Obedience to Authority

by Stanley Milgram

  The upper left-hand corner of the generator was labeled SHOCK GENERATOR, TYPE ZLB, DYSON INSTRUMENT COMPANY, WALTHAM, MASS. OUTPUT 15 VOLTS-450 VOLTS.

  Sample Shock

  Each subject was given a sample shock from the shock generator, prior to beginning his run as teacher. This shock was always 45 volts and was applied by pressing the third switch of the generator. The shock was applied to the wrist of the subject, and the jolt strengthened the subject’s belief in the authenticity of the generator.

  Shock Instructions

  The subject was told to administer a shock to the learner each time he gave a wrong response. Moreover—and this is the key command—the subject was instructed to “move one level higher on the shock generator each time the learner gives a wrong answer.” He was also instructed to announce the voltage level before administering a shock. This served to continually remind the subjects of the increasing intensity of shocks administered to the learner.1 If the subject reached the 30th shock level (450 volts), he was instructed to continue the procedure using this maximum voltage. After two further trials, the experimenter called a halt to the experiment.2

  Experimenter Feedback

  At various points in the experiment the subject would turn to the experimenter 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 prods were made in sequence: Only if Prod 1 had been unsuccessful could Prod 2 be used. If the subject refused to obey the experimenter after Prod 4, the experiment was terminated. The experimenter’s tone of voice was at all times firm, but not impolite. The sequence was begun anew on each occasion that the subject balked or showed reluctance to follow orders.

  Special Prods

  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.)

  Feedback from the Victim

  In all conditions the learner gave a predetermined set of responses to the word-pair test, based on a schedule of approximately three wrong answers to one correct answer.

  The victim’s responses were a key element in the situation and needed to be worked out carefully. In pilot studies, at first no vocal feedback from the victim was used. It was thought that the verbal and voltage designations on the control panel would create sufficient pressure to curtail the subject’s obedience. However, this proved not to be the case. In the absence of protests from the learner, virtually every subject in the pilot study, once commanded, went blithely to the end of the board, seemingly in-different to the verbal designations. This deprived us of an adequate basis for measuring obedient tendencies. A force had to be introduced that would strengthen the subject’s resistance to the experimenter’s commands, and reveal individual differences in terms of a distribution of breakoff points.

  This force took the form of protests from the victim. Initially, mild protests were used, but these proved inadequate. Subsequently, more vehement protests were inserted into the experimental procedure. To our consternation, even the strongest protests from the victim did not prevent many subjects from administering the harshest punishment ordered by the experimenter. But the protests did lower the mean of the maximum shocks somewhat and created some spread in the subject’s performance; therefore, the victim’s cries were recorded on tape and incorporated into the regular experimental procedure.

  The situation did more than highlight the technical difficulties of finding a workable experimental procedure: it indicated that subjects would obey authority to a greater extent than we had supposed. It also pointed to the importance of feedback from the victim in controlling the subject’s behavior.

  Each vocal response of the victim was coordinated to a particular voltage level on the shock generator. It is difficult to convey on the printed page the full tenor of the victim’s responses, for we have no adequate notation for vocal intensity, timing, and general qualities of delivery. In general, however, the victim indicated no discomfort until the 75-volt shock was administered, at which time there was a little grunt. Similar reactions followed the 90- and 105-volt shocks, and at 120 volts the victim shouted to the experimenter that the shocks were becoming painful. Painful groans were heard on administration of the 135-volt shock, and at 150 volts the victim cried out, “Experimenter, get me out of here! I won’t be in the experiment any more! I refuse to go on!” Cries of this type continue with generally rising intensity, so that at 180 volts the victim cried out, “I can’t stand the pain,” and by 270 volts his response to the shock was definitely an agonized scream. Throughout, from 150 volts on, he insisted that he be let out of the experiment. At 300 volts the victim shouted in desperation that he would no longer provide answers to the memory test.

  At this juncture, it was found, subjects would usually turn to the experimenter for guidance. The experimenter would instruct the subject to treat the absence of a response as a wrong answer, and to shock the subject according to the usual schedule. He advised the subjects to allow 5 to 10 seconds before considering no response as a wrong answer, and to increase the shock level one step each time the learner failed to respond correctly. At 315 volts, after a violent scream, the victim reaffirmed vehemently that he was no longer a participant. He provided no answers, but shrieked in agony whenever a shock was administered. After 330 volts he was not heard from, nor did his answers reappear on the four-way signal box.

  Measures

  The main measure for any subject is the maximum shock he administers before he refuses to go any further. In principle this may vary from 0 (for a subject who refuses to administer even the first shock) to 30 (for a subject who administers the highest shock on the generator).

