When I was working on the first edition of the textbook, I read carefully the original report on the experiment by Zimbardo and his colleagues (the same report that Whitbourne summarizes), and I also read a methodological critique of it by Ali Banuazizi and Siamak Movahedi.[2,3] After reading both, and thinking carefully about them, I decided that I could not in good conscience present the experiment with its usual interpretation. I decided to omit it completely (a decision I made also concerning many other research studies that in my mind were poorly conceived and improperly interpreted). In the years since then, I have found no reason to change my mind and some new evidence that validates my original decision.
I’ll present briefly here the standard interpretation of Zimbardo’s experiment—the interpretation given by Whitbourne, by Zimbardo himself, and by most psychology textbook authors. Then I’ll summarize the critique.
The Standard Story
In order to gain insight about the behavior of prisoners and guards in real prisons, Zimbardo and his colleagues constructed a simulated prison in the basement of the psychology building at Stanford University. Then they recruited 21 psychologically healthy male college students and randomly assigned 10 of them to be prisoners and the other 11 to be guards. The prisoners were to be held captive in the simulated prison around the clock for two weeks, and the guards were to serve duty in the prison on eight-hour daily shifts, so there were always at least 3 of them in the prison at any given time. For this, they would be paid $15 a day (a decent amount for students at that time, in 1971).
In order to simulate conditions of real imprisonment, the researchers arranged for the real Palo Alto city police to pick them up at their homes at a designated time, charge them, handcuff and search them, take them to the police station, fingerprint them, place them in a detention cell, and then take them blindfolded to the mock prison at Stanford. Once there, “each prisoner was stripped, sprayed with a delousing preparation (a deodorant spray) and made to stand alone naked for a while in the cell yard. Then each prisoner was given a uniform consisting of a loose smock with an identifying number on the front and back. A chain and lock were placed around one ankle, and their hair was covered with a nylon stocking made into a cap, to simulate having their hair shaved off. Prior to the experiment, the guards were given instructions that I will refer to later as part of the critique. Throughout the experiment, the guards wore reflective sunglasses, so the prisoners couldn’t see their eyes.
The results, in brief, were that the guards behaved toward the prisoners in “negative, hostile, affrontive and dehumanizing” ways [2, p 80]. The prisoners behaved alternately in rebellious and passive ways. At one point, on the second day, they ripped off their clothing and identification numbers while shouting curses at the guards. Later, five of the prisoners reacted with such extreme emotions that they were removed from the study before the end of five days. By the end of the sixth day the behavior of the guards and prisoners had spiraled to the point that Zimbardo decided to end the experiment early.
The standard interpretation, somewhat simplified, is this. The conditions of a prison, where one group has power over another and the powerless group are stripped of their individual identities, creates extreme, maladaptive responses that are characteristic of the responses often seen in real prisons. Those in power become abusive, and those subject to that power become immature, passive, and rebellious. These effects do not have to do with differences in original personality (because in the experiment the subjects were randomly assigned to roles), but result from the situation in which people find themselves. In the original paper and in subsequent writings, Zimbardo has repeatedly emphasized his view that this experiment reveals much that is significant in understanding what happens in real prisons.
In a nutshell, here’s the criticism, somewhat simplified. Twenty-one boys (OK, young men) are asked to play a game of prisoners and guards. It’s 1971. There have recently been many news reports about prison riots and the brutality of guards. So, in this game, what are these young men supposed to do? Are they supposed to sit around talking pleasantly with one another about sports, girlfriends, movies, and such? No, of course not. This is a study of prisoners and guards, so their job clearly is to act like prisoners and guards—or, more accurately, to act out their stereotyped views of what prisoners and guards do. Surely, Professor Zimbardo, who is right there watching them (as the Prison Superintendent) would be disappointed if, instead, they had just sat around chatting pleasantly and having tea. Much research has shown that participants in psychological experiments are highly motivated to do what they believe the researchers want them to do. Any characteristics of an experiment that let research participants guess how the experimenters expect or want them to behave are referred to as demand characteristics. In any valid experiment it is essential to eliminate or at least minimize demand characteristics. In this experiment, the demands were everywhere.
