The Milgram Experiment — Obedience to Authority

In the summer of 1961, while Adolf Eichmann was on trial in Jerusalem for orchestrating the Holocaust, a young Yale psychologist named Stanley Milgram set out to answer a question that obsessed the postwar world: how could ordinary Germans have followed orders to commit mass murder? Were they monsters — or could anyone be turned into one with the right setup?

His answer, published in 1963 under the dry title "Behavioral Study of Obedience," became the most famous and most disturbing experiment in the history of social psychology.

The Setup

A volunteer arrives at a Yale lab, where an experimenter in a gray lab coat introduces them to another "participant" (actually a confederate, an actor named Jim McDonough). Lots are drawn — rigged so the real volunteer always gets the role of "teacher" and the actor plays the "learner."

The learner is strapped into a chair in another room, wired with electrodes. The teacher sits at an imposing shock generator labeled with 30 switches in 15-volt increments, from 15V ("Slight Shock") up to 450V ("Danger: Severe Shock — XXX").

The teacher reads word pairs over an intercom. For every wrong answer, the teacher administers a shock — and is told to increase the voltage each time. The shocks are fake; the learner's screams, complaints about a heart condition, pleas to be released, and eventual silence are scripted and recorded.

When the teacher hesitates, the experimenter delivers one of four standardized prods:

1. "Please continue."
2. "The experiment requires that you continue."
3. "It is absolutely essential that you continue."
4. "You have no other choice; you must go on."

The Result That Shocked Everyone

Before running the study, Milgram polled Yale psychology students, professors, and psychiatrists. They predicted that no more than 0.1% — one in a thousand — would deliver the maximum 450V shock. Sadists, essentially.

The actual figure: 65% of participants (26 of 40) in the baseline condition went all the way to 450V. Many were visibly distressed — sweating, trembling, laughing nervously — but they kept flipping switches. Not one stopped before 300V.

It was a thunderclap. The implication seemed clear: ordinary Americans, asked politely by a man in a lab coat, would do something they believed was killing a stranger.

Variations That Mapped the Effect

Milgram ran more than 20 variations between 1961 and 1963. The factors that mattered most:

Variation	Obedience (% reaching 450V)
Baseline (separate rooms, voice feedback)	65%
Learner visible in same room	40%
Teacher had to physically force learner's hand onto shock plate	30%
Experimenter gave orders by telephone	21%
Experiment run in a rundown office in Bridgeport, not Yale	48%
Two confederate "teachers" defy the experimenter	10%
Ordinary man (not lab coat) gave orders	20%
Two experimenters disagree	0%

The lessons drawn at the time: physical proximity to the victim reduces obedience, peer rebellion massively reduces it, prestigious settings increase it, but the situation overall dominates the individual personality. This became the textbook "situationism" view, later reinforced by Zimbardo's Stanford Prison Experiment.

The Ethical Firestorm

Milgram's participants were deceived, distressed, and in many cases traumatized. Audio recordings capture men begging to stop, then continuing anyway. Diana Baumrind's 1964 critique in American Psychologist called the study unethical and accused Milgram of damaging his subjects for theatrical effect.

Modern Institutional Review Boards (IRBs), partly created in response to studies like this and the Tuskegee experiment, would never approve Milgram's protocol today. The standard view is that the deception and lack of fully-informed consent crossed lines that science should not cross.

Milgram defended himself by surveying participants afterward; most said they were glad they took part. Critics argued this was hardly the point.

The Reanalysis That Changed Everything

For 50 years the textbook narrative stood. Then in 2013, Australian psychologist Gina Perry published Behind the Shock Machine: The Untold Story of the Notorious Milgram Psychology Experiments, based on years digging through Milgram's archived audio tapes and notes at Yale.

What she found was damaging:

• The "65%" figure cherry-picks one condition out of many; obedience rates across conditions ranged from 0% to 92%. Milgram's writing emphasized the most shocking number.
• Experimenters frequently deviated from the four-prod script, improvising additional pressure, threats, and reassurances — sometimes a dozen or more interventions.
• Many participants suspected the shocks were fake. Debriefing notes and follow-up interviews show a significant fraction never believed the learner was really being hurt. They acted accordingly.
• Some subjects who "obeyed" were defying the script in their own way — administering shocks lightly, signaling sympathy to the experimenter, trying to game what they took to be a test of the experimenter's nerve.
• Milgram did not always debrief properly. Some participants left convinced they had genuinely hurt someone, and some learned the truth only years later via mailed letters.

Subsequent work by Reicher, Haslam, and Smith (2012) and Gibson (2024) emphasizes that the prods worked not as commands but as appeals to the participant's identification with the scientific project. When the experimenter said "the experiment requires that you continue," it was an invitation to be a co-author of important science — and that worked better than direct orders.

