Thought and Imagery
James Rowland Angell
IN his admirable treatise on Analytic Psychology, the distinguished editor of Mind has advocated a theory, touching the relation of thought and imagery, to which I have already taken exception in a previous issue of this REVIEW. I know well how easy it is to misinterpret an author concerning a matter of this kind, and I shall on that account be the more ready to adopt, so far as possible, Mr. Stout's own courteous avoidance of mere polemic, while traversing briefly the intrinsic merits of the doctrine at stake, i.e., the doctrine that we have imageless thoughts.
The following sentences contain Mr. Stout's most striking statement of his position. " . . . An imageless thought is no absurdity, however opposed such a conception may be to the hardened prejudice of those who have accustomed themselves to regard consciousness as a kind of picture gallery, or as a magic lantern in which the slides displace each other in rapid succession. There is no absurdity in supposing a mode of presentational consciousness which is not composed of visual, auditory, tactual, and other experiences derived from and in some degree resembling in quality the sensations of the special senses; and there is no absurdity in supposing such modes of consciousness to possess a representative value or significance for thought, analogous in some degree to that which attaches to images, just as revived images may have a representative value in some degree comparable to that of sense-perceptions, in spite of very great differences in respect of distinctness, vividness, and quality." These words occur in connection with a discussion of the psychological process involved in the apprehension of a word, it being asserted that such apprehension is clearly cognitive in nature and is with equal certainty unaccompanied in many instances by imagery. Obviously, therefore, the issue involved does not in any way hinge upon the question as to whether the cognitive aspect of consciousness ever becomes zero, so that affective or conative aspects alone remain in evidence. The very essence of Mr. Stout's contention rests upon the cognitive and rational characteristics of his imageless thought, and we are accordingly left with the task of weighing the evidence for a presentational form of consciousness entirely divorced from imagery.
( 647) Now, if we examine the processes characterizing the perception of a spoken or written word, we find psychologists practically agreed that the sense-impression in such a case is accompanied by " . . . an escort of revived sensations, the whole aggregate of actual and revived sensations being solidified or 'integrated' into the form of a percept. . . ." At this rate, then, any complete act of perception depends on the revivability of old imagery. I cannot convince myself just how far Mr. Stout accedes to this doctrine, but, whatever his view, it is another phase of the matter which most enlists his interest at this point. Even granting that in the sense just indicated imagery is really involved in the specific act of perception or apprehension, it still remains true that the plain man never thinks for an instant of referring his immediate perception of a word to any such revived imagery. The word seen or heard is for him a purely objective thing, and has, as immediately felt, no reference whatever to the past. Any imagery that may be present, is clearly of the implicit type. When, however, it comes to the question of an explicit and definite understanding of a word, then one finds that, whereas, if pressed to define what a word means, various images, auditory, visual, motor, and what not, may be discerned crowding in on the mind, it is none the less patent that, when the same word is heard or seen in connection with other words in a sentence, in a majority of instances no such definite imagery appears to attach to it, or, if it does, it is often totally irrelevant. Despite this asserted lack of imagery, and in the face of this irrelevant, sporadic imagery, the word is nevertheless understood, whereupon it is concluded that for certain cognitive processes, at least, imagery is a needless luxury. This conclusion appears to me to overlook certain essential facts and to misinterpret others. A brief statement of the processes which introspection reveals in my own case — and I have reason to think my experience is common to a large number of persons —will at least reveal the inadequacy and incompleteness of such an account.
