The Act of Creation

by Arthur Koestler

  In this example visual exploration and mental exploration are actually indistinguishable; the observational data derived from looking at the rock face, and the lessons derived from previous experience combine into one. In other situations, the exploratory process may be confined to the inner landscape, to the exclusion of all stimuli from the world outside. The poet's or the mathematician's trance-like condition while he concentrates on a problem, the vivid fantasies of the day-dreamer, the delusions of the insane, the dreams of the sleeper, are products of widely different games of the mind; but they all have this in common, that the beam of focal awareness is exploring the inner environment, and ignoring the input from the senses. The features on which the beam alights are images of a pictorial or verbal nature, memories in abstracted, conceptualized, or distorted shape; in a word, they are past experiences internalized. The inner landscape may be regarded as a kind of private, miniature model -- or caricature -- of the world in the subject's brain-mind (see Book Two).

  Thus the objects of the scanning process are ultimately the individual's past experiences (including his prenatal past) incorporated in one form or another into his mental landscape. And the rules which control the scanning process (the pattern of 'mental eye motions', as it were) are also derived from past experiences by abstraction and generalization; they are the results of learning compressed into the operational codes of thinking skills.

  As an example, take the parlour game 'Towns with M' (see page 38). The moment I start playing it, a fixed code takes control of my mental processes, and their freedom is whittled down to strategic choices. These may be based on exploring an imagined geographical map, or on the 'tuning-fork' method. The mental map is a blurred, hazy, and distorted rephca of what I learned in school and on travels; but as I proceed to scan it, from west to east with the mind's eye, name after name emerges from the misty twilight: Manchester, Munich, Moscow, Murmansk, Michigan, etc. If, on the other hand, I apply the tuning-fork method, Manchester will call out Mannheim, Madrid, Madras, and so forth. All of these names were learned in the past; all of them were members of the 'M' matrix (otherwise they could not have been summoned on the 'wavelength' of that particular code); all of them were unconscious or pre-conscious the moment before the beam of focal awareness alighted on them. The beam was guided firstly by the rule of the game ('find towns with M, not rivers with S'), and secondly by strategy ('move from west to east'). The rule was fixed, the strategy variable. A further point to note is (though it does not concern us yet) that strategy operates by a kind of feed-back from the lie of the land: I was searching for towns with 'M' between Munich and Moscow, but found none: so I moved on. Other factors enter: I might have remembered Mannheim, but did not because of an unpleasant experience there: emotional disturbances interfering with 'mechanical logic'. Incidentally, the forming of a sentence in ordinary conversation follows a similar pattern. Instead of scanning a map for towns with 'M', you must scan your vocabulary for words which will fit a given meaning.

  Take an even simpler practical example. I live in London and have to spend a day in Paris some time next week to see my French publisher. If this were a pleasure-trip the fringes of my consciousness would at once be crowded with half-remembered, floating images of bistros, streets, galleries, métro stations; but, as it is a business trip, a different code enters into action and the matrix is cluttered with timetables, appointment books, galley proofs, and dustcovers, which strategic planning must co-ordinate into the proper sequence.

  Purposive thinking, even of this ordinary, humdrum kind, proceeds in several steps. First, the code of rules appropriate to the task is 'tuned in' -- by dint of analogy with similar tasks encountered in the past. As a result, a matrix will emerge, a kind of patterned mental grid or chessboard, which provides a preliminary selection of permissible moves, a first guidance for the exploratory process. Next comes strategy, dependent upon the particulars of the situation.

  Each step involves processes more or less removed from focal awareness. The code which guides the focal beam of consciousness functions more or less unconsciously. (It could not be otherwise, for if the beam were guided by the beam, we would be landed in the paradox of a little man inside the brain with a little man inside his brain, and so forth.) The codes of grammar and syntax function unconsciously; the meaning you wish to express provides the strategy for selecting the proper word. The words -- just like the towns with 'M' -- were lying in darkness before the beam searched them out and lit them up for a fleeting moment; then they sink back into darkness again.

  Thus all reasoning, even of a trivial order, is steeped in unconscious processes. But when the task is of a more complex order, thinking may run into difficulty at each of the steps which I have outlined. A situation may share certain features with other situations encountered in the past, yet the code of rules which enabled us to cope with them proves mysteriously inadequate in the new situation. Bleeding and purging the patient proved beneficial in a number of cases, so it came to be regarded as an all-cure; why did it not work? We can bisect an angle with compass and ruler, so it was assumed that we can trisect angles by the same method; but it did not work. Sound waves are propagated in thin air, so it was assumed that light waves are propagated in a thin ether; but the analogy provided the wrong matrix. Circles turning upon circles yielded an adequate description of heavenly motions, until Tycho perfected the methods of observation; the new data disrupted the pattern, and the matrix was blocked.

  When a situation is blocked, straight thinking must be superseded by thinking aside -- the search for a new, auxiliary matrix which will unblock it, without having ever before been called to perform such a task. The essence of discovery is to hit upon such a matrix -- as Gutenberg hit on the wine-press and Kepler on the sun-force.

