The Zimbardo experiment
Rebecca Lann
by Philip G. Zimbardo -- By the 1970s, psychologists had done a series
of studies establishing the social power of groups. They showed, for
example, that groups of strangers could persuade people to believe
statements that were obviously false. Psychologists had also found
that research participants were often willing to obey authority
figures even when doing so violated their personal beliefs.
The Yale studies by Stanley Milgram in 1963 demonstrated that a
majority of ordinary citizens would continually shock an innocent man,
even up to near-lethal levels, if commanded to do so by someone acting
as an authority. The "authority" figure in this case was merely a high-
school biology teacher who wore a lab coat and acted in an official
manner. The majority of people shocked their victims over and over
again despite increasingly desperate pleas to stop.
In my own work, I wanted to explore the fictional notion from William
Golding's Lord of the Flies about the power of anonymity to unleash
violent behavior. In one experiment from 1969, female students who
were made to feel anonymous and given permission for aggression became
significantly more hostile than students with their identities intact.
Those and a host of other social- psychological studies were showing
that human nature was more pliable than previously imagined and more
responsive to situational pressures than we cared to acknowledge. In
sum, these studies challenged the sacrosanct view that inner
determinants of behavior?? personality traits, morality, and religious
upbringing?? directed good people down righteous paths.
Missing from the body of social-science research at the time was the
direct confrontation of good versus evil, of good people pitted
against the forces inherent in bad situations. It was evident from
everyday life that smart people made dumb decisions when they were
engaged in mindless groupthink, as in the disastrous Bay of Pigs
invasion by the smart guys in President John F. Kennedy's cabinet. It
was also clear that smart people surrounding President Richard M.
Nixon, like Henry A. Kissinger and Robert S. McNamara, escalated the
Vietnam War when they knew, and later admitted, it was not winnable.
They were caught up in the mental constraints of cognitive
dissonance?? the discomfort from holding two conflicting thoughts??
and were unable to cut bait even though it was the only rational
strategy to save lives and face. Those examples, however, with their
different personalities, political agendas, and motives, complicated
any simple conceptual attempt to understand what went wrong in these
situations. I decided that what was needed was to create a situation
in a controlled experimental setting in which we could array on one
side a host of variables, such as role-playing, coercive rules, power
differentials, anonymity, group dynamics, and dehumanization. On the
other side, we lined up a collection of the "best and brightest" of
young college men in collective opposition to the might of a dominant
system. Thus in 1971 was born the Stanford prison experiment, more
akin to Greek drama than to university psychology study. I wanted to
know who wins?? good people or an evil situation?? when they were
brought into direct confrontation.
First we established that all 24 participants were physically and
mentally healthy, with no history of crime or violence, so as to be
sure that initially they were all "good apples." They were paid $15 a
day to participate. Each of the student volunteers was randomly
assigned to play the role of prisoner or guard in a setting designed
to convey a sense of the psychology of imprisonment (in actuality, a
mock prison set up in the basement of the Stanford psychology
department). Dramatic realism infused the study. Palo Alto police
agreed to "arrest" the prisoners and book them, and once at the
prison, they were given identity numbers, stripped naked, and
deloused. The prisoners wore large smocks with no underclothes and
lived in the prison 24/7 for a planned two weeks; three sets of guards
each patrolled eight-hour shifts. Throughout the experiment, I served
as the prison "superintendent," assisted by two graduate students.
Initially nothing much happened as the students awkwardly tried out
their assigned roles in their new uniforms. However, all that changed
suddenly on the morning of the second day following a rebellion, when
the prisoners barricaded themselves inside the cells by putting their
beds against the door. Suddenly the guards perceived the prisoners as
"dangerous"; they had to be dealt with harshly to demonstrate who was
boss and who was powerless. At first, guard abuses were retaliation
for taunts and disobedience. Over time, the guards became ever more
abusive, and some even delighted in sadistically tormenting their
prisoners. Though physical punishment was restricted, the guards on
each shift were free to make up their own rules, and they invented a
variety of psychological tactics to demonstrate their dominance over
their powerless charges.
Nakedness was a common punishment, as was placing prisoners' heads in
nylon stocking caps (to simulate shaved heads); chaining their legs;
repeatedly waking them throughout the night for hourlong counts; and
forcing them into humiliating "fun and games" activities. Let's go
beyond those generalizations to review some of the actual behaviors
that were enacted in the prison simulation. They are a lesson in
"creative evil," in how certain social settings can transform
intelligent young men into perpetrators of psychological abuse.
[Prison Log, Night 5]
The prisoners, who have not broken down emotionally under the
incessant stress the guards have been subjecting them to since their
aborted rebellion on Day 2, wearily line up against the wall to recite
their ID numbers and to demonstrate that they remember all 17 prisoner
rules of engagement. It is the 1 a.m. count, the last one of the night
before the morning shift comes on at 2 a.m. No matter how well the
prisoners do, one of them gets singled out for punishment. They are
yelled at, cursed out, and made to say abusive things to each other.
