Milgram’s Experimental View of Authority
The experimenter (E) orders the teacher (T), the subject of the experiment, to give what the latter believes are painful electric shocks to a learner (L), who is actually an actor and confederate. The subject is led to believe that for each wrong answer, the learner was receiving actual electric shocks, though in reality there were no such punishments. Being separated from the subject, the confederate set up a tape recorder integrated with the electro-shock generator, which played pre-recorded sounds for each shock level.
Obedience to Authority: An Experimental View. By Stanley Milgram. (New York: Harper Colophon Books, 1975). n the early 1960s Professor Stanley Milgram then of Yale IUniversity, conducted a series of experiments on the behavior of human beings whose purpose was to examine the phenomenon of obedience to authority. The results were startling, since they showed that human beings would inflict needless suffering on innocent fellows simply because an apparently legitimate authority commanded them to do so. These results explained according to Milgram and other observers how it was possible for Hitler to command the obedience of men to torture, kill, and destroy millions. The results also showed the consequent necessity for citizens to be far more careful about obeying the commands of their governments and, presumably, any other authorities or authority figures.
Milgram’s argument, therefore, utilized the results of a scientific experiment and scientific methodology and concepts generally to prove an explicitly political point about how men ought to regard authority. This combination of empirical research in the social
sciences and of political philosophizing lends interest to Milgram’s work as well as a certain degree of complexity. The present essay will explore both the scientific and political aspects, and how they interact. This will be done under three major headings: Exposition (I); Responses and Criticisms (II); Authority and Science (III).
Obedience to Authority is the record of a series of experiments carried out over four years, plus commentary and an attempt to explain the phenomena Milgram uncovered. The commentary and explanation are as important as the experiments themselves, or nearly so, because they reveal, often in ways Milgram does not intend, the expectation, mind-set and attitude that provide the assumptions and influence the conclusions of the experiments. That is, because the ex-
periments are supposed to tell us something to do about people and society, it is necessary to know what non-experimental values and assumptions Milgram brings to these experiments.
262 THE POLITICAL SCIENCE REVIEWER
The book itself is not very long, only 224 pages including two appendices, notes, references, and index. There are 15 chapters; the first three introducing the experiments, chapters four through nine describing the experiments themselves and the final six chapters containing explanation and commentary. One of the appendices is worthy of note because it deals with the ethical problems raised by the methodology of Milgram’s experiments with human subjects. There are also statistical tables, charts, photographs and diagrams
throughout the book but mainly in the experimental chapters. Despite these and the difficulty of the subject matter, Obedience to Authority is a well written, remarkably clear and sometimes compelling book.
Chapter 1, “The Dilemma of Obedience” states the rationale for Milgram’s series of experiments and gives a brief overall description of the experimental procedure. Revealingly, the very first paragraph
puts the problem of obedience in the context of Hitler’s destruction of the European Jews, for Milgram throughout Obedience to
Authority will mix straight description of scientific experiments with commentary that applies his research findings to what he calls “the dilemma of obedience.” “Obedience,” he says, “as a determinant of behavior, is of particular relevance to our time,” (1) because obedience to the orders of totalitarian regimes is what made the Nazi death camps as well as other monstrous acts of inhumanity and war possible in the twentieth century.
Milgram’s experiments were designed to take a closer look at this phenomenon of obedience to orders from authority in an experimental setting. In this design, volunteers are asked to take part in an experiment in which the use of pain as a teaching technique is to be studied. The role of the volunteer is to be a “teacher” who will administer electric shocks to the “learner” every time the learner answers a question incorrectly. The learner is first given several matched pairs of words by the teacher, and subsequently the first word of each pair is repeated to the learner. The learner is expected to repeat the corresponding word. When he gives an incorrect answer, the teacher then administers an electric shock by throwing a switch. After each incorrect answer, the intensity of the shock is increased, until at some point the learner complains. This is the crucial point of Milgram’s experiment, for the real purpose of the experiment is to see if the teacher will stop obeying, i.e. will break off the experiment when it seems he is inflicting real pain on the learner. In reality, the learner is an actor who is uncannily good at
MILGRAM AND AUTHORITY 263
simulating the outward expressions of someone in great pain. No electric shock in fact is being administered to the learner. The real subject of the experiment is the teacher who must make increasingly difficult choices of whether to break off the experiment in the face of the orders from the scientist who has ostensibly arranged the experiment on one hand, and the complaints, protests and finally screams and pleas of the learner on the other.
According to Milgram, “the results of the experiment are both surprising and dismaying” for most of the “teachers” in fact continued to administer the shocks to the “learner” despite the learner ‘ s protests and despite the teachers’ own obvious feelings of stress and their complaints to the authority figure conducting the experiment. Many subjects will obey the experimenter no matter how vehement the protest of the person being shocked, and no matter how much the victim pleads to be let out. (5) The fact that almost two-thirds of the teachers could be classified as “obedient subjects” strikes Milgram as remarkable, calling for comment and explanation.
The first point Milgram makes about the untoward results of his experiments is that there is no avoiding them. The experimental subjects, the “teachers,” were not sadists and were in fact a fairly representative sample of the population at large, i.e., ordinary people. Secondly, even though many of the teachers protested, and hence knew what they were doing was morally wrong, they continued to administer shocks. From this, Milgram concludes that the moral sense of the individual is easily overcome by social pressures and that ethical rules such as the Fifth Commandment are not an inherent part of human psychic structure. On the other hand, it is not true that the individual loses his moral sense when following orders that contradict his conscience. Rather, Milgram says, his moral concern shifts from his act itself to how well he does his job, wanting to live up the expectation of the authority figure. Further, he looks to the wider context of the purposes of society for legitimizing his actions. One unexpected result of the experiment is that the teachers blame the learner’s stupidity and stubborness for the punishment inflicted on him. Milgram concludes by making the point that evil acts commanded by social authority today are fragmented in such a way that no one person is ever fully responsible for them. Rather, a number of people each take a small part so that there is a long chain
of actions from the initial command to the completion of the evil act. This tendency which reduces responsibility for evil acts Milgram sees as the “most common characteristic of socially organized evil in modern society.” (11)
264 THE POLITICAL SCIENCE REVIEWER
In Chapter 2, “Method of Inquiry,” Milgram gives details of how the experiments were carried out, as well as providing an explanation of the experiments’ rationale, which was to simplify the com-
plex phenomena of obedience. To do this he matched the strength of obedience against a countervailing factor, in this case the moral
principle that one should not inflict suffering on a helpless person who is neither harmful nor a threat. Since the electric shocks which the teacher must administer are graded in 30 levels of increasing intensity, ranging from 15 to 450 volts, when and if the teacher breaks off the experiment, it will be at one of 30 discrete levels. “Behavior prior to this rupture is termed obedience. The point of the rupture is termed disobedience.” (14) Thus, the procedure allows for quantifiable measurement and for control of variables.
Among the details of the experimental procedure presented are how participants were obtained for the study, locale of the experiment (at the ” elegant Interaction Laboratory of Yale University ” ), the instructions given to the teacher on the pretext that this was an experiment about learning, and samples of the word pairs which the learner was supposed to learn. Great attention was paid to verisimilitude, especially regarding the process of administering the supposed shock. The shock generator had a series of 30 switches for inducing increasing levels of shock, with a red light and a buzzer sound every time a “shock” was administered. The teachers were each given a sample shock, applied to their wrist, and they were assured that even though the shocks “can be extemely painful, they cause no permanent tissue damage.” (19) The experimenter under whose authority the pretextual learning experiment was being run,
wore a laboratory technician’s coat, assumed a professional manner and used a preestablished series of verbal prods to encourage the teachers when they resisted shocking the victim. Special attention was also paid to the manner in which the learner answered the questions. He gave approximately three wrong answers for every correct one, and at the 75 volt level in the series began a pre-set series of responses, starting with grunts, then proceeding to complaints, then to demands to be released, then to refusal to cooperate, and finally “an agonized scream.” (23) At the conclusion of the experiment, each teacher was debriefed and told the learner had in fact not received any shocks. This was done in a supportive manner so as not to embarrass the teacher who was also sent a follow-up questionnaire and a report about the experiment in which he had been a subject.
