. 52
( 115 .)


It would be a relief if those carrying out acts of destructive obedience were deviant indi-
viduals predisposed to antisocial behavior. Unfortunately, history tells us that those who
perpetrate evil are often quite ordinary. William Calley, who was in command of the
platoon that committed a massacre at the Vietnamese village of My Lai, was ordinary
before and after My Lai. So too was Mohammad Atta, the leader of the 9/11 hijackers.
So was Adolph Eichmann, one of the architects of the Holocaust and the Nazi of¬cer
responsible for the delivery of European Jews to concentration camps in World War II.
Eichmannʼs job was to ensure that the death camps had a steady ¬‚ow of victims.
He secured the railroad cattle cars needed to transport the human cargo. His job was
managerial, bureaucratic; often he had to ¬ght with competing German interests to
get enough boxcars. When the war was over, Eichmann, a most-wanted war criminal,
escaped to Argentina. From 1945 to 1961, he worked as a laborer outside Buenos Aires.
His uneventful existence ended in 1961 when he was captured by Israeli secret agents,
who spirited him to Israel. There he stood trial for crimes against humanity. After a long
trial, Eichmann was found guilty and was later hanged.
Social Psychology

The Israelis constructed a special clear, bulletproof witness box for Eichmann to
appear in during the trial. They were afraid that someone in Israel might decide to mete
out some personal justice. What did the man in the glass booth look like? Eichmann was
a short, bald man whose glasses slipped down his nose now and then. You could walk
past him a hundred times on the street and never notice him. During the trial, Eichmann
portrayed himself as a man anxious to please his superiors, ambitious for advancement.
Killing people was a distasteful but necessary part of his job. Personally, he had no real
hatred of the Jews. He was just following orders.
Philosopher and social critic Hannah Arendt observed Eichmann in the dock. She
was struck by the wide gap between the ordinariness of the man and the brutal deeds for
which he was on trial. In her book, Eichmann in Jerusalem: A Report on the Banality of
Evil (1963), Arendt essentially accepted Eichmannʼs defense. Her analysis of Eichmann
suggested that evil is often very commonplace. Those who carry out acts of destructive
obedience are often ordinary people, rather like you and me.
People were shocked by Eichmann and by Arendtʼs analysis. They had expected
a Nazi war criminal to be the epitome of evil. There was a prevailing belief that evil
Eichmann™s fallacy deeds are done by evil people, a belief referred to as Eichmannʼs fallacy (Brown,
The belief that evil deeds are 1986). Sometimes individuals who perpetrate evil deeds are quite ordinary, as Eichmann
done only by evil people.
apparently was.
As you might expect, not everyone subscribes to the general idea of the banality
of evil. For example, Calder (2003) argues that a person can have an “evil character”
and still have an ordinary appearance and demeanor. However, Calder admits that it is
possible for ordinary individuals to commit acts of evil even in the absence of an evil
character. In an interesting distinction, Calder suggests that some people, such as Adolph
Hitler, carry out evil deeds on their own, without direction from anyone else (autonomous
evil). Calder classi¬es individuals in this category as moral monsters. Moral monsters
like Hitler are singled out for special condemnation because of their active roles in ini-
tiating and directing evil acts (Calder, 2003). Others, such as Adolph Eichmann, carry
out evil at the behest of others (nonautonomous evil). Individuals in this category are
moral idiots. We may be more inclined to label moral monsters as truly evil than moral
idiots. However, it is possible to label the actions of moral idiots as truly evil if those
acts are particularly heinous and show a consistent pattern.
Our discussion of the nature of evil leads us to a central question: Are evil deeds the
product of an evil character (internal attribution), or are they driven more by aspects of
the social situation (external attribution)? This brings us to the main question we shall
consider in the sections to follow: Do evil deeds always lead us back to an evil person?
Although it might make us feel better if the answer to this question were yes, we see in
this chapter that things are not, unfortunately, so simple.