In order to assess the degree to which participants in the experiment could guess what Zimbardo expected to happen, Banuazizi and Mohavedi presented some of the details of the experimental procedure to a large sample of college students who had not heard of the experiment and asked them to write down what they thought the researchers wanted to prove and to describe how the guards and prisoners were likely to behave. The great majority guessed the results. In various words, they said that the purpose of the experiment was to prove that normal people placed into the position of prisoner or guard would act like real prisoners and guards, and they predicted that the guards would act in hostile, domineering ways and the prisoners would react in either passive or defiant ways or both.
Subsequent revelations about the experiment—published since the first edition of my textbook—reveal that the guards didn’t even have to guess how they were supposed to behave; they were largely told how by Zimbardo and his associates. In his relatively recent book, The Lucifer Effect [4, p 55] Zimbardo describes in the following terms what he told the guards at the outset of the study:
"We cannot physically abuse or torture them," I said. "We can create boredom. We can create a sense of frustration. We can create fear in them, to some degree. We can create a notion of the arbitrariness that governs their lives, which are totally controlled by us, by the system, by you, me, [Warden] Jaffe. They'll have no privacy at all, there will be constant surveillance -- nothing they do will go unobserved. They will have no freedom of action. They will be able to do nothing and say nothing that we don't permit. We're going to take away their individuality in various ways. They're going to be wearing uniforms, and at no time will anybody call them by name; they will have numbers and be called only by their numbers. In general, what all this should create in them is a sense of powerlessness. We have total power in the situation. They have none. ..."[5, see also here,]
Is this not an overt invitation to be abusive in all sorts of psychological ways? And, when the guards did behave in these ways and escalated that behavior, with Zimbardo watching and apparently (by his silence) approving, would that not have confirmed in the subjects’ minds that they were behaving as they should? They were doing this all for the sake of an experiment, for the good of science, and apparently they were doing the right thing—so they continued, and did even more of it, until the experiment was stopped. They may also have been motivated by the same ideological passion that motivated Zimbardo to conduct the experiment in the first place--to prove to the world that prison guards are abusive because of the situation they are in, or to prove that prisoners are hurt by such abuse.
What would have happened if Zimbardo had said to the guards, at the outset, that the purpose of the experiment was to prove that it is possible to be both a guard and a decent human being, or in some way implied that the goal was to prove that guards can be kind? I bet the results would have been entirely different. The “situation” would have been the same, but the “demand” would have been different and the results would have been different.
As part of their preparation for the experiment, Zimbardo and his colleagues consulted with an ex convict named Carlo Prescott, an African American who had served 17 years in San Quentin for attempted murder and who had spoken before Congress on issues of prison reform. Many years later, upon reading about the various ways the experiment was being used (misused, according to Prescott) by Zimbardo and others to explain real prison atrocities, Prescott wrote an article for the Stanford Daily entitled “The Lie of the Stanford Prison Experiment” (reprinted here).
In the article, Prescott expressed great regret for his involvement in the study and said that it was he, not the guards in the mock prison, who came up with the ways of psychologically humiliating and harassing the prisoners. Here are his words:
“…ideas such as bags being placed over the heads of prisoners, inmates being bound together with chains and buckets being used in place of toilets in their cells were all experiences of mine at the old “Spanish Jail” section of San Quentin and which I dutifully shared with the Stanford Prison Experiment braintrust months before the experiment started. To allege that all these carefully tested, psychologically solid, upper-middle-class Caucasian “guards” dreamed this up on their own is absurd. How can Zimbardo … express horror at the behavior of the “guards” when they were merely doing what Zimbardo and others, myself included, encouraged them to do at the outset or frankly established as ground rules? At the time, I had hoped that I would help create a valid, intellectually honest indictment of the prison system. In hindsight, I blew it. I became an unwitting accomplice to a theatrical exercise that conveniently absolves all comers of personal responsibility for their abominable moral choices.” 
In fairness, I must add that in their original article Zimbardo and his colleagues discussed the possibility that demand characteristics may have played a role in the participants’ initial behavior in the mock prison, but they contended that the situation quickly became a new reality for the subjects and then they were responding to the immediate situation. There may be some truth to that, but the degree to which it is true is very much debatable.
