seems to follow that before sense perceptions begin, a work resembling that of memory must have been performed on the unnoticed impressions which, in the process of perception-making, successively reach us from outside. Time must somehow have been cheated; and a temporal series must have been compressed into a single sensation, as the events of a day or of a decade are compressed,2 however imperfectly, into a flash of memory. The analogy, no doubt, is in some respects remote. But I think it is real.
It seems then that, so far as the external world is concerned, we never live in the present at all. A moving present there always is; but we never feel it, or know it, or remember it. Our experiences are always of the past. And this not merely on account of the lag in the causal process of perception, due to transmission through space. This I have already mentioned, and for the moment put on one side. My present contention depends on the fact that the physical causes of perception can only give rise to their characteristic experiences by cumulative action. And since they (e.g. a train of light waves) proceed in single file at a finite speed, they must always occupy time, they must always belong more to the past than the present, they must always require something resembling memory to compound their products into a sensation.
Quite apart, however, from these psychophysical considerations, it is plain that if we regard the present as the dividing-point between past and future, a moving limit and no more, it cannot contain anything which we are able to isolate in reflection, or treat as an experience. It is true that through this momentary present stream all the influences that reach us from the outer world. But it is only when these are collected, compressed, endowed with a certain unity, and with what M. Bergson calls “thickness,” that they crystallise into perceptions, take their place in what we loosely (and erroneously) call the present, and count for something in our self-conscious life.
Now this is the work of memory, or something like it. It is at least the work of a faculty which preserves, prolongs, and combines, which is in constant touch with the true present, and is not wholly severed from the distant past. Evidently, then, I was greatly within the mark when, at the beginning of this chapter, I observed that without the aid of memory, sense perception could give us no information about such fundamental facts as continuity and motion, duration and change. We must, it seems, go much further. We must admit that without memory sense perception tells us nothing. Memory does more than occasionally revive what is old; it continually creates what is new. It creates, among other things, what has been called the specious present, that brief and indeterminate period which the true present steals, as it were, from the immediate past, and to which it gives its name. The crack of a whip, the flash of a gun, unquestionably take time. But it would be on the verge of pedantry to say, at the end of one of those events, that we only knew its beginning by the aid of memory! Yet this, I suppose, is the fact.
It must be owned that the whole process is somewhat elusive. It looses its precision of outline at the very moment we hope to grasp it. This is the more disconcerting since if there be a subject about which we should expect our ideas to be, not only clear but obvious, it is surely the Present. The past may easily be forgotten, the future is always unknown, but doubt or difficulty about what lies between them might well appear impossible. Yet it is not so. Analyse the idea of Time, and the real present narrows to a vanishing point. Consider the practice of life, and the “specious” (or working) present includes a fragment of the past. The content of the first is never separately perceived, nor separately recollected. The smallest element of the second already owes something to memory. There is no sensation so brief but it fills more than a moment in the making.
We must thus accept the view that the causal process of perception, which begins (for our purposes) in the object perceived, cannot issue in beliefs unhelped by memory. What reaction has this conclusion upon our estimate of the cognitive process which takes us back, or ought to take us back, by some trustworthy road, intuitive or inferential, to a knowledge of the object? To this question I need attempt no detailed reply. It is sufficient to say that if there were difficulties in the simpler theories of sense perception discussed in Chapters VII and VIII, they cannot but be aggravated by the fresh complications which the present argument has disclosed. To all the perplexities which obscure our theories of external experience are added the mysteries inseparable from memory. None of the old riddles have been solved, and surely some new ones present themselves for solution. It could hardly indeed be otherwise. We must expect that with every additional stage introduced into the processes by which sense perceptions are produced, it becomes easier to understand why things as perceived so greatly differ from things as they are, and more difficult to justify the procedure by which truth of fact is extracted from ambiguity or falsity of appearance. No satisfactory solution of these and other allied problems is provided, so far as I am aware, by any of the philosophies, still less by any of the sciences, which have dealt with our perceptual access to the world of matter. Our methodological doubts remain for the present unresolved. We must be content to assume what we can neither prove nor even render intellectually plausible—namely, the general validity of our familiar beliefs about the relation of man to the independent and enduring environment whose secrets he is so successfully unveiling.