Like many social scientists back in the early 1960s, Stanley Milgram wondered about the obedience displayed by the German people during the Nazi era. How did the demented Nazi ideology and the brutality and genocide it spawned come to be embraced by an entire nation? It is not hard to imagine that Adolf Hitler was a very disturbed, hateful, narcissistic man because of some combination of genetic predispositions, childhood upbringing, and stressful life experiences. We might conclude the same of his closest Nazi allies. But how could the majority of an entire large nation participate in such egregious atrocities? Could millions of people be that deranged or evil?

259

Milgram did not think that Germany was filled with evildoers, but he did think that perhaps the German people were particularly prone to obedience because they were raised in an unquestioning environment that encouraged obedience to authorities. He speculated that living in this authoritarian society might have led citizens down the tragic path of war, genocide, and national disgrace. To understand this mass obedience, Milgram developed a laboratory situation to test people’s willingness to harm another just because an authority figure told them to do so. Because of the astounding, surprising nature of the findings, these studies have become the best-known, most widely taught research produced by social psychology: the Milgram obedience studies. We refer to this research as a set of studies rather than a set of experiments because Milgram actually did not manipulate an independent variable or randomly assign participants to conditions. Rather he conducted a series of 18 demonstrations that revealed something very fundamental about human nature—something that, before this research was conducted, no one had fully realized.

In his first study, conducted at Yale University, Milgram (1963) recruited 40 ordinary men, ranging in age from 20 to 50, from the New Haven, Connecticut area to participate in a study of learning. When each man arrived at the lab, he received $4.50 and was told it was for coming to the experiment and was his no matter what happened from that point on. He and another apparent participant were greeted by an experimenter who explained that the study concerned the effects of punishment on learning. However, the apparent participant was actually a confederate working with the experimenter. The participant and the confederate chose slips of paper, ostensibly to determine randomly who would be assigned to be the teacher and who would be assigned to be the learner. This drawing was actually rigged so that the participant would always be the teacher.

The learner was then strapped into a chair and an electrode was attached to his wrist. The experimenter explained that the electrode was attached to a shock generator in the next room. The teacher’s task was to read pairs of words to the learner and then test whether the learner could remember which words were paired together by choosing the correct paired word from four possible words. The learner would choose his answer by pressing one of four switches in front of him; his responses would be indicated by a light in an answer box in the adjacent room (see FIGURE 7.7a). The participant was then escorted into that room, which housed the shock generator and the answer box. The generator had 30 switches, labeled from 15 to 450 volts from left to right, in 15 volt increments (see FIGURE 7.7b). The switches for voltage levels above 180 were labeled “Very Strong Shock”; those above 240 were labeled “Intense Shock”; those above 300 were labeled “Extreme Intensity Shock”; and those above 360 were labeled “Danger: Severe Shock.”

The participant was instructed to read the word pairs and then test the learner for each word pair. Whenever the learner made an error, the participant was to give a shock to the learner. With each error, he was to increase the voltage of the shock by 15 volts. Thus, the participant would administer a 15-volt shock for the first error, a 30-volt shock for the second error, a 45-volt shock for the third error, and so forth. The participant was given a sample shock of 45 volts to convince him that the shock generator was real. What the participant didn’t know was that the shock generator was not connected to the electrode attached to the learner’s wrist. How long the participant would continue administering escalating levels of shock before refusing to continue was the indicator of level of obedience.

260

As the study proceeded, the learner used a predetermined set of three wrong answers to every correct one. In this initial study, the learner was silent when the shocks were administered, but if the participant continued to the point of delivering 300 volts, the learner pounded on the wall. In subsequent variations of the study, starting with Study 2 of the series, the learner grunted, complained, and screamed as the shocks escalated, but these reactions did not affect the level of obedience exhibited by the participants. After the 300-volt shock, the participant was confronted with silence as the learner no longer provided any answers. At this point, participants usually asked the experimenter for guidance. The experimenter told them to treat no responses after 10 seconds as a wrong answer and to continue with the appropriate level of shock. If the participant expressed unwillingness to continue, the experimenter used a set of four verbal prompts ranging from “Please continue” to “You have no other choice, you must continue.” But, of course, the point is that participants did have a choice: Should they obey these commands, given by respected researchers at Yale University? Or should they refuse and perhaps ruin the experiment but save the learner from additional pain?