  Interview and Debriefing

  An important aspect of the procedure occurred at the end of the experimental session. A careful postexperimental treatment was administered to all subjects. The exact content of the session varied from condition to condition and with increasing experience on our part. At the very least every subject was told that the victim had not received dangerous electric shocks. Each subject had a friendly reconciliation with the unharmed victim and an extended discussion with the experimenter. The experiment was explained to defiant subjects in a way that supported their decision to disobey the experimenter. Obedient subjects were assured that their behavior was entirely normal and that their feelings of conflict or tension were shared by other participants. Subjects were told that they would receive a comprehensive report at the conclusion of the experimental series. In some instances, additional detailed and lengthy discussions of the experiment were also carried out with individual subjects.

  When the experimental series was complete, subjects received a written report which presented details of the experimental procedure and results. Again, their own part in the experiments was treated in a dignified way and their behavior in the experiment respected. All subjects received a follow-up questionnaire regarding their participation in the research, which again allowed expression of thoughts and feelings about their behavior.

  Recapitulation

  In this situation the subject must resolve a conflict between two mutually incompatible demands from the social field. He may continue to follow the orders of the experimenter and shock the learner with increasin
g severity, or he may refuse to follow the orders of the experimenter and heed the learner’s pleas. The experimenter’s authority operates not in a free field but against ever-mounting countervailing pressures from the person being punished.

  Shock generator.

  Victim is strapped into chair.

  Subject receives sample shock.

  Subject breaks off experiment.

  Fig. 3

  This laboratory situation gives us a framework in which to study the subject’s reactions to the principal conflict of the experiment. Again, this conflict is between the experimenter’s demands that he continue to administer the electric shock and the learner’s demands, which become increasingly insistent, that the experiment be stopped. The crux of the study is to vary systematically the factors believed to alter the degree of obedience to the experimental commands and to learn under what condition submission to authority is most probable and under what conditions defiance is brought to the fore.

  What the experimental situation does is to condense the elements present when obedience occurs in the larger world such that the essential ingredients are brought into a relatively narrow arena where they are open to scientific scrutiny. The power of the situation derives from the fact that we are able to see, and the subject can experience, the concurrent operation of antagonistic forces that in the larger world rarely impinge on him at the identical instant.

  CHAPTER

  3

  Expected Behavior

  Too often, the value of a work in social science is played down by asserting the self-evident character of the findings. But rarely do we have clear information on exactly how people expect behavior to unfold in a given set of circumstances. If we acquire this information, we will be able to compare it with the actual outcome of the study. It will provide us with a benchmark from which to see how much or little we learn through the experiment. Further, should there be a disparity between what people expect and what actually occurs, we are left with the interesting problem of accounting for the gap. For the expectations then come to have the character of an illusion, and we must ask whether such an illusion is a chance expression of ignorance or performs some definite function in social life.

  The procedure for getting at such expectations is straightforward. In each case the respondents consist of an audience that has come to hear a lecture on the topic of obedience to authority. The experiment is described in detail without, however, disclosing the results in any way. The audience is provided with a schematic diagram of the shock generator, showing verbal and voltage designations. Each respondent is asked to reflect on the experiment, then privately to record how he himself would perform in it. Predictions were made by three groups: psychiatrists, college students, and an audience of middle-class adults of varied occupations.

  Fig. 4. Diagram of control panel.

  Table 1 summarizes the breakoff points predicted by these groups.

  Each one of the 110 respondents sees himself disobeying the experimenter at some point in the command series, and typical explanations offered to explain their disobedience are:

  “. . . I myself am scared stiff of shocks and I couldn’t give shocks to people that were protesting because I was causing the pain. I couldn’t deliberately hurt a perfect stranger.” (Maximum: 90 volts)

  “I do not think any experiment is worth inflicting strong shock on another human being. The fact that the subject responded as he did indicates to me that he was suffering from severe discomfort. I could not be the one to inflict this pain on him. The moment the subject insists he wants to leave the experiment, I respect his judgment.” (Maximum: 135 volts)

  “. . . I can’t stand to see people suffer. If the learner wanted to get out, I would free him so as not to make him suffer pain.” (Maximum: 150 volts)

  “. . . since the subject volunteered, and I volunteered, and since the experimenter has no physical authority over me, the subject should be released when he is at such a point that he asks to be released, and I would release him.” (Maximum: 150 volts)

  “The reason I would break off is because the person would scream with pain. Knowing how I would feel if I even scraped my knee, I would then feel compassion for the learner.” (Maximum: 180 volts)

  Table 1.Individuals Predict Their Own Breakoff Points

  These subjects see their reactions flowing from empathy, compassion, and a sense of justice. They enunciate a conception of what is desirable and assume that action follows accordingly. But they show little insight into the web of forces that operate in a real social situation.

