image copyrightBBC/Serenity Strull
In June 1961, an ad appeared in the Connecticut newspaper New Haven Register, recruiting people for an innocuous-sounding experiment at Yale University: "We will pay you $4 for one hour of your time. People needed for a memory study."
Factory workers, barbers, businessmen and a wide variety of other professionals were wanted, according to the advertisement. No special education or experience was needed. Those interested had to send their data by mail to Stanley Milgram, professor in the Yale Department of Psychology.
Forty men ended up signing up for the initial trial of the experiment. They entered a laboratory and, under the guise of a word learning task, were asked to administer electric shocks to a stranger, which could reach a potentially lethal level.
Most of the men - 26 of the 40, or 65% - ended up delivering the maximum, lethal shock. In subsequent studies with other volunteers, most applied the maximum shock even if the victim presented himself as having a heart problem or could be heard pleading to be released.
The men had no idea that the shocks were actually fake and that the entire study was rigged. It wasn't about memory at all. Instead, it was a study in obedience and cruelty, which would become one of the best-known examples of ordinary people doing brutally unthinkable things when ordered to do so in a laboratory.
Since then, researchers have used milder, modified versions of Milgram's experiment to study one of the most disturbing aspects of human nature: how, why, and when we obey an immoral order.
Their findings, along with details from Milgram's original experiments, not only shed light on some of humanity's worst impulses, but also offer surprising insights into what can help us resist.
"We were speechless"
It is difficult to imagine an ordinary person inflicting extreme harm on a human being. Most people are convinced that they themselves would not do it, and that they would resist in an experiment like Milgram's. Even psychologists who research this behavior can be surprised when it happens before their very eyes.
End of Most read
All the news you need is on BBC World Service
Download it here
End of Ad app
"It was pretty impressive to see these people, who seemed perfectly normal, happy, wonderful individuals, pull those levers and not object," says Jerry Burger, professor emeritus of social psychology at Santa Clara University in California.
In 2009, Burger conducted a modified version of Milgram's study, stopping the volunteer when he applied a minor (false) shock. He expected his results to be approximately the same as Milgram's. But seeing the participants on a monitor, he and his team were still surprised. "At first we were speechless: I can't believe this person is doing this!" he recalls.
In Burger's 2009 study, 70% of participants obeyed until the end of the experiment. Unlike Milgram, Burger tested both men and women and found no significant differences.
"We like to think of ourselves as independent, moral people who would resist an order if it conflicted with our values," says Tomasz Grzyb, a professor of social psychology at Poland's SWPS University, who conducted a partial replication study in 2015. "But the evidence suggests we should be much less sure of that. Under enough pressure, in the right social situation, people can find themselves doing things they don't really want to do."
image copyrightGetty Images
In Grzyb's replication study, which also included lower maximum shock levels than Milgram's study, 90% of participants complied until the end. There were no significant differences between men and women.
After the experiment, a psychologist talked to the volunteers about their experience to make sure they were okay. A young woman "started crying very intensely," says Grzyb.
"Through tears, she said something like, 'This is the kind of person I am. I do what people tell me to do. And only now, after participating in this experiment, have I realized it. That's why things in my life happen the way they do.'"
What explains such obedience, even when it conflicts with our personal values?
Disobeying consumes more brain resources
Emilie Caspar, a professor of experimental psychology at Ghent University in Belgium, studies how obedience influences human brain activity. In their studies, volunteers administer real, mild electric shocks to another volunteer.
Caspar and his team use brain imaging methods, such as functional magnetic resonance imaging (fMRI), to observe what happens in the volunteer's brain when they deliver a shock on command; and, separately, when you manage a download because you freely choose to do so.
"The shocks are real, they are not fake like in Milgram, but they are calibrated according to the victim's pain threshold," he explains. "It's real and painful, but tolerable for the victim."
Caspar and his team have discovered that several brain processes are different when a person obeys a command compared to when they freely choose.
For example, when people deliver shocks under command, regions associated with empathy, guilt, and a sense of agency or responsibility are less active compared to when they deliver shocks freely.
When they are under an order, they also find the shocks they are delivering to be less painful. This may explain how obedience clouds morality and allows people to cause harm to others, Caspar suggests.
In fact, in his own experiments, "people are extremely compliant," Caspar says. Of about 800 volunteers in their studies, in different countries, only 21 (2.6%) refused to deliver the shocks. He recalls one participant who did not deliver any shocks when he had free will, but completely complied with all of his commands: "He said, 'When you started giving me orders, it was so easy to obey the order.'"
"That's actually one of the key things (about obedience)," he says, referring to the volunteer's feeling that it's easier than disobeying. He points out that empathy, guilt and responsibility are brain processes that require energy. When we follow orders and these feelings are reduced, it can feel less effortful than disobeying.
"The brain is an organ that does not want to spend too many resources when it is not necessary," he explains. "Disobeying consumes more brain resources than simply following orders."
While we might think that individual personality traits, like empathy, would help resist an evil order, research tells a different story.
"Personality and temperament are surprisingly poor predictors of who will obey and who will refuse in a given situation," Grzyb says.
Empathy may, in fact, be a double-edged sword, the research suggests: People may empathize with someone who gives immoral orders and not want to disappoint them.
However, there is something we can do to help us make better moral decisions, the research indicates: understand the powerful social and environmental factors that influence our decisions, not just in a laboratory, but in everyday life.
"If they sue us, who is responsible?"
Burger highlights four specific aspects of Milgram's study—and especially, the way it was organized—that help explain why so many people complied.