MILGRAM AND AUTHORITY 265
Chapter 3, “Expected Behavior,” describes the results of a survey
in which members of three audiences, one consisting of
psychiatrists, another of college students, and a third of middle class adults, were asked to predict how the teachers would respond to the experiment. All three audiences predicted that almost none of the teachers would complete giving the series of shocks, and that only one or two percent, a pathological fringe, would give the maximum levels of shock. Milgram says this survey provided him with a bench mark to see how much we can actually learn from his experiment on obedience by contrasting actual with predicted behavior. He also elaborates on the assumptions that lay behind the predictions given on the surveys: first, that most people are decent; and second, that people act in certain ways because they themselves choose to do so,
i.e., their behavior “flows from an inner core of the person,” (38) apart from the physical or social setting.
Chapter 4, “Closeness of the Victim,” describes the initial set of actual experiments that Milgram conducted. There were four such experiments in which 160 subjects were tested to determine how the level of obedience varied according to the physical proximity of the learner to the teacher. The four variable conditions were: first, that the learner was hidden from the teacher in another room and communicated his pretextual distress by banging on the wall; second, that the learner was hidden but his voice was relayed by a microphone; third, that the learner was in the same room as the subject; fourth, that the teacher held the learner’s arm while administering the shocks. In general, these experiments showed that the closer the learner was to the teacher, the less obedient the teachers were, i.e., they broke off giving shocks earlier in the series and in greater proportion. Thus, 35% were disobedient-i.e. re-
fused to continue the shocks to the end-when the learner was hidden and could communicate only by banging the wall; 37.5% were disobedient when they could hear the learner’s voice, though not see him; 60 % were disobedient when the learner was in the room with them; 70% were disobedient when in physical contact with the learner.
Milgram offers six possible factors to explain why proximity, especially physical proximity where the teacher is in the same room with the learner, increases the level of disobedience, and asserts that any theoretical model of obedience will have to take this fact into account. (40)
There were two unexpected results from this initial set of ex-
266 THE POLITICAL SCIENCE REVIEWER
periments which continued to characterize the remainder as well. First, the experimental subjects were far more obedient than anyone predicted, either in the surveys mentioned in the previous chapter or among observers of the experiment itself. Second, the subjects exhibited a striking degree of emotional tension during the experiment, a fact which they confirmed when they were interviewed afterward. Milgram points out that this tension was the result of the
conflict between two opposing tendencies-obedience to a person in authority, and not harming an innocent person. He also points out that the degree of tension also indicates how real the situation is for the subject. “Normal subjects do not tremble and sweat unless they are implicated in a deep and genuinely felt predicament.” (8)
Chapter 5, “Individuals Confront Authority,” consists of descriptions of five of the subjects who acted as “teachers” in the first set of experiments described in Chapter 4. Milgram explains that the purpose of focusing on individuals is to provide a personal dimension to the experiment and to find clues to understanding the process of obedience. However, Milgram warns that since the subject is unaware of the forces that control him and of the variation in experimental conditions, the subject lacks a full understanding of the courses of his behavior. “A line must be drawn between listening carefully to what a subject says and’ mistaking it for the full story.” (7)
Each vignette consists of a brief personality sketch of the subject, along with excerpts from interviews and descriptions of the subject’s
responses during the experiment. Milgram also adds his own evaluations. Thus, he notes that the social worker giggled nervously when he pressed the switches, that the welder who pressed all the switches to the end indicated that he learned something but “he does not tell us what” (47), that the industrial engineer who broke off thinks that
psychology is more important than engineering (52), that the professor of “Old Testament liturgy” who broke off the experiment manifests “excessive fastidiousness” (48), and that the black drill
press operator who pressed all the switches expressed “total faith in the experimenter.” (50)
Chapter 6, “Further Variations and Control,” describes seven experiments in which Milgram sought to vary the experimental situation and explore the conceptual limits of his previous experiment on obedience. In the first of these experiments, two changes were introduced. First, besides the programmed responses which the learner would already give at various shock levels, he would also state he had a heart condition before the experiment began and then
MILGRAM AND AUTHORITY 267
claim his heart was bothering him at the 10th, 13th, and 22nd shock levels. Second, the experiment would move to a different laboratory setting, a more modest one with bare steampipes and a concrete floor instead of rugs. The learner was not in the room with the teacher, but gave his responses via a loudspeaker as in the second experiment in the first set. Despite these alterations there was no significant difference in results, 26 out of 40 pressing the shock board to the end for an obedience rate of 65%, compared with 25 out of 40 in the previous experiment. This set of conditions was used, with alterations, as the base line condition for subsequent experiments.
The second experiment in this group changed the personality styles of the experimenter who gave the orders to the teacher, and the learner whom the teacher shocked. A professional and aloof experimenter was replaced by an avuncular, soft man, and an affable learner was replaced by a man with a hard face “who looked as if he would do well in a scrap.” The results, says Milgram, “indicate that the change in personnel had little effect, (59) The table which summarizes the results of the experiment in Chapter 6 however shows that the percentage of the obedient subjects was 50% as opposed to
65% in the previous base line experiment. (60-61)
In the third experiment in this set, orders were given to the teacher by phone and the experimenter was not physically present. Obedience dropped sharply to 20.5%; further, several of the teachers administered lower shocks than were required.
The fourth variation was to use women as teachers only (thus far the experiment had been confined to male subjects). The results were virtually identical to the base line experiment, 65% obedience, although the women subjects manifested a greater level of tension.
The fifth variation introduced the notion of a contract, for the learner indicated before signing the standard consent forms that because of his heart condition, he must impose a condition on his compliance with the experimental procedure; namely, the experimenter must stop when the learner demanded it. There was some decline in the obedience level, but, writes Milgram, “it could easily represent a chance variation. ” (66) The summary table shows an obedience level of 40% compared to the base line experiment’s 65%.
The sixth variation was to move the experiment entirely out of the confines and sponsorship of Yale University which some of the teachers held in respect and even awe. A new laboratory was set up
268 THE POLITICAL SCIENCE REVIEWER
in the second story offices of a commercial building in a run-down shopping area of Bridgeport, and with the name of Yale University absent as sponsor. The level of obedience, according to Milgram, “although somewhat reduced, was not significantly lower.” (69) The summary table shows an obedience level of 47.5 % compared to the base line 65% .
The seventh variation was to allow the teacher to pick any shock level he wished rather than having to give increasingly more powerful shocks. Only 2.5% pressed the final, most powerful shock, the remaining 95.5 %o (38 out of 40 teachers) not pressing any shock level above the 10th level when the learner first vehemently protested. The result, says Milgram, invalidates the theory that autonomously generated aggression caused the subjects to press the maximum
Chapter 7, “Individuals Confront Authority II” like Chapter 5, consists of descriptions of five of the teachers who took part in the set of experiments described in the previous chapter. Again, Milgram offers evaluations of the responses of the various participants along with personality sketches and conversations that took place during the experiment and in subsequent interviews. Of the five vignettes offered in this chapter, four describe people who pressed all the switches, and his comments about them are fairly harsh. Typical of Milgram’s evaluation is this one describing the conversation with the unemployed man who pressed all the switches on the shock board: “The subject’s objections strike us as inordinately weak and inappropriate in view of the events in which he is immersed. He thinks he is killing someone, yet he used the language of the tea table.” (77)
Chapter 8, ” Role Permutations, ” describes six experiments in which Milgram attempted a more radical analysis of the elements which constituted the experimental situation. Up to now, the relationship and functions among the experimenter, the learner and the teacher were invariant. Thus, only the experimenter has ordered the teacher to shock the learners, not the learner himself, or someone else who is not a researcher. By altering this and other aspects of the fundamental relationship Milgram hoped to examine the roots of obedience as a form of social behavior.
In the first of these experiments, after the shock level at which the learner first protests, the experimenter calls off the experiment, but the learner demands that it continue. No teacher, however, administered any shocks after the experimenter demands they be stopped. From this, Milgram concludes that it is not the “substance
MILGRAM AND AUTHORITY
of the command,” i.e. the command itself, that is of decisive importance, but its ” source in authority ” (92), i.e. whether or not a person in authority issues the command.