Ultimately, Who Is Responsible for Evil Deeds?
After World War II, the Allies tried many of the high-ranking Nazis who, like Eichmann,
claimed innocence. Their principal defense was to shift responsibility to their superiors:
They were only following orders. More recently, a former East German border guard,
Ingo Heinrich, was brought to trial for his role in preventing East German citizens
from escaping to the west during the height of the cold war. Heinrich, along with his
fellow border guards, had orders to shoot to kill anyone attempting to escape over the
Berlin Wall. Heinrich did just that. But some of his comrades, under the same orders,
shot over the heads of escapees. After the fall of the Berlin Wall and the reuni¬cation
of Germany, Heinrich was arrested and charged with murder. He was eventually con-
victed and sentenced to 3.5 years in prison.
Chapter 7 Conformity, Compliance, and Obedience 259

The cases of Eichmann and Heinrich raise some important issues about responsibil-
ity. Is “I was only following orders” a valid defense? Does it erase personal responsi-
bility? Or should individuals be held accountable for their behavior, even if they were
following orders? On the surface it would appear that Eichmann and Heinrich were
personally responsible for their behavior. However, a deeper examination of author-
ity and its effects on behavior suggests a more complex picture, a picture with many
aspects. These issues and questions served as the catalyst for what are probably the
most famous experiments on obedience.

Milgram™s Experiments on Obedience
How does one test destructive obedience in a laboratory setting? The late Stanley
Milgram devised a simple yet powerful situation. Before we look at it, letʼs consider the
sociohistorical “climate” in the United States at the time. The year was 1962. Vietnam
was but a blip on the back pages of the newspapers. The Kennedy assassinations had
not yet occurred, nor had the murder of Martin Luther King, Jr., Watergate, or the riots
in the streets of Newark, Detroit, and Watts. This was America before the real 1960s
began, still holding on to some of the innocence, however illusory, of the 1950s. This
context is important to consider because it may have in¬‚uenced how people behaved
in Milgramʼs experiments.

The Participantʼs Perspective
Letʼs begin by considering what these experiments looked like from a participantʼs
perspective (Elms, 1972). Imagine you are living in New Haven, Connecticut. One
day you notice an ad in the paper asking for volunteers for an experiment on learning
and memory at nearby Yale University. The researchers are clearly seeking a good rep-
resentation of the general population. The ad piques your curiosity, and you decide to
sign up for the experiment.
When you arrive for the experiment, a young man, Mr. Williams, Dr. Milgramʼs
associate, writes out a check to each of you for $4.50. Williams tells you that little is
known about the impact of punishment on learning, and that is what this experiment is
about. You become a bit concerned when Williams says that one of you will be a learner
and the other will be a teacher. Your fears about getting punished soon evaporate when
you draw lots to see who will be the learner and you draw the role of the teacher.
Preliminaries out of the way, Williams leads you both into a room past an
ominous-looking piece of equipment labeled “Shock Generator, Thorpe ZLB . . .
Output 15 volts”450 volts” (Milgram, 1974). The learner, Mr. Wallace, is told to
sit in a straight“backed metal chair. Williams coolly tells you to help strap Wallaceʼs
arms down to prevent “excessive movement” during the experiment, which you do.
Williams then applies a white paste to Wallaceʼs arms, which he says is electrode
paste “to avoid blisters and burns.” Wallace is now worried, and he asks if there is
any danger. Williams says, “Although the shocks can be extremely painful, they cause
no permanent tissue damage” (Elms, 1972, p. 114).
In front of the learner is a row of switches that he will use to respond to your ques-
tions. Williams tells you that a light panel in the other room will register the learnerʼs
responses. If his answers are correct, you, the teacher, tell him so. If incorrect, you
deliver an electric shock from the shock generator.
Itʼs time to start the experiment. You leave Wallace strapped to the shock genera-
tor and follow Williams into the next room. He places you before a control panel that
has 30 levers, each with a little red light and a big purple light above. The lights have
signs above them reading 15 volts, 30 volts, 45 volts, and so on, up to 450 volts. There
Social Psychology

