Twilight of the Stanford Prison Experiment

The infamous experiment was even more deeply flawed than previously suspected.

Posted Sep 27, 2019

The infamous Stanford Prison experiment (SPE), conducted in 1971—in which Philip Zimbardo recruited young men to become either "prisoners" or "guards" in a mock prison, with disastrous results—has long drawn criticism for its sloppy methodology and the exaggerated conclusions about the psychology of evil that Zimbardo drew from it.

LightField Studios/Shutterstock
Source: LightField Studios/Shutterstock

In the subsequent decades, Zimbardo has repeatedly claimed that the SPE illustrated the “power of the situation” in driving good people to behave in cruel and dehumanizing ways. However, critics of the SPE have long suspected that Zimbardo’s analysis was far from the truth.

In 2018, archival data about the SPE was made available online, which permitted a much more thorough investigation than has previously been possible of what really happened during this experiment. An analysis of this data by Le Texier (2019) shows that Zimbardo misrepresented his findings and that his conclusions about the “power of the situation” are untenable.

Zimbardo claimed that his experiment demonstrated that “individual behavior is largely under the control of social forces and environmental contingencies, things that occur, rather than some vague notions of personality traits, character, willpower, or other empirically unvalidated constructs” (cited by LeTexier, 2019). He even went so far as to claim that the SPE extended the work of Milgram, whose famous obedience experiments showed that people could be induced to deliver electric shocks to someone against their will at the behest of an authority figure.

Zimbardo actually claimed that the SPE showed that an authority figure was not even necessary to induce people to behave badly. Simply “having participants embedded in a social context where the power resided in the situation” was enough, as participants in his experiment supposedly adopted the roles they were assigned and acted accordingly. According to Zimbardo, aggression by those assigned to be guards was “emitted simply as a ‘natural’ consequence of being in the uniform of a ‘Guard’ and asserting the power inherent in that role” (Le Texier, 2019).

However, LeTexier’s analysis shows that Zimbardo had actually decided in advance what conclusions he wanted to demonstrate. For example, on only the second day of the experiment, he put out a press release stating that prisons dehumanize their inmates and therefore need to be reformed. Moreover, contrary to his repeated claims that participants in the experiment assigned to the role of guards were not told how to treat the prisoners and were free to make up their own rules, the archival data clearly show that the guards were told in advance what was expected of them, how they were to mistreat the prisoners, and were given a detailed list of rules to follow to ensure that prisoners were humiliated and dehumanized.

Furthermore, Zimbardo and his research team were highly assertive in ensuring that participants acted as “tough guards,” contrary to Zimbardo’s claims that they just fall naturally into their roles. For example, in the orientation session for guards on the first day of the experiment, Zimbardo’s assistant David Jaffe, who acted as a prison warden, even read out a list handwritten by Zimbardo entitled: “Processing in—Dehumanizing experience,” that included instructions like, “Ordered around. Arbitrariness. Guards never use names, only number. Never request, order.” This contradicts Zimbardo’s claims that dehumanizing behavior like calling the prisoners by their numbers rather than their names was something the guards came up with themselves. Additionally, after the experiment, some of the guards stated that either Zimbardo or Jaffe had directed them to act in specific ways at various times during the study.

Not surprisingly, some of the guards did mistreat the prisoners. Indeed, one guard, nicknamed “John Wayne” by the prisoners went out of his way to do so. More interesting though, is that some of the guards resisted mistreating the prisoners, despite being under considerable pressure to do so. For example, the archive contains an audio recording of a formal meeting between Warden Jaffe and a guard, John Mark, who had been unwilling to treat the prisoners harshly (Haslam, Reicher, & Van Bavel, 2019).[1]

During the meeting, Jaffe repeatedly remonstrates with Mark, telling him that for the experiment to “work,” they need all the guards “to be what we call a tough guard.” Jaffe makes clear that the aim of the experiment is to simulate a prison in which all the guards are tough, so they can study how this affects the prisoners, and without the guards’ cooperation, “the experiment falls apart.” This interaction makes clear that the experimenters were actively attempting to persuade the participants to behave in a certain way by convincing them that mistreating the prisoners had a valid scientific purpose. Despite this, Guard Mark did not seem very convinced by Jaffe’s arguments, as he did not accept the stereotype of the tough guard and did not see himself getting into this kind of role. 

In their analysis of the archival data, Haslam et al. (2019) have argued that this shows that, contrary to Zimbardo’s role-based account of how participants behaved, it is actually quite difficult to induce people to mistreat others and that whether the guards mistreated the prisoners or not depended to a great extent on whether they were persuaded to follow the leadership of the experimenters—specifically, their appeals to the scientific importance of the experiment.

Hence, their behavior was not an automatic response to the “power of the situation” but was to some extent a reflection of their own values and desires. While some guards, like “John Wayne,” enthusiastically embraced this leadership, others, like John Mark, chose to actively resist it. This accords with the argument I made in a previous article critiquing the SPE that, far from showing that situational forces outweigh or overpower individual factors, such as personality, how people behave is a function of both environmental demands and individual characteristics of the person.

One thing I find striking about the SPE is that the experimenters worked very hard to produce a predetermined result, to push people to behave a specific way, and yet they were only partially successful. Similarly, in Milgram’s obedience experiments, even when participants were under considerable pressure to obey the experimenter’s commands to continue shocking the learner against his will, about a third or more of participants refused to do so. Hence, even strong situations are not quite as powerful as certain social psychologists have asserted.

In a previous article, I discussed the psychology of “moral rebels”: people who resist situational pressure to behave in ways that go against their moral values. The study I discussed in that article (Sonnentag & McDaniel, 2013) suggested that highly moral individuals may regard morality as more central to their identity than less moral individuals, and this "trait integration" may give them the courage of their convictions that allows them to resist pressure to go against their own values. This is worth considering when discussing the psychology of evil.

Much of Zimbardo’s discourse on this subject has amounted to absolving people of responsibility for their own choices. For example, he has often drawn parallels between prisoner abuse at Abu Ghraib and what happened in the SPE, stating, “I believed our soldiers were good apples that someone had put into a very bad barrel in that prison dungeon.” In other words, it’s not the person, but the situation that is responsible. On the contrary, a close examination of the SPE shows that it takes a lot more than simply putting on a uniform to make someone act badly. In fact, even when assertively pressured to mistreat others, it is still possible to resist if one is willing to do so.

Considering moral rebels as an example suggests that supposedly “empirically unvalidated constructs” like character may actually make an important difference. It could be that people who are very clear about their moral values may be more able to resist attempts to persuade them to behave in inhumane ways, even if these are justified by appeals to some alleged higher purpose.

[1] The tape can be accessed online here; the interview starts at 8:38 minutes. A written transcript is available here.


Haslam, S. A., Reicher, S. D., & Van Bavel, J. J. (2019). Rethinking the nature of cruelty: The role of identity leadership in the Stanford Prison Experiment. The American Psychologist.

Le Texier, T. (2019). Debunking the Stanford Prison Experiment. The American Psychologist.

Sonnentag, T. L., & McDaniel, B. L. (2013). Doing the Right Thing in the Face of Social Pressure: Moral Rebels and Their Role Models Have Heightened Levels of Moral Trait Integration. Self and Identity, 12(4), 432–446.