The Act of Creation

by Arthur Koestler

  Some writers identify the creative act in its entirety with the unearthing of hidden analogies. 'The discoveries of science, the works of art are explorations -- more, are explosions, of a hidden likeness', Bronowski wrote. [15] But where does the hidden likeness hide, and how is it found? Sultan's branch could literally be seen as a stick -- though even in this case, a change of the perceptual frame was required to discover the likeness. But in most truly original acts of discovery the 'seeing' is in fact imagining; it is done in the mind's, and mostly the unconscious mind's eye. The analogy between the life of one kind of microbe inside a cow and another kind of microbe in a forgotten culture tube was not 'hidden' anywhere; it was 'created' by the imagination; and once an analogy has been created, it is of course there for all to see -- just as a poetic metaphor, once created, soon fades into a cliché.

  Analogy, in logic, means a process of 'reasoning from parallel causes'; in common parlance it means that two situations or events are similar in some respects, but not in all respects. The rub is in the words 'parallel' and 'similar'; the latter, in particular, has bedevilled psychology ever since the term 'association by similarity' was invented (by Bain, I believe) as an explanation of how the mind works. A Chinaman who collects stamps is 'similar' to a Negro in that both are males; he is similar to a Chinese girl in that both are Chinese; and he is similar to other stamp-collectors of any nationality. Mathematics began, wrote Bertrand Russell, when it was discovered that a brace of pheasants and a couple of days have something in common: the number two.

  'Similarity' is not a thing offered on a plate (or hidden in a cupboard); it is a relation established in the mind by a process of selective emphasis on those features which overlap in a certain respect -- along one dimensional gradient -- and ignoring other features. Even such a seemingly simple process as recognizing the similarity between two letters a written by different hands, involves processes of abstraction and generalization in the nervous system which are largely unexplained.

  Thus the real achievement in discoveries of the type mentioned in this section is 'seeing an analogy where no one saw one before'. The scientist who sets out to solve a problem looks at it from various angles, through glasses of different colours, as it were -- in the jargon of the present theory, he experiments with various matrices, hoping that one will fit. If it is a routine problem of a familiar type, he will soon discover some aspect of it which is similar in some respect to other problems encountered in the past, and thus allows him to come to grips with it. Some of the mental operations involved in such routine cases we have already encountered in discussing the solving of witty riddles (pp. 84-6): extrapolation, interpolation, transposition. These are 'rules of the game' which enter as sub-codes into any complex mental skill. To put it in a different way: solving a problem means bridging a gap; and for routine problems there usually exist matrices -- various types of prefabricated bridges -- which will do the job; though it may require a certain amount of sweat to adjust them to the terrain.

  But in original discoveries, no single pre-fabricated matrix is adequate to bridge the gap. There may be some similarities with past situations, but these may be more misleading than helpful, and lure the victim into fruitless experimentation based on traditional rules of the game. Here the only salvation lies in hitting on an auxiliary matrix in a previously unrelated field -- the larvae of starfish or the Holy Ghost. One may call the process which follows after the hit 'reasoning from a parallel case' -- but the real achievement was to 'appoint', as it were, the larva as a parallel case to the pus, and the action of the Holy Ghost as 'similar' to the action of gravity. It is an achievement much closer to the birth of a poetic simile than to a logical production. After all, the Walrus too was arguing by analogy when he talked Of shoes -- and ships -- and sealing wax / Of cabbages -- and kings.

