The famous “obedience” experiments by Stanley Milgram: what did they really show?
This year marks the 50th anniversary of the publication of Stanley Milgram’s experiments on getting ordinary Americans to apparently torture others in obedience to a stranger in a white lab coat.
Many felt that Milgram’s work was unethical, as Gina Perry clearly does in her Behind the Shock Machine: The Untold Story of the Notorious Milgram Psychology Experiments (2012). Certainly it would not be permitted today.
Context is useful. 1963 was less than two decades after the Holocaust had become general knowledge. Key player Adolf Eichmann claimed at his 1961 trial that he was only following orders. Was he? In any event, the Cold War threatened to institutionalize such behaviour, famously termed “the banality of evil”. Could it? New Yorker Milgram (1933–1984), then teaching at Yale, wanted to find out.
Psychology, as always, was striving for the certainty of a hard science. One of Milgram’s high school classmates was Philip Zimbardo, the psychologist who developed the Stanford Prison Experiment and he also studied under Solomon Asch, developer of a (non-traumatic) conformity experiment involving perception.
It was also an age when Candid Camera hoaxed ordinary people by creating bizarre situations, broadcast on national TV. So, while the experiments were controversial even back then, mid-century culture was the backdrop.
So what did Milgram do? In 1961, he recruited participants to come to his Yale lab for an experiment on the effects of punishment on learning. Each subject was assigned the role of “teacher.” An apparent fellow volunteer was assigned to be the “learner,” who was audible but not visible. Whenever the learner, made an error in reciting a list of word pairs, the teacher was to give him an electric shock via a realistic-looking machine, and increase the voltage on each error. The levels were labeled from "SLIGHT SHOCK" to "DANGER—SEVERE SHOCK" and finally "XXX." The learner, a Milgram associate of course, screamed convincingly from pretend shocks and begged to be released, all according to protocol. The “teacher” (the subject) was repeatedly urged to continue anyway.
Milgram reported that 65 percent of his subjects continued to shock the learners even when the latter protested, cited heart problems, and then went silent. He later described the typical subject as one who “divests himself of responsibility” becoming an “agent of external authority.” (One hitch was, obedience cratered when the participants knew each other.)
As Tom Bartlett puts it in Chronicle of Higher Education, “Most of us, in other words, are potential Nazis.” Indeed, that is the version taught to introductory psychology classes to this day.
But there is another, more nuanced version of the story. First, contrary to the “shocked, shocked” claims one often hears, Milgram was reporting what everyone in his day who had read George Orwell’s then-futurist classic 1984 (1948) actually expected to hear.
But… was it true?
Australian psychologist Gina Perry, who diligently studied the original experiment data and the still living subjects, says that he misrepresented his findings, that the “zombie-like, slavish obedience” wasn’t what he’d observed.
First, the experiment was messy, like a pilot project, rather than the carefully designed, systematic study that the public was led to assume. There were 24 variations involving 780 people, virtual mini-dramas, featuring different scripts, actors and set up. Violations of the research protocol were routine.
Significantly, only the widely reported variation, involving 40 subjects, achieved 65% compliance. Perry notes, “By examining records of the experiment held at Yale, I found that in over half of the 24 variations, 60% of people disobeyed the instructions of the authority and refused to continue.” This despite the bullying and coercion that, she reports, comes out in the tape recordings she transcribed. In any event, between a quarter and a half of subjects suspected a ruse.
In short, a more accurate summary would read that more than half the time, fewer than half the subjects complied, despite pressure. And one quarter to one half of those who did so doubted they were hurting anyone.
So Milgram never really demonstrated that Americans of his day would easily accept Big Brother in a lab coat. That’s still not great news for human nature, considering that the subject were citizens of a free society not at war, and they had not been taught to demonize the “learners.”
And, as Wall Street Journal reviewer Carol Tavris notes, his findings were replicated elsewhere.
Replicated all too often, one fears, by real life. Today, Sabina Cehajic-Clancy, a social psychologist, struggles to make sense of Sbrenica, where the Serbian army killed 7000 Bosnian Muslims (Bosniaks) in 1995. Many people alive today are to some extent complicit in the worst mass murder in Europe since World War II—but most people she contacts don’t want to talk about it and have little to say when they do.
Tom Bartlett, who profiles her, cites theories about the evolved behaviour of macaques toward outgroup members, by way of explanation. Is this really an improvement over the shadow of Big Brother? One get the distinct sense that, faced with the mystery of evil, social science over the decades does not have so much to tell us as was once fondly hoped.
Denyse O'Leary is co-author of The Spiritual Brain.