Milgram asked Yale senior psychology majors, graduate students, faculty, and psychiatrists to predict whether they would continue obeying all the way to 450 volts (about four times the voltage of a standard electrical outlet). He also asked people from these same groups to predict the percentage of participants in the study who would. No one predicted that they would do so themselves, and each of these groups predicted that fewer than 2% of participants would obey fully by continuing to administer shocks until the maximum voltage was reached.

261

They were quite wrong in their predictions. In fact, 26 of the 40 participants, 65%, obeyed fully to the point of agreeing to deliver a dangerous 450-volt shock to the learner. Furthermore, not a single participant refused to continue until the shock level reached 315 volts. The level of obedience went far beyond what anyone predicted. As Milgram noted, this was a particularly remarkable level of obedience to engage in morally reprehensible actions because, unlike in many real-life situations, the authority figure here (the experimenter) had no real power to enforce his commands, and no significant penalty would be incurred from disobedience.

In subsequent studies, Milgram used the same shock-the-learner paradigm with one or more aspects changed each time (see FIGURE 7.8 for rates of obedience of some of Milgram’s different variations). From this series of follow-up studies, Milgram learned about variables that contributed to the high level of obedience found in the original study. The most important variable was the extent to which the person giving commands was perceived to be a legitimate authority figure. For example, if the study was run in an unimpressive-looking building in downtown New Haven rather than at the prestigious Yale University, full obedience was reduced to 47.5%. If the person directing the real participant to deliver the shocks seemed to be another participant in the study instead of the experimenter (who left, supposedly to take a telephone call), full obedience dropped to 20%.

These findings help explain why so many Germans contributed to the heinous actions during the Nazi era once Hitler became a legitimate authority figure in 1933. However, Milgram (1974) emphasized two points to make it clear that this proclivity to obey authority is a potential danger in any culture and in any era. First, he pointed out that horrific acts are not limited to dictatorships and fascist states. Once in power, duly elected officials in democracies are legitimate authorities who often demand actions that conflict with conscience. Second, he noted that the Nazi era was far from the first or the last time that obedience has led people to engage in egregious, destructive actions:

[T]he destruction of the American Indian population, the internment of Japanese Americans, the use of napalm against civilians in Vietnam, all are harsh policies that originated in the authority of a democratic nation, and were responded to with the expected obedience. . . . [W]hen lecturing . . . I faced young men who were aghast at the behavior of experimental subjects and proclaimed they would never behave in such a way, but who, in a matter of months were brought into the military and performed without compunction actions that made shocking the victim seem pallid. In this respect, they are no better and no worse than human beings of any other era who lend themselves to the purposes of authority and become instruments in its destructive processes. (1974, pp. 179–180)

262

In addition, Milgram explored the closeness of the victim. Recall that in the original study, the victim was in a different room and could be heard but not seen. In a variation in which the victim was in the same room, full obedience dropped to 40%. And in another version of the study where participants physically had to place the learner’s hand on a shock plate, obedience dropped to 30%. This is a substantial decrease in obedience, but it is also quite remarkable and disturbing that 3 in 10 participants would obey the repeated commands even when doing so meant physically compelling the shock. The plausibility of this particular variation could be called into question, however, because the confederate had to act out receiving the shocks, crying out, screaming, and so forth. Milgram (1974) did not provide enough details for us to assess this potential problem fully.

These and other variation studies indicate that the more psychologically remote the victim, the greater the obedience to doing the victim harm. It is interesting that in modern warfare, most of the killing is done very remotely. In aerial bombing, the soldiers dropping the bombs neither see nor hear their victims. The first aerial bombings occurred during the Spanish Civil War of the 1930s. Ever since then, the highest civilian casualties in war have come from this form of violence, whose victims are physically and psychologically remote.

