Thought Before Art

The art of thought, like the art of running, or the actor’s art of significant gesture, is an attempt to improve by conscious effort an already existing form of human behaviour. Men ran for countless generations before they invented or handed down the few expedients which constitute the art of running as taught by professional athletic trainers; they revealed their feelings by gestures long before there were any schools of dramatic art; and they thought for thousands of years before they had a name for thinking. In all these cases, therefore, the rules of art must be based on the most exact knowledge which we can obtain of the behaviour which the art is to modify. Sometimes that behaviour is completely “natural”; the teacher, for instance, of running, or of breathing-exercises, starts from behaviour which is mainly directed by the sympathetic nervous system, and which contains hardly any “acquired” elements. But when the co-operation of the higher nervous system is involved, it is, under the conditions of modern civilization, almost impossible to observe any instance of human behaviour which is entirely free, and extremely difficult to observe any instance which is even approximately free from acquired elements. Many, indeed, of the innate tendencies of the higher nervous system, such as the tendency to speak, represent rather a power and an inclination to learn to behave in a certain way, than a direct instinct to behave in that way. Such learning may proceed rather by half-conscious imitation than by conscious effort; and the result even of repeated conscious effort may be a habit which it is not easy for an observer either of his own or of other people’s behaviour to distinguish from a natural tendency.

An actor, for instance, can only with the greatest difficulty form any estimate as to how far his movements while he acts are “natural,” and how far they are due to acquired modifications of nature. Sometimes he can be helped by observing the gestures of less sophisticated persons than himself; he can watch the behaviour of children; and he could, before the cinema had soaked whole populations in third-rate theatrical conventions, go down to the East End of London on a Saturday night, and watch the comparatively natural behaviour of uneducated people who were under the influence of rage or jealousy, and some of whose acquired social habits had been temporarily weakened by alcohol. Or he may try to recall his own behaviour on some occasion when he was “off his guard”; or, if he has unusual powers of imagination, introspection and inhibition, he may stand before a looking-glass trying to believe that he is Othello or Lear, and to inhibit all acquired elements in the gestures which follow from that belief.

The thinker, when he is trying to observe thought in its most natural form, is faced with even greater difficulties than the actor who is trying to observe natural gesture. Some of the most important steps in the process of thought are normally unconscious or half-conscious; and our unconscious or half-conscious thought, even if we succeed in observing it, is not necessarily “natural.” The subject-matter, again, even of our least conscious thought is mainly derived from past experience, and is deeply influenced by intellectual and emotional habits; and thought at all grades of consciousness makes large use of language with its innumerable acquired associations. The student, therefore, of the art of thought has to choose a more or less arbitrary point from which he shall assume the conscious effort of the art to begin. I myself shall, in this and the following six chapters of my book, assume that I am addressing young adults who have already learnt, at home or at school, to speak, read, attend, and memorize, but who, though they do in fact constantly reach new ideas by using their brains, have never yet attempted to acquire or to apply a conscious art of thought. I shall postpone till the last three chapters of the book the problem of that preliminary training in the art of thought which may be given by teachers and others to children and adolescents.

The young adults whom I imagine myself to be addressing will, in spite of differences in their acquired experience, be alike in that the essential element in their inventive thought is the process by which, as I have already stated, one psychological event calls up another in the “telephone exchange” of the upper brain. It is this process of “association” which their art will attempt to improve, and which they must first try to observe and understand. The process of association has been observed introspectively by two methods—the observer has either remembered a train of association after it has occurred, or he has watched it while it is occurring. The first method is by far the easier, and up till our own time has been almost exclusively used. Aristotle, indeed, in the earliest recorded discussion of the association-process, treats association as a section of the problem of memory. He asks himself why the memory of one experience calls up the memory of another. “For,” he says, “experiences habitually follow one another, this succeeding that, and so, when a person wishes to recollect, he will endeavour to find some initial experience to which the one in question succeeded.” He concludes that experiences call each other up, sometimes because they succeeded each other in time, sometimes because the experiences were similar, or contiguous in place, or were connected logically as are the steps in a mathematical proof.