  The essence of discovery is that unlikely marriage of cabbages and kings -- of previously unrelated frames of reference or universes of discourse -- whose union will solve the previously unsoluble problem. The search for the improbable partner involves long and arduous striving -- but the ultimate matchmaker is the unconscious. I have discussed several tricks which qualify it for that role: the greater fluency and freedom of unconscious ideation; its 'intellectual libertinage' -- as one might call the dream's indifference towards logical niceties and mental prejudices consecrated by tradition; its non-verbal, 'visionary' powers. To these must be added, in our present context, the dream's tendency towards creating unusual analogies. These may be verbal puns, or 'optic puns' or visual symbols; but there is another type of vague and cloudy analogy generated in the dream and half-dream, which disintegrates on awakening and cannot be put into words -- except by muttering 'something reminded me of something, but I don't know what reminded me of what, and why'. Some dreams have a way of dissolving in the wakening mind like solid crystals melting in a liquid; and if we reverse the process we get at least a speculative pointer to the manner in which those 'somethings' vaguely reminding me of other 'somethings' condense into a nascent analogy. This may be a hazy, tentative affair -- the dance of Poincaré's unhooked atoms; and its shape may be changing from camel to weasel, as Hamlet's cloud. The unconscious regions of fertile minds must be pullulating with such nascent analogies, hidden likenesses, and the cloudy forms of things unknown. But most clouds form and dissolve again; only a few intuitions reach the stage of 'seeding the cloud' which results in the formation of verbal drops; and cloud-bursts are a rarity.

  Two final examples may serve to illustrate the actual process of discovering hidden analogies. The first is related to clouds in a literal sense -- Franklin's invention of the lightning conductor.

  Benjamin Franklin became interested in electricity in 1746 when he was forty, and began playing about with Leyden jars -- a kind of electrified bottle which gave one fearful shocks. Within the next three years he rediscovered by himself virtually everything that was known about electricity to that date, and added several fundamental discoveries of his own.

  In 1749 he noted in his diary that he thought lightning and thunder to be electrical phenomena.* He also found that when brought near to an electrified body, a pointed object, like a finger, will draw a much stronger spark than a blunt one. 'To know this power of points', he musingly wrote, 'may possibly be of some use to mankind, though we should never be able to explain it.' He then drew an analogy between a cloud and an electrified body, and concluded that lightning was an electrical discharge phenomenon. But if that was the case, mankind could protect itself against this cosmic scourge:

  I say, if these things are so, may not the knowledge of this power of points be of use to mankind, in preserving houses, churches, ships & cont. from the stroke of lightning, by directing us to fix on the highest parts of those edifices, upright rods of iron made sharp as a needle, and gilt to prevent rusting, and from the foot of those rods a wire down one of the shrouds of a ship, and down her side till it reaches the water? Would not these pointed rods probably draw the electrical fire silently out of a cloud before it came nigh enough to strike, and thereby secure us from that most sudden and terrible mischief? [16]

  However, before he could convince mankind to put 'Franklin rods' on their houses he had to prove his fantastically sounding notion that thunderclouds were in fact giant Leyden jars floating in the air. He waited for some time hopefully for the erection of a tall spire at Philadelphia, intending to fix a pointed rod on top of it, and so to bring down the electricity from a passing thundercloud. But the difficulties of the project proved insurmountable; it was during this period of impatient waiting and restless searching for a simpler method to prove his theory that he hit on the fantastic yet at the same time astonishingly simple idea of the kite.

  How did it happen? Franklin was an expert swimmer. On his first sojourn in London, at the age of nineteen, he swam from Chelsea to Blackfriars, a distance of three miles, 'performing on the way many feats of activity both upon and under the water' -- and was advised by some English gentleme
n, who watched him, to open a swimming school. He did not do that, but he devised a new method of learning to swim: 'Choosing a place where the water deepens gradually, walk coolly into it till it is up to your breast, then turn around, your face to the shore, and throw an egg into the water between you and the shore.' The learner then must 'boldly retrieve the egg' -- and in the act of retrieving acquires the art of swimming.

  Even earlier on he had devised another aquatic sport: as a boy he used to drift for hours on a lake, floating on his back, and towed by the string of a kite. He suggested that this method might be utilized by swimmers to cross the Channel from Dover to Calais -- with the judicious addendum: 'The packet-boat, however, is still preferable.'