In the second experiment, a fourth person enters the experimental situation, a layman who is an apparent volunteer like the teacher. In reality, he too is playing a role which is to run the experiment once the experimenter is called away on a pretext. No orders regarding shock levels are given by the experimenter before he leaves, and it is only due to the enthusiasm of the other layman that the experiment continues in the usual way. That is, he gives the orders, gives prods to the teacher and, in effect, runs the experiment. The result was a sharp drop in compliance, only 4 of 20 teachers pressing all the switches for an obedience rate of 20 % .
The third experiment was a variation of the second. In this one the new volunteer took over administering the shocks himself once
the teacher refused to do so. Out of 16 teachers tested in this experiment, 12 stopped and 4 continued the shocks under the orders of the new lay person. All of the 12 teachers who stopped the experiment either protested the continued administering of shocks or physically attempted to stop the experiment by interfering with the shock
generator or with the actions of the new volunteer himself.
In the fourth experiment, a pretext is found to reverse the roles of the learner and the experimenter so that the experimenter receives the shocks while the learner runs the experiment. The experimenter, however, protests at the 150 volt level just as the learner had in the previous experiments. The result was that none of the teachers tested administered shocks beyond the point at which the experimenter,
now in the role of learner, protested. Milgram notes that many teachers explained their disobedience on humanitarian grounds, not recognizing that they were “simply following the boss’s orders.” (104) Milgram also notes that this and the previous three experiments confirm the essential fact that the teachers are responding to the authority of the experimenter, not to the order itself. “It is not what subjects do but for whom they do it that counts.” (104)
In the fifth experiment two experimenters instead of one give contrary commands. At the point where the learner first protests, one directs that the experiment continue, the other directs it to stop. In all cases, the teachers stopped the experiment, almost always at this point of conflict of authority. Since conflict between authorities is far more effective in stopping the experiment than the pleas and screams of the victim, Milgram concludes that “action flows from the higher end of a social hierarchy to the lower.” (107) That is, the subject will respond to signals from persons on a higher level than his own, but not to those below it. Also, Milgram notes that at the point of conflict, teachers tried to determine which experimenter had the greater authority.
In the sixth experiment, there are again two experimenters, but no learner. Consequently, one of the experimenters volunteers to be the learner and protests at the usual points in the progression of shock intensity. The result is striking, because the rate of obedience (65%) is the same whether the learner is a scientist (as in this experiment) or a layman (as in the previous experiments). The fact that a person identified as an experimenter became the learner made no difference compared to the base line experiment.
Chapter 9, “Group Effects,” describes two experiments in which there are more than one teacher in order to discover the effect of conformity to peer pressure on the rate of obedience. Milgram begins the discussion by distinguishing conformity from obedience. While both conformity and obedience affect a person’s behavior, conformity is implied and voluntary, imitative of the actions of
peers, while obedience is explicit, following the orders of someone in authority.
In the first experiment, there are three teachers, only one of whom is a naive subject. One of the teachers who is in on the experiment sits at the shock board with the naive subject who has the job of pressing the switches. At the 150 volt level, when the learner gives his first strong protest, the non-naive teacher breaks off participation in the experiment, but stays in the room leaving the naive subject alone at the shock board. The result was that fully 90% of the naive teachers broke off the experiment before the end, a 10% rate
of obedience. In the second experiment, there are two teachers, one of them naive, who however, does not have the job of pressing the switches on the shock board. This instead is done by the second teacher who is in on the experiment. In this situation, only 3 of 40 subjects tested refused to participate to the end.
From these two experiments, Milgram concludes that “the mutual support provided by men for each other is the strongest bulwark we have against the excesses of authority” and that “any factor that will create distances between the subject and the victim, will . . . lessen disobedience.” (121)
Chapter 10, “Why Obedience? An Analysis,” attempts to explain why the experimental subjects obeyed the demands of the ex-
MILGRAM AND AUTHORITY
perimenter, knuckling under to authority and performing actions that were callous and severe. Milgram’s analysis is theoretical, of a high level of generality and is comprised of five parts.
Milgram starts with an analysis of the concept of hierarchy, pointing out that dominance structures are found throughout nature among all kinds of animals. Hierarchy is found among human beings as well, mediated by symbols rather than based on physical
strength. From an evolutionary point of view, hierarchical social organizations help species survive in coping with the physical environment, warding off attacks from other species, and by defining clearly the status of each member of the group so that internal conflict is minimized.
From this evolutionary viewpoint, Milgram next proceeds to a “cybernetic viewpoint” which he maintains will provide us with a model to alert us to the changes that logically must occur when hitherto autonomous entities are brought into a hierarchical structure. Automata living in isolation can be described by using a “homeostatic” model in which the automata comprise an open system requiring input from their environment to maintain their internal states. A lack or a need felt within the automata makes them do something to its environment to fill that lack or need, e.g. eating something when hungry. However, such automata when brought into a social organization need regulation if they are not to treat other automata as part of their environment, e.g. eat them. Hence, the need for ” an inhibitor that prevents automata from acting against each other.” (127) If such an inhibitor does not evolve, the species will perish. In human beings, this inhibitor is their conscience.
A more powerful form of social organization can be achieved by a hierarchical ordering in which there are subordinate elements controlled by superordinate ones. By combining many such hierarchical arrangements, we arrive at the typical pyramidal form of social organization. However, Milgram states, an inhibitor of conscience is not sufficient to allow control of one automaton by another, so the conscience itself must be secondary to the need to cede control to the superordinate element. Conscience, while necessary for autonomous functioning within a social context generally, cannot override the demands of the social organization when it becomes hierarchical. Next, Milgram points out that the variability characteristic of individual organisms, but of human beings especially, must be overcome if the group is to function. That is, it is not a case of limiting the activity to the lowest or preferred level of any one of the individuals, but of internally modifying each individual so that they all operate at the socially ordained level. The basic reason why this occurs is rooted not in individual but in organizational needs. (131)
There is a process of internal modification Milgram terms “the agentic shift,” which takes place when an individual enters a hier-
archical social arrangement, and directs the individual’s behavior toward obedience. An individual is either obedient or not within the social arrangement according to whether the agentic shift takes place. Milgram’s explanation of obedience is in reality a two-valued logic. “Where in a human being shall we find the switch that controls the transition from an autonomous to a systematic mode?” Milgram asks. “Hierarchical inhibitors and disinhibitors alter the probability of certain neural pathways and sequences being used.” This chemoneurological cause is reflected in an internal psychological state, for the person entering an authority system no
longer views himself as acting on his own but as an agent executing the wishes of his superior. This attitude Milgram calls “the agentic state” which is “the master attitude from which the observed behavior flows.” (133)
Chapter 11, “The Process of Obedience: Applying the Analysis to the Experiment,” is not a detailed analysis of Milgram’s series of experiments but of the concept of the agentic state with reference to the experiments. He considers three aspects: first, the antecedent conditions that move a person from an autonomous to an agentic state; second, the consequences of this shift behaviorally and psychologically for the person himself; third, the binding factors that keep a person in the agentic state.
There are a number of factors that make up the antecedent conditions of the agentic state, which, in effect, program the individual to be obedient to authority. In the context of a person’s lifetime, these are family, institutions such as school, employment, and the
military as well as rewards such as job promotions. The net result of this experience is the internalization of the social order. Within a specific situation there are more immediate factors that lead to the agentic state. The individual must perceive that he has entered a social situation in which an authority figure is appropriate and that one such figure is identifiably present. Other immediate factors are the subject’s perception that he has entered an authority system, the coordination of specific commands with the particular purpose of an authority, and an overarching ideology which justifies the authority situation.
