Aim: To investigate how readily people would conform to the roles of guard and prisoner in a role-playing exercise that simulated prison life.
Zimbardo (1973) was interested in finding out whether the brutality reported among guards in American prisons was due to the sadistic personalities of the guards or had more to do with the prison environment.
Procedure: Zimbardo used a lab experiment to study conformity.
To study the roles people play in prison situations, Zimbardo converted a basement of the Stanford University psychology building into a mock prison. He advertised for students to play the roles of prisoners and guards for a fortnight. 21 male college students (chosen from 75 volunteers) were screened for psychological normality and paid $15 per day to take part in the experiment.
Participants were randomly assigned to either the role of prisoner or guard in a simulated prison environment. The prison simulation was kept as “real life” as possible. Prisoners were arrested at their own homes, without warning, and taken to the local police station.
Guards were also issued a khaki uniform, together with whistles, handcuffs and dark glasses, to make eye contact with prisoners impossible. No physical violence was permitted. Zimbardo observed the behavior of the prisoners and guards.
Here they were treated like every other criminal. They were fingerprinted, photographed and ‘booked’. Then they were blindfolded and driven to the psychology department of Stanford University, where Zimbardo had had the basement set out as a prison, with barred doors and windows, bare walls and small cells. Here the deindividuation process began.
When the prisoners arrived at the prison they were stripped naked, deloused, had all their personal possessions removed and locked away, and were given prison clothes and bedding. They were issued a uniform, and referred to by their number only. Their clothes comprised a smock with their number written on it, but no underclothes. They also had a tight nylon cap, and a chain around one ankle.
There were 3 guards to the 9 prisoners, taking shifts of eight hours each (the other guards remained on call)
Findings: Within a very short time both guards and prisoners were settling into their new roles, the guards adopting theirs quickly and easily.
Within hours of beginning the experiment some guards began to harass prisoners. They behaved in a brutal and sadistic manner, apparently enjoying it. Other guards joined in, and other prisoners were also tormented.
The prisoners were taunted with insults and petty orders, they were given pointless and boring tasks to accomplish, and they were generally dehumanized.
The prisoners soon adopted prisoner-like behavior too. They talked about prison issues a great deal of the time. They ‘told tales’ on each other to the guards. They started taking the prison rules very seriously, as though they were there for the prisoners’ benefit and infringement would spell disaster for all of them. Some even began siding with the guards against prisoners who did not conform to the rules.
Over the next few days the relationships between the guards and the prisoners changed, with a change in one leading to a change in the other. Remember that the guards were firmly in control and the prisoners were totally dependent on them.
As the prisoners became more dependent, the guards became more derisive towards them. They held the prisoners in contempt and let the prisoners know it. As the guards’ contempt for them grew, the prisoners became more submissive.
As the prisoners became more submissive, the guards became more aggressive and assertive. They demanded ever greater obedience from the prisoners. The prisoners were dependent on the guards for everything so tried to find ways to please the guards, such as telling tales on fellow prisoners.
One prisoner had to be released after 36 hours because of uncontrollable bursts of screaming, crying and anger. His thinking became disorganized and he appeared to be entering the early stages of a deep depression. Within the next few days three others also had to leave after showing signs of emotional disorder that could have had lasting consequences. (These were people who had been pronounced stable and normal a short while before.)
Zimbardo (1973) had intended that the experiment should run for a fortnight, but on the sixth day he closed it down. There was real danger that someone might be physically or mentally damaged if it was allowed to run on. After some time for the researchers to gather their data the subjects were called back for a follow-up, debriefing session.
Conclusion: People will readily conform to the social roles they are expected to play, especially if the roles are as strongly stereotyped as those of the prison guards. The “prison” environment was an important factor in creating the guards’ brutal behavior (none of the participants who acted as guards showed sadistic tendencies before the study). Therefore, the roles that people play can shape their behavior and attitudes.
After the prison experiment was terminated Zimbardo interviewed the participants. Here’s an excerpt:
‘Most of the participants said they had felt involved and committed. The research had felt "real" to them. One guard said, "I was surprised at myself. I made them call each other names and clean the toilets out with their bare hands. I practically considered the prisoners cattle and I kept thinking I had to watch out for them in case they tried something." Another guard said "Acting authoritatively can be fun. Power can be a great pleasure." And another: "... during the inspection I went to Cell Two to mess up a bed which a prisoner had just made and he grabbed me, screaming that he had just made it and that he was not going to let me mess it up. He grabbed me by the throat and although he was laughing I was pretty scared. I lashed out with my stick and hit him on the chin although not very hard, and when I freed myself I became angry."’
Most of the guards found it difficult to believe that they had behaved in the brutalizing ways that they had. Many said they hadn’t known this side of them existed or that they were capable of such things. The prisoners, too, couldn’t believe that they had responded in the submissive, cowering, dependent way they had. Several claimed to be assertive types normally. When asked about the guards, they described the usual three stereotypes that can be found in any prison: some guards were good, some were tough but fair, and some were cruel.
Ethics: The study has received many ethical criticisms, including lack of fully informed consent by participants and the level of humiliation and distress experienced by those who acted as prisoners.
The consent could not be fully informed as Zimbardo himself did not know what would happen in the experiment (it was unpredictable). Also, participants playing the role of prisoners were not protected from psychological and physical harm. For example, one prisoner had to be released after 36 hours because of uncontrollable bursts of screaming, crying and anger.
Haney, C., Banks, W. C., & Zimbardo, P. G. (1973) A study of prisoners and guards in a simulated prison. Naval Research Review, 30, 4-17.