The Act of Creation

by Arthur Koestler

  The example further illustrates a point mentioned before: a matrix on the n-level is represented on the n+1 level by its code. There is, under normal conditions, no direct commerce between its members on the n-1 level, and the co-ordinating agency on the n+1 level. If the latter interferes directly with the former, routines become disorganized, and we get the 'paradox of the centipede'. Loss of direct control over automatized processes on lower levels of the body hierarchy is part of the price paid for differentiation and specialization. The price is of course worth paying so long as the species lives in an environment that is fairly stable. It is after all not part of the normal destiny of the salamander to encounter Dr. Paul Weiss.

  NOTES

  To p. 431. The validity of Coghill's findings for a whole range of other species -- cat, bird, man -- was demonstrated in a general way by authors like Coronios (1933), Herrick (1929), Kuo (1932). However, some geneticists (e.g. Windle and his associates) have maintained that functional co-ordination in higher species is the result of additive chaining of specialized local reflexes. As against this, Hooker (1950) has pointed out the undisputed fact that motor nerves are functional before the sensory nerves, and that the sensory and intercallated neurons in the reflex arc are the last to become functional, which amounts to an indirect refutation of the reflex summation view. For a summary of this controversy see Thorpe (1956) pp. 20 ff. and p. 45; also Barton (1950) and Hooker (1950).

  On the other hand, Tinbergen has shown that in some patterns of complex instinct behaviour (e.g. nest building, Kortlandt, 1940), the part-performances which go into the total pattern emerge at different times, following 'a fixed time pattern just as with growing morphological structures' (Tinbergen, 1951, p. 136). Thus, for instance, fastening of twigs in the nest precedes searching for twigs. The existence of 'internal clocks' which regulate the serial activation of the various sub-codes of the integrated performance is entirely in keeping with the total pattern view. Thorpe concludes: 'Embryological studies now suggest that ontogenetically, complex muscle co-ordinations resembling fixed action patterns [see below] precede responses of the simple reflex type in mammals.' Cf. next note.

  To p. 432. This confusion may have been a contributing factor in the controversy mentioned in the previous note. Since myogenic muscle contractions can be produced in embryos by electrical or mechanical stimulation before neuro-muscular integration is established, the 'isolated reflex-school' assumed such pseudo-reflexes to be true reflexes and the primary elements of adult behaviour. See, e.g. Thorpe, loc. cit.

  To p. 434. The Hixon Symposium was one of the most fertile exchanges ever held between leading experts in various disciplines. Among its partialpants were H. Kluver, Wolfgang Köhler, K. S. Lashley, W. S. McCulloch, John V. Neumann, R. W. Gerard, Lorente de Nó, Paul Weiss, Linus Pauling, etc., to mention only a few. No wonder that Weiss, carried away by enthusiasm after hearing Lashley's brilliant paper on 'The Problem of Serial Order in Behaviour', concluded with expressing his hope that today's 'discussion will mark a turning point in the building of neurological theories'.

  To p. 436. To make matters a little more complicated, we may remind ourselves in passing that muscle-contractions serve not only motility but also maintenance of tone and temperature in the organism; sometimes they serve only the last function alone (in shivering) since muscle is a main source of animal heat.

  To p. 436. At the time of writing this theory still has certain difficulties to overcome; among them the fact that the energy supply of the fibre depends not only on ATP itself but also on creatin phosphate.

  III

  DYNAMIC EQUILIBRIUM AND REGENERATIVE POTENTIAL

  Acting and Reacting

  'The organism', to quote Coghill once more, 'acts on the environment before it reacts to the environment.' This statement seems to apply to every level and every aspect of organic life. The lowliest creature and the highest, the moment it is hatched or born, lashes out at the environment, be it liquid or solid, with cilia, flagellae, or contractile muscle fibre; it crawls, swims, glides, pulsates; it kicks, yells, breathes, feeds, and sucks negative entropy from its surroundings for all it's worth.

  The patterns of these built-in motor activities we saw to be to a large extent autonomous; 'the structure of the input does not produce the structure of the output, but merely modifies it.' Moreover, the input itself is actively controlled and modified by the central nervous system from the moment it impinges on the peripheral receptor organs; and recent developments have caused, at least among an unorthodox minority of psychologists, a distinct 'shift from the notion that an organism is a relatively passive, protoplasmic mass whose responses are controlled by the arrangement of environmental stimuli to a conception of an organism that has considerable control over what will constitute stimulation.' [1]

  Even below the level of the single cell, organelles such as the mitochondria and kinetosomes carry on their autonomous activities; their shadowy patterns under the electron-microscope are a reminder that the emergence of life means the emergence of spontaneous, organized exertion to maintain and reproduce originally unstable forms of equilibrium in a statistically improbable system in the teeth of an environment governed by the laws of probability. The live organism succeeds in this by creating an inner environment with which to confront the outer environment -- and in which the law of entropy seems to be reversed, biological clocks replace astronomical clocks, and hierarchic order reigns supreme.