MILGRAM AND AUTHORITY
The general consequences of a person entering the agentic state is that he becomes “something different from his former self, with new
perspectives not easily traced to his usual personality. ” (143) The reasons for this are that once a person is in an authority situation and
subordinate to an authority figure, he tunes into the signals emanating from the authority figure; he is attentive to each word and responds willingly to them because the authority situation has
re-defined the circumstances and details in such a way as to effectively dictate reality for the person. As a result, the person experiences a loss of personal responsibility for whatever actions he is to perform since he is now concerned instead with how well he performs in the authority situation. The person’s self-image, Milgram says, is no longer involved with the actions he is commanded to perform. Commands, which both describe an action to be performed and demand that it be done, effect the action of obedience in the
person. The agentic state is not just another word for obedience; rather, it is that state of mental organization which enhances the likelihood of obedience. Obedience is the behavioral aspect of the
There are necessarily binding agents which keep a person in the agentic state. Otherwise any disturbances would eliminate the
tendency to obey authority despite internal disagreement with its orders and produce the tension manifested by some of the subjects in the experiments. These binding agents are first, the sequential nature of the actions since once a person starts obeying he becomes implicated in the procedure and accepts the expectation that he will continue; second, the fact that disobedience is, in effect, a social gaffe, which would cause an embarrassing situation to arise between the person and the authority figure; and third, the anxiety produced when a person contemplates disobedience.
Chapter 12, “Strain and Disobedience,” offers an explanation of the subjects’ choice of obedience or disobedience to the experimenter in non-moral terms. Instead of choice, the term “strain” is introduced to explain why some subjects were obedient and some were not. From the cybernetic viewpoint, strain arises because the demands on an autonomous entity are different when the person is
by himself than when in a hierarchical situation; that is, the “design requirements of an autonomous unit are quite different from those of a component . . . designed for systems functioning.” (153) In effect, a design compromise has been reached, and the compromise does not always work very well. Necessarily there will be mechanisms for the resolution of strain; disobedience will arise when the mechanisms are unable to cope and the level of strain overcomes the binding factors that retain a person in the agentic state.
Strain is seen in an actual sense in the tension experienced by the experimental subjects. The reason they felt tension is that transformation to the agentic state is only partial for some persons. Unlike the more potent authority systems of totalitarian governments, the authority system of the laboratory is less pervasive, allowing “residues of selfhood” to remain beyond the experimenter’s authority. The agentic state is like sleep, numbing the capacity for moral judgment just as sleep numbs the capacity to hear; however, a person may be awakened from both if a stimulant is loud enough. The experimental subjects experience several sources of strain including cues of pain from the victim, moral objections to inflicting pain, fear of retaliation, demands to stop the experiment from the victims, and a conflict between their actions and their self-image. Features that reduce the closeness between the subjects’ actions and their consequences also reduce the level of strain and are called “buffers.” (157) There are several mechanisms for the resolution of strain, including avoidance, denial, physical conversion (in which psychological
strain shows as tics, laughter, etc.), minimal compliance such as pressing the switches for a short period, subterfuge, searching for reassurance, blaming the victim, and noninstrumental dissent, i.e., protesting but obeying anyway. All of those mechanisms, however, allow the authority relationships to remain intact. Disobedience is the ultimate means of resolving tension, but is a difficult act, which arises as a series of stages; inner doubt, externalization of doubt, dissent, threat, and finally disobedience. While it destroys the experiment, disobedience is nevertheless a positive act.
In the next two chapters, Milgram presents objections to the validity of the experiments and his refutation of them. Chapter 13, “An Alternative Theory: Is Aggression the Key?,” deals separately with the thesis that the subjects in the experiment obeyed because the scientific setting allowed the release of their latent aggression. This explanation is based on Freud’s notion that destructive forces are present in the personality of all individuals but usually remain suppressed. Presumably, the experiment, by giving legitimacy to the expression of aggressive behavior, allowed the release of destructive instincts. Milgram denies the validity of this alternative explanation, however, stating that obedience and not aggression is the key to why men kill in war, for example. He also cites experimental evidence for
MILGRAM AND AUTHORITY
this view, noting that when the teachers were allowed to choose their own shock level, very few went beyond the learner’s first protest.
In Chapter 14, “Problems of Method,” Milgram observes that many people upon learning of the experiments seek to deny their
validity because they hold an image of man that does not allow for the kind of behavior exhibited in the experiments. Most people, these critics assert, would most often disobey authority rather than
hurt an innocent person. Therefore, there must be something wrong with the experiments. There are three defects usually cited: 1) that the experimental subjects were not typical, 2) that the subjects did not believe they were really hurting the victim, and 3) that the laboratory setting is so special that nothing can be inferred from it about real life situations. Milgram answers the first objection by
pointing to the wide variety of people who volunteered to be subjects for the experiment. In answer to the charge that the recruitment of volunteers is self-selecting, he cites a study that showed that people who volunteer for psychological experiments are less “authoritarian” than people who do not volunteer. Milgram answers the second objection by citing the results of a questionnaire
given to the subjects a year afterwards which showed that the great majority of them believed the experiment to be real and the shocks genuine. He also points out the tension they showed during the experiment.
He answers the third objection by stressing that however dissimilar the laboratory may be compared to living under Nazi government, the psychological process of obedience is invariant. In authority situations, i.e., those composed of subordinate and superordinate roles, the person responds not because of the content of what he is ordered to do, but because of his relationship to the authority figure.
Chapter 15, “Epilog,” generalizes on how obedience to authority lies at the heart of the grossest evils of our time and the consequent moral dilemmas they produce. The process of obedience to authority is the same in Nazi Germany, Vietnam and the Andersonville prison camp. An interview with an American soldier who participated in the My Lai massacre is included to emphasize these points.
(II) Responses and Criticisms
(A) Responses. Milgram’s experiments are possibly the best known of all those carried out on the behavior of human beings. They have
276 THE POLITICAL SCIENCE REVIEWER
a shocking quality, as if a veil has been ripped aside, leaving us to stare at the stark nakedness of human nature, revealing details which horrify, shock, and embarrass us. The interest which they aroused in the world of empirical psychology and social science is revealed by the number of responses-some critical, some extending the experimental procedure to other areas-written by scientists themselves. There has been at least one book triggered by Milgram’s research (besides his own) and many articles have been written about it. Some of the articles deal with ethical issues raised both by Milgram’s conclusions and by the nature of the experiments themselves which involved deception of the experimental subjects, as well as subjecting them to great distress.
More remarkable than the academic response to Milgram’s experiments is that they have become known not only to wellinformed people who might be expected to be interested in such things, but to the general populace at large. There have been no less than two plays written and produced based on the experiments described in Obedience to Authority, one in England in a repertory setting by playwright Dannie Abse, and another in the United States on network television, starring the erstwhile captain of the starship “Enterprise. ” There has been an article in Esquire Magazine and Milgram himself, it should be pointed out, has enlarged the public knowledge and notoriety of his experiments by lecturing about them
before college audiences and an appearance on NBC television’s Today Show.’
There are two basic reasons for the genuine popular interest shown in Milgram’s experiments. In the first place, reports of them first appeared in psychological journals during the 60s, as did Milgram’s attempts to bring them to popular notice. Since Milgram ‘s experiments could readily be interpreted as dire warnings against obedience to malevolent authority, they fit easily into the anti-authoritarianism of that time, particularly in the context of op-
position to the war in Vietnam. Also Milgram’s own presentation portrayed his experiments as explaining the Nazi phenomenon at a time when world attention was focussed on the capture, trial, and execution of Adolf Eichmann. Indeed, the tie-in of his experiments with Nazism is made so frequently in Obedience to Authority and in
Milgram’s popular presentations that it finally becomes impossible
1. Paul Meyer, “If Hitler Asked You to Electrocute a Stranger, Would You?,” Esquire, 73 (Feb. 1970), 72 et seq.
MILGRAM AND AUTHORITY 277
to see his experiments as anything but an anodyne, an attempt to relieve the pain of the memory of the vast evil of the Nazi holocaust against the Jews; Milgram mentions the holocaust in the very first
paragraph of his book as if to say once we understand it or once we can explain it, then we may prevent it from ever happening again.