The best-known description of a train of association as seen in memory is that given by Hobbes in a classical passage of his Leviathan (Chapter III, written about 1650). The passage forms part of a discussion of the type of thinking which Hobbes calls a “train of thoughts, or mental discourse . . . ‘unguided,’ ‘without design’ and inconstant.” “And yet,” he says, “in this wild ranging of the mind, a man may ofttimes perceive the way of it, and the dependence of one thought upon another. For in a discourse of our present civil war, what could seem more impertinent, than to ask, as one did, what was the value of a Roman penny? Yet the coherence to me was manifest enough. For the thought of the war introduced the thought of delivering up the king to his enemies; the thought of that brought in the thought of the delivering up of Christ; and that again the thought of the thirty pence, which was the price of that treason; and thence easily followed that malicious question; and all this in a moment of time; for thought is quick.” In his explanation of the connection between link and link in such a train Hobbes is here less full than Aristotle, and confines himself to succession in time—“in the imagining of anything,” he says, “there is no certainty what we shall imagine next; only this is certain, it shall be something that succeeded the same before, at one time or another.”

In these passages, neither Aristotle nor Hobbes distinguishes between the relation to each other of psychological events, and the relation to each other of the external facts which give rise to the psychological events. The likeness between the treachery of Judas and the treachery of the Scottish leaders seemed a sufficient explanation of the calling up of one by the other, without asking why the mind of the speaker was interested, even during “unguided” thought, in that kind of likeness. This over-simplification of the problem was made easier by the fact that neither Aristotle nor Hobbes recognized the existence of psychological causes which were not conscious. And the over-simplification was increased when, after the publication of Locke’s Essay Concerning Human Understanding in 1690, psychologists came to use the term “The Association of Ideas” for the whole association process, and to define “ideas” as copies, in conscious memory, of events.

Hobbes himself, however, realized that the path of association might in some cases be directed, not merely by the external connection between remembered facts, but also by the drive of passion in the thinker himself. These cases he classed as “regulated” thought. ‘For the impression,” he says in the same chapter, “made by such things as we desire, or fear, is strong, and permanent, or, if it cease for a time, of quick return; so strong it is sometimes, as to hinder and break our sleep. From desire, ariseth the thought of some means we have seen produce the like of that which we aim at; and from the thought of that, the thought of means to that mean; and so continually, till we come to some beginning within our own power. And because the end, by the greatness of the impression, comes often to mind, in case our thoughts begin to wander, they are quickly again reduced into the way.” And, as I pointed out in Chapter I, many modern psychologists have tended to simplify the problem in another way, by treating Hobbes’s special class as universal, and by declaring that the mechanical drive of some one of a list of instincts is invariably requisite before connection can be made between one link of association and another.

The second method of observation, in which the observer watches the association-process while it is going on, instead of remembering and explaining it when it is past, is much the more difficult; but it is much less likely (if the observer can prevent himself from distorting his observations by theories as to their causes) to lead to an over-simplified conception of the association-process itself. Fifty years hence, students will have an ample supply of this kind of observation before them; but for the moment the supply which I have been able to discover is curiously small. The experimental association-trains which are deliberately started and observed in psychological laboratories are limited in range, and are often distorted by the “unnatural” conditions of their formation; and the clinical observations recorded by the professional psycho-analyst seem to me to lose most of their evidential value owing to the influence of the suggestion of the psycho-analyst upon his patients, and to his own conscious or sub-conscious determination to defend, against outside critics, the dogmas of his profession. The most useful (from the point of view of the would-be thinker) of modern introspective evidence on the “natural” association-process, which I have met with, is that contained in Dr. J. Varendonck’s Psychology of Day Dreams, written by himself in English, and published in 1921. Dr. Varendonck, who was attached as an interpreter to the British army in Belgium, was, before the war, a lecturer on pedagogic psychology. During the war he trained himself to observe, night after night, the “foreconscious” events in his mind which immediately preceded sleep, and which were not initiated or controlled by any conscious effort of will. He was able both to watch these trains of thought, which he calls day-dreams, without allowing them to be influenced by the fact that they were being watched, and also at the right moment to wake himself, by a strong effort, into complete consciousness, and record his observations. His day-dreams deal mainly with the hopes and fears and annoyances of camp life; and they are set down with a courage and candour which compel the admiration of any one who has tried, as I have, to do the same thing, and who, partly from want of equal courage, has failed. Varendonck’s first observation was that “there occur in most of our day-dreams risings and fallings” (p. 176), or “successive risings to the surface and sinkings into the unconscious” (p. 155). A day-dream, he says, may last for a considerable time, and during that time several such “risings and fallings” of consciousness may take place, before the process is interrupted either by sleep or by a return to complete consciousness. He gives (pp. 170-2) a description, written down immediately after its conclusion, of a day-dream which lasted fifty-five minutes, and in which “on six different occasions the association had risen close to consciousness,”