  In the trivial routines of thinking, we are exploring the shallows on the twilight periphery of awareness, guided by a more or less automatized scanning procedure. In creative thinking we are exploring the deeps, without any obvious guidance. Yet some guidance there must be -- unless all novelty is due to random hits produced by the patient monkey on the typewriter.

  The 'Hooked Atoms of Thought'

  Let me return once more to Henri Poincaré, who proposed a theory concerning the nature of this unconscious guidance. We have heard him describe how, on three different occasions, the solution of a problem popped up spontaneously and ready-made, as it were, from the depth of the unconscious. Further on in that famous lecture from which I have quoted (pp. 114-16) he tried to give an explanation of this phenomenon. His starting point was that mathematical discovery consists in a 'combination of ideas'; and his description of this process stresses the characteristic features of what I have called the bisociative act:

  Among chosen combinations the most fertile will often be those formed of elements drawn from domains which are far apart. . . . Most combinations so formed would be entirely sterile; but certain among them, very rare, are the most fruitful of all.

  Now these combinations are engineered by the unconscious or, as he calls it, the 'subliminal self'; but how? There are, he says, two possibilities. The first is that the unconscious 'is capable of discernment; it has tact, delicacy; it knows how to choose, to divine. What do I say? It knows better how to divine than the conscious self since it succeeds where that has failed. In a word, is not the subliminal self superior to the conscious self? I confess that, for my part, I should hate to accept this. . . . ' So he rejects this first hypothesis in favour of the second: the unconscious is an automaton which mechanically runs through all possible combinations:

  Figure the future elements of our combinations as something like the hooked atoms of Epicurus. During the complete repose of the mind, these atoms are motionless, they are, so to speak, hooked to the wall. During a period of apparent rest and unconscious work, certain of them are detached from the wall and put in motion. They flash in every direction through the space . . . as would, for example, a swarm of gnats, or if you prefer a more learned
comparison, like the molecules of gas in the kinematic theory of gases. Then their mutual impacts may produce new combinations. But two objections come to his mind. Firstly, is not the number of possible combinations infinite, and the chance of hitting on a favourable one infinitesimal? No, he answers, because during the conscious preparatory work which preceded the period of unconscious incubation, a first selection was already made of those atoms which are to be unhooked from the wall; and although no satisfactory combination of them was found, 'after this shaking up imposed upon them by our will, these atoms do not return to their previous rest. They freely continue to dance' -- until the one favourable collision in a million occurs. (This is rather like saying that the chances of the monkey on the typewriter hitting on a Shakespeare sonnet would be considerably improved by building a typewriter which uses whole words as keys instead of letters.)

  The second objection which occurred to Poincaré is as follows: although countless combinations are formed 'in consequence of the automatism of the subliminal self, only the interesting ones . . . break into the domain of consciousness'. But, if so, what is the nature of the mysterious sieve which rejects the useless combinations and allows only the lucky hits to pass into consciousness? Poincaré's answer is that the selection is done by 'the aesthetic sensibility of the real creator. The useful combinations are precisely the most beautiful, I mean those best able to charm this special sensibility'.

  This is certainly a more attractive answer than Taine's, who summons ideas from the ante-chamber 'in a mechanically logical way'; yet Poincaré himself felt its unsatisfactoriness. For it combines a mechanistic theory about the random collision of atomic ideas in the unconscious, with an aesthetic sensibility which resides in the conscious, and plays the part of a deus ex machina. We do not doubt that this kind of sensibility is present in the creative mind, and to inquire into its nature is precisely what we are after; but Poincaré lets the matter rest just where the problem starts.

  Particularly fascinating in this lecture, delivered in 1908, is the fact that Poincaré, after acknowledging his debt to the 'subliminal self' and singing its praises, confesses that he would 'hate to accept' that it might in some respects be superior to the conscious self, and relegates it to the role of an automatic mixing machine in the basement. He worked by intuition, but for all his modesty and open-mindedness he was unable to shake off the rationalist hubris of the nineteenth century.*

  Exploring the Deeps

  All we have gleaned from these excursions into the history of our subject, from Plotinus to Poincar´, is firstly, a negative insight into the narrow limitations of conscious thinking; and on the positive side, affirmations of the superiority of unconscious mentation at certain stages of creative work. But regarding the reasons for this superiority, and the process by which it manifests itself, we got merely a few vague intimations, or else unsatisfactory mechanistic hypotheses such as Galton's and Poincaré's. Nor, I may add here, had Freud or Jung much to say about the specific problem how unconscious processes lead to new discoveries.