Another reason for the interest in Milgram’s experiments is that they blatantly reveal a facet of human nature we would deny, namely, how easily we are influenced by others. It is noteworthy
that most people upon hearing of Milgram’s experiments are repelled by them. Denial of their validity or egregious underestimating of the number of people who will obey are typical responses, and Milgram is no doubt correct when he maintains that
objections about the ethicality of his experiments would not be half so vehement if most of the participants had quit them. Our concept of human nature from the time of the Renaissance is that we are autonomous and rational, ruled by reason. Lately, we have found out differently, but enough sense of our independence and individual reason survives in our culture that we are shocked
2 whenever the discoveries of psychology and sociology contradict it. The irrational inner urges that direct our personal behavior, and the fact that social pressures determine most of our opinions are things we do not wish to know. That men would obey another man simply because he appears in the guise of an authority figure and thereby inflict needless pain on another human being is also something that we would rather not recognize. But Milgram’s research has showed it to us, and in our shock and embarrassment lies also its compelling interest.
(B) Methodological Questions. The reluctance of many observers to accept the findings of Milgram’s research has led to criticisms which attempt to deny their validity. Milgram deals with several of them in his book, but significant questions remain. While we may readily admit that, as Milgram argued, his experiments are applicable beyond the laboratory and that most of his subjects did believe they were administering a real shock, some lingering ques-
2. “What we have learned under the guidance of studies in modern social psychology, with the dismaying spectacle before us of enlarging masses of insecure individuals seeking communal refuge of one sort or another, is that the rationalist image of man is theoretically inadequate and practically intolerable . . . . We know no conception of individuality is adequate that does not take into consideration the myriad ties which normally bind the individual to others from birth to death.” Robert A. Nisbet, The Quest for Community (New York: Oxford University Press, 1978), 229.
278 THE POLITICAL SCIENCE REVIEWER
tions remain about the selection process by which volunteers were garnered for the experiments. Did they not, in effect, constitute a self-selected pool more ready to accede to authority than the population at large? Surely, people who volunteer for whatever reason will more readily go along with what is demanded of them than those
who are forced. That is why particularly dangerous military missions are sometimes put on a volunteer basis and why elite forces who are required to perform such missions, such as the Green Berets and the Marines, are volunteer forces. Milgram denies this point, maintaining that the argument is circular as if it says that those who are willing to obey are willing to obey. The point is a tricky one but it is not entirely semantical, for the fact of the volunteering itself by the participants is the critical point which is beyond experimental
control or observation. Therefore, Milgram cannot say what effect it has on his experiment and it remains a point for conjecture and speculation.’
There are several other significant methodological difficulties which tend to weaken Milgram’s conclusions. One of them deals with the question of what constitutes disobedience. Is it simply the act of breaking off the experiment and refusing to participate further in it as Milgram would have it? Clearly, the subjects gave many other signs of distress and extreme tension; they attempted to
subvert the experiment by giving lower shock levels than demanded when the experimenter was out of the room, they gave shocks as brief as possible in duration, gave verbal clues to the learner, and argued with the experimenter. Yet, while Milgram (or his cohorts) carefully described all of these actions, he classified them as “noninstrumental” and did not include them in his measure of disobedience. In real life, however, such “non-instrumental” acts can lose
a man his job or even his life as they did the Russian poet Mandelstam. Surely, if such actions count in real life, they ought to count in some fashion in Milgram’s experiment, especially as he claims the experiment is applicable outside the laboratory.
Another question arises over the experimenters’ statement made to all the teachers in the experiments that no “tissue damage” would
3. The same holds true for research of the type done by Alfred Kinsey in the area of sexual behavior. If questionnaires regarding sexual behavior are randomly handed out, the population refusing to answer them may be defined by a type of sexual behavior which they are hesitant to write about, even anonymously, e.g. chastity or homosexuality. This, of course, would invalidate any attempt to generalize from the questionnaire results to the population at large.
MILGRAM AND AUTHORITY 279
result from the shocks, which amounted to a guarantee that the experiment was not dangerous. According to Milgram this statement was made only to lend authenticity to the experiment. Yet it operates not merely as a detail of Milgram’s stage management like the lights of the shockboard, but also as a factor in the rational decision of at least some of the subjects to remain obedient. We know this because they said so in their statements which are quoted in some of the interviews which Milgram provides. For instance, the “slow” drill press operator who showed far less tension than the usual subject cited it as a reason why he continued to press all the switches. He also, not incidentally, himself had experienced a large electric shock once on his job which, as he explained, had caused him pain but had not killed him. The point is that the presumed malevolence of an authority and the immorality of obedience to its command are of far less degree if carrying out the order causes temporary pain but not permanent damage. Significantly, the statement about “tissue damage” was repeated to encourage recalcitrant teachers even though it is not listed as one of the four “verbal prods” of increasing severity to be used by experimenters when the teachers refused to go on pressing switches. This, in turn, implies that Milgram himself understood, but chose to ignore, the importance to the teachers of the guarantee that their actions would not permanently injure the learners.
Another methodolgical issue has to do with Milgram’s own discounting of the results of three sets of experiments described in Chapter Six. In the first one, a stern experimenter and an affable learner were replaced by a friendly experimenter and a tough looking learner. In the second one, Milgram contracted for an explicit agreement which defined conditions under which the learner could discontinue the experiment due to his “bad heart”; in the third, he changed the experimental setting from Yale to a downtown location.
All three experiments resulted in significant decreases in the obedience rate, from 65% to 50% in the first, from 65% to 40% in the second, and to 47.5% in the third. Yet Milgram claimed that these reductions were a matter of statistical indifference, having “little effect” in the first, of “chance variation” for the second, and “not significantly lower” for the third, even though the first showed a 23% drop in the rate of obedience as defined by Milgram, the second a 38% drop, and the third a 27% drop. In previous experiments, variations of the same range or less were considered significant as in the experiments described in Chapter Four. These
280 THE POLITICAL SCIENCE REVIEWER
effects, particularly of the agreement with the learner, thus weaken
the strength of Milgram’s general conclusions about the willingness of ordinary people to obey malevolent authority, especially when combined with the effect of the statement that no tissue damage would result from the experiments.
The net effect of considering the methodological objections to Milgram’s experimental view of authority is to weaken it but not to invalidate it. That is, while it is true that people are far more com-
pliant to authority (and to social pressure in general) than we might like to admit, the situation is not as severe as Milgram concludes. Indeed, as we have seen, Milgram in some measure must discount the
results of some of his experiments, ignore certain phenomena, and define disobedience in very restrictive terms in order to sustain the degree of menace he sees in men’s relationships to social authority. In fact, his statements constantly placing his experimental results in a political context, particularly in connection with the Holocaust, make it apparent that Milgram has a political as well as a scientific reason for his view of authority. Milgram analyzes a problem, how can the mad horrors of the twentieth century have happened-the police state, the gas ovens, the Gulag, the A-bomb, the Cold War-and reaches the conclusion that we are far too compliant to authority, which also implies a solution-that men ought not to
obey authority as readily as they do. The problem, he asserts starkly, is not authoritarianism as a mode of political organization “but authority itself.” (179) But such an analysis either says too much or too little; too much because, as Milgram admits, we cannot live without authority, and too little because authority is so diffuse and omnipresent that we cannot attack it without further refinement of the issue. Thus, Milgram has, through his experiments, presented us with a dilemma from which we cannot escape. The result of taking what Milgram describes as “an experimental view” of obedience to authority is to condemn human nature for its inherent weakness
while, at the same time, encouraging the human race to a task that is well beyond its capacity or strength.
(C) Ethical Objections. One response to Milgram’s series of experiments, which appeared even before his book appeared and which was based on reports of the experiments published in psychological journals, was the appearance of serious ethical objections by Milgram’s own colleagues in experimental psychology. It is apparent from Milgram’s own description, as well as the observations of those who have seen films of some of the experiments, that
MILGRAM AND AUTHORITY 281
the teacher/subjects undergo severe stress. Milgram described this effect, in an earlier paper in a passage that did not appear in his book.
Many subjects showed signs of nervousness in the experimental situation and especially upon administering the more powerful shocks. In a large number of cases the degree of tension reached extremes that are rarely seen in sociopsychological laboratory studies. Subjects were
observed to sweat, tremble, stutter, bite their lips, groan and dig their fingernails into their flesh. These were characteristic rather than exceptional responses to the experiment. One sign of tension was the regular recurrence of nervous laughing fits…. The laughter seemed entirely out of place, even bizarre. Full blown uncontrollable seizures were observed in three subjects. On one occasion, we observed a seizure so violently convulsive that it was necessary to call a halt to the experiment.”