Varendonck carefully analyses a few of his longer day-dreams, or, as he sometimes calls them, in Freudian language, “phantasies.” The most interesting for our purpose belonged to the type which is often called “mental trial and error”; that is to say they were part of an automatic mental attempt to solve a difficulty by imagining successive solutions. And just as the muscular “trial and error’ process comes to its conclusion when some one among a series of movements is successful, so the “mental trial and error’ process found, in Varendonck’s case, its normal conclusion in the mental recognition that the solution thought of would be successful if tried. In thinking of this recognition I find that I tend to use the word “click,” in the slang sense common among English school-boys and soldiers. ‘This term spread during the war, when every young soldier was being trained to use, by a process containing a large element of muscular “trial and error,’ a number of machines from rifle-locks upwards. The “click” was the sound made by the machine when the successful movement was made, and the verb “to click” meant, for instance, to succeed in such matters as obtaining by verbal ingenuity an irregular week’s leave. In its origin, the click feeling must have been due to the fact that a foreconscious or unconscious train of association had led to a point which revealed the need of action, and therefore the need of full consciousness. It is the same thing as the extremely painful shock which occurs when a casual train of association suddenly reveals to us that we are on the point of missing an important engagement.[1] The feeling itself varies from joy or horror to a mild recognition that the search is over.

But the process of association may lead not only to a recognition that an imagined solution of the situation will result in successful action, but also to a preceding series of recognitions that other imagined solutions will not succeed. Varendonck, therefore, speaks of the “dialogue form” of sections of his day-dreams, in which successive proposed solutions presented themselves, and were met by successive objections, until some solution appeared against which no valid objection suggested itself.[2] ‘A foreconscious chain of thought,” he says (his argument shows that he means a foreconscious chain of thought comparatively near full consciousness), “is a succession of hypotheses and rejoinders, of questions and answers, occasionally interrupted by memory hallucinations” (p. 179).

He gives a long and amusing analysis of a day-dream in which his foreconscious mind attempted to deal with the situation created by the fact that he desired to get an impertinent orderly punished. The orderly was attached to a Belgian Field Hospital, of which a certain formidable Lady V. was matron, and where “the chief medical officer was practically at her mercy” (pp. 64-76). Varendonck had already reported the orderly to his own superior officer, Major H. But the orderly had threatened him with Lady V.’s vengeance. Varendonck imagines in succession such expedients as writing to Lady V. before she can hear the orderly’s story, lending weight to his letter (which he begins to compose) by accompanying it with his visiting card (with his civilian professional status indicated on it), or getting a friendly Belgian captain to send the letter to Lady V. by his corporal. Some of these expedients were mentally accepted, and some rejected; until Varendonck remembered, in consequence of a detailed visual picture of the comfortable room where Lady V. used to sit, that she had a telephone, and finally decided to call upon Major H. himself, before Lady V. telephoned her version of the story to him.

