The Act of Creation

by Arthur Koestler

  Regeneration and Psychotherapy

  Less extreme cases are neurotics who react to their traumatic experiences by elaborating defence systems which enable them to find some kind of modus vivendi with the world. One may call such behaviour-patterns 'faulty integrations' -- like the newt's whose forelegs move backwards. Psychotherapy aims at undoing faulty integrations by inducing a temporary regression of the patient to an earlier level, in the hope that he will eventually reintegrate into a more stable pattern. Neuro-surgery, shock-therapy, and related methods aim at releasing philogenetically older centres of the brain from cortical restraints. In a less drastic form, Freudians, Jungians, etc., try to make the patient revert to unconscious and infantile planes of experience, and to regenerate, as it were, into a more or less new-born person.

  Thus psychotherapy may be called an experiment in artificially induced regeneration. It relies on the same basic process of reculer pour mieux sauter, which we see operating on every level: from the flatworm which replaces a lost head, through the crab which adjusts its gait to the loss of a leg, to the rat which, unable to turn to the right, makes a three-quarter turn to the left. We found the same pattern repeated on the level of human creativity: the scientist, faced by a perplexing situation -- Kepler's discrepant eight minutes' arc, Einstein's light-traveller paradox -- must plunge into a 'dark night of the soul' before he can re-emerge into the light. The history of the sciences and arts is a tale of recurrent crises, of traumatic challenges, which entail a temporary disintegration of the traditional forms of reasoning and perception: a de-differentiation of thought-matrices, a dismantling of its axioms, a new innocence of the eye; followed by the liberation from restraint of creative potentials, and their reintegration in a new synthesis.

  The Routine of Dreaming

  There is also a mental equivalent for the less spectacular routine regeneration of tissues, designed to compensate for wear and tear. The analogue process is the maintenance of 'mental tissues' exposed to the wear and tear of diurnal stresses, by the regenerative effect of nocturnal regressions to the primitive levels of the dream. Experimental evidence seems to indicate that the restorative powers of sleep are primarily derived from the process of dreaming. Experimental subjects who were woken up each time their EEG waves indicated the onset of dreaming, displayed symptoms of fatigue and nervous disorder; long periods of dreamless sleep could not compensate for dream-deprivation. 'Man cannot persist long in a conscious state,' wrote Goethe, 'he must throw himself back into the unconscious, for his root lives there.'

  We have seen (Book One, VII, VIII) that these periodic plunges into the unconscious are accompanied by the temporary disintegration of matrices of logical thought. But they also entail a partial loss of identity, a de-differentiation of the personality -- as indicated by the remarkable degree of uniformity in the contents of dreams shared by people of very different character, and by the relatedness of these contents to mythological themes and symbols. These shared patterns led Jung to postulate a 'collective' -- that is, individually undifferentiated -- level of the unconscious. On that level, members of the same culture seem to share some degree of psychic equipotentiality expressed in 'archetypal symbols'. These are supposed to be condensations of basic experiences of the race in the distant past; hence their great emotion-rousing potential.

  To recapitulate: the fact that art and discovery draw on unconscious sources indicates that one aspect of all creative activity is a regression to ontogenetically or philogenetically earlier levels, an escape from the restraints of the conscious mind, with the subsequent release of creative potentials -- a process paralleled on lower levels by the liberation from restraint of genetic potentials or neural equipotentiality in the regeneration of structures and functions. The scientist, traumatized by discordant facts, the artist by the pressures of sensibility, and the rat by surgical intervention, share, on different levels, the same super-flexibility enabling them to perform 'adaptations of a second order', rarely found in the ordinary routines of life. ### Cf. Learning III ###

  Regeneration and Creativity

  I must enlarge a little on this seemingly sweeping analogy, and try to show that it is in fact based on homologous principles, traceable on all levels of the hierarchy, and preserving their basic pattern throughout them.