Such signs of psychological discomfort are obvious indications of severe mental stress. Several psychologists condemned the research for this reason, especially Diana Baumrind who pointed out that stress and embarrassment were frequent results of recent forms of human research in social-psychology. Milgram’s response to this charge is found in an appendix to Obedience to Authority and in an exchange of letters published in Dannie Abse’s book version of his
play based on Milgram’s experiments, The Dogs of Pavlov. He writes here that the degree of stress was not anticipated, but that once he began the experiments, he felt impelled to their completion; that despite their distress at the time of the experiment, the subjects sus-
tained no permanent injury; that many of the subjects said of the experiments that they thought they were good, approved of them, or had learned something about themselves. Milgram also states that only the subjects may judge the experiment, and particularly cites the case of the young man who was a subject and who later refused military service during the Vietnam war. Another aspect of the ethical debate is the issue of whether it is fair to lie to the subjects in effect by telling them that the purpose of the experiment is about learning rather than obedience. Milgram claims that the deception is legitimate, no different from a play or a movie.’
4. `Behavioral Study of Obedience,” Journal of Abnormal Psychology, 67 (1963), 371-78.
5. Dannie Abse, The Dogs of Pavlov (London: Vallentine-Mitchell, 1973), 37-44, 126-27.
THE POLITICAL SCIENCE REVIEWER
The real issue however would seem to be whether or how seriously the subjects were damaged psychologically by going through the experiment. Here, there is some debate, with Milgram claiming that
post-interviews showed no permanent damage. Yet, one of the subjects described by Milgram who was “obedient,” i.e., had pressed all the switches, revealed that after he had described his actions in the
experiment to his wife, she had responded, “You can call yourself Eichmann.” Such announcements from a spouse are not easily forgotten. As Professor Baumrind wrote, “From the subject’s point of view procedures which involve loss of dignity, self-esteem and trust in rational authority are probably most harmful in the long run “8
Whatever the final resolution of the ethical debate, one thing is certain: Milgram’s most obedient subjects acted no worse than he did. They inflicted imaginary pain at the behest of apparently legitimate authority; but Milgram, in the name of science, permitted the subjects to undergo real psychic pain at the behest of no one at all despite the protest of his colleagues.
(III) Authority and Science
(A) Milgram’s Scientific Method. Milgram states at the outset of describing his experiments that, “Simplicity is the key to effective scientific inquiry” (13) and, indeed, this is true. Like Galileo, Newton and Darwin, Milgram sought, when confronted with a highly variable and epistemologically dense set of phenomena, to reduce them to their empirical essentials. This process of scientific reduction involves selecting some elements for study and discarding others, and then explaining the entire set of phenomena in terms of the selected elements. A good example from the history of biology, and one easily understood, is that of Mendel’s discovery of the genetic control of somatic characteristics. Among all the variety and multitude of cases of inherited characteristics, Mendel chose one
plant, the pea plant and seven easily observed and measurable characteristics including height, color, etc. From a series of experiments in which Mendel carefully controlled the mechanism of genetic transmission, he derived a series of mathematical laws which
6. Diana Baumrind, “Some Thoughts on Ethics of Research; After Reading Milgram ‘ s
Behavioral Study of Obedience,' " American Psychologist, 19 (1964), 421-23. Quoted in Abse.authoritarianism,’ as a mode of political organization or a set of psychological attitudes but authority itself.” (179)
MILGRAM AND AUTHORITY 283
accurately described and predicted how characteristics would be
transformed from one generation to the next. He also inferred an entity, one which he could not see, by which genetic transfer took place, i.e. the gene. In this way, Mendel reduced the complexity of genetic phenomena to a set of mathematical laws and posited the ex-
istence of a then unobservable element which caused the phenomena. Mendel's research is nearly an ideal example of how scientific research proceeds, by reducing chaos to simplicity.
Milgram hoped to do the same thing, but choosing the elements on which to concentrate is a tricky business. (Mendel was either extraordinarily lucky or possessed of marvelous insight.) Obviously, it is possible to oversimplify, and this, regrettably, is what Milgram has in fact done. The use of the laboratory setting itself is, let it be said, a brilliant stroke, though not entirely original with Milgram. But, as he notes, critics have doubted the validity of generalizing from this setting to the world at large, something that is never said about, e.g., Mendel's experiments with the pea plant. While Mendel's explanation can be extended from peas to other plants, and then to animals, Milgram's experiments cannot be extended in like fashion. The main reason is that the variety of authority relationships far exceeds the variety even of genetic transmissions, and the modes of disobedience are infinitely variable, as anyone who has raised an active child can testify. Thus, authority relationships in child rearing, employment, military, religious, and public order situations are all very different from each other. While there are resemblances (or we could not identify them all as authority situations), the differences are tremendous, and the applicability of Milgram's experiments must be in doubt, unless, that is, Milgram were to systematically apply his theories to all these kinds of situations, something he does not even attempt to do. As Milgram notes, "Psychological matter, by its nature, is difficult to get at and likely to have many more sides to it than appear at first glance." (13)
Milgram's research yields results typical of most research in the social sciences-theories of a high level of generality that are mechanical and unconvincing; middle level theories, such as the agentic shift, which are partially successful as explanations; finally,
low level theories or "laws" which are simply generalizations from the evidence. Milgram's high level explanation of obedience is a curious blend of an attitudinal anti-authoritarianism and psychosocial theorizing combined with a clumsy amalgam of various mechanistic formal explanations, including evolutionary,
THE POLITICAL SCIENCE REVIEWER
cybernetic, neurological, and chemical. In the end, Milgram contradicts himself by the assertion that, "[a]n element of free choice determines whether the person defines himself in this way or not," a statement which comes at the end of the chapter (10) in which he puts forth his evolutionary, cybernetic, and chemical modes. (134) But an element of free choice relieves the need for "chemical inhibitors" or any comparable explanation in which human behavior is driven by non-intentional causes.
Milgram is on somewhat firmer ground when offering theories of a middle level of generality which more directly rest on empirical
observations. Thus, he posits an "agentic shift" when the individual yields his conscience into the control of an authority figure which, Milgram states, must happen as an individual enters a hierarchical situation. The notion that a person will perfom acts under the command of authority that he would refuse to do otherwise is a perfectly valid inductive generalization from Milgram's experiments, albeit one he could have made beforehand. This generalization is the basis for the theory of the "agentic shift" (which is roughly comparable to Mendel's "genes") However, while there is some basis for supposing that such a psychological phenomenon as the agentic shift exists and
while Milgram describes it in some detail, there is no question that he pictures its operation in terms which are too black and white. That is, Milgram supposes that before the agentic shift occurs the person is a fully independent and rational agent, in control of his faculties, who would refuse to perform any act contrary to his conscience. After the shift occurs, this same person is now ready to perform acts comparable to those performed by the agents of the Nazi SS. There is no room for qualified assent of the sort that often is the case when adults obey authority, or for understanding the different kinds and degrees of obedience one gives, for instance, to policemen, parents, military officers, government officials, bosses or spouses. The ancient Stoic obeying the laws of Rome exemplifies the same empiric law as the actions of a guard at Auschwitz for Milgram.
In short, what is lacking in Milgram's theoretical explanation is the capacity to explain satisfactorily the nearly infinite degree of kinds of human relationships, for the authority relationship is not the same in all social hierarchies, nor is it even the same within a
single hierarchy. Thus, Milgram says that the obedience of the German soldier to his sergeant follows the same empirical laws that describe the obedience the Nazi generals gave to Hitler. (130) Yet it is not true, for we know that some Wehrmacht generals not merely
MILGRAM AND AUTHORITY 285
obeyed like automatons, but aided and advised Hitler, and willingly carried out his orders, while others did not. Thus, General Jodi, Hitler's Chief of Staff, was tried and shot at the end of the war, whereas Admiral Doenitz was sentenced to a relatively short prison term. Finally, it was Prussian military officers, such as Count Von Stauffenberg, the very archetype of authoritarians, who tried to kill Hitler, which indicates, if nothing else, the variability of the "agentic shift."
