The Act of Creation

by Arthur Koestler

  * * *

  E. Self-Pity. A little boy is beaten up by a gang of bullies. For a while he tries to fight back, to hit, scratch, and kick, but his tormentors immobilize him, and at last he begins to cry in 'impotent rage'.

  But the expression is misleading. Anybody who has watched children fight knows that weeping will start only after the victim has given up struggling and wriggling and accepted defeat. After a while new outbursts of rage may renew the struggle, but, each time this happens, weeping is interrupted. It is not an expression of rage (although the two may overlap) but an expression of helplessness after rage has been exhausted and a feeling of being abandoned has set in -- a yearning for love, sympathy, consolation. In other words, the tears once more signify a frustration of the participatory emotions; and if no sympathy is forthcoming, self-pity will provide a substitute -- a mild dissociation of the personality, in which the self is experienced almost as an alien object of loving commiseration.

  Similar considerations apply to so-called 'crying in pain'. In states of violent physical pain, as in acute states of rage, the organism is fully occupied coping with the emergency and has no time for tears.

  'Great pain', wrote Darwin, 'urges all animals, and has urged them during endless generations, to make the most violent and diversified efforts to escape from the cause of sufferings. Even when a limb or other separate part of the body is hurt, we often see a tendency to shake it as if to shake off the cause, though this obviously be impossible. Thus a habit of exerting with the utmost force all the muscles will have been established whenever great suffering is experienced.' [3]

  Cannon has shown that the Bodily Changes in Pain, Hunger, Fear, and Rage (the title of his classic work) all follow the same basic pattern, that they are emergency responses of the sympathico-adrenal system. Violent pain seems to be experienced by the unconscious mind as an aggression, whether it is inflicted by an outside agent or not. When the aggressor is a tooth or a cramp in the stomach we are apt to say 'it hurts', as if the offending organ were not part of oneself, and we try to shake the aggressor off, as animals do, by writhing, or pressing against it. Only when the pain has abated to a tolerably steady, 'dull' level do we accept it as part of ourselves -- we 'have' a headache or 'are' under the weather -- at the same time admitting that nothing can be done about it; writhing and struggling cease in the admission of defeat, as in the case of the child in the grip of its tormentors. 'Weeping in pain' starts only when the specific pain-behaviour stops, as 'weeping with rage' starts when rage-behaviour stops, and for precisely the same reasons: it is an abandoning of defences, an expression of helplessness, a craving for sympathy, and -- if accompanied by vocal cries -- an appeal for help.

  Another misconception is that children 'cry with fear', if 'crying' is used as a synonym for weeping. A child may cry out, in the literal sense, when suddenly frightened; it may run away, and if it cannot, strain away from the threatening apparition, lift his hands in protection, and distort his face into a mask of terror. Once more, the tears will come only after the acute fright and the specific strained fright-reactions have ceased; they do not mean 'I am frightened' but 'I was so frightened, and maybe still am a little, and now I want to be comforted.'

  Consider what happens when a little boy, running along a gravel path, suddenly stumbles and falls. The fright-reaction consists in the protective outstretching of hands, and related muscle-reflexes. Once the contact with the earth is made and the first shock overcome, the acute scare ebbs away, the muscles relax in surrender, the facial expression changes from fear to the sympathy-begging grimace of incipient weeping. If there is no witness to the drama, self-pity will again provide the overflow. If it is witnessed by the mother, who makes a fuss and betrays her anxiety, this will increase the child's craving for tenderness and its tears will ask for more. If, on the other hand, she gently but firmly debunks the drama, then, after a moment of puzzlement, the child may break into rather hesitant laughter -- the residue of the scare, and even the slight pain, are denied by reason and worked off, while at the same time the sympathy-craving emotions are nipped in the bud by the mother's matter-of-fact attitude.

  Lastly, 'crying in hunger'. A baby never weeps from hunger -- it cries to signal hunger. The proof is that crying instantaneously stops when the bottle or breast is offered, before hunger can have ceased; furthermore, once the child is weaned from breast and bottle, hunger ceases to be expressed by crying or weeping. [4]

  Needless to say, when a baby cries to attract attention, to signal that it is hungry or in distress, if often breaks into tears at the same time. Yet in such situations we say 'the baby is crying', not 'the baby is weeping', because the essence of the performance is the vocal protest or appeal for help; the shedding of tears is merely an accompaniment. The baby's bawling, kicking, and tossing is a typical and impressive emergency-reaction in 'pain, hunger, fear, and rage' of a dramatically self-asserting kind. The simultaneous overflow of the tear-glands may be 'genuine' weeping -- longing for affection and tenderness -- as an accompaniment to the bawling; it may also be due to physiological causes. Watering of the eyes can be induced as a purely physiological defence reflex against the intrusion of a foreign body -- a piece of grit or the molecules which carry the smell of onions. (Lachrymation caused by such local irritation is, by the way, unilateral -- it occurs initially in the affected eye only). [5] It can also be caused by coughing, sneezing, vomiting, and after prolonged fits of laughter. The physiological mechanism is still somewhat obscure, except that all these violent exertions affect the mucous membranes of the nose and throat, and tend to dry them; the lachrymal glands may have the function of restoring lubrication through tears entering the nose. [6] When one sees a baby cry its head off with dry eyes until it gets hoarse, one intuitively feels that tears would be a relief -- both psychologically and physiologically. The same applies to adults in situations of extreme distress.