  Let us at this stage follow the advice we have so often heard repeated, and 'think aside' -- by turning, for a moment, from scientists to poets. If we were to apply Poincaré's hypothesis we would come to the conclusion that the poet has a conscious mind endowed with aesthetic sensibility, and an unconscious mind equipped with an automatic rhyme-computer (built on the principle of rhyming lexicons), and also with an image computer (a kind of magic lantern with an automatic slide-changer). Out of the hundreds of rhymes and similes produced per minute the vast majority would, of course, be valueless, and the aesthetic censor in the conscious mind would have a full-time job rejecting them -- until he went out of his mind.

  It seems neither an economical nor an inspired procedure. Now let us listen to Coleridge's celebrated description of the genesis of Kubla Khan. He is speaking of himself in the third person singular:

  In the summer of the year 1797, the Author, then in ill health, had retired to a lonely farm-house between Porlock and Linton. . . . In consequence of a slight indisposition, an anodyne had been prescribed, from the effects of which he fell asleep in his chair at the moment that he was reading the following sentence, or words of the same substance, in Purchas's Pilgrimage: Here the Khan Kubla commanded a palace to be built, and a stately garden thereunto. And thus ten miles of fertile ground were enclosed with a wall. The Author continued for about three hours in a profound sleep, at least of the external senses [sic] during which time he has the most vivid confidence, that he could not have composed less than from two to three hundred lines; if that indeed can be called composition in which all the images rose up before him as things with a parallel production of the correspondent expressions, without any sensation or consciousness of effort. On awakening he appeared to himself to have a distinct recollection of the whole, and taking his pen, ink, and paper, instantly and eagerly wrote down the lines that are here preserved.

  This, of course, is an extreme case of unconscious production -- even if, in all likelihood, it did not originate in a dream, but in an intense daydream or hypnogogic state. (In another, and probably earlier, statement Coleridge gives a different version: 'This fragment with a good deal more, not recoverable, composed in a sort of Reverie brought on by two grains of Opium, taken to check a dysentery.' The 'reverie' version is strengthened by the words 'in a profound sleep, at least of the external senses' -- which point towards some intermediary kind of 'waking dream'.)

  But whether he was asleep or half asleep is unimportant; the point to note is the emphasis he puts on visual images 'which rose up as things'. Unfortunately, no sooner had he started on the actual writing down of the poem than he was interrupted 'by a person on business from Porlock, and detained by him above an hour, and on his return to his room, found, to his no small surprise and mortification . . . that with the exception of some eight or ten scattered lines and images, all the rest had passed away like the images on the surface of a stream into which a stone has been cast'. This incidental metaphor suddenly sets off in its author another chain of visual imagery which illustrates how the dream version of 'Kubla Khan' was lost, thanks to the gentleman from Porlock, but reconstructed later on out of the remaining fragments. After the 'stone had been cast':

  . . . all the charm Is broken -- all that phantom-world so fair Vanishes, and a thousand circlets spread, And each misshape the other. Stay awhile, Poor youth! . . . The stream will soon renew its smoothness, soon The visions will return! And lo, he stays, And soon the fragments dim of lovely forms Come trembling back, unite, and now once more The pool becomes a mirror.

  The whole poem, with its rather striking allegory, grew out of a hackneyed metaphor, which was meant to serve only as a visual illustration to a verbal narrative. But all at once the servant becomes master, the illustration takes over from the text; visual association, the logic of the eye are in command, and the words must follow their lead. . . .

  We further note that the whole sequence of 'not less than from two to three hundred lines' of the Kubla Klan dream itself was triggered off by a passage read in Purchas's Pilgrimage, as indifferent as the simile of the stone cast into the stream: 'Here the Khan Kubla commanded a palace to be built', etc. But at that point his imagination caught on, the opium took effect, visual thinking took over, and images 'rose up as things'.

  Thinking in pictures dominates the manifestations of the unconscious -- the dream, the hypnogogic half-dream, the psychotic's hallucinations, the artist's 'vision'. (The 'visionary' prophet seems to have been a visualizer, and not a verbalizer; the highest compliment we pay to those who trade in verbal currency is to call them 'visionary thinkers'.)

  But, on the other hand, pictorial thinking is a more primitive form of mentation than conceptual thinking, which it precedes in the mental evolution of the individual and of the species. The language of the primitive (and of the child) is, to borrow Kretschmer's simile, 'like the unfolding of a picture-strip: each word expresses a picture, a pictorial i
mage, regardless of whether it signifies an object or an action'. In Golding's novel The Inheritors the Neanderthal men always say 'I had a picture' when they mean 'I thought of something'; and anthropologists agree that for once a novelist got the picture right.

  Thus the poet who reverts to the pictorial mode of thought is regressing to an older and lower level of the mental hierarchy -- as we do every night when we dream, as mental patients do when they regress to infantile fantasies. But the poet, unlike the dreamer in his sleep, alternates between two different levels of the mental hierarchy; the dreamer's awareness functions on one only. The poet thinks both in images and verbal concepts, at the same time or in quick alternation; each 'trouvaille', each original find, bisociates two matrices. The dreamer floats among the phantom shapes of the hoary deep; the poet is a skindiver with a breathing tube.