  Differentiation and specialization of the parts are necessary for the normal functioning of the whole; abnormal conditions call for radical measures which may include a retreat of the over-exerted part to a structurally less differentiated, functionally less specialized stage, if the whole is to survive. The 'part' may be the newt's amputation stump, or the unsolved problem in the scientist's mind which tortures and obsesses him. We have seen that such regressions are mostly pathogenic, but under favourable conditions they may redress the situation by re-activating potentials which had been operative in the past but are inhibited in the adult -- such as the regulative powers of the embryo in the womb or the undifferentiated total-pattern-responses of its nervous system. The period of incubation is a similar retreat, if not into the womb, at least into long-outgrown forms of ideation, into the pre-verbal, pre-rational games of the unconscious, the wonderland-logic of the dream. The challenge which sets the process going is in all cases a traumatic experience: physical mutilation or mental laceration -- by data which do not fit, observations which contradict each other, emotions which disrupt approved styles in art: experiences which create mental conflict, dissonance, perplexity. The 'creative stress' of the artist or scientist corresponds to the 'general alarm reaction' of the traumatized animal; the anabolic-catabolic sequence of de-differentiation and reintegration corresponds to the destructive-constructive sequence in the creative act. The 'physiological isolation' of the over-excited part which tends to dominate, corresponds to the single-minded and obsessive preoccupation with the idée fixe -- Kretschmer's 'over-valued idea', Kepler's pursuit of a chimera -- which monopolizes the whole mind; it will either lead to its reorganization by giving birth to a new system, or to the cancerous proliferation of a degenerate tissue of ideas.

  Over-excitation of an organ or part is one of the four causes of 'physiological isolation'. The other three were: growth of the whole beyond a critical limit; senescence; and (partial) blockage of communication (pp. 452 f.). Each situation has its parallels on the mental plane -- of the individual, or the history of thought. The unmanageable size of the total body of human knowledge -- or even a single province thereof -- created that dissociated phenomenon, the specialist mind; senescent cultures produce degenerate art-forms; blocked communications between Ptolemaic astronomy and the main body of the physical sciences led to the untramelled proliferation of epicycles in a closed system, divorced from reality.

  'It is wonderful to see how analogies can blossom when they are given a little affection', wrote the authors of a book I have repeatedly quoted. [10] Particularly, one might add, if they have solid roots in the earth. So let me carry analogy one step further. In Book One (Chapters V-VIII) I have described various aspects of the Eureka process; each of these re-structurings of thought has its obvious correlate in regenerative processes on lower levels. The 'displacement of emphasis' to a previously irrelevant part or aspect of experience corresponds to the sudden dominance of a hitherto subordinate part of an organism -- such as the crab's second leg which becomes a pacemaker. The 'reversal of logic' (or of the figure-background relation) has its parallel in the reversal of physiological gradients during regeneration. When psychological textbooks describe Duncker's experiments as 'detaching' part of a visual percept from the context in which it is 'embedded', and 'attaching' it to the new context of the problem to be solved (pp. 189 f.) this description itself is based on analogies from physiological processes. During incubation, the intuitive groping of ideas towards the 'good combination', and their guidance by 'gradients in the unconscious', reminds one of the biochemical gradients in morphogenesis, or the 'contact-guidance' of out-growing nerve-processes towards their end-organ. Lastly a 'nasce
nt', unverbalized analogy may be compared to an unarticulated organ-primordium.

  But these genetic skills operate only in the embryonic stage of development; in the adult they are superseded by the integrative action of the nervous system -- unless the embryonic potentials are reactivated by regenerative needs. Similarly, the adult's mental co-ordination relies on conscious, verbalized, 'logical' codes; not on the quasi 'embryonic' (infantile, pre-causal) potentials of the unconscious; again unless these are revived under the creative stress. Physical regenerations strike us as 'spectacular pieces of magic' because they derive from prenatal skills; and creative inspirations are equally mysterious because they derive from levels which predate the conscious mind. As Polànyi wrote (in a different context): 'The highest forms of originality are far more closely akin to the lowest biotic performances than the external circumstances would indicate.' [11]

  Regeneration and Evolution

  These rather fancy diagrams are solely meant to indicate in a crude way the complementary factors in the reculer pour mieux sauter phenomenon. In A, increase in tissue-differentiation entails a reciprocal decrease of genetic multipotentiality. In B, an analogous reciprocity prevails between unconscious intuitions and automatized routines -- or, if you like, between fluid imagery and 'misplaced concreteness'. R indicates the 'regenerative span'. (The curve in A should of course have breaks and a series of discrete steps.)

