Janus: A Summing Up

by Arthur Koestler

  Nor does every group need a personal leader or 'father-figure' in whom authority is vested, as discussed under the previous heading. Religious and political movements need leaders to get under way; once established they still benefit from efficient leadership; but the primary need of a group, the factor which lends it cohesion as a social holon, is a credo, a shared system of beliefs, and the resulting code of behaviour. This may be represented by human authority, or by a symbol -- the totem or fetish which provides a mystic sense of union among the members of the tribe; by sacred icons as objects of worship; or the regimental flag to which soldiers in battle were supposed to hang on even at the price of their lives. The group-mind may be governed by the conviction that the group represents a Chosen Race whose ancestors made a special covenant with God; or a Master Race whose forebears were blond demi-gods; or whose Emperors were descended from the sun. Its credo may be based on the conviction that observance of certain rules and rites qualifies one for membership in a privileged elite in after-life; or that manual work qualifies one for membership in the elite class of history. Critical arguments have little impact on the group-mind, because identification with a group always involves a certain sacrifice of the critical faculties of the individuals which constitute it, and an enhancement of their emotional potential by a kind of group-resonance or positive feedback.

  Let me repeat that in the present theory the term 'group' is not confined to a crowd assembled in one place, but refers to any social holon, governed by a set code of rules (e.g., language, traditions, customs, beliefs, etc.) which defines its corporate identity, lends it cohesion and a 'social profile'. As an autonomous holon, it has its own pattern of functioning and is governed by its own code of conduct, which cannot be 'reduced' to the individual codes which govern the behaviour of its members when acting as autonomous individuals and not as parts of the group. The obvious example is the conscript who as an individual is forbidden to kill, as a disciplined member of his unit is in duty bound to do so.

  Thus it is essential to distinguish between the rules which govern individual behaviour and those which guide the behaviour of the group as a whole.*

  * In a paper on 'The Evolution of Systems of Rules of Conduct' Professor F. A. von Hayek defines as his aim 'to distinguish between the systems of rules of conduct which governs the behaviour of the individual members of a group (or of the elements of any order) and the order or pattern of actions which results from this for the group as a whole . . . That [they] are not the same thing should be obvious as soon as it is stated, although the two are in fact frequently confused.' [1]

  The group, then, is to be regarded as a quasi-autonomous holon, not simply as a sum of its individual parts; and its activities depend not only on the interactions of its parts, but also on the group's interactions, as a whole, with other social holons on a higher level of the hierarchy. These interactions will again reflect the polarity of the holon's self-assertive and integrative tendencies, oscillating between competition and/or cooperation with other groups. In a healthy social holarchy the two tendencies are in equilibrium; but when tensions arise, this or that social holon may tend to get over-stimulated and impose upon its rivals or usurp the role of the whole. History provides a never-ending list of such tensions, confrontations and conflicts.

  Several factors responsible for this chronic disequilibrium have already been mentioned in earlier pages -- such as the unique range of diversity in our species with regard to racial characteristics and national temperament, or the divisive effect of the multiplicity of languages -- which, in their ensemble, have always made the disruptive forces in mankind prevail over the cohesive forces on a local or global scale. An even more important cause of trouble is that the code of conduct of a social holon includes not only the rules which govern the behaviour of its members, but also moral precepts and imperatives with a claim to universal validity. These imperatives carry a high emotional charge, and the group-mind tends to react violently to any threat -- real or imaginary -- to its cherished beliefs.

  All that has been said points to the conclusion that in the group-mind the self-assertive tendencies are more dominant than on the level of the average individual; and that, by identifying himself with the group, the individual adopts a code of behaviour different from his personal code. The individual -- pace Lorenz -- is not a killer, the group is; and by identifying with it, the individual is transformed into a killer.