  It is easy to imagine how, in a moment of weariness and 'thinking aside' from that wretched spire in Philadelphia, a pleasant childhood memory rose like a bubble to the surface of his consciousness: drifting on the lake attached to the kite in the sky. Eureka! With the enthusiastic assistance of his young son, Franklin fabricated a kite out of a cross of cedar wood and a silk handkerchief. All he needed now were a few good thunderclouds -- which conveniently appeared in June 1752. Father and son sent up the kite and, with due precaution, drained the clouds' electric charge into a Leyden jar; 'by the electric fire thus obtained spirits were inflamed and other experiments performed'.

  Such was the excitement caused all over the world that one of Franklin's imitators, a certain Monsieur Riehmann, was killed in St. Petersburg by the lightning discharge he drew from a cloud. He was worshipped as a hero and found many would-be imitators; among them the German inventor Herr Boze. Even Joseph Priestley, one of the great British scientists of the century, rhapsodized about 'the sentiments of the magnanimous Mr. Boze, who with a truly philosophic heroism, worthy of the renowned Empedocles, said he wished he might die by the electric shock, that the account of his death might furnish an article for the memoirs of the French Academy of Sciences. But it is not given to every electrician to die the death of the justly envied Riehmann.' [17]

  There are two successive Eureka processes involved in this story. In the first, the bisociative link was what Franklin called 'the power of points'; it gave rise to the analogy: pointed finger discharges Leyden jar, pointed rod discharges cloud. It may have been attained by ideation on a relatively conscious level, probably with the aid of visual imagination. The second stroke of genius was the use of the kite to reach the thunderbolt. It illustrates the argument I have put forward earlier in this chapter: one can hardly say that a hidden analogy was pre-existent in the universe between a kite used as a sail by a boy floating on a lake, and a lightning conductor. What actually happened was that Franklin was desperately searching for a means to make contact with a thundercloud, thinking in habitual terms of tall spires, long iron rods, and perhaps the Tower of Babel. But all these approaches proved impracticable, and the matrix was blocked -- until in a moment of lassitude and day-dreaming the previously unrelated memory-train of swimming, egg-retrieving, and kite-sailing was brought to bear on it.

  The last example that I shall quote in this section is a particularly impressive illustration of the unconscious in the role of matchmaker. I am referring to the discovery, in 1920, of the chemical transmission of nerve-impulses by Otto Loewi. Since the matter is somewhat technical, I shall give a simplified account of it.

  Before Loewi's discovery it was generally believed that nervous control of bodily functions was exercised by a direct transmission of electrical impulses from nerve-terminal to muscle or gland. But this theory failed to account for the fact that the same type of electric impulse travelling down a nerve had an excitatory effect on some organs, an inhibitory effect on others. Now certain drugs were known to have precisely the same effect. In a discussion with a friend in 1903, it occurred to Loewi that the chemical agents which were contained in these drugs may also be present at the nerve-terminals; the electric impulse would initiate chemical action, which in its turn would act on the muscle or gland. But Loewi could not think of an experimental method to test the idea -- and forgot it for the next seventeen years.

  Fifteen years later, for quite different purposes, he designed an experiment. He made preparations of two frogs' hearts which were kept beating in salt solutions to see whether their activities gave out any chemical substance. In the sequel he forgot all about the experiment.

  Another two years passed until the critical event:

  The night before Easter Sunday of that year [1920] I awoke, turned on the light, and jotted down a few notes on a tiny slip of thin paper. Then I fell asleep again. It occurred to me at six o'clock in the morning that during the night I had written down something most important, but I was unable to decipher the scrawl. The next night, at three o'clock, the idea returned. It was the design of an experiment to determine whether or not the hypothesis of chemical transmission that I had uttered seventeen years ago was correct. I got up immediately, went to the laboratory, and performed a simple experiment on a frog heart according to the nocturnal design. . . .

