agony. They implored the researcher to let them stop. When he refused,
they went on, but in the process they trembled, they perspired, they
shook, they stammered protests and additional pleas for the victim’s
release. Their fingernails dug into their own flesh; they bit their lips
until they bled; they held their heads in their hands; some fell into fits
of uncontrollable nervous laughter. As one outside observer to the ex-
periment wrote:
I observed a mature and initially poised businessman enter the
laboratory smiling and confident. Within twenty minutes he was
reduced to a twitching, stuttering wreck who was rapidly approach-
ing a point of nervous collapse. He constantly pulled on his earlobe
and twisted his hands. At one point he pushed his fist into his
forehead and muttered: “Oh, God, let’s stop it.” And yet he con-
tinued to respond to every word of the experimenter and obeyed
to the end.^1
In addition to these observations, Milgram has provided even more
convincing evidence for the obedience-to-authority interpretation of
his subjects’ behavior. In a later study, for instance, he had the researcher
and the victim switch scripts so that the researcher told the Teacher to
stop delivering shocks to the victim, while the victim insisted bravely
that the Teacher continue. The result couldn’t have been clearer; 100
percent of the subjects refused to give one additional shock when it was
merely the fellow subject who demanded it. The identical finding ap-
peared in another version of the experiment in which the researcher
and fellow subject switched roles so that it was the researcher who was
strapped into the chair and the fellow subject who ordered the Teacher
to continue—over the protests of the researcher. Again, not one subject
touched another shock lever.
The extreme degree to which subjects in Milgram’s situation were
attentive to the wishes of authority was documented in yet another
variation of the basic study. In this case, Milgram presented the
Teacher with two researchers, who issued contradictory orders; one
ordered the Teacher to terminate the shocks when the victim cried out
for release, while the other maintained that the experiment should go
on. These conflicting instructions reliably produced what may have
been the project’s only humor: In tragicomic befuddlement and with
eyes darting from one researcher to another, subjects would beseech
the pair to agree on a single command they could follow: “Wait, wait.
Which is it going to be? One says stop, one says go. Which is it!?” When
the researchers remained at loggerheads, the subjects tried frantically
to determine who was the bigger boss. Failing this route to obedience
with the authority, every subject finally followed his better instincts and
Robert B. Cialdini Ph.D / 161