Throughout his book, Varendonck indicates certain correlations between the rising and falling of consciousness and the rising and falling of other elements in the association-process. One such correlation will remind English readers of the description of falling from day-dreaming into sleep-dreaming at the beginning of Lewis Carroll’s Alice in Wonderland. As the level of consciousness sank, he found that the critical power which the civilized human being acquires from education and experience sank with it; the steps from one link to another in association often became such as in his fully waking state he would have at once recognized as absurd, and his objections to them might be equally absurd. He describes, for instance, a “bombardment” phantasy, in which his mind assumed that after losing both legs he would be compelled to continue his military service (p. 114), and a “flea phantasy,” in which he hit upon the expedient of using a garden roller for killing a flea in the cracks of his bedroom floor. Freud, also, has shown, though with a serious amount of exaggeration, that as consciousness sinks toward sleep the links in the train of association may become more instinctive and animal; and the vague and generalized tendencies which the writers of his school call “sex” or “libido” may appear in symbolic forms. A similar correlation, therefore, takes place, according to Varendonck, between increasing consciousness and increasing rationality: “The upward movements have for consequence the introduction into the concatenations of elements proper to conscious thought, namely elements of critical thought-activity” (p. 176).

Varendonck is, I think, less successful in indicating some further correlations between sinking and rising consciousness and the other elements of the association process. He argues, for instance, that decreasing and increasing consciousness is accompanied by a decrease or increase in the use of words, and by a corresponding increase or decrease in the use of visual images. “At one end of the series my foreconsciousness thinks in words with a few [visual] illustrations distributed at random; at the other end this ideation seems to proceed by means of pictorial images with occasional verbal expressions” (p. 61). In a passage in which he speaks of his mind in its less-conscious state as his “second self,” he says that this “second self” “operates distinctly by means of optical images, and I have reason to think that most persons share this peculiarity with me” (p. 57). The facts, however, as to the interrelation of verbal and visual imagery with rising and falling consciousness seem to me, even on Varendonck’s own evidence, much more complicated. In actual sleep-dreams, and in the deeper forms of foreconsciousness, we often use words which, if we happen afterwards to remember them, are found to belong to the type which psychologists call “glossolaly,” and which may be less rational than the most absurd dream-images.[1] At a stage of consciousness well below rationality, a coherent but almost meaningless jingle of words may form itself in our minds, and appear to be perfectly satisfactory. Mr. Robert Graves, for instance, describes his delight with a dream-poem consisting of the words:

“It’s Henry VIII, it’s Henry VIII,

He is leading his armies over to France.” t And a relative of mine woke one morning with the conviction that she had achieved immortality by the lines:

‘Leave there thy steed,

And let it feed

On more than meets the eye.”

And, though it may be true that the use of visual and other “images” plays on the average a larger part (when compared with the use of words) in less conscious than in more conscious thought, some of the most completely conscious and most rational thought may be carried on entirely by the use of wordless images. A very able and rapid financial thinker, who was trained as a mathematician, told me that even when his thought is most conscious and rational he thinks, like a chess-player, in terms of seen or felt wordless “situations.” The wordless images of such a “situation” may be purely “kinetic,” with little or no visual element. The chess correspondent of the Observer (Feb. 8, 1925) writes that the great chess-player Alekhin said that “he does not see the pieces in his mind, as pictures, but as force-symbols; that is as near as one can put it in words.”