  No lesser person that Walter B. Cannon, the discoverer of adrenalin, has described this noctural design as 'one of the nearest, simplest, and most definite experiments in the history of biology'. Loewi again isolated two frog hearts, the first with its nerves, the second without. He stimulated the vagus nerve of the first heart for a few minutes. The vagus has an inhibitory effect on the heart, and its beats slowed down. Loewi now removed the salt solution from the first heart and applied it to the second. It slowed down just as if its own (no longer existent) vagus had been stimulated. . . . He repeated the experiment, this time stimulating the accelerator nerve of the first heart. When the liquid was transferred to the second heart it accelerated. . . . He concludes:

  These results unequivocally proved that the nerves do not influence the heart directly but liberate from their terminals specific chemical substances which, in their turn, cause the well-known modifications of the function of the heart characteristic of the stimulation of its nerves. The story of this discovery shows that an idea may sleep for decades in the unconscious mind and then suddenly return. Further, it indicates that we should sometimes trust a sudden intuition without too much scepticism. If carefully considered in the daytime, I would undoubtedly have rejected the kind of experiment I performed. It would have seemed likely that any transmitting agent released by a nervous impulse would be in an amount just sufficient to influence the effector organ. It would seem improbable that an excess that could be detected would escape into the fluid which filled the heart. It was good fortune that at that moment of the hunch I did not think but acted immediately.

  For many years this nocturnal emergence of the design of the crucial experiment to check the validity of a hypothesis uttered seventeen years earlier was a complete mystery. [18]

  In 1955 -- thirty-five years after the discovery, which earned him the Nobel Prize -- Loewi had to compile a bibliography:

  I glanced over all the papers published from my laboratory. I came across two studies made about two years before the arrival of the nocturnal design in which, also in search of a substance given off from the heart, I had applied the technique used in 1920. This experience, in my opinion, was an essential preparation for the idea of the finished design. In fact, the nocturnal concept represented a sudden association of the hypothesis of 1903 with the method tested not long before in other experiments. Most so-called 'intuitive' discoveries are such associations suddenly made in the unconscious mind. [19]

  Let me briefly recapitulate the three stages of this drama. The first is the sudden emergence, during a conversation in 1903, of the hunch that his problem could be solved by switching from a 'spark theory' to a 'soup theory' (in neurological jargon, 'spark' refers to electrical, 'soup' to chemical transmission of nerve impulses). But a hunch of this kind as often as not turns out to be a fallacious over-simplification; so the idea went into the incubator for the next seventeen years, till 1920.

  Act Two. In 1918, fifteen years after t
he hunch, Loewi performs certain experiments for which purpose he has to design a technique for the detection of fluids secreted by the frog's heart. He then forgets all about it.

  On the night before Easter Sunday the two previously unrelated memories meet; but their meeting place is so deep underground that the next morning he can remember nothing, and cannot even decipher his own scribbled note. He has to wait until the next night for another underground excursion -- which takes place at 3 a.m., followed by the rush to the laboratory.

  After the event one wonders, of course, why one idea had to wait for seventeen years, the second for two years, and then choose such a secret place for their final rendezvous that the identity of the second was only revealed another twenty-five years later. The first was a theory of the transmission of nerve impulses to organs by a fluid; the second was a technique for tracing fluids in an organ; what could be more logical than that the twain should meet? Yet they did not meet through all those years because mortal minds, even those of genius, are not governed by logic but by habit, and the two ideas were embedded each in its own habitual context. Wallace, too, had been thinking of evolution for two years, and had read Malthus many years before the two fused -- during an attack of tropical fever. It seems that encounters of this kind can occur only when the normal rules of the game are suspended and the unconscious match-maker enters into action. Loewi's inability to read his own note, and other cases of 'snowblindness' which I shall mention, indicate the stubborn resistance of habit against such breaches of the rules and illicit liaisons.