PSYCHOLOGY 63 question arises : How now are we to distinguish between imagining and remembering, and again, between imagining and expecting ? It is plainly absurd to make the difference depend on the presence of belief in memory and expecta- tion and on its absence in mere imagination ; for the belief itself depends on the difference instead of constituting it. One real and obvious distinction, however, which Hume pointed out as regards memory, is the fixed order and posi- tion of the ideas of what is remembered or expected as contrasted with " the liberty " of the imagination to trans- pose and change its ideas. This order and position in the case of memory are, of course, normally those of the original impressions, but it seems rather naive of Hume to tell us that memory " is tied down to these without any power of variation," while imagination has liberty to transpose as it pleases, as if the originals sat to memory for their portraits, while to imagination they were but studies. Such correspondence being out of the question as Hume takes care to state as soon as it suits him all we have, so far, is this fixity and definiteness as contrasted with the kaleidoscopic instability of ideation. In this respect what is remembered or expected resembles what is perceived : the grouping not only does not change caprici- ously and spontaneously, but resists any mental efforts to change it. But, provided these characteristics are there, we should be apt to believe that we are remembering, just as, mutatis mutandis, with like characteristics we might believe that we were perceiving : hallucination is possible in either case. This fixity of order and position is, however, not suffi- cient to constitute a typical remembrance where the term is exactly used. But remembering is often regarded as equivalent to knowing and recognizing, as when on revisit- ing some once familiar place one remarks, " How well I remember it ! " What is meant is that the place is re- cognized, and that its recognition awakens memories. Memory includes recognition ; recognition as such does not include memory. In human consciousness, as we directly observe it, there is, perhaps, no pure recognition : here the new presentation is not only assimilated to the old, but the former framing of circumstance is reinstated, and so perforce distinguished from the present. It may be there is no warrant for supposing that such redinte- gration of a preceding field is ever absolutely nil, still we are justified in regarding it as extremely vague and meagre, both where mental evolution is but slightly advanced and where frequent repetition in varying and irrelevant circumstances has produced a blurred and neutral zone. The last is the case with a great part of our knowledge ; the writer happens to know that bos is the Latin for "ox" and bufo the Latin for "toad," and may be said to remember both items of knowledge, if "remember" is only to be synonymous with "retain." But if he came across bos in reading he would think of an ox and nothing more ; bufo would immediately call up not only " toad " but Virgil's Georgics, the only place in which he has seen the word, and which he never read but once. In the former there is so far nothing but re- cognition (which, however, of course rests upon retentive- ness) ; in the latter there is also remembrance of the time and circumstances in which that piece of knowledge was acquired. Of course in so far as we are aware that we recognize we also think that remembrance is at any rate possible, since what we know we must previously have learned, recognition excluding novelty. But the point here urged is that there is an actual remembrance only when the recognition is accompanied by a reinstatement of portions of the memory-train continuous with the previous presentation of what is now recognized. Summarily stated, we may say that between knowing and remembering on the one hand and imagining on the other the difference primarily turns on the fixity and completeness of the grouping in the former ; in the latter there is a shifting play of images more or less "generic," reminding one of " dissolving views." Hence the first two approximate in character to perception, and are rightly called recogni- tions. Between them, again, the difference turns primarily on the presence or absence of temporal signs. In what is remembered these are still intact enough to ensure a localization in the past of what is recognized ; in what is known merely such localization is prevented, either because of the obliviscence of temporal connexions or because the reduplications of the memory-train that have consolidated the central group have entailed their suppression. There is further the difference first mentioned, which is often only a difference of degree, viz., that remembrances have more circumstantiality, so to say, than mere recognitions have : more of the collateral constituents of the original concrete field of consciousness are reinstated. But of the two characteristics of memory proper (a) concreteness or circumstantiality, and (6) localization in the past the latter is the more essential. It sometimes happens that we have the one with little or nothing of the other. For example, we may have but a faint and meagre representa- tion of a scene, yet if it falls into and retains a fixed place in the memory-train we have no doubt that seme such experience was once actually ours. On the other hand, as in certain so-called illusions of memory, we may suddenly find ourselves reminded by what is happening at the moment of a preceding experience exactly like it some even feel that they know from what is thus recalled what will happen next ; and yet, because we are wholly unable to assign such representation a place in the past, instead of a belief that it happened, there arises a most distressing sense of bewilderment, as if one were haunted and had lost one's personal bearings. 1 It has been held by some psychologists 2 that memory proper includes the representation of one's past self as agent or patient in the event or situation recalled. And this is true as regards all but the earliest human experience, at any rate ; still, whereas it is easy to see that memory is essential to any development of self-consciousness, the converse is not at all clear, and would involve us in a needless circle. Intimately connected with memory is expectation. We Expecta- may as the result of reasoning conclude that a certain tion. event will happen ; we may also, in like manner, conclude that a certain other event has happened. But as we should not call the latter memory, so it is desirable to distinguish such indirect anticipation as the former from that expectation which is directly due to the interaction of ideas. Any man knows that he will die, and may make a variety of arrangements in anticipation of death, but he cannot with propriety be said to be expecting it unless he has actually present to his mind a series of ideas ending in that of death, such series being due to previous associations, and unless, further, this series owes its re- presentation at this moment to the actual recurrence of some experience to which that series succeeded before. And as familiarity with an object or event in very various settings may be a bar to recollection, so it may be to expectation : the average Englishman, e.g., is continually surprised without his umbrella, though only too familiar with rain, since in his climate one not specially attentive to the weather obtains no clear representation of its successive phases. But after a series of events A B C D E . . . has been once experienced we instinctively expect the recurrence 1 Any full discussion of these very interesting states of mind belongs to mental pathology. 2 As, e.g. , James Mill (Analysis of the Human Mind, ch. x.), who treats this difficult subject with great acuteuess and thoroughness.