are also printed descriptions of the shock levels above the labels, reading Slight Shock,
Moderate Shock, Strong Shock, Intense Shock, Extreme Intense Shock, and ¬nally, over
the last few switches, in red, Danger: Severe Shock XXXXX. At this point, you hope
that Wallace is brighter than he looks (Elms, 1972).
Before you begin the experiment, Williams gives you a sample shock of 45 volts,
which gives you a little jolt. Next, you are told that your task is to teach Wallace several
lists of word pairs, such as blue“box, nice“day, wild“duck. You read the entire list
of word pairs and then test him, one pair at a time, by providing the ¬rst word from
each pair.
At ¬rst the test is uneventful; Wallace makes no errors. Then he makes his ¬rst
mistake, and you are required to give him a 15-volt shock. Williams tells you that for
every error after that, you are to increase the shock by 15 volts. On subsequent trials
Wallace makes frequent errors. When you get to 105 volts, you hear Wallace yell through
the wall, “Hey, this really hurts!”
Williams, cool as ever, doesnʼt seem to notice. You certainly do. At 150 volts, the
moaning Walace yells, “Experimenter, get me out of here! I wonʼt be in the experiment
anymore. I refuse to go on!” (Elms, 1972, p. 115). You look at Williams. He says softly
but ¬rmly, “Continue.”
Williams brings you more word-pair lists. You begin to wonder what you and
Wallace have gotten into for $4.50. You are now at 255 volts, Intense Shock. Wallace
screams after every shock. Whenever you ask Williams if you can quit, he tells you to
continue. At 300 volts, you wonder if Wallace is going to die. “But,” you think, “they
wouldnʼt let that happen at Yale . . . or would they?”
“Hey, Mr. Williams,” you say, “whose responsibility is this? What if he dies or is
seriously injured?” Williams does not bat an eye: “Itʼs my responsibility, not yours, just
continue with the experiment.” He reminds you that, as he told you before, the labels
apply to small animals, not humans.
Finally it is over. There are no more shock switches to throw. You are sweaty, uneasy.
Wallace comes in from the other room. He is alive and seems okay. You apologize. He
tells you to forget it, he would have done the same if he had been in your shoes. He
smiles and rubs his sore wrists, everybody shakes hands, and you and Wallace walk
out together.

Predicted Behavior and Results in the Milgram Experiment
How do you think you would behave in Milgramʼs experiment? Most people think they
would refuse to obey the experimenterʼs orders. Milgram was interested in this ques-
tion, so he asked a wide range of individuals, both expert (psychiatrists) and nonexpert
(college students and noncollege adults), how they thought participants would behave
in this situation. They all predicted that they would break off the experiment, defying
the experimenter. The psychiatrists predicted that participants would break off when the
learner began to protest, at the 150-volt level. So, if you believe that you would defy the
experimenter and refuse to in¬‚ict pain on another person, you are not alone.
Another study, independent from Milgramʼs, investigated the role of several vari-
ables in predicting obedience in a Milgram-type experiment (Miller, Gillen, Schenker,
& Radlove, 1974). Miller et al. provided participants with verbal descriptions and a slide
show depicting Milgramʼs experiment. Miller et al. looked at two classes of variables:
Perceiver variables (gender and normative information [some participants were provided
with the results of Milgramʼs baseline experiment and others were not]) and stimulus
person variables (gender and physical attractiveness). The dependent variable was the
Chapter 7 Conformity, Compliance, and Obedience 261

predicted shock level that would be administered in the situation. The results showed that
participants believed that males would administer higher shock levels than females and
that unattractive individuals would administer higher shock levels than attractive indi-
viduals. The latter ¬nding was true mainly for female shock administrators. Interestingly,
males showed greater consistency between predictions of another personʼs obedience
behavior than did females. Female participants believed they themselves would admin-
ister lower levels of shock than would another person in the same situation.
The underlying assumption of these predictions is that individual characteristics
will be more powerful determinants of behavior than situational factors. The predic-
tions of Milgramʼs participants re¬‚ect the notion that moral knowledge predicts moral
behavior; in other words, if you know what is right, you will do it. However, the results
of Milgramʼs ¬rst “baseline” experiment (in which there was no feedback from the
victim) donʼt support these rosy predictions. A majority of participants (65%) went all
the way to 450 volts. In fact, the average shock level delivered by the participants in
this ¬rst experiment was 405 volts! We can infer from this result that under the right
circumstances, most of us probably also would go all the way to 450 volts.
Of course, no electric shock was ever given to Wallace, who was, in fact, a profes-
sional actor, playing out a script. However, Milgramʼs participants did not know that
the entire situation was contrived.