  It could be objected that structural regenerations merely restore the status quo ante whereas mental reorganization leads to an advance. But in the first place this is not always the case. Psychotherapy aims at correcting 'faulty integrations' caused by traumatic experiences -- at restoring normality. In the second place the biological evolution of the species with which we are concerned has to all intents and purposes come to a standstill, whereas mental evolution continues, and its vehicle is precisely the creative individual. The Eureka process is a mental mutation, perpetuated by social inheritance. Its biological equivalent are the genetic mutations which carried the existing species up the evolutionary ladder. Now a mutation -- whatever its unknown cause -- is no doubt a re-moulding of previous structures, based on a de-differentiation and reintegration of the otherwise rigid genetic code. The transformations of fins into legs, legs into arms, arms into wings, gills into lungs, scales into feathers, etc., while preserving certain basic structural patterns (see, for instance, d'Arcy Thompson's On Growth and Form) were eminently 'witty' answers to the challenges of environment. It seems obvious that the dramatic release, at periods of adaptative radiations, of unexplored morphogenetic potentials by a re-shuffling of molecules in the genetic code, resulting in the de-differentiation and reintegration of structures like limbs into wings, is of the very essence of the evolutionary process. After all, 'ontogenesis and regenesis are components of a common mechanism', [12] which must have a phylogenetic origin.

  In Book One, Chapter XX, I have mentioned the perennial myth of the prophet's and hero's temporary isolation and retreat from human society -- followed by his triumphant return endowed with new powers. Buddha and Mohamed go out into the desert; Joseph is thrown into the well; Jesus is resurrected from the tomb. Jung's 'death and rebirth' motif, Toynbee's 'withdrawal and return' reflect the same archetypal motif. It seems that reculer pour mieux sauter is a principle of universal validity in the evolution of species, cultures, and individuals, guiding their progression by feedback from the past.

  NOTES

  To p. 454. It is still an open question, however, whether or how much undifferentiated 'reserve cells' (as in lower animals) contribute material to the blastema.

  To p. 454. It seems that the initial role of the nervous system is to determine the main axis of the regenerate -- that it acts, not as an inductor, but as a trophic agent. At the later, anabolic stages of the process no nerve supply is needed -- as denervation experiments show.

  To p. 456. This, actually, is the only clearly demonstrated case of metaplasia among higher animals.

  To p. 459. About the ways how this is achieved, cf. McCulloch in the Hixon Symposium, p. 56.

  V

  PRINCIPLES OF ORGANIZATION

  Before we turn to adult behaviour, a pause for stocktaking may be in order.

  In a four-dimensional continuum, embryonic development would be represented by an ascending hierarchy of spatial levels perpendicular to the time axis. In spite of the perplexing diversity of phenomena on different levels -- cleavage, gastrulation, induction, neuro-genesis -- certain basic principles were seen to operate on every level throughout the hierarchy. Principles (or 'laws of nature') can only be described in symbolic language of one kind or another. The language used in the present theory is based on four key-concepts: motivation, code, matrix, and environment. Since these are assumed to operate in a hierarchic framework, the dichotomy of self-asserting and participatory tendencies of behaviour on all levels need not be separately postulated, but derives logically, as it were, from the dual character of every sub-whole as a sub-ordinate and supra-ordinate entity. Let me now recapitulate some of the main points which have emerged from the previous chapters, taken in conjunction with Book One:

  1. Motivation in embryonic development is a subject for the metaphysician. J. Needham's tongue-in-the-cheek phrase about 'the striving of the blastula to grow into a chicken' indicates the directiveness of the morphogenetic process and its equifinal, regulative properties, which become particularly evident under adverse conditions. These properties represent the genetic precursors of the motivational drives, needs, and goal-directedness of the adult animal; during maturation, the former shade into the latter, and there is no sharp dividing line between them.