Milgram also wondered what would happen to the level of obedience if the real participant saw two other supposed participants defy the experimenter. As Asch found with conformity, this seemed to reduce greatly the impact of social influence, because the full obedience level dropped to 10%. Those who disobey make it easier for others to disobey, something the Nazis seemed to understand. A vivid portrayal of their common response to any act of disobedience by those on the way to or in a concentration camp can be seen in the classic film Schindler’s List: a bullet to the neck.

Thus, psychological distance from the authority, psychological closeness to the victim, and witnessing defiance all reduced obedience. But one important variation led to even more obedience than the original study. In this version, the real participant didn’t physically flip the switch on the shock generator. Rather, the real participant delivered the memory test, while another supposed participant dutifully delivered shocks up to 450 volts. In this version, 92.5% of the participants obeyed fully. This is an especially chilling finding. Very few Germans actually pushed Jews and other “undesirables” into the gas chambers or shot them, but many, many people participated in indirect ways, conducting the trains, spreading hatred of Jews and other groups, arresting them, processing paperwork, building the camps, designing mobile gas chambers at the Volkswagen automobile plants, and so on.

263

We can see from the Milgram research that at least a minority of people will eventually disobey when they feel that their own actions are physically causing harm. But when people contribute to but are not physically causing the harm, virtually all resistance to participating in atrocities sanctioned by authorities seems to vanish. Consistent with this reluctance to be directly responsible for causing sanctioned harm, it is common practice during executions for more than one individual to pull the lever, inject the serum, flip the switch, or take aim and fire. Thus no single individual knows for sure if he or she was actually physically responsible for executing the person condemned to death. The willingness to obey when one is not certain that one is physically causing the actual harm seems virtually limitless.

The lesson here is reminiscent of the oft-quoted adage (attributed to the statesman Edmund Burke), “The only thing necessary for the triumph of evil is for good men to do nothing.”

In addition, if the harm is not severe physical pain and won’t occur until after the participant leaves the situation, resistance to obedience again is virtually absent. In a series of studies in the Netherlands, when participants were commanded to give negative evaluations of a job applicant’s test performance that would result in the applicant’s not being hired at a later date, over 90% of the participants obeyed these instructions (Meeus & Raaijmakers, 1995).

264

What about now? The original studies were done during the early 1960s, a time when people perhaps had more faith in science and authority and the civil unrest that characterized the later part of the decade had yet to arise. Would obedience levels be lower now? Well, it turns out that it’s hard to answer this question definitively, because institutional review boards now generally do not allow people to use the Milgram paradigm. When Milgram published his initial research, there was quite an uproar over the ethicality of commanding study participants to engage in behaviors that they believed would seriously harm another person (e.g., Baumrind, 1964). Some critics focused on the stress that was imposed on participants; during the task, many displayed signs of stress, such as twitches and nervous laughter. Other critics contended that the most egregious ethical problem was that the study led most participants to discover something about themselves that could harm their self-image: that they were capable of seriously harming another human being simply because they were told to do so.

Milgram responded to these concerns first by noting his elaborate and thorough debriefing procedures. At the conclusion of each session, participants met the learner and saw that he was unharmed. They were fully informed of the study’s purpose in examining the powerful effects of the situation on behavior. Milgram also conducted a follow-up study showing that over 83% of participants were glad that they had been in the experiment and fewer than 1% of participants were sorry they had been in it. He also had participants examined by an experienced psychiatrist one year later and found no signs that any of the participants had been harmed by the experience. Milgram and others suggested that the ethical uproar might have been a reaction to the unpleasant implications of the findings rather than to the ethicality of the procedures, and some research supports this claim (Schlenker & Forsyth, 1977). Debate regarding this matter continues.

In Burger’s voice-feedback variation, the learner began grunting in response to the 75-volt shock; after ostensibly receiving the 150-volt shock, the learner protested that he wanted to quit, was in pain, and was worried about his heart. Burger found that the percentage of participants willing to proceed past the 150-volt level was no different from what it had been over 40 years earlier. And Milgram found that 79% of those who continued past the 150-volt point were fully obedient through 450 volts. So as best we can surmise, the susceptibility to obedience to legitimate authority has not changed.