  An organism is said to be 'well balanced' or 'well adapted' or 'in dynamic equilibrium' if it has established a modus vivendi between its internal and external environment. This, of course, is a more complex form of balance than mechanical or chemical equilibrium; it implies metabolic processes required for the maintenance of form and function in an open system in perpetual flux -- Bertalanffy's (1941) Fliessgleichgewicht; it implies self-regulating devices which keep irritability and motility within a safe standard range; and it also implies the slow, cyclic changes of morphogenesis, maturation, and reproduction, regulated by biological clocks. If all these processes are to be lumped together under the portmanteau word 'adaptation', then we must call it adaptation of a special kind, on the organism's own terms; after all, the perfect adaptation of an organism to the temperature and chemistry of the environment is to die. In fact, the animal does not merely adapt to the environment, but constantly adapts the environment to itself. It eats environment, drinks environment, fights and mates environment, burrows and builds in the environment; and even in observing environment, it modifies, dismantles, analyses, and reassembles it after its own fashion, converting 'noise' into 'information'. 'Perception', Woodworth wrote, 'is always driven by a direct, inherent motive which might be called the will to perceive.'

  Thus the terms 'adaptation', 'environment', 'equilibrium' will be used in the following pages not in their usual passive connotations, but with active overtones, as it were. Instead of treating an animal as a 'relatively passive, protoplasmic mass whose responses are controlled by the arrangement of environmental stimuli' -- a Pavlov-dog in its restraining harness -- we shall regard it as a relatively self-contained organism, deploying spontaneous activities simultaneously on various levels of its constituent functional hierarchies -- activities which are triggered and modified, but not created by the environment.

  What is Equilibrium?

  An organism can be said to function normally so long as the stresses between internal and external milieu do not exceed a certain standard range. To simplify the argument, let the term 'internal milieu' embrace all processes within the organism, and let us lump together the nature, intensity, and duration of environmental excitations in a single variable. We shall then be able to distinguish between (a) 'normal', (b) 'paranormal' or 'traumatic', and (c) destructive environmental conditions -- though, needless to say, the boundaries between them cannot be sharply defined.

  The term 'dynamic equilibrium' shall apply only to a normal organism functioning under 'normal' conditions. Under these conditio
ns the organ-systems, organs, and organ-parts of the animal perform their specific, autonomous functions as sub-wholes, at the same time submitting to the regulative control imposed by the higher centres. The control is exercised by excitatory and inhibitory processes, but the latter play a vastly greater part. From the moment of conception, the genetic potentials of the individual cell are further restrained with every step in differentiation; and on every level of the growing and mature organism inhibitory blocks, negative feedbacks, growth-inhibiting hormones are at work. In the nervous system, in particular, there is censorship at every step -- to prevent overloading of the information channels and overshooting of responses. Without this hierarchy of restraints, the organism would instantly blow its fuses in a kind of delirium agitans and then collapse.

  Under normal conditions the part will not tend to escape the restraining influence of the whole. Under paranormal conditions the balance is upset. Thus the term 'balance' or 'equilibrium' takes on a special meaning in the context of an organic hierarchy: it is not meant to refer to relations between parts on the same level of the hierarchy, but to the relation of a part to its controlling centre on the next higher level. The stresses arise not between inputs 'competing for the final common path,' as the expression goes, not between 'antagonistic drives' or 'conflicting impulses' (which do not directly communicate with each other and cannot 'fight it out among themselves') -- but between the excited part and the whole, whose attention it is trying to monopolize: in other words, between the self-assertive tendencies of the part and the restraints imposed by the controlling centre. Equilibrium is maintained in the organism by rules comparable to the procedure in a law court where the opposing parties address themselves not to each other, but to the judge.

  This interpretation of equilibrium in a hierarchy was suggested in my Insight and Outlook (p. 139 seq.), and independently proposed by Tinbergen. In discussing the competition between various 'fixed response patterns' in innate behaviour, Tinbergen wrote: 'It should be emphasized that it is quite possible that these interconnections [between the competing centres] do not in reality run directly from one centre to the other, but go by way of the superordinated centre.' [2] Thorpe has expressed similar ideas. [3]

  Super-Elasticity and Regenerative Span

  An organism lives by constant transactions with the environment. As a result, stresses are set up in the parts or organs which have been aroused to carry out the transaction. The excited part may tend to 'get out of control', i.e. to assert its autonomy against the restraints imposed on it; it may tend to act to the detriment of the whole. In a 'normal' environment, these tensions between part and whole are of a transitory nature, and equilibrium is restored with the completion of the transaction. Under paranormal conditions -- traumatic challenges -- this is not the case, and only what one might call 'adaptations of the second order' can restore the balance. The animal's capacity to recover from such traumatic challenges is its regenerative potential.

  A stable, monotonous environment tends to produce stereotyped and automatized reaction-patterns. A variable environment calls for flexible strategies, for behavioural matrices with sufficient degrees of freedom to cope with the changing conditions. Paranormal challenges call for a kind of super-flexibility, for adaptations of a second order which enable the animal to carry out major reorganizations on several levels of its structural or functional hierarchies. The range of this ability constitutes the animal's 'regenerative span'.