Modern Replications

Jerry Burger (2009), Santa Clara University. A partial ABC News replication, stopping at 150V (the point where most participants in the original who were going to quit, quit). 70% continued past 150V — within range of Milgram's numbers, suggesting the basic effect is real even with modern ethical safeguards.

Polish replication (Doliński et al. 2017). 90% obedience rate (72 of 80 participants pressed the 10th button).

Virtual reality versions (Slater et al., 2006). Participants shocking obvious avatars still showed strong stress responses and many "obeyed" — suggesting the social-pressure mechanism operates even when no one is actually being harmed.

The basic phenomenon — people will inflict apparent harm under expert authority — appears robust. The original 65% number is not the right number to memorize, but the effect is not a fabrication.

Comparison: Classic vs Modern Interpretations

Aspect	Milgram's 1963 framing	Modern reanalysis
Headline figure	65% obedience	One condition; range 0–92% across variations
What participants believed	Believed shocks real	Many suspected fakery
Mechanism	Robotic obedience to authority	Engaged identification with scientific mission
Experimenter prods	Four standardized	Often improvised, many additional
Ethical standing	Defensible by author	Would be rejected today
Theoretical lesson	Power of situation	Power of identification with a cause
Status of effect	Universal trait	Real but context-dependent

Why It Still Matters

Even with all the caveats, Milgram's work changed how we think about evil. Hannah Arendt's "banality of evil" — her phrase for Eichmann's bureaucratic mediocrity — found laboratory confirmation. Real-world echoes are everywhere:

• Nuremberg defense — "just following orders" — is still attempted in war crimes trials.
• Abu Ghraib (2003) — soldiers torturing prisoners under ambiguous command.
• Corporate scandals — Enron, Wells Fargo, Volkswagen Dieselgate — where employees executed clearly unethical orders.
• Healthcare hierarchies — junior nurses overriding their own judgment when senior doctors give instructions, contributing to fatal medication errors.

The Milgram studies are routinely used in medical ethics training, military leadership courses, and business ethics curricula — not as proof that obedience is automatic, but as a warning that authority structures change the moral calculation of normal people.

Related Reading

• The Stanford Prison Experiment is the natural companion — also famous, also troubled.
• See our pieces on how the human brain works and what is consciousness for the cognitive backdrop.
• Depression and mental health covers the psychological aftermath effects participants reported.

FAQ

Q: Were the shocks real? A: No. The shock generator was a prop. The learner was an actor named Jim McDonough whose screams were pre-recorded. Many participants believed the shocks were real; some suspected they were fake.

Q: Is the 65% number reliable? A: That number is from one specific condition (baseline, separate rooms, voice feedback). Across Milgram's variations the rate ran from 0% to 92%. Modern replications find similar broad effects, though never with the exact same conditions.

Q: Would the experiment be allowed today? A: No. Modern IRBs would reject it for inflicting psychological distress and using deception that risked lasting harm. Burger's 2009 replication stopped at 150V specifically to comply with current ethics rules.

Q: Did Milgram falsify his results? A: He did not invent the shocks or the screams. But the archival reanalysis by Gina Perry shows he overstated the consistency of his findings, downplayed deviations from the script, and emphasized the most shocking single number. Most experts now describe his reporting as selective rather than fraudulent.

Q: Does the experiment prove anyone can be turned into a Nazi? A: That was Milgram's framing. The modern view is narrower: under specific conditions of legitimacy, gradualism, and identification with an authority's project, ordinary people can be led to inflict harm they would otherwise refuse. Genuine murderous obedience requires far more than 30 minutes in a lab.

Q: What's the difference between Milgram and the Stanford Prison Experiment? A: Milgram studied obedience to a direct authority figure issuing orders. Zimbardo's SPE studied what happens when ordinary people are assigned social roles (guard vs prisoner) with minimal direct supervision. Both shaped situationist psychology; both have been heavily criticized in recent reanalyses.

Sources

• Milgram, S. (1963). "Behavioral Study of Obedience." Journal of Abnormal and Social Psychology 67, 371.
• Milgram, S. (1974). Obedience to Authority: An Experimental View.
• Perry, G. (2013). Behind the Shock Machine: The Untold Story of the Notorious Milgram Psychology Experiments.
• Burger, J. M. (2009). "Replicating Milgram: Would People Still Obey Today?" American Psychologist 64, 1.
• Haslam, S. A. & Reicher, S. D. (2012). "Contesting the 'Nature' of Conformity." PLoS Biology 10.
• Doliński, D. et al. (2017). "Would You Deliver an Electric Shock in 2015?" Social Psychological and Personality Science 8, 927.
• Gibson, S. (2024). "Reinterpreting Milgram." British Journal of Social Psychology.