  We shall see in a moment that this paradox can be observed not only on the battlefield or among lynching mobs, but also in austere psychological laboratories. Its paradoxical nature derives from the fact that the act of identification with the group is a self-transcending act, yet it reinforces the self-assertive tendencies of the group. Identification with the group is an act of devotion, of loving submission to the interests of the community, a partial or total surrender of personal identity and of the self-assertive tendencies of the individual. In our terminology, he relinquishes his 'wholeness' in favour of his 'partness' in a larger whole on a higher level of the holarchy. He becomes to some extent depersonalized, i.e., unself-ish in more than one sense. He may become indifferent to danger; he feels impelled to perform altruistic, even heroic actions to the point of self-sacrifice, and at the same time to behave with ruthless cruelty towards the enemy -- real or imagined -- of the group. But his brutality is impersonal and unselfish; it is exercised in the interest, or supposed interest of the whole; he is prepared not only to kill, but also to die in its name. Thus the self-assertive behaviour of the group is based on the self-transcending behaviour of its members; or to put it simply, the egotism of the group feeds on the altruism of its members.

  The 'infernal dialectics' of this process is reflected on every level of the various social holarchies. Patriotism is the noble virtue of subordinating individual interests to the interests of the nation; yet it gives rise to chauvinism, the militant expression of those higher interests. Loyalty to the clan produces clannishness; esprit de corps blossoms into arrogant cliquishness; religious fervour into zealotry; the Sermon on the Mount into the Church militant.

  Let us now turn to the experimental confirmation of our theoretical schema which has recently been provided, in a rather surprising manner, by the psychological laboratories in Yale and other universities.

  3

  The series of highly original experiments, which I propose to describe in some detail, were started by Dr Stanley Milgram at the Psychology Department in Yale University, and repeated by various experimental laboratories in Germany, Italy, Australia and South Africa. The purpose of the experiments was to discover the limits of the average person's obedience to authority, when ordered to inflict severe pain on an innocent victim in the interests of a noble cause. Authority was represented by a figure of professional appearance in a laboratory coat; I shall call him the Prof. The noble cause was Education; more precisely, the experiment was purportedly designed to provide answers to the problem whether punishing the pupil for his mistakes had a positive effect on the learning process. It involved three people: the Prof, who was in charge of the proceedings; the learner or victim; and the experimental subject, who was asked by the Prof to act as teacher and to punish the learner each time he gave the wrong reply. Punishment was by electric shocks of growing severity, administered by the 'teacher' on the Prof's orders. The 'learner' or victim was strapped into a kind of electric chair, with an electrode attached to his wrist. The 'teacher' was seated in front of an impressive shock-generator which had a key-board of thirty switches, ranging from 15 volts to 450 volts (i.e., a 15 volt increment from one switch to the next). There were also verbal inscriptions on the machine ranging from

  SLIGHT SHOCK to INTENSE SHOCK to DANGER -- SEVERE SHOCK.

  In fact the whole gruesome set-up was based on make-believe. The 'victim' was an actor hired by the Prof. The shock-generator was a dummy. Only the 'teacher', at whom the experiment was aimed, believed in the reality of the shocks he was ordered to administer, and of the
shrieks of pain and cries for mercy uttered by the 'victim'.

  The 'teachers' -- i.e., the real subjects of the experiment -- were volunteers from all walks of life between the ages of twenty and fifty, who came to the Yale laboratory attracted by newspaper advertisements to participate in 'a scientific study of memory and learning' (they were paid a modest four dollars per hour). Typical subjects were postal clerks, high-school teachers, salesmen, engineers and manual labourers. Altogether, more than a thousand volunteers were tested in Yale alone.

  The basic procedure of the experiment was as follows. The 'pupil' was given to read a long list of paired words, e.g., blue box -- nice day -- wild duck -- etc. Then, in the 'examination' he was given one test-word, for instance, 'blue', with four alternative answers, e.g., ink, box, sky, lamp, and had to indicate which was the correct answer. The 'teacher' was instructed by the Prof to administer a shock each time the pupil gave a wrong response, and moreover 'to move one level higher on the shock-generator each time the learner gives the wrong answer.