My own guess is that the behavior of the guards was largely if not entirely the result of their doing what they were told to do and what they believed they were supposed to do. The behavior of the prisoners in the first day or two, when they were pretending to riot and pretending to plot escapes, was probably also playacting of stereotyped concepts of what prisoners do. But their subsequent wearing down, passivity, and apparently genuine desire to get out of the prison may very well have been a direct response to what the guards were doing to them (coupled, I imagine, with their lack of sleep--the guards were on shifts but they were not).
From this point of view, there is nothing in the results of the study that should surprise us. Once we say that the guards humiliated and oppressed the prisoners because they were told to do that and felt it was important to do it for the sake of science, and once we say that the prisoners initially “rioted” because they felt that this is what they were supposed to do but later really did want out because they were humiliated and oppressed, what is there left to be surprised about? Did Zimbardo and his colleagues really need to do this experiment to “prove” such an obvious result? And does this really tell about the behavior of real guards and prisoners, who are not there for a two-week experiment (for the good of science), but are there because that is how they make their living or because they are being punished for a crime? There is no way to simulate the real experiences of being a guard or a prisoner.
Too often, in our psychological research labs, we trivialize reality. Zimbardo’s prison experiment is a good example of that. If I had included Zimbardo’s experiment in my textbook, it would have been to make that point.
What do you think? Do you still find value in the famous prison experiment, and if so what is that value? This blog is a forum for discussion, and your comments and questions are valued and treated with respect by me and other readers. As always, I prefer if you post your comments and questions here rather than send them to me by private email. By putting them here, you share with other readers, not just with me. I read all comments and try to respond to all serious questions, if I feel I have something useful to say. Of course, if you have something to say that truly applies only to you and me, then send me an email.
Addendum, Feb. 18, 2014 -Peter Gray
Just today I happened to find this quotation, from John Mark, who had been one of the guards in Zimbardo's "experiment." It's from the July/Aug, 2011, issue of the Stanford Alumni magazine, and you can find the whole article here. Here's the quote:
"During the day shift, when I worked, no one did anything that was beyond what you'd expect in a situation like that. But Zimbardo went out of his way to create tension. Things like forced sleep deprivation—he was really pushing the envelope. I just didn't like the whole idea of constantly disturbing people and asking them to recite their prisoner numbers in a count. I certainly didn't like when they put a guy in solitary confinement.
"At that time of my life, I was getting high, all day every day. I got high before I went to the experiment; I got high on my breaks and lunch. I got high afterwards. I brought joints with me, and every day I wanted to give them to the prisoners. I looked at their faces and saw how they were getting dispirited and I felt sorry for them.
"I didn't think it was ever meant to go the full two weeks. I think Zimbardo wanted to create a dramatic crescendo, and then end it as quickly as possible. I felt that throughout the experiment, he knew what he wanted and then tried to shape the experiment—by how it was constructed, and how it played out—to fit the conclusion that he had already worked out. He wanted to be able to say that college students, people from middle-class backgrounds—people will turn on each other just because they're given a role and given power.
"Based on my experience, and what I saw and what I felt, I think that was a real stretch. I don't think the actual events match up with the bold headline. I never did, and I haven't changed my opinion."
Notes and References
 Peter Gray (1st edition, 1991; 6th edition, 2011) Psychology. Worth Publishers. [The7th edition, which will appear soon, has a terrific new co-author, David Bjorklund, who has taken over the full task of revisions.]
 Haney, C., Banks, C., & Zimbardo, P. (1973). Interpersonal dynamics in a simulated prison. International Journal of Criminology & Penology, 1(1), 69-97.
 Banuazizi, A., & Movahedi, S. (1975). Interpersonal dynamics in a simulated prison: A methodological analysis. American Psychologist, 30, 152-160.
 Zimbardo, P. (2007). The Lucifer Effect: Understanding How Good People Turn Evil. Random House.
 Prescott, C. (April 28, 2005). The lie of the Stanford Prison Experiment. Stanford Daily, in News section.