  Home they brought the warrior dead; She nor swooned nor uttered cry. All her maids, watching said, 'She must weep or she will die'. (Tennyson, The Princess)

  Lastly, weeping may start in the child as a genuine, spontaneous overflow-reflex, but once the power of tears has been consciously or unconsciously recognized, the flow may be initiated automatically, or even voluntarily, as a weapon more subtle and more effective than mere cries of complaint or protest.* 'We seem to acquire specific visceral habits just as we pick up characteristic verbal and manual habits,' Kling has remarked, [7] and we ought to include in 'visceral habits' the exercise of the lachrymal glands. Weeping may be recruited into the service of hysteria, emotional blackmail, and even courtly behaviour (as a proof of sensibility less strenuous than swooning); it may be associated with convulsions, shrieks, and agitated display; but its true character is manifested by the person who weeps alone -- helpless in her surrender to an emotion which, by its nature, can find no other outlet, whether it is caused by the thunder of a church organ, or the fall of a sparrow.

  NOTES

  To p. 271. So scant are the references of any significance to the subject in the technical literature, that I thought it would be useful to future students to list what I could find under a separate heading at the end of the bibliography. My indebtedness to those who helped in this is acknowledged in the Preface.

  To p. 273. Romain Rolland describing the character of religious experience in a letter to Freud -- who regretfully professed never to have felt anything of the sort.

  To p. 274. 'The characteristic anatomical organization of the parasympathetic is correlated with absence of unitary action in this system. It is not surprising therefore that the adrenal medulla . . . has no counterpart in the parasympathetic system, and that no parasympathicomimetic hormone capable of acting extensively upon organs innervated by this system is liberated in the body.' (Macleod, ed. Bard, 1941 ed.) ' . . . In contrast to the sympathomimetic hormones, the vagus substance is rapidly destroyed, and therefore produces very localized response. These effects are in line with the general beh
aviour of the sympathetic and parasympathetic systems of nerves.' (White and Smithwick, 1941, 2nd. ed.)

  'All the viscera can be influenced simultaneously in one direction or the other by varying, up or down, the . . . tonic activity of the sympathetic division. And any special viscus can be separately influenced . . . by varying . . . the tonic activity of the special nerve of the opposed cranial or sacral [parasympathetic] division. . . . The sympathetic is like the loud and soft pedals, modulating all the notes together; the cranial and sacral [parasympathetic] innervations are like the separate keys.' (Cannon, 1929, 2nd. ed.)

  In the years since this has been written, the significance for psychology of the anatomical and physiological contrast between the two branches of the autonomic nervous system has become more evident, to the extent that 'rage is called the most adrenergic, and love the most cholinergic reaction' (Cobb, 1950). A further correspondence between patterns of emotive behaviour and modes of interaction between the two branches of the autonomic nervous system emerged when it was shown that the vagoinsulin system may act, in different circumstances, as an inhibitory or a catalytic agent in the glucose-utilization process and may also produce overcompensatory after effects (Gellhorn, 1943, and 1957). Hebb (1949) suggested that a distinction should be made between two categories of emotions, 'those in which the tendency is to maintain or increase the original stimulating conditions (pleasurable or integrative emotions)' and 'those in which the tendency is to abolish or decrease the stimulus (rage, fear, disgust)'. Whereas the latter have a disruptive effect on cortical behaviour, the former have not. A few years later, Olds (1959 and 1960) and others demonstrated the existence of 'positive' and 'negative' emotive systems by electric stimulation, and further showed that they were activated respectively by the parasympathetic and sympathetic centres in the hypothalamus.

  These hints all seem to point in the same direction, but in fairness to the general reader I ought to point out that, while there is ample experimental proof that the hunger-rage-fear emotions are mediated by the sympathico-adrenal division, there is no direct evidence for the symmetrical correlation proposed here. Such proof can be forthcoming only when emotions outside the hunger-rage-fear class will be recognized as a worthwhile object of study by experimental psychology -- which at present is not the case.

  To p. 282. A psychoanalyst friend of mine, after reading the manuscript of the preceding section, objected that his patients frequently weep during the analytical hour 'in anger and frustration'. But he agreed that anger alone would not have produced the tears, and that the frustration was due, metaphorically speaking, to the analyst's refusal 'to give the patient the breast and sing a lullaby'.

  XIII

  PARTNESS AND WHOLENESS

  Stepchildren of Psychology

  The self-transcending emotions* are the stepchildren of contemporary psychology. One of the reasons is perhaps that they do not tend towards observable muscular activity but towards quietude; grief, longing, worship, raptness, aesthetic pleasure are emotions consummated not in overt but in internalized, visceral behaviour, with weeping as its extreme manifestation. But even the shedding of tears is not so much an activity but rather a 'passivity'.