Milgram has been particularly successful in discovering low level theories or "laws" because his experiments clearly highlight the effects of such causal factors as proximity to the victim, closeness of authority, divided authority, institutional context, the nature of
commands, and peer rebellion on obedience. His experiments also show, implicitly but no less clearly, how effective authority based on science is, since one of the unintended results of Milgram's research is to have demonstrated experimentally that scientific authority takes its place alongside political, legal, military and religious authority in modern society. It is unlikely that the volunteers would have subjected the victim to presumed shock at the behest of a minister, policeman or military officer in that context, coming in off the street, as it were. That is, while people will perform difficult or even cruel tasks under, e.g. military authority, the indoctrination into military life is presumed to require an intensive period of "basic training." But this is not true of scientific authority in this case, for the teachers in the experiment gave shocks after only a minimal indoctrination.
(B) Milgram's Concept of Authority. Milgram never explicitly defines his concept of authority, presuming instead that an experimental view will be morally neutral and will present the issue in contemporary and immediate terms. (xi) Milgram's actual views about authority and obedience to it are not difficult to discern, however, for there are attitudinal elements which appear throughout Milgram's explanation and commentary and which intimate very clearly that Milgram perceives obedience to authority as a severe danger to our civilization, that he views the exercise of authority by a social agent as often destructive, and that he virtually equates the systematic use of authority with a form of fascism. His many references to Nazis and the concentration camps which appear throughout the book from beginning to end and his specific statements about the danger of the all-too-human tendency to obey malevolent authority make his attitude apparent. But a more reveal-
THE POLITICAL SCIENCE REVIEWER
ing comment appears in his statement about one of the experimental subjects, the Professor of Old Testament theology. This subject had stopped giving the victim shocks shortly after the first protest, thus standing as one of the disobedient subjects. Asked afterwards how most effectively to strengthen resistance to inhumane authority, the Professor said, "If one had as one's ultimate authority, God, then it trivializes human authority." (49) But this insight, worthy of an Aquinas, a Bonhoeffer, or a Maimonides, is not sufficient for Milgram's purposes. He comments, "Again, the answer for this man lies not in the repudiation of authority but in the substitution of good-that is, divine-authority for bad. (49) Again, what Milgram
seeks is the destruction, apparently, of authority itself. "For the problem is not
Although Milgram concedes the need for authority, (212) he does not in fact really understand or accept its function in human life, for Milgram’s prescientific attitude, i.e., the one he held before he ran the experiments, is a simple but fervent rejection of authority itself. The degree to which men are compliant to authority no doubt frightened him, but his reaction is to portray the condition as far worse than it is. It is as if Milgram wants to shock his readers by exposing the weakness of human nature itself. Here lies the explanation for the repugnance toward his experiments: Milgram not only disdainfully exposes the weaknesses of his individual human sub-
jects, but the weakness of our common human nature as well, and he does this with the attitude of a fifth grade teacher lecturing her class for writing dirty words on the desk-tops.
What Milgram has missed is how deep the need for authority is, how important role models, parents, teachers and social authorities are in the formation of our characters as children, how, even as adults, we still need to look to bosses, peers, spouses, public figures and religion for direction and support in the daily business of con-
ducting our lives. As a result, he has egregiously misdiagnosed the basic causes of the evils of the modern era, for it can be plausibly argued that the real evil of our time is not that men too willingly obey authority, but that they are not willing enough to obey it.’ The issue of social authority and the degree to which men do and should
7. As an illustration of this point, the guest interviewed on the “Today Show” the day after Milgram appeared on it was an expert whose concern was the ominous increase in the rate of forcible rape. The more immediate danger to society thus came from men who disobeyed social authorities rather than from those who obeyed them.
MILGRAM AND AUTHORITY 287
obey it is far more complex than Milgram acknowledges or implies. We cannot settle the question here, but we will add in passing the view of sociologist Robert Nisbet who argues that it is the destruction of intermediate social authorities which stand between the individual and the state that is the main cause of the evils of modern times.
We are prone to see the advance of power in the modern world as a consequence, or concomitant, of the diminution of individual freedom. But a more useful way would be to see it in terms of the retreat of authority in many of the areas of society within which human beings commonly find roots and a sense of the larger whole. 8
The basic human requirement for authority and authority structures that Nisbet points out must somehow be balanced against the dangers of misuse of authority that Milgram points out. This is the
true dilemma of authority of our time.
(C) Anti-Authoritarianism and the Priority of Theory. Milgram’ s attitude toward authority has directly influenced his attempt at ex-
plaining and evaluating his research results. That is, Milgram’s actual procedure has not been to analyze a complex phenomenon by running a series of experiments and then arriving at theories to explain the results. Rather, he began with a highly defined concept of authority and then set up his experiments and conducted his research guided by that attitude. This is not the same as saying that he set out to prove that his concept of authority was true and twisted the evidence in any way to suit his preconceived notions. It is to say that Milgram set up his experiments and read his research results with a definite attitude toward authority in mind and that it influenced the way he literally saw authority in light of his experiments. For, as we have seen, Milgram began with the idea that authority was a phenomenon not only somehow mixed up with the turmoils and vast evils of the twentieth century, but as a primary cause of those turmoils and evils.
The projection of Milgram’s anti-authoritarianism into the processes of the research itself is shown clearly in the manner in which
he treats the results of his observation and experiments, namely, his anti-authoritarianism has led him to emphasize the degree to which
people in general are compliant to authority. This attitude has led
Milgram in turn to deny the validity of three sets of experiments
8. Nisbet, op cit., xiv.
THE POLITICAL SCIENCE REVIEWER
which showed how easily disobedience to authority could be increased by a change in personnel, by a prior agreement with the victim and a change of setting from Yale. More significantly, it has also led him to discount many manifestations of resistance to authority, denying them even status as evidence in the experiments. These
manifestations were the so-called “non-instrumental” ones such as verbal protest, not pressing the switch at the next highest level of intensity, or pressing the switches for as short a time as possible. Had each of these non-instrumental manifestations been given a relative value and included as evidence of resistance to authority along with “disobedience” itself-i.e. the act of breaking off the experiment-a much different picture would emerge, one more subtle and hence more applicable to the complexities of authority relationships in the
real world, and one more hopeful about the potentialities of human nature. (For instance, each subject could have been assigned a “score,” 10 for disobedience, 5 for verbal protest, 2 for not pressing the proper switch level, etc. Then a curve, or sliding scale would result with relative levels of disobedience, rather than a “sheep versus goat” phenomenon. But this did not fit Milgram’s understanding of how authority works.)
The fact that Milgram had a definite concept or attitude about authority which influenced the manner in which he ran his ex-
periments and interpreted their results is not unusual. Such a procedure is characteristic of the social sciences, where the phenomena are complex and highly charged with meaning before the scientist even approaches them. The social scientist, no less than the layman, has ideas about such things as war, politics, economics, family life and authority, which when articulated constitute a theory in scien-
tific terms. This fact is well understood in reference to social science, but what is not so well understood is that holding the theory before analyzing the evidence is characteristic of the physical sciences as well. Indeed, the insight most characteristic of twentieth century philosophy of science is just this point, that it is the practice of scientists to have a theory in mind before proceeding to experiment and that, contrary to Bacon, the scientist’s mind is not a tabula rasa in
which explanatory conceptions arise only after he has examined his research results. Scientific observation, wrote N.R. Hanson, is a “theory-laden” undertaking, for observation of a particular object in an experiment is shaped by previous knowledge of that object.°
9. Norwood R. Hanson, Patterns of Discovery (Cambridge: Cambridge University Press, 1965), 19. The same point is also made by Popper, Kuhn, Einstein, and Duhem.
MILGRAM AND AUTHORITY 289
The priority of the theoretical to the particular is true of all fields of empirical science not just the social sciences, although it stands out more in the social sciences because the prior theoretical commitments are likely to be ideological or political in nature. But then
this highlights a fundamental dilemma for Milgram and for the social sciences in general; if experimental research is influenced and directed by theories held prior to the experiment, and if the experimental results themselves are literally seen in terms of the theory, then how can Milgram and social scientists in general guarantee that their research results and the theories explaining them are true or objective?