Milgram's setup was simple. Upon entering the laboratory, each volunteer was greeted by an "experimenter," a scientist in a lab coat (actually an assistant acting out the role). You were asked to select your task at random (but it was already predetermined) and you were assigned the role of “teacher” in a rote learning task.
He was then instructed to evaluate a "trainee" (another assistant acting out that role), who was sitting in another room, apparently hooked up to an electric shock machine. The teacher was instructed to shock him every time he gave an incorrect answer by pressing a series of buttons.
In both the original study and the replications, the experimenter took full responsibility for anything that happened to the victim.
This, the experimenter taking responsibility, was crucial, says Burger.
"We now know from research that if you make people anonymous, or say it's someone else's responsibility, they can engage in very bad behavior," he says.
In Burger's own study, one participant turned to the experimenter and asked, "There's not going to be any demand, is there?" The experimenter responded, "If something happens to him, I'm responsible." The woman then said, "That's what I needed to hear!"
"And she turned around and pushed the buttons," Burger says.
In fact, independent research has shown that there is a modern illustration of this anonymity effect: online behavior. Anonymity is linked to cyberaggression, according to studies from 2023 and 2026.
"People say and do all kinds of horrible things (on the Internet)," says Burger. "If you sat face to face with someone and took the consequences of your words, you would be less likely to do it. Anonymity allows you to get rid of personal responsibility."
In Milgram's study, after the experiment the volunteers were asked who was responsible for the person receiving the electric shocks.
Those who obeyed tended to attribute primary responsibility to the experimenter. Those who refused often said that the main responsibility fell on themselves.
Milgram also analyzed what happened when the victim was in the same room as the volunteer. In that situation, only 40% obeyed the order until the end; and when the victim's hand touched the volunteer's, only 30% obeyed.
Milgram suggested that proximity may have made the electrical discharges less abstract. As Burger points out, it's harder to say or do mean things when you're face to face with a person.
"Am I the kind of person who does this?"
Burger highlights two other factors: In Milgram's experiment, people found themselves in a situation that most had never experienced before. And then they didn't have time to think about what to do. Whenever they hesitated, the experimenter encouraged them with phrases such as: "You have no choice, you must continue."
When we have to make a decision in a new situation, we tend to look for someone close to us with more experience and take references from their behavior, explains Burger. In Milgram's study, the only other person in the room was the experimenter, who insisted on continuing. That, along with the rapid sequencing, made it difficult for volunteers to form their own opinions, Burger says.
If the experimenter had said, "why don't you take 10 minutes to think about what to do?" there probably would have been more disobedience, Burger believes. A pause would have given the volunteers a chance to ask themselves: "Is this right or wrong, am I the kind of person who does this?" he argues.
The gradual escalation of electrical shock levels was also a factor, Burger adds. If a person accepts a small request, they are more likely to accept larger requests as well, he says, a phenomenon known as the "foot in the door" effect.
In real life, one way to counteract these effects may be to think about a situation in advance, Burger suggests, and a novel situation, and try to take some time to reflect.
Grzyb suggests that cultivating certain habits of mind can be helpful, such as "creating psychological distance from the authority figure, recognizing when responsibility is shifting away from us, asking whether an instruction is legitimate, and giving ourselves permission to stop."
But a key lesson is that "we should build systems that make it difficult to abuse authority," Grzyb says. "If one person is given power without control over another, sooner or later there is a serious risk that that power will be misused."
"We can always choose another option"
There is another powerful source of moral courage: seeing others act morally.
Caspar has conducted research on former perpetrators, witnesses and actual rescuers in the 1994 genocide against the Tutsi community in Rwanda.
"One of the main factors that distinguishes rescuers from perpetrators of genocide is the presence of role models in their early childhood," he says, based on his team's research. "For example, rescuers are more likely to have witnessed their parents or other family members helping other human beings."
He points out that even if a small minority disobeys an immoral order or rescues people in an actual genocide, it shows that it can be done.
"Rescues are a good reminder that we can always choose another option," he says.
In fact, even in Milgram's original study, there were those who resisted.
"No, this is not right. It's the devil's experiment. The guy is suffering in there. No, I don't want to continue," said a disobedient volunteer. Told he had no choice but to continue, another snapped: "If this were Russia, maybe, but not in the United States."
Joe Dimow, a disobedient participant in Milgram's original study, wrote an essay in 2016 about his experience. He said his upbringing in a family that questioned authority, as well as his own political activism and run-ins with authorities, likely helped him resist the experimenter's orders.
"I had an unorthodox understanding of authority and was not likely to be impressed by a white coat," he wrote.
What blind obedience cannot explain
Caspar's research also highlights the crucial phases in which we might make a moral decision, long before we actually find ourselves in a situation where obedience seems the only way out.
This fills an important gap left by Milgram. Milgram argued that his findings could help explain how seemingly ordinary people ended up committing atrocities, in particular how so many Germans conspired and carried out the murder of millions of Jews in the Holocaust.
But as many have pointed out, blind obedience doesn't even begin to explain the complexities of actual genocide. In the Holocaust, for example, persecution occurred largely without direct orders and involved acts of pure cruelty and greed.
"Genocide is usually described as a process of stages, where there is discrimination, categorization, dehumanization (in an escalating sequence)," says Caspar. "The obedience part is more in the final phase, the extermination phase. It is like a trigger, telling people that they can go ahead (with the slaughter)."
Refusing to demonize a group of people and cultivating a belief in shared humanity can help combat persecution in those early stages, he says.
A line from Milgram's transcripts, quoting one of the disobedient volunteers, perhaps puts it best: "I'm sorry, I can't do that to a man," the volunteer insisted, and stopped.
Discover more
Video
End of Video