  The word 'emotion' is derived from 'motion'; and an emotion which tends to calm down motion seems to be a contradiction in terms. Yet the aesthetic or religious experiences which we call 'moving' are precisely those which induce passive contemplation, silent enjoyment. When the experimental psychologist talks of 'emotive behaviour', however, he nearly always refers to rage, fear, sex, and hunger, whereas emotions which do not beget overt activity are slurred over as 'moods' or sentiments -- with the implication that they are a suspect category of pseudo-emotions unworthy of the scientist's attention. This is probably a hangover of the great ideological currents of the nineteenth. century stressing the biological struggle for existence, the survival of the fittest, the acquisitive and competitive aspects of social behaviour. The ambiance of this 'Darwinistic psychology' is reflected in passages like the following, from Crile's The Origin and Nature of the Emotions, published in 1915:

  When the business man is conducting a struggle for existence against his rivals, and when the contest is at its height, he may clench his fists, pound the table, perhaps show his teeth, and exhibit every expression of physical combat. Fixing the jaw and showing the teeth in anger merely emphasize the remarkable tenacity of philogeny . . .

  It must be admitted, though, that the social climate of the nineteenth century did not favour the contemplative life, nor the arousal of genuine self-transcending emotions. The Victorian versions of religion, patriotism, and love were so thoroughly impregnated with prudery and hypocrisy that the experimental psychologist, devoted to measuring sensory threshholds and muscle twitches could hardly be expected to take such attitudes seriously, and to put them on a par with the sex and hunger drives. Around the turn of the century, the so-called James-Lange theory of emotions emphasized the importance of visceral processes, but it was nevertheless taken for granted that the 'true' or 'major' emotions were characterized by impulses to muscular action -- mainly to hit, run, or rape. When Cannon showed that hunger, pain, rage, and fear were, so to speak, variations on a single theme, it was tacitly taken for granted that all emotions worthy of that name were of the active, adreno-toxic, hit-run-mate-devour kind. Laughter and tears, awe and wonder, religious and aesthetic feeling, the whole 'violet' side of the rainbow of emotions was left to the poets to worry about; the so-called behavioural sciences had no room for them. Hence the paucity of the literature on weeping for instance -- although it is certainly an observable behavioural phenomenon.

  The emotions of the neglected half of the spectrum are as real as rage and fear; that much we know for certain from everyday experience. The theory which I have proposed assumes that they form a class, characterized by certain shared basic features. These are partly negative: the absence of adreno-sympathetic excitation alone puts them in a category apart from the emergency responses. On the positive side, emotional states as different as mourning and aesthetic enchantment share the logic of the moist eye: they are passive, cathartic, dominated by parasympathetic reactions. From the psychological point of view, the self-asserting emotions, derived from emergency reactions, involve a narrowing of consciousness; the participatory emotions an expansion of consciousness by identificatory processes of various kinds.

  There exist, however, considerations of a more precise and at the same time, more general nature on which this theory of the emotions is based. These are discussed in Book Two, but I must briefly allude to them. In that wider context, the polarity between the self-asserting and participatory tendencies turns out to be merely a particular instance of a general phenomenon: namely, that every member of a living organism or social body has the dual attributes of 'wholeness' and 'partness'. It acts as an autonomous, self-governing whole on its own subordinate parts on lower levels of the organic or social hierarchy; but it is subservient to the co-ordinating centre on the next higher level. In other words it displays both self-assertive and participatory tendencies.

  The Concept of Hierarchy

  The word 'hierarchy' is used here in a special sense. It does not mean simply 'order of rank' (as in the 'pecking hierarchy' of the farmyard); it means a special type of organization (such as a military hierarchy) in which the overall control is centralized at the apex of a kind of genealogical tree, which branches out downward. At the first branching-out, the commanders of the land-, sea-, and air-forces would correspond to the coordinating centres of, say, the digestive, respiratory, and reproductive organ-systems; each of these is subdivided into units or organs on lower levels of the hierarchy with their own coordinating centres, C.O.s and N.C.O.s; the organs in turn are subdivided into organ-parts; and so the branching-process goes on down to the cellular level and beyond.

  But each sub-organization, regardless on what level, retains a certain amount of autonomy or self-government. Without this delegation of powers, the
organization could not function effectively: the supreme commander cannot deal with individual privates; he must transmit strategical orders through 'regulation channels', which at each level are translated into tactical and sub-tactical moves. In the same way, information on what is happening in the various fields of operation (the sensory input) is selectively filtered on each level before being transmitted to the higher echelons. A living organism or social body is not an aggregation of elementary parts or elementary processes; it is an integrated hierarchy of semi-autonomous sub-wholes, consisting of sub-sub-wholes, and so on. Thus the functional units on every level of the hierarchy are double-faced as it were: they act as wholes when facing downwards, as parts when facing upwards.

  On the upper limit of the organic hierarchy, we find the same double-aspect: the individual animal or man is a whole relative to the parts of his body, but a part relative to the social organization to which he belongs. All advanced forms of social organization are again hierarchic: the individual is part of the family, which is part of the clan, which is part of the tribe, etc.; but instead of 'part' we ought in each case to say 'sub-whole' to convey the semi-autonomous character and self-assertive tendency of each functional unit.