Basic Instincts: The Science of Evil

Why do ordinary people sometimes commit unthinkable acts?


Jan. 3, 2007 — -- Most of us have struggled to understand how seemingly ordinary people can sometimes do morally questionable things.

Two years ago, the photos of young American soldiers smiling while torturing Iraqi prisoners at Abu Ghraib horrified the world and raised the question of who was to blame.

Some of the soldiers defended themselves by claiming they were just doing what their superiors had instructed. But the smiling faces in the photos seemed to imply that they followed the orders without protest.

Are those soldiers inherently bad people? Or is it more complex than that? Do you have to be an evil person to do evil things?

In 1961, social psychologist Stanley Milgram asked those same questions. That was the year Nazi Adolf Eichmann, on trial for his war crimes, denied responsibility for his actions by saying he was simply doing what his superiors told him to do.

Contemplating this rationalization, Milgram came up with a famous and controversial experiment to examine what happens when ordinary people are faced with morally questionable orders. What he learned shocked not only him but the entire world.

In the experiment, conducted at Yale University over a period of months in 1961, an authority figure -- "the experimenter" -- dressed in a white lab coat and instructed participants to administer what they believed were increasingly painful electric shocks to another person.

Although no one was actually receiving shocks, the participants heard a man screaming in pain and protest, eventually pleading to be released from the experiment. When the subjects questioned the experimenter about what was happening, they were told they must continue.

And continue they did: Two-thirds of Milgram's participants delivered shocks as they heard cries of pain, signs of heart trouble, and then finally -- and most frightening -- nothing at all.

The response to the experiment was enormous, and in 1975, strict guidelines about regarding psychological experiments on humans shelved any further potential replications. Since then, scientists have been stymied in efforts to replicate Milgram's study.

"Primetime" wanted to know if ordinary people today would still follow orders, even if they believed their actions were causing someone else pain. Would as many follow the seemingly dangerous and painful orders as in the original experiment? After contacting respected psychologist Jerry Burger at Santa Clara University in California, ABC News was able to replicate Milgram's study in a modified way.

After the American Psychological Association provided feedback on the testing protocol, this collaboration between "Primetime" and Santa Clara University marks the first time in decades that the famous study has been re-created.

Burger said, "People have often asked the question, 'Would we find these kinds of results today?' and some people try to dismiss the Milgram findings by saying, 'That's something that happened back in the '60s. People aren't like that anymore.'"

After placing an ad in the paper looking for participants for "a learning and memory study" and putting the respondents through psychological screening, "Primetime" found 70 people lined up for the experiment.

One of the first participants in the study was Troy, a 39-year-old electrician. Like all the participants, he was paid $50 and was told that the money would be his to keep, even if he quit the experiment early. Brian, in the role of the "experimenter," informed Troy that he was taking part in a learning and memory study and would be teaching word pairs to Ken, who was really a plant in the experiment.

If Ken got a word pair wrong, Troy was instructed to punish him with an electric shock from another room. The more word pairs Ken answered incorrectly, the more intense the shocks seemed to become. After getting a few wrong, at 75 volts, Troy heard what he thought was Ken shouting in pain -- but it was really an automatic audio cue that was set to go off at that voltage.

Each shock after that triggered a similar audio cue of pain. At 105 volts, Troy became uncomfortable. At 150 volts, he heard Ken plead, "That's all. Get me out of here. I told you I had heart trouble. My heart's starting to bother me. … Let me out!" Troy looked questioningly at the experimenter, who told him he must continue. Though he was clearly uncomfortable, Troy continued with another word pair before the experiment was stopped.

After the experiment, Troy said, "I was not comfortable. I cannot tell you why I listened to him and kept going. I should have said no."

When asked why he didn't stop administering the shocks, Troy explained, "I was doing what I was supposed to do, and I'm there to help conduct an experiment, so I'm just doing my part."

Troy's response is easy to understand, according to Burger. "The typical response is to turn toward the experimenter and if not to say something, at least give a look that says, 'What should I do?' And of course, when an expert tells them, 'Not a problem. This is nothing to worry about, continue.' The rational thing to do in that situation is to continue."

Milgram's original experiment tested just a handful of women, but "Primetime's" sampling was approximately half men and half women. Would the "gentler" sex be more reluctant to shock someone? And what about the people who refused to continue to shock the subject after hearing his demand to be released? What made them choose to stand up to authority?

In ABC News' version of the Milgram experiment, we tested 18 men, and found that 65 percent of them agreed to administer increasingly painful electric shocks when ordered by an authority figure.

22 women signed up for our experiment. Even though most people said that women would be less likely to inflict pain on the learner, a surprising 73 percent yielded to the orders of the experimenter.

Out of the 30 people we tested with an additional accomplice acting as a moral guide, 63 percent still inflicted electric shocks, even though the accomplice refused to go on.

Our subjects had an unusually high level of education. 22.9 percent had some college, 40 percent had bachelor's degrees and 20 percent had master's degrees.

The group was also ethnically diverse with 54.3 percent (white), 18.6 percent (Asian), 12.9 percent (Latin/Hispanic), 8.6 percent (Indian-Asian) and 4.3 percent (African -American).