"Tell him he's a prick," yells one guard. And each prisoner says that
to the next guy in line. Then the sexual harassment that had started
to bubble up the night before resumes as the testosterone flows freely
in every direction.
"See that hole in the ground? Now do 25 push-ups [expletive] that
hole! You hear me!" One after another, the prisoners obey like
automatons as the guard shoves them down. After a brief consultation,
our toughest guard (nicknamed "John Wayne" by the prisoners) and his
sidekick devise a new sexual game. "OK, now pay attention. You three
are going to be female camels. Get over here and bend over, touching
your hands to the floor." When they do, their naked butts are exposed
because they have no underwear beneath their smocks. John Wayne
continues with obvious glee, "Now you two, you're male camels. Stand
behind the female camels and hump them." The guards all giggle at this
double-entendre. Although their bodies never touch, the helpless
prisoners begin to simulate sodomy by making thrusting motions. They
are then dismissed back to their cells to get an hour of sleep before
the next shift comes on, and the abuse continues.
By Day 5, five of the student prisoners have to be released early
because of extreme stress. (Recall that each of them was physically
healthy and psychologically stable less than a week before.) Most of
those who remain adopt a zombielike attitude and posture, totally
obedient to escalating guard demands.
[Terminating the Torment]
I was forced to terminate the projected two-week-long study after only
six days because it was running out of control. Dozens of people had
come down to our "little shop of horrors," seen some of the abuse or
its effects, and said nothing. A prison chaplain, parents, and friends
had visited the prisoners, and psychologists and others on the parole
board saw a realistic prison simulation, an experiment in action, but
did not challenge me to stop it. The one exception erupted just before
the time of the prison-log notation on Night 5.
About halfway through the study, I had invited some psychologists who
knew little about the experiment to interview the staff and
participants, to get an outsiders' evaluation of how it was going. A
former doctoral student of mine, Christina Maslach, a new assistant
professor at the University of California at Berkeley, came down late
Thursday night to have dinner with me. We had started dating recently
and were becoming romantically involved. When she saw the prisoners
lined up with bags over their heads, their legs chained, and guards
shouting abuses at them while herding them to the toilet, she got
upset and refused my suggestion to observe what was happening in this
"crucible of human nature."
Instead she ran out of the basement, and I followed, berating her for
being overly sensitive and not realizing the important lessons taking
place here. "It is terrible what YOU are doing to those boys!" she
yelled at me. Christina made evident in that one statement that human
beings were suffering, not prisoners, not experimental subjects, not
paid volunteers. And further, I was the one who was personally
responsible for the horrors she had witnessed (and which she assumed
were even worse when no outsider was looking). She also made clear
that if this person I had become?? the heartless superintendent of the
Stanford prison?? was the real me, not the caring, generous person she
had come to like, she wanted nothing more to do with me.
That powerful jolt of reality snapped me back to my senses. I agreed
that we had gone too far, that whatever was to be learned about
situational power was already indelibly etched on our videos, data
logs, and minds; there was no need to continue. I too had been
transformed by my role in that situation to become a person that under
any other circumstances I detest?? an uncaring, authoritarian boss
man. In retrospect, I believe that the main reason I did not end the
study sooner resulted from the conflict created in me by my dual roles
as principal investigator, and thus guardian of the research ethics of
the experiment, and as the prison superintendent, eager to maintain
the stability of my prison at all costs. I now realize that there
should have been someone with authority above mine, someone in charge
of oversight of the experiment, who surely would have blown the
whistle earlier.
By the time Christina intervened, it was the middle of the night, so I
had to make plans to terminate the next morning. The released
prisoners and guards had to be called back and many logistics handled
before I could say, "The Stanford prison experiment is officially
closed." When I went back down to the basement, I witnessed the final
scene of depravity, the "camel humping" episode. I was so glad that it
would be the last such abuse I would see or be responsible for.
[Good Apples in Bad Barrels and Bad Barrel Makers]
The situational forces in that "bad barrel" had overwhelmed the
goodness of most of those infected by their viral power. It is hard to
imagine how a seeming game of "cops and robbers" played by college
kids, with a few academics (our research team) watching, could have
descended into what became a hellhole for many in that basement. How
could a mock prison, an experimental simulation, become "a prison run
by psychologists, not by the state," in the words of one suffering
prisoner? How is it possible for "good personalities" to be so
dominated by a "bad situation"? You had to be there to believe that
human character could be so swiftly transformed in a matter of daysnot
only the traits of the students, but of me, a well-seasoned adult.
Most of the visitors to our prison also fell under the spell. For
example, individual sets of parents observing their son's haggard
appearance after a few days of hard labor and long nights of disrupted
sleep said they "did not want to make trouble" by taking their kid
home or challenging the system. Instead they obeyed our authority and
let some of their sons experience full-blown emotional meltdowns later
on. We had created a dominating behavioral context whose power
insidiously frayed the seemingly impervious values of compassion, fair
play, and belief in a just world.