When listening to a speaker, as well as when engaged in reading myself, observation discloses as the content of my consciousness a series of cognitive states closely resembling those which Professor James has so tellingly denominated 'substantive' and 'transitive.' In just the same way in which it has so clearly been shown that the perception of a single word does not necessitate the separate perception of each of the constituent letters, does the apprehension of a clause or sentence come more or less as a whole with a background
( 648) of imagery, which belongs not to the understanding of any single word in isolation from the others, but, rather, to the understanding of the sentence as a whole. The imagery involved when the word is experienced apart from other words, is often quite different from that characterizing it as a member of a sentence, if, indeed, it chances in the latter case to grave any imagery attaching predominantly to it, in distinction from the whole of which it is a part. Only logically, if at all, can the word be referred to as the same word under such diverse conditions as these. Psychologically, there are many reasons why it is by no means the same word. When the process is that of apprehending a sentence, I find in my own case the imagery involved is frequently constituted by a matrix of vague, shifting, auditory word images, in which some significant word is likely to be most prominent, and which is accompanied by a tingling sense of irradiating meaning, which, if the sentence comes to a full stop, is likely to work itself out in associated images of a fairly definite type. Into this vague matrix of auditory, or sometimes visual, imagery, an image apparently suggested by some individual word in the sentence, as distinct from the suggestion of the sentence as a whole, will occasionally inject itself, and such imagery is perhaps more often than not irrelevant, as Mr. Stout says. But it is precisely these images attaching to individual words which we ought not to look for, when seeking the imagery functionally involved in apprehending phrases or sentences. I find, accordingly, that waxing and waning imagery of the kind described, interpretative or symbolic of the sentences, constitutes the relatively substantive or static states of mind, while the transitive states appear to consist in a complex of sensations of delicately adjusted strains and tensions in the muscles of the chest, neck, and face. The dim perceptions of these peripheral conditions apparently furnish almost the whole cognitive content of consciousness during the subsidence of one group of imagery and the emergence of the next. I need hardly add that these periods of relative prominence of peripheral conditions are marked by a distinctively anticipatory attitude, which enables them to blend with the subsequent imagery without any sense of irrelevancy. Indeed, it is just this fixating of attention preliminary to the appearance of the significant imagery, which constitutes the intrinsic function of these transitive states. At this point least of all, however, could the conscious conditions be spoken of as presentational states underived from the sensations of the special senses.
In the case of apprehending language, whether through the eye or
( 649) the ear, one obvious reason that we do not detect the images readily, is because those most frequently employed in such connections are the word images themselves. If the sentence supplies one or two words capable of serving in apprehension as symbolic of the sentence, then, because the words are actually present in a perceptual act, or, at all events, were present an instant before, the difficulty of recognizing the presence of the word as a real, independent image, and not a mere percept, is immensely enhanced. As a test of this, we may notice that the instant we stop reading and begin to reflect upon what we have read, we are beset by a cloud of images of various kinds. This difficulty is greatly increased by the very function of the word, which brings it about that as an independent thing it possesses a most numerous body of associates, each of which in its tendency to redintegration is likely to inhibit others, so that a relatively static condition may be induced in which we are left for an instant with the bare word itself. In those cases where we hang upon the dying sound of the word or its fading visual characteristics, without clear-cut imagery dissevered from the perceptual process itself, there is often present, as a highly important accompaniment, a definite (quasi-affective) attitude of familiarity with the word, and a feeling of placid conviction that at any moment the explicit associates which give it meaning could, if necessary, be summoned before us. Meanwhile we have certainly not been confronted by a presentational form of consciousness underived from the senses. In my own case, at least, such conditions approximate the transitive states already described. Nor am I, in my statement of what I find introspectively to be true of myself, denying the force which belongs in some instances to Hobbes' contention (cited with disapproval by Mr. Stout) that much of our use of language is essentially automatic, and as such involves only the dimmest and most evanescent imagery. At all events, I can discover no reason why this group of activities employed in connection with language, should form an exception to the general tendency toward automatic habits. This is not to say that a rigid automatism could ever supervene, so long as the function of language remains what it is at present. Only let it not be overlooked that, just in the degree in which unconscious automatic processes interpolate themselves in the stream of our consciousness, do we lose all right to speak of thought at all in the proper sense, and a fortiori of presentational forms of thought.