Situational Determinants of Obedience
Milgram himself was surprised at the levels of obedience observed in his ¬rst experi-
ment. He and others conducted several additional experiments investigating the situ-
ational factors that in¬‚uence levels of obedience. In the following sections, we explore
some of these situational factors.

Proximity of the Victim In his ¬rst series of experiments, Milgram tested the limits of
obedience by varying the proximity, or closeness, between the teacher and the learner
(victim). The conditions were:

1. Remote victim. The teacher and the learner were in separate rooms. There was
no feedback from the victim to the teacher. That is, Wallace didnʼt speak, moan,
or scream.
2. Voice feedback. The teacher and the learner were in separate rooms, but Wallace
began to protest the shocks as they became more intense. This is the experiment
just described. In one version of the voice-feedback condition, Wallace makes
it clear that he has a heart condition. After receiving 330 volts he screams,
“Let me out of here. Let me out of here. My heart is bothering me” (Milgram,
1974, p. 55).
3. Proximity. The teacher and the learner were in the same room, sitting only a few
feet apart.
4. Touch proximity. The teacher and the learner were in the same room, but the
learner received the shock only if his hand was placed on a shock plate. At one
point the learner refused to keep his hand on the plate. The teacher was told
to hold the learnerʼs hand down while delivering the shock. The teacher often
had to hand-wrestle the victim to be sure the hand was properly placed on the
shock plate.
Social Psychology

These four conditions decrease the physical distance between the teacher and the learner.
Milgram found that reducing the distance between the teacher and the learner affected
the level of obedience (Figure 7.7). In the remote-victim condition, 65% of the partici-
pants obeyed the experimenter and went all the way to 450 volts (the average shock
intensity was 405 volts). As you can see from Figure 7.7, obedience was not substan-
tially reduced in the voice-feedback condition. In this condition, obedience dropped
only 2.5%, to 62.5%, with an average shock intensity of 368 volts.
Thus, verbal feedback from the learner, even when he indicates his heart is bother-
ing him, is not terribly effective in reducing obedience. Signi¬cant drops in the rates of
obedience were observed when the distance between the teacher and the learner was
decreased further. In the proximity condition, where the teacher and the learner were
in the same room and only a few feet apart, 40% of the participants went to 450 volts
(with an average shock intensity of 312 volts). Finally, when the teacher was required to
hold the learnerʼs hand on the shock plate in the touch-proximity condition, only 30%
obeyed and went to 450 volts (the average shock intensity was 269 volts).
Why does decreasing the distance between the teacher and the learner affect obedi-
ence so dramatically? Milgram (1974) offered several explanations. First, decreasing the
distance between the teacher and the learner increases empathic cues from the learner,
cues about his suffering, such as screaming or banging on the wall. In the remote-victim
condition, the teacher receives no feedback from the learner. There is no way for the
teacher to assess the level of suffering of the learner, making it easier on the teacherʼs
conscience to in¬‚ict harm. In the feedback conditions, however, the suffering of the
learner is undeniable. The teacher has a greater opportunity to observe the learner in
voice-feedback, proximity, and touch conditions than in the remote-victim condition.
It is interesting to note, however, that even in the touch-proximity condition, a sizable
percentage of participants (39%) were willing to fully obey the experimenter. It is
apparent that there are some among us who are willing to discount empathic cues and

Figure 7.7 The effect of
moving the learner closer to
the teacher. In the remote
condition, obedience was
highest. Adding voice
feedback did not reduce
obedience signi¬cantly.
It was only when the
learner and teacher were
in the same room that
obedience dropped. The
lowest level of obedience
occurred when the teacher
was required to touch


. 52
( 115 .)