The research team used local newspaper ads to recruit 18 male students. Participants were told they’d be paid $15/hour to participate in a 2-week study about the psychological effects of incarceration.
Half of the 18 students were randomly assigned the role of prison guards, and the other half were assigned the role of prisoners.
The “prisoners” were notified of their roles when Palo Alto police showed up at their homes to arrest, handcuff and search them before taking them to the precinct to book them on charges of armed robbery or burglary.
The prison guards were given no training. They were asked to follow three directives: don’t let anyone escape, maintain law and order and avoid physical violence.
The 9 prisoners stayed in the prison around the clock, while the 9 guards rotated in three 8-hour shifts — so there were always 3 guards overseeing the 9 prisoners.
Zombardo’s research team was wholly unprepared for how swiftly and thoroughly both groups took to their roles. By Day Two, when the prisoners started to rebel, the prison guards didn’t hesitate to increase their punishments and apply more force.
By Day Six, things were so out of hand, they decided to shut the experiment down eight days before schedule.
If you’re interested in learning more about why and how these 18 Stanford students so easily assumed these uncharacteristic personas with such conviction, it’s really worth exploring the narrative timeline of the Stanford Prison Experiment.
An intercom system allowed us to secretly bug the cells to monitor what the prisoners discussed, and also to make public announcements to the prisoners. There were no windows or clocks to judge the passage of time, which later resulted in some time-distorting experiences.
Because the first day passed without incident,
The guards were very much angered and frustrated because the prisoners also began to taunt and curse them. And now the problem was, what were we going to do about this rebellion?
When the morning shift of guards came on, they became upset at the night shift who, they felt, must have been too lenient. The guards had to handle the rebellion themselves, and what they did was fascinating for the staff to behold.
They got a fire extinguisher which shot a stream of skin-chilling carbon dioxide, and they forced the prisoners away from the doors.
The guards broke into each cell, stripped the prisoners naked, took the beds out, forced the ringleaders of the prisoner rebellion into solitary confinement, and generally began to harass and intimidate the prisoners. One of the three cells was designated as a “privilege cell.” The three prisoners least involved in the rebellion were given special privileges. They got their uniforms back, got their beds back, and were allowed to wash and brush their teeth. The others were not. Privileged prisoners also got to eat special food in the presence of the other prisoners who had temporarily lost the privilege of eating. The effect was to break the solidarity among prisoners.
After half a day of this treatment, the guards then took some of these “good” prisoners and put them into the “bad” cells, and took some of the “bad” prisoners and put them into the “good” cell, thoroughly confusing all the prisoners. Some of the prisoners who were the ringleaders now thought that the prisoners from the privileged cell must be informers, and suddenly, the prisoners became distrustful of each other.
“In only a few days, our guards became sadistic and our prisoners became depressed and showed signs of extreme stress.”
~Phil Zimbardo, Stanford Psychology Professor
Our ex-convict consultants later informed us that a similar tactic is used by real guards in real prisons to break prisoner alliances. For example, racism is used to pit Blacks, Chicanos, and Anglos against each other. In fact, in a real prison the greatest threat to any prisoner’s life comes from fellow prisoners. By dividing and conquering in this way, guards promote aggression among inmates, thereby deflecting it from themselves.
The prisoners’ rebellion also played an important role in producing greater solidarity among the guards. Now, suddenly, it was no longer just an experiment, no longer a simple simulation. Instead, the guards saw the prisoners as troublemakers who were out to get them, who might really cause them some harm. In response to this threat, the guards began stepping up their control, surveillance, and aggression.
Every aspect of the prisoners’ behavior fell under the total and arbitrary control of the guards. Even going to the toilet became a privilege which a guard could grant or deny at his whim. Indeed, after the nightly 10:00 P.M. lights out “lock-up,” prisoners were often forced to urinate or defecate in a bucket that was left in their cell. On occasion the guards would not allow prisoners to empty these buckets, and soon the prison began to smell of urine and feces — further adding to the degrading quality of the environment.
Less than 36 hours into the experiment, Prisoner #8612 began suffering from acute emotional disturbance, disorganized thinking, uncontrollable crying, and rage. In spite of all of this, we had already come to think so much like prison authorities that we thought he was trying to “con” us — to fool us into releasing him.
During the next count, Prisoner #8612 told other prisoners, “You can’t leave. You can’t quit.” That sent a chilling message and heightened their sense of really being imprisoned. #8612 then began to act “crazy,” to scream, to curse, to go into a rage that seemed out of control. It took quite a while before we became convinced that he was really suffering and that we had to release him.
NOTE: Visit the Stanford Experiment website to read and watch what happened over the next four days. Prison guards react harshly to rumors of an escape plot, two prisoners become unhinged and have to leave the study, parents visit their children and become part of the story and a prisoner stages a hunger strike.
There were three types of guards. First, there were tough but fair guards who followed prison rules. Second, there were “good guys” who did little favors for the prisoners and never punished them. And finally, about a third of the guards were hostile, arbitrary, and inventive in their forms of prisoner humiliation. These guards appeared to thoroughly enjoy the power they wielded, yet none of our preliminary personality tests were able to predict this behavior. The only link between personality and prison behavior was a finding that prisoners with a high degree of authoritarianism endured our authoritarian prison environment longer than did other prisoners.
Even going to the toilet became a privilege which a guard could grant or deny at his whim. Indeed, after the nightly 10:00 P.M. lights out “lock-up,” prisoners were often forced to urinate or defecate in a bucket that was left in their cell. On occasion the guards would not allow prisoners to empty these buckets, and soon the prison began to smell of urine and feces — further adding to the degrading quality of the environment.
Prisoners coped with their feelings of frustration and powerlessness in a variety of ways. At first, some prisoners rebelled or fought with the guards. Four prisoners reacted by breaking down emotionally as a way to escape the situation. One prisoner developed a psychosomatic rash over his entire body when he learned that his parole request had been turned down. Others tried to cope by being good prisoners, doing everything the guards wanted them to do. One of them was even nicknamed “Sarge,” because he was so military-like in executing all commands.
WHY ZIMBARDO ENDED THE STUDY EARLY
By the end [Day Six], the prisoners were disintegrated, both as a group and as individuals. There was no longer any group unity; just a bunch of isolated individuals hanging on, much like prisoners of war or hospitalized mental patients. The guards had won total control of the prison, and they commanded the blind obedience of each prisoner.
At this point it became clear that we had to end the study. We had created an overwhelmingly powerful situation — a situation in which prisoners were withdrawing and behaving in pathological ways, and in which some of the guards were behaving sadistically. Even the “good” guards felt helpless to intervene, and none of the guards quit while the study was in progress. Indeed, it should be noted that no guard ever came late for his shift, called in sick, left early, or demanded extra pay for overtime work.
I ended the study prematurely for two reasons. First, we had learned through videotapes that the guards were escalating their abuse of prisoners in the middle of the night when they thought no researchers were watching and the experiment was “off.” Their boredom had driven them to even more pornographic and degrading abuse of the prisoners.
Second, Christina Maslach, a recent Stanford Ph.D. brought in to conduct interviews with the guards and prisoners, strongly objected when she saw our prisoners being marched on a toilet run, bags over their heads, legs chained together, hands on each other’s shoulders. Filled with outrage, she said, “It’s terrible what you are doing to these boys!” Out of 50 or more outsiders who had seen our prison, she was the only one who ever questioned its morality. Once she countered the power of the situation, however, it became clear that the study should be ended.
And so, after only six days, our planned two-week prison simulation was called off.
What do you think?