The Stanford Prison Experiment and the doubts that followed

What actually happened, and what's disputed

In August 1971, the psychologist Philip Zimbardo built a mock prison at Stanford University and ran what became one of the most cited studies in psychology, with a famous claim: that ordinary people turn cruel simply because they are assigned the role of guard. The study itself is real, and it did end early. What is disputed is Zimbardo's own explanation for it. Archives and interviews gathered since 2018 show the guards were coached and at least one dramatic breakdown was staged, which is why Real or Legend stamps this record disputed rather than confirmed.

Building the mock prison

Zimbardo drew his volunteers from more than 75 people who answered newspaper ads in the Palo Alto Times and The Stanford Daily, paying each one $15 a day. Twenty-four male college students were chosen: 12 assigned to play prisoners, nine of them plus three alternates, and 12 assigned to play guards, also nine plus three alternates, according to the study's own site, prisonexp.org. The plan was to run the simulation for two weeks.

Audio recordings later placed in Stanford's archives caught a member of the research staff, identified by Prison Legal News as Zimbardo's assistant David Jaffe, instructing guards to "be tougher with the prisoners." In a separate excerpt examined by the psychologists Jay Van Bavel, Alex Haslam and Stephen Reicher for a 2019 Scientific American analysis, a staff member tells one guard the team wants him "active and involved," adding that "every guard is going to be what we call a tough guard," and repeats the word "we" 57 times in that one conversation.

Six days instead of two weeks

By the study's own account, the guards grew sadistic and the prisoners grew depressed and showed signs of extreme stress, so Zimbardo ended the study after six days instead of the planned two weeks. One prisoner, Douglas Korpi, had a breakdown so dramatic that it became the experiment's best-known scene.

Two accounts of one breakdown

The sharpest criticism of the study concerns that scene. Korpi, later a forensic psychologist in Oakland, told the writer Ben Blum, in reporting carried by Inside Higher Ed, that the breakdown was acting: "Anybody who is a clinician would know that I was faking. If you listen to the tape, it's not subtle." He said the outburst was really about losing time to study for an upcoming graduate school admissions test, not fear of the guards.

Zimbardo disputes that reading. In a written response posted on prisonexp.org, he argues researchers have a duty to treat an apparent breakdown as genuine at the moment it happens, and notes that Korpi went on to spend years working inside the prison system: "The Stanford Prison was a very benign prison situation, and it still caused guards to become sadistic."

What the guards say now

Guard Dave Eshelman, a Chapman University student at the time who was later nicknamed "John Wayne," told Stanford's alumni magazine that he had been "running my own experiment in there," pushing to see "how far can I push these things and how much abuse will these people take," and modeling his manner on the warden in the film Cool Hand Luke. He later recognized parallels to the Abu Ghraib prisoner-abuse scandal.

John Mark, a Stanford junior who worked the calmer day shift, remembered it differently: "No one did anything that was beyond what you'd expect in a situation like that." He went on to question whether Zimbardo had, in words reported by Inside Higher Ed, tried to shape the experiment to fit a conclusion he had already worked out before it began.

A rival explanation: identity, not roles

The researcher Thibault Le Texier examined the Stanford archives and published a book-length critique in 2018 called "History of a Lie," which Zimbardo names in his own online response to critics. Zimbardo has also dismissed a 2002 BBC-televised restaging of the study, run by the psychologists Alex Haslam and Stephen Reicher, as unscientific because of constant researcher intervention and television cameras. Haslam and Reicher argue instead that cruelty follows when people identify with leaders who frame harm as necessary for a larger cause, not from the assigned role alone, and note that some guards in the Stanford recordings actively resisted staff pressure to be brutal.

Zimbardo's defense

Zimbardo has answered his critics point by point on prisonexp.org. He says he told guards to be "firm" and "in the action" but banned physical force, and argues that because some guards stayed decent throughout the study, cruelty was still each guard's own choice rather than a scripted outcome. He also says an op-ed once attributed to his prison consultant, Carlo Prescott, was ghostwritten by a screenwriter, Michael Lazarou, who was seeking film rights.

Ethics, method and where the study stands now

The study's early publications were never peer-reviewed, and full methodological detail was not made public for nearly 50 years, according to Prison Legal News. Zimbardo has said that gap reflected funding deadlines and an editor's invitation, not an attempt to dodge review. A 2007 study by the researchers Thomas Carnahan and Sam McFarland found that the original ad's wording, which named "prison," would tend to draw in people "already interested in the authoritarian nature of prison settings."

Many introductory textbooks still leave this record out, and the ethics of putting students through it are part of the debate. A 2017 academic paper found that 95 percent of criminologists who cited the study did so without skepticism, and psychology textbooks have typically presented it with little critique, Prison Legal News reported. That is changing: the psychologist Simine Vazire has called the study "anti-scientific" and urged that it be dropped from textbooks, and Haslam has said the original work "has no credibility whatsoever now."

Questions people ask