  Fig. 5. Psychiatrists’ predictions of behavior in Voice-Feedback Experiment.

  Perhaps the question posed to them was unfair. People like to see themselves in a favorable light. So we asked also a somewhat different question to eliminate the bias induced by vanity. We asked them to predict how other people would perform. (And more specifically, we requested that they plot the distribution of breakoff points of one hundred Americans of diverse ages and occupations.) Psychiatrists, graduate students and faculty in the behavioral sciences, college sophomores, and middle-class adults responded to the question, and there is remarkable similarity in the predictions of the several groups. They predict that virtually all subjects will refuse to obey the experimenter; only a pathological fringe, not exceeding one or two per cent, was expected to proceed to the end of the shockboard. The psychiatrists’ predictions are shown in detail in Figure 5. They predicted that most subjects would not go beyond the 10th shock level (150 volts, when the victim makes his first explicit demand to be freed); about 4 percent would reach the 20th shock level, and about one subject in a thousand would administer the highest shock on the board.

  What are the assumptions that underlie these predictions? First, that people are by and large decent and do not readily hurt the innocent. Second, that unless coerced by physical force or threat, the individual is preeminently the source of his own behavior. A person acts in a particular way because he has decided to do so. Action takes place in a physical-social setting, but this is merely the stage for its occurrence. The behavior itself flows from an inner core of the person; within the core personal values are weighed, gratifications assessed, and resulting decisions are translated into action.

  Most people start with presuppositions of this sort when asked to think about the obedience experiment. They focus on the character of the autonomous individual rather than on the situation in which he finds himself. With this view, they are likely to expect few subjects to go along with the experimenter’s orders.

  CHAPTER

  4

  Closeness of the Victim

  Now let us look at the actual performance of subjects in the experiment. To begin, we shall consider the results of the Remote-Feedback variation (Experiment 1). It differs slightly from the usual situation in that no vocal complaint is heard from the victim. He is placed in another room where he cannot be seen by the subject, nor can his voice be heard; his answers flash silently on the signal box. However, at 300 volts the laboratory walls resound as he pounds in protest. After 315 volts, no further answers appear, and the pounding ceases.

  It is instructive to reprint the subjects’ remarks at the point of defiance, as transcribed from the tape recordings:

  “I think he’s trying to communicate, he’s knocking. . . . Well it’s not fair to shock the guy . . . these are terrific volts. I don’t think this is very humane. . . . Oh, I can’t go on with this; no, this isn’t right. It’s a hell of an experiment. The guy is suffering in there. No, I don’t want to go on. This is crazy.” (Subject refuses to administer more shocks.)

  “He’s banging in there. I’m gonna chicken out. I’d like to continue, but I can’t do that to a man. . . . I’m sorry I can’t do that to a man. I’ll hurt his heart. You take your check. . . . No really, I couldn’t do it.”

  Of the 40 subjects, 26 obeyed the orders of the experimenter to the end, proceeding to punish the victim until they reached the most potent shock availa
ble on the generator. After the 450-volt shock was administered three times, the experimenter called a halt to the session.

  Subjects were frequently in an agitated state. Sometimes, verbal protest was at a minimum, and the subject simply got up from his chair in front of the shock generator, indicating that he wished to leave the laboratory. Some obedient subjects expressed reluctance to administer shocks beyond the 300-volt level and displayed tensions similar to those who defied the experimenter.

  After the maximum shocks had been delivered, and the experimenter called a halt to the proceedings, many obedient subjects heaved sighs of relief, mopped their brows, rubbed their fingers over their eyes, or nervously fumbled cigarettes. Some shook their heads, apparently in regret. Some subjects had remained calm throughout the experiment and displayed only minimal signs of tension from beginning to end.

  Bringing the Victim Closer

  An experiment differs from a demonstration in that in an experiment, once an effect has been observed, it becomes possible to alter systematically the conditions under which it is produced, and in this way to learn the relevant causes.

  What we have seen thus far applies only to a situation in which the victim is out of sight and unable to communicate with his own voice. The recipient of the punishment is thus remote, nor does he indicate his wishes very clearly. There is pounding on the wall, but this has an inherently ambiguous meaning; possibly, some subjects did not interpret this pounding as evidence of the victim’s distress. The resulting obedience may be attributable to this. Perhaps there will be no obedience when the victim’s suffering is more clearly communicated; when the victim is given a sense of presence, and he is seen, heard, and felt.