(D) The Authority of Science. One aspect of authority that Milgram misses is the way that science operates as a kind of authority in comtemporary society. It is deeply ironic that Milgram does not comment on this fact anywhere in his book because the authority of science is the effective basis of his experiments. That is, the motivating “force” which moved the experimental subjects to “shock” the victim and the force against which they were to rebel in order to generate a quantifiable measurement of disobedience, was the authority of empirical science. Indeed, Milgram was meticulous in reinforcing the impression that the subjects were helping carry out a scientific experiment, including such details as white laboratory coats, electric lights on the shockboard, the text of the newspaper advertisement soliciting test subjects, the laboratory setting and, above all, the actual indoctrination at which the subjects were told that they were helping carry out an experiment. It was against this carefully stage-managed array of cues to scientific authority that the subjects were to rebel or with which they were to
The use of science as the source of authority was in no way accidental, since it was the one authority that Milgram could most
easily manipulate and the one which was nearest to him. As a bona
fide research scientist, Milgram is an authority figure on a par with a minister of the Gospel, a high ranking military officer, or a police
official, and thus could influence people’s behavior in an experimen tal context. So deeply does the authority of science root in the mores of our society that only a cursory indoctrination was necessary to involve the subjects and to get them to perform an act which no other kind of authority could so quickly and so easily motivate, i.e. inflic-
ting pain on an innocent person. The reason that science was so effective as a source of authority in the experiments did not rest,
290 THE POLITICAL SCIENCE REVIEWER
ultimately, on anything Milgram himself did, but was the result of a
long tradition in the West of accepting science as an authority, which extends back to the Enlightenment of the eighteenth century. Science was, so to speak, the avatar of reason, the one human activity that could produce intellectual certainty and whose practical results could improve the lot of mankind. Science and Reason, so conceived, arose from the release of the human intellect from the fetters of religion, metaphysics, superstition, fear, tradition and pas-
sion, in short from all the forces of irrationality. Science was seen to be disinterested and benevolent, a model as well as a method for understanding the nature of man and society. The widespread suc-
cess of science in the elaboration of theories, in the number and importance of discoveries about the physical universe, and in enabling man to control his environment reinforced its status as a source of authority in our culture. It was on this tradition of authority that Milgram depended for effecting the obedience of his experimental subjects.
Two contradictions result from Milgram’s use of science as the source of authority in his experiments. The first derives from the implication that science is or can be a malevolent authority, for this weakens the value of the very authority, namely science, by which
Milgram attempts to prove something about authority. That is, Milgram seeks to prove that human nature is too compliant towards authority by means of scientific experimentation, at the same time that the authority he uses in his experiment is scientific authority. If
we then accept Milgram’s conclusion as experimentally proved that we are too complacent towards authority, we further ought to question the truth of his conclusion, since it assumes our full acceptance of the authority of science. In conclusion, if we accept Milgram’s premises, then we both ought and ought not to accept his conclusions, which in the purely formal terms of logic means that at least one of his premises is incorrect. Which of these premises is incorrect is difficult to say, but most likely Milgram has overstated the utility of the scientific method in explaining social phenomena such as obedience to authority. In less formal terms, the real effect of Milgram’s
series of experiments is to weaken science as an authority source, not only because they can be interpreted to weaken all authority sources, but also because of the serious ethical question they raised as to whether Milgram should have lied to his subjects and put them
through the extreme stress that many of them experienced.
The second contradictory result of Milgram’s experiments is that
MILGRAM AND AUTHORITY 291
they revealed something about human behavior which contradicts the rationalistic tradition in which science plays an important part, i.e. they revealed that men tend to obey authority even when it violates their personal notions of morality and that men’s values and identities are more often the result of their social environment than their personal choice. In short, science as a rational method of discovery has determined that man’s behavior is irrational. This discovery seems to have depressed and angered Milgram; however it is worth noting that other research conducted by social scientists suggests that other aspects of human nature contradict Enlightenment expectations, e.g., that man is aggressive by nature, that he is territorial, that sex roles and intelligence are genetically influenced if not determined. Science has authority because we expect that it tells us the truth. Now science, through the discoveries of
psychology, sociology and anthropology, is re-affirming things about human nature which Enlightenment rationalism denied, but
which older traditions such as Christian theology and Greek philosophy affirmed.
(E) Conclusion. The combination of empirical science and political philosophizing is not new in the history of Western thought since the Renaissance. Indeed, from Bacon’s New Atlantis to Skinner’s Walden Two this combination has been a potent mixture in our culture, our politics, and our thinking. The difference is that in the twentieth century, the social sciences have developed to the point where they can sometimes give the same kinds of control and information regarding individual human beings and social institutions that empirical science had hitherto given us about physical substances and living matter. Thus, while Hobbes, for example, could only speculate about how the nature of man and society ap peared from a scientific point of view, social scientists today have at their disposal a large amount of precise, empirical knowledge from which to develop what Milgram calls “an experimental view” of both man and society. However, what the addition of scientific theory and experimental information adds to our understanding of man and society and how science transforms our understanding by the addition of these elements is still an open question.
Milgram makes the point that while many critics of the social sciences claim they don’t discover anything not already known, that is not true for his series of experiments. (27) Yet, the facts that individuals are easily influenced by others, and often will obey malevolent authority was well known prior to Milgram; in fact, they
292 THE POLITICAL SCIENCE REVIEWER
were known to the ancients. Milgram’s “discovery” that men will perform deeds under the command of legitimate authority that they otherwise would not perform-which includes acts of heroism and bravery, not just cruelty-is but one specific aspect of a more general fact now amply demonstrated by social psychology, namely, that individual personality contains large elements that are socially and culturally induced. To repeat, this contradicts our cherished Enlightenment illusion that we are both rational and in full command of ourselves, and it is ironic that modern science provides the contradictory evidence. Yet, it was Aristotle who defined man as a “social animal” and the fact of our social nature was well known to those close observers of life in the Greek city-state, Socrates, Plato, Thucydides, Euripides and the Sophists. Indeed, to the Greeks, men were simply citizens of their cities, not autonomous individuals cooly picking and choosing their allegiances according to a logical rubric.
The re-affirmation of traditional concepts of human sociality in the social sciences indicates the difference between the physical and the social sciences. While Aristotle’s Physics is a mystery to modern physicists and of no use in the study of physical phenomena, the Ethics and Politics are still read today for their insight into social phenomena. As science has developed since the seventeenth century, its method has more easily dealt with physical phenomena than social phenomena and we can speculate that the reason is that the complexity of science’s subject matter has overmatched the range of its method as it has approached the areas of the social sciences. The final implication is that if we are to obtain adequate knowledge of social reality, it must come from other sources as well as from social science. This is demonstratedkby the fact that the human race has historically had such knowledge long before the advent of social science, i.e., before the application of empirical methodology to social phenomena. An adequate expression of the reality of social phenomena must be looked for in religion, philosophy, history and one’s own personal experience in addition to empirical science, since these other accesses to reality are necessary to gain a full and balanced view of social phenomena.
But if men are social animals whose ideals and courses of action are frequently prescribed by their social environment, then the dangers of malevolent social authority are intense and severe. Nor is this merely a deduction; it is an observation only too often made by citizens of those states subject to the risk of misrule by tyrants,
MILGRAM AND AUTHORITY 293
demagogues and autocrats. Indeed both Aristotle and Plato describe these dangers in great detail, and both prescribe a remedy, Aristotle a mixed constitution (really a division and balance of power) and Plato an education based on ideals for the guardian class. Yet, the last voice we hear on the subject of the dangers of malevolent authority, and the all too human ease with which we accede to it, should be a current one, for while the ancients had their special travails, we have had ours as well. In this sense, Milgram’s experimental view of authority serves us well and makes a positive contribution. The science may be faulty-the research not correctly evaluated, the theories incomplete and contradictory-but science is the idiom of our time, just as myth was for the Greeks. What Milgram has really done is to illustrate the danger of authority in properly scientific fashion for our time, for what was well known throughout recorded history, we will accept only under the authority of science.
Brown University Graduate Center JOHN C. CAIAZZA