  The regenerative span of a species thus provides it with an additional safety device in the service of survival, which enters into action when the limits of dynamic equilibrium are exceeded -- as the shock-absorbers of a motor-car take over when the range of elasticity of the suspension springs is exceeded. But it is more than a safety device. Regeneration has been described as 'one of the more spectacular pieces of magic in the repertoire of living organisms'. [4] That may be the reason why it is so difficult to find a satisfactory definition which would embrace the whole range of phenomena to which the term is applied.

  These include (a) the replication of entire individuals by asexual reproduction (fission and budding); (b) the reconstitution of a whole organism from its broken-up fragments, or from a single fragment. Sponges and hydra can be disintegrated into small clumps or even single cells by forcing them through a fine filter mesh, yet will reorganize themselves into normal, complete individuals. A single tentacle of a sweet-water polyp is capable of regenerating a complete individual; and transverse slices of a flatworm, taken from any part of its body, will regenerate the whole animal -- including, brains, eyes, genitalia, and other complex organs which the segment did not contain. (c) Among the higher animals, crustaceans are capable of regenerating single organs exposed to accidental damage (antennae, stalk-eyes, etc.); among vertebrates, salamanders and newts are capable of regenerating limbs, eyes, tails, and some inner organs (lungs and gonads); the process in these cases follows closely the processes of embryonic development. (d) Equipped with high regenerative powers, some animals practise autotomy -- the self-amputation of an exposed structure in the grasp of an enemy, which is subsequently replaced. Lizards let go of their tails; crabs, insects, and spiders of their legs or antennae, starfish cast off an arm. Self-amputation is facilitated by a 'breaking plane' of weakened structure somewhere near the base of the expendable appendage -- rather like the perforations between stamps. (e) Among mammalia, regeneration is generally limited to the repair of damaged bone, muscle, skin, and peripheral nerves. (f) Lastly, the term 'physiological regeneration' is used for the routine replacement of tissues used up by ordinary wear and tear.

  Thus regeneration appears to serve two different functions: on the one hand normal, asexual reproduction, on the other, the restoration of organs and structures lost by accidental mutilation or by wear and tear. But the two functions are in fact continuous; they shade into each other, and are often undistinguishable. If a flatworm sponttaneously sheds its tail, then grows a new tail and the shed tail grows a new head, this is called asexual reproduction; if it is sliced into two in the laboratory it is called regeneration; and the same goes for budding, which is the natural way of reproduction of some marine coelenterates, but can be artificially induced by laceration of the body wall. The 'regenerative field' in the salamander's amputated leg-stump obeys the same type of code as the morphogenetic field of embryonic primordia. And vice versa: the development of twins or triplets following accidental fragmentation might as well be called a regenerative process. Hence, ontogenesis may be described as the regeneration of a complete individual from a fragment specially set aside for that purpose. But this 'setting apart' of undifferentiated embryonic cells which 'specialize in non-specialization' occurs in regenerative processes too -- for instance, in annelids, hydra, and flatworms, which store 'reserve cells' or 'regeneration cells' in various parts of their bodies and mobilize them when the need arises. Sexual reproduction thus appears merely as an added twist to asexual regeneration -- though a twist with momentous consequences. Instead of replicating a single genetic code ad infinitum, the bisociation of two genetic codes is the basic model of the creative act.

  Although closely related species on the same level of the evolutionary hierarchy may differ widely in their regenerative power, it is nevertheless true that, in a general way, this power decreases as we proceed from lower to higher organisms. The essence of organic regeneration is a release of genetic cell potentials which are normally inhibited in adult tissue.

  Physiological Isolation

  These genetic potentials are the residues of the cell's erstwhile totipotentiality before differentiation set in -- its original power to create a whole new organism. Some of that power is reactivated when the regeneration tissue -- the part designed to replace the lost organ or limb -- is released from the controls which under normal conditions keep it under restraint. For this partial or total secession of the part from the whole, C. M. Child coined the useful concept of 'physiological isolation'. [5]

&nbs
p; Physiological isolation may be regarded as a drastic form of disequilibrium between the part and the whole. Its consequences may be beneficial or deleterious. Child distinguishes four causes for it. (a) Growth of the whole beyond a critical limit may make it ungovernable so that parts of it find themselves outside the range of central dominance and control. This may lead in lower organisms to reproduction by fission or budding: the isolated part is either shed (as, for instance, the planarian's tail) to form a new organism, or it may de-differentiate and reintegrate into a complete organism by budding. (b) Decline of the organism's powers of control (through senescence, metabolic or hormonal disorders) may, in combination with other causes, lead to a pathological regression of cells and tissues with untramelled proliferation and without reintegration, resulting in malignant growths. (c) Partial obstruction or total blockage of (nervous and chemical) communications, and (d) persistent local excitation beyond a critical limit, may release the part from its normal controls and activate, for better or worse, its latent potentialities.