The situation won; humanity lost. Out the window went the moral
upbringings of these young men, as well as their middle-class
civility. Power ruled, and unrestrained power became an aphrodisiac.
Power without surveillance by higher authorities was a poisoned
chalice that transformed character in unpredictable directions. I
believe that most of us tend to be fascinated with evil not because of
its consequences but because evil is a demonstration of power and
domination over others.
[Current Relevance]
Such research is now in an ethical time capsule, since institutional
review boards will not allow social scientists to repeat it (although
experiments like it have been replicated on several TV shows and in
artistic renditions). Nevertheless, the Stanford prison experiment is
now more popular then ever in its 36-year history. A Google search of
"experiment" reveals it to be fourth among some 132 million hits, and
sixth among some 127 million hits on "prison." Some of this recent
interest comes from the apparent similarities of the experiment's
abuses with the images of depravity in Iraq's Abu Ghraib prison?? of
nakedness, bagged heads, and sexual humiliation.
Among the dozen investigations of the Abu Ghraib abuses, the one
chaired by James R. Schlesinger, the former secretary of defense,
boldly proclaims that the landmark Stanford study "provides a
cautionary tale for all military detention operations." In contrasting
the relatively benign environment of the Stanford prison experiment,
the report makes evident that "in military detention operations,
soldiers work under stressful combat conditions that are far from
benign." The implication is that those combat conditions might be
expected to generate even more extreme abuses of power than were
observed in our mock prison experiment.
However, the Schlesinger report notes that military leaders did not
heed that earlier warning in any way. They should have?? a
psychological perspective is essential to understanding the
transformation of human character in response to special situational
forces. "The potential for abusive treatment of detainees during the
Global War on Terrorism was entirely predictable based on a
fundamental understanding of the principles of social psychology
coupled with an awareness of numerous known environmental risk
factors," the report says. "Findings from the field of social
psychology suggest that the conditions of war and the dynamics of
detainee operations carry inherent risks for human mistreatment, and
therefore must be approached with great caution and careful planning
and training." (Unfortunately this vital conclusion is buried in an
appendix.) The Stanford prison experiment is but one of a host of
studies in psychology that reveal the extent to which our behavior can
be transformed from its usual set point to deviate in unimaginable
ways, even to readily accepting a dehumanized conception of others, as
"animals," and to accepting spurious rationales for why pain will be
good for them.
The implications of this research for law are considerable, as legal
scholars are beginning to recognize. The criminal-justice system, for
instance, focuses primarily on individual defendants and their "state
of mind" and largely ignores situational forces. The Model Penal Code
states: "A person is not guilty of an offense unless his liability is
based on conduct that includes a voluntary act or the omission to
perform an act of which he is physically capable." As my own
experiment revealed, and as a great deal of social-psychological
research before and since has confirmed, we humans exaggerate the
extent to which our actions are voluntary and rationally chosen?? or,
put differently, we all understate the power of the situation. My
claim is not that individuals are incapable of criminal culpability;
rather, it is that, like the horrible behavior brought out by my
experiment in good, normal young men, the situation and the system
creating it also must share in the responsibility for illegal and
immoral behavior.
If the goals of the criminal system are simply to blame and punish
individual perpetrators?? to get our pound of flesh?? then focusing
almost exclusively on the individual defendant makes sense. If,
however, the goal is actually to reduce the behavior that we now call
"criminal" (and its resultant suffering), and to assign punishments
that correspond with culpability, then the criminal-justice system is
obligated, much as I was in the Stanford prison experiment, to
confront the situation and our role in creating and perpetuating it.
It is clear to most reasonable observers that the social experiment of
imprisoning society's criminals for long terms is a failure on
virtually all levels. By recognizing the situational determinants of
behavior, we can move to a more productive public-health model of
prevention and intervention, and away from the individualistic medical
and religious "sin" model that has never worked since its inception
during the Inquisition.
The critical message then is to be sensitive about our vulnerability
to subtle but powerful situational forces and, by such awareness, be
more able to overcome those forces. Group pressures, authority
symbols, dehumanization of others, imposed anonymity, dominant
ideologies that enable spurious ends to justify immoral means, lack of
surveillance, and other situational forces can work to transform even
some of the best of us into Mr. Hyde monsters, without the benefit of
Dr. Jekyll's chemical elixir. We must be more aware of how situational
variables can influence our behavior. Further, we must also be aware
that veiled behind the power of the situation is the greater power of
the system, which creates and maintains complicity at the highest
military and governmental levelswith evil-inducing situations, like
those at Abu Ghraib and Guant?namo Bay prisons.
~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
Not for commercial use. Solely to be fairly used for the educational
purposes of research and open discussion. Content and/or links do not
necessarily reflect the views of the poster.
~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~