Stanford Prison Experiment Debunked
Philip Zimbardo's 1971 Stanford Prison Experiment — long taught as proof that ordinary people turn sadistic when given power — was substantially debunked by 2017-2019 evidence showing guards were coached, the famous breakdown was faked, and the 'study' was closer to improvised theater than science.
The **Stanford Prison Experiment** (SPE) was conducted by psychologist Philip Zimbardo in August 1971 at Stanford University. Student volunteers were assigned as 'guards' or 'prisoners' in a mock prison; the study was cut short after six days amid reported escalating brutality. For decades it was taught as definitive evidence that 'ordinary people' become cruel when given institutional power. By 2017-2019, archival investigation had established that most of this narrative was wrong. ## Debunking evidence - **Guards were coached.** Archival tape has researcher David Jaffe telling a guard: 'We really want you to get active and involved because the guards have to know that every guard is going to be what we call a tough guard.' Guards were explicitly instructed to induce the behaviors Zimbardo later attributed to emergent dynamics. - **The breakdown was faked.** Douglas Korpi (Prisoner #8612), whose dramatic collapse became the iconic event, admitted in 2017 he faked it to get released to study for his GREs. 'Anybody who is a clinician would know I was faking.' - **The 'sadistic' guard was performing.** Dave Eshelman (the 'John Wayne' guard) studied acting and said: 'I took it as a kind of improv exercise. I believed I was doing what the researchers wanted me to do.' - **Zimbardo wasn't neutral.** He was 'superintendent' and actively ran the scenario, including briefing guards. - **Thibault Le Texier's 2019 archival investigation** (*Histoire d'un Mensonge*) showed conclusions were largely written in advance, guards were given precise instructions, and data collection was biased and incomplete. - **BBC replication (2006).** The BBC Prison Study (Reicher & Haslam) used similar setup without coaching and the prisoners overthrew the guards — the opposite of Zimbardo's result. - **Selection bias.** SPE ads read 'psychological study of prison life,' attracting participants scoring higher on aggression and authoritarianism per a 2007 study. ## What actually happened SPE is now understood as a demonstration of **demand characteristics** — participants inferring what the experimenter wanted and performing it. This matters: demand-characteristic effects are real, but they don't show that authority spontaneously produces cruelty. They show that people coached and incentivized to perform cruelty will sometimes do so. ## Milgram by contrast Stanley Milgram's obedience experiments have held up better, partially replicated (Burger 2009). But with caveats: people conditionally defer to authority they identify with and believe is legitimate, not blindly. Revisionist work (Reicher, Haslam) frames Milgram as engaged followership rather than raw obedience. ## Why debunking matters The 'human nature' reading of SPE has real-world costs. It framed institutional cruelty (prison abuse, interrogation programs, Elan School and the Troubled Teen Industry) as inevitable when under-supervised people get power — absolving structure. Real-world institutional cruelty, when examined, always points to specific fixable conditions: method (Synanon-derived attack therapy), training, peer enforcement, political protection, absent oversight. Those are engineering problems with engineering solutions. The SPE narrative let institutions off the hook. The actual evidence from places like Elan points the other direction.