  To make sure that the 'teacher' was aware of what he was doing, the actor who played the role of the victim uttered complaints which increased in stridency according to the voltage, from 'mild grunts' starting at 75 volts, in a crescendo, until at 150 volts the victim cried out 'Get me out of here! I won't be in the experiment any more! I refuse to go on.' (Remember that the 'teacher' believed that the victim too was a volunteer.) 'At 315 volts, after a violent scream, the victim reaffirmed vehemently that he was no longer a participant. He provided no answers, but screamed in agony whenever a shock was administered. After 330 volts he was not heard from. . . .' Yet the Prof instructed the subject to treat no answer as a wrong answer and to continue to increase the shock level according to schedule. After three shocks of 450 volts he called off the experiment.

  How many people, in an average population, do you think would obey the command to carry on with the task of torturing the victim to the limit of 450 volts? The answer seems to be a foregone conclusion: perhaps one in a thousand, a pathological sadist. Before starting his experiments, Milgram actually asked a group of psychiatrists to predict the outcome. 'With remarkable similarity they predicted that virtually all subjects would refuse to obey the experimenter.' The consensus of the thirty-nine psychiatrists who answered the questionnaire was that 'most subjects would not go beyond 150 volts (i.e., when the victim asks for the first time to be released). They expected that only 4% would reach 300 volts, and that only a pathological fringe of about one in a thousand would administer the highest shock on the board.' [2]

  In actual fact, over 60 per cent of the subjects at Yale continued to obey the Prof to the very end -- the 450 volt limit. When the experiment was repeated in Italy, South Africa and Australia, the percentage of obedient subjects was somewhat higher. In Munich it was 85 per cent.

  Before going any further, let me clarify a few points relating to the experimental set-up.

  First, the Prof had no power over his volunteer subjects comparable to that of an army officer or an office boss or even a school teacher. He had no power to punish the subject who refused to administer further shocks, nor did he have any financial or other incentives to offer. (It was understood that volunteers would only be employed on a single occasion.)

  How then did the Prof impose his authority on the 'teacher, and induce him to continue with his gruesome task? There was no bullying, nor any eloquent persuasion. The Prof's procedure was rigidly standardized:

  At various points in the experiment the subject would turn to the experimenter [the Prof] for advice on whether he should continue to administer shocks. Or he would indicate that he did not wish to go on. The experimenter responded with a sequence of 'prods', using as many as necessary to bring the subject into line. Prod 1: Please continue or Please go on. Prod 2: The experiment requires that you continue. Prod 3: It is absolutely essential that you continue. Prod 4: You have no other choice, you must go on. The experimenter's tone of voice was at all times firm, but not impolite. If the subject asked if the learner was liable to suffer permanent physical injury, the experimenter said: 'Although the shocks may be painful, there is no permanent tissue damage, so please go on.' (Followed by prods 2, 3 and 4, if necessary.) If the subject said that the learner did not want to go on, the experimenter replied: 'Whether the learner likes it or not, you must go on until he has learned all the word pairs correctly. So please go on.' (Followed by Prods 2, 3 and 4, if necessary.)3

  One could hardly call this technique brain-washing. And yet it worked on nearly two-thirds of all experimental subjects, regardless of country and of the method of soliciting volunteers. It worked even when the 'victim' complained of a heart condition and the maximum shocks seemed to constitute a danger to his life. That humane people are capable of committing inhuman acts when acting as members of an army or a fanatical mob has always been taken for granted. The importance of the experiments was that they revealed how little was needed to push theni across the psychic boundary which separates the behaviour of decent citizens from dehumanized SS guards. The fragility of that boundary -- which two-thirds of the subjects crossed -- came as an utter surprise even to psychiatrists, whose recorded predictions turned out to be totally -- though understandably -- wrong.