Why is this willingness to obey a part of human nature? That’s a bigger question and we’ll address it in the next section.

265

Milgram also noted that a considerable portion of the socialization process involves teaching children to obey first their parents, then teachers, other adults, doctors, police, and a host of other legitimate authority figures within the culture. So we all have been taught to obey legitimate authority figures. And by and large we are rewarded when we do obey and punished, often severely, when we don’t. Obedience is the norm in all cultures and becomes a remarkable and negative phenomenon only when authority figures tell us to do things that end up causing great harm.

Besides the innate predispositions and learning experiences we humans share, Milgram also pointed to some specific factors in the paradigm he created that may have contributed to the levels of obedience in his studies. One is the gradual increase in the severity of the actions in which the participants were commanded to engage. Fifteen volts is barely a noticeable tickle, a 30-volt shock is very tolerable, and so on. Thus, one aspect of the process was its gradual nature. Once the participant delivered 30 volts, why not proceed to 45? Once 45, why not 60? And if one has delivered 330 volts, why resist moving on to 345?

Can you think of any theories we have covered which could explain how actions once taken can increase commitment to further actions along similar lines? Recall self-perception theory (chapter 5; also covered earlier in this chapter), which suggests that we infer our attitudes from our own actions, and cognitive dissonance theory (chapter 6), which proposes that people often shift their attitudes to justify their prior actions. Either or both of these theories could help explain why someone who has delivered 315 volt shocks likely would be okay with delivering 330 volts. For both self-perception and dissonance processes to occur, though, people must have some perception of choice. Although the participants in the Milgram studies did have a choice—indeed, some chose not to continue—it is not clear if they had enough perception of choice for self-perception and dissonance processes to have contributed to their decision.

However, in real life, perception of choice usually is involved when we act. Self-perception and dissonance processes are likely to be involved in phenomena such as the rise of Nazism in Germany. Imagine being a non-Jewish German during the early 1930s. Perhaps a friend coaxes you into going to a Hitler rally. Once you are there, others greet you with “Heil Hitler.” You find yourself following suit. Sometime later, the Nazi propaganda machine exhorts you to boycott Jewish shops. If you support that, why not support the deportation of Jews to concentration camps where they will have to work for the Third Reich? The point is that historically significant atrocities often start with small acts that escalate to more severe ones over time.

266

Another aspect of this paradigm that Milgram noted was that, in the studies with the highest levels of obedience, the participants had to tell the authority figure to his face that they refused to continue. Milgram posited that it is very difficult to defy a legitimate authority figure in this manner. Why? Milgram based his explanation on Erving Goffman’s analysis of self-presentation (1959). Defying legitimate authorities challenges the definition of the situation and disrupts the working consensus by which we all live. The participants had agreed to take part in the study, and they may have viewed their agreement as a contract they needed to honor. Thus, normative pressures worked to keep the participants on track with obeying the authority. Another, broader way to view the difficulty of defying authority is that legitimate authority figures are valued representatives of the prevailing cultural worldview (Solomon, Greenberg, & Pyszczynski, 1991). In fact, the legitimacy of their authoritative position is given to them by the culture. So to defy a scientist running a study, a doctor in an examining room, a police officer who has pulled you over, or a teacher in a classroom is to go against the very worldview on which you predicate your meaningful view of the world and your own self-worth. So for this reason as well, we obey.

267

The existential perspective provides some answers. According to Ernest Becker (1973) and the terror management theory (e.g., Greenberg et al., 2008), new leaders emerge when the prevailing worldview of a culture no longer provides its members with compelling bases of meaning and self-worth. In this context, people need a more secure belief system that provides them with a sense of enduring significance, especially if death-related concerns are heightened by prevailing political or economic factors. In such circumstances, an individual who takes bold action and who very confidently espouses an alternative worldview that does seem to offer a better basis of meaning and self-worth can gain followers. A worldview that portrays the in-group as representing the greater good and as on a heroic mission to vanquish evil is particularly suited to providing such a sense of purpose and enduring significance. A leader who exhibits these attributes—boldness, self-confidence, and a vision that inspires and meets the psychological needs of followers—is known as a charismatic leader.

269