Turning to the more complex stages of cognitive process, we again meet with various forms of imagery. Inasmuch as judgment and
( 650) reasoning are by common consent concerned with the manipulation and development of concepts, we may content ourselves with an examination of conception as representative of the higher cognitive activities. Here, too, Mr. Stout distrusts the indispensability of the image, for he says: "On the other hand, the conceptual synthesis itself is relatively independent of the image, which may either be extremely vague or wholly absent." Speaking once more for myself, I find my concepts, as concrete psychological events occurring at certain definite times, constituted by images clear or vague as the case may be. Occasionally they are blended with processes involving perception of peripheral conditions, as described in connection with transitive states. They often shift, and are by no means the same at different times. Indeed, there is no necessity that they should be, representing as they do simply a medium for the performance of a certain function, i.e., the function of meaning a definite thing. Thus the image which serves me in using the concept ' justice' is sometimes a visual image of a statue of justice and sometimes simply the auditory word image.
It happens that I have for some little time been gathering statistics on the imagery of concepts. Provisional as my present material is, I find an overwhelming preponderance of the persons examined able to discover images underlying their concepts. The few remaining cases I have thought fairly attributable to one of two causes, either poorly trained introspective powers, or failure to recognize the psychological content as a concept because imbedded in a quasi-perceptual state. An illustration will make this last point clearer. I know a person whose imagery is largely motor, for whom the concept 'equation,' as a concrete entity existing at a given moment of time and apart from its function as meaning a certain experience, is largely made up of sensations of muscular strains by means of which a form of stable equilibrium is temporarily established, not wholly unfit to serve as a symbol of the balance of the equation. If such an account seems fantastic and unreal, one has only to recall the complicated number-forms by means of which many persons accomplish their numerical computations, to realize how utterly disparate function and content may be. In such a case as that of our illustration, where introspection may long look in vain for pure imagery divorced from immediately present sensuous elements, there is still nothing that can by any stretch of imagination or terminology be spoken of as a presentational form of consciousness underived from the senses.
In working out a troublesome problem, it is often the periods representing the transition states with their content of sensations of adjustment that are most intense. But this simply means that the interstices of thought, as representative, are filled in by processes involving mere awareness of bodily conditions. The actual advance is always marked by a resolution of these strains and inhibitions characterizing the bodily attitude in an image or group of images temporarily satisfactory.
To bring our considerations together, it appears, then, that even in the case upon which Mr. Stout seems most willing to rest his doctrine and the one which led to Burke's famous passage upon the matter, we shall (1) find imagery, if we look in the right place. This imagery will be of an independent type, and not merely such as enters into the solidified percept, but such as must be sought in connection with the apprehension of the perceptual units, be they words, phrases, or sentences. If we look to the isolated, individual words in the sentence, we shall probably meet with irrelevant imagery or else find nothing at all. The introspective process will be baffled. (2) The fact that so much of the imagery involved in the case under consideration is itself word imagery, which is swamped in the mere perceptual process of seeing or hearing, accounts for the difficulty of detecting it introspectively. This difficulty is augmented by the tendency of the multitudinous associates of any one word to inhibit each other, and so leave only the dim image of the word itself, plus the feeling of familiarity with it. (3) Our positive ground for assuming the presence of the image is twofold : first, the very origin of cognitive consciousness through the action of the senses ; second, the fact that accurate introspection always reveals the image explicit or implicit. (4) In so far as an image is simply a symbol of certain experiences, great dimness in it is still compatible with its correct use within limits. One of the simplest of such cases is realized in mere apprehension or recognition, which, as experience develops, may readily tend under certain limiting conditions toward automatism. (5) We can attribute no intelligible origin to a form of presentation underived from the senses, nor does it appear how an individual who might by chance be fitted out with only such presentational material as is afforded by the senses, could ever get a glimmering of what the former is like. A Kantian noumenon turning up in the midst of a well-behaved lot of phenomena could not be a greater anomaly, nor more out of place.
JAMES ROWLAND ANGELL.
UNIVERSITY OF CHICAGO.