Again, when reading Varendonck, one has always to remember that almost the whole of his first-hand evidence is derived from introspection during the process of sinking towards, or rising away from sleep, and that that fact limits the validity of his conclusions about thought carried on under other conditions. He says, for instance, that “the chains of thought which occupy our minds during our distractions in waking life are wholly similar to the phantasies that arise in the somnolent state” (p. 34). He illustrates this by a description of the process in which thoughts and counterthoughts arise in our mind and are accepted and rejected when we are in a state of “full awareness.” If, for instance, one has to compose and send a painful letter: “One thinks about the letter; and in one’s mind it has already been composed over and over again before one writes it down; every argument that one can think of has been put forward and criticized, dropped or retained, until in the end the letter is present in the mind before it is confided to paper” (p. 139). In a later book, Varendonck points out that “the orator prepares his speech in the same way, while his mind is absent during a purely physical occupation; the business-man unintentionally ponders over his affairs in the train, or as he walks to the office; the journalist has his article in his mind before reaching his office.” I myself, however, believe that though less-conscious thought during our hours of full wakefulness has many analogies with the less-conscious thought which occurs while the main nervous system is sinking into natural sleep or the hypnotic trance, there is a difference between the two processes, which is of great importance in the higher and more difficult forms of intellectual creation (see below, chapter ix). And the process of “distraction” in waking life, when a fully-conscious train of association is broken in upon by a call for our attention to another subject, is different from the process in which Alice in Wonderland ceased to think about cats and bats, and sank into a region where it seemed quite natural that a white rabbit should carry a watch in his waistcoat pocket.

Finally, the student who desires to use Varendonck’s evidence as a working description of “natural” thought, must remember that Varendonck was originally attracted to the whole question by reading Freud’s Interpretation of Dreams; and that although he is obviously a man of high intellectual integrity, and much less liable to the involuntary distortion of his introspective observations by loyalty to his master than are most of the followers of Freud, yet even he seems to feel bound to ascribe every train of association to the driving force of some “wish” or “instinct” or “affect.” He uses, indeed, the term “affective thinking” as synonymous with “foreconscious thinking” (e.g., p. 19). He can do so with less violence to the facts, because the “day-dreams” which he describes are almost all the result of intense anxiety either as to his position in the army, or his professional future, or his intended re-marriage. But nevertheless in conscientiously analysing his thought-trains he has to use the word “affect” in many different senses, and sometimes, apparently, in hardly any sense at all. He says, indeed, frankly, that “I am quite aware that this same word affect has been used in my various arguments to denote very different notions, such as wishes, emotions, etc.” (p. 245).?

One has to be similarly on one’s guard in using the description of his thought-processes given by H. Poincaré in the celebrated chapter on “Mathematical Invention” in his Science and Method (translated 1914). Poincaré is also dealing with the complicated and still insufficiently analysed problem of “emotion” as a frequent directing force in the association process, and as a still more frequent accompaniment of that process. And he, too, under the influence of the general tendency in psychological theory which I have called the “mechanist” view, simplifies that relation by ascribing the direction of all association trains to the drive of some instinctive emotion. He asks what is the selective force, the “sieve,” which chooses the apparently right solution of a mathematical problem and brings it into full consciousness, while rejecting the apparently wrong solution. He answers that the cause is “sensibilité”— an extremely ambiguous French term which may either be translated “feeling” or merely anglicized as “sensibility.” ‘More commonly,” he says, ‘the privileged unconscious phenomena, those that are capable of becoming conscious, are those which, directly or indirectly, most deeply affect our sensibility” (p. 58). He adds that, in the case of his own mathematical discoveries, the sensibility concerned is that which arises from the esthetic instinct. ‘It may appear surprising that sensibility should be introduced in connection with mathematical demonstrations, which, it would seem, can only interest the intellect. But not if we bear in mind the feeling of mathematical beauty, of the harmony of numbers and forms, and of geometric elegance. It is a real esthetic feeling, that all true mathematicians recognize. . . . The useful combinations are precisely the most beautiful” (p. 58).

Poincaré’s authority is sufficient to assure us that in his case the instinctive appreciation of elegance did play a real part in stimulating and guiding many of his sub-conscious trains of thought, and in deciding which of many subconsciously imagined solutions should produce the “click” of conscious success. He may even be right in saying that without a rather high degree of this esthetic instinct no man will ever be a great mathematical discoverer (p. 60). But it is extremely unlikely that the esthetic instinct alone was the “power” driving the “machine” of his thought. He must have possessed some of that “public spirit” which is an almost indispensable condition of a lifetime spent in intellectual toil. He must have had many ambitions and loyalties and habits of thinking and feeling. Above all he had a brain which without scope for its self-activity would have been as restless as a wild hawk in a cage. And each of these factors must have played its part in the thought-processes that went on in the varying levels of his consciousness. One almost fears that if Poincaré had been a friend of Freud, instead of being a friend of Boutroux who was a friend of William James, he might have become certain that libido was the sole and sufficient “sieve” of his thoughts.