  A comfortable way to evade the uncomfortable problem with which these results confront us, is to put the blame on the repressed aggressive impulses of the subjects, for which the experiments provide a socially respectable outlet. This interpretation is in the traditional line of Freud's 'urge to destruction', or Lorenz's 'killer-instinct' -- a view which, as I have argued before, is contradicted by both the historical and psychological evidence. Milgram found an elegant method to refute this facile explanation, and to demonstrate that

  ... the act of shocking the victim does not stem from destructive urges but from the fact that the subjects have become integrated into a social structure and are unable to get out of it. Suppose the experimenter instructed the subject to drink a glass of water. Does this mean the subject is thirsty? Obviously not, for he is simply doing what he is told to do. It is the essence of obedience that the action carried out does not correspond to the motives of the actor but is initiated in the motive system of those higher up in the social hierarchy.'

  To prove his point, he carried out a further series of experiments in which the 'teacher' was told that he was free to inflict on the learner any shock level of his own choice on any of the trials --

  ... the highest levels on the generator, the lowest, any in between, or any combination of levels ... [5]

  Though given full opportunity to inflict pain on the learner, almost all subjects administered the lowest shocks on the control panel, the mean shock level being 54 volts. [Remember that the victim's first mild complaint came only at 7S volts.] But if destructive impulses were really pressing for release, and the subjects could justify their use of high shock levels in the cause of science, why did they not make the 'learners' suffer? There was little if any tendency in the subjects to do this. One or two, at most [out of 40 subjects]*, seemed to derive any satisfaction from shocking the learner. The levels were in no way comparable to that obtained when the subjects were ordered to shock the victim. There was an order-of-magnitude difference.'

  * The experimental series consisted of batches of 40 subjects of mixed ages and professions.

  In the original experiments, when the teacher acted on the Prof's orders, an average of 25 out of 40 subjects administered the maximum shock of 450 volts. In the free-choice experiment 38 out of 40 did not go beyond 150 volts (victim's first loud protest) and only two subjects went up to 325 and 450 respectively.

  To clinch the argument, Milgram quotes other experiments, carried out by his colleagues Buss and Berkowitz in a similar set-up.

  In typical experimental manipulations, they frustrated the subject to see whether he would administer higher shocks when angry. But the effect of these manipulations was minuscule compared with the levels obtained under ob
edience. That is to say, no matter what these experimenters did to anger, irritate or frustrate the subject, he would at most move up one or two shock levels, say from shock level 4 to level 6 [90 volts]. This represented a genuine increment in aggression. But there remained an order-of-magnitude difference in the variation introduced in his behaviour this way, and under conditions where he was taking orders. [7]

  The vast majority of the experimental subjects, far from deriving any pleasure from shocking the victim, showed various symptoms of emotional strain and distress. Some broke into a sweat, others pleaded with the Prof to stop, or protested that the experiment was cruel and stupid. Yet two-thirds nevertheless went on to the bitter end.

  What made them persist in a task that was obviously distasteful to them and in blatant contradiction to their individual standards of ethics? Milgram's analysis, apart from some differences in terminology, is on the same lines as the theoretical considerations set out in previous chapters. He recognizes the profound implications of the hierarchic concept*: to wit,

  that . . . when individuals enter a condition of hierarchic control, the mechanism which ordinarily regulates individual impulses is suppressed and ceded to the higher-level component . . . [8] The individuals who enter into such hierarchies are, of necessity, modified in their functioning . . . [9] This transformation corresponds precisely to the central dilemma of our experiment: how is it that a person who is usually decent and courteous acts with severity against another person within the experiment? . . . [10] The disappearance of a sense of responsibility is the most far-reaching consequence of submission to authority . . . [11] Most subjects in the experiment see their behaviour in a larger context that is benevolent and useful to society -- the pursuit of scientific truth. The psychological laboratory has a strong claim to legitimacy and evokes trust and confidence in those who perform there. An action such as shocking a victim, which in isolation appears evil, acquires a totally different meaning when placed in this setting . . . [12] Morality does not disappear, but acquires a radically different focus: the subordinate person feels shame and pride depending on how adequately he has performed the actions called for by authority. Language provides numerous terms to pinpoint this type of morality: loyalty, duty, discipline . . . [13] * I was gratified by the generous references in his book to the hierarchic mode proposed in The Ghost in the Machine.