  2. 'A part is a whole is a part'. Each sub-whole is both a 'sub' and a 'whole'. Facing downward or outward in the hierarchy, it behaves as an autonomous whole; facing upward or inward, it behaves as a dependent part which is inhibited or triggered into action by higher controls. One might call this the 'Janus principle' in organic (and social) hierarchies.

  3. The 'whole'-aspect of the sub-whole is manifested in its autonomous and spontaneous activities. The principle of autonomy is asserted on every level; from cell-organelles functioning as power plants or motor units, through the self-differentiating activities of the morphogenetic field, to the autonomous regulations of organs and organ system. In the motor hierarchy, it is reflected at every stage, from the muscle's selective response to specific excitation-patterns, through the stubborn behaviour of the reversed newt limb, to the unalterable features in a person's gait or handwriting. I have briefly mentioned (Book One, Chapters XIII, XXI), and shall discuss in more detail later, autonomous mechanism in perceptual organization -- visual constancies, the automatic filtering, analysing, generalizing of the input. Lastly, thinking and communicating are based on hierarchically ordered, autonomous patterns of enunciation, grammar, logic, mathematical operations, universes of discourse.

  The dynamic aspect of the part's autonomy is manifested in its apparently spontaneous, unprovoked rhythmic activities which are 'modified but not created by the environmental input' -- a statement which equally applies to morphogenesis, to intrinsic motor patterns, to the spontaneous discharges of unstimulated sensory receptors, the electric pulses of the unstimulated brain, to drives in the absence of external stimuli or to communications addressed to imaginary audiences.

  Autonomy and spontaneity taken together constitute what I have called the 'self-assertive' aspect of part-behaviour.

  4. The opposite aspect is the part's dependence on supra-ordinate controls which may be said to represent the interests of the whole vis-à-vis the part in question. The controls are largely of an inhibitory or restraining character, to prevent overloading of information channels, over-shooting of responses, confusion and redundancy in general; while the activation of the part is effected by signals of the trigger-release type. During morphogenesis, control is exercised by the suppression of unrequired, and the release of required genetic potentials; in the mature organism by
interlacing multiple hierarchies of nervous and circulatory processes, and biochemical gradients.

  5. Sub-wholes on any level of the hierarchy function (a) autonomously in supra-ordination to their own parts, (b) in sub-ordination to their controlling agency, and (c) in co-ordination with their environment.

  'Environment' is a relative term. On the level of the single cell, the environment of the nucleus is cytoplasm; on the level of the morphogenetic field, the environment of one cell-population is another cell-population; each organ in the adult animal is bathed in body-fluids which constitute its environment. The structure and function of any sub-whole is determined by (a) and (b) -- its intrinsic pattern and its controls in the 'vertical' hierarchy to which it belongs; but it is also affected by inputs and feedbacks from its 'horizontal' environment, as it were -- the lie of the land. The difference between (a) and (b) on the one hand, and (c) on the other is that the former determine the invariant pattern of the operation, (c) only its variable details. To mention a few examples: feedback from the cytoplasm to the genetic blue-print co-determines into which variety of specialized cell that particular unit will develop; but it does not alter the blue-print itself. Inductor substances in the immediate environment of a tissue will promote its differentiation into an organ -- but only within the limits of the tissue's 'competence'. Environmental hazards decide the neuro-muscular connections in the grafted salamander-limb, but its functional co-ordination remains unaffected by it: it is controlled by its 'vertical' hierarchy, and the lie of the land in the scar-tissue determines only the local 'tactics' of the outgrowing nerve-filaments.