But, while the introspective evidence both of Varendonck and of Poincaré is, I believe, presented in a setting of over-simplified theory, it is nevertheless possible for the student after reading their books to form, with the help of his own introspection, a fair working conception of those “natural” thought-processes, which, however much influenced by experience and habit, are not, at the time of thinking, voluntarily controlled by any rules of the thinking art. He will observe in his own mind automatic trains of associated ideas, some ending with a remembered positive or negative decision, some broken and at once forgotten. Some of these trains may belong to that primitive type which has given rise to the “mechanist” conception of the relation between “instinct” and “reason.” That is to say, behind the train may be the urge of a strong and simple instinct, love, or hatred, or fear, driving the train onward, judging its results, and bringing it back again and again to the same starting-point. Or the connecting cause may be some habit of thought; or, again, the upper brain may be acting on its own initiative, or in obedience to a “curiosity” which may only be another name for more or less independent brain-activity. And, at any moment, a passionless association may lead him to a conclusion that wakens a vehement passion, or a train of thought driven by passion may fade into a passionless reverie. Flickering over all these processes, as the searchlights flicker along a line of cliffs, there may be a hundred different conscious or less-conscious “feelings,” alternating with, and fading into each other. He may be dimly aware of some brooding “sentiment.” The hormones discharged by the obscure processes of the endocrine glands may pervade him with vague “euphoria” or “dysphoria”—elation or discomfort, energy or inertia. Actual sensations, and the more or less vivid memories of sensations, may play their part. Or, he may be able afterwards to remember sudden flashes of emotional experience almost too momentary for description, feelings of queerness, or surprise, or recognition, or amusement, or the craftsman’s delight in his own skill and success.

Notes

1 Aristotle, De Memoria, II, 12. This difficult passage has been admirably translated and explained by Prof. Howard C. Warren in his History of the Association Psychology (1921), pp. 25 and 26.

? Professor H. C. Warren points out (History of the Association Psy- chology, p. 5) that: “When Locke speaks of the association of ideas he has reference to possible connections between all sorts of mental content; whereas from the time of David Hume onward the phrase refers to connections between representative data only. . . . This permanent fixing of the expression association of ideas with an altered meaning given to the term idea has exerted some influence on the development of the doctrine itself.”

1It would be interesting if some student of comparative psychology could discover whether the process of cerebral association is ever sufficiently advanced in an intelligent dog, to produce any indication of the “click” phenomenon. Does the dog’s whole organism ever recognize that his mind has discovered the need of immediate action? Do, for instance, the dog’s endocrine glands ever discharge their hormones into the blood-streams as the result of a train of association, starting, as human “day-dreams” may start, in the mind, or only when the train of association is started by a sensation—the sight of a rat, or the smell or step of his master—as Lord Shaftesbury’s train of association was started by the sight of a book? (above, p. 42). ? At the threshold between dreaming and foreconscious thinking the critical faculty may have a strong negative power over the train of association with no positive power of directing the train. I, as a child, used often to continue my dreams into a foreconscious state which preceded full awakening. I used then to notice that if I was vaguely aware that a lion was about to appear round the corner of the street in which I was walking, I could prevent the appearance of the lion, but could not cause anything else chosen by me to appear.

1 See Varendonck (Day Dreams), p. 331.

* On English Poetry (1922), p. 16.

1 Varendonck, Evolution of the Conscious Faculties (1923), p. 108.

1 Varendonck’s later book, The Evolution of the Conscious Faculties (1923), is much less valuable than his Day Dreams (1921), because it contains fewer of those introspective records in which he excels, and more of the psychological generalizations in which he seems to me to be weak,